-

~ TLAS OIF
Second Edition

ATLAS OF

Regional and Free Flaps for

Head and Neck Reconstruction:
Flap Harvest and Insetting
Mark L. Urken, MD, FACS Keith E. Blackwell, Mo
Professor Prafessor
Department of OtominolaryngoloiJ'f---Haad and Neck Surgery Diviaion af Head and Neck Surgery
Albert Einstein College of Medicine Department af Surgery
Chief of Head and Neck Surgical Oncology David Geffen School of Medicine
Continuum Cancer Cantars of New Yorlt University of Caltfornia, Loa Angeles
Division of Head and Neck Surgery Director
Department of OtDiaryngology Head and Neck Surgery SaJVica
Beth Israel Medical Center Department of Surgery
New York. New York Ronald Reagan UCLA Medical Center
Loa Angelea, California
Mack L. Cheney, MD
Professor Jeffrey R. Harris, MD FRCS (C)
Department of Otolaryngology-Head and Neck Surgery Associate Prafessor
HalYard Medical School Department of Surgery
Director of Facial Plastic and Reconatructive Surgery Division of Otolaryngology-Head and Neck Surgery
Department of Otolaryngology-Head and Neck Surgery University of AI berta
MassachusettJ Eya and Ear Infirmary Chief
Boii1Dn, MaaaachusetiB Department of Otolaryngology-Head and Neck Surgery
University of Alberta Hospital
Neal Futran, MD, DMD EdmoniDn, Alberta
Profassor and Chair of OtDiaryngology-HNS Canada
Director af Head and Neck Surgery
Univeraity of Wuhington
Seatlla, W11hington Forewsrdby
Tessa Hadlock, MD Shan R. Baker, MD
Director
Facial Ne!Ve Center
Department of OtDiaryngology-Head and Neck Surgery Illustrator
MaaaachuaetiB Eye and Ear Infirmary Sharon Ellis
Aaaociate Professor New York. New York
HalYard Medical School
BolltDn, MaaaachusetiB

I
. Wolters Kluwer Lippincott Williams & Wilkins
Health
Philadelphia Baltimore New Yorir. L.ondon
Buenos Alru Hona ICon& Sydlll)' Tokyo
Acquisition Editor: Robert Hurley
Product Manager: Dave Mmphy
Marketing Manager: Angela Panetta
Design Manager: Holly McLaughlin
Manufacturing Manager: Benjamin Rivera
Produaion Service: SPi Global

Copyright 2012 Lippincott Williams & Wilkins, a Wolters Kluwerbusiness

351 West Camden Street Two Commerce Square
Baltimore, MD 21201 2001 Market Street
Philadelphia, PA 19103

Printed in China

All rights reserved. This book is protected by copyright. No part of this book may be reproduced or transmitted in any form
or by any means, including as photocopies or scanned-in or other electronic copies, or utilized by any information storage
and retrieval system without written permission from the copyright owner, except for brief quotations embodied in critical
articles and reviews. Materials appearing in this book prepared by individuals as part of their official duties as U.S. government
employees are not covered by the above-mentioned copyright. To request permission. please contact Lippincott Williams &
Wilkins at Two Commerce Square, 2001 Market Street, Philadelphia, PA 19103, via email at permissions@lww.com, or via
website at lww.com (products and services).

Library of Congress Cataloging-in-Publication Data

Atlas of regional and free flaps fur head and neck reconstruction: flap harvest and insetting I Mark L Urken ... [et al.); illustrator,
Sharon Ellis. - 2nd ed.
p.;cm.
Includes bibliographical references and index.
ISBN 978-1-60547-972-9
1. Head-smgery-Atlases. 2. Neck-surgery-Atlases. 3. Flaps (Smgery)-Atlases. 4. Surgery, Plastic-Atlases. I. Urken,
Mark L, 1954-
[DNLM: 1. Head-surgery-Atlases. 2. Neck-surgery-Atlases. 3. Surgery, Plastic-Atlases. 4. Surgical Flaps-Atlases.
WE 17]
RD52l.A846 2012
617 .5' 100223-dc23
2011022763

DISCLAIMER
Care has been taken to confirm the accuracy of the information present and to describe generally accepted practices. However,
the authors, editors, and publisher are not responsible for errors or omissions or for any consequences from application of the
information in this book and make no warranty, expressed or implied, with respect to the currency, completeness, or accuracy
of the contents of the publication. Application of this information in a particular situation remains the professional responsibil-
ity of the practitioner; the clinical treatments described and recommended may not be considered absolute and universal rec-
ommendations.
The authors, editors, and publisher have exerted every effort to ensure that drug selection and dosage set forth in this text
are in accordance with the current recommendations and practice at the time of publication. However, in view of ongoing re-
search, changes in government regulations, and the constant flow of information relating to drug therapy and drug reactions,
the reader is urged to check the package insert for each drug for any change in indications and dosage and for added warnings
and precautions. This is particularly important when the recommended agent is a new or infrequently employed drug.
Some drugs and medical devices presented in this publication have Food and Drug Administration (FDA) clearance for
limited use in restricted research settings. It is the responsibility of the health care provider to ascertain the FDA status of each
drug or device planned for use in their clinical practice.

To purchase additional copies of this book, call our customer service department at (800) 638-3030 or fax orders to (301) 223-
2320. International customers should call (301) 223-2300.

Visit Lippincott Williams & Wilkins on the Internet: http://www.Iww.com.l.ippincott Williams & Wilkins customer service
representatives are available from 8:30 am to 6:00 pm, EST.

9 8 7 6 5 4 3 2 1
 Dedication

To Laura, for the countless ways that she has enriched my life and whose endless
love and devotion provide the fuel for all of my professional endeavors.
To Gabrielle, Grant and Hannah who provide me with indescribable joy on a daily
basis and are the source of great pride for the remarkable individuals that they have
become.
To the Board of Directors of the Thane Foundation that has provided generous
support and placed their faith in me to fight thia cruel disease in novel ways that could
not be accomplished without them.
To the Levy, Lane, Dawson and Plaut families who have always gone the extra mile
to provide their extraordinarily generous suppon that was critical to achieving success
in ao many of the initiatives of the Thane foundation.
To Elliot, for his friendship and suppon, and for putting his faith in me to help him
to combat a devutating illness, and which he has done with grace and courage.
Finally, to my fellows and residents, who have endured long hours in the operating
room and helped to provide the highest level of patient care.
Marie L Urletm
To Paul.
Mack L Cheney
To Paul Ward, MD and Mark Urken, MD for their mentonhip and teaching, and to
my wife Julie and son Ryan, for their support and understanding.
Keith E. BlacktJJell
To aU my friends and fim:Wy for their ongoing support and especially to Jody,
Jacbon, and Mackenzie who continue to show me what is truly important in life.
Jeffrey R. Harrit
For Bruce, who I am so fortunate to share life with, and our children, Rob, Kent,
McKenzie, and Forrest.
'lbsa A . Haillock
To my wonderful wife Margi, and children Alaandra, Jordan, and Evan, whose love,
devotion, and constant suppon have allowed me to pursue this endeavor, and have
enriched my life beyond compare.
Neal Furran

v
Contributors

Keith E. Blackwell, MD Jeffrey R. Harris, MD FRCS (C)

Professor Associate Professor
Division of Head and Neck Surgery Deparbnent of Surgery
Department of Surgery Division of Otolaryngology-Head and Neck Surgery
David Geffen School of Medicine University of Alberta
University of California, Los Angeles Chief
Director Department of Otolaryngology-Head and Neck Surgery
Head and Neck Surgery Service University of Alberta Hospital
Department of Surgery Edmonton, Alberta
Ronald Reagan UCLA Medical Center Canada
Los Angeles. California
Arjun S. Joshi, MD
Mack L. Cheney, MD Assistant Professor
Professor Division of Otolaryngology-Head and Neck Surgery
Department of Omlaryngology-Head and Neck Surgery The George Washington University
Harvard Medical School Attending Physician
Director of Facial Plastic and Reconstructive Surgery Daparbnent of Surgery
Department of Omlaryngology-Head and Neck Surgery The George Washington University Hospital
Massachusetts Eye and Ear Infirmary Wahingmn, DC
Boston, Massachusetts
Robin W. Lindsay, MD
Neal D. Futran, MD, DMD Assistant Professor
Professor and Chalr of Otolaryngology-H NS Daparbnent of Surgery
Direcmr of Head and Neck Surgery Uniformed Services University of the Health Sciences
University of Washington Facial Plastic and Reconstructive Surgeon
Seattle, Washington Daparbnent of Otolaryngology-Head and Neck Surgery
National Naval Medical Centar
Allan M. Goldstein, MD Bethesda, Maryland
Associate Professor of Surgery
Harvard Medical School Mark L. Urken, MD, FACS
Associate Visiting Surgeon Professor
Department of Pediatric Surgery Department of Otorhinolaryngology-Head and NeckS urgery
Massachusetts General Hospital Albert Einstein College of Medicine
Boston, Massachusetts Chief of Head and Neck Su11ical Oncology
Continuum Cancer Centers of New York
Tessa A. Hadlock, MD Division of Head and Neck Surgery
Diracmr Department of Otolaryngology
Facial Nerve Centar Beth Israel Medical Canter
Department of Omlaryngology-Head and Neck Surgery New York, New York
Massachusetts Eye and Ear Infirmary
Associata Professor
Harvard Medical School
Boston, Massachusetts

vi

D esearch in vascular surgery was markedly enhanced
~s a result of the work of Carrell and Guthrie in
the early 1900s in which they performed replantations
and transplantations of several composite tissues. Hepa-
rin was discovered in 1916. The ability to control blood
cloning was an essential step forward in the develop-
ment of microvascular surgery, as was the use of the
operating microscope that was initiated by Nylen and
Holmgren for ear and eye surgery in the early 1920s at
the Karolinska Medical School in Stockholm, Sweden.
Jacobson and Suarez in 1960 wt:re the first to use the
operating microscope to perform anastomoses of 3-mm
arteries using 7-0 braided silk. Thus was bom microvas-
cular surgery. Technical improvements in microsurgi-
cal instruments, suture, and the operating microscope,
including coaxial illumination, motorized zoom, and
binocular viewing enabled multiple surgeons to simul-
taneously begin to investigate experimental extremity
replantations throughout the world.
Success with replantations led to efforts to perform
reconstructive microvascular surgery. During the 1960s,
Buncke experimented with replanting and transplant-
ing tissues in laboratory animals. He developed many
important principles and techniques and is considered
by some to be the founding father of microvascular sur-
gery. The first reported experimental microsurgical skin
flap transplantation was reported by Krizek and associ-
ates. Abdominal skin baaed on the superficial epigastric
vascular pedicle was successfully transferred in dogs.
During the decade of the 1970s, there were numer-
ous advances in experimental microsurgical tissue trans-
fer and later in the decade, the emergence of clinical
microvascular surgery. Although human tissue transfer
was accomplished as early as 1957, when Sam and Sei-
denberg reconstructed an esophagus with a free jejunal
segment, it was not unti11972 that the first human free
skin flap transfer ofscalp tissue using microvascular sur-
gery was reported by Harii and colleagues. A year later,
Daniel and Taylor reported the successful transfer of a
groin flap. The first revascularized fibular transfer was
reported byTaylor and colleagues in 1975. The decades
Foreword
of the 1980s and 1990s brought many more surgical
innovations and new microsurgical flap designs includ-
ing such important ones as the scapular flap, fibula oste-
ocutaneous :flap, and the radial forearm flap.
A number of regional flaps were coming into use
for head and neck reconstruction at the same time as
the emergence of clinical microvascular surgery. Most
important was the pectoralis major myocutaneous :flap.
The use of microsurgical tissue transfer for reconstruc-
tion of the head and neck was undoubtedly delayed
somewhat by the popularity of the pectoralis major
myocutaneous flap and the belief that it represented the
answer to virtually all of the reconstructive challenges
in head and neck reconstruction. The failure of the pec-
toralis major flap to reliably transfer vascularized bone
for mandibular reconstruction and the difficulties in
tubing this flap to reconstruct the pharyngoesophageal
segment led to renewed interest in the role of free tissue
transfer as a potential solution.
It was in the 1970s that otolaryngologists who per-
formed reconstructive surgery, like myself, became
interested in microvascular surgery. As a 2nd year house
officer in the Department of Otolaryngology, Maxillo-
facial Surgery at the University of Iowa, William Panje,
MD, traveled to New York to take a 1-week surgical
course to learn the techniques of microvascular surgery.
When he returned from the course, he was kind enough
to teach me the techniques. We dissected donor :flaps
in cadavers and practiced anastomosing 1-mm blood
vessels in rats. Once we were consistently successful in
anastomosing these small blood vessels, we attempted
the transfer of a free groin flap based on the superficial
circu.m:fla iliac artery and ~in to the anterior floor-of-
mouth in order to repair a defect resulting from removal
of a large squamous cell carcinoma. The vascular pedi-
cle was only 3 em in length. Much to our delight, the
flap survived and it was reported in 1975 as the world's
first successful intraoral microsurgical flap. Bill and I
were in our 3rd year of residency at the time. We contin-
ued to perform other microsurgical reconstructive cases
together during our senior year of residency at Iowa.
vii
VIII FOREWORD
In 1977, I completed my residency and began an
academic career in the Department of Otolaryngol-
ogy-Head and Neck Surgery at the University of
Michigan. At this time, there was concern among some
head and neck surgeons that otolaryngologists were not
being trained to perform microvascular surgery and
that this aspect of head and neck reconstruction would
be lost to our specialty. Dr. Krause, Chairman of the
Department of Otolaryngology in Michigan, convened
an impromptu meeting of young academic otolaryngol-
ogists with an interest in head and neck reconstruction
at the fall meeting of the AAO-HNS in 1978. Attend-
ees concluded that the most effective way of securing
microvascular surgery as pan of the expertise of otolar-
yngologists was to teach young academic otolaryngolo-
gists the techniques of microvascular surgery and the
anatomy of the various donor flaps. These individuals
could, in turn, teach fellows and house officers in their
respective residency programs. The first microvascular
training course for otolaryngologists was directed by
myself and Charles J. Krause, MD, in 1979 at the Uni-
versity of Michigan. In 1989, I published the first text
book solely devoted to microsurgical aspects of head
and neck reconstruction.
Dominant among these early surgical leaders is Mark
L. Urken, MD, who has had a prominent role in teach-
ing microvascular surgery to otolaryngologists through
courses he conducts, the fellowship progmm he directs,
and textbooks he has written and edited. Young leaders
such as those who have contributed to the second edi-
tion of this outstanding adas continue to innovate and
expand the clinical applications of microsurgical recon-
struction of the head and neck.
The second edition of the Atlas of Regional and Free
Flaps for Head and Neck Reconstruction represents the
culmination of the experience and knowledge in micro-
vascular and regional flap reconstruction of the head
and neck gained over the last 30 years. The contem-
porary head and neck surgeon must have a thorough
understanding of the anatomy of microvascular and
regional flaps and their applications for head and neck
reconstruction. They must also be familiar with the
donor site morbidity resulting from their use. Micro-
vascular surgery requires specialized surgical skills and
detailed understanding of anatomy. This atlas provides
a highly illustrated and in-depth discussion of both
normal and abnormal anatomy of donor sites, the most
common designs of flaps, and their major applications
for head and neck reconstruction. The book details the
surgical techniques of dissecting and transferring flaps,
potential pitfalls when harvesting flaps, and preopera-
tive and postoperative surgical care. The Adas provides
a comprehensive review of flaps used for head and
neck reconstruction and is divided into two pans. Part
I discusses regional flaps including muscle and mus-
culocutaneous and fascial and fasciocutaneous flaps.
An additional chapter explores the palatal island flap.
Part II deals with a large selection of microsurgical flaps
and nerve graft donor sites. Photogmphs of detailed
flap dissections in fresh cadavers are supplemented with
beautiful anatomic illustrations. New to this edition and
of great value is the inclusion of detailed cadaveric dissec-
tions that show the insetting and orientation of flaps to
reconstruct some of the most common defects encoun-
tered by reconstructive surgeons. The book is a valuable
and timeless contribution to the medical literature.
Microvascular surgery has revolutionalized head
and neck reconstruction over the last three decades.
Surgeons can replace missing segments of the mandi-
ble and maxilla with vascularized bone grafts in which
osteointegrated dental implants can be incorporated.
Patients undergoing partial or total glossectomy are
reconstructed with innervated musculocutaneous flaps.
Sophisticated techniques in transplanting skin, muscle,
bone, mucous membrane, and nerve enable the head
and neck surgeon to repair patients suffering from
great deformities and dysfunction as a result of soft tis-
sue and bony defects of the head and neck. The level
of functional and aesthetic restoration of such patients
was not possible before the development of microvas-
cular surgery. This Atlas clearly provides an insight to
the impressive gains made in head and neck reconstruc-
tion by enumerating and carefully describing the many
microsurgical and regional flaps available to the con-
temporary head and neck surgeon. It will be an invalua-
ble resource for both the beginning and the experienced
head and neck surgeon.
Shan R. Baker, MD
Universi'ljl of Michigan
 n head and neck surgery, there are very few develop-
I ments that have had as great an impact as the abil-
ity to transport healthy tissue from regional and distant
sites for the purpose of restoring patient form and func-
tion following ablative surgery. The impact of free tis-
sue transfer has been monumental in promoting wound
healing despite scarring, radiation damage, and salivary
contamination. The ability to reliably transport healthy
tissue into a head and neck defect following an exten-
sive resection has greatly streamlined the ablative and
reconsuuctive program for both the patient and the sur-
geon; this simplification has dramatically decreased the
incidence of prolonged hospitalizations and the need for
multiple surgical procedures to achieve a suboptimal
final result. Finally, the predictable nature of Bap trans-
fers permined the development of a myriad of smgical
and restorative refinements in our approaches to man-
aging defects in the head and neck. The degree of detail
involved in contemporary treatment planning is a direct
result of the ability to transfer well-vascularized hard-
and soft-tissue components, with varying thicknesses
and the potential for both sensory and motor recovery.
The inclusion of dental implants as a part of the com-
prehensive restorative process has permitted significant
advances in function and quality oflife and has justified
patient expectations for recovery levels that far more
closely approximate their predisease state of Mrmaky.
Advances in the treatment of head and neck cancer
continue to demand modification of the approaches
that we take in the management of patients afilicted
with these devastating illnesses. With the evolution of
nonsurgical "organ sparing'' strategies, as wclJ. as tran-
soral resections, with or without robotics, the art of
head and neck reconstruction is no longer a part of the
mainstream of head and neck residency ttaining. This is
problematic with respect to the experience level of the
next generation of surgeons who are no longer skilled
in the transfer of even basic regional flaps and may
not be familiar with the more sophisticated free tissue
transfers that are currently employed in centers that are
performing head and neck reconsuuctive surgery with
high frequency. In addition, the more widespread use of
radiation and chemotherapy as the primary modalities
for treating many upper aerodigestive tract malignancies
has introduced a new set of treatment-related problems
that can only be managed through the introduction of
healthy, nonirradiated tissue to replace the damaged
sttuctures in the :field of radiation. Osteoradionecrosis
and pharyngoesophageal stenosis are examples of such
problems.
The writing of the second edition of this book was
therefore motivated by the recognition that many of
the basic skill sets in flap harvest and utilization are no
longer a part of the mainstream of head and neck surgi-
cal education. Understanding what can be done is vitally
important in order to ensure that clinicians are able to
make the proper choices in advising patients as to what
their best options are for managing their particular dis-
ease process.
It was important, in the design of this edition, to
thoroughly present the full range of both regional and
free tissue transfers that are utilized in contemporary
head and neck reconstruction. This involved the elimi-
nation of certain donor sites from the first edition, in
particular the masseter muscle and lateral thigh flap
chapters. In their place, several new donor sites have
been included, which are the submental flap, paramed-
ian forehead flap, anterolateral thigh flap, posterior tib-
ial flap, ulnar forearm flap, serratus anterior flap (with
nb), and the radius osteocutaneous flap. In addition to
defining the anatomy and flap harvest techniques of
these new donor sites, the authors felt that the demon-
stration of flap insetting techniques, in a variety of dif-
ferent clinical situations, would be helpful to the reader
in order to understand the methods for using the tissue,
once it was successfully harvested. This book does not
cover all of the issues related to defect analysis, which is
the subject of the book titled, Multidisciplinary Head and
Neik Rubmt:ruet:Um: A Dejeet 0Nnt8d Approach, which
is considered a companion text for this second edition.
The inclusion of descriptions of new donor sites that
are presented in detail in this edition is a reflection of
ix
X PREFACE
the evolution of the field of reconstructive surgery and
a plea to surgeons to avoid complacency in using just a
limited range of techniques that they have grown com-
jorrable in performing. Failure to continue to expand the
range of donor sites stifles the creativity and the drive
to do more for our patients and to continue to push
the envelope, which is the process that led to the many
advances that have dramatically changed our manage-
ment of head and neck cancer over the past three dec-
ades.
Chapters 1 through 4 provide the reader with the
anatomy and harvest techniques for the most important
regional muscle and musculocutaneous flaps that are
still vitally important in the head and neck surgeon>s
armamentarium of reconstructive options. Along with
the latissimus dorsi donor site, which is covered in
Chapter 20 as a free flap, these four donor sites provide
immediately available soft tissue that can be transferred
without microvascular surgical skills. The sternocleido-
mastoid muscle flap is not a reliable carrier for the over-
lying cervical skin, but its use in airway reconstruction,
in particular in the management of invasive thyroid can-
cer, is demonstrated through the illustrations that have
been added in this revised chapter.
In the section on regional cutaneous and fascia-
cutaneous flaps, Chapters 5 through 8, there are two
new donor sites that are presented, the submental and
paramedian forehead. These flaps are vitally important
sources of color-matched skin that play a very significant
role in contemporary reconstruction of facial defects.
For total and subtotal defects of the cheek, the posterior
scalp flap remains an excellent technique, which, when
coupled with pre-expansion, provides a large surface
area of skin to restore virtually the entire side of the
face. Finally, the deltopectoral flap, although less com-
monly used, was maintained as a part of this second
edition because there are still circumstances where this
is the donor site of choice for resurfacing cervical skin
defects. The technique of the island deltopectoral flap is
highlighted in this revised chapter and represents a use-
ful strategy for a single-stage transfer of color-matched
skin from the upper chest.
The palatal island flap, presented in Chapter 9, has
emerged as the primary reconstructive option for full-
thickness palatal defects with dimensions that are con-
ducive to this technique. In particular, the surface area
of the remaining palatal mucoperiosteum must be suf-
ficient to provide coverage of the defect with overlap of
the edges of the bone.
The rectus abdominus and gracilis flaps, Chapters
10 and 11, have very important roles in contemporary
head and neck reconstructive surgery. In addition to
the conventional method of harvest, the rectus abdomi-
nus perforator flap has been added to this edition. A
more detailed description of flap insetting has been
introduced into the gracilis chapter to highlight its role
in facial reanimation surgery.
There are three new chapters added to the section of
Fascial and Fasciocutaneous Free Flaps. Both the ulnar
and anterolateral thigh flaps have a proven track record
in head and neck surgery, while the posterior tibial is a
relatively new but promising source of thin skin from
a remote part of the body. Reconstructive techniques
for pharyngoesophageal reconstruction and lengthening
of the thoracic trachea are demonstrated with extensive
cadaveric dissections.
While the fibula, iliac crest, and scapular compos-
ite flaps are the mainstay of reconstructive surgery for
defects of the maxillomandibular skeleton, there are two
new donor sites introduced in this edition: the radius
osteocutaneous and the serratus anterior/rib flaps. The
technique for closure of the abdominal wall with a syn-
thetic mesh following harvest of the iliac crest-internal
oblique osteomusculocutaneous flap is presented in
this edition of the atlas. In addition, the techniques for
insetting of these flaps for restoration of both mandib-
ular and maxillary defects are demonstrated in detail.
In particular, the selection of side of harvest and its
impact on the location of the skin flap as well as the
vascular pedicle has been extensively covered through
numerous cadaver dissections. Other vitally important
surgical problems have been presented in these revised
chapters, such as the management of the missing con-
dyle and the creation of a double-barreled fibular flap
reconstruction of the mandible. Reconstruction of the
palatomaxillary complex with the scapular tip, the fib-
ula, and the iliac-internal oblique flaps is demonstrated
in great detail in the relevant chapters.
The chapters on the free jejunal autograft and
the gastroomental flap have been extensively revised
through step-by-step cadaver dissections of the harvest
technique. In addition, the most common application of
these visceral flaps for reconstructing the laryngophar-
yngectomy defect is presented in a detailed step-by-step
fashion.
Very little has changed in the harvest of the medial
antebrachial and sural nerves other than the intro-
duction of endoscopic, minimally invasive techniques.
These donor sites continue to be a very important part
of head and neck reconstruction, particularly in the
management of facial nerve disorders. With advances
in cross-face nerve grafting techniques and the use of
innervated free muscle to restore dynamic facial anima-
tion, the need for nerve grafts has grown substantially.
Chapter 28, on recipient vessel selection, has been
significantly altered with the introduction of surgical
dissections that demonstrate the harvest of the thora-
coacromial artery and cephalic vein as well as the inter-
nal mammary artery and vein. These recipient vessels
play a very important role in expanding the opportunity

to apply free tissue transfer to patients who have under-
gone prior surgery and radiation and present the chal-
lenges inherent in the vessel-depleted neck. Finding
suitable recipient vessels that permit the performance
of free tissue transfer is vital to the decision to offer life-
preserving and quality-of-life enhancing surgery.
I will end this preface in a similar way to my ending
of the preface in the first edition. While my coauthors
and I have attempted to provide detailed photographs
of dissections and illustrations for the performance of
innumerable surgical techniques in this edition, they are
not a replacement for each surgeon to go to the labora-
tory to painstakingly practice these techniques prior to
attempting them in the operating room. While the lack
of availability of appropriate laboratories in one's own
institution may serve as a barrier, the expansion of sur-
gical courses that provide hands on harvesting experi-
ence allows ample opportunity to expand one's range of
surgical skills. The field of head and neck reconstruction
continues to be a dynamic and exciting career choice.
However, it is imperative to never lose sight of the fact
that surgery is a discipline that requires that each surgeon
reaches the requisite level of skill and experience prior
to embarking on a surgical technique in a live patient.
PREFACE FROM THE FIRST EDITION
The most attractive and challenging feature of head and
neck reconstruction is the complexity of the anatomy
and function of this region. The range of tissue types
that must be duplicated is arguably greater than any
other site in the body. Therefore, it is no surprise that a
growing desire to achieve a higher level of rehabilitation
has caused dissatisfaction with conventional regional
cutaneous and musculocutaneous flaps. The ability to
transfer flaps that are thinner, are more pliable, contain
vascularized bone, and have both motor and sensory
potential has driven the era of free flap surgery. How-
ever, the availability of free tissue transfer must not
mean the abandonment of conventional techniques.
Regional donor sites provide a valuable source of tis-
sues that were ideal for many types of reconstruction.
There are many different factors that enter into the
decision regarding the optimum reconstruction for a
particular patient and a particular defect. The adage,
simpler is better, certainly applies to the selection of a
donor site. However, the desire for simplicity by using a
regional flap must be weighed against the quality of the
end result that can be achieved when free tissue from a
distant site is utilized.
Contemporary head and neck reconstruction involves
a thorough appreciation of both regional and free flaps.
This book covers a spectrum of donor sites and spans
the innovations in technique from the 1960s through
the early 1990s. The art of head and neck surgery with
PREFACE XI
the diversity in reconstructive options has become a
true creative endeavor. A mastery of different donor site
options provides the surgeon with the confidence to find
a solution for virtually every reconstructive problem,
regardless of the complexity of the defect or the tech-
niques that had been previously utilized in a particular
patient.
In addition, there is a growing appreciation that one
donor site may not suffice in the most complicated
defects. We have often resorted to the use of multiple
free flaps or the combination of a free flap and a regional
flap to achieve the final result. Once again, expertise
with many different flaps allows the surgeon to combine
flaps as the situation dictates.
This book was conceived, in large part, out of the
requests of participants at an annual reconstructive
course that my colleagues and I have given at Mount
Sinai Medical Center over the past several years. The
course has in many ways mirrored the evolution of head
and neck reconstruction with an ever-increasing cur-
riculum that reflects the expansion of available recon-
structive options and an ever-increasing enrollment that
reflects the growing interest and enthusiasm for this dis-
cipline. We realized that there was no single book that
provided the head and neck surgeon with a detailed
description of the anatomy and harvesting techniques
for the major regional and free flap donor sites currently
employed in head and neck reconstruction. We chose
the medium of fresh cadaver dissections to provide the
most realistic portrayal of the step-by-step details that
would give the resident and attending surgeon a thor-
ough understanding of each donor site. Since attention
to detail is so vital to successful surgery, the descriptions
in this book reflect that detail as closely as possible.
A thorough understanding of anatomy is the cor-
nerstone of all surgery, and reconstructive surgery is
certainly no exception. With an understanding of the
intricate details of a donor site, the surgeon can creatively
mold the tissue to fit the needs of the patient and the
particular defect. Each chapter includes details of nor-
mal donor site anatomy as well as anatomic variations.
In every section of the book, the most important designs
of each flap are presented as are the major applications
to which that flap has been applied. With the tools of
anatomy and surgical technique, the surgeon's imagina-
tion is the only limitation to solving a particular problem.
Chapters 23 and 24 detail the anatomy and harvest
of nerve grafts from the sural and medial antebrachial
nerves. With an emphasis on restoring function to the
head and neck, sensory and motor reinnervation are key
components and the head and neck surgeon will find it
valuable to be well versed with these two donor sites.
By providing a discussion of anatomy, flap design
and utilization, anatomic variations, preoperative and
postoperative care, potential pitfalls, and harvesting
techniques for each donor site, this book is oriented
XII PREFACE
toward the resident as well as the practicing head and
neck surgeon. However, it is not meant as a substi-
tute for the essential painstaking learning processes of
working in a microsurgical laboratory and in a cadaver
dissection laboratory to master the techniques before
applying them in clinical practice.
Just as the oncologic management of head and neck
neoplasms will continue to evolve, so too will the recon-
struction and rehabilitation of these patients. New
donor sites will undoubtedly be introduced that further
expand the range of tissue that is available. There will
certainly be new techniques that may totally revolu-
tionize this discipline. It is imperative that the surgeon
approaches these innovations with an open mind.
Flexibility will permit change to occur and offer new
hope to our patients.
Mark L. Urken. MD. FACS
 umerous pioneers in head and neck surgery are
N responsible for laying the foundations that have
led to the advances in reconstruction that are presented
in this second edition of the adas. William Panje, Shan
Baker, Hugh Biller, Sebastian Arena, and John Conley
are but a few of those individuals who had the vision
and the insight to make countless advances, without
their efforts, the field of head and neck reconstruction
would have never reached the level that is reflected in
the pages of this book. M.u: Som was one of the great
head and neck surgeons who had the foresight and the
courage to perform the first free tissue transfer tech-
nique on a human in 1958. It was that groundbreaking
procedure that led to the explosion in this field that we
currendy enjoy some six decades later.
There are numerous individuals that I would like
to thank for their help in making this second edition
a reality. I would first like to express my gratitude to
my coauthors on this endeavor. Mack L. Cheney was a
coauthor on the first edition and he continues to be one
of the most creative surgeons that I have had the pleas-
ure of interacting with as a valued colleague through-
out my career. Mack and Tessa A. Hadlock have greatly
advanced the management of facial reanimation sur-
gery and have provided a level of science and creativity
to this field that is a remarkable achievement. Three
former fellows, Neal Futran, Keith Blackwell, and Jeff
Harris have advanced what I have taught them and
moved the field of head and neck reconstructive sur-
gery to new heights. Each has made numerous con-
tributions and started their own fcllowships to ensure
the quality of the next generation of head and neck
surgeons. In addition to being a source of great pride
for me for their numerous individual accomplishments,
their contributions to this edition as coauthors was
vital to providing a final product that we can all be very
proud of.
Four individuals played a very important role in
the painstaking dissections and photography that were
required to produce the countless changes in this edi-
tion. Dr. Arjun Joshi, Dr. Allan M. Goldstein, and
Dr. Robin W. Lindsay each contributed to the writing
of multiple chapters in this edition. In addition, their
efforts, along with Dr. Matthew Bak, contnbuted to
the efficiency and success of those long arduous days in
the laboratory that led to the tremendous quality of the
images that provide clarity to the written word.
Sharon Ellis is a medical illustrator who has an
unparalleled understanding for the need for anatomic
detail, portraying illustrations from a surgeons perspec-
tive, and providing an aesthetically pleasing product.
I have had the pleasure ofworking with her on three sep-
arate projects and my appreciation of her professional-
ism, timeliness, and artistic accuracy continues to grow.
I would like to express my gratitude to Bob Hurley,
Dave Murphy, Eileen Wolfberg, and Franny Murphy at
Lippincott Willi.anul & Wilkins for understanding the
need for a second edition of this book and their com-
mitment to delivering the best possible product for our
readership. They shared our collective vision for the
type of resource that we wanted to provide and never
wavered in their support.
Finally, I and my coauthors would like to gM: a
very special thanks to Synthes CMF Corporation for
the financial and material support that they provided
for this project. They understood the educational value
of this book and its imponance for the next generation
of head and neck reconstructive surgeons and enthusi-
astically commined critical resources that allowed this
project to be completed.
xiii
Contents

Contributors vi Chapter & Anterior and Posterior Scalp ........... 89

Foreword vii Mack L Cheney and Mark L Urkan
Preface ix Flap Harvest Techniques
Acknowledgements xiii Anterior Scalping Flap Dissection ................................. 94
Posterior Scalping Flap Dissection ................................ 98
PART 1 REGIONAL FLAPS 1 Chapter 7 The Submental Island ..................... 102
MarkL Ulten
Muscle and Musculocutaneous Flaps Flap Harvest Techniques
Chapter 1 Pectoralis Major .................................. 3 Submental Flap ............................................................ 111
Mark l. Ulten
Flap Harvest Techniques Chapter 8 Paramedian Forehead .................... 122
Tessa A. Hadlock. Robin W. undsay, and Mack L Cheney
Pectoralis Major Flap ..................................................... 20
Flap Harvest Techniques
Chapter 2 Trapezius System ............................... 27 Paramedian Forehead Flap ........................................... 124
Mark l. Ulten Calvarial Bane Graft ..................................................... 127
Flap Harvest Techniques
Superior Trapezius Flap .................................................. 37 Mucosal Flaps
Latera/Island Trapezius Flap ......................................... 39 Chapter! Palatal Island .................................... 130
Lower Trapezius Island Musculocutaneous Flap ........... 42 MarkL Ultan
Flap Harvest Techniques
Chapter 3 Temporalis ........................................... 47 Palata/Island Rap ........................................................ 133
Tessa A. Hadlock. Robin W. Lindsay, and Mack L. Cheney
Flap Inset Tee hniq ues
Muscular Anatomy of the Face 52
Palata/Island Flap ........................................................ 137
Flap Harvest Techniques
Temporalis Muscle ......................................................... 53
Temporalis Muscle Tendon ............................................ 56
PART 2 FREE FLAPS 139
Chapter 4 Sternocleidomastoid ......................... 59
Mark l. Ulten Muscle and Musculocutaneous Flaps
Flap Harvest Techniques Chapter 10 Rectus Abdominis ......................... 141
Sternocleidomastoid Flap .............................................. 72 Mark L Ulten and Keith E. Blackwell
Flap Harvest Techniques
Cutaneous and Fasciocutaneous Flaps Rectus Abdominis Flap (Extended Deep Inferior
Chapter 5 Deltopectoral ...................................... 76 Epigastric Flap) ........................................................ 152
Mark L. Ulten Rectus Abdominis Flap (Deep Inferior Epigastric
Flap Harvest Techniques Perforator Flap) ........................................................ 158
Deltopectoral Flap .......................................................... 86

xiv
 CONTENTS XV

Chapter 11 Gracilis ............................................ 162 Chapter 17 Ulnar Forearm Free Flap ............... 272
Tessa A. Hadlock, Robin W.lindsay. and Mack l. Cheney Jeffrey A. Harris and Arjun Joshi
Flap Harvest Techniques Flap Harvest Techniques
Gracilis Flap .................................................................. 168 Ulnar Forearm Free Flap ............................................... 282
Flap Inset Techniques
Gracilis Muscle for Facial Reanimation ....................... 171 Composite Free Flaps
Anastomosis of Cross Facial Nerve Graft to Chapter 18 Subscapular System ..................... 288
Obturator Nerve ................... ........... ........... .............. 172 Mark l. Urken
Isolation of Masseteric Motor Nerve .... .... .... ... .... ... .... 173
Reinnervation of the Masseteric Nerve and Chapter 19 Scapular and Parascapular
Bilateral Gracilis Muscle Transfer .......................... 174 Fasciocutaneous and Osteofasciocutaneous
and Subscapular Mega Flap ............................... 292
Fascial and Fasciocutaneous Flaps Mark l. Urken
Flap Harvest Techniques
Chapter 12 Radial Forearm ............................... 176
Mark l. Urken and Jeffrey R. Harris Scapular Osteocutaneous Flap .................................... 301
Flap Harvest Techniques Scapular-Latissimus Dorsi Mega Flap" .................... 309
Radial Forearm Fasciocutaneous Flap ......................... 187 The Scapular Tip Based on the Angular Branch
Beavertail Modification of the Radial and the Parascapular Flap ....................................... 313
Forearm Flap ............................................................ 192 Flap Inset Techniques
Flap Inset Techniques Inset of Scapular Osteocutaneous Flap-Latissimus
Insetting of the Radial Forearm Flap for Dorsi Flap for Mandibular Reconstruction .............. 320
Pharyngoesophageal Reconstruction ........ .............. 196 Reconstruction of the Hemipalatal Shelf
Inset of the Radial Forearm Cutaneous Flap with a Half of the Scapular Tip ............................... 322
for Reconstruction of the Circumferential Reconstruction of the Total Palatal Defect
Pharyngoesophageal Segment ................................ 198 with the Scapular Tip Osteocutaneous Flap ........... 323
Inset of the Radial Forearm Flap for Lengthening
Chapter 20 Latissimus Dorsi and Serratus
of the Mediastinal Trachea ..................................... 201
Anterior ................................................................... 326
Mark L. Urken and Keith E. Blackwell
Chapter 13 Lateral Arm ..................................... 206
Mark l. Urken Flap HarvestTechniques
Flap Harvest Techniques Latissimus Dorsi Myocutaneous Flap .......................... 343
Lateral Arm Flap ........................................................... 212 Flap Harvest and Inset Techniques
Latissimus Dorsi-Serratus Anterior-Rib
Chapter 14 Temporoparietal Fascia ................ 219 Osteomyocutaneous Flap for Oromandibular
Mack l. Cheney, Robin W.lindsay. and Tessa A. Hadlock Reconstruction ......................................................... 351
Flap Harvest Techniques
Tempoparietal Fascial Flap .......................................... 227 Chapter 21 Iliac Crest Osteocutaneous
Tempoparietal Fasciocutaneous Flap .......................... 231 and Osteomusculocutaneous ............................. 359
Mark L. Urken
Chapter 15 Anterolateral Thigh Free Flap ...... 234 Flap Harvest Techniques
Keith E. Blackwell Iliac Crest-Internal Oblique Flap .................................. 372
Flap Harvest Techniques Synthetic Mesh Closure of the Abdominal Wall ......... 385
Anterolateral Thigh Flap .............................................. 243 Closure of the Abdominal Wa/1 .................................... 389
Flap Inset Techniques Iliac Crest Osteocutaneous Flap .................................. 390
Inset of the Anterolateral Thigh Flap for Flap Inset Techniques
Circumferential Pharyngoesophageal Inset of Right Iliac Crest-Internal Oblique Composite
Reconstruction ......................................................... 249 Flap to Reconstruct a Right Oromandibular
Defect ...................................................................... 393
Chapter 16 Posterior Tibial Artery Inset of Left Iliac Crest-Internal Oblique
Free Flap ........................................................................ 257 Musculocutaneous Flap to Reconstruct an
Jeffrey R. Harris Anterolateral Mandibular Defect ............................ 395
Flap HarvestTechniques Reconstruction of a Right Total Maxillectomy Defect
Posterior Tibial Artery Flap .......................................... 266 with a Right Iliac Crest-Internal Oblique Flap ......... 398
XVI CONTENTS

Chapter 22 Fibular Osteocutaneous ............... 404 Chapter 25 Free Omentum and

Mark L. Urken and Neal Futran Gastro-Omentum ................................................... 475
Flap Harvest Techniques Mark L. Urken and Allan Mo Goldstein
Fibular Osteocutaneous Flap oooooooo ooooooooo ooooooooo oooooooo 413 Flap Harvest Techniques
Sensate Fibular Osteocutaneous Flap 414
0000000000000000000000000
Tubed Gastro-Omental Free Flap 481
000000000000000000000000000000000

Fibular Osteocutaneous Flap 415

0000000000000000000000000000000000000000
Flap lnsetTechniques
Flap Inset Techniques Reconstruction of Circumferential
Orientation of the Fibular Osteocutaneous Flap Pharyngoesophageal Defect oooooooooo oooooooooooooooooooo oo 484
Relative to the Laterality of the Leg of Harvest
and the Position of the Cutaneous Paddle and the PART 3 NERVE GRAFT DONOR
Donor Vascular Pedicle 419
00000000000000000000000000000000000000000000
SITES 489
Contour and Rigid Fixation of the Fibular Free Flap to
Reconstruct a Right Hemimandibular Defect 421
0000000000
Chapter 26 Medial Antebrachial Cutaneous
Management of the Resected Mandibular Nerve Graft ............................................................. 491
Condyle 428
00000000000000000000000000000000000000000000000000000000000000000000
Robin W. Lindsay, Tessa A. Hadlock, and Mack L. Cheney
Contouring and Inset of a Right Double Barreled Fibular Flap Harvest Techniques
Flap to Reconstruct a Right Hemimandibulectomy Medial Antebrachial Cutaneous Nerve 00000000000000000000000 494
Defect 434
0000000000000000000000000000000000000000000000000000000000000000000000

Contouring and Fixation of a Fibular Osteocutaneous Chapter 27 Sural Nerve Graft .......................... 497
Flap for Reconstruction of a Right Infrastructure Robin W. Lindsay, Tessa A. Hadlock, and Mack L. Cheney
Maxillectomy Defect 436
00000000000000000000000000000000000000000000000
Flap Harvest Techniques
Sural Nerve Graft 000000000000000000000000000000000000000000000000000000000 500

Chapter 23 Osteocutaneous Radial Forearm

Free Flap ................................................................. 445
PART 4 TECHNICAL
Neal Futran CONSIDERATIONS IN
Flap Harvest Techniques FREE TISSUE TRANSFERS 503
Radial Forearm Osteocutaneous Flap ooooo ooooooooo oooooooo 450
Chapter 28 Recipient Vessel Selection
in Free Tissue Transfer to the Head
Visceral Flaps
and Neck ................................................................ 505
Chapter 24 Free Jejunal Autograft .................. 455 Mark L. Urken
Mark L. Urken and Allan Mo Goldstein Recipient Vessel Harvest Techniques
Flap Harvest Tee hniq ues Harvest of the Lingual Artery in Lessers Triangle 510 0000000

Harvest of Free Jejunal Autograft 466

0000000000000000000000000000000

Harvest of the TACSystem of Recipient Vessels 512 00 0000 000

Flap Inset Techniques Harvest of the Internal Mammery Artery and Vein 000000516
Reconstruction of a Circumferential
Pharyngoesophageal Defect oooo ooooooooo ooooooooo oooooooo 468 Index 521
""rbe pectoralis major muscle has been applied to the
~ reconstruction of a variety of chest wall defects
since 1947 when Pickerel et al. (42) reported its use
as a turnover flap. Sisson et al. (52) used the pecto-
ralis major as a medially based Bap to provide great
vessel protection and obliteration of dead space follow-
ing mediastinal dissection for recurrent cancer of the
laryngostoma after total laryngectomy. In 1977, Brown
et al. (13) descnoed the technique of bilateral island
pectoralis major Baps for the reconstruction of a mid-
line upper chest and lower neck defect. The muscle was
completely isolated on its neurovascular pedicle follow-
ing trans section of its origins and insertions. A skin graft
was used for epithelial coverage after bilateral muscle
advancement. In 1968, Hueston and McConchie (26)
reported a case in which the pectoralis major was used
as a carrier for the overlying skin in reconstruction of an
upper sternal defect. The authors designed this Bap with
a broad base at the shoulder, which limited its arc of
rotation. In addition, they performed a delay procedure
to ensure the vascularity of the skin.
It was not until the latter part of the 1970s that Ari-
yan and Cuono (2) and Ari~ (1) recognized the tre-
mendous potential of the musculocutaneous unit based
on the pectoralis major for the reconstruction of a large
number of head and neck defects. nus discovery was of
paramount importance because it enabled the single-
stage ttansfer of large amounts of wcll-vascularized skin
formauy ofthe ablative and traumatic defects ofthe upper
aerod.igestive tract, face, and skull base, which heretofore
could only be restored with s~d procedures, and inef-
fectively at that. In addition, the hardiness of the vascular
3
4 CHAPTER1

supply permitted the creation of two skin paddles by de-
epithelialization of an intermediate segment of akin so that
the inner and outer lining could be 1ran8ferred with a sin-
gle flap for re<:onstruction of compla, composite defe<:ts.
The impact of this new reconstructive technique on
head and neck sw:gery was recognized almost immedi-
ately. It rapidly replaced many ofthe existing reconstruc-
tive methods, and large series of cases from a variety of
different medical centers were reported as testimony to
the reliability, versatility, and ease of harvesting this flap.
Although various modifications of the original descrip-
tion of this fiap have been reported, along with a recog-
nition of its shortcomingB, it is still a mainstay of head
and neck reconstruction (11,48). One ofits major appli-
cations is to provide coverage of the vital neurovascular
structures in the neck in patients who have undergone
prior radiation with or without chemotherapy.
The pectoralis major is a large fan-shaped muscle
that covers muc:h of the anterior thomcic wall. To a vari-
able extent, it overlies the pectomlia minor, subclavius,
serratus anterior, and intercostal muscles. The origins of
the pectoralis major are divided into two or, sometimes,
three portions. The cephalad segment arises from the
medial third of the clavicle. The central, or sternocostal,
portion has a broad origin from the sternum and the
cartilages of the first six nbs. The third origin of this
muscle, from the aponeurosis of the external oblique
muscle, is variable in size. The fibers of this broad mus-
cle cODVerge to form a tendon that passes deep to the
deltoid and inserts into the crest of the greater tubercle
of the humerus. As it narrows in its course toward the
humerus, it forms the anterior uillary fold (Fig. 1-1).
The medial aspect of the deltoid muscle is almost insep-
arable from the muscle fibers of the pectomlis major.

RGURE 1-1. The pectoralis major is described as having three different heads of origin: clavicu-
lar, sternal-manubrial, and external oblique. The clavicular portion is distinct from the central and
inferior portions of the muscle, both in function and in its neurovascular supply. The central portion
of the muscle originates from the manubrium, the stemum, and the cartilages of the first six ribs.
The pectoralis major both adducts and medially rotates the arm. The relationship of the cephalad
portion of the rectus abdominis muscle to the caudal part of the pectoralis major should be noted.

The cleavage between these two muscles is referred to
as the deltopectoral groove, through which runs the
cephalic vein, which is a constant anatomic landmark.
The pectoralis major is surrounded by a layer of deep
fascia. However, this is separate from the clavipectoral
fascia that surrounds the pectoralis minor and extends
cephalad from that muscle to the clavicle. Prior to
attaching to the undersurface of the clavicle, this fas-
cia splits to envelop the subclavius muscle. Both the
vascular and nerve supply to the pectoralis major pass
through the clavipectoral fascia en route to the deep
surface of the muscle (see Fig. 1-12).
The action of the pectoralis major is to adduct and
medially rotate the arm. It becomes active in internal
rotation of the arm only when working against resist-
ance. The upper muscle fibers help to flex the arm to the
horizontal level; the lower fibers assist in arm extension.
Contraction of the pectoralis major helps to extend
the arm to the individual's side, but it plays no role in
hyperextension beyond that point.
The loss of the dynamic activity of the pectoralis
major appears to be well tolerated, although the true
impact on brachial function has not been studied exten-
sively in any of the large series of pectoralis major mus-
culocutaneous flap transfers. The extent to which the
humeral attachments are transsected undoubtedly leads
to a variable impact on the functional loss resulting
from harvest of this muscle. The additional morbidity
of combining the loss of pectoralis major function and
a radical neck dissection has also not been investigated
in a systematic fashion. Much of the adductor activity is
compensated for by the powerful latissimus dorsi mus-
cle, which makes up the posterior axillary fold.
FLAP DESIGN AND UTILIZATION
The major advantages of the pectoralis major muscu-
locutaneous flap that distinguished it from the three
major cutaneous flaps (deltopectoral, nape of neck, and
forehead) that were in use at the time that the pectoralis
major flap was introduced are the following:
1. Rich vascularity.
2. Large skin territory.
3. Ability to transfer without prior delay.
4. Improved arc of rotation.
5. Increased bulk.
6. Primary donor site closure.
7. Well-vascularized tissue coverage of the carotid
artery in the event of a salivary fistula or cervical
skin necrosis.
8. Ease ofharvest in the supine position.
9. Ability to transfer two epithelial surfaces for inner
and outer lining.
PECTORALIS MAJOR 5
The total skin territory of the pectoralis major is often
greater than 400 cm2 However, it is rare for the entire
skin territory to be required to satisfy the demands of
the ablative defect. As the cumulative experience in the
use of this flap has increased, its limitations have been
identified, and modifications have been described to
help overcome them. The major modifications are dis-
cussed according to these problem categories.
Methods to Improve the Arc of Rotation
Early in the history of this flap, it was recognized that a
distal skin paddle placed over the caudal extent of the
muscle was not only well vascularized but it also per-
mitted a greater arc ofrotation (4). Ariyan's (1) original
description of this flap incorporated a long segment of
skin that extended from the clavicle to the caudal extent
of the muscle. The skin component was oriented over
the course of the pectoral branch of the thoracoacromial
artery. The excess skin resulting from this flap design
often required secondary trimming. An additional ben-
efit to placing the skin paddle over the lower portion of
the muscle was that it permitted the deltopectoral flap to
be preserved for simultaneous or later use (55). Magee
et al. (32) described the placement of the skin paddle
over the lower portion of the pectoralis major, with an
extension overlying the rectus abdominis muscle. Not
only did this skin placement lead to less disfigurement
of the breast in female patients, but, as noted earlier, it
also provided a mechanism to achieve a greater arc of
rotation of this flap to more cephalad defects. Magee
et al. described an array of vessels on the surface of the
rectus sheath that he believed contributed to the vascu-
larity of a distally based skin paddle. Incorporation of
those vessels necessitated the harvest of this fascia to
ensure the blood supply to the overlying skin. Although
it is widely recognized that a portion of "random skin"
can be harvested, it is also recognized that it may be
unreliable. The foundation for the claim of Magee et al.
that a segment of skin could be harvested entirely distal
to the pectoralis is tenuous. The general belief is that a
significant portion of the skin paddle must overlie the
pectoralis major to capture a sufficient number of mus-
culocutaneous perforators (Fig. 1-2). The blood supply
to the skin is discussed later in detail.
Additional measures that have been used to enhance
the arc of rotation are related to the method of trans-
fer of the muscular component of the flap. In the vast
majority of cases, muscle is transposed over the clavicle
and tunneled deep to the cervical skin. This is helpful to
provide coverage of the carotid artery and to augment
the soft tissue deficit following radical neck dissection.
When a radical neck dissection is not performed, the
bulk of the muscle may be problematic, requiring the
use of a skin graft to achieve coverage. In Ariyan's (1)
early description, the muscle was completely exteriorized
6 CHAPTER1

Skin

Anterior
rectus
sheath

Rectus
Linea alba
abdominis m.

FIGURE 12. A skin paddle has been designed over the caudal aspect of the pectoralis major
and the cephalad portion of the rectus abdominis. A portion of the anterior rectus sheath that
is beneath the skin flap is incorporated to enhance the skin's vascularity. A sufficient por-
tion of the skin flap should overlie the pectoralis to ensure capture of the musculocutaneous
perforators.

and later removed after neovascularization of the skin
had occurred. Fabian (21,22) and later Lee and Lore
(31) proposed the removal of a segment of the clavicle
to gain up to 3 em oflength..As a further modification of
this approach, Wilson et al. (62) reported tunneling the
muscle pedicle deep to the clavicle in a subperiosteal
plane. They warned of the potential risk related to vas-
cular compression. De Azevedo (19) described a similar
modification by passing the flap through a subclavicu-
lar tunnel. In addition, he reported the preservation of
the clavicular portion of the mwcle by harvesting only
a distal island of muscle beneath the desired skin pad-
dle. The neurovascular supply to the proximal muscle
was preserved with this technique, which reportedly led
to improved brachial function. In particular, he noted
that patients were able to move their arms forward and
downward against resistance. However, this technique
has not been evaluated in a systematic fashion.
Methods to Deal with Excessive Bulk
The body habitus of most patients with head and neck
cancer rarely leads to concern about excessive bulk in a
Bap. However, this may be a problem in certain patients,
especially when tubing of the skin is required to recon-
sttuct the pharyngoesophagus, or the inttoduction of
excess tissue in the oral cavity results in interference
with normal tongue movement (22). To reduce the bulk
of the skin and subcutaneous tissue, Sharzer et al. (47)
descnbed harvesting a vertically oriented "parasternal"
skin paddle that extended across the sternum to the
opposite internal mammary perforators. Although the
skin paddle had a substantial portion overlying the mus-
cle, the skin extension overlying the sternum achieved a
considerable reduction in bulk (Fig. 1-3).
Alternative solutions to the problem of excessive
Bap bulk were achieved by eliminating the skin paddle
entirely. Murakami et al. (39) described a two-stage
 PECTORALIS MAJOR 7

FIGURE 1-3. A parasternal skin paddle may be designed that crosses over to the opposite side
of the sternum. The skin overlying the stemum markedly reduces the bulk of1his flap. There
must be a sufficient amount of the flap designed to capture perforators from the ipsilateral
pectoralis major muscle.

procedure in which a split-thickness skin graft was placed
over the muscle and then followed, 3 to 4 weeks later,
by the harvest of the muscle-skin graft unit (Fig. 1-4).
They used this thinner Bap for the reconstruction of the
hypopharynx in four women in whom flap thiclaless was
particularly problematic. This concept was extended
by Robertson and Robinson (46) who reported the use
of a quilted skin graft over the pectoralis major in a
one-stage reconstruction of the pharyngoesophagus.
Small mucosal defeas pose the additional problem of
requiring only small segments of skin for reconstruction.
By reducing the size of the skin paddle, there is a greater
risk of missing a sufficient number of musculocutaneous
pe:rfw:ators to achieve adequate Sap vascularity.To prevent
the necessity ofincluding a larger skin paddle than needed,
Johnson and Langdon (28) reported their experience
with seven patients whose oral defects were recOD.StrUcted
with the pectoralis major alone. Re-epithelialization of
the muscle was found to be rapid and produced a satis-
factory long-term result. However, it is apparent to the
experienced surgeon that this approach introduces some
degree of unpredictability due to soft tissue contracture
that occurs over the raw surface of the muscle.
Methods to Achieve Two Epithelial Surfaces
for Reconstruction of Compound Defects
The reconstruction of compound defects involving the
mucosa and overlying skin can be challenging. Early
in the development of the pectoralis major Bap, it was
recognized that the rich vascularity of the skin permitted
the design of two epithelial surfaces by removing the
intervening bridge of skin (11). 'This design placed an
added requirement that the Bap be of sufficient length
to allow it to be folded upon itself. This also produced
additional bulk, which was either advantageous or
8 CHAPTER1

FIGURE 14. A two-stage procedure may be performed in which a skin graft is initially placed
over the muscle. This prefabricated composite flap is then transferred after a 2-week period.
allowing the skin graft to heal to the muscle.

disadvantageous, depending on the location ofthe defect.
Weaver et al. (60) described a bilobed "Gemini"' Bap
in which two separate skin paddles were harvested side
by side to achieve opposing epithelial surfaces. These
authors split the intervening skin and underlying muscle
to achieve more complete separation between the two
skin paddles. Tobin et al. (58) atended this concept one
step further by raising two separate musculocutaneous
units from the same pectoralis major: one based on the
lateral thoracic artery and the other based on the pecto-
ral branch of the acromiothoracic artery (Fig. 1-5).
Preservation of the ipsilateral deltopectoral Bap
allows the ttansfer of a musculocutaneous and a fascio-
cutaneous flap from the same side of the chest to achieve
inner and outer lining (33). The benefit of the added
vascularity of the deltopectoral flap makes it necessary
to place a skin graft on the donor site overlying the del-
toid muscle at the tip of the deltopectoral flap. Bunkis
et al. (14) reported the combination of these two flaps to
reconstruct full-thickness defects of the cheek. In those
situations in which the deltopectoral Bap is preserved
but not primarily ttansferred, a delay procedure can be
performed by making parallel incisions along the upper
and lower limbs of the deltopectoralflap and raising the
intervening skin to allow transfer of the pectoralis Bap
(18). Either simultaneous ttansfer or delay of the del-
topectoral Bap requires the preservation of the internal
mammary perforators while harvesting the pectoralis
major flap. Further details about the delay of a delto-
pectoralflap are presented in Chapter 5.

RGURE 1-5. Two separate musculocutaneous units may be harvested with one based on the
pectoral branch of the thoracoacromialartery and the other supplied by the lateral thoracic
artery.
Dennis and Kashima (20) introduced the "Janus"'
flap as a solution to the problem of reconstructing a
defect that requires both inner and outer lining. These
authors reported a two-stage procedure in which a skin
graft was placed on the deep surface of the pectora-
lis muscle and allowed to heal .After 1 to 2 weeks, the
musculocutaneous Bap was harvested with the muscle
sandwiched between the skin graft and the skin paddle
(Fig. 1-6).
Methods to Include Vascularized Bone in the
Musculocutaneous Flap
The incorporation of vascularized bone with the pec-
toralis major musculocutaneous flap was intended to
expand the use of this technique for the reconstrUction
PECTORALIS MAJOR 9
of composite defects of the head and neck. Experimen-
tal work in the early 1970s demonstrated the advantage
of using vascularized bone in a contaminated and irra-
diated field (36,41). Cuono and Ariyan (17) were the
first to report the use of the pectoralis osteomusculocu-
taneous flap for oromandibular reconstrUction. They
demonstrated the viability of the transferred 5th rib
through Buorescence microscopy (Fig. 1-7). Pulse labe-
ling with different color markers showed the deposition
of new osteoid and, hence, indicated active metabolism.
However, the tenuous nature of the blood supply was
refiected by additional investigators who used this com-
posite flap and reported bone failure rates of 21% (30),
28% (9), and 75% (ll).Additional complications asso-
ciated with nb ha.rvt:st included pneumothorax and
pleural effusion.
10 CHAPTER1

Skin
paddle-

Skin
graft

FIGURE 1-6. The transfer of a Janus flap is achieved through a two-stage procedure in which
a skin graft is initially placed on the deep surface of the muscle. After a 2-week delay, the mus-
culocutaneous flap is harvested with a skin paddle on one side and a skin graft on the other side
of the muscle.

An alternative source of vascularized bone for tranafer
with the pectoralis major is the sternum. Green et al. (25)
described the transfer of the outer cortex of the sternum
with a parasternal skin paddle. Although the harvest of
this composite flap was associated with fewer pulmonary
complications tlw:L with rib harvest, this teclmique has
not been embraced with much enthusiasm (Fig. 1-8).
With the emergence offree tissue transfer of osteocu-
taneous and osteomyocutaneous free flaps from a vari-
ety of different donor sites, a technique that is highly
reliable and reproducble, the use of regional flaps to
transfer bone for mand:&bular and ma:zillary reconstruc-
tion has all but disappeared.
Additional Flap Modifications to Manage the
Muscular Pedicle in the Neck
In most cases, the pectoralis major muscle provides
coverage to the carotid artery and augments the radical
neck dissection contour deformity. Depending on the
muscular development of the individual patient, there
may be a significant bulge as the muscle passes over the
clavicle. Tra1111ection of the medial and lateral pectoral
nerves helps to promote muscle atrophy. In patients
who have not undergone prior radical neck dissection
or in those patients with heavily irradiated cervical
skin, it may be difficult to achieve primary closure of
the skin of the neck over the muscle. In these cases,
the cervical skin may be split and a skin graft placed
over the exposed muscle. Alternatively, the muscle can
be completely exteriorized and then resected after a
2- to 3-week period to allow neovascularization of the
skin paddle. As noted previously, exteriorization of the
muscle can provide additional length to the vascular
pedicle.
Wei et al. (61) described an alternative solution by
harvesting a skin paddle over the sternocostal pol'-
tion of the muscle. The blood supply to that portion
of the muscle was distinct from the vascular supply
to the clavicular portion of the muscle. These authors

FIGURE 1-7. A segment of the 5th rib can be 1ransferred as a vascularized bone, composite
flap. The blood supply to the rib is derived from the periosteal feeders coming from the muscle.
noted that the blood supply to the clavicular portion
was derived &om the acromial, deltoid, and clavicular
branches of the acromiothoracic pedicle; the pectoral
branch supplies the sternocostal segment. Hence, the
sternocostal portion of the muscle could be isolated,
and either it could be tunneled under the clavicular por-
tion, or the latter could be divided. In so doing, the bulk
of tissue crossing the clavicle is greatly diminished to
only that tissue surrounding the vascular pedicle.
The ultimate solution to the problem of muscle bulk
and limited reach was proposed by Reid et al. (45)
who used microvascular surgery to transfer a compos-
ite fiap based on the clavicular head of the pectoralis
major with a skin island and a segment of the medial
clavicle. They reported the successful use of this free
flap in four patients with oral cancer and one patient
with a post-traumatic defect in the tibia. The thinness
and mobility of the skin overlying the clavicle was par-
ticularly advantageous for intraoral restoration. In addi-
tion, the authors discussed the potential for a sensate
PECTORALIS MAJOR 11
flap through the harvest of the supraclavicular sensory
nerves.
FLAP DESIGN AND UTILIZATION
The enthusiasm surrounding the inttoduction of the
pectoralis major musculocutaneous flap led to its appli-
cation to most of the major reconstructive challenges that
had not been adequately solved by the awilable tech-
niques. The early ez.perience with this fl.ap included the
reconstruction of mucosal defects of the oral cavity and
pharynx and cutaneous defects of the neck (4). Ariyan
and Cuono (2) and Ariyu~. (1) reported the successful
application of this flap to the reconstruction of skull base
defects following temporal bone resection and orbito-
mazillary resection. Full-thickness defects of the pharym:
and cheek were easily reconstructed by any of the tech-
niques descn"bed previously that achieve two epithelial
surfaces, including the use of the ipsilateral deltopectoral
12 CHAPTER1

J.

FIGURE 18. The outer table of the sternum may be transferred as a vascularized bone com-
posite flap. The design of a parasternal skin paddle provides a thin soft tissue component.

flap (14). In 1970, Snyder et al. (54) described anum-
ber of techniques to transfer vascularized hone to the
head and ned: using regional cutmeous flaps. The use
of vascularized bone for primary reconstruction of the
mandible led to a flurry of activity, using the composite
osteomusculocutaneous pe<:toralis major flap. However,
advances in microvascular surgery that ocCUITed in the
latter part of the 1970s and early 1980s demonstrated
that vascularized hone could be tl'&l:lSferred from anum-
ber of distmt sites to achieve a more reliable and accu-
rate restoration of mandibular continuity (59).
The pectoralis major fl.ap was used to restore form
and function to the crippled oral cavity. Conley and
Parke (15) reported its use to augment the chin follow-
ing glossectomy and anterior mand:&bulectomy. They
noted that the bulk of the soft tissue component alone
provided an improved external profile ofthe mentum hut
warned that, over time, gravity would lead to a distortion
of the external contour when a flap of too great dimen-
sions was used. This could be overcome through primary
or secondary mand:&oular reconstruction. In 1981, Con-
ley et al. (16) reported their experience in reconstruct-
ing total glossectomy defects with the reinnervated
pectoralis major musculocutmeous flap. The pectoral
nerves were anastomosed to the stump of the hypoglos-
sal nerve. Reinnervation of the muscle could he demon-
strated through ele<:tromyographic recordings. However,
although atrophy of the muscle could be prevented,
meaningful coordinated m<m:ment of the "new tongue"
could not he restored. This technique was invt:stigated in

the rat model in which a pectoralis muscle flap was rein-
nervated through anastomosis to the hypoglossal nerve.
The restoration of contractile activity was confirmed by
electromyographic recordings and the measurement of
isometric contractions. The use of horseradish peroxi-
dase confirmed that the hypoglossal nerve was the source
of the central motor neuron activity (29).
Another use for the dynamic activity of the pectoralis
major is in facial reanimation. Milroy and Korula (37)
transferred the clavicular head of the pectoralis major
in a two-stage procedure, with the first stage being the
placement of a cross-facial nerve graft. The clavicular
head was based on a separate neurovascular pedicle
rather than the one supplying the sternal head of the
muscle. They reported the restoration of dynamic facial
reanimation by using this technique in one patient.
The problem of pharyngoesophageal reconstruction
continued to plague head and neck surgeons because
of the necessity for multistaged procedures when using
either the tubed deltopectoral flap or the Wookey tech-
nique (7,64). In 1980, Theogaraj et al. (57) published
their experience with the pectoralis major flap in seven
patients of whom six underwent secondary reconstruc-
tions and one a primary reconstruction of the pharynx
and esophagus. In five cases of pharyngoesophageal
stricture, the pectoralis major flap was used to aug-
ment the lumen after opening the stricture and preserv-
ing the posterior mucosal strip. Circumferential tubing of
the pectoralis major musculocutaneous flap to recon-
struct the total pharyngoesophageal segment was dif-
ficult because of the bulk of the subcutaneous tissue.
As noted previously, Murakami et al. (39) overcame this
problem by placing a skin graft on the muscle and then,
at a second stage, creating a new pharyngoesophagus by
tubing the skin-grafted muscle. Baek et al. (5) advised
extending the skin paddle over the sternum to harvest
thinner skin to facilitate tubing of the flap. Fabian (21)
described a new technique for reconstructing the cir-
cumferential pharyngoesophageal segment by placing
a skin graft along the prevertebral fascia and using a
partially tubed pectoralis major flap to resurface the
anterior and lateral walls. In 1988, he updated his expe-
rience in 22 patients who underwent this form of recon-
struction and noted a success rate of 88%, with one flap
failure and one stenosis (22). Lee and Lore (31) modi-
fied this technique by placing a dermal graft along the
posterior wall of the reconstructed pharynx.
Reconstruction of the upper thoracic and lower cer-
vical defects following ablative surgery for stomal recur-
rence was considered a risky procedure and fraught
with complications as a result of the exposure of the
great vessels. In addition, the reconstructive techniques
that were used for this defect, prior to the pectoralis
flap, did not obliterate the dead space in the mediasti-
num. In 1981, Biller et al. (10) reported the successful
application of the pectoralis flap to this defect in seven
PECTORALIS MAJOR 13
consecutive patients. The skin paddle was designed with
a semilunar shape when resurfacing stomal recurrences
that involved the superior margin of the tracheostoma.
When circumferential skin defects were created, the new
stoma was formed by placing the opening in the center
of the pectoralis skin paddle (Fig. 1-9). By suturing the
distal trachea to the opening in the skin, a portion of the
depth of the new stoma was composed of the involuted
pectoralis skin paddle. A redundant skin paddle was
needed to accommodate the surface area required for
the involuted portion. This technique not only solved
the problem of the "short trachea," but it also permit-
ted great vessel coverage and dead space obliteration.
Sisson and Goldman (53) confirmed the value of this
reconstructive technique for stomal recurrence in their
report of seven cases in 1981.
As an extension of this technique, Fleischer and
Khafif (23) described a tubed pectoralis major muscu-
locutaneous flap to reconstruct the trachea following
total laryngectomy and tracheal resection for a recur-
rent thyroid carcinoma. The tracheal resection left only
one cartilaginous ring above the carina. One end of the
pectoralis skin tube was sutured to the trachea, and the
other end was sutured to the skin, creating a new stoma.
This technique is particularly useful when the depth of
the cut end of the trachea relative to the level of the skin
and the remaining sternum makes placement of a fenes-
trated pectoralis major flap difficult. The depth of the
funnel that is created places additional tension on the
tracheal suture line. We have used a trapezoid design of
the tubed pectoralis major musculocutaneous flap for
this purpose. This technique creates a larger opening at
the skin level of the chest wall and, therefore, facilitates
visualization of the depth of the airway while helping to
prevent stenosis. Alternatively, we have more recently
approached this particular problem with a radial fore-
arm flap that is more readily tubed in order to achieve
the desired tension-free reconstruction of a short tra-
cheal stump (see Figs. 12-45-12-51).
NEUROVASCULAR ANATOMY
According to the classification scheme of Mathes and
Nahai (35), the pectoralis major is a type V muscle with
one major vascular pedicle from the thoracoacromial
artery and secondary segmental parasternal perforators
that arise medially from the internal mammary artery.
The thoracoacromial artery is a branch from the second
part of the axillary artery (Fig. 1-10). It commonly
divides into four major branches: deltoid, acromial, cla-
vicular, and pectoral (Fig. 1-11). It is the latter branch,
which descends medial to the tendon of the pectoralis
minor, that supplies the pectoralis major.
The lateral thoracic artery is not commonly believed
to contribute significantly to the vascularity of the
14 CHAPTER1

FIGURE 1-9. The rich vascularity of the pectoralis major flap allows it to be modified for recon-
struction following ablative surgery for stomal recurrent cancer. The opening in the center of
the flap is sutured to the end of the trachea and, therefore, solves both the problems of the short
trachea as well as coverage of the great vessels in the mediastinum.

pectoralis major. However, Freeman et al. (24) reported
information to the contrary. In a cadaveric study in
which they examined the vascular supply to the pecto-
ralis major, they found that the lateral thoracic artery
was present in all 17 specimens that were eumined. It
arose from the axillary artery and pierced the clavipec-
toral fascia lateral to the tendon of the pectoralis minor.
In its course within the muscular fascia, it provided a
significant vascular contribution to the pectoralis major
and the major cutaneous supply to the female breast.
Through the injection of contrast material, followed by
xeroradiography, the authors reported that the pectoral
branch of the thoracoacromial artery supplied the clav-
icular and upper sternal portion of the muscle, while the
lateral thoracic artery perfused the inferior and medial
portions.
In their investigation of 10 aortic arch arte-
riogram& and detailed dissections of 35 pectoralis
major muscles, Moloy and Gonzales (38) corrobo-
rated these findings. These authors reponed that,
in all cases, the diameter of the lateral thoracic
anery was equal to or greater than the diam-
eter of the pectoral branch of the thoracoacro-
mial artery. Manktelow et al. (34) reported that a
branch of the lateral thoracic artery, approximately
1 mm in diameter, entered the inferior one fifth of
the muscle in more than 70% of their dissections.
Although the lateral thoracic artery is sacrificed by
most surgeons to improve the arc of rotation of the
pectoralis major musculocutaneous flap, these ana-
tomic studies suggest that it may provide an impor-
tant contribution to the vascular supply of this flap.
 PECTORALIS MAJOR 15

Subclavian a. Lateral

Thoracoacromial a.

Lateral
thoracic a.

thoracoacromial a. pectoral n.

FIGURE 110. The primary vascular supply to the pectoralis major muscle arises from the
thoracoacromial artery. which is a branch of the second part of the axillary artery. The lateral
thoracic artery also supplies some degree of vascularity to the pectoralis major muscle. the
extent of which is controversial. The lateral thoracic artery is variable in size and its contribu-
tion may be completely replaced by the lateral intercostal perforators. The medial and lateral
pectoral nerves supply motor innervation to different regions of the muscle. The clavicular head
is primarily supplied by the lateral nerve; the sternocostal head is supplied by the medial nerve.

Reid and 'Thylor (44) performed the most atensive
study of the vascular supply of the pectoralis major mus-
cle that has been reported in the literature. Their study
included 50 dissections in fixed cadavers and 50 dissec-
tions in fresh cadavers. In the latter group, injections of
the arterial tree included both ink and barium contrast
medium.Although their study focused on the acro.mio-
thomcic axis, they reported no significant contribution
from the lateral thoracic artery. They found that the
pectoralis major had a regional distribution of its blood
supply, with the pectoral artery supplying the sterno-
costal portion and the deltoid artery supplying the cla-
vicular head. They reported only one instance of a very
small pectoral branch and none of complete absence of
this branch. Ink-injection studies of the pectoral artery
revealed staining of the skin overlying the lateral and the
sternocostal portion of the muscle. The clavicular head
of the muscle was not stained until the deltoid branch of
the acro.miothoracic uis was injected. The deltoid mus-
cle and its overlying skin were also stained by ink injection
of the deltoid branch. There were two other interesting
obserwtions in this study. The first was that a significant
zone in the medial aspect of the pectoralis major was not
stained with injections of either the pectoml or deltoid
branches. This zone was thought to be the primary ter-
ritory of the internal mammary perforators. The second
observation was that the major vessels supplying the
skin in the territory of the pectoralis major were actu-
ally fasciocutaneous perforators that ran a course around
the free lower and lateral border of the muscle. These
16 CHAPTER1

Thoracoacromial a.

Clavicular branch

Acromial
branch

Deltoid ~..,.---....;;_---,....--+--Pectoral
branch branch

Lateral thoracic a.

FIGURE 1-11. The thoracoacromial axis classically divides into four main branches: the clavic-
ular, deltoid, pectoral, and acromial arteries. The lateral thoracic artery may also arise from this
system but, more commonly, does so separately from the axillary artery. The thoracoacromial
artery commonly divides into two major branches: the pectoral and deltoid. The acromial and
clavicular arteries variably arise from either division. The deltoid artery runs in the deltopectoral
groove with the cephalic vein, supplying both the pectoralis major and deltoid muscles. It gives
off a cutaneous perforator in the mid portion of the deltopectoral groove. The acromial branch
contributes to a vascular plexus along with branches from the deltoid, suprascapular, and
posterior humeral circumflex vessels. The clavicular branch runs a cephalad and medial course
toward the sternoclavicular joint. The pectoral branch pierces the clavipectoral fascia and then
runs a cephalocaudal course on the deep surface of the pectoralis major, which it supplies.

fasciocutmeous vessels were considerably larger than the
musculocutaneous perforators exiting from the muscle.
The superior thoracic artery provides a small vascu-
lar supply to the pectoralis major. The parasternal inter-
nal mammary perforators perfuse the medial aspect of
the muscle, which allows it to be used as a turnover flap
for reco1111truction of midline chest wall defects.
The pectoral branch of the thoracoacromial arterY
and the lateral thoracic artery penetrate the clavipectoral
fascia along with the medial and lateral pectoral nerves
(Fig. 1-12). The two arteries are both accompanied by
their venae comitantes.After penetrating the clavipectoral
fascia, they run in a cephalocaudal direction before ente!'-
ing the pectoralis major; either the pectoral branch of the
thoracoacromial artery or the lateral thoracic artery sup-
plies branches to the pectoralis minor near the clavicle.
This explail18 the avascular plane of dissection between
the pectoralis major and minor. The deltoid branch ofthe
thoracoacromial artery accompanies the cephalic vein in
the deltopectoral groove. Either the acromial or the del-
toid branch gives off a direct cutaneous vessel at the most
cephalad enent of the deltopectoral groove. In addition,
the deltoid artery commonly gives rise to a cutaneous
perforator in the midportion of the groove.
The application of the angiosome concept to the
blood supply of the anterior chest wall helps to explain
the observatio1111 related to the vascularity of the skin
of the pectoralis major musculocutaneous flap. Taylor
and Palmer (56) defined an angiosome as a segment
of tissue supplied by a single-source artery and vein.
A system of "choke" arteries was descn"bed that connect
adjacent angiosomes. Based on clinical observations and
injection studies, it appears that an adjacent angiosome
can be reliably "captured" after intenupting its source
artery. However, when the area of tissue that is to be
harvested is enended to the subsequent angiosome, or
 PECTORALIS MAJOR 17

the "angiosome once removed,'" necrosis becomes more
likely. Taylor and Palmer surmised that this phenomenon
was caused by the pressure gradient across the choke ves-
sels that connect angiosomes. There is a greater reduc-
tion in the pressure to the distal angiosomes when more
skin territory is harvested based on a single-source vessel.
The pectoralis major and itll overlying skin can be
divided into va.sc:ular territories or &!l,giosomes. There
appears to be some controversy as to whether the lateral
portion of the muscle is supplied by the pectoral bnmclt or
by the lateral thoracic artery. It seems clear, however, that
the medial portion of the muscle is supplied by the inter-
nal mammary perforators. The skin overlying the rectus
sheath is part of the &!l,giosome of the superior epigastric
artery and vein. When based on the pectoral branch, it is
no surprise, therefore, that capture of this skin in the upper
abdomen is tenuous because it is part of an &!l,giosome
that is once removed from the primary &!l,giosome. This
hypothesis maintains that the reduction of the pressure
gradient from the pectoral artery, as it traverses the sys-
tem of choke vessels that surround the internal mammary
tenitory, leads to a tenuous blood supply in the skin oveJ:~
lying the upper abdomen. nus was evident by the poor
staining of skin in this region following ink injections of
the peaoral artery. Reid and 'Thylor (44) noted the stain-
iDa of a netWork of vessels on the surface of the rectus
sheath, which gives credence to the suggestion that this
layer should be harvested along with skin e:xtensions dis-
tal to the territory of the pectoralis major. These authors
also advised great caution in the technique utilized when
interrupting the internal mammary perforators on the
undersurface of the muscle. They warned that the internal
mammary branches should be either ligated or controlled
with bipolar cautery. Excessive use of unipolar cautery
may lead to ascending trauma to the vessels in the internal
mammary angiosome, which would further jeopardize the
flow across this &!l,giosome to the distal skin (56).
The nerve supply to the pectoralis major is from the
lateral (C5 to C7) and medial (C8 toTl) pectoral nerves.
Manktelow et al. (34) identified multiple nerves entering
different parts of the pectoralis major, which numbered
from 4 to 10 individual nerves entering the sternocostal

Subclavius m. Pectoralis major m.

Clavipectoral fascia

Subclavian a.

Pectoralis minor m.

FIGURE 112. The clavi pectoral fascia surrounds the pectoralis minor muscle. The fascial
layers from the posterior and anterior surfaces of this muscle converge to form a single fascial
sheath that runs cephalad toward the clavicle. Before reaching the clavicle, the clavipectoral
fascia again splits to surround the subclavius muscle. The thoracoacromial artery also traverses
this fascia before dividing into its terminal branches. The pectoral branch sends an artery to the
pectoralis minor before forming the primary pedicle of the pectoralis major.
18 CHAPTER 1
portion of the muscle alone. This muscle has been trans-
ferred as a dynamic free flap through anastomosis of
these motor nerves to recipient motor nerves (27,37).
ANATOMIC VARIATIONS
Congenital absence of the pectoralis major is rare. In
a clinic population, this anomaly was observed with a
frequency of approximately 1:11,000 (12). Congenital
absence of the sternocostal head of the pectoralis major
was first reported by Alfred Poland (43) in 1841. This
anomaly was described in conjunction with ipsilateral
syndactyly, and this combination bears the name Poland's
syndrome or Poland's anomaly. It is reported to occur with
an incidence of I :25,000. The potential causes for this
condition include abortion attempts and leukemia (8,63).
The variability in the vascular supply to the pecto-
ralis major was studied by Moloy and Gonzales (38).
They evaluated 10 aortic arch arteriograms and 35 fresh
cadaver dissections. The study revealed that the lateral
thoracic artery was equal to or larger in diameter to the
pectoral branch of the thoracoacromial artery in 90% of
cases. They found extensive collateral flow between these
two vessels in all cases. There was only one instance of a
nonvisualized thoracoacromial system in a patient with
extensive atherosclerosis in the subclavian artery.
POTENTIAL PITFALLS
The overall reliability of the pectoralis major musculocu-
taneous flap is attested to by the low incidence of com-
plete flap failure. In several large series, the incidence
of total flap necrosis was reported to be 1.0% (40),
1.5% (6), 3% (51), and 7% (62). This low incidence of
total flap necrosis is a reflection of the constancy of the
anatomy and the ease of flap harvest. Partial flap necro-
sis, however, has been reported at a much greater rate.
Schusterman et al. (50) noted a 14% incidence of flap
loss involving greater than 50% of the skin surface area.
Other large series have reported partial necrosis rates in
the range of 4% (40) to 7% (6). Partial necrosis rates
were probably a function of the caudal extent of the skin
paddle design. Shah et al. (51) reviewed their compli-
cations in 211 pectoralis major flaps during a 10-year
period. Although they reported a 29% incidence of
partial flap necrosis, they did not break down this fig-
ure according to the number of skin paddles that were
"placed" at the risk of partial necrosis by virtue of their
caudal extension over the rectus sheath. In their series,
the authors identified a number of patient-related, sta-
tistically significant factors for the development of flap
necrosis: age older than 70 years, female sex, overweight,
albumin level less than 4 gldL, and oral cavity defects, in
particular subtotal or total glossectomy. In addition, a
variety of systemic diseases were also associated with an
increased risk of necrosis. Although many of the compli-
cations in this series did not require additional surgical
procedures, they did lead to prolonged hospitalization.
The potential pitfalls in harvesting the pectoralis
major musculocutaneous flap begin with flap selec-
tion. The use of this donor site to resurface defects that
extend more cephalad on the face or scalp calls for skin
paddles designed over the more caudal aspects of the
chest wall and upper abdomen. As noted previously, this
may result in high rates of partial flap failure. Excessive
bulk may be problematic, not only from a functional
point of view, but also in terms of wound healing. In
wounds that are likely to pose problems with healing
as a result of prior radiation and/or poor nutrition, the
effect of gravity can be extremely detrimental and may
require the selection of an alternative nondependent
donor site (3). Finally, the design of a small skin island
may pose problems with incorporation of a musculocu-
taneous perforator and may require the use of a Doppler
to identify the perforator upon which to center the flap.
Pedicle compression may result from external causes,
such as tracheostomy tapes or circumferential dressings.
The creation of an inadequate tunnel for the pedicle
may also cause vascular compromise. Shearing of the
skin paddle through excess tension of the skin relative to
the muscle may disrupt the musculocutaneous perfora-
tors, leading to partial or complete necrosis.
Donor site problems are rare. Hematomas usually
occur because of a failure to control bleeding adequately
following transsection of the humeral head of the mus-
cle. The use of a large skin paddle may lead to excess
wound tension in donor site closure. Necrosis of the skin
of the chest wall may result. In theory, excess tension in
closure may also lead to restrictive pulmonary disease,
although this is rare. The incidence of radiologically
evident and clinically significant pulmonary atelectasis
was investigated by Schuller et al. (49) who selected
two groups of patients with head and neck cancer who
underwent ablative surgery for their disease. One group
underwent reconstruction with a pectoralis major flap,
and the other did not. Both groups were subdivided into
patients with and without preexisting pulmonary dis-
ease. In addition, the patients who underwent pectoralis
flap reconstruction were divided, based on whether the
cutaneous paddle was greater than or less than 40 cm2
Although there was a fairly high rate of radiographic
atelectasis in all patient groups, the incidence of clini-
cally significant pulmonary complications was low. The
group of patients with preexisting pulmonary disease
and flaps greater than 40 cm2 had the highest incidence
of both major radiographic signs of atelectasis and clini-
cal pulmonary symptoms. However, no statistical analy-
sis was reported in this study. It should be noted that
the development of postoperative pulmonary complica-
tions is probably multifactorial, with the preoperative

nuuitional status being a potentially important factor
not considered in this study. When bilateral pectoralis
major flaps are harvested, it is not uncommon that clo-
sure of the second side may require a skin graft. Expo-
sure of the costochondral cartilage may lead to serious
infections, including chondritis (60). We have experi-
enced one such case in a patient who developed necro-
sis of the chest wall skin following closure with excess
tension. Debridement of the affected rib and coverage
with the ipsilateral latissimus dorsi muscle led to suc-
cessful resolution of this problem. In women, distortion
of the breast following donor site closure may be mini-
mized with an infmmammary skin paddle (Fig. 1-13).
The use of this donor site in male patients may lead
to problems with excessive hair growth in the oral cavity
or pharym:. When radiation is given postoperatively, this
problem is usually remedied. Finally, Schuller (48) raised
concern about the ability to detect recurrences in the
neck in a timely fashion because of the bulk of the muscle
FIGURE 113. The infra mammary skin paddle provides thinner skin and leads to less distortion
of the female breast by avoiding medial displacement following closure.
PECTORALIS MAJOR 19
pedicle. He also pointed out the significant morbidity to
the shoulder when the pectoralis major is utilized on the
aide of a denervated trapezius. As noted previously, this
parameter has not been adequately studied.
POSTOPERATIVE CARE
The use of a suction drainage system in the chest wall
donor site is imperative to help avoid the formation
of a seroma. Passive and active range of motion and
strengthening exercises for the shoulder are instituted
within a few days after surgery.
Acknowledgments
The authors would like to acknowledge the conuibu-
tions of Dr. Hugh F. Biller to the writing of this chapter
in the first edition of this book.
Z0 CHAPTER1

Pectoralis Major Flap

FIGURE 1-14. The clavicle and lateral borders

of the sternum are marked on the chest wall. The
approximate course of the dominant vascular
pedicle is marked along an axis drawn from the
acromion to the xiphoid process. The paraster-
nal perforators to the deltopectoral flap are also
marked.

FIGURE 1-15. Askin paddle has been marked

over the caudal, medial portion of the chest wall.
The upper limb of the pectoralis major skin paddle
corresponds to the lower border of the deltopec-
toral flap, which is preserved. Various skin paddle
shapes and sizes may be harvested, depending
on the requirements of the defect.

FIGURE 1-16. The lateral border of the

pectoralis major is identified through wide
undermining ofthe skin of the lateral chest wall.
Obtaining this exposure early in the dissec-
tion allows the surgeon to evaluate the caudal
extent of the muscle and, therefore, the extent
of the random component of the skin paddle.
It is evident that the skin flap that has been
designed completely overlies the pectoralis
major without any distal random component.
 PECTORALIS MAJOR 21

Pectoralis Major Flap

FIGURE 1-17. A circumferential incision around

the skin paddle has been completed, along with
exposure of the entire pectoralis major. The
deltopectoral flap is elevated to the level of the
clavicle without violating its parasternal blood
supply. Although tacking sutures were originally
placed between the skin and muscle to help
prevent shearing forces and injury to the muscu-
locutaneous perforators, this is no longer thought
to be necessary. However, care must be taken
in handling this flap to prevent devascularization
as a result of excess distortion of the skin paddle
relative to the muscle.

FIGURE 118. The pectoralis major is elevated

off the chest wall by blunt and sharp dissection.
Intercostal perforators entering the deep surface
of the muscle must be ligated or coagulated. The
deep plane of dissection along the intercostal
muscles must be respected to prevent entry into
the thoracic cavity.

FIGURE 119. The medial attachments to the

sternum are then transsected up to the level of the
clavicle. Careful attention must be paid to stay lat-
eral to the internal mammary perforators in the 2nd
and 3rd intercostal spaces in order to preserve
the blood supply to the deltopectoral flap. Internal
mammary perforators in the lower interspaces
must be identified and controlled.
Z2 CHAPTER1

Pectoralis Major Flap

FIGURE 1-20. The plane of dissection between

the pectoralis major and pectoralis minor is
avascular, and separation can be done largely
by blunt dissection. The cuff of muscle that is left
attached to the sternum in the region of the 2nd
and 3rd interspaces preserves the vascular sup-
ply to the deltopectora I flap.

FIGURE 1-21. The pectoral branch of the thora-

coacromial artery lsmsllsrraws) is easily visual-
ized on the undersurface of the pectoralis major.
The vascular pedicle is usually located along the
medial aspect of the pectoralis minor. In addition,
one of the pectoral nerves Iarrow) is seen exiting
the pectoralis minor and must be transsected to
achieve additional mobilization of the muscle.

FIGURE 1-22. The muscular attachments to the

humerus are transsected while keeping the vas-
cular pedicle in full view to prevent injury to the
nutrient supply. It is imperative to obtain good
hemostasis as the lateral portion of the muscle
is transsected. This is the most common location
for postoperative bleeding to occur.
 PECTORALIS MAJOR 23

Pectoralis Major Flap

FIGURE 1-23. A close-up view of the undersur-

face of the muscle reveals the vascular pedicle
and transsected muscle fibers coursing across
the axilla to insert on the humerus.

FIGURE 1-24. Atunnel is created for the pas-

sage of the pectoralis flap into the neck. Ade-
quate undermining must be achieved to prevent
compression of the vascular pedicle. The ability
to comfortably pass four fingers into this tunnel
is usually deemed adequate. A distal incision
has been made in the deltopectoral flap for the
purpose of delay to improve the vascular supply
in the event that it is needed. A delay procedure
may also be performed by elevating the delto-
pectoral flap without a distal incision to avoid
committing it to a predetermined length.

FIGURE 1-25. The pectoralis flap has been

transferred into the neck, superficial to the clav-
icle. It is important to avoid twisting or placing
excess tension on the pedicle in this maneuver.
Z4 CHAPTER 1
Pectoralis Major Flap
FIGURE 1-26. Donor site closure has been
accomplished by wide undermining of the chest
wall skin. Suction drains are utilized to prevent
seromas and hematomas.
REFERENCES
1. Ariym S:The pectoralis major myocutaneous flap. Aver-
satile flap for rec:onatruction in the head and nec:k. Plast
&consw Svrg 1979;63:73.
2. Ariym S, Cuono C: Use of the pectoralis major myocu-
taneow flap for rec:oostruction of large cervical facial or
cranial defects. Am J Surg 1980; 140:503.
3. Aviv J, Urkcn ML, LawsonW. Biller HF:The superior tra-
pezius myoc:utaneow tlap in head and nec:k reconstruc-
tion. Arch OrolaryrlfOI Head Nede Surg 1992; 118:702.
4. Baek S, Biller HF, Krespi Y, Lawson W: The pectoralis
major myoc:utaneow island flap for reconstruction of the
head and neck. Head Necle 1979;1:293.
5. Baek S, Lawson W. Biller HF: Reconstruction of h~
ph.aryu and c:ervic:al esophagus with pectoralis major
island myocutancous flap. Amt P'ltm SJ1Tg 1981;7:18.
6. Baek S, Lawson W, Biller HF: An analyaia of 133 pec-
toralis major myoc:utaneoua flap;. .Pftur 1Ucmua SJ1Tg
1982;69:460.
7. BakamjianVA: A two-6tage method for pharyngoesopha-
geal rec:onatruction with a primary pectoral skin flap.
Plast Reumm S1WK 1965,;36:173.
8. Beals R, Crawford S: Congenital absence of the pecto-
ral muscles. A review of twenty-five patients. Glin Orthop
1976;119:166.
9. Bell M, Barron P: The rib-pectoralis major osteomusc:u-
locutaneous tlap. .Amt Pla11t Surg 1981;6:347.
10. Biller HF, Baek S, Lawson W, Krespi Y. Blaugrund S:
Pectoralis major myocutaneous island flap in head and
nec:k rrurgery. Analysis of complications in 42 cases .Arch
Or.olaf'Y"'''l Head Necle SJ1Tg 1981,;107:23.
11. Biller HF, K:rcspi Y. Lawson W. Back S: A one-6tage flap
reconstruction following resection for stomal recurrence.
Or.olaf'Y"'''l Head Necle Sll1g 1980;88:357.
12. Bing R: Ueber angeborene Muakekiefecte. Vm:horvs Arch
1902;170:175.
13. Brown R, FlemingW. Jukiew.icz M: An island flap of the
pectoralis major muscle. Br J Pfast Surg 1977;30:161.
14. Bunkis J, Mulliken J, Upton J, Murray J: The evolution
of tec:bniquct1 for reconstruction of full thickness cheek
d~cts. Plan Recmutr SJ1Tg 1982;70:319.
15. Conley J, Parke R: Pectoralis myocutaneou& flap
for chin augmentation. Orolmyngol Head Nede Surg
1981;89:1045.
16. Conley J, Sac:hs M, Parke R: The new tongue. Otolaryngol
Head Nldt Surg 1982;90:58.
17. Cuono C, Ariyan S: Immediate rec:onstruction of a com-
posite mandibular defect with a regional osteomusc:ulo-
c:u~cow; flap. Plast lUetmstr Surg 1980;65:477.
18. Davis 1<, Price J: Bipedicled delay of the dcltopecton.l
flap in raiaing the pectoral myocu1aneous flap. ~
st.epe 1984;94:554.
19. De Azevedo JF: Modified pectoralis major myoc:utaneous
flap with partial preservation of the muscle: a study of
55 cases. Hlad Nede SllTK 1986;8:327-331.
20. Dennis J, Kaahima H: Introduction of the Janus flap.
A modified pecton.lis major myocutaneous ftap for cervi-
cal esophageal and pharyngeal reconstruction .Arch Oro-
laryyil Head Nede Surg 1981;197:431.

21. Fabian R: Reconstruction of the laryngopharynx and
cervical esophagus. Laryngoscope 1984;94: 1334.
22. Fabian R: Pectoralis major myocutaneous flap recon-
struction of the laryngopharynx and cervical esophagus.
Laryngoscope 1988;98: 1227.
23. Fleischer A. Khafif R: Reconstruction of the mediastinal
trachea with a tubed pectoralis major myocutaneous flap.
Plast Recomtr Surg 1989;84:342.
24. Freeman J, Walker E, Wilson J, Shaw H: The vascular
anatomy of the pectoralis major myocutaneous flap. Br J
Plast Surg 1981;34:3.
25. Green M, Gibson J, Bryson J, Thomson E: A one-stage
correction of mandibular defects using a split sternum
pectoralis major osteomusculocutaneous transfer. Br J
PlastSurg 1981;34:11.
26. Hueston J, McConchie I: A compound pectoral flap. Ausx
N ZJSurg 1968;38:61-63.
27. IkutaY, Kubo T, Tsuge K: Free muscle transplantation by
microsurgical technique to treat severe Volkmann's con-
tracture. Plan &comr.r Surg 1976;58:407.
28. Johnson M, Langdon J: Is skin necessary for intraoral
reconstruction with myocutaneous flaps? Br J Oral Max-
illofac Surg 1990;28:299-301.
29. Katsantonis G: Neurotization of pectoralis major myo-
cutaneous flap by the hypoglossal nerve in tongue
reconstruction: clinical and experimental observations.
Laryngoscope 1988;98:1313.
30. Lam K, Wei W. Sui K: The pectoralis major costomyocu-
taneous flap for mandibular reconstruction. Plast &comr.r
Surg 1984;73:904.
31. Lee K, Lore J: Two modifications of pectoralis major
myocutaneous flap (PMMF). Laryngoscope 1986;96:363.
32. Magee W, McCraw J, Horton C, Mcinnis W: Pectoralis
"paddle" myocutaneous flaps. The workhorse of head
and neck reconstruction. Am J Surg 1980; 140:507.
33. Maisel RH, Liston SL: Combined pectoralis major myo-
cutaneous flap with medially based deltopectoral flap for
closure of large pharyngocutaneous fistulas. Ann Otol
Rhinol Laryngo/1982;91:98-100.
34. Manktelow R, McKee N, Vettese T: An anatomical
study of the pectoralis major muscle as related to func-
tioning free muscle transplantation. Plast Recomr.r Surg
1980;65:610.
35. Mathes S, Nah.ai F: Clinical Applications for Musde and
Musculocutaneous Flaps. St. Louis: CV Mosby; 1991.
36. McCullough D, Fredrickson J: Neovascularized rib grafts
to reconstruct mandibular defects. Can J Otolaryngol
1973;2:96.
37. Milroy BC, Korula P: Vascularized innervated transfer
of the clavicular head of the pectoralis major muscle in
established facial paralysis. Ann Plast Surg 1988;20:75--81.
38. Moloy P, Gonzales F: Vascular anatomy of the pectoralis
major myocutaneous flap. Arch Otolaryngol Head Neck
Surg 1986;112:66.
39. Murakami Y, Saito S, lkari T, Haraguehi S, Okada K,
Maruyama T: Esophageal reconstruction with a skin
PECTORALIS MAJOR 25
grafted pectoralis major muscle flap. Arch Otolaryngol
Head Neck Surg 1982;108:719.
40. Ossoff R, Wurster C, Berktold R, Krespi Y, Sisson G:
Complications after pectoralis major myocutaneous flap
reconstruction of head and neck defects. Arch Otolaryngol
Head Neck Surg 1983; 109:812.
41. Ostrup L, Fredrickson J: Reconstruction of mandibular
defects after radiation using a free, living bone graft trans-
ferred by microvascular anastomoses: an experimental
study. Plast &comr.r Surg 1975;55:563.
42. Pickerel KL, Baker HM, Collins JP: Reconstructive sur-
gery of the chest wall. Surg Gynecol Obstet 1947;84:465.
43. Poland A: Deficiency of the pectoral muscles. Guy's Hosp
Rep 1841;6:191.
44. Reid C, Taylor GI: The vascular territory of the acromio-
thoracic axis. Br J Plan Surg 1984;37: 194.
45. Reid C, Taylor GI, Waterhouse N: The clavicular head of
pectoralis major musculocutaneous free flap. Br J Plan
Surg 1986;39:57.
46. Robertson M, Robinson J: Immediate pharyngoesopha-
geal reconstruction. Use of a quilted skin grafted pecto-
ralis major muscle flap. Arch Otolaryngol Head Neck Surg
1984;11 0:386.
4 7. Sharzer lA, Kalisma M, Silver CE, Strauch B: The para-
sternal paddle: a modification of the pectoralis major
myocutaneous flap.Plast&comr.rSurg 1981;67:753-762.
48. Schuller D: Limitations of the pectoralis major myocuta-
neous flap in head and neck reconstruction. Arch Otolar-
yngol Head Neck Surg 1980;1 06:709.
49. Schuller D, Daniels R, King M: Analysis of frequency of
pulmonary atelectasis in patients undergoing pectoralis
major musculocutaneous flap reconstruction. Head Neck
1994;16:25.
50. Schusterman M, Kroll S, Weber R, Byers R, Guillamon-
degui 0, Goepfert H: Intraoral soft tissue reconstruc-
tion after cancer ablation: a comparison of the pectoralis
major flap and the free radial forearm flap. Am J Surg
1991;162:397.
51. Shah JP, Haribhakti V, Loree TR, Sutaria P: Complica-
tions of the pectoralis major myocutaneous flap in head
and neck reconstruction. Am J Surg 1990; 160:352-355.
52. Sisson G, Bytell D, Becker S: Mediastinal dissec-
tion-1976: indications and newer technique. Laryngo-
scope 1977;87:751.
53. Sisson G, Goldman M: Pectoral myocutaneous island
flap for reconstruction of stomal recurrence. Arch Otolar-
yngolHead Neck Surg 1981;107:446.
54. Snyder C, Bateman J, Davis C, Warder G: Mandibulo-
facial restoration with live osteocutaneous flaps. Plan
RecomtrSurg 1970;45:14.
55. StrelzowV, Finseth F, FeeW: Reconstructive versatility of
the pectoralis major myocutaneous flap. Otolaryngol Head
Neck Surg 1980;88:368.
56. Taylor G, Palmer J:The vascular territories (angiosomes)
of the body: experimental study and clinical applications.
Br J Plast Surg 1987;40: 113.
26 CHAPTER 1
57. Theogaraj S, Meritt W, Acharya G, Cohen I: The pecto-
ralis major musculocutaneous island flap in single-stage
reconstruction of the pharyngoesophageal region. Plan:
Recunst.r Surg 1980;65:267.
58. Tobin G, Spratt J, Bland K, Weiner L: One-stage pha-
ryngoesophageal and oral mucocutaneous reconstruction
with two segments of one musculocutaneous flap. Am J
Surg 1982; 144:489-493.
59. Urken ML: Composite free flaps in oromandibular
reconstruction: review of the literature. Arch Otolaryngol
Head Neck Surg 1991;117:724.
60. Weaver A, Vandenberg H, Atkinson D, Wallace J: Modi-
fied bilobular ("Gemini") pectoralis major myocutaneous
flap. Am J Surg 1982; 144:482.
61. Wei W, Lam K, Wong J: The true pectoralis major
myocutaneous island flap: an anatomical study. Br J Plast
Surg 1984;37:568.
62. Wilson J, Yiacaimettis A, O'Neill T: Some observations
on 112 pectoralis major myocutaneous flaps. Am J Surg
1984;147:273.
63. Wolfson R: Syndactyly, a review of 122 cases. Proceed-
ings of the Western Orthopaedic Association. J Bune Joint
SurgAm 1971;53A:395.
64. Wookey H: Surgical treatment of carcinoma of the
pharynx and upper esophagus. Surg Gynecol Obstet
1942;75:499.
""rbere are three distinct muaculocutaneous flaps that
~ can be harvested from the ttapezius muscle, mak-
ing it unique among the regional muscle fiaps that are
used in head and neck reconstruction. Conley (4), in
1972, is credited with being the first to report the use of
the uapezius muscle as a carrier for skin.The skin design
of this flap was similar to the one that was reported by
Mutter (17) in 1842. Mutter used this cutaneous fiap,
which was based at the midline of the upper back and
extended onto the shoulder, to release burn conttac-
tures of the neck. In 1957, Zovickian (26) reported
using a "mastoid-occiput-based shoulder flap" to close
pharyngeal fistulas. He staged these cutaneous fiaps by
putting a skin graft on the undersurface for lining, and a
skin graft on the recipient bed to close the donor defect.
The fiap was staged one more time prior to transfer.
Conley (4) reported using the same skin design but
incorporated the uapezius muscle in a nondelayed flap.
In addition, he reported that the uapezius muscle could
be used as a vehicle to transfer a segment of vascularized
clavicle to reconstruct the ma:xillofacial skeleton. Ariyan
(1) and McCraw and D:ibbell (16) popularized the flap
design that we now refer to as the superior ttapezius
fiap, which is an extension of Conley's (4) original work.
The superior ttapezius flap, based on the paraspinous
perforators, is a highly reliable flap, although limited in
its utility because of its short arc of rotation.
In 1978, Demergasso (6) reported a bipedicled tra-
pezius fiap based on both the paraspinous perforators
and the transverse cervical artery (TCA) and the trans-
verse cervical vein (TCV). In the subsequent year, at
the international meeting of the American Academy of
27
28 CHAPTERZ

Facial Plastic and Reconstructive Surgery, both Demer-
gasso (6,7) and Panje (21) introduced the unipedicle
lateral island trapezius flap, based solely on the TCA
and TCV. This musculocutaneous flap was useful but
limited because of its short arc of rotation and variable
vascular anatomy, which precludes the transfer of this
flap in a significant percentage of patients.
The third musculocutaneous flap, the lower trape-
zius island musculocutaneous flap (LTIMF), was intro-
duced by Baek et al. (3) in 1980. The transfer of a skin
island overlying the lower portion of the muscle pro-
vides an increased arc of rotation, which is independent
of the variable vascular anatomy of the TCA and TCV
in the posterior triangle of the neck. However, the need
to place the patient in the lateral decubitus position for
harvest has limited the widespread use of the LTIMF.
MUSCLE ANATOMY
The trapezius muscle is a broad, thin, triangular mus-
cle that covers much of the upper back and posterior
neck (Fig. 2-1). Its major action is to raise the lateral
angle of the scapula, which is important for adduction
of the arm. It is helpful to divide this muscle into three
functional anatomic units. The cephalad unit arises
from the superior nuchal line, external occipital pro-
tuberance, and ligamentum nuchae. The upper fibers
insert into the lateral third of the clavicle, defining the
lateral boundary of the posterior triangle of the neck.
The function of the upper trapezius fibers is to elevate
the tip of the shoulder.
The middle portion of the trapezius muscle takes
its origin from the seventh cervical and the upper six
thoracic vertebrae. These muscle fibers have a trans-
verse orientation and insert into the acromion and the
upper border of the scapular spine. The major activity
of the midportion of the muscle is retraction of the
shoulder.
The caudal fibers of the trapezius muscle originate
from the lower six thoracic vertebrae and course in an
oblique cephalad direction to insert into the medial
aspect of the scapular spine. This portion of the tra-
pezius muscle overlaps the upper medial border of
the latissimus dorsi muscle. The caudal portion of the
trapezius assists in the functional activity of the upper
portion by its downward pull on the root of the scapular
spine, which helps in the rotation of the scapula.
NEUROVASCULAR ANATOMY
The blood supply to the trapezius muscle is probably the
most confusing of any of the regional flaps. Mathes and
Nahai (13) classified the vascular pattern to the trape-
zius as a type IT muscle with a dominantTCA and TCV
and with minor pedicles from the occipital artery and
vein and the perforating posterior intercostal vessels of
the cervical and thoracic regions. However, this classi-
fication does not recognize the contributions to the dis-
tal muscle from the dorsal scapular artery (DSA). Even
though the DSA and theTCA commonly arise from the
same parent vessel, they usually enter the trapezius mus-
cle at separate locations; their separate contributions
to different regions of the muscle have been described
(Fig. 2-1) (19). In reporting the "potential pitfalls" of
the trapezius musculocutaneous flap, Nichter et al. (20)
described a case in which an "accessory vessel," arising
at the level of the scapular spine, was ligated to achieve
greater mobilization of the muscle. However, the distal
portion of the muscle and overlying skin showed signs
of ischemia and became necrotic soon after interruption
of this blood supply, despite the fact that the TCA and
TCV were intact. In an effort to clarify this situation, it
is easiest to begin by providing the classic description
of the anatomy of these vessels before discussing the
numerous variations.
As classically described, theTCA arises from the thy-
rocervical trunk and courses along the posterior trian-
gle of the neck toward the trapezius muscle (Fig. 2-2).
The TCA divides into a superficial branch, which passes
over the levator scapulae to run on the undersurface of
the trapezius muscle, and a deep branch, which passes
under the levator scapulae, descending along the medial
aspect of the scapula, deep to the rhomboid minor mus-
cle (Fig. 2-1). The superficial branch of the TCA divides
into descending and ascending branches. The former
runs a caudal course on the undersurface of the mus-
cle, and the latter runs a more cephalad course, sup-
plying the upper portions of the trapezius along with
the occipital artery. The deep branch oftheTCA, which
we will refer to as the DSA, sends a significant branch

FIGURE 2-1. The trapezius is a broad thin muscle that arises from the superior nuchal line, the external occipital protuberance,
the ligamentum nuchae, and the spinous processes of the vertebrae of C7 through T12. The insertions of the trapezius muscle are
to the lateral third of the clavicle, the medial border of the acromion, and the entire length of the scapular spine. There is some
variability in the cephalad and caudal extent ofthe origin of the trapezius muscle, with the upper part failing to reach the skull
and the lower part arising from the vertebrae from T8 to Ll. The muscles lying deep to the trapezius include the levator scapulae,
rhomboid minor, and rhomboid major. In its lateral extent. the trapezius also overlaps the supraspinatus and infraspinatus. The
upper portion of the trapezius muscle is supplied by the TCA,. which exits the posterior triangle superficial to the levator scapulae.
The DSA supplies the caudal portion of the trapezius muscle. It emerges between the rhomboid major and minor muscles or less
commonly between the rhomboid minor and levator scapulae {dotted line*). Additional arterial supply to the trapezius muscle is
derived from the occipital artery and the intercostal perforating arteries, which emerge in the paraspinous region.
 TRAPEZIUS SYSTEM 29

Levator scapulae m. Trapezius m.

Rhomboid minor m.

Rhomboid major m.

Levator
scapulae m.

Rhomboid --+.P.P
minor m.

Rhomboid--+~
major m.
30 CHAPTER2

---f.--Anterior
Ascending - 4-........._...,.;-__, scalene m.
branch

_ _ .__ Common
carotid a.

:---+-- Thyrocervical
trunk

-+---Subclavian a.

DSA TCA

B C
FIGURE Z-2. The anatomy of the TCA and the DSA in the posterior triangle is highly variable. A: The TCA is classically
described as arising from the thyrocervical trunk and running across the posterior triangle of the neck. It divides into a
superficial branch, which crosses over the levator scapulae, and a deep branch, which runs deep to the levator scapulae.
The superficial branch divides into an ascending branch and a descending branch, which supply the upper and lower por-
tions of the trapezius muscle, respectively. The deep branch of the TCA, the DSA. runs deep to levator scapulae and then
gives rise to a superficial branch that arises between either the levator scapulae and rhomboid minor or, more commonly,
between the rhomboid major and minor, supplying the distal portion of the trapezius muscle. B: A common anatomic variation
is shown in which the DSA arises separately from the second or third part of the subclavian artery. The TCA may also arise
directly from the subclavian artery. C: In some cases, the DSA and the TCA may run a course below or intertwined in the
brachial plexus. This variation is most important to identify when harvesting a lateral island trapezius flap in which mobiliza-
tion of the TCA is critical to achieving an adequate arc of rotation.

to the caudal aspect of the trapezius muscle, which
emerges between the rhomboid major and minor and
less commonly between rhomboid minor and levator
scapulae (Fig. 2-1).
Variations in the origin of the TCA and the DSA
are the rule, rather than the exception (Fig. 2-2). Both
branches may arise independently from the second or
third part of the subclavian artery. The importance of
this variation is that the vessels may then run a circui-
tous course, intertwined in the brachial plexus, before
passing out of the posterior triangle either over (TCA)
or under (DSA) the levator scapulae. This variation has
no bearing on the superior trapezius flap or the LTIMF.
However, the utility of the lateral island flap depends
greatly on the complete mobilization of the TCA and
TCV, which is impossible when the artery courses
through the brachial plexus.
Netterville and Wood (19) studied the relationship
between the TCA and DSA in supplying the trapezius
muscle. They found that, in most cases, there was a
reciprocal relationship between these two vessels, with
either one or the other being dominant. In 50% of
their dissections, the DSA was dominant, and theTCA
was a branch of the DSA. In 30% of the dissections,
the TCA was dominant, and the DSA was a branch
of the TCA. In the remaining 20% of cases, the DSA
and TCA appeared to be of equal dominance and size
and had a separate takeoff from the subclavian artery.
In addition, ink-injection studies of the TCA and the
DSA revealed that the former supplied the skin overly-
ing the trapezius above the rhomboid minor and the
latter supplied the skin below the rhomboid minor.
These findings conflict with the results of selective
intra-arterial injections of prostaglandin E 1 by Maruy-
ama et al. (11). Following selective catheterization of
the TCA, the authors reported that injection of pros-
taglandin E1 led to flushing of the skin over the entire
territory of the trapezius muscle. These findings can be
explained by assuming that the DSA was a branch of
the TCA, and therefore, both the proximal and distal
blood supplies were probably injected in Maruyama's
study.
The venous anatomy is equally variable. Goodwin
and Rosenberg (9) identified three major patterns of
TCV anatomy. In the majority of cases, the TCV is a
single vessel, but it may be a dual system. TheTCV exits
the trapezius muscle on its deep surface, close to the
point of entry of the TCA, which is 2 to 5 em above
the clavicle. Although the TCA always runs deep to the
omohyoid muscle, theTCV may be superficial in 25% of
cases. In 60% of cases, the authors found that theTCV
traveled with the TCA; in 15%, it followed a course
under or through the brachial plexus. In the remain-
ing 25%, the TCV ran a more caudal course beneath
the clavicle, terminating in the subclavian vein. In the
majority of cases, the TCV enters the medial subclavian
TRAPEZIUS SYSTEM 31
vein. It can enter the lower portion of the external jugu-
lar vein in one-third of cases.
The accessory nerve, cranial nerve XI, provides motor
innervation to the trapezius muscle after supplying
innervation to the sternocleidomastoid muscle. There
are contributions to the nerve supply of the trapezius
from C2 through C4, but the exact nature of this addi-
tional innervation is uncertain. It is speculated that there
may be proprioceptive sensory fibers coursing through
the cervical contributions to the spinal accessory nerve.
SUPERIOR TRAPEZIUS FLAP
The superior trapezius flap is an extremely reliable
source of coverage for defects of the posterolateral por-
tion of the neck that extend no further medially than
the midline. In our review of the literature on this flap,
we found no instances of total flap failure. In my expe-
rience, it has proven to be extremely reliable, with no
instances of either partial or total necrosis in more than
30 cases (2).
This flap is usually transferred as a peninsula of skin
and muscle, which is based at the midline of the back.
However, an island of skin, overlying the lateral aspect
of the muscle, may also be transferred. The primary
blood supply to this flap is derived from the paraspinous
perforators, with some contribution from the occipital
artery. This flap is unique among the trapezius flaps
in that its blood supply is unaffected by a prior radi-
cal neck dissection with transsection of the transverse
cervical vessels. In fact, the vascularity of the distal por-
tion of this flap may be enhanced through a delay phe-
nomenon when the transverse cervical vessels have been
previously interrupted.
The rationale for this hypothesis is based on the
angiosome concept. Taylor et al. (24) proposed that the
delay phenomenon is caused by the opening up of choke
vessels between angiosomes located in series as a result
of interrupting the source artery in an adjacent angio-
some. Under normal circumstances, without a delay,
it was hypothesized that only one adjacent angiosome
could be captured, but not an angiosome once removed.
This hypothesis can be applied to the superior trapezius
flap by dividing it into its component angiosomes. The
primary angiosome, which has its base at the midline
posteriorly, is supplied by the paraspinous perforators.
The adjacent angiosome overlying the lateral aspect of
the muscle is supplied by the transverse cervical vessels.
Finally, the third angiosome in line, or the angiosome
once removed, which overlies the deltoid muscle, is sup-
plied by a branch of the thoracoacromial artery (Fig.
2-3). Following interruption of the transverse cervical
vessels during a radical neck dissection, the choke sys-
tem of vessels between the three angiosomes becomes
dilated. This allows a more favorable pressure gradient
32 CHAPTER2

FIGURE 2-3. Angiosomes of the superior trapezius flap. The superior trapezius flap is primarily supplied by the paraspinous
perforators that exit in the posterior cervical region. The primary angiosome Ill is shown in yellow; the adjacent angiosome (II),
supplied by the TCA. is shown in blue. Finally, the third angiosome (Ill) in the series, the angiosome once removed, is supplied by
a branch of the thoracoacromial system, which is the primary blood supply to the deltoid. Interruption of the TCA leads to a delay
phenomenon of the skin overlying the deltoid by opening up the choke vessels that separate these three angiosomes. The third
angiosome in the series can be more reliably captured by improving the hemodynamic pressure gradient across the middle zones.

by which the skin overlying the deltoid can be reliably
captured by the medial angiosome supplied by the par-
aspinous perforators (2_,24).
The major use for this flap is to resurface cutaneous
defects of the posterior and lateral aspects of the neck.
Following a radical neck dissection, the transfer of this
flap is not only safe but also causes no further functional
deficit because the muscle is already denervated. It is
especially advantageous for the coverage ofheavily irradi-
ated wounds, including those in which the carotid artery
is exposed. It is unique among the regional musculocuta-
neous flaps in that it is superiorly based; therefore, gravity
does not cause the flap to pull away from the recipient
bed as readily as is the case with other regional flaps with
a dependent muscle supply. The success of this flap in the
"problem wound" is enhanced by inserting the flap along
its entire path to the site of the defect, even if intervening
skin must be excised to do so. The poorer aesthetic result
of wrapping the flap around the neck is counterbalanced
by the increased chances of successful wound healing.
More often than not, a skin graft is required for closure
of the donor site. Secondary correction of the "dog-ear"
deformity below the auricle is often necessary.
LATERAL ISLAND TRAPEZIUS FLAP
The lateral island trapezius flap is the least reliable of
the three musculocutaneous flaps because its arc of
rotation is dependent on favorable anatomy and metic-
ulous mobilization of the TCA and TCV. Preliminary
exploration of the posterior triangle of the neck is essen-
tial to assess the suitability of these vessels. Because the
musculocutaneous island is completely isolated on the
nutrient vascular pedicle, there are no alternative efilu-
ent routes through secondary venous channels, such
as might occur in a musculocutaneous flap in which
the muscle is not completely detached. It is therefore
imperative that a patent TCV is present along with the
artery. The likelihood of both vessels being present fol-
lowing radical neck dissection is small, and therefore,
both the lateral island flap and the LTIMF should not
be selected in such patients.
The primary use of the lateral island flap is for exter-
nal defects of the lateral and anterior neck. It may also
be used for mucosal defects of the pharynx and oral
cavity. Panje (21) described an extension of the lateral
island flap, which he classified as the trapezius muscu-
locutaneous island paddle flap. In this design, a small
island of muscle is used as a carrier for an extended
island of skin that is harvested well beyond the lower
lateral border of the muscle in the direction of the axilla.
The proposed advantage of this flap is the large area of
thin skin that can be harvested, which improves the arc
of rotation without completely interrupting shoulder
function (18).
TRAPEZIUS SYSTEM 33
Ryan et al. {23) described a novel use of the lateral
island trapezius flap to achieve dynamic facial reani-
mation in a variety of situations of facial paralysis. The
surgical technique involved the transfer of an inner-
vated and vascularized segment of the trapezius muscle
to the paralyzed side of the face. The muscle was inset
into the corner of the mouth and the temporalis fascia.
In some cases, the vascular pedicle was not long enough
to reach the defect, and the pedicled muscle flap was
converted to a free muscle flap. This technique was also
used for composite cheek defects by transferring an
innervated musculocutaneous flap. By maintaining the
accessory nerve intact, there was no chance for den-
ervation atrophy to occur. However, the disadvantage
of this technique is that facial movement requires a
conscious effort by the patient to tense the ipsilateral
shoulder.
LOWER TRAPEZIUS ISLAND
MUSCULOCUTANEOUS FLAP
The skin paddle of the LTIMF is designed over the infe-
rior aspect of the trapezius muscle between the midline
vertebrae and the medial border of the scapula. The
harvest of this flap is facilitated by placing the patient
in the lateral decubitus position with adduction and
internal rotation of the ipsilateral arm to increase the
space between the medial edge of the scapula and the
midline of the back. The lower extent of the flap design
is somewhat controversial; some authors report reliable
skin vascularity up to 15 em below the inferior border
of the scapula (22).
The angiosome concept provides some insight into
what the safe caudal extent of this skin flap should be.
The blood supply to the trapezius muscle allows it to be
divided into three separate angiosomes. The TCA sup-
plies the angiosome of the lateral cephalad portion of
the muscle; the cervical paraspinous perforators supply
the medial cephalad angiosome. The lower portion of
the trapezius is supplied by the DSA, which enters the
deep surface of the muscle at the upper border of the
rhomboid major. The flaps that extend below the lower
border of the trapezius muscle fall into the angiosome
of the latissimus dorsi which, in this region, is supplied
by the intercostal arteries (Fig. 2-4) (24).
By applying the principles of the angiosome concept
to the LTIMF, the safe lower border of this skin pad-
dle becomes readily apparent. When the skin paddle
extends beyond the lower border of the scapula, into
the medial angiosome of the latissimus dorsi, then this
inferior portion of the skin is in the angiosome imme-
diately adjacent to the one supplied by the DSA. The
skin overlying the latissimus dorsi angiosome should be
readily captured if the dorsal scapular vessels are pre-
served. Attempts to capture the skin of an angiosome
34 CHAPTER2

Ill

FIGURE 24. The primary angiosomes of the trapezius muscle are divided between the paraspinous perforators, which
supply the medial aspect of zone I and the TCA which supplies the lateral aspect of zone I. The DSA supplies zone II. The
zone I angiosome also receives contributions from the occipital artery. The division between zones I and II is signified by the
underlying division between rhomboid minor and major, through which the contributions from the DSA enter the undersur-
face of the caudal aspect of the trapezius muscle. The division between the cephalad angiosomes I and II and the caudal
angiosome Ill is located at the transition point between the end of the trapezius muscle and the territory of the medial aspect
of the latissimus muscle, supplied by the paraspinous perforators. In harvesting the LTIMF. the position of the skin paddle
may extend caudal to the lower border of the trapezius muscle into the medialangiosome of the latissimus dorsi muscle.
This region can be reliably captured by the trapezius flap by incorporating the DSA. However if the DSA is interrupted and
the flap is based solely on the TCA. then the skin of the third angiosome in the series is less reliable. This explains some of
 TRAPEZIUS SYSTEM 35

once removed, without a delay, are often met with com-
plications. However, we and others have transferred
distal skin paddles with success, despite the increased
incidence of ischemia. Alternatively, a skin island that
does not extend beyond the confines of the trapezius
muscle can be reliably transferred on the TCA-TCV
pedicle alone, through capture of the adjacent angio-
some of the DSA and dorsal scapular vein (DSV) (24).
In routine cases in which the skin paddle is confined
to the territory of the TCA and DSA angiosomes, I do
not preserve the DSA pedicle. However, if a large DSA is
encountered, then temporary occlusion of the DSA may
be accomplished with a microvascular clamp. Observa-
tion of the color and quality of the dermal bleeding in the
distal skin allows the surgeon to decide whether the DSA
pedicle needs to be preserved in order to prevent distal
ischemia (25). Because preservation of the DSA severely
limits the arc of rotation, this pedicle can be mobilized by
cutting a cuff ofrhomboid minor on either side of the ves-
sel to improve the arc of rotation. When this maneuver is
performed, the distal portion of the DSA that travels deep
to the rhomboid major must be ligated (see Fig. 2-20).
The tremendous arc of rotation of the LTIMF makes
it the most versatile of the three trapezius musculocu-
taneous flaps. Mobilization of the entire muscle may be
achieved by pedicling it solely on the TCA and TCV in
the posterior triangle of the neck, similar to the lateral
island flap. I have not found such extensive dissection
to be necessary in the routine application of this flap to
defects of the lateral skull, the midface, the neck, and
the oral cavity. The primary advantages of this flap, aside
from the arc of rotation, are the thinness and pliability of
the tissue compared with that of other regional muscu-
locutaneous flaps. The donor defect is also well camou-
flaged on the patient's back. Preservation of the function
of the upper trapezius muscle fibers can often be accom-
plished by mobilizing only that portion of the muscle
needed to transfer the skin to the defect ( 12). The major
disadvantage of the LTIMF is the necessity to place the
patient in a lateral decubitus position. This flap is most
useful in patients who require reconstruction of lateral
skull and cheek defects in which the ablative procedure
can also be performed in the lateral decubitus position.
TRAPEZIUS OSTEOMUSCULOCUTANEOUS
FLAP
Although Conley (4) and Dufresne et al. (8) reported
the transfer of a portion of the clavicle with the trape-
zius flap, the most commonly transferred segment of
bone is the spine of the scapula. Cadaveric injection
studies indicated that the TCA provides a periosteal
circulation to the spine. Approximately 10 to 14 em of
bone can be harvested while preserving the acromion
to minimize shoulder and upper arm dysfunction (15).
The scapular spine is most effectively transferred with
the lateral island flap design. The transfer of bone with
the superior trapezius flap, although feasible, is limited
in its reach and the flexibility of positioning the skin
relative to the bone. The quality of the blood supply to
the scapular spine is probably comparable to the vas-
cularity of the rib transferred with the pectoralis major
flap.
Bone-containing composite free flaps offer several
distinct advantages that favor their use for oromandibu-
lar reconstruction as follows: (a) a rich vascular supply
to the bone; (b) a flexible relationship of the soft tis-
sue to the bone, allowing a more accurate restoration of
normal anatomy and function; and (c) a complete free-
dom to position and contour the bone of defects involv-
ing the symphysis and contralateral body.
POTENTIAL PITFALLS
Each of the three trapezius flaps has its own potential
problems, which may cause an unsuccessful outcome.

the variability that I have encountered in harvesting this flap and the questionable reliability of this skin paddle that has been
reported in the literature. On the left hand side of the figure, 2 different skin paddles are drawn. The solid line denotes a
skin paddle that would be theoretically readily captured by the TCA because of the fact that it lies almost entirely within the
territory of the adjacent angiosome (II). The datted line indicates a skin paddle that partially overlies the lower portion of the
trapezius and extends into the territory of the latissimus dorsi (Ill). The more caudal skin island has a greater arc of rotation.
However, the reliable transfer of this more caudal segment of skin would require preservation of the DSA. The disadvantage
of this approach is that it requires the harvest of a cuff of the rhomboid minor muscle to achieve an adequate arc of rotation.
The harvest of that muscle cuff allows the surgeon to incorporate the proximal portion of the DSA on the undersurface of this
musculocutaneous flap. The advantage of this approach is that it preserves the upper fibers of the trapezius muscle, which
helps to stabilize the shoulder and preserve its function. Our usual approach to harvesting skin paddles that extend more
than 5 em below the scapular border is to place a temporary microvascular clamp on the dorsal scapular vessels, occluding
the flow to the distal trapezius muscle, and to observe the blood supply to the skin paddle to determine whether interruption
of the DSA is safe.
36 CHAPTERZ
The superior trapezius flap is probably the least
problematic if its limited arc of rotation is respected and
the extent of the defect for closure does not cross the
midline anteriorly.
The lateral island flap is technically easy to harvest,
but it is imperative that the posterior triangle be carefully
explored to ensure that the anatomy of the TCA and
TCV is favorable. Failure to identify and carefully iso-
late both an artery and vein will lead to inevitable fail-
ure. Particular attention must be taken in preserving the
TCV, which may be in jeopardy because of its course
superficial to the posterior belly of the omohyoid and
its entry into the external jugular vein. Both of these
venous patterns must be sought when dissecting in this
region to avoid inadvertent injury (9).
The most common error performed in harvesting
the LTIMF is the failure to raise the trapezius mus-
cle in the plane superficial to the rhomboid major and
minor muscles. This is best accomplished by identi-
fying the lateral border of the caudal portion of the
trapezius muscle. Meticulous dissection in the plane
deep to the trapezius muscle allows the surgeon to
identify the fibers of the rhomboid major muscle that
run in a more transverse orientation and insert into
the medial border of the scapula. Krespi et al. (10)
described the combined rhombotrapezius flap, which
reportedly enhanced vascularity to the overlying skin,
provided added bulk, and allowed transfer of vascu-
larized bone from the medial border of the scapula.
The additional vascularity is undoubtedly a result of
including the dorsal scapular system by dissecting in
the plane deep to the rhomboids. However, the harvest
of the rhomboid major is not critical to achieving this
end. The bulk that is obtained by incorporating the
thin rhomboid muscles is minimal after denervation
atrophy occurs. Finally, the medial border of the scap-
ula is a thin bone that is not suitable for functional
mandibular reconstruction. The morbidity of a winged
scapula that results from total disruption of the medial
muscle group outweighs the limited benefits of this
flap design.
The LTIMF has been labeled an unreliable recon-
structive technique, with complication rates caused by
partial or total flap necrosis ranging from 0% (14) to
57% (5) among the larger series of cases in which this
donor site was used (24). The interpretation of these
complications must be placed in the context of the flap
designs that were utilized in each series. Mathes and Ste-
venson (14) reported a 0% rate of complications when
using the LTIMF for the repair of 13 posterior neck and
skull defects. Although these authors divided the dorsal
scapular pedicle, the arc of rotation to the defects of
the posterior neck and skull allowed the skin paddle to
be placed over the distal Trapezius muscle. Significant
mobilization of the muscle was not required. Cummings
et al. (5) reported a 57% incidence of flap necrosis.
Although these authors noted that they extended the
distal skin paddle beyond the lower border of the mus-
cle in some patients, they did not analyze their compli-
cations with this variable in mind.
In the largest series of 45 LTIMFs reponed by Urken
et al. (25), there was a 6.5% incidence of major compli-
cations defined as a greater than 20% flap loss. There
was similarly a 6.5% incidence of minor (20%) flap loss.
In no case in this series was the dorsal scapular pedi-
cle preserved. The lower border of the skin paddle was
not extended beyond 5 em below the scapular border in
most patients. Of note was the fact that all cases of flap
necrosis except one occurred in patients who underwent
flap transfer on the side in which a prior neck dissection
had been performed. It is evident from this finding that
a prior radical neck dissection should be considered a
contraindication to harvesting an ipsilateral LTIMF.
Donor site problems are rarely significant. However,
seroma formation is common, and long-term suction
drainage is recommended.
 TRAPEZIUS SYSTEM 37

Superior Trapezius Flap

FIGURE 25. The superior trapezius flap is

based on the para spinous perforators of the
lower cervical region. The flap is outlined over
the upper portion of the trapezius muscle with
the anterior incision of the flap placed along
the anterior border of the trapezius muscle.
The posterior border of the flap is a transverse
incision parallel to the anterior incision. The
width of the flap is determined in part by the
width of the defect and by the necessity to
incorporate several perforators. The arc of
rotation of this flap is limited by the posterior
inferior attachment (arrow). This rotation may
be improved slightly by extending the incision
across the midline in a cephalad direction. This
modification was introduced by Panje (21) and
helps to add additional length to this flap. The
distal portion of the skin paddle may extend
several centimeters beyond the acromion pro-
cess. In my experience, this flap is extremely
reliable, and up to 8 to 10 em of random skin
beyond the distal lateral extent of the trapezius
muscle may be safely incorporated. The actual
dimensions of the flap have not been fully
defined through injection studies to determine
reliable parameters for flap design.

FIGURE 26. The posterior incision is made

through the skin and through the trapezius
muscle. The firm attachments of the trapezius
to the spine of the scapula must be incised
to maintain the proper depth of dissection.
The TCA and TCV are encountered in this
portion of the dissection and may or may not
be ligated and transsected. In theory, if the
anatomy of the TCAJTCV is favorable, and
these vessels can be mobilized and preserved
with the flap, then the distal end of the flap
should retain an enhanced vascularity. The
deep plane of dissection is between the trape-
zius and supraspinatus. In the medial aspect
of the dissection, the trapezius is elevated off
the levator scapulae and the rhomboid minor.
38 CHAPTER2

Superior Trapezius Flap

FIGURE 27. The incisions along the anterior

border of the flap is made to coincide with the
anterior border of the trapezius muscle. The
distal portion ofthe skin paddle is elevated
along the plane just superficial to the deltoid
fascia. On reaching the lateral aspect of the
trapezius muscle, the plane of dissection is
then deepened to incorporate that muscle.
The TCA and TCV are usually ligated and
transsected when encountered in the dissec-
tion along the anterior border. As noted above,
if the anatomy of these vessels is favorable
and they can be adequately isolated and
mobilized in the posterior triangle of the neck,
then this rich vascular supply to the trapezius
muscle should be preserved, thus preserving
the direct flow to the third angiosome. If the
clinical situation permits, and a longer supe
rior trapezius flap is desirable, then a delay
phenomenon can be performed whereby the
incisions around the flap are made and then
sutured into pia ce. In addition, the tip of the
flap may be raised and sutured. Two weeks
later, the entire flap can be elevated and
transferred.

FIGURE 28. Demonstration of rotation of the

superior trapezius flap. This flap can be used
to close defects that do not extend beyond the
midline of the neck anteriorly. Closure of the
donor site is achieved by wide undermining.
In most cases, a skin graft is needed to cover
the wound. The area of skin grafting may be
reduced by using retention sutures to lessen
the area of the defect.
 TRAPEZIUS SYSTEM 39

Lateral Island Trapezius Flap

FIGURE 29. Harvest of the Lateral Island Trapezius

Flap. The selection of this donor site is based on two
fundamental assessments. The first is the location
and size of the defect, which is usually limited in its
cephalad extent by the mastoid and lower border of the
mandible. The defect may extend either to or across
the midline anteriorly. The second is an anatomic
determination of the TCA/TCV to ensure that they are
favorable for flap rotation. The lateral island trape-
zius flap is designed as an island of skin overlying the
lateral aspect of the cephalad portion of the trape-
zius muscle where it inserts into the clavicle and the
acromion process of the scapula. The anterior border
of the trapezius muscle is marked at its insertion on
the distal one third of the clavicle. The skin island may
be designed over the approximate boundaries of the
trapezius muscle and with random portions extending
more distally. The dimensions of the flap are limited
by the redundancy of the tissue in this region, which
would permit primary closure of the defect.

FIGURE Z-10. The dissection begins by exposure of

the inferior aspect of the posterior triangle. The dis-
section begins by exposure of the inferior aspect of the
posterior triangle. This is best accomplished by making
the anterior incision of the flap. The supraclavicular
fossa is carefully dissected to identify the TCA and
vein. When a neck dissection is performed at the same
time as harvest of a lateral island flap, particular atten-
tion must be paid to preserving the TCV.
40 CHAPTER2

Lateral Island Trapezius Flap

FIGURE 211. The anterior border of the tra pe-

zius (large arrow} has been identified. The poste-
rior belly of the omohyoid has also been isolated
(small arrows}. The TCA runs along the floor of
the posterior triangle. The TCV may run a more
superficial course relative ta the omohyoid and
the artery. The incisions around the skin paddle
are made after the anatomy of the vessels has
been detennined and the surgeon has ensured
that the TCA and TCV are not intertwined with
the roots ofthe brachial plexus. Variations in
the entry site of the TCA and TCV along the
anterior border of the trapezius may cause the
skin paddle to be altered so that it is centered
on the vascular pedicle. After the skin paddle is
outlined, the incisions are made circumferentially
through the skin, subcutaneous tissue, and trape-
zius muscle.

FIGURE 212. Mobilization of the latera I island

flap has been completed. Ugation of the distal
TCA and TCV, as they descend along the more
caudal aspect of the trapezius muscle, must be
accomplished when the distal incision through
the flap is made. The DSA may arise as a branch
of the TCA. Beeause of its course deep to the
levator scapulae muscle, the DSA must be
ligated and transsected.
 TRAPEZIUS SYSTEM 41

Lateral Island Trapezius Flap

FIGURE Z-13. Transposition of this flap into the

recipient site is now completed. Greater mobilization
of this flap can be achieved by dissection along the
vessels in the medial aspect of the posterior triangle
and extending into level IV of the neck toward the
origin of the thyrocervical trunk. Transection of
the posterior belly of the omohyoid muscle may be
required. Wide undermining, followed by layered
closure, is performed to manage the donor site
defect.

FIGURE 214. Harvest of the Lower Island Trape-

zius Flap. The patient is placed in a lateral decubitus
position with an axillary roll in the contralateral
axilla. The approximate position of the scapula is
outlined on the back along with that of the TCA,
which courses over the shoulder on the under-
surface of the trapezius muscle. The DSA enters
the deep surface of the trapezius along the medial
scapular border. Adduction and internal rotation of
the ipsilateral arm is helpful to lateralize the scapula
and open up the space between the scapula and the
midline of the back.
42 CHAPTER2

Lower Trapezius Island Musculocutaneous Flap

FIGURE Z-15. A skin paddle has been outlined

between the medial border of the scapula and the
midline of the back. The inferior extent of the skin
paddle may be reliably placed up to 5 em below the
inferior border of the sea pula. A Ia rger skin paddle
placed over the mare cephalad portion of the Tra-
pezius muscle helps to ensure capture of a greater
number of musculocutaneous perforators.

FIGURE Z-16. The dissection begins by incising

the skin paddle and a vertical line drawn from the
proximal tip of the skin paddle toward the posterior
triangle of the neck. This incision is carried down to
the level of the trapezius muscle, and then, skin flaps
are elevated both medially and laterally to expose the
full extent of the trapezius muscle.

FIGURE 2-17. The lateral border of the trapezius

(arrows) muscle has been identified and elevated,
with the overlying skin paddle, in the plane between
the trapezius and rhomboid muscles.
 TRAPEZIUS SYSTEM 43

Lower Trapezius Island Musculocutaneous Flap

FIGURE Z-18. The muscle attachments to the midline

vertebrae are then transsected sharply to mobilize the
muscle from distal to proximal. Para spinous perfora-
tors Iarrow) must be ligated and transsected. A suture
has been placed around the dorsal scapular pedicle.

FIGURE 219. At the junction of the rhomboid major

and rhomboid minor. the DSA and DSV are identi-
fied as 1hey enter the undersurface of 1he trapezius
muscle.
44 CHAPTER2

Lower Trapezius Island Musculocutaneous Flap

Rhomboid minor m. Trapezius m. Skin paddle

(cui)

Rhomboid major m.

Descending branch of
dorsal scapular artery
and vein transsected

FIGURE Z-ZO. If the DSA and DSV are to be preserved, and further mobilization of the trapezius
muscle is required to reach the donor site, then dissection deep ta the rhambaid muscles must
be carried out in order to transsect the distal branches of the dorsal scapular pedicle. A cuff of
rhambaid miner must be harvested to allaw the DSA and DSV to be mobilized.
 TRAPEZIUS SYSTEM 45

Lower Trapezius Island Musculocutaneous Flap

FIGURE 221. The DSA has been preserved

through harvest of the cuff of the rhomboid
minor muscle. With more proximal dissec-
tion, the TCA and TCV are identified on
the undersurface of the trapezius muscle.
Preservation of both of these pedicles to
ensure vascularity to the distal portion of the
muscle may be carried out, depending on the
location of the recipient defect If a tempo-
rary microvascular clamp applied to the DSA
reveals no disturbance in the circulation to
the distal skin paddle, then the DSA can be
ligated and transsected. The arc of rotation
is enhanced by transsecting the insertions of
the trapezius muscle along the scapular spine
and the medial attachments to the vertebrae.
The flap has been completely mobilized. The
cephalad extent to which the skin paddle can
be used is evident by its position relative to
the auricle. Wide undermining is critical for
closure of the donor defect. Adduction of the
arm helps to achieve a tension-free repair. A
suction drain must be placed with an exit site
along the midaxillary line.

FIGURE 222. Closure of the donor site has

been completed. The muscular pedicle may be
tunneled under the intervening skin to provide
access to the recipient defect In select situ-
ations, exteriorization of the trapezius muscle
may be preferable. The second-stage trans-
section of the muscle must then be carried out
in approximately 2 to 3 weeks.
46 CHAPTERZ
REFERENCES
1. Ariyan S: One-stage repair of a cervical esophagostome
with two myocutaneous flaps from the neck and shoulder.
Plast Recomtr Surg 1979;63:426.
2. Aviv J, Urken ML, Lawson W. Biller HF:The superior tra-
pezius myocutaneous flap in head and neck reconstruc-
tion. Arch Orolaryngol Head Neck Surg 1992;118:702.
3. Baek SM, Biller HF, Krespi YP, Lawson W: The lower
trapezius island myocutaneous flap. Ann Plast Surg
1980;5:108-114.
4. Conley J: Use of composite flaps containing bone for
major repairs in the head and neck. Plan: Reconnr Surg
1972;49:522.
5. Cummings C, Eisele D, Coltrera M: The lower trapezius
myocutaneous island flap. Arch Orolaryngol Head Neck
Surg 1989;115:1181.
6. Demergasso F:The lateral trapezius flap. Presented at the
Third International Symposium of Plastic and Recon-
structive Surgery, New Orleans, Louisiana, Apri129--May
4, 1979.
7. Demergasso F, Piazza M: Colgajo cutaneo aislada a
pediculo muscular en cirugia reconstruction por cancer
de cabeza y cuello: tecnica original. The 47th Congreso
Argentine de Cirugia Forum de Investigaciones. &fJ
Argent Chir 1977;32:27.
8. Dufresne C, Cutting C, Valouri F, Klim M, Colen S:
Reconstruction of mandibular and floor of mouth defects
using the trapezius osteomyocutaneous flap. Plast Recon-
str Surg 1987;79:687.
9. Goodwin WJ, Rosenberg G: Venous drainage of the lat-
eral island trapezius musculocutaneous island flap. Arch
Orolaryngol Head Neck Surg 1982; 108:411.
10. KrespiY, Oppenheimer R, dud Flanyer J:The rhombotra-
pezius myocutaneous and osteomyocutaneous flaps. Arch
Orolaryngol Head Neck Surg 1988;114:734.
11. Maruyama Y, Nakajima H, Fujino T, Koda E: The defini-
tion of cutaneous vascular territories over the back using
selective angiography and the intra-arterial injection of
prostaglandin E 1: some observations on the use of the lower
trapezius myocutaneous flap. Br J Plast Surg 1981;34:157.
12. Mathes S, Nahai F: Muscle flap transposition with
function preservation: technical and clinical consider-
ations. Plast &comtr Surg 1980;66:242.
13. Mathes S, Nahai F: Clinical Applicatiom for Muscle and
Musculacutaneous Flaps. St. Louis: CV Mosby; 1982:50.
14. Mathes S, Stevenson T: Reconstruction of posterior neck
and skull with vertical trapezius musculocutaneous flap.
AmJ Surg 1988;156:248.
15. Maves M, Phillippsen I.: Surgical anatomy of the scapu-
lar spine in the trapezius-osteomuscular flap. Arch Orolar-
yngol Head Neck Surg 1986; 112:173.
16. McCraw JB, Dibbell DG: Experimental definition of
independent myocutaneous vascular territories. Plast
&comtrSurg 1977;60:212.
17. Mutter J: Cases of deformities of burns, relieved by oper-
ation. Am J Med Sci 1842;4:66.
18. Netterville J, Panje W, Maves M: The trapezius myocuta-
neous flap: dependability and limitations. Arch Orolaryn-
gol Head Neck Surg 1987;113:271.
19. Netterville JL, Wood D: The lower trapezius flap: vascular
anatomy and surgical technique. Arch Orolaryngol Head
Neck Surg 1991;117:73.
20. Nichter L, Morgan R, Harman D, et al.: The trapezius
musculocutaneous flap in head and neck reconstruction:
potential pitfalls. Head Neck 1984;7:129.
21. PanjeWR:The island (lateral) trapezius flap. Presented at
the Third International Symposium of Plastic and Recon-
structive Surgery, New Orleans, Louisiana, April 29-May
4, 1979.
22. Rosen H: The extended trapezius musculocutaneous flap
for cranio-orbital facial reconstruction. Plan: Recomtr
Surg 1985;75:318.
23. Ryan R, Waterhouse N, Davies D: The innervated trape-
zius flap in facial paralysis. Br J Plast Surg 1988;41 :344.
24. Taylor GI, Palmer JH, McManamny D: The vascular
territories of the body (angiosomes) and their clinical
applications. In: McCarthy JG, ed. Plastic Surgery. Vol. 1.
Philadelphia:WB Saunders; 1990:329.
25. Urken ML, Naidu R, Lawson W, Biller HF: The lower
trapezius island musculocutaneous flap revisited.
Report of 45 cases and a unifying concept of the vascu-
lar anatomy. Arch Otolaryngol Head Neck Surg 1991;
117:502.
26. Zovickian A: Pharyngeal fistulas: repair and prevention
using mastoid-occiput based shoulder flaps. Plast Recomtr
Surg 1957;19:355.
~e temporalis muscle, one of the muscles of
~ mastication, has been used for a wriety of
reconstructive problems in the maxillofacial region
(5,8,15,18,22,23,25,28,32). Described in 1898 by Golo-
vir&e (15), the temporalis is one of the earliest reported
muscle flaps (19). It was initially described for use in oblit-
eratiD,g the dead space created by OJ.biw exenteration and
was felt to be an excellent choice based upon its bulk and
its prazimity to the OJ.bit (4,11, 15,27,28,35). In the 1930s,
Gillies (14) introduced the teclmique oftemporalis lJ."aDS.-
position as a method for rehabilitation of the paralyzed
face. Sheehan (34) also conttibuted useful modifications
of this technique, by describing the reduction or removal
of the zygomatic arch, increasing the arc of rotation of
the ttansposed muscle and minimizing the potential prob-
lem of excessive bulk in the midface. In 1961, Andersen
modified the Gillies technique by using temporalis fascia,
instead of fascia lata, to reconstruct the eyelids in patients
with facial paralysis (32). In the 1970s and 1980s, numer-
ous SUJ:geOD.S (12,2~26,32,33) expanded the use of the
temporalis muscle in the management of the paralyzed
face. Rubin (30--32) and Rubin et al. (33) also clarified
the application of this flap in oral commis!IUl'e reanima-
tion, by carefully categOrizing human smile pane.rns, and
detailing the anatomic relationship between the orbicula-
ris oris and the facial muscles. These reports established
the temporalis as a logical option for reanimation of the
paralyzed face. Further refinements in the transfer of the
temporalis increased its clinical usefulness in managing
contour defects follaw:ing maxillofacial resections (4,5,29)
and in eyelid (17) and intraoral reconstruction (6).
FLAP DESIGN AND UTILIZATION
The temporalis flap has gained acceptance for a variety
of clinical pu:rposes, including the augmentation of
47
48 CHAPTER3
regional tissue deficiencies and the elimination of scar
contractures. It may also serve as a vascular surface for
skin grafting, to protect the carotid artery, as a myo-
osseous flap, and also to provide dynamic rehabilitation
of the paralyzed face (8,12,23-26,31-33).
The temporalis muscle may be transferred as a seg-
ment or in its entirety, depending on the specific recon-
structive demand. The dimensions of the muscle vary,
with the thicker aspect of the muscle located in the ante-
rior third of the temporal fossa; the middle and posterior
thirds of the muscle are consistently thinner and slightly
longer. The muscle is longest in its middle third, ren-
dering this portion ideal for use in rehabilitation of the
paralyzed face. Although some authors have relied upon
the muscle for ocular and midfacial rehabilitation, most
surgeons favor independent reconstruction of these two
important functional zones of the face.
The temporalis is the only widely employed regional
muscle option for rehabilitation of the paralyzed face.
Although it does not produce spontaneous mimetic
movement, the transferred temporalis permits immediate
reanimation and repositioning of the paralyzed face, and
may be used when the potential for facial nerve recov-
ery exists. The muscle has a contraction capability of 1 to
1.5 em, and the midportion of the muscle has sufficient
strength to adequately mobilize the face and resist the
forces of soft-tissue contracture (22,23). It is innervated
in a segmental pattern by the branches of the trigeminal
nerve (branch V3), permitting independent segments of
the muscle to be designed for use in distinct regions of
the face (i.e., one slip in the orbit and another in the mid-
face). The zygomatic arch can be used effectively as a ful-
crum, to provide the transferred muscle with a fixed point
of origin following its transposition into the midface.
Although the temporalis is firmly attached to the
coronoid process and ramus, the point of attachment of
the distal transposed end of the muscle can be varied to
individualize the procedure to the particular character-
istics of the patient's smile, as analyzed on the normal
side (8,26,29,30).
There are several limitations when using the tempo-
ralis muscle for facial reanimation. Muscle contracture
is initiated by the fifth cranial nerve and is therefore
not mimetic with the contralateral face. This drawback
can be minimized by early and regular physical therapy.
The other significant concern with this technique has
been management of the donor site. When the muscle is
transferred over the zygomatic arch, the contour of the
temporoparietal scalp and the midface can become dis-
torted. By limiting the amount of muscle that is trans-
ferred to a 2-cm-wide strip of muscle from the middle
third, the amount of bulk over the zygoma is minimized.
The secondary depression in the infratemporal fossa
had historically been managed with synthetic implants,
yielding an unnatural feel and a susceptibility to extru-
sion. As an alternative, the temporoparietal fascial flap
(TPFF) can be elevated as an independent flap, based
upon the superficial temporal artery and vein, and used
to reestablish the contour of this region (8).
In the past decade, reports have emerged that describe
an advancement technique of the muscle rather than a
transposition over the zygomatic arch (7, 10,20). In this
approach, the tendinous attachments of the muscle to
the coronoid process are released and advanced to the
region of the modiolus and/or nasolabial fold. In some
techniques, the belly of the muscle is elevated off of its
bony deep surface attachment and then rotated and
advanced anteriorly to produce the desired vector pull.
While some descriptions involve removal and replace-
ment of the zygomatic arch, others describe coronoid-
ectomy without removal of the zygomatic arch.
The temporalis muscle has been used to reconstruct a
variety of midfacial defects by designing it as a turnover
flap, with the point of rotation based at the zygomatic
arch. Because the flap has a rotational radius of8 em, it is
possible to cover defects of the mastoid, cheek, pharynx,
and palate. The muscle is longer and thinner than the
masseter, and therefore can be placed throughout the
midface, providing bulk to anatomic locations that are
not within the rotational range of other regional muscle
flaps (9, 16). The arc of rotation can be improved by pass-
ing the temporalis deep to the zygomatic arch, a maneu-
ver that is often made simpler by osteotomies to remove
and then replace the bone. The muscle readily accepts
split-thickness skin grafts, a feature that makes it useful
in the management of full-thickness defects of the middle
third of the face. It can provide adequate bulk to oblit-
erate full-thickness defects of the orbit and the buccal sur-
face of the oral cavity (18), and in the closure of oroantral
fistulas, or defects of the lateral maxilla and skull base (6).
Craniofacial surgical procedures often produce a
communication between the anterior cranial fossa and
the nasal or paranasal sinus cavities. Separation of these
two regions is critical to minimize the incidence of cer-
ebrospinal fluid leak, epidural abscess, and meningitis.
The temporalis muscle has been successfully used for
this purpose (22,29), and does not interfere with the
vascularity of the overlying scalp, as it is based upon
the deep temporal vascular system, and does not require
transfer or disruption of the superficial temporal vessels.
The temporalis muscle has been described as a car-
rier of vascularized outer calvarial bone for palatal (13),
orbital rim, and orbital floor reconstruction (3,13,22),
though more modem techniques usually involve either
free calvarial bone grafts or other composite myo-
osseous free flaps. The temporalis muscle can also be
transferred along with the coronoid process for recon-
struction of the orbital floor. Although described for
use in segmental defects of the lateral mandible, the
limited bone stock of this donor site is inferior to the
bone stock of other donor sites currently in use for oro-
mandibular reconstruction.

NEUROVASCULAR ANATOMY
The temporalis muscle is broadly based, arising from the
superior temporal line. It fills the entire temporal fossa
and narrows as it inserts onto the coronoid process of
the mandible, via a thick tendinous sheath (Fig. 3-1).
It is covered superficially by the temporalis muscular
fascia. The temporalis functions in conjunction with the
masseter and pterygoid muscles, to elevate and retract
the mandible during mastication. The vascular supply to
the temporalis muscle is provided by the deep temporal
artery and vein, which arise from the internal maxillary
system, deep to the zygomatic arch. The deep tempo-
ral vessels penetrate the undersurface of the temporalis,
providing a segmental vascular pattern (Fig. 3-2) (18).
The muscle is classified as having a type m pattern of cir-
culation (two dominant vascular pedicles), as described
by Mathes and Nahai (21).An additional arterial supply
to the muscle arises from the middle temporal artery,
which sends minor branches through the superficial
aspect of the muscle. The middle temporal artery arises
from the superficial temporal artery and crosses over the
zygomatic arch to provide a separate vascular supply to
the temporalis muscular fascia.
The temporalis muscular fascia inserts on the superior
temporal line. Approximately 2 em above the zygomatic
arch it divides into two layers. The deep and superficial
muscular fascial layers insert on the medial and lateral
aspects of the arch, and are separated by a fat pad. The
muscular fascia fuses with the periosteum of the arch to
form a very dense fibrous layer. The temporal and zygo-
matic branches of the facial nerve cross the zygomatic arch
in the temporoparietal fascia, superficial to the muscular
fascia-periosteal layer. The fat pad that separates the two
layers of the temporalis muscular fascia may be used to
protect the facial nerve branches. By starting at the root
of the zygomatic arch, and incising the superficial layer of
the temporalis muscular fascia, the fatty plane is entered.
H this incision is made at a 45-degree angle in the antero-
superior direction, and the zygomatic arch is uncovered
in a subperiosteal plane, the facial nerve branches can be
protected by reflecting this fascial-periosteal layer in an
anterior and inferior direction (1).
POTENTIAL PITFALLS
Dynamic temporalis muscle transfer relies heavily on
capturing an adequate vascular supply, as well as an
intact neural supply, for use in facial reanimation. In
patients who have undergone extensive skull base or
neck surgery, disruption of the neurovascular supply
eliminates the temporalis as a viable regional option.
When transferring the temporalis muscle flap, a
TPFF is raised independently to obliterate the donor
site defect. Secondary alopecia of the overlying scalp
TEMPORALIS 49
may occur following dissection of the scalp skin from the
superficial surface of the TPFF. This problem is more
common among patients who have undergone regional
radiation therapy, and/or in whom a scalp incision has
been used to gain access to the skull base. Patients who
have undergone occipital approaches to the skull base,
involving temporal incisions around the auricle, are not
good candidates for the use of the TPFF because partial
devascularization of the auricle occurs in a significant
percentage of patients.
The frontal branch of the facial nerve is located in
the temporoparietal fascia, superficial to the tempora-
lis muscular fascia (Fig. 3-1, inset). It crosses the zygo-
matic arch approximately 2.4 em from the tragus, along
a line drawn from the tragus to the lateral canthus (2),
and branches somewhat variably into anterior, middle,
and posterior rami in the region from the zygomatic
arch to the lateral border of the frontalis muscle. This
nerve branch should be identified and avoided during
the anterior dissection and elevation of the muscle. The
auriculotemporal nerve, a branch of the third division of
the trigeminal nerve, courses under the zygomatic arch
and then runs in a cephalad direction posterior to the
superficial temporal artery and vein. It supplies sensa-
tion to the anterior auricle, the external auditory mea-
tus, and the scalp of the temporal region, and should be
identified and preserved to maintain sensation to these
areas.
PREOPERATIVE ASSESSMENT
The neurovascular integrity of the temporalis muscle is
assessed by asking the patient to clench his or her teeth,
and palpating the area for the appropriate bulging,
ensuring that the muscle exhibits normal strength and
tone. This is particularly important in patients who have
undergone skull base procedures where the viability of
the fifth cranial nerve is in question, and in edentulous
patients who may suffer from disuse atrophy of the mus-
cles of mastication. Asymmetric wasting of the temporal
fossa is a telltale sign of denervation atrophy of the tem-
poralis muscle. During the preoperative assessment, it is
also important to establish the patency of the superficial
temporal artery and vein via Doppler auscultation, to
assess the utility of the temporoparietal fascia for donor
site obliteration (8).
POSTOPERATIVE WOUND CARE
At the completion of the temporalis transfer procedure,
a suction drain is placed in the temporoparietal scalp for
24 to 36 hours. A bulky compressive dressing is used for
the first 24 hours postoperatively to prevent hematoma
formation and reduce facial swelling.
50 CHAPTER3

,\lftolii'+\~--+-Temporalls
musdefasda

Temporalls
musc:fe --+----,;..r+~

1/1/H:H+---~Temporoparfetal
fa8da
(wi1h superficial
temporal artery)
Superficial
temporal
fat pad --+-+~
Zygomatic
arch ---4-~ll

musc:fe

gland
RGURE 3-1. The temporalis originates from the surface of the calvarium on the lateral aspect of
the skull. The superior attachment forms a gentle arc that is referred to as the superior temporal
line. The temporalis occupies the entire temporal fossa and inserts into the coronoid process and
the anterior aspect of the mandibular ramus. The temporalis is covered by athick fascial layer.
termed the temporalis muscular fascia. The temporalis fascia is adherent to the skull atthe supe
riortemporalline where it is continuous with the pericranium that covers the remainder of the skull.
Inferiorly, the temporalis muscular fascia splits into a deep and superficial layer approximately2 em
cephalad to the arch {see inset), with a fat pad between these layers. These two layers merge with
the periosteum of the medial and lateral surfaces of the zygomatic arch, respectively. The superfi-
ciallayer of the temporalis muscular fascia is continuous with the masseteric muscular fascia.
 TEMPORALIS 51

Deep temporal
arteries and veins

Superficial temporal
artery and vein

Internal maxillary a.- - -

FIGURE 3-Z. The temporalis muscle is supplied by the anterior and posterior deep tempo-
ral arteries, which arise from the internal maxillary artery and enter the muscle anterior and
posterior to the coronoid process, respectively. These two vessels enter the muscle on its deep
aspect. The anterior artery tends to enter the muscle at a more caudal point than the posterior
artery does, but both vessels usually enter the substance of the muscle by the upper edge of the
zygomatic arch. The nerves to the temp ora lis also enter on its deep surface and are typically
three but, sometimes, four in number. The temporal nerves run between the superior and infe-
rior heads of the lateral pterygoid muscle, crossing over the superior head along with the deep
temporal arteries, to enter the temporal is.
52 CHAPTER!

FIGURE 3-3. A lateral view of the face

demonstrates the relationship between the
mimetic muscles and the temporalis and mas-
seter muscle bodies. Note the key character-
istics of the facia I muscles; they have a bony
origin, with a soft tissue or muscular insertion,
allowing them to move the overlying integu-
ment of the face and, therefore, produce
changes in facial expression.

FIGURE 3-4. A close-up frontal view of a facial

dissection demonstrates the position of the
midfacial mimetic musculature. The importance
ofthe orbicularis oris to facial expression is
evident; the mimetic muscles insert into or
adjacent to this circular sphincteric muscle.
The small and delicate nature of the mimetic
muscles can be appreciated.
 TEMPORALIS 53

Temporalis Muscle

FIGURE 3-5. The approach to the temporalis

muscle is made through a scalp incision, with a
vertical component extending from the midpor-
tion of the superior auricular helix to approxi-
mately 2 em above the superior temporal line.
This incision allows full exposure of the muscle
and its overlying fascia. The incision can be
extended into the preauricular crease to gain
exposure of the superficial temporal artery and
vein. Preservation of the vascular pedicle to the
temporoparietal fascia permits its use for donor
site obliteration.

FIGURE 3-6. The deep muscularfascia is

exposed by elevating scalp flaps approximately
6 em anteriorly and posteriorly. The width and
orientation of the portion of the temporalis
transfer is determined. A 2- to 3-cm strip har-
vested from the mid portion of the muscle region
is most commonly used. This segment provides
adequate length and exhibits active contractile
properties that are ideal for facial reanimation.
If a larger portion of the muscle is required for
reconstruction of a midfacial or an oral cavity
defect the incisions can be modified to elevate
as much of the muscle as is needed.
54 CHAPTER!

Temporalis Muscle

FIGURE 3-7. The temporalis and its averlying

fascia are raised down to the zygomatic arch.
In the caudal aspect of this dissection, the
neurovascular supply is at risk. so both blunt
dissection and avoidance of electrocautery are
necessary, to prevent injury to the neurovascu-
lar pedicle as it enters the undersurface of the
muscle.
 TEMPORALIS 55

Temporalis Muscle

FIGURE 38. There are three accepted approaches to medial inset of the temporalis muscle
for facial reanimation, as shown. In the vermilion border incision, a 1.5-cm incision is made
along both the upper and lower vermilion borders, meeting at the commissure. In patients with
extensive facial rhytids, a nasolabial fold incision yields an acceptable cosmetic result. Another
option is to extend the tern poral and preauricular incision inferiorly below the angIe of the man-
dible. This latter option is the most lengthy, but avoids any direct facial incision. It is favored in
young patients and those opposed to direct facial incisions.
56 CHAPTER!

Temporalis Muscle Tendon

FIGURE 3-9. Mattress sutures are used to

secure the temporalis muscle and fascia to
the lateral border of the orbicularis oris. Direct
contact between the two muscles is thought ta
be important to maximize postoperative facial
mavement. To achieve aral cammissure and
nasalabial fald symmetry at rest, the pull of the
temporalis must be exaggerated at the time of
surgery. This overcorrection will account for the
known stretching and relaxation of the trans-
ferred muscle over time.

RGURE 310. An alternate approach totem-

poralis transfer can be accomplished with use
of the temporalis tendon. The insertion of the
temporalis tendon onto the caronoid process
of the mandible is shown (zygomatic arch and
masseter muscle removed). The line deman-
strates the osteotomy site for coronoidectomy.
 TEMPORALIS 57

Temporalis Muscle Tendon

FIGURE 3-11. The tendon has been dissected

from the coronoid process, and secured to
the modiolus. The oral commissure should be
modestly overcorrected, with exposure of the
first molar.

FIGURE 3-12. The muscle is rotated anteriorly, and secured to a remnant of temporalis fascia at
the anterior aspect of the temporal line. The solid line represents the true origin of the temporalis
muscle along the superior temporal line. The dashed line indicates the neo-origin after transfer,
encompassing only the anterior 40% of the fascial remnant at the superior temporal line.
58 CHAPTER3
REFERENCES
1. Al-Kayat A. Bramley P: A modified preauricular approach
to the temporomandlbular joint and malar arch. Br J Oral
Surg 1978;17:91-103.
2. Ammirati M, Spallone A, Ma J, Cheatham M, Becker
0 : An anatomicosurgical study of the temporal
branch of the facial nerve. Neurosurgery 1993;33(6):
1038-1043.
3 . Antonyshyn 0, Colcleugh RG, Hurst l.N, Anderson C:
The temporalis myo-osseous flap: an experimental study.
Plast Reconstr Surg 1986;77 :406.
4 . Antonyshyn 0, Gruss JS, Birt BD: Versatility of temporal
muscle and fascial flaps. Br J Oral Surg 1988;41: 118.
5. Bakamjian V, Souther S: Use of the temporal muscle flap
for reconstruction after orbito-maxillary resections for
cancer. Plasc Reconm Surg 1975;56:171.
6. Bradley P, Brockbank J: The temporalis muscle flap in
oral reconstruction. A cadaveric, animal and clinical
study.J Maxillofac Surg 1981;9: 139.
7. Byrne PJ, Kim M, Boahene K, Millar J, Moe K: Tempo-
ralis tendon transfer as part of a comprehensive approach
to facial reanimation. Arch Facial Plast Surg 2007;9:
234-241.
8. Cheney ML, McKenna MJ, Ojemann RG, Nadol JB:
Early temporalis muscle transposition for the manage-
ment of facial paralysis. LaryngosaJpe 1995;105(9 Pt 1):
993--1000.
9. Conley J, Patow C: Flaps in Head and Neck Surgery. New
York: Thieme; 1989.
10. Contreras-Gareis R, Martins PO, Braga-Silva J: Endo-
scopic approach for lengthening the temporalis muscle.
Plast Reconstr Surg 2003; 112: 192-198.
11. Deitch RD, Callahan A: Temporalis muscle transplant
for tissue defects about the orbit. Am J Ophthalmol
1964;58:849.
12. Edgerton MT, Tuerk DB, Fisher JC: Surgical treatment
of Moebius syndrome by platysma and temporalis muscle
transfers. Plast Reccmstr Surg 1975;55:305.
13. Ewers R: Reconstruction of the maxilla with a double
musculoperiosteal flap in connection with a composite
calvarial bone graft. Plast Reccmstr Surg 1988;3:431.
14. Gillies HD: Experience with fascia lata grafts in the
operative treatment of facial paralysis. Proc R Soc Med
1934;27:1372.
15. Golovine SS: Procede de cloture plastique de l'orbitc:
apres l'e:xenteration.J Fr Ophtalmol1898;18:679.
16. Habel G, Henscher R: The versatility of the temporalis
muscle flap in reconstructive surgery. BrJ Oral MaxilltJfac
Surg 1986;24:96.
17. Hallock GG: Reconstruction of a lower eyelid dc:fect
using the temporalis muscle. A7171 Plast Surg 1984; 13: 157.
18. Hollinshead WH: Textbook of Anaw1ey. 3rd ed. Hager-
stown: Harper and Row; 1974.
19. Holmes AD, Marshall KA: Uses of the temporalis
muscle flap in blanking out orbits. Plast Reconstr Surg
1979;63:336.
20. Labbe D, Huault M: Lengthening temporalis myoplasty
and lip reanimation. Plast Reconstr Surg 2000;105:1289-
1297; discussion 1298.
21. Mathes S, Nahai F: Clinical Applications for Muscle and
Musculocutaneous Flaps. St. Louis: CV MosbyYearbook;
1982:40.
22. Matsuba HM, Hakki AR, Little )W, Spear SL: The tem-
poral fossa in head and neck reconstruction: twenty-
two flaps of scalp, fascia and full thickness cranial bone.
Laryngoswpe 1988;98:444.
23. May M: Muscle transposition for facial reanimation. Arch
Owlaryngol1985;110:184.
24. May M: Facial reanimation after skull base trauma. Am J
Owl (Nov. Suppl.):62~7 .
25. May M: The Facial Nerve. New York: Thieme; 1986.
26. McKenna MJ, Cheney ML, Borodic G, Ojemann RG:
Management of facial paralysis after intracranial surgery.
Contemp Neura/1991;13:519.
27. Naquin HA: Orbital reconstruction utilizing temporalis
muscle. Am J Ophchalmol 1956;41 :519.
28. Reese AB, Jones IS: Exenteration of the orbit and repair
by transplantation of the temporalis muscle. Am J Oph-
chalmol1961;51:217.
29. Renner G, Davis WE, Templer J: Temporalis pericranial
muscle flap for reconstruction of the lateral face and
head. Laryngoswpe 1984;94:1418.
30. Rubin LR: Reanimation of the Paralyzed Face: New
Approaches. St. Louis: Mosby Yearbook; 1977.
31. Rubin LR: The anatomy of a smile: its importance in
the treatment of facial paralysis. Plast R econstr Surg
1974;53:384.
32. Andersen JG: Surgical treatment of lagophthalmos in
leprosy by the Gillies temporalis transfer. Br J Plast Surg
1961;14:339-345.
33. Rubin LR, Mishiki Y, Lee G: Anatomy of the nasolabial
fold: the keystone of the smiling mechanism. Plast ReaJn-
szr Surg 1989;83: 1.
34. Sheehan )E: The muscle nerve graft. Surg Clin North Am
1935;15:471.
35. Tessier P, Krastinova 0 : La transposition du muscle
temporal dans l'orbite anophtalme. Ann ChiT Plast Esthet
1982;27:212.
""rbe first reported we of the sternocleidomastoid
~ (SCM) muscle in head and neck reconstruction
was by Jiano (20) in 1908 in which it was transposed to a
paralyzed face to restore dynamic reanimation. Schotts-
taedt et al. (33) used the SCM to replace the masseter
muscle in a child who had developed paralysis in the
disttibution of the ttigeminal nerve, which resulted &om
poliomyelitis. Additional cases were reported by Ding-
man et al. (10) and Hamacher (15) who ttansferred
a segment of the SCM muscle with its intact motor
nerve and vascular supplies for replacement of the con-
genitally absent or paralyzed masseter. Owens (29), in
1955, is credited with being the first to report a mus-
culocutaneous flap based on the SCM. He ttansferred
a superiorly based flap but maintained a broad cutane-
ous attachment of the skin in the region of the mastoid
(Fig. 4-1). Owens incorporated the platysma and the
SCM to enhance the blood supply to the skin. Bakam-
jian (4) modified Owens flap by extending the skin ter-
ritory below the level of the clavicle. Littlewood (24)
reported additional experience in wing the extended
SCM flap and identified the conttibutions of the occipi-
tal and posterior auricular arteries to the vascular supply
of the muscle. O'Brien (28) is credited with being the
first to transfer an island of skin overlying the caudal
aspect of the neck with the SCM pedicled superiorly
(Fig. 4-2). Finally, Ariyan (2,3) identified the inferior
vascular supply from the thyrocervical trunk and suc-
cessfully transferred an inferiorly based flap (Fig. 4-3).
The SCM flap has been extensively studied but not
widely wed. It has been criticized on oncologic grounds,
which are related to the safety of preserving this muscle
59
60 CHAPTER4

Internal jugular v.

Common carotid a.

Omohyoid m. Sternocleidomastoid m.

FIGURE 4-1. The original SCM musculocutaneous flap described by Owens (29) had a broad
attachment superiorly at the level of the mastoid and included the SCM and the platysma
muscles. Bakamjian (4) modified this design by extending the skin paddle below the clavicle.

when there are regional lymphatic metastases. The
limited aize of the musculocutaneous flap restricts its
use to amall defecta. Finally, thia flap has been criticized
for the unreliability of the skin paddle and for the con-
tour deformity in the neck following flap transfer. These
iaauea are addreaaed in this chapter.
The SCM ia a round muscle that originates from the
manubrium and the medial aspect ofthe clavicle. It nms
an oblique course in the neck to insert on the mastoid
process and the superior nuchal line (Fig. 4-4). Con-
traction of the SCM leads to tilting ofthe head, bringing
the ipsilateral ear closer to the shoulder. The superficial
layer of the deep cervical fascia splits to aWTOUild the
SCM on both its deep and superficial surfaces (17).
FLAP DESIGN AND UTILIZATION
The evolution in flap design since the inttoduction of
the broad superiorly baaed musculocutaneous flap of
Owens (29) was outlined in the introduction to this
chapter. The superiorly based and inferiorly based
island musculocutaneous flaps, as weD as the SCM used
aa a muscle flap only, are presently the moat commonly
used SCM flaps. Although Bakamjian (4) and Little-
wood (24) extended the stin territory below the clavi-
cle with the superiorly based "peninsular'' flap, there
has been little reported that substantiates the maximum
dimensions when the flap is harvested as an island of
stin. In addition to transfer of the musculocutaneous
 STERNOCLEIDOMASTOID 61

SCM branch

FIGURE 4-Z. The superiorly based SCM island musculocutaneous flap transfers skin from the
caudal aspect of the neck that overlaps the distal third of the muscle and the medial portion of
the clavicle. The primary blood supply to this flap arises from the posterior auricular, the occipi-
tal, and the superior thyroid arteries.
62 CHAPTER4

Branch of
thyrocervical a.

FIGURE 4-1. The inferiorly based SCM island musculocutaneous flap transfers a segment
of skin overlying the upper third of the SCM muscle. The primary blood supply to this flap arises
from the thyrocervical trunk and the superior thyroid artery.
 STERNOCLEIDOMASTOID 63

--,..,_

FIGURE 4-4. The SCM muscle originates from the manubrium and the medial aspect of the
clavicle. It inserts into the mastoid process and the superior nuchal line. The spinal accessory
nerve supplies the motor innervation to the SCM and the trapezius muscles.

unit, there have been numerous reports of using the
SCM to transfer clavicular periosteum (12,37,38) and
segments of the clavicular bone for reconstruction of
the mandible (5,36).
The problem of the donor site contour deformity was
addressed by Alvarez et al. (1) who reported the use of
the split SCM musculocutaneous flap in 1983. Trans-
position of the entire belly of the SCM produced an
obje<:tionable bulge in the midne<:k and a concavity in
the lower ned::. Alvarez et al. (1) described a series of
cases in which either the sternal head or the clavicular
head of the muscle was transferred to the recipient site.
They cautioned that this longitudinal split could only
be carried through appro.Umately two thirds of the mus-
cle's belly in its longitudinal direction.
The ability to transfer the SCM with the preserva-
tion of its vascular and neural supply led to the appli-
cation of this fl.ap to reconstructive problems requiring
dynamic activity. The early report by Jiano (20) in
restoring mimetic activity to the paralyzed face was one
such ez.ample. O,Brien (28) used the SCM to recon-
struct a total lower lip defect, with the skin island pro-
viding the inner lining. The dynamic activity of the
SCM was preserved and believed to have functional
value in restoring oral competence. In the introduc-
tion to this chapter, it was noted that this flap was also
applied to the problem of dynamic restoration of the
masticator muscle sling (10,14,32). Finally, Matulic et
al. (27) reported the combination of the SCM muscle
flap with a forehead cutaneous flap to reconstruct the
64 CHAPTER4
oral cavity following performance of a glossectomy. The
forehead flap provided the inner lining, and the SCM
was transposed to provide dynamic tongue movement.
As in many reconstructive techniques that purport to
restore motion, the documentation by electromyogra-
phy of electrical or contractile activity does not nec-
essarily translate into coordinated functional activity.
An exception to this statement is in facial reanimation
in which muscle transposition has been shown to be an
effective means of restoring mimetic activity. The SCM
has been supplanted by the temporalis muscle because
of the improved axis of pull of this muscle in producing
a symmetric smile. More recently, there are numerous
reported strategies for transfer offree innervated muscle
flaps that produce a more natural reanimation triggered
by the facial nerve rather than by the act of biting.
The SCM muscle flap has also been used to restore
a normal lateral facial contour following parotidectomy
and mandibular reconstruction. Hill and Brown ( 16)
transposed a superiorly based muscle flap over a free
iliac bone graft to achieve a more satisfactory lower
facial contour in secondary mandibular reconstruction.
Bugis et al. (7) reported their experience with the use
of the SCM muscle flap to restore the facial contour in
31 patients following parotidectomy. In addition, they
reported the successful application of this flap in two
patients who had postoperative salivary fistulas. Despite
these findings, the SCM muscle flap was not an effective
method to prevent Frey syndrome in an extensive series
of patients reported by Kornblut et al. (21,22). A group
of 35 patients who underwent parotidectomy and SCM
muscle transposition into the parotid bed were com-
pared with a control group of 35 patients who under-
went comparable ablative procedures but no muscle
transposition. The rationale for transposing the muscle
was to interfere with the presumed mechanism of Frey
syndrome, which is the misdirection of auriculotempo-
ral secretomotor fibers from their normal end organ,
which is the salivary tissue. It is thought that the trans-
sected nerves are rerouted to the sweat glands of the
overlying skin, thereby producing "gustatory sweating."
Komblut et al. reported no difference in the incidence
of Frey syndrome in the two study groups.
The SCM musculocutaneous flap has been used for
oral and pharyngeal mucosal defects since Bakamjian's
(4) initial report of the use of this flap to reconstruct
the palate following radical maxillectomy. As noted pre-
viously, Bakamjian used an extended peninsular skin
muscle flap that was transferred through the posterior
oral cavity. At a second stage, the flap's pedicle was
transsected, with closure of the orostoma. Ariyan (3)
reported 14 cases of either superior or inferior muscu-
locutaneous flaps used in the oral cavity or pharynx. He
noted "partial epithelial loss" in seven cases, but only
one developed a salivary fistula. Reepithelialization of
the denuded areas of the oral cavity was reported, and
biopsies from the healed reconstructed site demonstrated
preservation of the dermal layer. In this report, Ariyan
also described primary closure of the donor site defect
through cutaneous advancement flaps rather than by
application of a skin graft. Additional reports on the
SCM musculocutaneous flap cited varying degrees of
skin viability. Sasaki (32) used four inferiorly based and
one superiorly based flap to reconstruct the floor of the
mouth and tonsillar regions. The skin of one of the supe-
riorly based flaps underwent total necrosis; partial skin
necrosis was reported in two of the remaining inferiorly
based flaps. Despite these complications, there were
no cases of salivary fistulas, which Sasaki attributed
to the viability of the underlying SCM muscle. Marx
and McDonald (25) reported a more favorable expe-
rience with the superiorly based flap in eight patients
in whom they noted distal skin necrosis 2 em from the
tip of the flap. These 8 cases of oral cavity reconstruc-
tion represented a subset of the 16 reported cases of
SCM flaps also used for a variety of other indications.
These authors emphasized the necessity of maintain-
ing the vascular contributions from the superior thyroid
artery and vein. The importance of this contribution
from the superior thyroid pedicle is discussed later in
detail. Finally, Ariyan (2) reported closure of a cervical
esophagostoma by using a superiorly based SCM flap
for inner lining followed by a superior trapezius flap for
outer cutaneous coverage.
The use of vascularized segments of the clavicle sup-
plied by adjacent soft tissue was introduced in the early
1970s as a solution to the frustrating problem of restor-
ing bone continuity following segmental mandibulec-
tomy. Siemssen et al. (36) referred to two of the earliest
reports that used portions of the clavicle to reconstruct
the mandible, which dated back to the beginning of the
20th century. They credited Rydygier (31) with being
the first to transfer an osteocutaneous flap containing
a portion of the clavicle. This was followed by Blair's
(6) description of composite flaps containing clavicle
and rib. Snyder et al. (37) are credited with reviving this
concept with a report issued in the current era in which
vascularized bone was used to restore bony defects of
the maxillofacial skeleton. They reported several cases of
vascularized bone transfer based on regional cutaneous
flaps. They transferred either full- or split-thickness seg-
ments of clavicle with the overlying skin in a two-stage
procedure. This publication was followed by Conley's
(9) report in 1972 of a series of 50 regional bone-con-
taining flaps for mandibulofacial reconstruction. This
series was composed of a variety of different composite
flaps, including the deltopectoral acromion flap, the tra-
pezius-scapular flap, the temporalis--calvarial flap, and
the SCM-clavicular flap. Although Conley reported
three complete flap failures in this series, there were few
details regarding the actual techniques used for each
of these donor sites. He warned about the potential

shoulder morbidity resulting from segmental defects
of the clavicle and advised that a sagittal split be per-
formed to transfer only the outer cortex.
Siemssen et al. (36) reported on a series of 18 patients
who underwent mandibular reconstruction with either
split or segmental segments of the clavicle pedicled on
the clavicular head of the SCM. Although seven bone
flaps were transferred in the primary setting, internal
lining was achieved with either a forehead or a deltopec-
toral flap. There were significant complications in the
group of five patients in this series who underwent split
clavicle transfers, with fractures occurring at both the
donor and recipient sites. In the remaining patients in
this series, a full segment of the clavicle was harvested
anterior to the attachment of the trapezius muscle
(Fig. 4-5). The authors reported little to no shoulder
morbidity in these patients. In addition, the clavicle was
pedicled on the clavicular head of the SCM and passed
FIGURE 4-5. A segment of 1he clavicle, pedicled on the clavicular head of the SCM, may be
transferred as a vascularized bone graft.
STERNOCLEIDOMASTOID 65
under the sternal head to avoid a contour deformity in
the neck. Finally, they speculated about the possibil-
ity of reconstructing a near-total mandibular defect by
transferring the anterior portions of both clavicles with
an intervening segment of the manubrium pedicled on
both SCM muscles.There was only one total Bap necro-
sis in this series.
Barnes et al. (5) reported a similar favorable experi-
ence with the use of this musculoclavicular flap in four
primary and one delayed mandibular reconstruction.
The viability of the neomandible was confirmed with
postoperative technetium scans.
The SCM musculoclavicular Bap has also been used
for rigid support in laryngotracheal reconstruction for
the correction of stenotic segments. Schuller and Par-
rish (34) reported the successful use of vascularized split
clavicle grafts to provide rigid support of the cervical
airway. Approximately one half the circumference of the
6& CHAPTER4
clavicle was harvested and "hollowed out,. with a bone
curette to create a rigid lumen. The bony shell waa then
lined with a free mucosal graft. Alternatively, Tovi and
Gittot (38) described a myoperiosteal flap to achieve
a similar result (Fig. 4-6). The clavicular periosteum,
pedicled on the SCM, waa used to repair noncirCUIIlieJ:..
ential defecta of the larynx and trachea in three patients.
A stent was placed in one patient. All three patients were
successfully decannulated, and at the time of follow-up
endoscopy, the reconsttucted portion of the airway was
relined with normal-appearing respiratory epithelium.
Friedman et al. (12) examined the SCM myoperiosteal
Bap in dogs that underwent tracheal reconattuction. The
growth of new bone from the transplanted periosteum
was documented at the 6- and 9-month follow-ups.
In addition, the patency of the lumen waa preserved.
RGURE 4-6. The SCM myoperiosteal flap transfers vascularized periosteum for use in airway
reconstruction.
Following research on the SCM Bap performed in
dogs, Friedman later reported his series ofusing the SCM
periosteal Bap to reconsttuct defects of the trachea and
subglottic region in humans, resulting from resection of
inwsive thyroid cancer.nus one-stage technique involved
the transfer of periosteum from the clavicle or the manu-
brium, based on the SCM as a carrier. A stent was left
in place for several weeks to facilitate restoration of an
adequate lumen. Ten of eleven patients were successfully
decannulated using this tec:hnique (12,13). It was unclear
from this report as to how large a segment of the trachea
can be restored with this technique. The SCM muscle
or musculoperiosteal Bap can be used to close a window
defect in the trachea and larym, while the same muscle
Bap can be used to bolster a primary tracheal repair by
suturing the muscle over the suture line (Fig. 4-7A,B).
 STERNOCLEIDOMASTOID 67

A
FIGURE 4-7. A:. The inferiorly based SCM can be utilized in tracheal repair following resection
for an invasive thyroid cancer. In limited defects of the trachea, up to one third of the circumfer-
ence, the window can be repaired with an SCM muscle flap, used as a patch. {continued)
68 CHAPTER4

B
FIGURE ~7. (continued) B: Fcllcwing primary repair cf segmental tracheal defects with
end-toend anastomosis, the closure can be augmented with an SCM muscle flap placed over
the suture line. This technique helps to prevent and limit the extent of a leak.

An alternative teclmique for the repair of laryngotta-
cheal stenoses was described by Eliachar and Moscona
(11) who used a SCM island musculocutaneous flap.
This flap augmented the lumen after resection of the
stenotic framework. A T.-tube, with a laryngeal stent,
was kept in place for 4 to 6 weeks after surgery.
NEUROVASCULAR ANATOMY
The vascular supply to the SCM muscle and its overly-
ing skin is arguably the most confusing of any flap used
in head and neck reconstruction. This is one explana-
tion why this flap has not been embraced with a sig-
nificant amount of enthusiasm. The SCM has a type II
vascular supply, according to the classification scheme
----
Occipital a.
(SCM branch)
External carotid a.
Internal jugular v.
Branch of
thyrocervical a.
FIGURE 4-8. The dominant arterial supply to the SCM muscle is from the occipital artery. Minor
vascular contributions arise from branches of the posterior auricular artery, the superior
thyroid artery, and the thyrocervical trunk.
STERNOCLEIDOMASTOID 69
of Mathes and Nahai (26). There is one dominant
pedicle arising superiorly from the occipital artery and
vein and three minor pedicles: a branch of the poste-
rior auricular artery and vein, a branch of the superior
thyroid artery and vein, and a branch of the thyrocervi-
cal trunk (Fig. 4-8). As noted previously, the segmental
nature of the vascular supply allows this muscle to be
pedicled either superiorly or inferiorly. The motor sup-
ply to the SCM is from a branch of the accessory nerve,
which continues across the posterior ttiangle of the neck
to innervate the trapezius muscle as well. There remains
some conttoversy as to whether the contribution to the
SCM's innervation from C2 and C3 is motor or sensory.
The successful ttansfer of skin as a musculocutane-
ous flap requires preservation of the vascular supply
to the muscle and capture of the musculocutaneous
Branches of
superior thyroid a.
70 CHAPTER4
perforators that exit the superficial surface of the mus-
cle. The relationabip of the SCM to the overlying cervi-
cal skin variea, depending on whether one is looking at
the caudal aspect of the neck or the region below the
mastoid tip. The reason for this difference is the pres-
ence of the platysma muscle, which is a sheetlike muscle
of varying thickness that runs in the superficial fascia
of the neck (Fig. 4-9). It arises below the clavicle from
the muscular fascia overlying the pectoralis major and
the deltoid. It courses obliquely across the neck at right
angles to the SCM to blend with the muscles inserting
on the lower lip. The paired platysma muscles are defi-
cient in the midline of the neck; laterally, they overlap
the lower portion of the SCM only to approximately
the midlevel of the neck. Therefore, the caudal half of
the SCM is separated from the overlying skin by a layer
of platysma; the cephalad half has no such intervening
Sternocleidomastoid m.
FIGURE 4-9. The platysma muscle originates from the muscular fascia overlying the pectoralis
major and runs an oblique course across the neck. It completely overlaps the caudal aspect of
the SCM muscle.
layer. This difference can readily be felt by assessing the
relative mobility of the lower neck skin compared with
the tightly adherent skin of the upper neck.
The platySma muscle is a vestige of the panniculus
carnosus in lower animals. The skin is tightly adherent
to this muscle, and it has long been recognized that cer-
vical skin flaps are more viable when the platysma is
included. The platySma has been successfully used as
a carrier of cervical skin, as introduced by Futrell et
al. (14) in 1978. The platysma is primarily supplied by
the submental branch of the facial artery. The platysma
musculocutaneous unit has been shown to be reliable in
subsequent reports (8).
What is unique about the superiorly based SCM
island flap is that a successful outcome requires capture
by a deeper muscle (SCM) of a more superficial muscu-
locutaneous unit (platysma). There is no other flap used

in the head and neck or perhaps elsewhere in the body
in which the blood supply to the skin must traverse two
distinct muscle layers. The cephalad portion of the mus-
cle appears to be a more favorable donor site to harvest
skin because of the lack of the intervening muscle layer.
However, the inferiorly based flap is at a disadvantage
as a result of the smaller vascular pedicle entering the
caudal aspect of the muscle.
Studies that have investigated the vascular supply to
the cervical skin are helpful in shedding light on this
problem. One of the earliest reports that looked at the
vascular contributions from the SCM to the cervical
skin was by Jabaley et al. (19). In a series of cadaver
dissections, these investigators reported an extreme
paucity of musculocutaneous perforators arising from
the lower two thirds of the SCM. They did, however,
identify a direct cutaneous branch from the transverse
cervical artery that penetrates the platysma to supply
the supraclavicular skin.
Two publications from the University of Pittsburgh
reported on a series of fresh cadaveric studies that also
examined the blood supply to the cervical skin. A sum-
mary of the findings in these two studies is enlightening
(18,30). These investigators corroborated the observa-
tions ofJabaley et al. (19) that there are few musculocu-
taneous perforators from the SCM and those that were
present are extremely small. Direct cutaneous perfora-
tors were identified from a number of sources, includ-
ing the occipital, posterior auricular, and superior
thyroid arteries, which were the most consistent. These
three branches of the external carotid artery, therefore,
supply feeders to the SCM muscle and direct perfora-
tors to the skin. In 80% of the cadaver dissections, a
large cutaneous vessel from the superior thyroid artery
was identified coursing around the anterior border of
the SCM, which supplied the platysma and skin ofthe
midneck. It is likely that the success reported by Marx
and McDonald (25) in their series of superiorly based
SCM flaps was directly related to the preservation of
the superior thyroid branches. Ink-injection studies of
the cutaneous branch of the superior thyroid artery
caused staining of the skin of the middle and lower
cervical regions. In one dissection, the direct cutane-
ous branch from the superior thyroid system traveled
on the undersurface of the SCM and then entered the
platysma and the overlying skin between the sternal
and clavicular heads of the SCM (30).
This review of the vascular anatomy points out
potential pitfalls in regard to both the superiorly and
the inferiorly based island flaps. The superiorly based
flap preserves the dominant blood supply to the mus-
cle but is problematic because of the intervening layer
of platysma. The inferiorly based flap relies on the
nondominant contributions to the muscle from the
caudal aspect of the neck after transsecting the domi-
nant cephalad muscular branches. However, the skin
STERNOCLEIDOMASTOID 71
overlying the cephalad portion of the SCM appears
to be more favorably related to the muscle because of
the absence of the platysma. The peninsular skin flap,
as described by Owens (29), should have an excel-
lent chance of viability because of the preservation of
the dominant blood supply to the muscle and direct
cutaneous feeders to the skin entering from the occipi-
tal and posterior auricular branches. Muscle only or
muscle plus periosteum, with or without clavicle, also
appear to be reliable flaps.
POTENTIAL PITFALLS
Many of the potential complications of this donor site
have been discussed in this chapter. The viability of the
skin of either a superiorly or an inferiorly based flap
is questionable. However, the experience of Marx and
McDonald (25) suggests that preservation of the supe-
rior thyroid artery may be extremely important for the
reasons mentioned. It may be possible to mobilize the
superior thyroid pedicle to enhance the arc of rotation.
One of the other major criticisms of this donor site
is its intimate relationship to the region of most com-
mon nodal metastases from the head and neck pri-
mary malignancies. The necessity for a formal radical
neck dissection eliminates this flap as a surgical option.
Modified neck dissections may allow preservation of
the SCM, but its vascular supply is placed in jeopardy.
Transfer of a SCM flap from the contralateral neck may
be feasible. Arguments against violating a potential site
of regional metastases have been raised (23). However,
the opposing point of view in this controversy is that,
in raising the SCM, the posterior fascial layer does not
need to be violated and, therefore, the envelope of deep
cervical fascia that encloses the lymph node-bearing tis-
sue can be preserved.
The largest published series of SCM flaps was
reported by Sebastian et al. (35) in 1994. A total of 121
superiorly based SCM flaps were utilized in 120 patients
with clinically NO necks. The branches to the SCM
muscle arising from the occipital artery were preserved.
Total flap loss occurred in 7.3% of patients while super-
ficial skin loss was reported in 22.7%. Orocutaneous
fistulas were noted in 11.8% of patients. The authors
noted a significantly higher incidence of flap complica-
tions in patients who were previously irradiated. Finally,
nodal recurrence occurred in 5.7% of ipsilateral necks
that were pathologically NO and in 17.4% of ipsilateral
necks that demonstrated pathologically positive nodes.
Acknowledgments
The author would like to acknowledge the contribu-
tions of Dr. Hugh F. Biller to the writing of this chapter
in the first edition of this book.
72 CHAPTER4

Sternocleidomastoid Flap

FIGURE 4-10. The left SCM muscle is outlined

over the left neck with the approximate posi-
tions of the superior, middle, and inferior blood
supplies.

FIGURE 4-11. The skin paddle of a superiorly

based island flap is outlined over the caudal
aspect of the muscle.

FIGURE 4-12. An incision is made around the

perimeter of the skin flap. The caudal aspect
of the flap overlies the clavicle. The incision is
carried through the skin, subcutaneous tissue,
and platysma muscle to expose the superfi-
cial layer of the deep cervical fascia, which
encompasses the SCM.
 STERNOCLEIDOMASTOID 73

Sternocleidomastoid Flap

FIGURE 4-13. The random caudal extension

of the skin is elevated off the clavicular perios-
teum. The inferior attachments of the SCM to the
clavicle are transsected. Care must be taken to
avoid causing any shearing forces between the
SCM and the overlying skin. If a segment of the
clavicle is to be harvested, osteotomies would
be made at this time.

FIGURE 4-14. A vertical incision has been

made to expose the cephalad portion of the
SCM. The inferior blood supply from the thyro-
cervical trunk has been transsected. The middle
blood supply from the superior thyroid artery and
vein is shown entering the deep surface of the
SCM just above the omohyoid muscle.

FIGURE 4-15. Preservation of the superior

thyroid pedicle limits the arc of rotation but
improves the vascular supply to the superiorly
based SCM flap.
74 CHAPTER4

Sternocleidomastoid Flap

FIGURE 4-16. The dominant superior blood

supply arising from either the occipital artery
or directly from the external carotid artery
has been isolated. This branch usually runs a
course cephalad to the hypoglossal nerve.

FIGURE 4-11. By transsecting the caudal

and middle blood supply, the arc of rotation
is greatly improved. This flap can be used for
intraoral, facial, and pharyngeal defects.

REFERENCES 5. Barnes D, Ossoff R, Pecaro B, Sission G: Immediate

1. Alvarez G, Escamilla J, Carranza A: The split ster-
nocleidomastoid myocutaneous ftap. BT J Plasr Surg
1983;36: 183--186.
2. Ariyan S: One-stage repair of a cervical esophagostoma
with two myocutaneous flaps from the neck and shoul-
der. Plast Reconm Surg 1979;63:426--429.
reconstruction of mandibular defects with a composite
sternocleidomastoid musculoclavicular graft. Arch Our
laryngol Head Neck S1ng 1981;107:711-714.
6. Blair VP: S1ngery and Diseases of the Mouth and Jaws.
St. Louis: CV Mosby; 1918.
7. Bugis S, Young J, Archibald S: Sternocleidomastoid flap
following parotidectomy. Head Neck 1990;12:430--435.

3. Ariyan S: One stage reconstruction for defects of the 8. Coleman J, Jurkiewicz M, Nahai F, Matthes S: The pla-
mouth using a sternomastoid myocutaneous flap. PltJn tysma musculocutaneous flap: experience with 24 cases.
Reamstr SUTg 1979;63:618-625. Plast RecMUtr Surg 1983;72:315--321.

4. Bakamjian V: A technique for primary reconstruction 9. Conley J: Use of composite flaps containing bone for
of the palate after radical maxillecoomy for cancer. Plast major repairs in the head and neck. Plast Reconsrr Surg
Reamstr S1ng 1963;31: 103--117. 1972;49:522-526.

10. Dingman RO, Grabb WC, O'Neal RM, Ponitz RJ:
Sternocleidomastoid muscle transplant to masseter area:
case of congenital absence of muscles of mastication.
Plast Reconstr Surg 1969;43:5-12.
11. Eliachar I, Moscona AR: Reconstruction of the laryngo-
tracheal complex in children using the sternocleidomas-
toid myocutaneous flap. Head Neck 1981;4:16-21.
12. Friedman M, Grybaieskas V, Skolnick E, Toriumi D,
Chills T: Sternomastoid myoperiosteal flap for recon-
struction of the subglottic larynx. Ann Otol Rkinol
Laryngo/1987;96:163-169.
13. Friedman M, Toriumi D, Owens R. Grybauskas VT.
Experience with the sternocleidomastoid myoperiosteal
flap for reconstruction of subglottic and tracheal defects:
modification of technique and report oflong-term results.
Laryngoscope 1988;98: 1003-101 1.
14. FutrellJ,Johns M, Edgerton M, Cantrell R. Fitz-Hugh GS:
Platysma myocutaneous flap for intraoral reconstruction.
AmJ Surg 1978;136:504--507.
15. Hamacher E: Sternocleidomastoid muscle transplants.
Plast Reconstr Surg 1969;1: 1--4.
16. Hill H, Brown R: The sternocleidomastoid flap to restore
facial contour in mandibular reconstruction. BT J Plast
Surg 1978;31:143-146.
17. Hollinshead WH: AnatomY joT Surgeons: The Head and
Neck. New York: Harper and Row; 1982:446.
18. Hurwitz D, Rabson J, Futrell ]W: The anatomic basis
for the platysma skin flap. Plast &constr Surg 1983;72:
302-314.
19. Jabaley M, Heckler F, Wallace W, Knott L: Sternocleido-
mastoid regional flaps: a new look at an old concept. BT J
Plast SuTg 1979;32:106-113.
20. Jiano J: Para.lizie faciale dupa extriparea unei tumori a
parotidee trata prin operatia dlui gomoue. Bull Mem Soc
Clin BuckaTest 1908:22.
21. Kornblut A, Westphal P, Michlke A: The effective-
ness of a sternomastoid muscle flap in preventing post-
parotidectomy occurrence of the Frey syndrome. Acta
OtolaTyngol1974;77:368-373.
22. Kornblut A. Westphal P, Michlke A: A re-evaluation
of the Frey syndrome following parotid surgery. ATck
OtolaTyngol Head Neck SuTg 1977;103:258-261.
23. Larson DI..., Goepfert H: Limitations of the
sternocleidomastoid musculocutaneous flap in head and
neck cancer reconstruction. Plast&constr SuTg 1982;3:328-
335.
STERNOCLEIDOMASTOID 75
24. Littlewood M: Compound skin and sternomastoid flaps
for repair in extensive carcinoma of the head and neck.
PlastReconstrSuTg 1967;20:403--419.
25. Marx RE, McDonald DK: The sternocleidomastoid
muscle as a muscular or myocutaneous flap for oral and
facial reconstruction. J OTal Maxillofac Surg 1985;213:
155-162.
26. Mathes S, Nahai F: Clinical Applications joT Muscle and
Musculocutaneous Flaps. St. Louis: CV Mosby; 1982:38-
39.
27. Matulic Z, Bartovic M, Mikolji V, Viras M: Tongue recon-
struction by means of the sternocleidomastoid muscle
and a forehead flap. BT J Plast SuTg 1978;31:147-151.
28. O'Brien B: A muscle-skin pedicle for total reconstruc-
tion of the lower lip: case report. Plast Reconstr SuTg
1970;45:395-399.
29. Owens N: A compound neck pedicle designed for the
repair of massive facial defects: formation, development,
and application. Plast Reconstr Surg 1955; 15:369-389.
30. Rabson J, Hurwitz D, Futrell J:The cutaneous blood sup-
ply of the neck: relevance to incision planning and surgi-
cal reconstruction. BT J Plast Surg 1985;38:208-219.
31. Rydygier LR: Zum Osteoplastischen ersatz nach
Unterkieferresektion. Zentralbl ChiT 1908;36: 1321.
32. Sasaki C: The sternocleidomastoid myocutaneous flap.
ATck OtolaTyngol Head Neck Surg 1980;106:74--76.
33. Schottstaedt E, Larsen I.., Bost F: Complete muscle
transposition. J Bone Joint Surg Am 1955;37:897-919.
34. Schuller D, Parrish RT: Reconstruction of the larynx
and trachea. ATck OtolaTyngol Head Neck Surg 1988;114:
278-286.
35. Sebastian P, Cherian T, Ahamed I, Jayakumar K, Sivara-
makrishnan P: The sternomastoid island myocutaneous
flap for oral cancer reconstruction. ATck OtolaTyngol Head
Neck SuTg 1994;120:629.
36. Siemssen S, Kirkby B, O'Connor T: Immediate recon-
struction of a resected segment of the lower jaw using
a compound flap of clavicle and sternomastoid muscle.
Plast Reconstr SuTg 1978;61 :724--735.
37. Snyder C, Bateman J, Davis C, Warden G: Mandibu-
lofacial restoration with live osteocutaneous flaps. Plast
Reconstr Surg 1970;45:14--19.
38. Tovi F, Gittot A: Sternocleidomastoid myoperiosteal flap
for the repair of laryngeal and tracheal wall defects. Head
Neck 1983;5:447--451.
"T""'be medially based deltopectoral Bap, also refened
~ to as the Bakamjian flap, was a major adwnce in
head and neck surgery when it was popularized in the
early 1960s by V. Y. Bakamjian (2) as a solution to the
problem of pharyngoesophageal reconstruction. The
design of this flap, with its pedicle based at the ster-
num, represented a divergence from the commonly held
belief that the midline of the body was a relatively avas-
cular territory (19). There remains some controversy as
to whether the Bap reported in 1917 by Aymard (1) for
nasal reconstruction was the first description of the del-
topectoral flap. This flap was again described by Joseph
(14) in the 1930s in his book on plastic surgery. Joseph
referred to Manc:hot's description of the vascular ter-
ritories of the body and clearly understood the nature
of the blood supply to this flap (10). For apprazimately
40 years, the deltopectoral Bap remained essentially
1&
dormant in the medical literature until Bakamjian
(2,3) described its versatility and wide application in
head and neck reconstruction. Along with the forehead
Bap introduced by McGregor (18), it was the primary
method for resurfacing cutaneous and mucosal defects
until the late 1970s when musculocutaneous flaps were
introduced. It remains a useful tool in the reconstruc-
tive surgeon's armamentarium, although it has primar-
ily been relegated for use in reconstructing external
cutaneous defects of the neck. The major disadvantages
of this flap include the requirement, in most cases, for
a skin graft to close the donor site, and the unreliability
of the distal portions of this flap when extended over
the deltoid region. I have found the design and harvest
of an island deltopectoral flap to be a useful method to
overcome the need to transfer the medial portion of the
Bap as a peninsula. The color match of the skin paddle
 OELTOPECTORAL 77

provides an excellent method to resurface the anterior

neck to achieve an aesthetically pleaaing result.
The deltopectoral flap is a faaciocutaneoua Bap baaed
on the perforating branches of the internal mammary
artery. Although originally described as having a pedi-
cle baaed on the first three perforators, it is now most
commonly baaed on the second and third. When a clearly
dominant perforator is present, the entire flap could
probably be based on that single vascular pedicle. Pri-
mary tmnafer of the deltopectoral flap may be performed
with a high degree of reliability provided that it does not
extend into the territory overlyiD,g the deltoid muscle.
Distal Bap necrosis occurs with significant frequency
when a Bap extending onto the shoulder is raised without
prior delay. Flap vascularity and the reliability of different
Bap designs are discussed in detail later in this chapter.

FLAP DESIGN AND UTILIZATION

A variety of different flap designs have been described to
reconstruct many different defects in the head and neck.
Greater length and greater diversity can be achieved
when a delay procedure is instituted. The body habitus
of the patient greatly influences the arc of rotation of the
medially based Bap. The optimal situation is a patient
with broad shoulders and a short neck.
There are several d.Uferent ways to ttansfer this flap to
the recipient site. The bridgiD,g portion of the flap can be
tubed over the clavicle and neck skin. A staged secondary
procedure is required to either return or excise the tubed
component. Alternatively, the intervening skin between
the defect and the clavicle may be excised to allow a one-
stage insertion of the entire length of the Bap. Finally, an
island flap can be created by de-epithelializing the proxi- FIGURE 5-1. A modified design of the deltopectoral flap
mal portion of the flap, which is then buried beneath the with the proximal portion de-epithelialized and buried
cervical skin betWeen the defe<:t and the clavicle. When beneath the intervening bridge of skin (see Fig.5-2). The fas-
buried in this fashion, a secondary procedure is not ciocutaneous nature of this flap ensures that this maneuver
required (Fig. 5-1) (13). An alternative approach to cre- will not impede vascularity to the distal portion. The skin
ating an island Bap is to raise the skin over the prozima1 island for this flap is more reliably placed more medially
portion of the island flap in a subdermal plane, which than shown, unless a delay procedure is performed.
facilitates the closure of the donor defect (Fig. 5-2).
Krizek and Robson (17) described the longitudinally
split Bap in which an incision is made through the distal In the initial delay procedure, the upper arm extension
end of the skin paddle creating two separate segments for was folded under the deltoid component to produce a
restoration of the inner and outer lining (Fig. 5-3). This buried skin Bap. The two epithelial surfaces were then
design places less stress on the vascularity to the tip than transferred at the time of the second procedure. An alter-
de-epithelializing a segment and folding the Bap on itself. native solution to the requirement for a double epithelial
A transverse fold in the tip of the flap allows the distal surface is the use ofa skin graft on the undersurface of the
portion to be used for the internal lining of composite flap. The graft may be buried at the time ofan initial delay
defects. However, this technique requires that a loDger procedure (22).
Bap be harvested, and the distal fold occurs in the por- East et al. (8) described the placement of a fenestra-
tion that is the least well vascularized. Bakamjian et al. tion in the distal portion of the deltopectoral ftap for
(5) described the !..-shaped design of the deltopectoral reconstruction of a ttacheostoma. Although I would
Bap, with the short limb of the L extending downward be wary of causing tip necrosis with this technique, the
along the upper arm. This Bap design was used to obtain authors advised that the short arc of rotation required to
an inner lining by using a two-stage procedure (Fig. 5-4). reach the ttacheostoma allows the design of a short flap,
78 CHAPTERS

FIGURE 5-2. The island deltopectoral flap can be harvested by elevating skin flaps over the
proximal portion of the fascial subcutaneous pedicle in a subcutaneous plane in order to allow
transfer of the island of skin. After incising the distal skin island. the proximal flap is elevated in
the plane deep to the fascia. The flap can be tunneled into the neck in a plane over the clavicle.

and, therefore, the fenestra can be placed in a relatively
well-vascularized portion of the flap.
The length of the deltopectoral flap that can be safely
transferred without a delay is somewhat controversial.
Kirkby et al. (15) reported that the end of the Bap could
be safely extended to the tip of the shoulder.When addi-
tional length was required, these authors re<:ommended
the creation of a back cut from the inferior limb of the
medial portion of the Bap across the sternum. They
descnoed the cephalad extension of the cut lateral to the
contralateral internal mammary perforators (Fig. 5-5).
However, the efficacy of this maneuver is somewhat con-
troversial in light of the contention of McGregor and
Jackson (19) that the arc of rotation of a deltopectoral
Bap is more limited by the upper limb of the Bap than
by the inferior limb, as is most commonly believed. The
rationale for this statement is based on the contention
that the skin of the anterior a:Dllary fold is intrinsically
the most redundant portion of the skin of the distal end
of the flap. The inferior limb of the Bap captures that
anterior uillary fold skin, while the superior limb of
this Bap, which is located parallel and just inferior to the
clavicle, captures the less redundant upper skin. It is easy
to demonstrate the relative redundancy of the skin in the
anterior axillary fold by raising an arm abovt: the head
and realizing the amount of skin laxity in that location.
Bilateral deltopectoral Bap transfers havt: been reported
for complex reconstrUctions or recurrent cancers (15).

FIGURE 5-3. The longitudinally split deltopectoral flap
provides two epithelial surfaces. This design is safer than
de-epithelializing a horizontal strip and folding the tip to
achieve an inner lining.
The deltopectoral fiap was t:r&l:ISferred as a microvas-
cular free flap, as first reported by Harii et al. (12) in
1974 and then by Fujino et al. (9) in the following year.
The publication ofFujino et al. described the tranafer of
a de-epithelialized dermis-fat fiap for augmentation of
contour deformities of the head and neck. Percutaneous
Doppler sonography was used to localize the dominant
perforator on whic:b. to base the flap. The free fiap is usu-
ally harvested on the second internal mammary perfora-
tor, which is most commonly the la:rgest. The vascular
pedicle for this fiap is quite short, and it is rarely used
for free tissue transfer due to the abundance of other
donor sites that are available.
David (7) introduced the concept of an inner-
vated deltopectoral flap for intraoral reconstruction
with sensory restoration reestablished through the
supraclavicular nerves of the cervical plexus (Fig. 5-6).
Although he noted excellent sensation when the nerves
OELTOPECTORAL 79
FIGURE 5-4. A delay procedure is required to improve the
chances of successfully transferring distal portions of the
deltopectoral flap, which extend down the arm and around
the shoulder.
were mobilized and not transsected, he did not report
sensory recovery when the nerves were transsected and
then reanastomosed to recipient sensory nerves in the
neck. The potential for tranafer of a sensate deltopeo-
toral free flap is readily apparent. The concept of tran5-
ferring sensate skin to the oral cavity and pharynx to
assist in postoperative rehabilitation was not pursued
until Urken et al. (26) reported the first sensate radial
forearm fiap in pharyngeal reconstruction.
80 CHAPTERS
/
FIGURE 5-5. The desire to improve the arc of rotation of the deltopectoral flap has led to a num-
ber of modifications, including the use of a back cut on the contralateral side of the sternum.
The deltopectoral flap has been applied to a wide
variety of re<:onstructive problems in the head and neck.
As noted previously, Bakamji&n (2) first descnoed this
flap as a solution to the problem of restoring continu-
ity to the gullet following laryngopharynge<:tomy. In
a landmark publication in 1965, he reported a two-
stage technique that involved the transfer of a tubed
deltopectoral flap (Fig. 5-7). Following the initial pro-
cedure, a conttol salivary fistula was created at the lower
end of the tube that was positioned lateral and inferior
to the tracheostoma, permitting a safer and more man-
ageable salivary egress (Fig. 5-8). The stump of the
esophagus was sutured in end-to-side fashion to the skin
tube. After a 3- to 5-week interval, the base of the delta-
pectoral flap was transsected and closed to complete the
pharyngoesophageal reconstruction (Fig. 5-9). Bakam-
jian and Holbrook (4) later described the use of a staged
Deltoid m.
Pectoralis major m.
secondary reconstruction of the pharyngoesophagus by
tubing the deltopec:toral flap on the chest wall prior to
transfer.
Additional experience with the deltopectoral flap led
to its application to intraoral reconsttuction of the tODgUe,
floor of the mouth, tonsil, and pharynx. It has also been
widely used for external defects of the neck, cheek, ear,
and mentum (19). Ingenious teclmiques, albeit through
staged procedures, of reconsttucting extensive mid and
upper facial defects have been reported by "waltzing"
the pedicle to more cephalad regions. Resw:facing hemi-
facial and OlbitomaziDary defects have been descnoed.
(6,23). A favorable body habitus and, more often, the
institution of a prior delay are critical to the use of this
flap for more cephalad defects of the face. McGregor and
Reid (21) descn"hed the combined use of the forehead
flap to achieve internal lining and the deltopeaoral
 OELTOPECTORAL 81

FIGURE 5-i. The supraclavicular sensory nerves arise from C3 and C4 and can easily be traced
through the posterior triangle fat pad to be mobilized or transsected and then anastomosed to
an appropriate recipient nerve in order to transfer sensate skin.
82 CHAPTERS
FIGURE 5-7. The use of1he deltopectoral peninsular flap
for reconstruction of the pharyngoesophageal segment
requires a two-stage procedure. A peninsula of skin is
harvested as a routine deltopectoral flap and passed under
the lower cervical skin into the neck.
flap for external lining when reconstructing through-
and-through defeas of the cheek. Babmjian and Poole (6)
descnoed the use of the deltopeaoral flap for reconstruct-
ing the palate following ablative surgery. In most cases,
e:xcept where an island Bap is created, the use of the delto-
pectoral flap for relining any part of the gullet required the
creation of a control salivary fistula that was subsequently
closed at the time ofreturning the pedicle to the dlestwaD.
NEUROVASCULAR ANATOMY
The blood supply to the deltopectoral flap is derived
from parasternal perforators of the internal mammary
anery and vein, which traverse the intercostal inter-
spaces. The 2-cm. zone lateral to the border of the
sternum should not be violated when raising this Bap
to avoid injury to these vessels. The second and third
perforators are usually the largest in size with external
diameters in the range of 1.2 mm. The venae comitantes
are usually equal or greater in diameter (Fig. 5-10).
The vessels of the deltopectoral Bap run in a plane
superlicial to the fascia overlying the pectoralis major and
deltoid muscles. Although this flap is most commoDly
Proximal portion
of esophagus -------1;...._.,~
FIGURE 5-8. In the first stage, the flap is tubed upon itself.
At the distal end, it is sutured to the pharyngeal opening at
the base of tongue. The flap is partially sutured to the proxi-
mal esophagus with creation of a controlled fistula.
harvested with this fascial layer to protect the circula-
tion, it is not an absolute requirement to do so (16).
A number of articles have been wrinen on the nature
of the vascular supply to the deltopectoral flap and the
implications for safely harvesting skin overlying the
deltoid muscle. A review of the vascular territories of
the upper chest provides a bener understanding of the
potential problems that may arise when using skin from
distal portions of this Bap. The angiosome concept may
be applied to this discussion by defining the source ves-
sels that supply the anterior thoracic skin. The primary
region of the internal mammary perforators extends
from the lateral border of the sternum to the delta-
pectoral groove. This territory is also supplied by mus-
culocutaneous perforators arising from the pectoralis
major. In the region of the deltopectoral groove, there is

RGURE 5-9. In the final stage, the fistula located at the flap to esophageal anastomosis is closed
in order to complete the repair of the pharyngoesophagus (blue srrovi}.
a direct cutaneous artery arising from the thoracoacro-
mial system, which supplies a small area of skin below
the clavicle. The skin of the deltoid territory, lying lat-
eral to the deltopectoral groove, is supplied by muscu-
locutaneous branches arising from the deltoid branch
of the thoracoacromial system and the anterior circum-
fl.ex humeral artery. It is therefore evident that, in rais-
ing a deltopectoral Bap, the skin overlying the deltoid
muscle and the deltopectoral groove, which were previ-
ously supplied by musculocutaneous vessels and direct
cutaneous vessels, respectively, must now be captured
and made exclusively dependent on the internal mam-
mary perforators. In the angiosome model descnoed by
Taylor et al. (25), the blood supply to skin in imme-
diately adjacent angiosomes is usually quite reliable.
However, the pressure gradient of the nutrient fl.ow
diminishes as one moves to the angiosome next in line,
or "once removed," from the primary source vessel. The
skin overlying the deltoid muscle is an angiosome once
OELTOPECTORAL 83
removed from the internal mammary angiosome, and,
therefore, that skin is at risk for partial or total necrosis.
It is possible that the variable pattern of reliability of
the tip of the deltopectoral fl.ap is a function of the size
of the thoracoacromial angiosome that is the middle
territory in this series.With a larger and more dominant
cutaneous branch from the thoracoacromial axis, the
deltoid skin may be rendered less reliable (Fig. 5-11).
The aDgiosome concept provides a &amework for
describillg delay procedures that are used to increase the
reliability ofthe deltoid skin (Fig. 5-4).To capture the blood
supply ofthat tenitory, it is e98elltial to reverse the direction
offl.ow in the adjacent thoracoacromial angiosome and the
third aDgiosome in line overlying the deltoid region. It is
critical that the direction of fl.ow aao98 the choke arteries
that connect adjacent aDgiosomes be uniformly oriented
from the sternum to the tip ofthe shoulder.The most prom-
ising delay procedures an: those that interrupt the source
arteries and veins in the :intermediate and distal angiosomes
84 CHAPTERii
Subclavian a.
Oeltopectoral g
Internal mammary a.
Intercostal
neurovascula
bundle
FIGURE ~10. The dettopectoral flap is supplied by internal mammary perforators, which emerge
from the 2nd and 3rd intercostal space in the parasternal region.
to allow revasa1 of flow and more favorable p.resslm: gra-
dients. 'This was demonstrated by the :fluorescein injec-
tion studies of McGregor and Mozgm (20). A successful
delay procedlm: for the deltopectoral flap muat interrupt
the direct cutmeoua branch of the thOiacoacr:oJ:n syati:Dl
and the distal musculocutaneous branches of the deltoid
adliewd by raising the tip ofthe flap lateral to the deltopeo-
toml gi'O(M! and underminillg in the infraclavicular foaaa.
The sensory nerve supply to the deltopecto.ral skin is
derived from the supraclavicular nerves of C3 and C4
and the anterolateral intercoatlll nerves ofT2, T3, and T4.
The ability to maintain the sensory supply intact lqely
depends on whether a radical neck dissection is pe:rformed.
As noted previously, the report by David (7) of a sensate
deltopectoml flap was the first succeaaful n::sto:ration of
sensation to the reconstructed oral lining (Fig. 5-6).
Internal mammary
perforators to skin
POTENTIAL PITFALLS
The technique of deltopectoral flap harvest is so
straightforward that it is rare to encounter problems
leading to total flap necrosis. The incidence of partial
tip necrosis has varied in different series, depending on
the lecgth of the ftap and the use of a delay procedure.
Park et al. (24) warned that factors contnbuting to flap
loss included diabetes, wound infection, and a radiated
recipient bed. In a series of51 deltopectoral flaps placed
in irradiated beds, Krizek and Robson (16) reported
only fi'VI! major complications. Kirkby et al. (15) noted
an overall totalfl.ap failure rate of 26%, which required
secondary reconstructive procedures. Higher rates of
flap failure were noted in flaps placed for intemallining
and for flaps used in an irradiated field. The total flap
 OELTOPECTORAL 85

FIGURE 5-11. The three major angiosomes of the upper chest and shoulder, moving from
medial to lateral, are the internal mammary, the acromiothoracic, and the deltoid angiosomes.
The regions that are marked represent the approximate territories of these source vessels.
To capture the skin in the deltoid angiosome, or the angiosome once removed, flow from the
internal mammary perforators must traverse the acromiothoracic angiosome, which causes a
pressure gradient prior to reaching the deltoid region.

failure rate of 26% was considerably greater than that
reported in other large series, e.g., 9% (5), 12% (17),
16% (24), and 14% (22) . .Minor complications that did
not require additional surgery ranged from 14% (5) to
26% (17).
.Although extension of the flap over the deltoid leading
to disw ischemia is the most common cause of partial
necrosis, there are a variety of other etiologic factors that
have been implicated, e.g., placement of the flap over a
mandibular K wire, folding of the flap for inner and outer
lining, head movement causing flap tension or kink-
ing, and inadequate oro- or pharyngostomal aper1.'Un:
through which to pass the deltopectoral flap for mucosal
replacement (11). The wide array of flaps available for
oral and pharyngeal defects have limited the current role
of the deltopectoral flap to reconstruction of cervical
cutaneous defects. For this purpose, the deltopectoral
flap should be considered a highly reliable teclmique.
Acknowledgments
The author acknowledges the conttibutions of Dr.
Hugh F. Biller to the writing of this chapter in the first
edition of this book.
86 CHAPTERS

Deltopectoral Flap

FIGURE 5-12. A deltopectoral flap is shown

outlined over the right upper chest. The upper
incision runs just inferior to the clavicle; the
inferior incision extends from the 4th or 5th
interspace, parallel to the upper incision. The
distal extent of the flap is determined by the
defect. The dominant pedicle to this flap arises
in the 2nd or 3rd interspace, and therefore to
ensure viability, the base should overlie these
two interspaces.

FIGURE 5-13. Incisions are made alang the

upper, lower, and distal margins. The incision
is made through the skin, subcutaneous tissue,
and deltopectoral fascia.

FIGURE 5-14. As the flap is elevated laterally

to medially in a subfascial plane, the deltopec-
toral groove {large arrows) is encountered. The
direct cutaneous branch arising from either
the deltoid or acromial branches (small arrow)
has been isolated in the cephalad aspect of the
groove.
 OELTOPECTORAL 87

Deltopectoral Flap

FIGURE 5-15. The medial extent of the dis-

section is usually to a point approximately
2 em lateral to the sternal border. Although an
internal mammary perforator (srrowt has been
isolated to demonstrate its position, these ves-
sels are not identified in the dissection for fear
of injuring the blood supply to the flap.

FIGURE 5-16. The deltopectoral flap has

been completely isolated and transposed over
the clavicle onto the anterior neck. Closure
of the donor site is accomplished by wide
undermining and the use of retention sutures.
Although primary closure may be achieved,
the use of a skin graft is the norm.

REFERENCES 6. BakamjianVY, Poole M: Muillofacial and palatal recon-

1. Aymard JL: Nasal re(;onsttuction with a note on nature's
plastic surgery. l..a'IIUf 1917;2:888.
2. BakamjianVY: A two-stage method for phar}lngoesopha-
geal re(;onsttuction wilh a primary peaoral skin flap.
Pfast Recorutr Surg 1965;36: 173.
3. Bakamjian VY: Total re<X~nstruction of pharynx with
medially based deltopectoral skin flap. NY St4U J Med
1968;1:2771.
4. BakamjianVY, Holbrook I..: Prefabrication techniques in
cervical phar}lngoesophageal reconstruction. Br J Plast
Swg 1973;26:214.
5. Bakamjian VY, Long M, Rigg B: Ezperience with me
medially based deltopectoral flap in reconstructive
surgery ofthe head and neck. BrJPfast SUYg 1971;24:174.
structions with 1he deltopeaoral flap. Br J Platt Swg
1977;30:17.
7. David JD: Use of an innervated deltopectoral flap for
intraoral reconstruction. Pfast Ruomtr SUYg 1977;60:377.
8. East C, Flcmm.ing A, Brough M: Tracheostomal r:on-
struction using a fenestrated deltopectoral skin flap.
J lAryngol Otcl1988;102:282.
9. Fujino T, Tanino R, Sugimoto C: Microvascular transfer
of free deltopectoral dermal-fat flap. Pkm ReCO'/UtT Swg
1915;55:428.
10. Gibson T, Robinson D: The mammary artery pectoral
flaps of Jacques Joseph. Br J Platt Sfi.Tg 1976;29:370.
11. Gingrass R., Culf N, Garrett W, Mladick R: Complica-
tions with lhe deltopectoral flap. Pkm &ccnm Surg
1972;49:501.
88 CHAPTERS
12. Harii K, Ohmori K, Ohmori S: Free deltopectoral skin
flaps. Br J Plast Surg 1974;27:231.
13. Jackson I, LangW: Secondary esophagoplasty after pha-
ryngolaryngectomy using a modified deltopectoral flap.
PlastReconszrSurg 1971;48:155.
14. Joseph J: Nasenplastik und sonstige Gesicktsplastik nebs
reinem Anhang uber Mammaplastik und einige weitere Oper-
ationen aus dem gebiek der ausseren Korperplastik. Leipzig:
Verlag von Curt Kabitzchl; 1931:673--677,811-819.
15. Kirkby B, Krag C, Siemssen 0: Experience with the dd-
topectoral flap. Scand J Plast Reconstr Surg 1980; 14:151.
16. Krizek T, Robson M: The deltopectoral flap for recon-
struction of irradiated cancer of the head and neck. Surg
Gynecol Obstet 1972; 135:787.
17. Krizek T, Robson M: Split flap in head and neck recon-
struction. Am J Surg 1973; 126:488.
18. McGregor I: The temporal flap in intraoral cancer: its use
in repairing the post-ex:cisional defects. Br J Plast Surg
1963;16:318.
19. McGregor I, Jackson I: The extended role of the
ddtopectoral flap. Br J Plast Surg 1970;23:173-185.
20. McGregor I, Morgan G: Axial and random pattern flaps.
Br J Plast Surg 1973;26:202.
21. McGregor I, Reid W: The use of the temporal flap in the
primary repair of full-thickness defects of the cheek. Plast
Reconstr Surg 1966;38: 1.
22. Mendelson B, Woods J, Masson J: Experience with the
deltopectoral flap. Plast Reconstr Surg 1977;59:360.
23. Nickell W, Salyer K, Vargas M: Practical variations in the
use of the deltopectoral flap. South MedJ 1974;67:697.
24. Park J, Sako K, Marchette F: Reconstructive experience
with the medially based deltopectoral flap. Am J Surg
1974;128:548.
25. Taylor GI, Palmer J: The vascular territories (angiosomes)
of the body: experimental study and clinical applications.
Br J Plast Surg 1987;40: 113.
26. Urken ML, Vickery C, Weinberg H, Biller HF: The neu-
rofasciocutaneous radial forearm flap in head and neck
reconstruction-a preliminary report. Laryngoscope
1990;100:161.
~e myriad of regional and free flaps that are pres-
~ ently available has made it possible to transfer skin
of virtually any size and shape to the facial region. How-
ever, the requirement of using skin of similar color and
texture to that of the native facial skin greatly limits the
available donor sites (2,5,6,12,13,14,18,20). In survey-
ing the available donor sites to accomplish that goal, the
skin of the forehead and the posterior neck perhaps come
closest to mimicking the facial skin. Tissue from both of
these regions can be transferred to the midface for cheek
and nasal reconstruction by using the scalp as a carrier.
In the 1930s, Gillies (14,15) developed the principles of
transferring forehead skin for nasal reconstruction.
.MacGregor (21) is credited with introducing the
forehead flap for intraoral reconstruction in 1963 (17).
Along with the deltopec:toral flap, this flap was the
primary source of tissue for bead and neck reconstruc-
tion until the development of the extensive range of
regional musculocutaneous and free flaps that began
in the 1970s and continued to the present. Although
MacGregor's (21) transverse forehead flap is rarely
used today because of the morbidity of the skin-grafted
donor site, the forehead continues to be widely utilized
for resurfacing cutaneous and soft tissue defects of the
nose and cheek (3). A variety of forehead flap designs
have been reported during the last six decades, includ-
ing the median, the paramedian, the sickle, and the
oblique patterns of forehead skin transfers.
DuringWorldWarn, Converse (7,8,9,10) introduced
the anterior scalping flap, which transferred the skin of
one half of the forehead. The flap is based on the con-
tralateral vascular supply of the scalp. It was originally
89
90 CHAPTERS
developed as a variation of Gillies "up-and-down flap"
and includes the forehead skin, the scalp, and galea
with its vascular supply derived from the vessels of the
forehead and anterior portion of the scalp. Despite
the necessity of a two-stage procedure, this technique
continues to have specific applications for the recon-
struction oflarge nasal, upper lip, and cheek cutaneous
defects for which color match and tissue pliability are
priorities. Its role in partial and total nasal reconstruc-
tion has been replaced by the paramedian forehead flap.
Arena (1) recognized that the posterior neck skin has
similar qualities to that of the facial skin and reported
a two-stage technique for transferring skin from this
region to the face. Using surgical principles similar to
those developed by Gillies (14) and Converse (7) for
transferring forehead skin to the nose, he took advan-
tage of the rich vascularity of the scalp to use it as a vehi-
cle for transporting favorable skin from the nape of the
neck to the midface. This technique may be considered
an extension of the flap developed byWashio (29,30) in
which postauricular skin is transferred for reconstruct-
ing limited facial and nasal defects (22). Closure of this
donor site was most often accomplished with a skin
graft or a scalp advancement flap. As a result of its pos-
terior location, the donor-site defect is more easily cam-
ouflaged than the deformity caused by an anterior scalp
flap. In addition, the pre-expansion of this flap with a
tissue expander not only enhances the vascularity of this
flap but also allows the surgeon to achieve primary clo-
sure while transferring a sufficient amount of tissue to
resurface virtually the entire cheek (23).
FLAP DESIGN AND UTILIZATION
The anterior scalp flap is most useful in the reconstruc-
tion of large nasal and cheek defects. The pliability of
the distal aspect of the flap allows it to be contoured
to recreate the anatomic details that are required to
satisfactorily reconstruct the nose ( 4,25). The anterior
scalping flap offers some distinct advantages compared
with other forehead flaps. The design of the flap pro-
vides an adequate pedicle length due to the extensive
undermining that can be safely performed. Because of
the limited tension in the forehead skin that is trans-
ferred, lower nasal and columellar reconstruction can
be safely performed when required. When columellar
reconstruction is needed, an adequate vertical length
is essential to allow the tip of the reconstructed nose
to be sufficiently projected (Fig. 6-1) (11). The area
of flap harvest is limited to one half of the forehead,
which facilitates concealment of the donor site. This
flap should be considered in patients whose foreheads
are narrow or who have a low hairline. In such patients,
a median or paramedian forehead flap would transfer
hair-bearing skin when resurfacing caudal nasal defects.
Alternatively, laser depilation can be performed either
before or after transfer to eliminate hair growth on the
reconstructed portion of the nose. One of the unique
features of the anterior scalp flap is that there is suf-
ficient length to fold the flap on itself to create an inner
lining for both the ala and the columella.
It is often helpful to fashion a template of the nasa-
facial defect that can be used to design the area of skin
to be transferred from the forehead. The pattern that is
created should be as accurate as possible to minimize
the amount of skin transferred and the necessity for sec-
ondary flap debulking (2,3,16).
Secondary flap division is customarily performed at
21 days unless the recipient bed has been compromised
by prior radiation or scar. At the time of the initial flap
transfer, the donor site is covered with a full-thickness
skin graft harvested from the postauricular or supracla-
vicular areas (24).
The anterior scalping flap's biggest drawback is the
donor-site defect. The aesthetic deformity can be mini-
mized by preserving innervated frontalis muscle. Place-
ment of the skin graft over this muscle improves the
contour of the forehead and preserves expressive move-
ment in this region (24). As a secondary procedure, the
donor site can be reduced by serial excisions or resur-
faced with a temporofasciocervical flap (19,26). This
technique may be particularly necessary in male patients
and in cases in which hyperpigmentation of the graft
develops. Additional options for donor-site camouflage
include changes in the patient's hairstyle and the use of a
tissue expander to allow advancement of the contralateral
forehead for full-thickness skin coverage of the defect.
When contemplating the use of a posterior scalp flap, it
is important to examine the texture and color of the pos-
terior neck skin to determine its suitability for replacing
skin in the midface region. In women who have longer
hairstyles, the skin of this region tends to be well protected
from the effects of the sun. In addition, longer hair makes
the camouflage of this defect much easier. The posterior
neck skin in patients who have spent considerable time
in the sun may be unsuitable for resurfacing cutaneous
defects of the face. In most individuals, however, the pro-
cess of photoaging tends to affect the skin in the posterior
neck in a manner similar to that of the skin of the face (28).
The scalp is the thickest skin in the human body. The
dermis and epidermis of the scalp region vary in thick-
ness from 3 to 8 mm. However, the skin in the postau-
ricular area and the posterior neck is much thinner and
more pliable. It is therefore suitable for the reconstruc-
tion of large defects of the nose, cheek, and orbital cav-
ity. Its use in the reconstruction of the upper and lower
lips has also been described (29,30). In addition to its
use in oncologic surgery, the posterior scalp flap may be
extended onto the posterior shoulder to provide a large
area of skin that may be utilized to replace areas of scar
contracture caused by trauma, irradiation, or burns.
 ANTERIOR AND POSTERIOR SCALP 91

FIGURE 6-1. The anterior scalp flap transfers up to one half of the forehead skin, using the
scalp as a carrier. Its primary advantages are that it provides skin of the closest color and tex-
ture to the skin of the cheek and nose. In addition, it has the viability and the length to achieve
a detailed reconstruction of the caudal portion of the nose. Inner lining of the caudal portion
of the nose may be achieved by folding this flap on itself. When significant defects of the nose
require reconstruction, then local or distant flaps may be required to achieve an adequate
lining. This factor is of paramount importance when replacing the architectural support of the
nose by the use of free bone grafts. A reliable inner lining, under these circumstances, is critical
for the protection and revascularization of these nonvascularized structural grafts.

NEUROVASCULAR ANATOMY artery should be identified prior to surgery by palpation

The scalp is supplied by a rich array of arteries, includ-
ing the superficial temporal, supraorbital, supratroch-
lear, occipital, and postauricular (Fig. 6-2). There are
significant anastomotic channels between the different
primary scalp vessels that make it possible to transfer
large areas of the scalp on a single arterial pedicle.
The anterior scalp Bap is supplied by the supratroch-
lear, supraorbital, and superficial temporal vessels of the
side opposite to that in which the forehead skin is har-
vested. The frontal branch of the superficial temporal
or Doppler sonography. This branch should be incol.'-
porated by designing the fl.ap so that the transverse and
vertical supraauricular limbs of the incision that crosses
over the scalp are placed behind this vessel.
The venous drainage to this area is reliable. The
supraolbiW veins run superficial to the fronWis muscle
and communicate with the frontal branch of the superli-
cial temporal vein and the supraorbital vein. AD these veins
contribute to the venous egress in the anterior scalping flap.
The vascular supply of the posterior scalping Bap is
similar to that of the anterior fl.ap. The contnbutions
92 CHAPTER li

Superficial temporal a. Superficial temporal a.

(parietal branch) (frontal branch)

Posterior
aurtcular a.

Splenius capitis m.

Galea aponeuro11ce.

Cranium
 ANTERIOR AND POSTERIOR SCALP 93

from the occipital artery are transsected in the process
of raising this flap, and, therefore, it is entirely depend-
ent on the blood flow from the anterior system. The
parietal branch of the superficial temporal artery is
preserved when harvesting the posterior scalp flap, and
this branch plays a significant role in ensuring an ade-
quate circulation to the posterior neck skin. The venous
supply to the posterior scalp flap parallels that of the
arteries. The full extent of the skin territory that can
be harvested with this flap is unknown. We have safely
harvested skin to the level of the scapular spine, which
increases both the surface area and the arc of rotation.
The sensory supply to the anterior scalp is primarily
derived from the supraorbital nerve, which is a branch
of the ophthalmic branch of the trigeminal nerve.
The auriculotemporal branch of the trigeminal nerve
supplies sensation to the temporoparietal scalp. Con-
tributions from the cervical plexus supply sensation to
the posterior scalp through the greater auricular and the
greater and lesser occipital nerves.
POTENTIAL PITFALLS
The rich vascularity of the scalp makes it uncommon
for ischemic complications to occur in either the pos-
terior or anterior flaps. These two flaps are unique in
the head and neck because of the fact that the pedicle
is located either cephalad or on an even plane to the
defect, and, therefore, the effects of gravity are not prob-
lematic in causing flap separation from a poor recipient
bed. The latter problem is often encountered in heavily
irradiated wounds when the flap pedicle is located in a
dependent position.
It is virtually impossible to push the limits of the vas-
cularity of the anterior scalping flap due to the fact that
the lower limit of that flap is the eyebrow, which can-
not be violated. Alternatively, the limits of the posterior
scalping flap are not as clearly defined. Extension of the
skin paddle laterally and inferiorly can push the limits
of the vascularity. Problems with tip ischemia, primarily
venous in nature have been encountered (23). Ifthe clin-
ical situation permits, a delay procedure, with or without
tissue expansion has proven to be extremely helpful in
avoiding tissue loss when a larger flap is required (27).
POSTOPERATIVE WOUND CARE
Both the anterior and posterior scalp flaps have the
disadvantage of requiring a two-stage procedure.
The nutrient vascular flow through the scalp must be
maintained for 2 to 3 weeks until neovascularization at
the recipient site has occurred. The interval between the
first and second procedures is uncomfortable for the
patient because of the cosmetic deformity of the dis-
placed scalp and the necessity for biologic dressings
over the denuded portion of the skull. The patient must
be advised preoperatively of these factors in order to be
psychologically prepared.

FIGURE 6-2. The scalp has a rich vascular supply that arises from the supratrochlear, the supraorbital, the two major
branches of the superficial temporal, the occipital, and the posterior auricular arteries. There are significant anastomoses
between all of these systems that allow long narrow flaps to be transferred if at least one of these major arteries is incorpo-
rated in the base of the flap. The posterior scalp flap involves the transfer of skin from the nape of the neck region overlying
the splenius capitis and trapezius. The major layers of the scalp are shown. The vascular channels are located in the galea
and subcutaneous tissue layers. The loose areolar layer that separates the galea from the periosteum is a relatively avascu-
lar plane that is responsible for the mobility of the scalp over the bone.
94 CHAPTER&

Anterior Scalping Flap Dissection

FIGURE 6-3. The primary vascular supply to

the anterior scalp flap is from the supratroch-
lear and supraorbital vessels in conjunction
with the frontal branch of the superficial tempo-
ral artery and vein.

FIGURE 6-4. The anterior scalping flap has

been outlined to transfer skin from the right side
of the forehead. It is often fabricated from a tem-
plate of the defect to transfer only that portion of
the forehead that is needed. However, the aes-
thetic result is improved by skin grafting a defect
that extends from the eyebrow to the hairline, as
well as by using a full-thickness graft. The exten-
sion of the incision across the vertexcfthe scalp
to the contralateral ear ensures vascularity
through the three dominant pedicles of this flap.
 ANTERIOR AND POSTERIOR SCALP 95

Anterior Scalping Flap Dissection

FIGURE 6-5. The skin of the forehead is

elevated over the frontalis on which a split- or
full-thickness skin graft is
subsequently applied.

FIGURE 6-6. It is important when making the

lateral incision to preserve the innervation to
the frontalis. After the upper limit of the frontalis
has been reached, the level of dissection is
changed to the supra periosteal plane, which is
carried over the remainder of the skull.
96 CHAPTER&

Anterior Scalping Flap Dissection

FIGURE 6-7. The anterior scalp flap has been

elevated. The large area of denuded skull is
noted. The transitional zone can be easily seen
on the undersurface of the flap. In addition the
vascular supply to this flap from both the super-
ficial temporal and the supraorbital system of
vessels is readily seen on the undersurface of
the flap. Undermining over the contralateral
forehead is performed to provide adequate
mobility to achieve caudal transposition of the
forehead skin.

FIGURE 6-8. A Ia rge quantity of thin pliable

skin can be transferred to reconstruct total or
near-total nasal defects. Further undermining
of the contralateral scalp allows the forehead
skin to be placed onto the upper lip or cheek as
needed. Following this stage of the procedure,
the denuded skull must be carefully covered
with a biologic dressing for the 2- to 3-week
period prior to the second stage.
 ANTERIOR AND POSTERIOR SCALP 97

Anterior Scalping Flap Dissection

FIGURE 6-9. The scalp is transferred back

to the donor site, leaving the forehead defect,
which was previously covered at the time of
the first procedure with a split- or full-thickness
graft. Smaller defects may be covered with
a skin graft harvested from the postauricular
region. Advancement of the contralateral fore-
head may be achieved by either serial excision
or use of a tissue expander.

RGURE 6-10. Harvest of the Posterior Scalp-

ing Flap. The dominant blood supply to the pos-
terior scalping flap is derived from the anterior
blood supply to the scalp through the supraor-
bital, the supratrochlear, and both branches of
the superficial temporal artery and vein. The
occipital and posterior auricular branches that
supply the posterior scalp are transsected in
the process of elevating this flap.
98 CHAPTER&

Posterior Scalping Flap Dissection

FIGURE 6-11. The incisions for raising the

posterior scalp flap are shown. The extension in
the postauricular sulcus is required to achieve
adequate mabilization of this flap. It is impera-
tive that the postauricular incision stop at the
superior attachment of the helix to avaid violat-
ing the vascular supply from the superficial tem-
poral vessels. The dotted line in the midline af
the scalp demonstrates a possible extension af
the incision, depending on the degree to which
the flap must be mobilized to achieve tension-
free closure of the defect. Anterior extension of
the midline incision can be performed without
concern about compromise of the flap vascular-
ity. However, the fulcrum point at the top of the
ear, is usually the limiting factor in rotation of
the flap over the cheek. The caudal and lateral
extension of the skin flap provides a greater arc
of rctation and a greater amount of non-hair-
bearing skin. The occipital hairline is a distinct
landmark for the upper limits of usable skin.

RGURE 6-12. The skin of the posterior neck is

elevated superficial to the trapezius, splenius
capitis, and levator scapulae. At the superior
nuchal line, the plane of dissection changes to
a supraperiasteallevel, which is continued over
the remainder of the skull.
 ANTERIOR AND POSTERIOR SCALP 99

Posterior Scalping Flap Dissection

FIGURE 6-13. Elevation of the scalp is contin-

ued until the posterior neck skin can be placed
in the desired recipient defect. The midface can
be easily reached and a tension-free closure
performed without doing extensive undermining.

FIGURE 6-14. The arc of rotation can be

increased to reconstruct the nose or upper lip
by extending the incision in the midline of the
scalp. Extensive undermining does not in any
way compromise the blood supply to the pos-
terior neck skin, which runs through the galea
and subcutaneous tissue layers.
100 CHAPTER&

Posterior Scalping Flap Dissection

FIGURE 6-15. The skin of the posterior neck

has been detached from the scalp pedicle. Dur-
ing the time interval prior to the second surgical
procedure, the denuded posterior skull must be
covered with a biologic dressing. A skin graft
is placed over the donor defect in the neck that
overlies the posterior neck muscles.

FIGURE 6-16. The posterior neck defect can

often be well camouflaged in individuals with
longer hairstyles and by the use of high-col
Ia red shirts. Over time, the aesthetic appear-
anee of this defect improves. The size of the
defect may be diminished by serial excision or
the use of a tissue expander.

REFERENCES
1. Arena S: The posterior scalping flap. L aryngoscope
1977;137:98-104.
2. BlairVP: Reconstructive surgery of the face. Surg Gynecol
Obstet 1922;34:70 1.
3 . Burget GC, Menick FJ: Aesthetic Reconnrucrion of the
Nose. St. Louis: CV Mosby; 1994:57-91.
4. Coiffman F: Total reconstruction of the nose. In: Stark
RB, ed. Plastic Surgery of the Head and Neck. Vol. 1.
NewYork: Churchill; 1986:704-705.
5. Coleman CC: Scalp flap reconstruction in head and neck
cancer patients. Plmt Reconnr Surg 1959;24:45.
6. Conley J: Regicnal Flaps of the Head and Neck. Stuttgart:
Georg Thieme Verlag; 1976.
7. Converse JM: A new forehead flap for nasal reconstruc-
tion. Proc R Soc Med 1942;35:811.
8. Converse JM: Reconstruction of the nose by scalping flap
technique. Surg Clin NorthAm 1959;39:335.
9. Converse JM: Clinical application of the scalping flap in the
reconstruction of the nose.PlastReconnrSurg 1969;43:247.
10. Converse JM: Full-thickness loss of nasal tissue. In:
Converse JM, ed. Reconsr:rucu"ve Plastic Surgery. Vol. 2.
Philadelphia: WB Saunders; 1977:1236.
11. Converse JM, McCarthy JG: The scalping forehead flap
revisited. Clin Plast Surg 1981;8:413.
12. Denneny EC, Denneny J III: Forehead and scalp recon-
struction. In: Papel ID, Nachlias NE, eds. Facial Plastic
and Reconstructiw Surgery. St. Louis: Mosby-Year Book;
1992:392-398.
13. Friedman M : Parietal occipital nape of neck flap. Arch
Otolaryngol Head Neck Surg 1986;112:309.
14. Gillies liD: Plastic Surgery of the Face. London: Oxford
University Press; 1920.
15. Gillies liD: The development and scope of plastic sur-
gery. Northwest Unif.J BuU 1935;35:1.
16. Gonzalez-Ulloa M: Restoration of the facial covering by
means of selected skin in regional aesthetic units. Br J
Plast Surg 1956;46:265.
ANTERIOR AND POSTERIOR SCALP 101
17. Hamaker RC, Singer MI: Regional flaps in head and neck
reconstruction. Otolaryngol Clin North Am 1982; 15:99.
18. Joseph}: Nasenplastik und sonstige GeisidJ.toplastik nebst
einem Anhang uber Mammaplastik und einige weitere
Operationen aus dem Gebiete der a.ussereu Korperplas-
tik. In: BinAtlas und Lehrbuch. Leipzig: Kabitzsh; 1931.
19. Juri J, Juri C, Cerisola J: Contribution to Converse's flap
for nasal reconstruction. Plast Reconstr Surg 1982;69:697.
20. Kazanjian VH: The repair of nasal defects with the
median forehead flap: primary closure of the forehead
wound. Surg Gynecol Obstet 1946;83:37.
21. MacGregor lA: The temporal flap in the intraoral defects:
its use in repairing postexcisional defects. Br J Plant Surg
1965;16:318-335.
22. Maillard GF, Montandon D : The Washio tempororetro-
auricular flap: its use in 20 patients. Plast Reconstr Surg
1982;70:550.
23. Mandell DL, Genden EM, Biller HF, Urken MI... Pos-
terior scalping flap revisited. Arch Otolaryngol Head Neck
Surg 2000; 126(3):303-307.
24. McCarthy JG, Converse JM: Nasal reconstruction with
scalping flap. In: Brent B, ed. TheArristry of Reconstructiw
Surgery. St. Louis: CV Mosby; 1987.
25. Millard DR: Total reconstructive rhinoplasty and a miss-
ing link. Plast Reconstr Surg 1966;37:167.
26. Schimmelbusch C: Bin neues Verfahren der Rhinoplas-
tik und Operation der Sattelnase. Verh Duch Ges Chir
1895;24:342.
27. Smet HT: Tissue Transfers in Reconstructifle Surgery.
NewYork: Raven Press; 1980:6-7.
28. Stark RB, Khoury F: Anatomy of the skull, scalp, and
brow. In: Stark RB, ed. Plastic Surgery of the Head and
Neck. Vol. 1. NewYork: Churchill Livingstone; 1987:3-6.
29. Washio H: Retroauricular-temporal flap. Plast R.eccnstr
Surg 1969;43: 162- 166.
30. Washio H: Further experiences with the retroauricular
temporal flap. Plast Reconstr Surg 1972;50:160.
~e desire to ttansfer tissue of similar color and te:z:ture
~ to the facial skin for aesthetic restoration ofcutmeous
de:feas of the lower and middle thirds of the face led to the
development of the submental island ftap by Mar1in et al.
in 1993 (12). They described this flap as either a regional
or a free flap and noted that it had the added benefit of a
well-camouflaged donor site scar.Aside from its advantage
of transferring skin of favm:able quality for facial recon-
struction, it has the versatility of t:rallSfe.rring haiJ:I..beariD,g
skin in males for beard and mustache restoration, as well as
the capacity to transfer vascularized bone from the lower
border of the mandlble (3,22). It has also been utili7.ed as a
vascul&rized subcutaneous tissue Bap for facial augmenta-
tion in cases of hemifacial microsomia (17).
This donor site has been utilized for reconstruc-
tion of the upper aerodigestive tract with mucosal
102
defects of the oral cavity and larynx being described
(5,20). When utilized in the reconstruction of intraoral
mucosal defects, the surgeon must be mindful of the
fact that the first echelon lymph nodes in the subman-
dibular and submental basins are likely to be subop-
timally dissected and removed when harvesting this
flap for cases involving malignancies that metastasize
to this region (20).
The submental fl.ap has also been utilized to repair
mucosal defects of the pharyngoesophagus for cases of
stenoses as well as the closure of pharyngocutaneous
fistulae (2,20).
The reverse fl.ow modification of this fl.ap has pur-
ponedly increased its cephalad reach in order to expand
its use for reconstruction of the upper third of the face,
including the periorbital region (4).

FLAP DESIGN AND UTILIZATION
Age plays a critical role in determining the vertical
dimension of skin that can be harvested from this donor
site and still achieve primary closure. The "pinch test"
is utilized to determine the amount of skin laxity that
exists in a vertical direction, in the anterior neck, that
permits upward to 6 to 8 em of skin to be harvested in
patients with significant skin redundancy. However the
vertical dimension for safe flap harvest is significantly
greater, as indicated by injection studies by Faltaous
and Yetman (4), which documented that a skin territory
of 10 em by 16 em can be safely based on a single sub-
mental artery. In a horizontal direction, the flap can be
designed from one mandibular angle to the contralat-
eral angle, and, therefore, a skin paddle with as much
as 16 em of length can be safely transferred based on a
unilateral blood supply ( 4). Transfer of a skin flap based
on a bilateral blood supply has been reported in order
to bring hair-bearing skin to the upper lip (9). How-
ever, the rich vascular supply from a unilateral arte-
rial supply is usually sufficient to perform a safe and
efficient transfer with a wider arc of rotation than can
be achieved with a bilateral blood supply. The smallest
flap dimension that can be harvested is the skin terri-
tory surrounding the dominant perforator located in
the region overlying the ipsilateral anterior belly of the
digastric muscle. Doppler localization of that perforator
can be performed in a manner similar to perforator flaps
harvested elsewhere in the body.
As noted above, the dimensions of the flap that can
be harvested are limited in the vertical direction by the
laxity of the anterior cervical skin that can be recruited
for tension-free primary closure. In the horizontal direc-
tion, the limits of safe harvest are usually determined by
the angles of the mandible, beyond which the vascularity
to the skin becomes compromised on both the ipsilat-
eral and the contralateral sides. Preharvest tissue expan-
sion of the anterior cervical skin in order to facilitate the
donor-site closure was alluded to by Martinet al. (12)
in their initial description of this donor site. However, to
date, there have been no reported cases of utilizing this
strategy. In order to preserve the blood supply to the
flap, it would be prudent to place the expander below
the hyoid bone in order to avoid compromise of the
perforators to the skin.
When harvesting a subcutaneous flap with the platy-
sma muscle, the dimensions of the flap can be broadened,
so long as the skin is elevated off of the subcutaneous tis-
sue and utilized in the anterior cervical closure. Smaller
skin islands can be harvested with subcutaneous exten-
sions, depending on the needs ofthe reconstruction (17).
Hair-bearing skin for use in upper lip and chin
reconstruction has been reported in males in order to
restore the mustache and beard (9,22). Alternatively,
THE SUBMENTAL ISLAND 103
the hair-bearing quality of the skin in this region may be
problematic when placing the skin in the midface region
or when using it to reconstruct mucosal defects of the
upper aerodigestive tract, unless the skin has already
been irradiated or the recipient site will be irradiated in
the postoperative setting.
Martin et al. (12) described the potential to harvest
vascularized bone from the inner table of the mandi-
ble at the level of the mentum. Use of a split mandible
harvest with preservation of the outer table maintains
the contour of the lower border of the jaw in this criti-
cal location. Yilmaz et al. (22) reported the successful
transfer of bone to the upper maxilla for restoration of
contour. Although they reported a successful transfer of
this composite flap, there was no long-term follow-up to
gage the viability of the bone. The shape of the anterior
mandible at the level of the symphysis is ideally suited
for reconstruction of the premaxilla. We have utilized
this composite flap for that purpose but have little long-
term follow-up, and the patient was not rehabilitated
with either a tissue-hom or an implant-born prosthesis.
Prefabrication and multiple flap combination strate-
gies have been employed in order to expand the ver-
satility of this donor site. Tan et al. (18) described a
two-stage procedure involving the initial placement of
a 3-em strip of costal cartilage between the skin and
the platysma muscle in order to reconstruct an exten-
sive defect in the columella. One month after the initial
stage, the composite flap was transferred. Subsequent
stages were required to thin the soft tissue in order to
achieve the final result (18).
Barthelemy et al. (1) reported the combination of
the submental flap with the temporoparietal fascial flap
for cheek reconstruction in patients suffering from oro-
facial noma. In the first stage of this reconstruction, a
split-thickness skin graft was placed over the temporo-
parietal fascia and left in place for 5 days in order to pre-
pare the flap for use as the inner lining of the cheek. At
the second stage of this procedure, the scarred tissue of
the cheek was excised and the prefabricated temporo-
parietal fascial skin graft composite flap was transferred
along with a submental flap for resurfacing the exterior
of the cheek. This technique was successfully employed
in five patients with extensive deformity of the cheek
caused by noma.
A novel approach for reconstruction of the oral com-
missure was devised and reported by Koshima et al.
(10). The submental flap was combined with the dor-
salis pedis flap from the first toe web space, in order to
provide both inner lining and cheek skin repair in two
cases of full-thickness loss. The thin pliable skin from
the dorsum of the foot was used to restore the mucosal
loss from the upper and lower lip, while the submental
flap was used to restore the cheek skin with color- and
texture-matched skin (10).
104 CHAPTER 7

Several strategies have been reported to lengthen the
reach of the submental flap in order to expand its appli-
cation. The caliber of the facial artery and vein makes
them very cooducive to transfer this flap as a microvas-
cular free flap in order to reach the upper third of the
face and scalp. Martin et al. (12) reported one case of a
free submental flap in a patient who sustained a gunshot
wound that destroyed the ipsilateral submental vessels.
I have utilized this technique in patients in whom the
submental flap was the ideal dooor site for recoostruc-
tioo, but the ipsilateral level 1 lymph nodes required
complete dissectioo to satisfy the oocologic needs of the
patient. Harvest of the flap on the contralateral facial
vessels permits both the oncologic and reconstructive
goals to be met.
The complete mobilization of the submental and
facial artery and vein is usually sufficient to reach the
middle third of the face and lower periorbital regioos
(Fig. 7-1). Extension of the venous pedicle can be
achieved by taking advantage of reverse flow in the
cQD.D.ecting venous branch between the retromandibu-
lar vein and the facial vein (Fig. 7 -2). While the venous
anatomy of the neck is susceptible to variations, the
classic descriptioo of the venous anatomy includes an
anastomosing branch betWeen the facial vein and the
retromand:&bular vein that combine to create the com-
mon facial vein. If the common facial vein is ligated and
divided, then reverse flow in the communicating branch
permits venous drainage to be established into the
external jugular system, thereby lengthening the venous
pedicle by several centimeters (Fig. 7-3).
The third optioo for lengthening the vascular pedi-
cle is to employ the reverse flow strategy by ligating
the antegrade blood supply and basing the flap on the

RGURE 71. Harvest of the submental flap is demonstrated. The facial artery and vein are
ligated at the level of the lower border of the mandible. The facial artery and vein are mobilized
to achieve a greater arc of rotation for antegrade flow into the flap.
 THE SUBMENTAL ISLAND 105

Postaurtcular
vein

"'HM.,.,,_- - + - - - Retromandlbular
vein

'-----+--- Posterior branch of

vein retromandlbular vein

------ vein
Extemaljugular

FIGURE 1-Z. The venous anatomy of the neck is quite variable. A common description of the
superficia l veins includes the retromandibular vein which divides into anterior and posterior
brsnche~. The anterior branch joins with the facial vein to become the common facial vein that
drains into the interns/ jugular vein. The posterior brsnch joins with the retraauricular vein to
form the external jugular vein.
10& CHAPTER 7
FIGURE 7-3. Greater length of the venous pedicle can be obtained by ligating the common facial
vein in order to establish reverse flow through the anterior branch of the retromandibular vein.
retrograde arterial and venous blood supply through the
distal facial artery and vein (Fig. 7-4). The arterial pres-
sure head through the distal facial artery is more than
adequate to supply the submental flap. This was rec-
ognized by Martinet al. (12) in their original descrip-
tion of this donor site and referred to it as a "distally
based flap" rather thm the nomenclature that has been
adopted as the "reverse flow flap."The potential to base
this flap on reverse flow in the facial artery was dem-
onstrated by Neligan et al. (13) in their description of
the use of the distal facial artery as a source of recipi-
ent vessels for free tissue transfer to the head and neck.
These authors measured the arterial pressure in the
distal stump of the facial artery and found it to he on
average 55 6.3 mm Hg. They alluded to the example
of the distally based radial forearm flap used in hand
reconstruction in which blood flow through the palmar
arch is sufficient to produce retrograde flow through the
radial artery (13). The reverse flow submental artery
flap represents a strategy for transfer of this flap to the
upper third of the face by ligating the proximal facial
artery and vein and rotating the flap under the marginal
mandibular branch of the facial nerve. Without this
mobilization under the branch of the facial nerve, the
fulcrum point for rotation of the flap would be the Ill.al'-
ginal branch and thereby limit the arc of rotation.
Karac:al et al. (7) reported the successful use of
the reverse flow strategy for periorbital defects in six
patients. These authors reported one case of temporary
paralysis of the facial nerve and one case of venous con-
gestion. The latter resolved spontaneously over the first
postoperative day. Kim et al. (8) used the reversed flow
submental flap in three cases of nasal reconstruction
and descnoed the superficial win as the basis for venous
outflow in the flap. That superficial vein is the facial vein
that drains into the common facial vein, in contradis-
tinction to the venae comitants that run parallel to the
submental artery. Sterne et al. (16) reported a case of
total flap loss in a patient who underwent transfer of a
submental flap using the reverse flow orientation. They
 THE SUBMENTAL ISLAND 107

FIGURE 7-4. Retrograde flow has been reported to enhance the arc of rotation of the flap for
use in defects on the upper third of the face. ligation of the proximal facial artery and common
facial vein establishes retrograde flow though the facial artery and vein. The skin paddle must
be swept under the marginal mandibular nerve in order to maximize the arc of rotation.

attributed this loss to progressive venous congestion
and speculated that it was due to a valve in the facial
vein located at the lower border of the mandible. On
the basis of this one adverse event, they advised that a
separate venous anastomosis should be performed (16).
The number of cases of reverse fl.ow submental fl.aps
reported in the literature is limited and therefore it is
difficult to make a definitive statement regarding the
safety of this approach with respe<:t to the risk ofvenous
congestion. It appears from the review of the experience
of the authors noted above, that this approach has been
used successfully with only one adverse event.
One further option that can be entertained for use in
expanding the cephalad reach of this fl.ap is to interpose
vein grafta on either the arterial, venous, or both limbs
of the vascular pedicle. In this approach, antegrade 1iow
is preserved, but there is an additional amount of work,
and potential risk, involved in performing the requisite
microvascular anastomoses, albeit to very large cabber
vessels in the facial artery and the common facial vein.
NEUROVASCULAR ANATOMY
There are several important anatomic points to be con-
sidered in discussing the neurovascular anatomy of the
submental fl.ap. One of the most important is related
to the location of the cutaneous perforators and their
relationship to the anterior belly of the digastric mus-
cle. There is a conttoversy that exists in the literature
regarding the necessity to include that muscle in the
harvest of the submental fl.ap. The submental artery not
only supplies the overlying skin in the submental and
submandibular regions but also gives off branches to
the lower lip, the periosteum, as well as the mylohyoid
and digastric muscles (Fig. 7-5).
Whetzel and Mathes (21) identified 14 arteries that
provide the major blood supply to each side of the face,
neck, and scalp.They described 11 anatomically distinct
cutaneous territories that are akin to the angiosomes
described by Houseman et al. (6). In a detailed ana-
tomic study, Houseman et al. described 13 angiosomes
of the head and neck; however, they did not divide the
territory supplied by the submental branch specifically
from the remainder of the territory supplied by the
facial artery.
Through latex injections of the dominant nutrient
arteries, Whetz:el and Mathes (21) were able to iden-
tify and characterize the dominant perforating vessels to
the overlying skin in the 11 territories that they identi-
fied in the head and neck region. While the perforating
branches from the ttansverse facial, zygomatico-orbital,
suprattochlear, supraorbital, and superficial temporal
vessels could be predicted to within a 2-cm region with
95% confidence limits, the perforators from the sub-
mental and descending branch of the posterior auricu-
lar arteries were found to be less consistent. Through
selective ink injections, they defined the cutaneous tel.'-
ritory supplied by the submental artery as being an area
5 em x 5 em extending superiorly over the mandible to
108 CHAPTER 7

Mental
neurova~Jar
bundle

FIGURE 7-5. The submental artery arises as a branch of the facial artery, which courses over
the superior surface of the submandibular gland. The facial artery continues over the mandible.
The submental artery runs anteriorly along the surface of the mylohyoid muscle and gives off
cutaneous perforators in direct proximity to the anterior belly of the digastric muscle. The nerve
to the mylohyoid muscle travels in close proximity to the submental artery.

about the level of the oral commissure, anteriorly to a
point 1.5 em lateral to the oral commissure, posteriorly
to a point 2 em anterior to the anterior border of the
sternocleidomastoid muscle, and inferiorly to a point
3 em below the mandibular border. This territory is
obviously smaller than the one reported by both Martin
et al. (12) and Faltaous andYetman (4).
Perforating vessels connect the deep fascial plane to
the subdermal arterial plexus. These vessels represent
fasciocutaneous perforators as opposed to the muscu-
locutaneous perforators seen in other parts of the body
and more commonly in the anterior portion of the face.
The submental perforator was reported by Whetzel
and Mathes (21) as the only vessel that emerged from
the swface of the platysma to traverse the subcutane-
ous tissue to the dermal layer in this region. Faltaous
and Yetman (4) described two major cutaneous perfo-
rators, with one arising proximal to the digastric and
the second is given off medial to that muscle. They also
descn'bed several smaller perforators arising directly
from the anterior belly of the digastric muscle. In the
detailed cadaveric dissections performed by Kim et al.
(8), only one reliable perforator was found to be perfus-
ing the unilateral submental territory in 87.5% of cases.
The same authors reported that there was a single arte-
rial perforator supplying the overlying skin in 56.3% of
their clinical dissections. The location of that perforator
was quite variable in that series. The authors noted that
the perforator was located at the medial border of the
anterior belly of the digastric muscle in 29% of cases,
at the lateral border in 37.5%, and on either side of the
anterior belly in the remaining 33%. There was little
correlation in the location of the perforator on the two
sides of the anterior neck (8).
The submental artery arises from the facial artery
and travels over the upper portion of the submandibular

gland in an anterior direction. The point of its branching
from the facial artery is usually after that vessel emerges
from its intraglandular course on the superior surface
of the submandibular gland (Fig. 7-5). After running
anteriorly in a groove on the upper surface of the gland,
the submental artery continues forward on the surface
of the mylohyoid muscle. Magden et al. (11) reported
that the submental artery was superficial to the sub-
mandibular gland in 69% of their dissections, while it
ran between the superior border of the gland and the
mandible in the majority of the remainder. In one case,
the submental artery was found to pass through the sub-
mandibular gland. It gives rise to numerous branches
that supply the submandibular salivary gland and the
platysma muscle (11). In the dissections performed by
Faltaous and Yetman (4), the submental artery was
found to run deep to the anterior belly of the digastric
in 70% of cases and superficial to that muscle in the
remainder. Very similar results were noted by Magden
et al. (11) who noted that 81% ofthe submental vessels
ran deep to the anterior belly of the digastric muscle.
The diameter of the submental artery ranges from
1.0 to 1.5 mm, while the diameter of the facial artery
is 2.0 to 2.8 mm. Alternatively, the facial vein diameter
is 2.2 to 3.2 mm and the common facial vein is slightly
larger at 2.4 to 3.5 mm (4).
Due to the variable point of origin of the perforators
and the variable relationship of the submental artery to
the anterior belly of the digastric muscle, it would seem
prudent to harvest the anterior belly of the digastric
muscle with this flap in order to ensure the capture ofthe
dominant blood supply to the overlying skin. However,
in their original description of this donor site, Martin
et al. (12) did not include this muscle and reported no
flap failures. The functional impact of the loss of the
anterior belly of the digastric is relatively small and
therefore the risk to benefit ratio would favor the sacri-
fice of that muscle. To take this one step further, Patel
et al. (14) described a modification of the harvest of this
flap by incorporating the mylohyoid muscle, which these
authors feel helps to avoid any potential injury to the
submental vessels or its perforating branches to the skin.
In my experience, the submental flap can be reliably
harvested without incorporating the mylohyoid muscle
in the flap, as long as the surgeon adheres to the strategy
of including the anterior belly of the digastric muscle.
In the anatomic study by Magden et al. (11), detailed
vascular information was reported. On average the
length of the facial artery from its origin to the takeoff of
the submental branch is 27.5 mm while that of the facial
vein is 19.3 mm. These distances represent the potential
gain in the arc of rotation by harvesting a flap on the
facial vessels rather than the submental artery and vein.
Magden et al. (11) reported that the vessel diameter of
the submental artery and vein were 1.7 and 2.2 mm,
which are very similar to the ones reponed by Faltaous
THE SUBMENTAL ISLAND 109
and Yetman (4). Demonstrable vascular connections
between the submental vascular systems on opposing
sides were reportedly seen in 92% of the cadaveric dis-
sections performed by Magden et al. (11).
There are numerous nerves in the submandibular
triangle, of which the marginal mandibular branch of
the facial nerve has already been mentioned. Its vari-
able course places its caudalmost dissent either below
or above the lower border of the mandible and therefore
the only safe way to preserve this structure is to iden-
tify it in order to avoid deformity of the lips. The motor
nerve to the mylohyoid muscle also passes through the
submandibular triangle and consistently lies deep to the
submental artery (Fig. 7-5). That nerve can, therefore,
be preserved in all cases provided that a meticulous dis-
section of this region is performed. Magden et al. (11)
reported that the submental artery crosses the nerve to
the mylohyoid muscle, on average, 16.8 mm from the
point of origin of the submental artery.
ANATOMIC VARIATIONS
In their original anatomic and clinical description of
the submental flap, Martinet al. (12) reponed one case
in which the submental artery arose as an independent
branch from the external carotid artery in close proxim-
ity to the takeoff of the facial artery. In this particular
case, they described the cutaneous perforating branch
as arising 8 em anterior to the mandibular angle, very
distal in the course of this unusual submental artery.
In three other cases, these authors noted a more proxi-
mal takeoff of the submental artery between 2 and
3 em from the origin of the facial artery from the exter-
nal carotid artery. Martin et al. (12) identified no sig-
nificant variations in the submental vein and noted a
consistent communicating branch between the facial
vein and the external jugular vein, allowing an alternate
basis for venous egress in all cases.
In one cadaveric dissection, Magden et al. (11)
described the lingual artery arising from the submental
artery. This was the only such anatomic variation of its
kind reported.
POTENTIAL PITFALLS
Perhaps the most likely risk associated with the harvest
of a submental island flap is injury to the marginal man-
dibular branch of the facial nerve. In most cases, the
nerves on both sides are at risk in the dissection. This
complication is best avoided by early identification of
those nerves soon after completing the upper incision
and beginning the cephalad portion of the dissection.
The incidence of marginal mandibular nerve injury in
reported series is very low. Pistre et al. (15) identified
110 CHAPTER 7
one case of a temporary weakness in 31 patients who
underwent this procedure. They reported that identifi-
cation of the nerve early in the procedure was not criti-
cal when the upper dissection is performed by staying
close to the undersurface of the platysma layer in the
region between the angle and the anterior body of the
mandible. Alternatively, Sterne et al. (16) reported 2 of
12 patients suffered marginal mandibular nerve palsy
for an incidence of 16%.
Partial flap loss has been reported in most larger
series, with 2 cases out of 31 described by Pistre et al.
(15) and 1 case of 12 patients reported by Sterne et al.
(16). The reverse flow orientation of flap transfer
appeared to be more hazardous with Sterne et al. (16)
reporting one complete failure, which they attributed to
a venous problem secondary to a valve in the facial vein
that led to intractable congestion.
The transfer of hair-bearing skin into the oral cav-
ity and other portions of the upper aerodigestive tract
can also be problematic as reported by Vural and Suen
(20) in a case of reconstruction of a hemilaryngectomy
defect. This problem is restricted to males and, in par-
ticular, to those who have not been irradiated.
The ease of closure of the donor site is a function
of the size of the flap and the laxity of the skin of the
anterior neck. Radiation to this region will place added
stress on the closure and mandates a flap design with
more conservative vertical dimensions. Wound break-
down and the need for a skin graft have been reported
in this population of patients who underwent transfer of
this flap following radiation (20).
The issue of safety of transfer of this flap in patients
who are at risk for lymph node metastases requires
that the surgeon be very careful in patient selection for
use of this donor site. Preoperative imaging will pro-
vide information regarding the presence of suspicious
macroscopic adenopathy, but microscopic disease in
submental and submandibular lymph nodes can be
problematic in patients with primary malignancies that
put them at risk. Although technically feasible, it is very
challenging to skeletonize the submental vessels to the
extent that a thorough level 1 lymph node dissection
can be performed without compromising the circula-
tion to the flap. I have been very reluctant to utilize this
donor site in that patient population and have chosen
alternative donor sites or elected to harvest the submen-
tal flap as a free flap based on the contralateral facial
vessels when the tissue from this donor site was clearly
superior to any other reconstructive option. In the lat-
ter strategy, the ipsilateral nodes can be safely removed
without placing the flap vascular supply at risk.
The placement of the upper skin incision approxi-
mately 1 em behind the lower border of the mandible
helps to avoid a conspicuous scar. In addition, it is
imperative for the surgeon to avoid undermining the
skin over the mandible in order to prevent postoperative
eversion of the lower lip. Restoration of the cervicomen-
tal angle is accomplished by suturing the anterior cer-
vical skin to the perihyoid soft tissue in order to avoid
blunting of that angle and altering the normal contour
of the neck.
PREOPERATIVE MANAGEMENT
There are very few preoperative measures that need
to be taken prior to flap harvest. The most important
is related to patient selection and the dimensions of
the flap required and the risk of metastases to level
1 lymph nodes. A careful determination of the amount
of skin that can be safely harvested is best made with the
patient in the neutral position, prior to extension of the
neck. The pinch test can be performed with the patient
in this position in order to determine a safe amount of
skin that can be harvested and still achieve a primary
closure.
POSTOPERATIVE CARE
Maintaining the head in a neutral or partially flexed
position during the postoperative period will help to
keep tension off the suture line, especially following
harvest of flaps with a larger dimension or in patients
who have been previously irradiated. A suction drain
is placed in order to avoid hematoma formation in the
widely undermined anterior neck. This complication
has been reported in most series (12,15,19).
 THESUBMENTALISLAND 111

Submental Flap

FIGURE 7-6. Harvest of a submental flap

begins with identification of the approximate
position of the facial artery and vein as well as
the marginal mandibular branch of the facial
nerve (yellow line).

FIGURE 7-7. The lower border of the mandible

(dotted line) is an important landmark to ensure
that the outline of the flap does not extend
superior to that line. The position of the hyoid
bone is also marked.

FIGURE 7-8. The approximate position of the

anterior and posterior bellies ofthe digastric
muscle is marked. The ipsilateral anterior belly
will be harvested to ensure capture of the cuta-
neous perforator.
112 CHAPTER 7

Submental Flap

FIGURE 19. A large submental flap has been

marked for harvest. The maximum horizontal
dimensions extend from the angle of the man-
dible on one side to the angle on the other.
The vertical dimension is limited by the pinch
test to determine the vertical height of the
flap that can be harvested and still achieve
primary closure. The smallest flap that can be
harvested requires more precise identification
of the cutaneous perforator in order to reduce
the vertical dimension of the flap.

FIGURE 7-10. The surgical procedure is initi-

ated by making the inferior incision.
 THE SUBMENTAL ISLAND 113

Submental Flap

FIGURE 711. The subplatysmal dissection is

elevated only to the point of identification of the
intermediate tendon of the digastric muscle.

FIGURE 712. The superior incision has been

made along the lower border of the mandible.
The attachment of the anterior belly of the
digastric muscle (blue srrowt has been identi-
fied, as well as 1he intermediate tendon (yellow
arrow).
114 CHAPTER 7

Submental Flap

FIGURE 713. The facial vein has been dis-

sected and isolated in its path over the sub
mandibular gland (blue srrowt. The submental
vein is identified as a branch coursing over the
top ofthe gland in an anterior direction (yellow
arrow). The marginal mandibular branch of the
facial nerve must be identified to avoid injury to
it and resultant deformity of the lower lip.

FIGURE 7-14. The facial artery (red arrow)

and facial vein (blue srrow) have been isolated
both proximally and distally as they continue
over the mandible {red snd blue arrows Mth
white outline). The submental artery and vein
have been isolated coursing in an anteromedial
direction under the flap (green arrow).

FIGURE 7-15. The submental artery and vein

(blue arrow) are shown following removal of the
submandibular gland. The nerve to the mylohy-
oid muscle runs deep to the vascular pedicle
and can be preserved (yellow arrow).
 THE SUBMENTAL ISLAND 115

Submental Flap

FIGURE 7-16. The anterior belly of the digas-

tric muscle insertion into the lower border of
the mandible has been severed.

FIGURE 7-11. The intermediate tendon of the

ipsilateral digastric muscle has been transected
to allow harvest of the anterior belly so as to
protect the cutaneous perforator supplying the
skin.

FIGURE 7-18. The flap is elevated from distal

to proximal or from the contralateral side
toward the side of the vascular pedicle. The
plane of dissection on the contralateral side is
over the submandibular gland fascia (yellow
srrow) and over the ipsilateral anterior belly of
the digastric muscle (blue arrow).
116 CHAPTER 7

Submental Flap

FIGURE 7-19. Flap elevation viewed from the

contralateral side with the submandibular gland
preserved. The marginal mandibular branch of
the facial nerve should be identified on both
sides to avoid injury.

FIGURE 7-20. As the dissection proceeds

medial to the contralateral anterior belly of
the digastric muscle, the plane deepens to the
fascia overlying the mylohyoid muscle (yellow
arrow). As the ipsilateral anterior belly of the
digastric muscle is approached, the cutaneous
perforator is identified on the undersurface of
the flap Iblue arrow).

FIGURE 721. Harvest of the ipsilateral

anterior belly ofthe digastric muscle is shown
to protect the perforator that may arise either
medial or lateral to that muscle.
 THESUBMENTALISLAND 117

Submental Flap

FIGURE 7-22. The vascular pedicle to the flap

is identified from the contralateral orientation
as the dissection proceeds deep to the ipsilat-
eral anterior belly of 1he digastric muscle . The
perforator to the skin has been demonstrated
(blue arrow). The submental artery and vein
have been isolated (green arrow). A branch of
the pedicle to 1he mylohyoid muscle {MM) must
be ligated (red arrowt.

FIGURE 7-23. A close-up view of the vascular

pedicle {red arrowt to the flap is shown with the
nerve to the mylohyoid muscle demonstrated
deep to the submental vessels (yellow srrowt.
A superior branch of the submental vessels
courses toward the lower border of the man-
dible and supplies the periosteum (blue srrow).

FIGURE 72.4.. The flap {yellow srrowt has

been isolated on the ipsilateral facial artery and
vein. The mylohyoid muscle is shown following
removal of the anterior belly of the digastric
muscle (blue arrow).
118 CHAPTER 7

Submental Flap

FIGURE 7-25. Extensive mobilization of the

facial artery and vein has been performed to
optimize the arc of rotation of the flap that has
been transposed to the midface regioll.

FIGURE 7-26. Harvest of the submental flap as

a free flap has been performed with isolation of
the facial artery and vein.

FIGURE 1-11. The undersurface of the

submental flap shows the anterior belly of the
digastric muscle (blue arrow).
 THE SUBMENTAL ISLAND 119

Submental Flap

FIGURE 728. Closure of the defect is per-

formed with wide undermining of the lower
anterior cervical skin in order to advance it
superiorly. The surgeon should avoid under-
mining the superior skin over the mentum in
order to avoid eversion of the lower lip.

FIGURE 729. Closure of the neck is per-

formed with a deep suture to the undersurface
of the flap that is suspended to the periosteum
ofthe hyoid bone. This maneuver is required
to restore the normal cervicomental angle.
120 CHAPTER7

Submental Flap

FIGURE 1-30. Multiple subJres should be

placed to the hyoid bone in order to ensure
creation of that angle.

FIGURE 1-31. Final skin closure has been

aceom plished with flexion of the neck to rei ieve
tension on the repair.

REFERENCES
1. Barthelemy I, Martin D, Sannajust J, et al: Prefabri-
cated superficial temporal fascia flap combined with
a submental flap in noma surgery. Plast Recomtr Surg
2002; 109:936-940.
2. Demir Z, Him V, Celebioglu S: Repair of pharyngocuta-
neous fistulas with the submental artery island flap. Plast
Reconm Surg 2005;115:38--44.
3. Demir Z, Kurtay A, Sabin U, et al.: Hair-bearing sub-
mental artery island flap for reconstruction of mustache
and beard. Plast Reromtr Surg 2003;112:423-429.
4. Faltaous A,Yetman R:The submental artery flap: an ana-
tomic study. Plast Reconstr Surg 1996;97:56--60.
5. Genden EM, Buchbinder D, Urken ML: The submental
island flap for palatal reconstruction: a novel technique.
J OralMaxillajac Surg 2004;62(3):387-390.
6. Houseman N, Taylor I, Pan W: The angiosomes of the
head and neck: anatomic study and clinical applications.
Plast Recomtr Surg 2000;1 05:2287-2313.
7. Karacal N, Ambarcioglu 0, Topal U, et al: Reverse-flow
submental artery flap for periorbital soft tissue and socket
reconstruction. Head Neck 2006;28:40-45.
8. Kim JT, Kim SK, Koshima I: An anatomic study and
clinical applications of the reversed submental artery
island flap. Plast Reconstr Surg 2002;109:2204-2210.
9. Kitazawa I<, Harashina T, Thira H, Thkamatsu A: Biped-
icled submental island flap for upper lip reconstruction.
Ann Plast Surg 1999;42:83.
10. Koshima I, Inagawa I<, Urushibara I<, Moriguchi T:
Combined submental flap with toe web for reconstruc-
tion of the lip with oral commissure. Br J Plast Surg
2000;53:616--631.
11. Magden 0, Edizer M, Tayfur V, Atabey A: Anatomic
study of the vasculature of the submental artery flap. Plast
Reconstr Surg 2004; 114: 1719-1 723.
THE SUBMENTAL ISLAND 121
12. Martin D, Pascal JF, Baudet J, et al: The submental island
flap: a new donor site: Anatomy and clinical applications
as a free or pedicle flap. Plast Reronm Surg 1993;92:867-
873.
13. Neligan P, She-)Ue H, Gullane P: Reverse flow as an
option in microvascular recipient anastomoses. Plast
Reconstr Surg 1997;100:1780--1785.
14. Patel U, Bayles S, Hayden R: The submental flap: a
modified technique for resident training. Laryngoscope
2007;117:186-189.
15. Pistre V, Pelissier P, Martin D, et al: Ten years of expe-
rience with the submental flap. Plast Reconstr Surg
2001;108:1576-1581.
16. Sterne G, Januskiewicz P, Hall P, Bardsley A: The sub-
mental island flap. Br J Plast Surg 1996;49:85--89.
17. Tan 0, Bekir A, Parmaksizoglu D: Soft tissue aug-
mentation of the middle and lower face using the
deepithelialized submental flap. Plast Reconstr Surg
2007;119:873-879.
18. Tan 0, Kiroglu Q, Atik B,Yuca K: Reconstruction of the
columella using the prefabricated reverse flow submental
flap: a case report. Head Neck 2006;28:653--657.
19. Uppin S, Ahmad Q, Yadav P, Shetty K: Use of the sub-
mental island flap is orofacial reconstruction-a review
of 20 cases. J Plast Reconm Aesrlzet Surg 2009;62(4) :514-
519.
20. Vural E, Suen J: The submental island flap in head and
neck reconstruction. Head Neck 2000;22:572-578.
21. Whetzel T. Mathes S: Arterial anatomy of the face: an
analysis of vascular territories and perforating cutaneous
vessels. Plast Reconstr Surg 1992;89:591-603.
22. Yilmaz M, Menderes A, Barutcu Q: Submental artery
island flap for reconstruction of the lower and mid face.
Ann Plast Surg 1997;39:30--35.
""rrte pedicled transfer of a segment of forehead tis-
~ sue has its roots in ancient history; descriptions
of rudimentary forms of the technique are found as
early as 700 BC (1). It appeared in the Indian litera-
ture, where punishment for certain crimes was nasal
tip amputation, and forehead flaps were used as a
means for restoring the nasal tip. European descrip-
tions of forehead flaps appeared starting in the 15th
century, and the operation achieved popularity by the
mid-1800s. In the 1930s, Kazanjian and Roopenian
(6) identified that the supratrochlear and supraorbital
arteries were in fact the primary blood supply of the
fl.ap, and this led to the flap becoming a mainstream
tool in the armamentarium of the facial reconstructive
surgeon.
122
FLAP DESIGN AND UTILIZATION
The paramedian forehead flap is most often designed
with the base centered over one of the supratrochlear
arteries (3,5,9,10). By capturing the dominant arterial
pedicle, it is possible to raise and transfer a large amount
of skin and subcutaneous tissue, on a pedicle as narrow
as 1.2 em (Fig. 8-1). Burget and Menick descnoed an
extension of the incisions for the pedicle down below the
bony orbital rim, to provide additional flap length, and
this method is in common use today (2,7,8). A signifi-
cant benefit of extending the incisions below the bony
orbital rim is the avoidance of transferring haiJ:I.bearing
scalp into the nose or midface, where hair growth is

undesirable. The narrow pedicle also provides increased
freedom of transposition around the pivot point and
more effective arcs of rotation. In addition, the narrow
pedicle permits primary closure of at least the inferior
portion of the donor site.
Paramedian forehead flaps are designed from
templates of the nasal, orbital, or midfacial defect that
requires coverage. For any of these areas, a template
of the defect is created and then drawn onto the fore-
head just underneath the hair-bearing skin. Planned
incisions are extended inferiorly to the orbital rim in
order to capture the supratrochlear artery, which is eas-
ily identified with the assistance of a Doppler probe.
Dissection proceeds in a cephalocaudal direction. While
the supraperiosteal plane is appropriate for the superior
50% of the flap, below the midforehead, the plane of
dissection deepens to the subperiosteal layer in order
to safely capture the pedicle (Fig. 8-2). Because this
area is nearly always able to be closed primarily, there
is no portion of the cranium that is exposed without
periosteal coverage. When the superior aspect of the
incision cannot be closed primarily, maintenance of
the periostium provides a bed for either skin grafting or
healing by secondary intention. Aggressive thinning of
the flap is possible during the inset step, based upon the
superficial location of the artery in the distal portion of
the flap (9).This operation is performed as a two-staged
procedure in nasal reconstruction, where the pedicle
is divided in the second stage, usually after a matura-
tion period of 3 weeks (Fig. 8-3). However, single-stage
paramedian forehead flap reconstruction is possible for
isolated midfacial defects as well as for orbital socket
reconstruction.
Over the past decade, Burget and Walton (4) have
popularized the paramedian forehead flap in conjunc-
tion with a free radial forearm free flap and free struc-
tural grafts (cartilage and calvarial bone) for total
nasal reconstruction (Figs. 8-4-8-6). Their results,
combining pedicled and free tissue in carefully exe-
cuted, staged reconstructive efforts, have revolution-
ized the way reconstructive surgeons think about the
total rhinectomy defect and about the functional and
esthetic results that are achievable. They describe
harvesting a radial forearm free flap containing multi-
ple skin islands, each for a separate component of the
internal reconstruction (Fig. 8-4). The skin islands
can then be rotated independently from one another
for reconstruction of the columella, nasal lining, nasal
floor and the lip. Free cartilage or bone grafts are then
fashioned to provide the necessary structural elements
of the nasal skeleton, and a paramedian forehead
flap is raised to cover the structural elements, sand-
wiching them between two well-vascularized tissue
layers. They emphasize the need for secondary pro-
cedures to achieve the desired functional and esthetic
results.
PARAMEDIAN FOREHEAD 123
NEUROVASCULAR ANATOMY
The anatomy of the paramedian forehead flap was best
described by Mangold, McCarthy, and Shumrick in the
1980s (3,5,9,10). They demonstrated that the forehead
skin is supplied by a terminal branch of the angular
artery, as well as the supratrochlear, supraorbital, and
superficial temporal arteries. While each of these vessels
provides a primary blood supply to a single region of the
forehead, all demonstrate significant interconnecting
anastomoses. Based upon this rich overlap of arterial
input, the surrounding skin is not at risk for necrosis
following the removal of the supratrochlear supply to
the region. Some experimental evidence suggests that
the vascular supply to the forehead is so rich that even
in the absence of the supratrochlear artery within the
pedicle, the forehead flap may still receive ample arte-
rial input from small tributaries in the pedicle to survive
elevation and transfer into a recipient bed (3).
The supratrochlear artery exits the superomedial
orbit approximately 2 em lateral to the midline, trav-
els for approximately 2 em vertically, and then begins
significant branching. The branching pattern includes
multiple branches across the midline, to anastomose
with the contralateral supratrochlear artery. As the
artery travels superiorly, it traverses planes and becomes
increasingly superficial until reaching the subdermal
plane at the hairline. This important fact permits the
extensive thinning of the flap at the inset stage, without
compromising distal flap vascularity (9).
ANATOMIC VARIATIONS
While the supratrochlear artery has been shown to
provide the dominant axial blood supply to the median
and paramedian forehead tissues, the rich anastomotic
network in the medial canthal region may permit sur-
vival of this flap even in the absence of the dominant
artery (3). Thus, even in situations where the arterial
anatomy differs from the normally expected location,
this flap remains a reliable reconstructive option. In such
circumstances, a delay procedure may be prudent to
enhance the vascular supply to the distal tip of the flap.
POTENTIAL PITFALLS
Several drawbacks to the paramedian forehead flap
must be recognized before it is selected as the recon-
structive modality in nasal, midfacial, and orbital socket
reconstruction. The first drawback is the requirement
for a second-stage surgical procedure for pedicle divi-
sion. It has been well established that the pedicle can
be divided as soon as 10 days following tissue trans-
fer, though most surgeons prefer to wait a full 3 weeks
(JUt conr.inwd on pa.g~ 129)
124 CHAPTERI

Paramedian Forehead Flap

FIGURE 8-1. A. B: The traditional paramedian

forehead flap is based upon the supratrochlear
vessels. There are adjacent supraorbital and
infratrochlear vessels. Capture of the vascular
pedicle requires a 1.2 em cutaneous base at
the orbital rim, centered on the supratrochlear
vessels. The distal cutaneous paddle can be
designed to repair a wide range of defects,
including the external lining for a total nasal
defect. The final flap design is based upon a
template reflecting the cutaneous nasal defect,
after the subunit principle has been applied. B

FIGURE 8-2. An appropriate pattern is

designed, followed by paramedian forehead
flap elevation, proceeding superiorly to
inferiorly. The superior aspect of the flap can
be elevated in either a supraperiosteal plane
or a subperiosteal plane and then debulked.
However, the inferior flap must be elevated in
a subperiosteal plane to protect the vascular
pedicle. The elevated flap is then rotated medi-
ally and inset.
 PARAMEDIAN FOREHEAD 125

Paramedian Forehead Flap

FIGURE 8-3. Final repair of 1he nasal defect

requires tension-free inset of the distal portion
of the flap.

Columella

Suprstrochlear
vessel

FIGURE 8-4. In total nasal reconstruction,

radial forearm tissue is used for internal lining
and can be customized depending upon the
defect. Multiple semi-independent skin paddles
may be transferred based upon the radial
artery.
126 CHAPTERI

Paramedian Forehead Flap

FIGURE 8-5. Various structural elements for

nasal reconstruction. Top: Rib (or conchal)
cartilage fashioned into relevant nasal skeletal
elements: nasal tip graft upper lateral cartilage
graft, paired lower lateral cartilages. Calvarial
bone graft fashioned into nasal dorsal shape.
Bonom: Harvested synchondrosis of the 7th
and 8th ribs, demonstrating ideal location for
cartilage harvest from chest wall.

Rib cartilage

/~~
u~ / -:.

FIGURE 8-6. Completed total nasal recon-

struction, employing free tissue for the inner
lining and columella, structural elements (not
shown), and paramedian forehead flap external
coverage. This trilayered approach optimizes
functional and esthetic outcome.
 PARAMEDIAN FOREHEAD 127

Calvarial Bone Graft

FIGURE 8-1. An outer calvarial bone graft is harvested and cantilevered over the dorsum to
provide structural support. Calvarial bone grafts are used when structural support is required
for nasal dorsal reconstruction.
128 CHAPTERI

Calvarial Bone Graft

FIGURE 8-8. A: The ideal donor site lies 1 em

medial to the superiortemporalline, as shown.
This site avoids the temporalis muscle, ensures B
adequate graft thickness, and possesses the
proper natural curvature to yield a natural
appearance to the reconstructed nasal dorsum.
B: Harvest ofthe outer calvarium is initiated by
drilling a circumferential trough that penetrates
the diploic layer of the calvarium without
violating the inner calvarium, using an otologic
cutting burr. Copious irrigation is required to
prevent thermal injury to the bone graft. The
outer edges of the drilled trough must be bev-
eled to permit introduction of the osteotome
parallel to the inner calvarium. C: An osteotome
is used to elevate the graft from the inner
calvarium. Meticulous elevation prevents graft
fracture. The donor site is managed by simple
contouring to prevent obvious step-offs. c

before pedicle division. In the intervening time between
flap inset and pedicle division, the wound is unsightly
and requires extensive patient counseling. A second pit-
fall, occurring when a large amount of cutaneous cover
is necessary, is failure of the wound donor site to close
with simple undermining and advancement techniques.
Options when this occurs include skin grafting or leav-
ing the wound to granulate on the bed of exposed peri-
osteum. While this method of handling the donor site
can yield a reasonable aesthetic result, it occasionally
results in a stellate scar, requiring revision via forehead
expansion. However, the inferior 50% of the incision
can almost always be closed primarily, so with appropri-
ate hairstyling techniques when possible, the superior
aspect may present less of an esthetic issue.
A third pitfall involves undesirable hair-bearing
skin on the forehead flap. If flap design requires the
incorporation of some hair-bearing skin, the hair will
continue to grow on the transferred cutaneous seg-
ment over the long term. This growth may be treated
with depilatory creams or electrolysis or with laser
hair removal. Hair removal requires multiple treat-
ments and is not effective for the villous hair that can
be found lower on the forehead. Villous hair is not
responsive to laser hair removal and must be consid-
ered in flap design.
POSTOPERATIVE CARE
The paramedian forehead flap is straightforward to har-
vest and can be performed under local anesthesia, intra-
venous sedation, or general anesthesia. Postoperative
care involves retaining moisture around the pedicle, so
that desiccation does not create thrombosis of the arterial
and venous structures. This is accomplished with the top-
ical use of bacitracin or loose wrapping of the pedicle in
petrolatum-impregnated gauze. The gauze is changed on
PARAMEDIAN FOREHEAD 129
a daily basis for the intervening 3 weeks, and pedicle divi-
sion may be accomplished in the office setting. Once the
paramedian forehead flap has completely healed, there
may be requirements for flap debulking and additional
contouring, which the transferred tissue tolerates easily.
REFERENCES
1. Baker SR: Interpolated paramedian forehead flaps. In:
Baker SR, ed. Local Flaps in Head and Neck Reconstruc-
tion. Philadelphia: Mosby; 2007:265-312.
2. Burget GC: Aesthetic restoration of the nose. Clin Plan:
Surg 1985;12:463-480.
3. Burget GC, Menick FJ: The subunit principle in nasal
reconstruction. Plan: Reromtr Surg 1985;76:239-247.
4. Burget GC, Walton RL: Optimal use of microvascular
free flaps, cartilage grafts, and a paramedian forehead
flap for aesthetic reconstruction of the nose and adjacent
facial units. Plast Reromtr Surg 2007;120:1171-1207;
discussion 1208-1116.
5. Kazanjian VH: The use of skin flaps in the repair of facial
deformities. Plast Reromtr Surg (1946) 1950;5:337-352.
6. Kazanjian VH, Roopenian A: Median forehead flaps in
the repair of defects of the nose and surrounding areas.
Tram Am Acad Opkthalmol Otolaryngo/1956;60:557-566.
7. Mangold U, Lierse W. pfeifer G: The arteries of the fore-
head as the basis of nasal reconstruction with forehead
flaps.ActaAnat (Basel) 1980;107:18-25.
8. McCarthy JG, Lorenc ZP, Cutting C, Rachesky M: The
median forehead flap revisited: the blood supply. Plan:
Reconm Surg 1985;76:866-869.
9. Menick FJ: Aesthetic refinements in use of forehead for
nasal reconstruction: the paramedian forehead flap. Clin
Plast Surg 1990; 17:607-622.
10. Shumrick KA, Smith TL: The anatomic basis for the
design of forehead flaps in nasal reconstruction. Arch
Otolaryngol Head Neck Surg 1992;118:373-379.
The reconstruction of mucosal defects of the oral cavity
following ttauma or ablative surgery can be a cllalleng-
ing problem. Primary closure, healing by secondary
intention, or the application of skin grafts are effective
techniques in most situations. It is tempting to borrow
mucosa from adjacent areas of the oral cavity, and a wide
variety of oral fiaps" have been described (3). In par-
ticular, the tongue has been subjected to assault as the
source of well-vascularized mucosa to resurface defects
&om the palate to the hypopharynx. However, when the
surgeon borrows tissue from the region that is being
reconstructed, it is imperative that a critical appraisal
be made of the potential deficits associated with bor-
rowing that tissue. The availability of a wide range of
alternative flaps from regional or distant sites makes it
generally unnecessary to use tissue from the tongue to
130
resurface the oral cavity or pharynx. The tongue is the
most critical structure in the oral cavity for postopera-
tive oral function. To interfere with its activity, in any
way, should be condemned.
However, this philosophy does not take away from
the desirability of using "like tissue,. to accomplish the
reconstruction. The transfer of well-vascularized, sen-
sate mucosa is particularly appealing. The palatal island
mucoperiosteal flap is an atttactive reconstructive option
for those reasons. Originally introduced by Millard (5)
in 1962, it was populari2:ed for use in ablative defects of
the posterior oral cavity by Gullane and Arena (1). The
laner authors expanded the utility of this flap by repon-
ing the safe transfer of vinuaUy the entire hard palate
mucoperiosteum on a single neurovascular pedicle (2).
Although the loss of mucosa from the palate creates a

donor-site defect of exposed bone, the healing of that
defect by secondary intention causes no functional mor-
bidity. The fact that the secondary defect overlies bone
ensures that it will heal without contraction.
The vascular pedicle of the palatal island mucoperi-
osteal flap is unique becauae it traverses a bony canal
and also because of ifll nondependent position in the
oral cavity, which eliminates the deuimental effects of
gravitational pull.
FLAP DESIGN AND UTILIZATION
Because of the small area of mucosa of the palate, there
is a limited range of flap designs. The island flap that was
reported by Millard (5) was harvested from one side of
the palate. Gullane and Arena (2) expanded the area of
transfer to include virtually the entire palatal mucoperi-
osteum, providing approximately 8 to 10 cm2 of tissue.
The flap island is created by incising the palate 1 em
medial to the teeth and 1.5 em anterior to the junction
of the hard and soft palate. The flap can be rotated 180
degrees for inserting it into defects of the retromolar
trigone and tonsillar fossa. To improve the arc of rota-
tion, the hook of the hamulus can be removed, thereby
providing an additional! em oflength.
The island palatal flap is ideal for resurfacing defects
of the hard and soft palate. Perhaps one of its great-
est applications in contemporary reconstructive algo-
rithms is the restoration of posterior palatal resections,
most commonly for minor salivary gland malignancies.
As long as the remaining palatal mucosa is larger than
the defect, this technique can be successfully applied
(see Figs. 9-12 and 9-13). While we uaually prefer to
have a two-layer closure with both an inner and outer
lining, the combination of the mucosal, submucosal,
and periosteal layers makes for a finn composite of tis-
sue that lends itself to a single-layer repair.
Previoualy used to restore defects of the retromo-
lar trigone, tonsil, and lateral pharyngeal wall, tumors
in these locations are now more often operated in the
salvage setting following radiation, which is a contrain-
dication to the use of the palatal flap for fear of osteo-
radionecrosis. When the situation arises that a defect in
these regions is created, then the palatal island flap is a
reasonable reconstructive choice. It is imperative that
early and frequent physical therapy be implemented to
avoid fibrosis and the development of trismus. The pala-
tal island flap is not as pliable as a cutaneous flap, and as
a result, it may form a firm band betWeen the mandible
and Jll.Uilla that restricts oral opening.
This donor site has been used to restore velopharyn-
geal competence in combination with a mucosal flap
from the posterior pharyngeal wall (6). I have used this
combination on several occasions for early-stage soft
palate cancers in an effort to avoid the need for a palatal
PALATAL ISLAND 131
obturator. Tilis flap is also useful in cleft palate repair
and for closure of oroanttal fistulas (2).
NEUROVASCULAR ANATOMY
A thorough understanding of the osteology of the palate is
crucial to raising the mucoperiosteal flap. The hard palate
is formed by the palatine processes of the maxillae and the
horizontal laminas of the palatine bones. There is a lon-
gitudinal suture that separates the palate in the midline
and a t:ransVerse suture that separates the maxillary shelf
from the palatine shelf posteriorly. The greater palatine
foramen is located in the lateral aspect of the t:ransverse
suture just opposite the second molar (Fig. 9-1). Postero-
lateral to the greater palatine foramen are the lesser pala-
tine foramina, of which there are UBUally two. The latter
foramina are located in the palatine bone and transmit
the lesser palatine artery and nerves. The hard palate is
covered by a mucosal layer, which is firmly adherent to
the periosteum. The periosteum is firmly attached to the
palatal bone through the fibrous pegs of Sharpey.
The blood supply to the palate is derived from the
descending palatine artery, which is a branch of the inter-
nal ma:xillary artery (Fig. 9-2). The descending palatine
artery gives offthe greater palatine branch, which emerges
through the greater palatine foramen with the greater pala-
tine nerve. The greater palatine artery runs forward on the
lateral aspect of the palate to supply the mucoperiosteum.
Greater
palatine a.
andn.
Greater
palatine
foramen
FIGURE 9-1. The greater palatine artery and nerve emerge
from the greater palatine canal through the greater palatine
foramen. The neurovascular pedicle runs forward on the
palate, and the artery then ascends through the incisive
canal to supply the nasal mucosa.
132 CHAPTER 9
Descendlng-~~~~~~~~~'~O~Im.
palatine a.
Greater
palatine a.
FIGURE 92. The greater palatine artery is a branch of the descending palatine artery, which in
turn, arises from the internal maxillary artery.
The descending palatine artery traverses the greater pala-
tine canal, which connects the pterygoma.illary fossa
with the hard palate. The lesser palatine artery, a branch
of the descending palatine artery, emerges through the
lesser palatine foramina to supply the soft palate. Addi-
tional blood supply to the palate comes through branches
of the ascending pharyngea, facial, and lingual arteries.
This collateral supply is primarily to the soft palate.
The vascular supply to the palatal island mucoperi-
osteal flap is the greater palatine artery and vein. Mter
rwming their posteroanterior course on the hard pal-
ate, these vessels ascend in the incisive canals to sup-
ply the nasal mucosa. The greater palatine vein drains
into the pterygoid plexus of veins. Despite the presence
of a midline longitudinal raphe that divides the palatal
mucosa in half, Gullane and Arena (1,2) demonstrated
that the entire palate could be supplied by one greater
palatine pedicle. They referred to the work reported by
Maher (4) in 1977, which showed an extensive arbori-
zation of the greater palatine vessels, which was termed
the "macronet." By arteriographic studies, Maher found
evidence of three vascular layers: mucosal, submucosal,
and periosteal. The arterial netWork crossed the mid-
line raphe to provide nutrient fl.ow through one pedicle
when the contralateral one was sacrificed.
POTENTIAL PITFALLS
Gullane and Arena (2) reported a 5% failure rate in a
series of 53 palatal flaps. They warned against the use of
this technique when any of the following three condi-
tions were present: (a) ligation of the external carotid or
internal ma:r.illary artery, (b) prior palatal surgery with
possible disruption of the greater palatine vessels, or (c)
prior radiation to the palate. The surgeon should also
be cautious about placing excess tension on the pala-
tal blood supply, which may be less forgiving than most
island flaps because of the course of the vessels through
a bony canal.
POSTOPERATIVE CARE
The exposed palatal bone is cleansed on a regular basis
with frequent oral irrigations. The ingrowth of mucosa
from the edges of the defect occurs fairly rapidly. It
has been our experience that the mucosal ingrowth
brings sensory nerve fibers, which reduces the donor-
site morbidity. The fact that the defect overlies bone
ensures that there is no scar contracture, which would
otherwise occur. Fabrication of a prosthesis to cover
the denuded palate may provide valuable pain relief
during the early postoperative period until mucosaliza-
tion has occurred.
Acknowledgments
The author acknowledges the contnbutions of
Dr. Hugh F. Biller to the writing of this chapter in the
first edition of this book.
 PALATAL ISLAND 133

Palatal Island Flap

FIGURE 9-3. The palatal island mucoperiosteal

flap is outlined with the approximate position of
the greater palatine neurovascular pedicles on
either side.

FIGURE 9-4. The mucoperiosteal flap is

elevated by sharp and blunt dissection, moving
in an anterior to posterior direction. The
mucosal layer is intimately associated with the
palatal periosteum.
134 CHAPTER!

Palatal Island Flap

FIGURE 9-5. The neurovascular arcade

(arrows) is visualized on the undersurface of
the periosteum. It is best to begin the dissectian
on the side opposite the pedicle that is to be
preserved.

FIGURE 9-6. The contralateral neurovascular

pedicle has been isolated and is now readyta
be transsected.
 PALATAL ISLAND 135

Palatal Island Flap

FIGURE 9-7. The mucoperiosteal flap is care-

fully elevated toward the nutrient neurovas-
cular pedicle. The fixed position of the vessels
exiting through the greater palatine foramen
provides little leeway in mobilizing the flap.
The contralateral pedicle has been ligated and
transsected.

FIGURE 9-8. The palatal flap is completely

isolated on its pedicle and can now be rotated
to resurface the mucosal defect.
13& CHAPTER!

Palatal Island Flap

FIGURE 9-9. A common use of the palatal flap

is to close defects of the tonsillar fossa and the
retromolar trigone. Early mobilization of the jaw
to prevent scar formation and resultant trismus
is very important.

FIGURE 9-10. The palatal island flap has been

rotated 180 degrees. Further mobilization can
be achieved by cutting the hook of the hamulus
and decompressing the posterior wall of the
greater palatine foramen.
 PALATAL ISLAND 137

Palatal Island Flap

~~-Flap
over
defect

FIGURE 9-11. The palatal flap is sutured into

the defect. It is important that undue tension
is not placed on the flap because the pedicle's
course through the greater palatine canal is
unforgiving.

Defect in
hard
palate

FIGURE 9-12. Use of the palatal island flap for

reconstruction of through-and-through oroan-
tral and oronasal defects created by resection
of palatal carcinomas is an ideal reconstruc-
tive technique. The surgeon must size up the
defect and the remaining palatal mucosa that is
available for reconstruction to ensure that there
is an adequate surface area of mucosa for the
size of the defect such that the flap can be
sutured to the palatal bone with some degree of
overlap.
138 CHAPTER!

Palatal Island Flap

FIGURE 9-13. Rotation of the flap is readily

accomplished to provide a complete repair of
the posterior defect. The leading anterior edge
of the flap is tacked to the palate at a distance
from the bony defect with the use of com-
mercially available implants that are placed
into holes that are drilled into the bone. These
implants have 8 suture with 8 needle attached
to it and provide an excellent method for secur-
ing the leading edge of the flap.

REFERENCES 4. Maher W: Distribution of palatal and other arteries in

1. Gullane P, Arena S: Palatal island flap for reconstruc-
tion of oral defects. Arch Otolaryngol Head Neck SUTg
1977;103:598.
cleft and non-deft human palates. Cleft Palau Cnmiojac
J 1977;14:1.
5. Millard DR: Wide: and/or short cleft palate:. Plast Reronm-
SU1't 1962;29:40.

2. Gullane P, ArenaS: Extended palatal island mucoperios-
teal flap. Arch Orolaryngol Head Neck SUTg 1985;111:330.
3. Komisar A. Lawson W: A compendium of intraoral flaps.
Head Neck 1985;8:91.
6. Millard DR. Seider H: The versatile palatal island flap: its
use in soft palate reconstruction and nasopharyngeal and
choanal atresia. BT J Plan SUTg 1977;30:300.
 rown et al. (3) are credited with being the first to
B use abdominal cutmeous flaps based on the perfo-
rators of the rectus abdominis muscle. However, Drev-
er's (7) report of the "epigastric island flap" was the first
to recognize the potential of transferring an island of
skin supplied by a segment of the underlying muscle.
He described a vertically oriented musculocutaneous
flap that was transferred to a defect of the chest wall
based on the deep superior epigastric vascular supply.
Pennington and Pelly (27) are credited with the first
report of transferring a free rectus abdominis muscu-
locutaneous flap based on the deep inferior epigastric
artery (DIBA) and the deep inferior epigastric vein
(DmV). These authors described the results of ink-
injection studies that demonstrated the rich vascularity
of the abdominal skin through the DIBA.
The rectus abdominis musculocutaneous flap has
assumed an important role in. head and neck reconstruc-
tion because of its ease of harvest, long vascular pedicle,
and tremendous reliability. Pedicle flaps, based on the
deep superior epigastric vascular supply to the rectus
abdominis, have been used extensively in. reconstruction
of the breast. Pedicled transposition flaps can also be
based on the deep inferior epigastric system for use in
reconstructing defects in. the groin. and upper thigh (20).
The DIEA and DIEV are much more useful for free
tissue transfers because of their greater diameter and
length and the larger skin territory that can be captured.
The rectus abdominis muscles occupy the paramed-
ian position of the anterior abdominal wall. Eac:b. mus-
cle spans the entire length of the abdomen, arising from
the pubis and inserting into the anteroinferior part of
141
142 CHAPTER 10

Linea alba

Tendinous e::;.-1-- - - --r::-r-.

inscriptions

Transversus
~~i~~lJ_ abdominis m.

FIGURE 10..1. The rectus abdominis arises from 1he pubis and runs the entire length of the
abdomen to insert on the 5th, 6th, and 7th costal cartilages and the xiphoid process. The
muscle is wider in its cephalad portion. Two to five tendinous inscriptions divide the muscle
transversely. These inscriptions are firmly adherent to the anterior, but not 1he posterior. rectus
sheath.

the thorax (Fig. 10-1). The primary action of the rec-
tus abdominis is to flex the trunk. The rectus abdominis
donor site is a useful source of vascularized muscle and
skin for a variety of ablative defects of the head and neck
(23). It offers several unique features compared with the
regional musculocutaneous flaps based on the pectoralis
major, trapezius, and latissimus dorsi. The area of skin
that can be reliably harvested with a single rectus muscle
encompasses a substantial portion of the abdomen and
lower chest. The size of the muscle component ranges
from the entire muscle to only a small portion in the
paraumbilical region where the dominant perforators
are located. The caudal portion of the muscle may be
trimmed to add length to the vascular pedicle. "Muscle
sparing" transverse rectus abdominus musculocutane-
ous (TRAM) flaps havt! been described that preservt!
the medial and lateral portions of the rectus muscle
in situ and harvest only the central portion of the muscle
with the cutaneous perforators. It is imperativt! that the
remaining segments of muscle maintain their vascular
and nerve supply in order to provide a meaningful advan-
tage for the function and integrity of the abdominal wall.
This "alternate" neurovascular supply arises laterally,
where the intercostal nerves and vessels course through
the layer between the internal oblique and the transversus
abdominis muscles. In addition to harvest of the central
muscle, perforawr flaps based on paraumbilical perfora-
tors have been described for use in breast reconstruction
{18). However, one of the great advantages of the rectus
abdominis muscle in head and neck reconstruction is the
ability to transfer large amounts ofwell-vascularized mus-
cle in addition to the large surface area of skin to solvt!
challenging reconstructive problems. Vascularized mus-
cle is invaluable for managing problem wounds that are
radiated and/or where separation of ccwities is required.
The thickness of the subcutaneous tissue varies from
being very thick in the lower abdomen to rather thin in
the region above the costal margin. Thinning of this flap
by excision of subcutaneous tissue can provide a thinner
flap, but it must be performed with caution to preserve
the blood supply to the skin. It is imperativt! to avoid
injury to the very small vessels ttavt!rsing the subcuta-
neous tissue, which may compromise the vascularity to
larger flaps, more so than smaller flaps. The rich vascu-
larity of the skin territory permits a greater flexibility in
the flap design, leading to more accurate contouring to
the surgical defects. Finally, the ability to harvt!st this
flap with the patient in the supine position greatly facili-
tates the use of a two-team approach.
FLAP DESIGN AND UTILIZATION
The rectus abdominis muscle may be transferred alone,
with overlying fascia and subcutaneous tissue, or as a
composite flap consisting of muscle, fascia, skin, and
RECTUS ABDOMINIS 143
subcutaneous tissue. Dye-injection studies by Boyd et
al. (2) showed vascularity to the 6th rib and, therefore,
introduced the possibility of incorporating bone in this
flap. The 6th through the 1Oth costal cartilages can be
harvested with the flap to provide bone-cartilage with a
length of up to 10 em and a width of 2 em. The segmen-
tal nerve supply to the muscle provides the potential for
a dynamic reconstruction, and we have successfully used
this flap for facial reanimation (36). Although there are
no reported cases of sensate rectus abdominis flaps, this
potential exists through the mixed motor-sensory nerves.
There are a multitude of flap designs that havt! been
reported that permit the contouring of this flap to vir-
tually any defect in the head and neck. The patient,s
body habitus may be a limiting factor with regard to
excess thickness of the subcutaneous tissue component.
However, the muscle alone may be transferred and then
resurfaced with a split-thickness skin graft. The skin of
a significant portion of the abdomen may be reliably
transferred because of the network of subcutaneous
vessels emanating from the musculocutaneous perfo-
rators in the paraumbilical region. These perforators
are located in a zone that extends from 2 em above to
3 em below the umbilicus. Boyd et al. (2) speculated
that the deep inferior epigastric flap may provide the
largest potential territory of vascularized skin of any
donor site in the body (Fig. 10-2).
There are many factors that enter into the decision
with regard to flap design. The defect's size and volume
play a significant role, along with its proximity to the
recipient vessels. Lengthening of the donor vascular
pedicle may be achieved by placing the skin paddle in
a more cephalad position on the abdominal wall. The
entire rectus abdominis may be transferred if needed.
As noted above, preservation of a portion of that muscle
at the donor site adds little to maintain the integrity of
the abdominal wall to prevent ventral herniation, unless
it has a good blood and nerve supply. As with any mus-
culocutaneous flap, the muscle component provides lit-
tle long-term bulk as a result of denervation atrophy.
The degree of atrophy may be diminished by reestab-
lishing a motor input through repair of the flap motor
nerves to suitable recipient motor nerves in the neck.
A more accurate method to provide bulk for contour is
to transfer vascularized subcutaneous tissue.
The most commonly used design for this donor site
is the TRAM flap (29). Popularized for use in breast
reconstruction, this design incorporates skin from the
entire lower abdomen. Four different skin zones have
been identified. Zone 1 refers to the skin overlying the
ipsilateral rectus muscle. Zone 2 denotes the skin of
the contralateral lower abdomen overlying the oppo-
site rectus muscle. The skin territory on the ipsilateral
side of the abdomen lateral to the linea semilunaris is
referred to as zone 3, and the skin lateral to the opposite
linea semilunaris is zone 4. The blood supply to zone
144 CHAPTER 10

I
Deep superior

~
epigastric artery

c
I
DCIA
(perforators) ' ~I ..)

Deep inferior
epigastric artery

SCIA
~~ ~

FIGURE 10-2. The versatility of the deep inferior epigastric system of flaps is largely the result of the periumbilical perfora-
tors, which send branches in all directions on the anterior abdominal wall like the spokes on a wheel. The DSEA and the
DIEA communicate through a system of choke vessels. Other vascular systems that contribute to the blood supply of the
abdominal wall are the DCIA and SCIA. Not shown in this illustration are the superficial inferior epigastric pedicle and the
superficial epigastric artery. An array of different flap designs are shown that may be combined to meet the needs of the
particular defect. A:. Transverse skin paddle placed below or above the umbilicus. It is imperative to capture the dominant
periumbilical perforators. B: The extended OlEA flap may be transferred with the entire rectus muscle. C: The extended DIEA
flap may be 1ransferred with only a small cuff of muscle in the region of the umbilicus. D: A longitudinal skin paddle oriented
over the entire length of the muscle provides a rich vascularity to the skin but a thicker flap due to the large muscle compo-
nent. E: Athinner flap can be achieved by harvesting the rectus abdominis with the anterior rectus sheath above the arcuate
line. F: A deep inferior epigastric perforator (DIEP) flap that contains skin and subcutaneous fat is harvested based upon
musculocutaneous perforators while preserving in situ the anterior rectus sheath and rectus abdominis muscle.

4 is the most tenuous. Investigations of the vascular sup-
ply to the TRAM flap reveal vessels that arise from one
rectus abdominis and cross the midline to supply the
skin of zone 3. The examination of these crossing ves-
sels and contrast-injection studies have confirmed the
poor blood supply to zone 4 (17). However, Takayanagi
and Ohtsuka {30) reported a technique that augmented
the vascular supply in the zone 4 skin when its viability
was deemed critical to the success of the reconstruc-
tion. They anastomosed the superficial epigastric or
the superficial circumflex iliac pedicle to enhance the
vascular supply to that region.
A vertically oriented skin paddle that overlies the
entire length of the rectus muscle may be harvested
from the pubis to the xiphoid. This design is reliable
but has the disadvantage of the additional buJk of the
muscle. An alternative flap design {referred to as the
thoracoumbilical flap or the extended deep inferior epi-
gastric flap) crosses the abdomen in an oblique fashion,
extending from the infraumbilical region to above the
ipsilateral costal margin (Fig. 10-2). The advantage of
this skin design is that it introduces a range of tissue
thickness that is at its greatest in the portion overlying
the muscle and least in the region that extends above
the costal margin. More accurate flap contouring may
be achieved by trimming the muscle component to
only that portion in the periumbilical region with the
greatest concentration of dominant musculocutaneous
perforators {32).
In patients with excessive amounts of subcutaneous
tissue in the anterior abdominal wall, a thinner flap may
be harvested by skin grafting the muscle or just using the
anterior rectus sheath above the arcuate line. We have
used the latter technique in reconstructing through-
and-through defects of the cheek. In such cases, the
external skin is restored with a skin paddle from the
periumbilical region, and the inner lining is resurfaced
with the anterior sheath overlying the cephalad por-
tion of the muscle. The muscle is folded upon itself to
achieve inner and outer epithelial surfaces. As noted
previously, the anterior sheath from the region above
the arcuate line can be harvested without fear of ventral
herniation. The anterior sheath provides a thin, tough
layer to achieve a watertight seal. We have also used this
technique in palatal reconstruction in which the rectus
sheath provides a suitable lining for the oral cavity fol-
lowing ablative procedures of the sinuses, nasopharynx,
or skull base. Chicarilli and Davey (6) used the rectus
abdominis flap to reconstruct a cranio-orbitomaxillary
defect. The anterior rectus sheath was sutured to the
bony margins of the cranial defect above the orbit to
serve as a hammock to support the intracranial contents.
An alternative technique to harvest a thinner flap
from this donor site was described by Koshima et al.
(18). They reported the transfer of "thinned, paraum-
bilical" flaps based solely on the musculocutaneous
RECTUS ABDOMINIS 145
perforators. The dominant perforators were dissected
through the muscle to the DIEA and DIEV. Branches to
the muscle were ligated, and no muscle was transferred
with the skin. The advantages of this flap included not
only its thinness but also the fact that the integrity of
the abdominal wall musculature was not disturbed.
The authors warned, however, that the dissection
through the muscle may be technically difficult. Along
similar lines, Akizuki et al. {1) described harvesting the
"extremely thinned" rectus abdominis free flaps. The
basic design was an extended deep inferior epigastric
flap based on a small segment of muscle in the peri-
umbilical region. The portion of the flap that extended
lateral to the muscle was thinned by removing all fatty
tissue deep to Scarpa's fascia, thereby preserving the
vascularity through the subdermal plexus. This tech-
nique is an important contribution because it provides
a method to utilize the rectus abdominis donor site in
individuals who might otherwise not be considered can-
didates because of an unfavorable body habitus.
Virtually, any combination of these flap designs may
be used. Two separate skin paddles may be oriented
over the longitudinal axis of the muscle to reconstruct
composite defects requiring inner and outer lining.
Although primary closure of the abdominal wall skin
is highly desirable, the application of a skin graft may
be performed and, certainly, represents a better option
when wound tension and respiratory compromise
become an issue.
We have used this donor site for a variety of head and
neck defects. It serves as an alternative source of skin for
external coverage when regional flaps are not available
or are unsuitable because of the defect's size or distance
from the donor site. Although the radial forearm and
anterolateral thigh flaps remain the workhorse flaps for
oral cavity reconstruction, the rectus abdominis is use-
ful to supply buJk following total glossectomy. The goal
in total tongue reconstruction is to supply sufficient soft
tissue height for an approximation of the neotongue
to the palate without the use of a palatal augmenta-
tion prosthesis. The rectus abdominis flap is an excel-
lent choice as a result of the tendinous inscriptions of
the anterior rectus sheath, which can be sutured to the
mandible to form a platform for the overlying skin pad-
dle. The tissue above this platform does not atrophy,
and, therefore, the shape and volume of the neotongue
can be precisely contoured (36).
Perhaps the greatest use of this flap has been in skull
base reconstruction; it has become the flap of choice
for many skull base defects requiring free tissue transfer
(35). Jones et al. {16) reported the use of this flap for
defects involving the middle and posterior cranial fossa;
Yamada et al. (37) described its application to defects
of the anterior cranial fossa. In the latter report, the free
rectus abdominis flap was used in patients who had
undergone prior surgery or radiation to help prevent a
146 CHAPTER 10
cerebrospinal fluid leak, to prevent ascending infection,
and to provide vascularity to free bone grafts used in
the periorbital region. I have used the rectus abdominis
free flap in the anterior cranial fossa in a patient who
had a postoperative collection of pus following resec-
tion of a recurrent frontal meningioma. The free rectus
muscle successfully achieved a functional separation of
the nasal and anterior cranial cavities, with complete
resolution of the infectious process (35). I normally do
not like to place a microvascular free flap in a grossly
infected field because of the extremely detrimental
effect that infection has on the microvascular pedicle.
However, this was a unique circumstance, and the abil-
ity of the rectus flap to survive in this environment is a
testimony to its hardiness. It also highlights the impor-
tance of vascularized muscle in the management of
"problem wounds," which are common occurrences in
the head and neck.
The versatility of flap design of the rectus abdominis
flap is extremely useful when trying to achieve a water-
tight seal of the various cavities that are opened follow-
ing many skull base procedures (10). This is particularly
challenging in the "three-cavity defect," which involves
the nasal, oral, and intracranial cavities in which multi-
ple epithelial surfaces are required for a successful out-
come. De-epithelialized portions of the flap can be used
to enhance the contour of regions such as the infratem-
poral fossa or the orbit following exenteration. Finally,
the extremely reliable nature of this flap is a critical fac-
tor in ensuring protective coverage to exposed portions
of the brain (35).
Many different free flaps have been used to recon-
struct defects of the scalp. Miyamoto et al. (25) reported
four cases of extensive scalp reconstructions using the
rectus abdominis free flap. Aside from the large surface
area, the length of the vascular pedicle was noted to be
a particular advantage for this donor site. The vascu-
lar pedicle can be lengthened by judicious skin paddle
placement and careful removal of the caudal portion of
the rectus muscle. These techniques help to avoid the
use of vein grafts.
NEUROVASCULAR ANATOMY
According to the classification system of Mathes and
Nahai (22), the rectus abdominis is a type III muscle
with two dominant vascular pedicles: the deep superior
epigastric artery (DSEA) and deep superior epigastric
vein (DSEV) and the DIEA and DIEV. The DSEA is a
continuation of the internal mammary artery; the DIEA
is a branch of the external iliac artery arising directly
opposite the deep circumflex iliac artery (Fig. 10-3). The
DSEA and DIEA pedicles arborize as they approach
each other in a longitudinal direction on the undersur-
face of the muscle. The two systems connect above the
umbilicus through a system of small-caliber vessels that
Taylor and Palmer (34) referred to as "choke" vessels.
Through cadaveric studies, the degree of arboriza-
tion of the DIEA and the DSEA has been classified into
three different types. In the type 1 pattern, the DIEA
does not divide, remaining a single vessel as it runs its
course on the undersurface of the muscle (29%). The
type 2 pattern refers to a DIEA that divides into two
dominant branches (57%). The type 3 pattern is a tri-
furcation of the DIEA (14%). The extent of branching
of the inferior system is mirrored by the DSEA (25).
The division of the vascular pedicle into two or three
branches is the basis for the "split muscle" transfer that
was reported by Sadove and Merrell (28).
The DIEA, measuring an average of 3 to 4 mm, is
roughly twice the diameter of the DSEA. The venous
supply of the inferior muscle is composed of paired
venae comitantes, which usually join to form a single
venous pedicle prior to their junction with the exter-
nal iliac vein. The DIEV is approximately 3.5 mm in
diameter. Extensive studies of the venous circulation
of the TRAM flap revealed a superficial and deep sys-
tem. The veins of the superficial system were above
Scarpa's fascia and communicated extensively across
the midline. The superficial veins drained into the deep
inferior epigastric system by way of the veins accompa-
nying the musculocutaneous arterial perforators. Valves
located in the connecting veins regulated the direction
of flow from the superficial toward the deep system. The
findings from this study confirmed the safety of thin-
ning the rectus abdominis musculocutaneous flaps by
removing fat from below Scarpa's fascia, as long as the
musculocutaneous perforators were preserved (5).
In addition to the size of the vascular pedicle, there
are a variety of compelling reasons why the inferior
pedicle is a better supply for free tissue transfer than
is the superior pedicle. The musculocutaneous perfora-
tors are direct branches of the DIEA and DIEV and
are therefore capable of supplying a much larger ter-
ritory of skin. Although the deep superior epigastric
system is capable of capturing these perforators, it does
so through a reversal of flow in the DIEA and DIEV,
across the choke system of vessels that connect the two
systems. Boyd et al. (2) studied the distribution of mus-
culocutaneous perforators exiting through the anterior
rectus sheath. By dividing the length of the muscle into
horizontal segments, they found that the dominant per-
forators were located in a zone close to the umbilicus,
with very few perforators arising in the most caudal or
cephalad portions of the muscle. These perforators were
also mapped according to a longitudinal division of the
muscle into thirds. The greatest concentration of large
perforators traversing the anterior sheath was located in
the middle and medial zones, with few exiting laterally.
The inferior pedicle runs an extraperitoneal course
in close proximity to the deep inguinal ring. It crosses
 RECTUS ABOOMINIS 147

Internal
mammary a.--:-----__,~~-:-~--11

.- + - - - Rectus
abdominis m.
(reflected)

----"l!~+f-"""+~;.._--- Posterior
Musculocutaneous -----~
rectus sheath
perforators

~~t-~----- Segmental
nerve supply

Arcuate line

.,._'1:~~~~~~-Segmental
nerve supply

line

FIGURE 10-3. The rectus abdominis has a type Ill vascular supply with two dominant pedicles, the DSEA and DSEV and
the OlEA and DIEV. These two systems communicate through a rich system of anastomoses located approximately halfway
between the xiphoid and 1he umbilicus. Musculocutaneous perforators arise in the paraumbilical region and send an array
of dominant branches that are oriented toward the inferior border of the scapula. The segmental nerve supply to 1he rectus
abdominis arises from the terminal branches of the lower six intercostal nerves, which run across the abdominal wall from
lateral to medial in the layer between the transversus abdominis and the internal oblique muscles. These nerves penetrate
the posterior rectus shea1h approximately 3 em medial to the linea semilunaris.
148 CHAPTER 10
the lateral border of the rectus muscle and pierces the
transversalis fascia approximately 3 to 4 em caudal to
the arcuate line. The DIEA gives off a number of smaller
branches to the pubis and the caudal aspect of the rec-
tus muscle prior to entering the rectus sheath.
Additional cadaveric studies by Taylor et al. (31)
demonstrated the rich connections between different
arterial supplies to the anterior abdominal wall skin.
The periumbilical musculocutaneous perforators of
the DIEA give off a series of radiating branches that
anastomose with the cutaneous branches of the fol-
lowing arteries: the superficial superior epigastric, the
intercostal, the deep and superficial circumflex iliac, the
superficial inferior epigastric (SIEA), and the puden-
dal. The dominant connections between these sys-
tems were found to occur within the subdermal plexus
(Fig. 10-2). As previously noted, there are also anasto-
motic connections in the intermuscular layer running
between the transversus abdominis and internal oblique
muscles. The primary connections occurring in this
layer are between the branches of the epigastric system
and the lower six intercostal vessels.
Taylor et al. {33) used the angiosome concept of
defining the vascular territories of the abdominal wall
to describe a technique of surgical delay to enlarge the
territory of skin that can be safely transferred on the
DIEA and DIEV. The angiosome theory proposes that
the region of skin that is most reliably harvested on the
deep inferior epigastric vascular system is defined by a
line that marks the interface between the DIEA and the
other source arteries of the abdominal wall. This line of
demarcation denotes the system of choke vessels that
connect two adjacent angiosomes. Thus, the system of
choke vessels between the DIEA and the DSEA is con-
sistently located above the umbilicus. The interface zone
with the contralateral DIEA is along the linea alba. In a
two-stage procedure, surgical delay can be achieved by
ligating the source artery in the adjacent territory, which
varies depending on the orientation of the desired skin
flap. This delay procedure causes a dilation of the choke
vessels connecting the two adjacent territories leading to
a more favorable hemodynamic gradient across the two
systems. Although the surgeon can usually capture the
skin in an adjacent angiosome without surgical delay,
it becomes progressively more difficult when more
than one system of choke vessels is traversed in series.
If a longer, vertically oriented flap is desired, then the
appropriate delay procedure would involve interruption
of the deep superior epigastric pedicle. Enhancement
of the vascularity to the skin of the TRAM flap can be
achieved by interruption ofthe other vascular supply to
the skin of the lower abdomen. On the ipsilateral side,
the delay procedure would involve ligation of the SIEA.
Improving the vascular supply to the skin of zones
2 and 4 can be achieved by interrupting the contralat-
eral DIEA and SIEA. Although the area of skin that can
be reliably harvested on a single DIEA is so great that
a delay phenomenon is rarely needed in head and neck
reconstruction, there are certain complex defects in
which this technique may be helpful.
The system of branches of the periumbilical perfo-
rators has the appearance of the radiating spokes of a
wheel with the hub located at the umbilicus, thus giv-
ing credence to the clinical observation that incorpora-
tion of the periumbilical perforators permits a skin flap
to be harvested with virtually any orientation from the
midline. However, the dominant orientation of these
branches is 45 degrees from the horizontal toward the
inferior scapular border (32). This explains the extreme
reliability of the oblique flap design that was previously
referred to as the extended deep inferior epigastric or
the thoracoumbilical flap. Taylor et al. {32) reported
that this design may safely incorporate skin above the
costal margin as far lateral as the midaxillary line.
The nerve supply to the rectus abdominis is derived
from the lower six intercostal nerves, which traverse the
plane between the transversus abdominis and the inter-
nal oblique muscles. These nerves are mixed motor and
sensory nerves, providing a segmental innervation to
the rectus muscle and sensory supply to the overlying
skin. The intercostal nerves enter the midportion of the
muscle on its posterior surface (9). By stimulating one
of the segmental nerves, it is possible to select a portion
of the rectus muscle for use in facial reanimation. We
have done this successfully in one patient who under-
went a composite reconstruction of the cheek follow-
ing radical parotidectomy. Good dynamic activity was
obtained by anastomosing two segmental nerves to the
upper and lower divisions of the ipsilateral facial nerve
(36). Hata et al. (15) reported successful facial reani-
mation in two patients with chronic facial paralysis.
A cross-facial nerve graft was initially placed and then
followed at 1 year with a free rectus abdominis transfer
in which several of the segmental nerves were anasto-
mosed to the cross-facial graft. The rectus abdominis
was considered a good muscle for facial reanimation
because of the ease of harvest and the length of the neu-
rovascular pedicle. A particular advantage that is unique
to the rectus abdominis is the tendinous inscriptions,
which allow placement of anchoring sutures.
ANATOMY OF THE RECTUS SHEATH
An understanding of the anatomy of the fascial enve-
lope of the rectus abdominis is perhaps more critical
than with any other flap. The prevention of herniation
depends on restoring the integrity of the abdominal wall
by effectively closing the fascial layers.
The aponeurotic extensions of the three muscles of
the anterior abdominal wall merge to form the anterior
and posterior sheaths of the rectus fascia (Fig. 10-4).
 RECTUS ABOOMINIS 149

Rectus abdominis m. Linea alba Anterior rectus sheath

External
oblique m.

Internal
oblique m.
Posterior Transversalis
~~~--- Transversus
rectus sheath fascia
abdominis m.

:ztj~~~~r:::::~~~~~~~~~~~~~$~~=----
~
Anterior
rectus
sheath

8 Transversalis fascia
Fat

FIGURE 11).4. Transverse sections through the anterior abdominal wall reveal the fascial
anatomy of the anterior and posterior rectus sheaths at two different levels in the abdomen.
A:. Above the arcuate line, the posterior sheath is composed of contributions from the aponeu-
roses of the transversus abdominis and the internal oblique muscles. The aponeurosis of the
internal oblique muscle splits to form part of the anterior rectus sheath with the external oblique
aponeurosis. B: Below the arcuate line, at about the level of the anterior superior iliac spine, the
aponeurotic extensions of all three muscle layers contribute to the anterior rectus sheath. The
posterior sheath is composed only of the transversalis fascia.

The compositions of these sheaths vary in different loca-
tions between the pubis and the :xiphoid. Above the cos-
tal margin, there is no posterior sheath, and the anterior
sheath is formed by an extension of the external oblique
aponeurosis. In the upper two thirds of the muscle, the
anterior sheath is formed by the external oblique plus a
contribution from the internal oblique aponeurosis. The
internal oblique aponeurosis also contributes to the pos-
terior sheath where it joins with the aponeurosis of the
transversus abdominis. An important transition occurs
in the posterior sheath at the arcuate line (semicircular
line or arch qj Douglas), which is approximately at the
level of the anterosuperior iliac spine. From this point to
the pubis, the posterior sheath is composed only of the
transversalis fascia. The strength of the posterior rectus
sheath above the arcuate line is sufficient to prevent an
abdominal bulge or herniation. Below the arcuate line,
these sequelae undoubtedly occur if the transversalis
fascia is not augmented. It is rarely necessary to design
a flap that requires harvest of the anterior rectus sheath
below the arcuate line. Maintaining the integrity of the
blood supply to the skin requires only that the anterior
rectus sheath is harvested in the paraumbilical region
where the dominant perforators are located. Although
it is probably not essential to do so above the arcuate
line, we routinely augment the posterior sheath by clos-
ing cuB's of the anterior sheath that are preserved both
medially and laterally. Taylor et al. (31) described a fa5-
cial sparing technique whereby cuts in the anterior reo-
tus sheath are made based on direct visualization of the
dominant perforators. The amount of anterior sheath
that is harvested may be minimized by this teclmique.
Augmentation of the anterior rectus sheath may also be
achieved with a synthetic mesh that adds greatly to the
strength of the repair.
Three other terms related to the fascia must be
defined (Fig. 10-1). The linea alba is the midline fas-
cial condensation that divides the two rectus muscles.
The linea semilunaris refers to the fascial condensation
that marks the lateral extent of each rectus muscle.
150 CHAPTER 10
The rectus abdominis is subdivided by two to five tendi-
nous inscriptions. The anterior sheath but not the poste-
rior sheath is firmly adherent to each inscription. These
fascial condensations do not extend to the posterior
sheath. Moon and Taylor (26) reported that 93.5% of a
series of 108 muscles had three tendinous inscriptions
with the most caudal one at the level of the umbilicus.
ANATOMIC VARIATIONS
In a series of 25 cadaver dissections reported by Boyd
et al. (2), the average diameter of the DIEA was 3.4 nun.
The vessel was slightly larger in instances in which the
DIEA was the source of an abnormal obturator artery.
The authors reported that the omv entered the exter-
nal iliac vein as a single trunk in 68% of cases and as a
double trunk in 32%. In a series of 115 cadaver dissec-
tions, Milloy et al. (24) reported no cases ofabsent DIEA
and DffiV and only three cases in which the DSEA and
DSEV could not be identified. In the vast majority of
cases, the DIEA-DffiV pedicle runs its usual course
along the deep surface of the rectus abdominis. We have
encountered two cases in which the pedicle tracked for
an unusual distance along the lateral aspect of the mus-
cle before taking a medial intramuscular route. Another
anomalous course of the deep inferior epigastric pedicle
has been described in which it winds around the medial
aspect of the muscle and does not give feeders to the
muscle until assuming a position along its superficial
surface. In this particular instance, the perforators to
the skin run directly through the anterior rectus sheath
without traversing the muscle (11).
POTENTIAL PITFALLS
The reliability of the rectus abdominis free flap is
reflected by the success rate of 93% in a large series of
cases in which this flap was used throughout the body
(23). In a review of all reported cases used for head
and neck reconstruction, we found only one failure in
73 free flaps transferred to the head and neck (36).
The preoperative assessment must include a careful
history and examination of the abdomen to be certain
that prior surgery will not interfere with flap harvest.
Most intraperitoneal procedures involve a longitudinal
incision through the linea alba, even though the skin
incision is transverse. A right subcostal incision for an
open cholecystectomy does not preclude the use of the
rectus muscle. This procedure almost invariably inter-
rupts the nerve supply to the cephalad portion of the
muscle. The surgeon should be aware of the potential
for denervation atrophy in the postcholecystectomy
patient. However, this does not interfere with the vascu-
larity of the atrophic muscle or its suitability for transfer.
As noted above, we have found that the DIEA and
omv run a longitudinal course along the lateral aspect
of the muscle prior to arborizing on the undersurface of
the muscle. Extreme care must be taken when making
the cuts in the anterior rectus sheath medial to the linea
semilunaris.
Although the requirements of the recipient defect
usually dictate the size and design of the abdominal skin
paddle, there are situations in which the surgeon has
options for planning the incisions and the approach to
the rectus muscle. Taylor et al. (31) and Hallock (12)
advocated a transverse suprapubic incision for the expo-
sure of the vascular pedicle. The resulting scar is well
camouflaged. Alternatively, Hallock (12) described the
use of an abdominoplasty approach but warned that
proper case selection was imperative.
Removal of one rectus abdominis muscle along with
a portion of the overlying fascia creates a potential
weakness in the anterior abdominal wall that may pre-
dispose the patient to ventral herniation or a midline
bulge. A preexisting hernia or diastasis recti may com-
plicate donor-site closure and mitigate against the use
of this flap. Taylor et al. (31) warned that divarication of
the recti must be recognized preoperatively to account
for the more lateralized position of the rectus muscles in
designing the flap.
The rectus abdominis muscles assist in flexion of
the torso and also provide static support to the anterior
abdominal wall. There is a tremendous amount of con-
troversy regarding the optimum method for closure of
the abdominal wall defect following rectus abdominis
flap harvest. Much of the data on donor-site complica-
tions are derived from large series of pedicled flaps that
were used for breast reconstruction. For the purposes of
this discussion, we will not cover all of the issues regard-
ing closure of the abdominal wall following harvest of rec-
tus abdominis muscles. There are two major camps that
are divided over the necessity of introducing a synthetic
mesh for closure of the anterior rectus sheath. Drever
and Hodson-Walker (8) described the technique of
placing a mesh of the exact dimensions to the area of
the anterior rectus sheath that was removed with the
flap. Using this method in 87 cases, they reported no
cases of ventral hernias and only 2 cases of abdominal
wall bulging. In a comparable group of 31 patients who
were closed primarily without a mesh, there was a 43%
incidence of bulging or hernias. These authors argued
that the mesh maintained the position of the remain-
ing abdominal wall musculature and did not cause an
increased resting tone resulting from direct approxima-
tion of the linea alba to the linea semilunaris. Lejour and
Dome (19) reported using a 4-cm wide double-layered
synthetic mesh between the posterior sheath and the
direct closure of the anterior sheath. They reported no
hernias or bulging in their series of unilateral flaps. Har-
trampf ( 13) espoused a different view of this controversy.

He argued for direct approximation of the residual ante-
rior fascial margins to achieve a centralization of the
remaining muscles. By so doing, he believes that the
mechanical advantage of the residual abdominal mus-
cles is restored. In 300 patients who underwent either
unilateral or bilateral rectus muscle transfers, there
was a 0.3% incidence of abdominal hernia and a 0.8%
incidence of abdominal wall laxity (14). In my experi-
ence of more than 300 unilateral rectus abdominis free
flap transfers in which the abdominal wall was closed
in this manner, there has been one patient who has
had an abdominal wall hernia that required repair. An
additional technique of anterior sheath closure using
autologous tissue should be mentioned. A "turnover,"
contralateral, anterior rectus sheath flap based at the
midline has been reported for donor-site closure (23).
The postoperative function of the abdominal wall is
generally believed to be unaffected by transfer of a single
rectus abdominis muscle. Bunkis et al. ( 4) warned that
in patients who are active in sports or who engage in
other physical activities, there may be an impact on their
lifestyle. However, there are few reports that quantify
the actual effects of removing a single rectus abdominis
muscle. Hartrampf and Bennett (14) conducted the
RECTUS ABDOMINIS 151
most extensive study of abdominal wall function in 300
patients, of whom the majority returned to their preop-
erative level of function, based on the parameters used
in their investigation. Lejour and Dome (19) reported
that follow-up studies in 57 patients revealed a signifi-
cant discrepancy between a patient's responses to a
questionnaire and objective findings that were recorded
by a physiotherapist. Although the majority of patients
reported either no disturbance or improvem.ents in
abdominal strength and sports activities following sur-
gery, the physiotherapist reported a marked decrem.ent
in the functions of the recti and the external oblique
muscles.
POSTOPERATIVE CARE
Because of retraction of the peritoneal cavity for dis-
section of the vascular pedicle, it is not uncommon for
patients to develop an ileus in the early postoperative
period. Interim feedings must be delayed for a short
period until this resolves. Early postoperative ambulation
is encouraged. Exercises that involve the abdomen may
be resumed in approximately 6 weeks following surgery.
152 CHAPTER 10

FIGURE 10-5. The tapographical anatomy of

the rectus abdominis flap is autlined an the
abdomen. The position of the palpable pulse
of the femoral vessel is shown. In addition, the
iliac crest and costal margins are outlined.
In the midline, the linea alba has been drawn
from the pubis to the xiphoid. The approximate
position of the linea semilunaris is outlined
by a dashed line at the mid paint between the
pubis and the anterior superior iliac spine.
The OlEA and DIEV and the DSEA and DSEV
are shown in their approximate course on the
undersurface of the rectus abdom inis.

FIGURE 10-6. An extended deep inferior epi-

gastric flap has been outlined on the left side of
the abdamen. This flap extends over the costal
margin and provides an abundance of thin
well-vascularized pliable skin. The vascularity
to this flap depends an capture of the dominant
periumbilical perforators. This flap may extend
across the midline, capturing well-vascularized
tissue to approximately the level of the con-
tralateral linea semilunaris. The arcuate line is
located at the approximate longitudinal level
on the abdominal wall of the anterior superior
iliac spine. The anterior rectus sheath should
not be harvested below this level.

FIGURE 10-7. The dissection begins in the

cephalad portion of the flap by incising the
skin and subcutaneous tissue to the level
of the fascia. The full breadth of the rectus
abdominis is identified by incising the anterior
rectus sheath to expose the rectus abdominis
from the linea alba to the linea semilunaris.
 RECTUS ABOOMINIS 153

Rectus Abdominis Flap (Extended

Inferior t:P1aa1r11

FIGURE 10-8. The dissection is continued infe-

riorly in a similar plane through the skin and sub-
cutaneous tissue to expose the full width of 1he
caudal portion of the rectus abdominis. This is
achieved by incising the anterior rectus sheath
and exposing the rectus abdominis muscle from
the linea alba to the linea semilunaris.

FIGURE 10-9. Having identified the linea semi-

lunaris in its cephalad portion and in its caudal
portion, the skin paddle can now be elevated off
the external oblique muscle and aponeurosis
to the linea semilunaris, which is identified by
the dashed line. Identification of the dominant
perforators to the skin may allow1he surgeon
to further limit the dimensions of the anterior
sheath harvest. Defatting of the portion of this
flap, which has been elevated, may be safely
performed deep to Scarpa's fascia.

FIGURE 1010. Meticulous dissection in 1his

plane superficial to the external oblique fascia
is performed to identify the first set of muscu-
locutaneous perforators (srrow). The anterior
sheath is then incised laterally, preserving these
dominant perforators.
154 CHAPTER 10

FIGURE 10-11. The medial dissection is performed

by elevating the skin and subcutaneous tissue in a
prefascial plane above the contralateral anterior
rectus sheath. The exact position of the linea alba
is marked by a dotted line, having been identified
in both the cephalad and caudad exposure. A cuff
of anterior rectus sheath is preserved by making a
longitudinal incision lateral to the linea alba, as indi-
cated by the dashed line.

FIGURE 10-12. The lateral aspect of the dissection

is completed by incising the anterior rectus sheath
medial to the linea semilunaris, preserving a small
cuff of fascia to facilitate closure of the anterior
sheath. Exposure of the caudal aspect of the rectus
fascia has been obtained through a vertical skin inci-
sion. However, the exposure may be achieved with-
out this incision and simply by retracting on the lower
abdominal flap. A corresponding vertical incision in
the anterior rectus sheath is then made (dashed line)
in order to expose the caudal portion of the museIe
for harvest
FIGURE 10-13. Full exposure of the rectus abdominis
has been achieved by longib.ldinally incising the ante-
rior rectus sheath interiorly in the midportion between
the linea alba and the linea semilunaris. The anterior
rectus fascia is elevated both medially and laterally to
obtain full exposure ofthe muscle. The skin paddle is
completely isolated, except for its attachments to the
anterior rectus sheath in the midportion of the muscle.
Only that portion of the anterior rectus sheath that is
immediately subjacent to the skin paddle needs to be
harvested. The cephalad portion of the rectus abdomi-
nis may be incorporated in this flap by gaining expo-
sure through elevation of the anterior rectus sheath
to the costal margin. The attachments of the anterior
sheath to the muscle at the level of the tendinous
inscriptions require sllarp dissection.
 RECTUS ABOOMINIS 155

Rectus Abdominis Flap (Extended

Inferior t:P1aa1r11

FIGURE 10-14. The dissection progresses from

cephalad to caudad by elevating the rectus abdomi-
nis off the posterior rectus sheath. This is achieved
by transecting the rectus abdominis above and
bluntly dissecting between the muscle and the
posterior sheath. Blunt dissection along the linea
semilunaris reveals the segmental nerve supply
(small arrows). In the caudal aspect of this exposure,
the deep inferior epigastric pedicle (lsrge srrowt is
identified.

FIGURE 10-15. A closer view of the undersurface

of the rectus abdominis reveals the segmental nerve
supply and the deep inferior epigastric pedicle.

FIGURE 10-16. The extended deep inferior epigas-

tric flap has been completely isolated on its vascular
supply, and the segmental nerves are shown against
the blue backgrounds. Proximal dissection of the
vascular pedicle is achieved by the use of deep
abdominal retractors. Despite the large caliber and
length of the DIEA. it is helpful to extend the dis-
section to the takeoff from the external iliac artery
and vein because the venae comitantes join in most
cases to create a single DIEV at a variable distance
from the external iliac vein.
15& CHAPTER 10

FIGURE 1D-17. Meticulous closure of the donor

defect is required to prevent weakening or hernia-
tion of the anterior abdominal wall.lt is impera-
tive to close the anterior rectus sheath below
the arcuate line. This can be readily achieved by
designing the skin paddle so that only that portion
of the anterior rectus sheath above the arcuate
line is harvested.

FIGURE 1D-11. Closure of the anterior rectus

sheath below the arcuate line has been accom-
plished. The integrity of the anterior abdominal
wall is fortified by suturing the anterior rectus
sheath to the posterior rectus sheath at the level
of the arcuate line (arrowheads). Although the
posterior rectus sheath cephalad to the arcuate
line is probably of sufficient strength, it can be
augmented by closing the preserved cuffs of ante-
rior fascia attached to the linea semilunaris and
the linea alba. A ribbon retractor is usually placed
along the posterior rectus sheath to prevent the
errant placement of a suture into the peritoneal
cavity.

RGURE 1019. The anterior rectus sheath has

bee11 closed. An alternative approach to fortifying
the anterior abdominal wall is to suture a mesh to
the linea alba and the linea semilunaris, above the
arcuate line.
 RECTUS ABOOMINIS 157

Rectus Abdominis Flap (Extended

Inferior t:P1aa1r11

FIGURE 10-20. Closure of the skin is accom-

plished by wide undermining.

FIGURE 10-21. The rectus abdominis muscu-

locutaneous flap provides a large area of skin,
a long vascular pedicle, and a segmental nerve
supply.

FIGURE 10-22. If desirable, because of excess

bulk or the necessity for a longer vascular
pedicle, the amount of muscle that is incorpo-
rated in this flap may be significantly reduced
by separating the loose attachments of the
DIEA and DIEV on the undersurface of the
muscle in its proximal portion. In so doing, the
caudal aspect of the rectus abdominis may be
removed. This provides further length to the
vascular pedicle and reduces the amount of
muscle to only that portion in the paraumbilical
region that harbors the dominant musculocuta-
neous perforators.
158 CHAPTER 10

Rectus Abdominis Flap (Deep

Inferior igastric Perforator

FIGURE 1D-Zl. Harvest of a deep inferior

epigastric perforator flap is initially outlined as
a vertical skin paddle on the anterior abdomi-
nal wall.

FIGURE 1D-24. Three dominant cutaneous

perforators are identified exiting through the
anterior rectus sheath to supply the overlying
skin (marked A B. and CJ.
 RECTUS ABOOMINIS 159

Rectus Abdominis Flap (Deep

Inferior igastric Perforator Fla

FIGURE 10-25. Dissection through the mus-

cle has been performed to harvest a perfora-
tor flap based on the DIEA/DIEV with minimal
to no muscle surrounding 1he pedicle.

FIGURE 10-2&. The perforatorflap is shown

based on a single dominant musculocutane-
ous perforator.
160 CHAPTER 10
REFERENCES
1. Akizuki T, Harii K, Yamada A: Extremely thinned infe-
rior rectus abdominis free flap. Plast Reconszr Surg
1993;91 :936--941.
2. Boyd JB, Taylor GI, Corlett R: The vascular territories of
the superior epigastric and the deep inferior epigastric
systems. Plart Reconm Surg 1984;73:1-14.
3. Brown R, Vasconez L, Jurkiewicz M: Transverse abdom-
inal flaps and the deep epigastric arcade. Plast Reconszr
Surg 1975;55:416--419.
4. Bunkis J, Walton R, Mathes S, Krizek J, Vascomez I.:
Experience with the transverse lower rectus abdominis
operation for breast reconstruction. Plast Reconszr Surg
1983;72:819-827.
5. Carramenh.a e Costa M. Carriquiry C, Vasconez L, Grot-
ring I, Herrera R, Windle B: An anatomic study of the
venous drainage of the transverse rectus abdominis mus-
culocutaneous flap. Plast Reconszr Surg 1987;79:208.
6. Chicarilli ZN, Davey LM: Rectus abdominis
myocutaneous free-flap reconstruction following a cra-
nio-orbital-maxillary resection for neurofibrosarcoma.
Plast Reconszr Surg 1987;80:726-731.
7. Drever J: The epigastric island flap. Plast Reconszr Surg
1977;59:343-346.
8. Drever J. Hodson-Walker N: Closure of the donor defect
for breast reconstruction with rectus abdominis myocuta-
neous flaps. Plast Reconszr Surg 1985;76:558-567.
9. DuchateauJ,Declety A, Lejour M: Innervation of the rec-
tus abdominis muscle: implications fur rectus flaps. Plast
Reconszr Surg 1988;82:223-227.
10. Ebihara H, Maruyama YU: Free abdominal flaps: varia-
tions in design and application to soft tissue defects of the
head. J Reconszr Microsurg 1989;5: 193--201.
11. Godfrey P, Godfrey N, Romita M: The "circummus-
cular" free TRAM pedicle: a trap. Plast Reconszr Surg
1994;93:178-180.
12. Hallock G: Aesthetic approach to the rectus abdominis
free tissue transfer.J Reconszr Microsurg 1989;5:69-73.
13. Hartrampf CR: Discussion of article by Drever JM,
Hodson-Walker N: Closure of the donor defect for breast
reconstruction with rectus abdominis myocutaneous
flaps. Plast Reconstr Surg 1985;76:563--565.
14. Hartrampf CR., Bennett GK: Autogenous tissue recon-
struction in the mastectomy patient. Ann Plast Surg
1987;205:508-519.
15. HataY,Yano K, Matsuka K, Ito 0, Matsuda H, Hosokawa
KI: Treatment of chronic facial palsy by transplantation
of the neurovascularized free rectus abdominis muscle.
Plast Reconszr Surg 1990;86: 1178.
16. Jones N, Sekhar L, Schramm V: Free rectus abdominis
muscle flap reconstruction of the middle and posterior
cranial fossa. Plast Reconstr Surg 1986;78:471--473.
17. Kaufman T, Hurwitz D, Boehnke M. Futrell J: The
microcirculatory pattern of the transverse-abdominal
flap: a cross-sectional xerographic and CAT scanning
study. Ann Plast Surg 1985;14:340.
18. Koshima I, Moriguchi T, Fukuda H, Yoshikawa YS-S:
Free thinned paraumbilical perforator-based flaps. J
Reconszr Microsurg 1991;7:313-316.
19. Lejour M, Dome M: Abdominal wall function
after rectus abdominis transfer. Plast Reconszr Surg
1991;87:1054.
20. Logan S, Mathes S:The use of a rectus abdominis myo-
cutaneous flap to reconstruct a groin defect. Br J Plast
Surg 1984;37:351.
21. Markowitz BL, SatterbergT, Calcaterra T, et al: The deep
inferior epigastric rectus abdominis muscle and myocu-
taneous free tissue transfer: further applications for head
and neck reconstruction. Ann Plast Surg 1991;27:577.
22. Mathes S, Nahai F: Clinical Applications for Muscle and
Musculacutaneous Flaps. St. Louis: CV Mosby; 1982:
44--45.
23. Meland N, Fisher J, Irons G, Wood M. CooneyW: Expe-
rience with 80 rectus abdominis free-tissue transfers.
Plast Reconszr Surg 1989;83:481.
24. Milloy F, Anson B, McAfee D: The rectus abdominis
muscle and the epigastric arteries. Surg Gynecol Obszet
1960;11 0:293--302.
25. Miyamoto Y, Harada K, Kodama Y, Takahashi H,
Okano S: Cranial coverage involving scalp, bone and
dura using free inferior epigastric flap. Br J Plast Surg
1986;39:483.
26. Moon H, Thylor GI: The vascular anatomy of rec-
tus abdominis musculocutaneous flaps based on the
deep superior epigastric system. Plart Reconszr Surg
1988;82:815-829.
27. Pennington D'l..ai M. Pelly A: The rectus abdominis
myocutaneous free flap. Br J Plast Surg 1980;33:277.
28. Sadove R, Merrell J: The split rectus abdominis free mus-
cle transfer. Ann Plast Surg 1987;18:179-181.
29. Schetlan M, Dinner M: The transverse abdomi-
nal island flap, II: surgical technique. Ann Plast Surg
1983;10:120-129.
30. Takayanagi S, Ohtsuka M: Extended transverse rectus
abdominis musculocutaneous flap. Plast Reconszr Surg
1989;83: 1057.
31. Taylor G, Corlett RJ, Boyd B: The versatile deep inferior
epigastric (inferior rectus abdominis) flap. Br J Plast Surg
1984;37:330--350.
32. Taylor G, Corlett R, Boyd J: The extended deep inferior
epigastric flap: a clinical technique. Plast Reconstr Surg
1983;72:751.
33. Taylor I, Corlett R, Caddy C, Zeit Z: An anatomic review
of the delay phenomenon, II: clinical applications. Plast
Reconszr Surg 1992;89:408.

34. Taylor G, Palmer }:The vascular territories (angiosomes)
of the body: experimental and clinical applications. Br J
Plan Surg 1987;40:113-131.
35. Urken MI.., Catalano PJ, Sen C, Post K, Futran N, Biller
HF: Free tissue transfer for skull base reconstruction:
analysis of complications and a classification scheme for
defining skull base defects. Arch Owlaryngol Head Neck
Surg 1993;119:1318.
RECTUS ABDOMINIS 161
36. Urken MI.., Turk J, Weinberg H, Vickery C, Biller HF:The
rectus abdominis free flap in head and neck reconstruc-
tion. Arch OUJlaryngolHeadNeikSurg 1991;117:857-866.
37. Yamada A, Harii I{, U eda K,Asato H: Free rectus abdom-
inis muscle reconstruction of the anterior skull base. Br J
Plast Surg 1992;45:302-306.
'"'J""he gracilis muscle was one of the first musculocu-
~ taneous flaps to be transferred by microvascular
teclm.iques. Harii et al. (4) introduced this free flap in
1976 and subsequently popularized ita use for dynamic
facial reanimation. The gracilis flap has also been used
extensively as a pedicled flap for defects of the peri-
neum, including the vagina, rectum, and even pres-
sure ulcers overlying the ischium (11-13). The gracilis
muscle has an easily identifiable vascular pedicle with a
large motor nerve that makes it suitable for transfer and
reinnervation.
The gracilis muscle is an adductor and a medial rota-
tor of the thigh. It is a long strap-like muscle that arises
from the pubic symphysis and ramus and inserts below
the knee onto the tibia. The powerful adductor longus
162
and adductor magnua muscles compensate for the func-
tional loss of the gracilis muscle (Figs. 11-1 and 11-2).
FLAP DESIGN AND UTILIZATION
The gracilis muscle is a long thin muscle, measuring
4 to 6 em in width. The neurovascular pedicle enters
the proximal portion of the muscle 8 to 10 em cau-
dal to the pubic tubercle (Fig. 11-3). The dominant
musculocutaneous perforator is located in the same
vicinity.
The primary use of the gracilis muscle in the head
and neck has been for facial reanimation; the mus-
cle is both revascularized and reinnervated to restore

-+--- Sartorius m.
-+--- Rectus femoris m.
FIGURE 11-1. The muscular anatomy of the medial1high is
shown from the anterior view. The gracilis arises from the
body and inferior pubic ramus. It crosses 1he knee to insert
into the medial aspect of1he upper end of the tibia. As it
runs its cephalocaudal course, it narrows significantly until
terminating at its tendinous insertion.
its contractile activity. In Harii et al.'s (5) original
description of this technique, a segment of the gra-
cilis muscle was transferred to the paralyzed side of
the face, and its motor nerve was coapted to the deep
temporal nerve. One end of the muscle was sutured to
the orbicularis oris at the lateral commissure, and the
GRACILIS 163
:..;..;..iOF.-fW-1- Adductor
magnus m.
m.
FIGURE 112. The surgical approach to 1he gracilis is
performed from the medial aspect of the thigh. The muscle
tendons that converge to make up the pes anserinus are the
sartorius, gracilis, semimembranosus, and semitendinosus.
The powerful adductor longus and mag nus compensate for
the loss of the gracilis muscle.
other end was sutured under tension to the temporal
fascia.
Over time, there have been several refinements
to Harii's original technique. To achieve synchro-
nous mimetic movement in the absence of the ipsi-
lateral facial nerve stump, a two-stage procedure was
described employing a cross-face sural nerve graft in
the first stage, followed by second-stage muscle ttans-
fer once axons populate the tip of the cross-face nerve
graft (3). The arrival of axons at the tip of the graft
is determined by the presence of Tinel sign, a tin-
gling in the donor muscle when tapping over the tip
of the nerve graft. The utilization of branches of the
164 CHAPTER 11

healthy conttalateral facial nerve is feasible based upon

the known. robust arborization of the facial nerve in
the midface. Selective neurectomy of donor branches
does not result in loss of smile function on the donor
' -- -- An tenor branch side. Kumar and Hassan (6) describe favorable results
obturator n.
combining the two stages into one in a small series
rw~~l-- Profunda femoris
artery and vein
of patients. After placing the obturator nerve across
the face, it was coapted directly to the contralateral
l.:r ..,.....,+-- Branch of
adductor a. facial nerve at the time of muscle inset. This technique
resulted in the recovery of more rapid muscly activity,
compared with a 2-stage transfer. However, most facial
reanimation surgeons agree that the critical element
to successful cross-face nerve grafting is to harvest the
...;.~H-- Adductor donor branches much more proximally than the obtu-
longus m. rator nerve is able to reach, and so the one-stage tech-
nique has not been widely adopted.
In addition to introducing cross-face nerve grafts to
achieve mimetic activity, an effort was made to reduce
the bulk of the muscle in the face. Harii (3) described
transfer of a gracilis muscle segment 10 em in length
and 3 em in width, approximately one-half the breadth
of the muscle belly. M&nktelow (8) investigated the fas-
cicular pattern of the motor branch of the obturator
nerve, which supplies the gracilis. Of the three fascicles
usually present in that nerve, one was usually responsi-
ble for the innervation of the anterior 25% to 50% of
the muscle. Interfascicular dissection and intraoperative
stimulation allow selective harvest of only a small por-
tion of the muscle innervated by a single or a pair of
fascicles. Through this technique, .Manktelow has trans-
ferred a smaller portion of the muscle while maintain-
ing a neurovascular pedicle suitable for microvascular
anastomosis (Fig. 11-4).
Profunda femoris
Harii et al. (5) reported a wide range of results in the
artery and vein quality of facial reanimation using the free gracilis mus-
cle transfer in 122 patients. The selection of the donor
Branch of adductor ~-:.-+.....,.
artery and vein nerve used to drive the transplanted muscle appeared to
have the greatest impact on the results of reanimation.
The most consistent results were achieved with coap-
tation of the motor nerve of the gracilis to the stump
Adductor ~- of the ipsilateral facial nerve. However, good to satis-
-'*"-~-+- Adductor
longus m. factory results were also achieved with the two-stage
magnus m.
cross-face nerve grafting method (see Fig. 11-14). Frey
FIGURE 11-3. The vascular supply to the gracilis muscle et al. discovered that results were relatively independent
is from a terminal branch of the adductor artery and vein, of the total number of neurons traversing the graft (2).
which arise from the profunda femoris artery and vein. The .Manktelow and Zuker and others have written
adductor artery usually arises from the profunda femoris extensively on the utility of the motor branch to the
artery, but it may also arise from the medial circumflex masseter muscle to supply neural input to the graci-
femoral artery. The gracilis branch of the adductor artery, lis muscle in one-stage facial reanimation procedures
accompanied by paired venae comitantes, passes between (1,6,7,9) and popularized the technique for the treat-
adductor longus and adductor magnus. The anterior divi- mentofMobiussyndrome (Figs.ll-17 and 11-18)(15).
sion of the obturator nerve supplies the motor innervation to They have found that using the masseteric nerve leads
the gracilis muscle. It enters the muscle in a more cephalad to increased average excursion compared with cross-
location and runs a more oblique course than the main face nerve grafting, and numerous groups describe
vascular pedicle. that cortical adaptation occurs, permitting voluntary
 GRACILIS 165

Gracilis harvested

FIGURE 11-4. The branching pattern of the anterior division of the obturator nerve allows the
gracilis muscle to be separated into at least two functional muscular units. A single fascicle
usually supplies the anterior 25% of the muscle; the remaining nerve fascicles supply the rest of
the muscle. A small portion of the muscle can be harvested with the main vascular pedicle and
the fascicle from the anterior branch of the obturator nerve.

amiling in the absence of teeth clenching. The current
hypothesis for this phenomenon involves neural plas-
ticity at the level of the motor cortex, whereby path-
ways connecting the "command" to initiate a voluntary
smile with the cortical zone responsible for mastication
become established, through repetition and visual and/
or EM.G biofeedback.
The gracilis musculocutaneous flap was originally
descn"bed by Harii et al. (5) with a skin island designed
over the prcmmal muscle. A longitudinally oriented skin
paddle overlying the proximal muscle was transferred
for a variety of defects (Fig. 11-S).An alternative trans-
verse skin paddle design was descn"bed by Yousif et al.
(14) and is believed to increase the reliability of the skin
paddle. However, given the superior reliability of other
musculocutaneous free fiaps, the gracilis musculocuta-
neous fiap is no longer a first-line choice in head and
neck reconstruction when skin is req_uired.
NEUROVASCULAR ANATOMY
The gracilis muscle has a type II vascular pattern,
according to the classification system of Mathes and
Nahai (10). The dominant pedicle is the terminal
branch of the adductor artery, which arises from the
profunda femoris. It runs a circuitous course between
the adductor longus anteriorly and the adductor brevis
and magnus posteriorly before entering the gracilis at
the junction of the upper third and lower two-thirds
(Fig. 11-3). This point of entry is consistently located
between 8 and 10 em inferior to the pubic tubercle.
The adductor artery arises from the profunda femo-
ris in the vicinity of the first perforator or from the
medial femoral circum1lex artery. It gives off branches
to adductor longus and brevis, which must be ligated
to obtain adeq_uate pedicle length, which is usually 6
em. The average arterial diameter is 2 mm. Although
the adductor artery may take its origin from the medial
circumflex femoral artery, the main vascular supply to
the gracilis muscle is not the medial circumflex femoral
artery itself.
The minor vascular pedicle arises from the supel.'-
ficial femoral artery and enters the lower third of the
muscle. An additional minor vascular supply arises from
the medial circumfiex femoral artery.
The major artery to the gracilis is accompanied by
two venae comitantes. These veins may either join or
drain separately into the profunda femoris vein. The
average diameter of the veins is 1.5 to 2.5 mm.
16& CHAPTER 11

Profunda femoris a.

Adductor brevis m.
FIGURE 11-5. The cross-sectional anatomy of the medial thigh shows the course of the
adductor artery arising from the profunda femoris artery and running between the adductor
muscles with the brevis and mag nus located posteriorly and the longus located anteriorly. The
musculocutaneous perforator enters the skin opposite the point where the main pedicle enters
the muscle.

The blood supply to the skin is derived through the
system of musculocutaneous perforators. The domi-
nant perforators exit the muscle in the upper third,
with few noted in the middle and lower portions. Yousif
et al. (14) described an additional skin supply through
septocutaneous vessels that ezited through the intermu5-
cularseptumbetween the gracilis and the adductor longus.
The orientation of the terminal bnmches of the septocu-
taneous vessels was also transVerse. The septocutaneous
vessels in the distal thigh are bnmches of the superficial
femoral artery, rather than the profunda system.
Motor input to the gracilis muscle is provided by
the anterior branch of the obturator nerve, which
enters the muscle obliquely, 2 to 3 em cephalad to the
entry point of the vascular pedicle. The nerve may be
traced proximally betWeen adductor longus and bre-
vis to gain additional length. The sensory supply to
the medial thigh skin is from branches of the obtura-
tor nerve, which may be dissected in the subcutaneous
tissues cephalad to the skin paddle, though there are
no published reports of sensory reinnervation with the
gracilis musculocutaneous fiap.
ANATOMIC VARIATIONS
The vascular and nerve supply to the gracilis mus-
cle is consistent. The major variability is noted in
the blood supply to the overlying skin, both in the
number and the size of musculocutaneous perfora-
tors. Yousif et al. (14) described several dissections
in which there were no musculocutaneous perfora-
tors exiting the gracilis, and the major skin supply
was derived from septocutaneous vessels or from the
inferior branch of the superior external pudendal
artery, which extended into the territory of the gra-
cilis. Because the skin paddle vascularity is less pre-
dictable than other musculocutaneous free flaps, it is
employed much more often as a muscle on{y flap for
dynamic facial reconstruction.

POTENTIAL PITFALLS
The morbidity of removing the gracilis muscle is limited
because of the strength of the remaining adductor mus-
cles. Due to the small average artery and vein diameter
compared with other free flaps, they are more techni-
cally demanding for the novice microvascular surgeon.
Occasionally, the vascular pedicle can enter the overlying
GRACILIS 167
adductor longus muscle, which, when unrecognized, can
compromise pedicle length and necessitate vein grafting
during inset. It is critical to recognize when the domi-
nant pedicle enters the adductor longus, because simple
dissection of the artery and venae comitantes through
a small segment of muscle will then permit continued
pedicle dissection toward the profunda femoris system,
with resultant harvest of appropriate length vessels.
168 CHAPTER 11

Gracilis Flap

FIGURE 116. This dissection demonstrates

the harvest of the left gracilis muscle flap.
With the leg flexed and abducted, a dotted line
is drawn between the pubic tubercle and the
medial condyle ofthe tibia. The superior edge
of the gracilis muscle lies approximately 1to
1.5 em posterior to this line. The neurovascular
pedicle enters the muscle on its undersurface
approximately 8 to HI em below the pubic
tubercle. If a musculocutaneous flap was
harvested, the skin paddle would be centered
over this point, oriented either in a transverse
or in a longitudinal direction.

FIGURE 117. The initial incision is carried

through the skin and subcutaneous tissue to
identify the midbelly of the gracilis muscle. The
sartorius is a good landmark to help identify
the gracilis in the midthigh. From the surgical
position, the sartorius is located immediately
above the gracilis. In the proximal thigh, the
adductor longus is immediately above the
gracilis, which is shewn at this point in this
dissection.

FIGURE 11-8. With the anterior edge of the

gracilis muscle reflected pasteriorly, the neura-
vascular pedicle is easily identified as it enters
the muscle on its deep surface. The neurovas-
cular pedicle exits between the adductar longus
above and adductor magnus below before
entering the gracilis.
 GRACILIS 169

Gracilis Flap

FIGURE 11-9. After the artery and vein are

identified, the anterior obturator nerve (arrow)
is found entering the muscle in a mare ablique
course and 2 to 3 em proximal to the vascular
pedicle.

FIGURE 11-10. Branches (arrows) from the

vascular pedicle to the adductor longus muscle
must be ligated ta maximize the length of the
pedicle.

FIGURE 11-11. The vascular pedicle can be

dissected medially to obtain an average of6 em
of length. The maximum diameter af the artery
is approximately 2 mm, and the vein may reach
a diameter of3 mm.
110 CHAPTER 11

Gracilis Flap

FIGURE 1112. The wound is closed by

reapproximating the fascia of adductor longus
and mag nus and then closing the skin in layers.

FIGURE 11-13. A segment of the gracilis

muscle has been harvested with the nutrient
artery and its paired venae comitantes. The
anterior obturator nerve is shown entering the
muscle in a more proximal location. The muscle
may be divided longitudinally to reduce the bulk
while preserving the neurovascular pedicle to
the anterior segment.
 GRACILIS 171

Gracilis Muscle for Facial Reanimation

Sural
Fac:llll artery and velA

FIGURE 11-14. The harvested gracilis is inset systematically. A pseudotendon is created on

both ends of the harvested muscle using a 3-0 vicryl running locking suture. The proximal inset
is performed using 0-vicryl at five carefully determined locations that duplicate the normal smile
pattern. Two sutures are placed in the upper lip, one at the modiolus and two in the lower lip,
utilizing a mattress suture technique. The distal inset is performed using 0-vicryl, securing the
muscle to the true temporalis fascia.
112 CHAPTER 11

Anastomosis of Cross Facial Nerve

Graft to Obturator Nerve

FIGURE 11-15. The distal end of the previously

placed cross-facial sural nerve graft is located
through a mucosal incision in the upper lip.
The obturator nerve is passed medially through
the intraoral incision prior to muscle inset, and
the neurorrhaphy is performed via an intraoral
approach.
 GRACILIS 113

Isolation of Masseteric Motor Nerve

/
I
I
~-

/
~~~~~~~--~~~+--~oow
1Uberc:le

TemporalIs
muscle

Coronoid
prOC888

FIGURE 1116. The masseteric branch of the trigeminal nerve (V3) can be used as the donor
nerve when patients desire a one-stage procedure or the contralateral facial nerve is not an
appropriate donor nerve {i.e., NF-2 or bilateral facial paralysis). It is located by first reflecting
the masseter muscle inferiorly through division of its fascial attachments to the zygomatic arch.
The masseter nerve is typically located 1 to 11 mm anterior to the articular tubercle and within
the body of the masseter muscle. The use of a lighted retractor and a nerve stimulator facilitates
this dissection.
114 CHAPTER 11

Reinnervation of the Masseteric Nerve and Bilateral Gracilis Muscle Transfer

/
I A

1'
I~

nerve

FIGURE 11-17. When using the masseteric

branch of V3, the vascular anastomoses are
performed, and the neurorrhaphy is executed
under the zygomatic arch prior to lateral muscle Facial artery and vein
inset.

RGURE 11-11. In bilateral facial paralysis,

simultaneous bilateral free gracilis muscle
transfer with oral sphincter reconstruction
can be performed for one-stage bilateral facial
reanimation. Two extended gracilis muscles
are harvested (14-16 em) and bivalved at one
end. Coaptation of the left and right gracili is
performed through vertical incisions made in
the midline of the upper and lower lip vermilion,
creating a complete oral sphincter.

REFERENCES
1. BaeYC, Zuker RM, Manktelow RT, Wade S: A comparison
of commissure excursion following gracilis muscle trans-
plantation for facial paralysis using a cross-face nerve graft
versus the motor nerve to the masseter nerve. Plart Reconm
Surg 2006;117:2407-2413.
2. Frey M, Happak W, Girsch W, Bittner RE, Gruber H: His-
tomorphometric studies in patients with facial palsy treated
by functional muscle transplantation: new aspects for the
surgical concept. Ann Plast Surg 1991;26:370-379.
3. Harii K: Microneurovascular free muscle transplantation.
In: Rubin 1.., ed. The Paralyzed Face. Philadelphia: Mosby
Yearbook; 1991:178-200.
4. Harii K, Olunori K, Sekiguchi J:The free musculocutane-
ous flap. Plast ReconstT Surg 1976;57:294-303.
5. Harii K, Ohmori K, Torii S: Free gracilis muscle transplan-
tation with microvascular anastomosis for the treatment of
facial paralysis. PlastReccnstrSurg 1976;57:133-135.
6. Kumar PA, Hassan KM: Cross-face nerve graft with free-
muscle transfer for reanimation of the paralyzed face: a com-
parative study of the single-stage and two-stage procedures.
Plast ReccnstT Surg 2002;109:451-462; discussion 463-454.
7. l..ifchez SD, Matloub HS, Gosain AI<: Cortical adaptation
to restoration of smiling after free muscle transfer inner-
vated by the nerve to the masseter. Plast ReconstT Surg
2005;115:1472- 1479.
GRACILIS 175
8. Manktelow R: Mic!ooarcular RecomtTUCrion: Anatong~,
Applications and Surgical Technique. New York: Springer-
Verlag; 1986.
9. Manktelow Kr, Tomat LR, Zuker RM, Chang M: Smile
reconstruction in adults with free muscle transfer inner-
vated by the masseter motor nerve: effectiveness and cere-
bral adaptation. Plast Reconm Surg 2006;118:885-899.
10. Mathes S, Nahai F: Clinical Application for Muscle and
Musculocutaneous Flaps. London: CV Mosby; 1982.
11. McGraw J, Massey F, Shanklin K, Horton C: Vaginal
reconstruction with gracilis myocutaneous flaps. Plast
ReconstT Surg 1976;58:176.
12. Pickrill K, Georgiade N, Maquire C, Crawford H: Graci-
lis muscle transplants for rectal incontinence. Surgery
1956;40:349.
13. Wingate G: Report of treatment of ischial pressure ulcers
with gracilis myocutaneous island flaps. Plast ReconstT
Surg 1978;62:245.
14. Yousif N, Matloub H, Kabachalam R, Grunert B,
Sanger J: The transverse gracilis musculocutaneous flap.
Ann Plasz Surg 1992;29:482-490.
15. Zuker RM, Goldberg CS, Manktelow RT. Facial ani-
mation in children with Mobius syndrome after seg-
mental gracilis muscle transplant. Plast RecomtT Surg
2000;106:1- 8; discussion 9.
~e first free fiap to be transferred that was based
~ on the radial artery was a segment of the superfi-
cial branch of the radial nerve and was performed by
'Iltylor in 1976 (48). However, the radial forearm fiap,
as a fasciocutaneous flap, was first introduced by Yang
et al. (57) in the Chinese litezature in 1981. This ini-
tial report of 60 radial forearm fiaps with only one fail-
ure was soon followed by additional publications from
China (40), and, hence, this flap became known as the
"Chinese fiap." Soutar's group ( 42-44) popularized the
radial forearm fiap for intraoral reconstruction through
a number of publications, the first of which appeared
in 1983.
Baaed on the radial artery and either the cephalic
vein or the venae comitantes, this fiap may be trans-
ferred as a composite fiap containing vascularized bone
( 43), vascularized tendon (34), the brachioradialis mus-
cle (36), vascularized nerve (19), or sensory nerves (20).
However, its thin pliable skin with its rich vascularity,
permitting a flexibility in design and a high degree of
176
reliability, is the key reason why the radial forearm flap
has assumed such an important place in head and neck
reconstruction.
FLAP DESIGN AND UTILIZATION
The skin of virtually the entire forearm, extending from
the antecubital fossa to the fiexor crease of the wrist,
may be harvested (Fig. 12-1). The thickness of this
fiap varies among individuals but tends to be thinner
in its distal portion. The fiap is also usually thinner in
male than in female patients. The degree and pattern of
hair-bearing skin also varies between individuals.
The radial forearm fasciocutaneous flap bas
been used more extensively and for more diverse
reconstructive problems than any other free flap.
Unquestionably, its greatest application is in the res-
toration of oral mucosal defects following ablative sur-
gery. It has been used in virtually every portion of the
 RADIAL FOREARM 117

I\
\

FIGURE 121. The size and shape of the radial forearm flap vary with the defect. The axis of the flap
should be centered over the course of the radial artery and the cephalic vein. The flap may extend
from the flexor crease of the wrist to above the antecubital fossa. There are regional differences in
the thickness of the subcutaneous tissue with the thinnest flaps harvested from the distal forearm.
The donor site is more easily camouflaged when the majority of it is located on the volar surface.

oral cavity (23,27,28,29,42,43,55).In 1994, Urken and
Biller (52) deacribed a bilobed design for the radial
forearm flap to help preserve the mobility of the residual
tongue following significant glossectomy. Toward that
end, this fl.ap provides thin and redundant tissue. The
bilobed design allows one lobe to be used to resurface
the tongue defect; the second lobe is placed in the Boor
of mouth. In so doing, the rest of the tongue remains
separate from the inner table of the mand:&ble. The deep
fascia and subcutaneous tissue can be harvested with-
out the overlying skin. Ismail (20) desenbed this fasci
subcutaneous fl.ap for extremity reconstruction and also
reported the improved aesthetic result of a straight-line
closure of the preserved forearm skin. This fascial flap
may be covered with a skin graft to resurface epithelial
defects. It is also highly effective in skull base surgery to
assist in dural repair; this thin well-vascularized tissue is
more easily inset in locations adjacent to the brain that
will not accommodate thicker flaps. Martin and Brown
(27) introduced the free radial forearm fascial Bap for
intraoral reconstruction. They described a rapid re-
epithelialization of the fascial-subcutaneous tissue layer
to achieve a mucosal surface. In addition, this teclm.ique
produced a thinner and less mobile layer over the man-
dible that was more conducive to the placement of a
dental prosthesis.The avoidance ofa skin graft produced
a more favorable donor-site appearance. Although this
technique has merit in resurfacing the alveolus, it may
lead to tethering in the Boor of the mouth as a result of
scar formation.
Another disadvantage of using a fascial free flap is
that it removes the sensory receptors preaent in the skin
and therefore leads to less predictable sensory restora-
tion following anastomosis of the antebrachial cutane-
ous nerves to recipient nerves in the head and neck. We
reported the first successful sensate radial forearm flap
in head and neck reconstruction (56). A young woman
with a pharyngeal defect underwent reconstruction
with a forearm Bap, and the sensory nerve was anasto-
mosed to the greater auricular nerve. The patient expe-
rienced sensation when driD.king hot and cold liquids
that was referred to her ear. Arguably, the greater auric-
ular nerve was not the best recipient nerve for a pharyn-
geal defect, but it provided a valuable result. It showed
178 CHAPTER 12
that sensate flaps could be successfully used in the head
and neck and that the mechanism for sensory recov-
ery was through the nerve anastomosis. Our experience
with sensate flaps has grown to include more than 60
cases, and our enthusiasm for this technique remains
high, based on the predictable level of sensory recov-
ery and the functional impact on patient rehabilitation
(52). The upper aerodigestive tract defects in which we
believe sensate flaps have their greatest applications are
mobile tongue and tongue base reconstructions, phar-
yngeal wall reconstruction, laryngeal reconstruction
following partial laryngectomy, and restoration of the
upper and lower lips.
In addition to defects of the lower half of the oral
cavity, the radial forearm flap has been used for palatal
reconstruction. Hatoko et al. (18) reported their favora-
ble experience with reconstructing the hard palate with
a folded double-layer forearm flap following maxil-
lectomy. One layer was used for the oral side and the
other layer for the nasal and sinus floors. These patients
were reportedly able to wear a maxillary denture. We
have had considerable success in reconstructing partial
and total soft palate defects with a folded radial fore-
arm flap. The two layers of the flap are used for the oro-
pharyngeal and nasopharyngeal sides of the defect. To
achieve velopharyngeal competence, the folded edge of
the flap is sutured to the posterior pharyngeal wall by
de-epithelializing opposing surfaces. Bilateral mucosa-
lined pons provide communication on either side of the
midline attachment.
The desire to add to the versatility of the forearm
donor site has led investigators to include additional
components to this flap. The use of the brachioradialis is
a product of that desire. It has been shown that although
the dominant muscular perforator to the brachioradialis
may arise from the radial artery (40%), the radial recur-
rent artery (33%), or the brachial artery (37%), it can
be reliably transferred with the radial artery because of
a series of secondary perforators (37). There are a large
number of musculocutaneous perforators that exit the
surface of the brachioradialis that allow a separate skin
paddle to be harvested with the muscle as a carrier. Not
only does this add bulk to the flap but it also permits
separate epithelial surfaces to be harvested for complex
defects. The brachioradialis may also be transferred as a
functional muscle unit if desired (36).
The brachioradialis has been used for total upper lip
reconstruction by suturing the two ends of the mus-
cle to the inferior mbicularis oris. Reinnervation of the
muscle was accomplished by suturing the motor nerve
of the brachioradialis to a buccal branch of the facial
nerve. Swanson et al. (45) reponed excellent functional
results and electromyographic evidence of electrical
activity in the muscle. Total lower lip reconstruction
with a radial forearm flap was reported by Sadove et al.
(36). The palmaris longus tendon was transferred with
the forearm skin to provide support to the lower lip
and to maintain the height of the lower lip. The com-
bination of the palmaris longus tendon with the sen-
sory nerve supply provides an elegant total lower lip
reconstruction.
There are occasions in head and neck reconstruction
in which multiple skin paddles are required. The radial
forearm flap has been divided into two epithelial sur-
faces separated by a de-epithelialized zone for providing
inner and outer linings (5). The intervening zone has
been divided down to the level of the fascia while still
providing adequate vascularity to the distal skin pad-
dle (3).YousifandYe (58) divided the perforators to the
forearm skin into three clusters of vessels along the lat-
eral intermuscular septum, each capable of supporting
a segment of skin.
A segment of radius, limited proximally by the inser-
tion of the pronator teres and distally by the insertion of
the brachioradialis, may also be harvested (Fig. 12-2).
The length of bone is no greater than 10 to 12 em, and
the bone stock is restricted to 40% of the circumference
of the radius. Soutar and Widdowson (44) reponed the
successful use of the osteocutaneous radial forearm
flap in 12 of 14 patients who undeiWent oromandibu-
lar reconstruction. Osteotomies were created in nine
patients to achieve a more favorable contour of the
neomandible. Although the composite osteocutaneous
flap is conceptually attractive for the reconstruction of
oromandibular defects, there are two major factors that
restrict its use. The dimensions of the bone that can be
safely harvested are limited by the necessity to maintain
the structural integrity of the remaining radius. Other
donor sites for vascularized bone provide much better
bone stock for functional mandibular reconstruction
(31,51). The second major factor mitigating against the
use of the radius is the potential morbidity resulting from
pathologic fractures that have occurred in up to 23%
of reported cases (51). Although specific techniques in
creating the osteotomies and prolonged postoperative
immobilization help to limit the incidence of fractures,
the potential morbidity and the poor bone stock make
this a less favored site for harvesting vascularized bone.
A more detailed discussion of the osteocutaneous radial
forearm flap is provided in Chapter 23.
As noted above, the superficial branches of the radial
nerve may be transferred as vascularized nerve grafts.
An isolated case report using this technique for the
repair of the facial nerve revealed excellent results (20).
However, the true value of vascularized nerve grafts
over nonvascularized grafts remains controversial.
The radial forearm flap has been used successfully to
cover large cutaneous defects of the head and neck, in
particular, those involving the scalp where thin coverage
is desirable (8). The radial forearm flap has also been
 RADIAL FOREARM 179

Cephalic v

Segment of radius

FIGURE 12-Z. The cross-sectional anatomy of the forearm reveals the radial artery with its
venae comitantes in the lateral intermuscular septum. The connection of the septum to the
radius provides vascularity through perforators that supply the periosteum. The amount of
radius that may be safely harvested is limited to 40% of the circumference.
180 CHAPTER 12
applied to the restoration of complex defects of the nose
and forehead (1,7).
Regional differences in the thickness of the skin of the
forearm can be used to achieve a more aesthetic result,
as dictated by the particular defect. More proximal skin
paddles offer thicker tissue and the potential for wider
flaps. Skin grafts placed over the proximal muscle bed
are more reliable than those that are placed over the
distal tendons. The disadvantage of a skin flap harvested
from a proximal location is that it significantly shortens
the arterial pedicle. Baird et al. (1) described a proximal
forearm flap for forehead reconstruction in which per-
fusion was maintained by retrograde flow through the
distal radial artery, which was attached to the superficial
temporal artery on one side of the head. The cephalic
vein was dissected proximally in the arm for additional
length and was anastomosed to the external jugular vein
in the contralateral neck.
The radial forearm flap has also been applied to
smaller defects in the head and neck. Tahara and Susuki
(46) reported favorable results when introducing a
radial forearm flap into the orbit to correct malignant
contracture of an irradiated enophthalmic eye socket.
The creation of an epithelium-lined socket permitted
the patient to wear an orbital prosthesis.
Another application for the radial forearm flap is in
circumferential pharyngoesophageal reconstruction.
Harii et al. (17) introduced the concept of a tubed
radial forearm flap to reconstruct a laryngopharyngec-
tomy defect. The thin pliable tissue from this donor site
is more readily tubed than is a thicker musculocuta-
neous flap. In addition, this method of reconstruction
offers distinct advantages over free jejunal autografts
because of the avoidance of a laparotomy. Two differ-
ent designs for a tubed radial forearm flap reconstruc-
tion of the pharyngoesophagus are demonstrated in this
chapter: one is a more longitudinal orientation of the
flap (Figs. 12-32 to 12-36) and the second a horizontal
design (Figs. 12-37 to 12-44). In addition, the tubed
radial forearm flap can be used to lengthen the short
tracheal stump following resection of stomal recur-
rences. The design of a trapezoid-shaped flap produces
a funnel-shaped tube that can bridge the gap between
the tracheal stump and the cervical skin (Figs. 12-45 to
12-51).
The versatility of the forearm donor site is further
reflected by its use in reconstructing defects of the larynx
and pharynx. Chantrain et al. ( 6) described the application
of the radial forearm flap to the reconstruction of a vt:rtical
hemipharyngolaryngectomy defect in three patients with a
pyriform sinus cancer and one patient with a transglottic
carcinoma.The tendon of the palmaris longus was included
with the flap and fixed anteriorly to the thyroid cartilage
and posteriorly to a hole drilled in the midline of the ros-
trum of the cricoid cartilage. This maneuver provided a
fixed position for the neocord. The cutaneous portion of
the flap was folded to line the hemilarynx and the medial
wall of the pyriform sinus. All four patients were able to eat
orally, and three of the patients were decannulated. The
full extent to which this technique can be applied in partial
laryngeal surgery has not y.:t been realized.
Hagen (14) presented a modified design of the radial
forearm flap for postlaryngectomy voice rehabilitation.
The forearm flap was tubed to create a skin-lined tube
that was sutured to the cephalad end of the trachea. An
epiglottis-like structure that was reinforced with autol-
ogous cartilage was sutured over the open end of the
tube positioned at the base of tongue. The advantage
of this form of alaryngeal speech was that there was no
prosthesis required, and the phonation pressure was
less than that with a tracheoesophageal prosthesis. Most
importantly, the seven patients in the series were able to
swallow without aspiration.
The use of the radial forearm flap in the oro- or
hypopharynx creates a problem for postoperative moni-
toring because of the limited access to these regions.
I introduced a new design for the buried radial forearm
flap in which there are two skin paddles: a distal one for
resurfacing the mucosal defect and a smaller proximal
one that is exteriorized in the neck (Fig. 12-3) (53). The
intervening fascial subcutaneous tissue is used to cover
the carotid artery and provide augmentation to the radi-
cal neck dissection defect. Because the superficial veins
and the radial artery are completely encompassed by
this vascularized subcutaneous tissue, it provides an
effective barrier for the pedicle in the event of a salivary
fistula.
A design of the radial forearm flap, described as the
beavertail modification, has been published by Seikaly
et al. (38). In their modification, the fat and fascia of the
upper part of the forearm are separated from the vas-
cular pedicle and left attached to the proximal portion
of the skin paddle to function as a random adipofas-
cial extension of the radial forearm flap (Fig. 12-21 to
12-31). The fat is then rolled and precisely placed into
the defect to provide bulk in the area required. Excellent
functional outcomes have been documented using this
technique for base of tongue reconstructions (35), and
this modification can be used in many situations where
strategically placed bulk is required in conjunction with
a radial forearm flap.
NEUROVASCULAR ANATOMY
The blood supply to the lower arm and the hand is
derived from the brachial artery, which divides into the
radial and ulnar arteries at the level of the antecubital
fossa. The radial artery gives rise to the deep palmar
arch; the ulnar artery terminates in the superficial pal-
mar arch. Harvest of the radial forearm flap requires
complete interruption of the radial artery and therefore
 RADIAL FOREARM 181

Lateral antebrachial cutaneous n.

FIGURE 12-3. A modified design of the radial forearm flap has a distal skin paddle for resurfac-
ing a mucosal defect and a proximal skin paddle 1hat is exteriorized in the neck to serve as a
monitor. The intervening subcutaneous tissue provides coverage of 1he great vessels and the
microvascular pedicle and also leads to augmentation of the radical neck dissection deformity.

total reliance on the ulnar system to maintain the
vascular supply to the hand (Fig. 12-4).
There are four arterial systeDUI that supply the
forearm skin through an array of septocutaneous and
musculocutaneous perforators. These four vessels are
the radial, ulnar, anterior, and posterior interosseous
arteries (24).
The fl.exor and extensor muscles of the forearm are
enclosed by a common fascial sheath. A condensation
of this fascia, referred to as the lateral intermuscular
septum, separates the brachioradialis and the flexor
carpi radialis in the forearm (Fig. 12-2). The radial
artery, with its two venae comitantes, runs in the lat-
eral intermuscular septum and gives off 9 to 17 fascial
branches in the forearm. 'Ihis fascial plexus supplies the
skin of virtually the entire forearm. There are few fas-
cial branches in the middle third of the forearm, and
in fact, the connections between the radial artery and
the deep fascia are attenuated because of the overlap
between the flexor carpi radialis and the brachioradialis.
The radial artery gives off a few fascial branches in the
prcmmal third of the forearm. There is one dominant
faaciocutaneous branch in the proximal forearm, the
inferior cubital artery, which has been used to supply a
prcmmally based fasciocutaneous fl.ap. This vessel may
arise from either the radial or the radial recurrent arter-
ies (24). Ink-injection studies revealed that the vessels
in the distal zone are capable of supplying a fasciocu-
taneous fl.ap that atends proximally to the elbow (48).
The maximum dimensions of the skin territory that can
be based on the radial artery are undetermined. Yang
et al. (57) reported a radial forearm fl.ap that measured
35 x 15 em. There is at least one case in the literature
in which the entire skin of an amputated forearm was
ttansferred as a free flap based on the radial artery (56).
In addition to supplying the forearm skin through
the fascial plexus, the radial artery sends branches to
the muscles of the flexor compartment, the palmaris
longus tendon, and the radial nerve. The lateral intel'-
muscular septum is attached to the distal radius, and it
is through this connection that it supplies branches to
the periosteum, allowing harvest of a segment of vas-
cularized bone (10). A longitudinal vascular arcade has
been described on the surface of the periosteum that
originates in close proximity to the insertions of fl.exor
pollicis longus and pronator quadratus (42).
The radial forearm flap has a deep venous supply
through the two paired venae comitantes, which run in
the intermuscular septUm as well as the larger superficial
veins, such as the cephalic vein. Both venous systems
have valves that permit unidirectional fl.ow. The veins
that supply the fascial plexus run with the branches of
the radial artery and drain into the venae comitantes.
The multiple connections between the venae comitantes
and the superficial veins form the basis for using either
of these two systems to drain the flap. The branching
patterns of the deep and superficial venous systems
have been classified into five different types (49). In the
type 1 pattern (20%), there is a wide communication
through an anastomotic vein between the superficial
182 CHAPTER 12

~lnara. Deep palmar arch Superficial palmar arch

FIGURE 12-4. The radial artery runs a course between the flexor carpi radialis and the bra-
chioradialis muscles before terminating in the deep palmar arch. The deep palmar arch sup-
plies the principal circulation to the thumb and index finger. The ulnar artery terminates in the
superficial palmar arch, which primarily supplies the third, fourth, and fifth digits and often also
the index finger.

and deep systems. In addition, the median cubital vein
splits into two large branches: the cephalic median vein
and the baailic median vein. The type 2 pattern (43%)
is the most common and is identical to type 1 except
that the median cubital vein does not bifurcate. The
type 3 pattern (18%) includes a conBuence of the two
venae comitantes to a sizable common trunk, but there
is no significant communication of this common trw::Lk
with the cephalic vein to form a median cubital vein.
In the type 4 pattern (5%), the venae comitantes do
not converge or join with the cephalic vein, but each is
of suitable caliber for microvascular anastomosis. The
type 5 pattern (15%) is similar to type 4 except that
one of the venae comitantea is dominant relative to the
other. I prefer to use the subcutaneous venous system
because of the larger caliber vessels and the thicker wall,
which permits an easier anastomosis. Often, there are
multiple subcutaneous veins suitable for anastomosis,
depending on the design of the radial forearm flap.
When the types 1 and 2 patterns are identified, I prefer
to use the median cubital vein for anastomosis based
on the theoretical advantage that both the deep and
superficial systems are being directly drained. However,
I have never encountered a problem of venous :insuffi.-
ciency when using either the superficial or deep systems
independently.
In addition, it has been my experience that in
patients who have been hospitalized for prolonged peri-
ods of time, the antecubital veins are often thrombosed.
In these patients the deep venous system is usually
enlarged and suitable for microvascular anastomosis.
The length of the arterial pedicle to this fiap is lim-
ited by the radial recurrent artery, which is the first
major branch of the radial artery following ita takeoff
from the brachial artery. The radial recurrent artery is
primarily a muscular branch. Alternatively, the cephalic
vein may be traced throughout its entire course in the
upper arm to ita junction with the subclavian vein in the
infraclavicular region. The additional length of vein that
can be harvested by extending the dissection above the
antecubital fossa may be helpful in skull base reconatruo-
tions in which vein grafts are frequently needed. This
flap has been used in the head and neck in a manner in
which the venous outfiow is maintained without intel'-
ruption of the cephalic vein while performing only an
anerial anastomosis (32). Using this technique of an
extensively mobilized but uninterrupted cephalic vein,
Bhathena and Kavarana (3) described using a proximal
 RADIAL FOREARM 183

forearm fiap that was revascularized through retrograde
arterial flow through the distal radial artery. The reli-
ance on retrograde flow through the radial artery is rou-
tinely used when the "reversed radial forearm flap" is
transferred as a pedicled fiap to the hand (21).
The cutaneous innervation of the forearm is derived
&om the medial, lateral, and posterior antebrachial
cutaneous nerves (Fig. 12-5). The lateral antebrachial
cutaneous nerve, the continuation of the musculocuta-
neous nerve, is the primary sensory nerve to the ter-
ritory of forearm skin most commonly harvested. nus
nerve usually runs in close proximity to the cephalic
vein in the upper forearm before ramifying in the distal
forearm and continuing onto the hand in the region of
the thenar eminence. The medial antebrachial cutane-
ous nerve arises &om the medial cord of the brachial
plexus, runs with the basilic vein, and supplies the skin
of the medial aspect of the forearm. The skin of the dor-
sum of the forearm is supplied by posterior branches
of the medial and lateral antebrachial cutaneous nerves
and by the posterior antebrachial cutaneous nerve, a
branch of the radial nerve. The medial and posterior
antebrachial cutaneous nerves are rarely encountered in
the dissection of a forearm flap, except when very large
fiaps are harvested.
The radial nerve is a mixed motor and sensory nerve.
It supplies most of the muscles of the extensor compart-
ment and the abductor pollicis longus and brevis. The
sensory distribution of this nerve includes portions of
the upper arm and the dorsum of the forearm through
the posterior antebrachial cutaneous nerve. The super-
ficial branch of the radial nerve courses in the forearm,
deep to the brac:hioradialis, and passes laterally to the
brachioradialis tendon where it becomes superficial and
then divides into the dorsal digital nerves supplying sen-
sation to the dorsum of the hand, thumb, and index and
middle fingers. Although the distal branching pattern is
highly variable, the major branches are always encoun-
tered in the wrist while elevating the radial forearm
flap. This nerve and all ita branches can be routinely
preserved to maintain sensation to the hand.
ANATOMIC VARIATIONS
The anatomy of the superficial venous system in the
forearm is highly variable. However, the pattern of veins
can be easily mapped prior to surgery by placing a tour-
niquet on the upper arm and tracing the course of the
engorged veins.
The greatest concern in harvesting the radial fore-
arm flap is the integrity of the ulnar arterial supply
to the hand through the superficial palmar arch. It is
therefore the anomalies of the arterial blood supply to
the hand that must be addressed. On the moat basic
level, the blood supply to the hand following radial
artery transsection relies on the presence of an ulnar
artery. In a cadaveric study of 750 upper extremities,
McCormack et al. (30) found the ulnar and radial arter-
ies to be present in all cases. The ulnar artery supplies
the hand through the superficial palmar arch, which is
either "complete,"' in the sense that it provides branches
to four fingers and the thumb, or "completed"' through
communications with the deep palmar arch. Cadaveric
studies by Coleman and Anson (9) revealed that the
ulnar artery supply to the third, fourth, and fifth digits
is rarely, if ever, compromised by anomalous patterns.
When harvesting a radial forearm flap, it is the vascular

Basilic v. Medial antebrachial cutaneous n.

FIGURE 12-5. The sensory nerve supply to the forearm skin consistently follows a course that
parallels the major subcutaneous veins. Depending on the size and orientation of the flap, either
the medial, the lateral, or both antebrachial cutaneous nerves would be incorporated.
184 CHAPTER 12
supply to the thumb and the index finger that is most
at risk by the combination of two concurrent arterial
variations. The first is an incomplete superficial arch
that does not send branches to the thumb and index
finger. The second anomaly, which must also be pre-
sent for there to be a problem of digital ischemia, is a
complete lack of communication between the superfi-
cial and deep arches. In 265 cadaveric specimens, Cole-
man and Anson found a complete superficial arch in
77.3% of cases. The coexistence of the two anomalies,
which would put the thumb at least at risk of ischemia
in the event of radial artery occlusion, occurred in 12%
of specimens. The problem of anomalous circulation to
the hand is reviewed in Chapter 17 and highlighted in
Figure 17-6.
A third source of blood supply to the hand may
arise from a persistent median artery, which has been
noted in up to 16% of cases (9). This vessel usually
joins the superficial palmar arch. It is therefore possi-
ble that this vessel may supply protective circulation to
the hand in those cases in which the thumb and index
finger are otherwise at risk following sacrifice of the
radial artery.
He den and Gylbert (19) reported an aberrant radial
artery that was encountered while raising a radial
forearm flap. A small branch of the radial artery was
found in the normal anatomic position, but the main
vessel ran a divergent course in the distal forearm
that was superficial to the extensor pollicis tendons
and entered the hand several centimeters radial to its
normal location. This anomaly has been reported to
occur in 1% of the population, and its importance is
evident when considering the implications for per-
forming an accurate Allen test as well as an effective
harvest of a radial forearm flap (19,26). Compression
of the rudimentary branch may not demonstrate an
abnormality in the palmar arch, putting the thumb at
risk for ischemia. If encountered during a dissection
of a forearm flap, it is imperative that the true aberrant
radial artery is temporarily occluded and the tourni-
quet released to assess the impact on the vascularity
of the hand.
Additional, although rare, anomalies of the radial
artery have been reported. Otsuka and Terauchi (33)
reported an aberrant dorsal course of the radial artery
that passed around Lister tubercle of the radius to enter
the hand above the extensor tendons. Small and Miller
(39) reported a radial artery that passed deep to the
pronator teres.
Anomalies of the ulnar artery are equally rare. How-
ever, Fatah et al. (12) described an ulnar artery that ran
a course superficial to the flexor muscles in the fore-
arm. The importance of recognizing such an anomaly
is essential to avoid catastrophic injury to the remaining
blood supply to the hand following harvest of the radial
forearm flap.
POTENTIAL PITFALLS
Although the radial forearm flap is a highly reliable
method of reconstruction, there are a number of donor-
site problems that may be encountered. The perfor-
mance of an accurate Allen test is the most important
consideration in avoiding the catastrophic complication
of an ischemic hand. There is one report in the literature
by Jones and O'Brien (22) in which insufficient flow to
the hand resulted from flap harvest despite a normal
Allen test result. The hand was salvaged by intraopera-
tive recognition of the problem and reconstruction of
the radial artery by an interposition vein graft. Bardsley
et al. (2) reported on a subset of 12 patients in their
total series of 100 patients who underwent radial fore-
arm flaps in whom reconstitution of the radial artery
was performed. This group constituted the earliest
patients in their series. Despite reestablishing flow in
the operating room, only six vein grafts remained pat-
ent over time. None of the patients with occluded grafts
suffered any complications. Vascular insufficiency may
also result from too tight a closure of the forearm skin or
compression from the forearm dressing. Careful obser-
vation of the hand both during and after surgery will
help to avoid such problems.
Poor take of the skin graft may result from shearing
forces of the underlying muscles as a result ofinadequate
immobilization of the hand. Failure to preserve the par-
atenon over the flexor tendons may also contribute to
problems with the skin graft.
Infection of the forearm donor site is uncommon.
However, the devastating effects of this complication
were reported by Hallock (16) who described a patient
with a frozen hand resulting from a descending sup-
purative tenosynovitis. Meticulous sterile technique
and avoidance of cross-contamination from the head
and neck by using a separate set of instruments are
absolutely imperative.
The aesthetic deformity of the skin-grafted donor site
is recognized as one of the major disadvantages of the
radial forearm free flap. A variety of different techniques
have been reported to modify the appearance of the
skin-grafted forearm defect. Direct closure of the flap
defect has been reported by the use of an ulnar transpo-
sition flap. Long fascial attachments provide a reliable
blood supply to these flaps, which can be used to close
small to medium-sized defects. In muscular individuals,
there is often a problem of closing the secondary defect
in the proximal forearm that results from using an ulnar
transposition flap. The usual technique of a V-to-Y
closure will not work, and in such cases, a skin graft can
be placed proximally while achieving full-thickness skin
coverage of the flexor tendons distally (2, 11). Poor take
of the skin graft over the distal flexor tendons can be
problematic, despite care in preserving the paratenon.

Coverage of the flexor carpi radialis tendon with tum-
over muscle flaps of the flexor pollicis longus and the
flexor digitorum superficialis was described to improve
the donor-site bed to accept a skin graft (13).
The use of tissue expanders has been reported to
achieve full-thickness coverage of the donor defect.
Masser (28) placed a tissue expander several weeks
prior to flap harvest to facilitate forearm coverage.
The expanders were placed deep to the forearm fas-
cia and deep to the radial artery. Careful monitoring
of the Doppler signal during the expansion phase was
critical to prevent occlusion of the radial artery. The
pre-expansion was believed to achieve a greater flap
surface area and to delay marginal areas that may have
been ischemic following transfer. In addition, Masser
reported reduced thickness of the forearm flaps.
As an alternative to this approach, Hallock (15)
placed tissue expanders under the residual forearm skin
at the time of harvesting a radial forearm flap. He waited
a minimum of 2 weeks to begin serial expansion. In
10 patients so treated, 5 had linear scars; the remaining
patients had a significant reduction in the size of the
skin-grafted area.
The transfer of a fascial flap alone avoids the problem
of a cutaneous donor-site defect. Fascial flaps can be
transferred and covered with a split-thickness skin graft.
To eliminate the necessity of using a separate donor
site for harvesting a skin graft, Kawashina et al. (23)
described the de-epithelialized forearm flap for resur-
facing the lining of the upper aerodigestive tract. The
forearm flap was harvested and then de-epithelialized
on a drum dermatone. The resulting split-thickness graft
was returned to the forearm, avoiding the deformity of a
second donor site for a skin graft. The de-epithelialized
flaps healed uneventfully and were rapidly covered by
an epithelial layer.
The most common neurologic problems in the fore-
arm are related to sensory loss following injury to the
superficial branches of the radial nerve and transsection
of the antebrachial cutaneous nerves. Anesthesia over
the region of the anatomic snuffbox and the thumb and
first finger can be troublesome for patients and every
effort should be taken to avoid this complication. Pain-
ful neuromas have not been a problem in my experience,
but the potential certainly exists.
The function of the hand following routine harvest of
a radial forearm flap is usually normal. The potential for
morbidity in the hand mounts when an osteocutaneous
flap is harvested. Fracture of the radius can have a sig-
nificant detrimental effect on supination, wrist flexion,
grip strength, and pinch strength ( 4). As noted previ-
ously, this potential morbidity and the poor bone stock
greatly limit the advisability of routinely using the oste-
ocutaneous flap. However, there are situations in which
a very small composite flap may be needed. Under these
circumstances, the method by which the osteotomy
RADIAL FOREARM 185
is created may be important in avoiding a significant
donor-site problem. Bardsley et al. (2) recommended
removal of as small a segment of bone as possible, and
in the process, they advised creating smooth ''boat-
shaped" bone cuts rather than right-angled ones, which
were prone to fracture. Prolonged immobilization and
serial radiographs of the healing donor site were also
advised to ensure adequate bone remodeling prior to
stressing the forearm.
Swanson et al. ( 45) tested the hypothesis that the
type of osteotomy influenced the mechanical strength
of the residual radius. They did not find a statistically
significant difference in the breaking force between a
right-angled bone cut and a beveled bone cut. There
was a significant difference in breaking strength (24%)
in the group of osteotomized radii compared with that
in intact controls. However, the authors did advise the
use of a beveled bone cut to reduce the concentration
of stress at the corners of right-angled cuts and to mini-
mize the amount of bone removed, which they believe
should not exceed one-third of the radial diameter. In
addition, they advised an above-elbow splint for 8 weeks
after surgery.
The institution of postharvest protective measures
with rigid fixation will be presented in Chapter 23
and has greatly reduced the incidence of donor-site
complications.
PREOPERATIVE MANAGEMENT
Factors, such as tissue thickness, hair-bearing skin,
and the distribution of superficial veins, can be
assessed preoperatively to plan the design of the fore-
arm flap. The flap,s dimensions and shape can usually
be accurately determined by direct laryngoscopy and
manual palpation. However, we usually wait to harvest
this flap until frozen sections on tumor margins have
been determined. The Allen test is the most important
preoperative evaluation to assess the adequacy of the
circulation to the hand through the ulnar artery. The
Allen test must be performed properly. Simultaneous
compression of the ulnar and radial arteries is applied
by the examiner while the hand is alternately opened
and closed. This pumping action causes the hand to
become pale as a result of mechanical exsanguina-
tion. The hand is then opened to a relaxed position
prior to the release of the ulnar artery. It is impor-
tant that the fingers are not held in a hyperextended
position, which can cause them to remain pale and
therefore lead to a false-positive result. Release of the
ulnar artery should cause reperfusion through a blush
of the hand within 15 to 20 seconds. Hthere is a delay
beyond this time, then this raises concern about the
ulnar circulation, and a radial forearm flap should not
be performed. In dark-skinned individuals in whom
186 CHAPTER 12
capillary blush is not easily assessed, the perfusion of
the hand can be confirmed by checking the capillary
refill of the nail bed on compression and release of the
fingernail. It is imperative that the examiner observe
the vascularity to the thumb for the reasons outlined
earlier. It is our preference to select the nondominant
arm for flap harvest. The Allen test should be repeated
to confirm the initial findings. A final check may be
performed in the operating room where a pulse oxime-
ter can be attached to the thumb to assess the wave-
form changes when the Allen test is repeated. Little et
al. (25) reported a 3% incidence of positive Allen test
findings in the general population. The reliability of
this test for screening individuals with poor ulnar cir-
culation is attested to by my experience with over 500
patients and Soutar's (41) experience with 200 radial
forearm flaps in which no cases of hand ischemia have
occurred.
On admission to the hospital for a radial forearm
flap, a bandage is placed over the donor forearm to pre-
vent anyone from using that arm for arterial or veni-
punctures. The patient must also be instructed to warn
all hospital personnel against violation of the forearm.
Patients who have been hospitalized for prolonged peri-
ods may have few patent superficial veins in their arms.
A forearm flap may be harvested in these patients based
on the venae comitantes, or perhaps more prudently, an
alternative donor site should be selected.
The patency of the radial artery is rarely an issue
except in those patients who have had a previous
indwelling radial artery catheter. Although flow in these
vessels is usually restored over time, I have encountered
a small number of cases in which this has not occurred.
The Allen test will obviously not detect an occluded
radial artery unless a reverse occlusion of the ulnar
were employed. Failure to detect a palpable pulse in the
radial artery should trigger concern or selection of an
alternative donor site. Doppler assessment will confirm
the integrity of flow through that vessel.
POSTOPERATIVE CARE
After applying a split-thickness skin graft, a volar plas-
ter splint is formed that extends from the fingers to the
antecubital fossa. A compressive wrap is then placed
over the splint, and the forearm is elevated. Immobili-
zation of the forearm is important to prevent shearing
of the muscles underneath the skin graft. It is impera-
tive that the vascularity of the hand be confirmed after
releasing the tourniquet. The circulation to the thumb
must be assessed to ensure once again that the collat-
eral circulation through the ulnar artery provides suf-
ficient vascularity. The dressing and the volar splint are
left in place for approximately 7 days following surgery.
During this time, the forearm is elevated, and monitor-
ing of the vascularity to the hand is continued to be
certain that the bandage does not cause compression of
the circulation as a result of the postoperative edema.
On the seventh day following surgery, the dressing is
removed, and the skin graft is observed. A conforming
elastic stocking is then used to assist in wound healing
and reduce edema in the hand resulting from the inter-
ruption of the venous and lymphatic supplies. Copious
padding must be placed in the operating room prior
to the compressive dressing in order to avoid pressure
ulceration over prominent points. The development of
pain in the wrist or hand under the volar splint should
trigger an urgent examination to ensure that pressure
ulceration has not occurred.
 RADIAL FOREARM 187

Radial Foreann Fasciocutaneous Flap

FIGURE 12-6. The design of the radial fore-

arm flap begins by outlining the path of the
dominant subcutaneous veins and the palpable
pulse of the radial artery. The paths of the
cephalic vein and the radial artery have been
drawn on the left forearm. In this dissection,
the approximate topographical anatomy of the
sensory nerves is outlined in orange. The lateral
antebrachial cutaneous nerve runs adjacentto
the cephalic vein, and the approximate course
of the superficial branches of the radial nerve
is shown as the branches terminate on the
dorsum of the hand.
FIGURE 127. A rectangular radial forearm
flap has been outlined on the distal forearm.
The axis of this flap is centered on the radial
artery and the cephalic vein. A curvilinear
dotted line indicates the incision in the proximal
forearm where skin flaps will be elevated to
provide access to the proximal portion of the
neurovascular pedicle. A larger skin paddle
can be harvested that extends proximally to the
antecubital fossa and virtually encompasses
the entire circumference of the forearm, except
for a bridge of skin along the ulnar aspect.

FIGURE 12-8. The dissection begins distally

after exsanguination of the forearm through the
use of an elastic bandage and raising the tour-
niquet to approximately 250 mm Hg. The distal
skin incision is made to gain exposure of the
radial artery and its adjacent venae comitantes.
188 CHAPTER 1Z

Radial Foreann Fasciocutaneous Flap

FIGURE 129. The radial artery is then ligated

and divided.

FIGURE 12-10. Dissection may begin either

from the ulnar or from the radial direction. In
this particular dissection, the skin flap has
been elevated, starting from the radial aspect.
The distal portion ofthe cephalic vein must be
ligated and transsected.

FIGURE 1211. The skin flap has been elevated

with the deep fascia to the level of the lateral
intermuscular septum marked by the border of
the brachioradialis. The superficial branches of
the radial nerve are isolated and preserved to
maintain sensation to the dorsum of the hand.
The dissection of the ra dia I nerve requires that
the subfascial plane of dissection be broken.
 RADIAL FOREARM 189

Radial Foreann Fasciocutaneous Flap

FIGURE 12-12. The ulnar dissection of the flap

is carried out in a subfascial plane, elevating
the flap off the tendons of the muscles in the
flexor compartment. It is imperative to maintain
the integrity of the paratenon when perform-
ing this dissection. The forearm flap has been
elevated in an ulnar direction to the border of
the flexor carpi radialis, which marks the posi-
tion of the intermuscular septum.

FIGURE 12-13. Skin flaps are elevated in

the dissection proximal to the skin paddle by
making an incision along the dotted line. The
skin flaps are elevated in a subcutaneous plane
to preserve the integrity of the subcutaneous
veins and the adjacent sensory nerves. In this
dissection, there are two subcutaneous veins,
and the lateral antebrachial cutaneous nerve
(arrowhead) is demonstrated lying adjacent to
the cephalic vein.

FIGURE 1214. Atthis point in the procedure,

the radial artery (arrowhead) is dissected
distally to proximally by transsecting and
cauterizing the deeper branches that supply the
muscles of the forearm and the radius.
190 CHAPTER 1Z

Radial Foreann Fasciocutaneous Flap

FIGURE 12-15. Dissection along the intermus-

cular septum is continued proximally until the
point of overlap (arrowhead) of the brachiora-
dialis and the flexor carpi radialis.

FIGURE 12-16. In the proximal forearm,

the radial artery courses deep to the
brachioradialis. Therefore, it is apparent that
the primary arterial inflow to the skin com-
ponent of the forearm flap arises through its
fasciocutaneous perforators, which are given
off in the distal third of the forearm.
FIGURE 12-11. Exposure of the proximal radial
artery and the venae comitantes is achieved by
separating the brachioradialis from the flexor
carpi radialis. The radial artery may be traced all
the way to the brachial artery. The forearm flap
is then reperfused by releasing the tourniquet
The vascularity is ensured through observation
of the color and dennal bleeding. In addition,
it is prudent to assess the vascularity of the
hand and in particular the thumb. At this point,
hemostasis is obtained, and the flap is prepared
for harvest when the recipient site is ready. If
a proximal monitor paddle has been designed
for monitoring of a buried flap, it is important to
assess the vascularity ofthe monitor paddle to
ensure that it will be reliable.
 RADIAL FOREARM 191

Radial Foreann Fasciocutaneous Flap

FIGURE 1218. The donor site is closed by

reapproximating 1he proximal skin flaps as
shown. The remainder of the defect is closed
wi1h a split-thickness skin graft.

FIGURE 12-19. As noted previously, it is imper-

ative to maintain the thin paratenon layer over
the tendons in the distal forearm to facilitate
skin grafting of this donor site.

FIGURE 12-20. The radial forearm flap is

shown wi1h 1he radial artery, the cephalic vein,
and the lateral antebrachial cutaneous nerve.
192 CHAPTER 1Z

Beavertail Modification of the Radial Foreann Flap

FIGURE 1221. The beavertail modification is

shown with the beavertail adipofascial paddle
indicated in green {yellow arrowj.

FIGURE 1222. The ulnar (upper) skin flap

is elevated in a subdermal plane to maintain
the maxi mum amount of fat and fascia in the
beavertail portion of the flap.

FIGURE 1223. The ulnar and radial skin flaps

have been elevated. Dissection ofthe radial
forearm flap proceeds, as shown in the prior
description with the inclusion of the large
beavertail portion of fat and fascia.
 RADIAL FOREARM 193

Beavertail Modification of the Radial Forearm Flap

FIGURE 12-24. The proximal portion of the flap

is shown with the beavertail component high-
lighted with a dashed yellow line. The antecubi-
tal vein is marked with a blue pin.

FIGURE 12-25. The proximal neurovascular

pedicle has been dissected and is shown with
a red pin to demonstrate the radial artery, the
blue pin indicates the antecubital vein, and the
yellow pin highlights the position of the ante-
brachial cutaneous nerve.

FIGURE 12-2&. The flap has been harvested

with a large beavertail paddle. The radial artery
is indicated by the red pin. The cephalic vein
and antebrachial cutaneous nerve are seen
lying in close proximity as indicated by the
blue and yellow pins. The communicating vein
between the deep and superficial systems is
indicated by the middle blue pin (green arrow).
194 CHAPTER 1Z

Beavertail Modification of the Radial Foreann Flap

FIGURE 1Z-Z7. In order to mobilize the beaver-

tail, the radial artery pedicle is dissected free of
the surrounding tissue, taking care not to divide
the perforatars to the skin paddle partion of the
flap.

FIGURE 1Z-Z8. With the radial artery and

venae comitantes having been separated from
the flap, the attention is turned to performing
the same maneuver for the venous supply to the
flap.

FIGURE 1Z-Z9. The cephalic vein is released

fram the beavertail to the level of the skin
paddle
 RADIAL FOREARM 195

Beavertail Modification of the Radial Forearm Flap

FIGURE 12-30. The flap is shown with the bea-

vertail portion dissected free of the pedicle. The
lateral antebrachial cutaneous nerve (yellow
pin), cephalic vein (blue pin), communicating
vein (green arrow), and radial artery (red pin)
are shown.

FIGURE 12-31. The beavertail portion has

been rolled to provide bulk to the distal portion
of the flap. This orientation would commonly
be used for base of tongue reconstruction or
other reconstructions where carefully placed
bulk is required to enhance the functional and
aesthetic result.
196 CHAPTER 1Z

Insetting of the Radial Forearm Flap for Pharyngoesophageal Reconstruction

Longitudinal fla orientation

FIGURE 12-32. Harvest of a longitudinally

oriented radial forearm flap for circumferen-
tial pharyngoesophageal reconstruction is
performed with a design as sllown in (A) and
(B), where the length of the defect from the
esophagus to the base of tongue is shown by
the yellow arrow. The width of1he proximal por-
tion of the flap is wider than tile distal flap due
to the larger caliber of the pharyngeal lumen
compared to the esophagus. A B

FIGURE 12-33. The radial forearm flap

has been harvested with a distal extension
(blue arrow) to be incorporated in the dis-
tal anastomosis in a tongue and groove..
configuration.

FIGURE 12-34. Following a total laryngophar-

yngectomy, the tracheostome (blue arrow) is
formalized and the esophageal stump is pre-
pared for anastomosis (yellowsrrow).
 RADIAL FOREARM 197

Insetting of the Radial Forearm Flap for Pharyngoesophageal Reconstruction

Longitudinal fla orientation

FIGURE 12-35. A cut is made in the esopha-

geal stump (blue srrow) in order to allow
insertion of the V-shaped extension of the flap,
designed to interrupt the distal anastomosis.

FIGURE 12-36. Insetting of the flap is per-

formed with the distal anastomosis performed
first followed by the proximal repair, and then,
tubing of the flap with a longitudinal suture line.
The monitor paddle is exteriorized in a suture
line in the neck.
 198 CHAPTER 1Z

Inset of the Radial Forearm Cutaneous Flap for Reconstruction of the

Circumferential Pharyngoesophageal Segment
Horizontal flap orientation
----
FIGURE 12-37. Harvest of the radial forearm
flap for circumferential pharyngoesophageal
reconstructian is perfarmed with a flap that
is designed with the width detennined by the
distance between the esophagus and the base
of tongue (yellow arrow). This design usually
requires that the flap extends beyond the
volar surface of the forearm. A monitor flap is
designed in the proximal portion of the forearm
(green arrow).

FIGURE 12-38. The flap has been harvested

with the portion of the flap slated for the proxi-
mal anastomosis indicated by line A-B. The
esophageal portion is shorter and indicated by
line C-O. Extensions ofthe radial forearm flap
to be inset into both the esophageal and the
pharyngeal repairs are demanstrated by the
green and blue arrows, respectively.

FIGURE 12-39. Similar to the cut in the proxi-

mal esophagus, an apening is created in the
pasterior wall of the pharynx for inset of the
V-sha ped partion of the radial fares rm flap.
 RADIAL FOREARM 199

Inset of the Radial Forearm Cutaneous Flap for Reconstruction of the

Circumferential Pharyngoesophageal Segment
Hotizontsl tis otientstion

FIGURE 12-40. The flap is inset into the

pharynx as well as the esophagus.

FIGURE 12-41. Once the posterior walls of the

proximal and distal repairs are completed, the
flap is tubed upon itself.

FIGURE 12-42. The completion of the skin-

lined tube is carried out over a salivary bypass
tube. Meticulous suturing is required to avoid a
sa Iiva ry fistula.
ZOO CHAPTER 12

Inset of the Radial Forearm Cutaneous Flap for Reconstruction of the

Circumferential Pharyngoesophageal Segment
Hotizontal flap otientation

FIGURE 12-43. The repair has been completed

and the flap revascularization is now set to be
performed.

FIGURE 12-44. With this flap design the vas-

cular anastomoses are usually performed to the
transverse cervical artery and the transverse
cervical or external jugular veins. The monitor
flap is readily placed in the apron incision for
easy access.
 RADIAL FOREARM Z01

Inset of the Radial Forearm Flap for Lengthening of the Mediastinal Trachea

FIGURE 12-45. A radial forearm flap with

similar shape to that of the horizontal orienta-
tion noted above is designed. A monitor is not
required since the flap will be visible in the
neck.

FIGURE 12-46. A radial forearm flap is

designed in a fashion so as to create a 'b.Jbed
funnel-shaped structure that bridges the gap
between the cut end of the mediastinal tra-
chea and the cervical skin. This 'b.Jbed flap is
designed with the tracheal end matching up
with the desired circumference of the cut end
of the trachea and the opposite end is Ion ger
so as to create a funnel, with the opening in the
neck considerably larger. The length of the flap
reflects the distance that the flap will have to
traverse in order to comfortably form a conduit
(yellow arrawj. A subcutaneous extension has
been harvested to allow a double-layer closure
of the tracheal anastomosis (blue arrow).

FIGURE 12-47. As is often the case with

stomal recurrences, the pharyngoesophageal
repair has been accomplished with a separate
flap. In this case, a free jejunum was utilized.
The manubrium and the clavicular heads have
been resected with exposure of the great ves-
sels in the mediastinum. The posterior wall af
the tracheal repair has been completed.
202 CHAPTER 1Z

Inset of the Radial Flap for Lengthening of the Mediastinal Trachea

FIGURE 12-41. The proximity of the innominate

artery and vein provides impetus for a secure
anastomosis of the skin flap to the tracheal
stump. The development of a leak in this loca
tion will result in seeding of the mediastinum
and a resultant mediastinitis. A double-layer
repair is performed with the skin to mucosal
anastomosis (blue arrow) and then the fascial
subcutaneous extension is sutured to the outer
wall of the trachea (green arrow).

FIGURE 1249. The skin to mucosal anastomo-

sis has been completed and the vertical limb
remains for closure. A circumferential two-layer
anastomosis has been performed.
 RADIAL FOREARM 203

Inset of the Radial Forearm Flap for Lengthening of the Mediastinal Trachea

FIGURE 12-50. The vertical skin to skin closure

has been completed with everting sutures. The
pedicle exits to the right side.

FIGURE 12-51. The position of the upper

opening to the neotrachea can be altered to
fit the contour of the cervical skin and the
remaining bony architecture of the upper
sternum and clavicles.
204 CHAPTER 12
REFERENCES
1. BairdW, Womom I, Culbertson J: Forehead reconstruc-
tion with a modified radial forearm flap: a case report.
J Reconstr Microsurg 1988;4:363.
2. Bardsley A, Soutar D, Elliot D, Batchelor A: Reducing
morbidity in the radial forearm flap donor site. Pfast
Reconm Surg 1990;86:287.
3. Bhathena H, Kavarana N: Bipaddled retrograde radial
extended forearm flap with microarterial anastomoses for
reconstruction in oral cancer. BrJ Plast Surg 1988;41 :354.
4. Boorman J, Brown J, Sykes P: Morbidity in the forearm
flap donor arm. Br J Pfast Surg 1982;40:207.
5. Boorman JG, Green MF: A split Chinese forearm flap for
simultaneous oral lining and skin cover. Br J Pfast Surg
1986;39: 179.
6. Chantrain G, Deraemaecker R, Andry G, Dor P: Wide
vertical hemipharyngolaryngectomy with immediate glot-
tic and pharyngeal reconstruction using a radial forearm
free flap: preliminary results. Laryngoscope 1991;101:869.
7. Chicarelli Z, Ariyan S, Cuono C: Free radial forearm
flap versatility for the head and neck and lower extremity.
J Reconstr Microsurg 1986;2:221.
8. Chicarelli Z, Ariyan S, Cuono C: Single-stage repair of
complex scalp and cranial defects with the free radial
forearm flap. Pfast Reconstr Surg 1986;77:577.
9. Coleman T, Anson B: Arterial patterns in the hand based
upon a study of 650 specimens. Surg Gynecol Obstet
1961;ll3:409.
10. Cormack G, Duncan MJ, Lamberty B:The blood supply
of the bone component of the compound osteocutane-
ous radial artery forearm flap--an anatomical study. Br J
Plan Surg 1986;39: 173.
11. Elliot D, Bardsley F, Batchelor A. Soutar D. Direct clo-
sure of the radial forearm flap donor site. Br J Pfast Surg
1988;41 :358.
12. Fatah M, Nancarrow J, Murray D: Raising the radial
artery forearm flap: the superficial ulnar artery "trap." Br
J Pfast Surg 1985;38:394.
13. Fenton 0, Roberts J: Improving the donor site of the
radial forearm flap. Br J Pfast Surg 1985;38:504.
14. Hagen R: Laryngoplasty with a radialis pedicle flap from
the forearm: a surgical procedure for voice rehabilitation
after total laryngectomy. Am J Otolaryngol1990; 11 :85.
15. Hallock G: Refinement of the radial forearm flap donor
site using skin expansion. Plast Reconstr Surg 1988;81:21.
16. Hallock G: Complication of the free flap donor site from
a community hospital perspective. J Reconstr Microsurg
1991;7:331.
17. Harii K, Ebihara S, Ono I, Saito H, Terui S, Thkato T:
Pharyngoesophageal reconstruction using a fabricated
forearm free flap. Pfast Reconstr Surg 1985;75:463-476.
18. Hatoko M, Harashina T, Inoue T, Thnaka I, Imai K:
Reconstruction of palate with radial forearm flap: a report
of 3 cases. Br J Plan Surg 1990;43:350.
19. Heden P, Gylbert L: Anomaly of the radial artery
encountered during elevation of the radial forearm flap.
J Microsurg 1990;6:139.
20. Ismail TI: The free fascial forearm flap. Microsurgery
1989;10:155-160.
21. JinY, GuanW. ShiT, QuiwanY, Xe L, ChangT: Reversed
island forearm fascial flap in hand surgery. Ann Pfast Surg
1985;15:340.
22. Jones B, O'Brien C: Acute ischemia of the hand resulting
from elevation of a radial forearm flap. Br J Pfast Surg
1985;38:396.
23. Kawashina T, Harii K. Ono I, Ebihara S, Joshizumi T:
Intraoral and oropharyngeal reconstruction using a deep-
ithelialized forearm flap. Head Neck 1989;11:358.
24. Lamberty B, Cormack G: The forearm angiosomes. Br
J Pfast Surg 1982;35:420.
25. Little J, Zylstra P, West J, May J: Circulatory patterns in
the normal hand. Br J Plast Surg 1973;60:652.
26. Loetzke HH, Kleinau W: Gleichzeitiges Vorkommen der
Aa. brachialis superficialis, radialis und antebrachialis
dorsalis superficialis sowie deren Aufzweigungen. Anat
Anz 1968;122:137-141.
27. Martin IC, Brown AE: Free vascularized fascial flap in
oral cavity reconstruction. Head Neck 1994;16:45.
28. Masser M: The pre-expanded radial free flap. Plast
Reconstr Surg 1990;86:295.
29. Matthews RN, Fatah F, Davies DM, Eyre J, Hodge RA,
Walsh-Waring GP: Experience with the radial forearm
flap in 14 cases. Scand J Pfast Reconstr Surg 1984;18:303.
30. McCormack L, Cauldwell E, Anson B: Brachial and
antebrachial arterial patterns, a study of 7 50 extremities.
Surg Gynecol Obsrez 1953;96:43.
31. Moscoso J, Keller J, Genden E, et al.: Vascularized bone
flaps in oromandibular reconstruction. A comparative
study of bone stock from various donor sites to assess
suitability for enosseous dental implants.Arch Otolaryngol
Head Neck Surg 1994;120:36.
32. Nakayama Y, Soeda S, Iino T: A radial forearm flap
based on an extended dissection of the cephalic vein.
The longest venous pedicle? Case report. Br J Pfast Surg
1986;39:454.
33. Otsuka T, Terauchi M: An anomaly of the radial
artery-relevance for the forearm flap. Br J Plast Surg
1991;44:390.
34. Reid CD, Moss AIR: One stage repair with vascularized
tendon grafts in a dorsal hand injury using the "Chinese"
forearm flap. Br J Pfast Surg 1983;36:473.
35. Rieger JM, Zalmanowitz JG, Li SY, et al.: Functional out-
comes after surgical reconstruction of the base of tongue
using the radial forearm free flap in patients with oropha-
ryngeal carcinoma. Head Neck 2007;29(11):1024-1032.
36. Sadove R, Luce E, McGrath P: Reconstruction of
the lower lip and chin with the composite radial
forearm-palmaris longus free flap. Plast Reconstr Surg
1991;88:209.

37. Sanger J, Ye Z, Yousif N, Matloub H: The brachioradialis
forearm flap: anatomy and clinical application. Presented
at the 8th Annual Meeting of the American Society
for Reconstructive Microsurgery, Scottsdale, Arizona,
November 8, 1992.
38. Seikaly H, Rieger J, O'Connell D,Ansari K, Alqahtani K,
Harris JR: Beavertail modification of the radial forearm
free flap in base of tongue reconstruction: technique
and functional outcomes. Head Neck 2009;31(2):213-
219.
39. Small J, Miller R: The radial artery forearm flap: an
anomaly of the radial artery. Br J Plast Surg 1985;38:50 1.
40. Song R, Gao Y, SongY, Yu Y, SongY: The forearm flap.
Clin Plast Surg 1982;9:21-26.
41. Soutar D: Radial forearm flaps. In: Baker S, ed. Micro-
surgical Reconstruction of the Head and Neck. Ne'W' York:
Churchill Livingstone; 1989.
42. Soutar DS, McGregor IA: The radial forearm flap in
intraoral reconstruction: the experience of 60 consecu-
tive cases. Plast Recomtr Surg 1986;78: 1.
43. Soutar DS, Scheker LR, Tanner NSB, McGregor IA: The
radial forearm flap: a versatile method for intraoral recon-
struction. Br J Plast Surg 1983;36: 1.
44. Soutar DS, Widdowson WP: Immediate reconstruction
of the mandible using a vascularized segment of radius.
Head Neck 1986;8:232.
45. Swanson E, Boyd J, Mulholland R: The radial forearm
flap: a biomechanical study of the osteotomized ramus.
Plast Reconstr Surg 1990;85:267.
46. Tahara S, Susuki T: Eye socket reconstruction with free
radial forearm flap. Ann Plast Surg 1989;23: 112.
4 7. Takada K, Sugata T, Yoshiga K, Miyamoto Y: Total upper
lip reconstruction using a free radial forearm flap incor-
porating the brachioradialis muscle: report of a case.
J Oral MaxU/tJjac Surg 1987;45:959.
RADIAL FOREARM 205
48. Taylor GI, Ham FJ: The free vascularized nerve
graft. A further experimental and clinical applica-
tion of microvascular techniques. Plast Reconstr Surg
1976;57:413.
49. Thoma A, Archibald S, Jackson S, Young J: Surgical
patterns of venous drainage of the free forearm flap
in head and neck reconstruction. Plast Reconstr Surg
1994;93:54.
50. Timmons M: The vascular basis of the radial forearm
flap. Plast Recomtr Surg 1986;77:80.
51. Urken ML: Composite free flaps in oromandibular
reconstruction: review of the literature. Arch Otolaryngol
Head Neck Surg 1991;117:724.
52. Urken MI.., Biller HF: A ne'W' bilobed design for the sen-
sate radial forearm flap to preserve tongue mobility fol-
lowing significant g]ossectomy. Arch Otolaryngol Head
Neck Surg 1994;120:26-31.
53. Urken MI.., Futran N, Moscoso J, Biller HF: A modified
design of the buried radial forearm free flap to exteriorize
a monitoring segment. Arch Otolaryngol Head Neck Surg
1994;120(11):1233-1239.
54. Urken MI.., Weinberg H, Vickery C, Biller HF: The
neurofasciocutaneous radial forearm flap in head and
neck reconstruction: a preliminary report. Laryngoscope
1990; 100:161.
55. Vaughan ED: The radial forearm free flap in orofacial
reconstruction. J Craniomaxillojac Surg 1990; 18:2.
56. Wakrhouse N, Moss A, Townsend Pl.: Lower limb sal-
vage using an extended free radial forearm flap. Br J Plast
Surg 1984;37:394.
57. Yang G, Chen B, Gao Y, et al.: Forearm free skin flap
transplantation. NatlMedJ China 1981;61:139.
58. Yousif NJ, Ye Z: Analysis of cutaneous perfusion: an
aid to lower extremity reconstruction. Clin Plast Surg
1991;18:559.
With the expansion of new donor sites for cutaneous
and fasciocutaneous free flaps that has occurred during
the past three decades of free tissue transfer, the upper
arm was a logical choice for harvest of well-vascularized,
apendable, intermediate thickness skin. The upper arm
has numeroua advantages to harvest of skin relative to
the forearm, not the least of which is the ability to place
the scar in a location that can be camouflaged with even
a short sleeve shirt. A variety of different flaps have been
described from this location. The medial arm flap, how-
ever, was found to have an unreliable vascularity. The
variability in the size of the nutrient septocutaneous per-
forators arising from the superior ulnar collateral artery
along the medial intermuscular (IM) septum (3) led to
an unpredictable vascular system upon which to base a
free flap. The deltoid flap was introduced by Franklin (4)
ZO&
and is based on the arteria deltoidea subcutanea, a con-
stant branch of the posterior circum1la: humeral artery.
However, a deep dissection with limited exposure is neo-
essary to obtain an adequate pedicle length.The vascular
pedicle also rests dangerously close to the u.illary nerve.
In 1982, Song et al. (17) introduced the lateral arm
fasciocutaneous flap, which has become a uaeful free
flap in head and neck and extremity reconstruction.
Prior to the introduction of the anterolateral thigh flap,
this donor site was often considered the second line flap
for harvest of a fasciocutaneous flap. The lateral ann
flap may be transferred with a segment of the humerus,
triceps tendon, and two nerves, one of which can serve
as a sensory supply and the other as a vasculari2:ed nerve
graft. It has similarities to the radial forearm flap but
offers the distinct advantages that its nutrient artery,
 LATERAL ARM 207

the profunda braclili, is not essential to the vascularity
of the distal upper extremity, and the donor defect can
most often be closed with a linear scar.
FLAP DESIGN AND UTILIZATION
The territory of the lateral ann flap has been investi-
gated through dye-injection studies. The maximum
dimensions of the cutaneous paddle have not been
determined; however, flaps as large as 18 x 11 em have
been reported (14). Rivet et al. (14) descnoed a "zone
of security" that extended 12 em proximal to the lateral
epicondyle and included one-third of the circumference
of the arm. They advised that flaps should incorporate
this zone to ensure vascularity and a successful recon-
struction. Katsaros et al. (7) reported that dye-injection
studies yielded areas of staining that ranged from 8 x
10 em to 15 x 14 em. In a review of 150 lateral ann flaps,
Katsaros et al. (8) reported the successful transfer ofskin
flaps that extended over a longitudinal direction both
10 em proximal to the deltoid insertion and 10 em distal
to the lateral epicondyle. These authors also speculated
that the profunda bracbii pedicle could support a com-
plete tube of skin from the shoulder to the midforearm.
In most cases, the width of the harvested skin is limited
to 6 to 8 em, or one-third of the arm's circumference, to
allow primary closure. However, larger flaps have been
harvested, with a skin graft placed over the donor site.
Alternatively, donor site closure following harvest of
larger flaps can be accomplished by making a longitu-
dinal counterincision along the medial surface of the
upper arm where placement of a skin graft is less notice-
able than along the lateral surface of the upper arm.
The axis on which the flap is usually designed is a
line drawn from the insertion of the deltoid to the lateral
epicondyle. Alternatively, a line connecting the tip of the
acromion and the lateral epicondyle has been advocated
(Fig. 13-1) (20).
The blood supply to the skin of the lateral ann flap
is derived from a series of four to five septocutaneous
perforators that arise from the posterior branch of the
radial collateral artery in the lateral 1M septum. Katsa-
ros et al. (7) described a technique to achieve a wider
segment of skin, by harvesting a long flap and dividing it
transversely, as long as adequate perforators are present
to perfuse both the proximal and distal portions. The
additional width was achieved by folding the distal seg-
ment so that it lay adjacent to the proximal one, thereby
doubling the width while still achieving primary closure.
Kuek and Chuan (9) investigated the distal limits of
the skin paddle through eosin injections and found that
the area of staining extended an average of7 .9 em (range
of 4.5 to 10.0 em) distal to the lateral epicondyle. The
additional length of the flap not only allows the flap to be
used for wider defects by folding it on itself but also pel'-
mits a distal skin paddle to be harvested that effectively
lengthens the vascular pedicle. In addition, the skin of

FIGURE 13-1. The topographical anatomy of the lateral arm flap is illustrated. Although a line
drawn between the deltoid insertion and the lateral epicondyle is classically used as the axis of
the flap, the lateral 1M septum is actually located 1 to 2 em posteriorly to that line. The central
axis of the flap should be adjusted for this as shown.
208 CHAPTER 13
the distal portion of the upper arm tends to be thinner
than the skin in the more proximal portion of this region.
The lateral arm flap may be harvested as a fascial
flap or as a fasciocutaneous flap. The use of a vascu-
larized fascial flap allows a much larger surface area of
tissue to be harvested without having an impact on the
primary closure of the donor site. A split-thickness skin
graft may be applied to the fascial flap to achieve epithe-
lial coverage. This is also an effective means to harvest
thin tissue from this donor site in patients who have a
thicker adipose layer in this region. Large segments of
well-vascularized fascia measuring 12 x 9 em have been
harvested while still achieving primary donor site do-
sure. Small islands of skin may be included to facilitate
postoperative monitoring (21). The fascial subcutane-
ous free flap has been used to augment contour defects
of the maxillofacial region (18).
This donor site usually provides a layer of adipose tis-
sue that is intermediate in thickness between that of the
radial forearm flap and the scapular flap. The body habi-
tus of the patient will influence the relative thickness of a
flap from this donor site compared to that of the antero-
lateral thigh. Yousif et al. (21) reported detailed descrip-
tions of the fascial envelope that surrounds the triceps,
brachialis, and brachioradialis. Portions of this layer fuse
to form the IM septum. The superficial layer of this enve-
lope is continuous with the fascial sheath that covers the
entire arm. The two layers are separated by adipose tis-
sue. The fascia anterior to the IM septum averages 0.41
mm in thickness compared with the 0.21 mm average
thickness of fascia posterior to the septum.
Shenaq (16) reported an alternative solution to the lim-
ited dimensions of the lateral free flap by using pretrans-
fer tissue expansion. He was able to harvest an 11 x 18 em
flap from a child's arm and still achieve primary closure.
The ability to tailor donor site properties through staged,
pretransfer expansion is a technique with significant
potential that has not been extensively explored.
An osteocutaneous flap may be harvested by includ-
ing a segment of humerus measuring 1 x 10 em. Septal
perforators extend to the periosteum in a manner similar
to the blood supply to the radius in an osteocutaneous
radial forearm flap. A muscular cuff of triceps and bra-
chioradialis is left attached to the lateral 1M septum to
protect the blood supply (8, 10).This segment ofbone has
been used in mandibular reconstruction, but its limited
bone stock imposes restrictions on the capacity to insert
endosteal implants for dental rehabilitation. Although
limited in dimensions because of concern about patho-
logic fracture of the residual humerus, this segment of
bone may be useful in midface reconstruction.
The posterior cutaneous nerve of the arm (PCNA)
and the posterior cutaneous nerve of the forearm
(PCNF) provide the potential for reneurotized lateral
arm flaps. Sensation can be restored to the transferred
skin by anastomosing the PCNA to a suitable recipient
nerve in the head and neck (19). Matloub et al. (11)
reported on six patients who underwent sensate lat-
eral arm flap restoration of the oral cavity. Two patients
who underwent reconstruction of partial glossectomy
defects were reportedly able to differentiate light and
deep touch and hot and cold stimuli. The PCNF has
been described by Rivet et al. (14) as a "nerve in tran-
sit." This nerve travels in the 1M septum to supply sen-
sation to the skin of the lateral forearm. It receives its
blood supply from the branches of the posterior radial
collateral artery (PRCA) and, therefore, may be used as
a vascularized nerve graft. Katsaros et al. (8) reported
using this nerve to bridge facial nerve gaps in four cases.
Katsaros et al. (7) reported harvesting a segment
of vascularized triceps tendon for use in extremity
reconstruction. The use of this tissue in head and neck
reconstruction is limited.
NEUROVASCULAR ANATOMY
The profunda brachii artery is the largest branch of the
brachial artery in the arm. It runs a course on the pos-
terior aspect of the arm that parallels the radial nerve
as it spimls around the humerus in a medial to lateml
direction. The profunda brachii divides into two termi-
nal branches (Fig. 13-2). The nomenclature of these
branches is somewhat confusing. The PRCA, which
is the main nutrient artery of the lateral arm flap, has
also been referred to as the middle collateral artery. This
vessel passes through the lateral 1M septum and anas-
tomoses with the interosseous recurrent artery. This
"flow-through" system of the PRCA to the interosseous
recurrent artery is the anatomic basis for the reverse-flow
lateml upper arm flap that has been used for coverage
of the elbow region. In this flap, the PRCA is ligated
proximally, and flow to the lateral arm flap is achieved
through the anastomotic channels of the interosseous
recurrent artery (2). The anterior radial collateml artery,
which has also been referred to as the radial collateml
artery, runs a divergent and more anterior course along
with the radial nerve between the origins of the brachialis
and brachioradialis muscles. The anterior radial collat-
eral artery anastomoses with the radial recurrent artery.
In the classic description, the profunda brachii also sup-
plies the main nutrient artery ofthe humerus, the deltoid,
and the three heads of the triceps muscles (Fig. 13-3).
The average diameter of the profunda brachii was
found to be 2.45 mm (range, l.7 5 to 2. 7 mm) at a dis-
tance of 1 em below its origin from the brachial artery
(14). In the region of the deltoid insertion, where it
enters the lateral IM septum, the artery has an average
diameter of 1.55 mm (range, 1.25 to 1.75 mm) (14).
Moffett et al. (12) described a technique to lengthen
the vascular pedicle by 6 to 8 em by extending the dis-
section proximally between the lateral and long heads
 LATERAL ARM 209

A
Biceps brachii m.

Anterior radial collateral a.

Profunda brachii a .
PCNA
Posterior radial collateral a.
c
Brachialis m.

Profunda brachii a .

Radial n.

Anterior radial collateral a.

Posterior radial collateral a.

Triceps brachii m.
lateral head
Brachioradialis m. Lateral 1M septum

FIGURE 13-2. A:. Anatomy of lateral arm musculature. The lateral 1M septum is located between
the triceps posteriorly and the brachia lis and the brachioradialis anteriorly. The actual position
of the lateral 1M septum can be seen to lie 1to 2 em behind the deltoid insertion. B: The profunda
brachii artery arises from the brachial artery and winds its way in the spinal groove along with
the radial nerve. Splitting of the long and the lateral heads of the triceps provides extended
exposure of the neurovascular pedicle. As the arterial pedicle descends in the upper arm, the
profunda brachii divides into the anterior and posterior radial collateral arteries. The anterior
radial collateral artery runs an anterior course with the radial nerve between the insertions of
the brachioradialis and the brachialis. C: Aclose-up view of the neurovascular anatomy of the
lateral 1M septum. The radial nerve and the anterior branch of the radial collateral artery are
seen diverging anteriorly between the brachial is and brachioradialis. The posterior branch of the
radial collateral artery supplies the lateral arm flap; the PCNA provides sensation to the flap.
210 CHAPTER13
Posterior cutaneous n.
of arm
PRCA and PRCV
Posterior cutaneous n.
of the forearm
Triceps brachii
lateral head
FIGURE 13-3. Cross-sectional anatomy of the upper arm reveals the lateral 1M septum with
the neurovascular pedicle running in close proximity to the humerus. PRCA, posterior radial col-
lateral artery; PRCV: posterior radial collateral vein.
of the triceps muscle. In this technique, the standard
dissection is performed until the fibers of the lateral
head of the triceps limit further dissection along the
spiral groove. A tunnel is created underneath the tri-
ceps insertion by working both from below and above
through the exposure gained by splitting the lateral
and long heads of the triceps. The takeoff of the pro-
funda brachii from the brachial artery can usually be
exposed by this approach. The authors caution that
the muscular branches from the radial nerve to the
triceps muscle must be identified and preserved.
They tested triceps strength following the extended
approach in four patients at 3- and 6-month intervals
(12). There was a slight deficit in both extension and
flexion relative to the contralateral, nonoperated arm.
This slight discrepancy could be attributed to postop-
erative disuse of the operated arm and to the fact that
the flaps were harvested from the nondominant arm.
The lateral arm flap has both a superficial and deep
venous system. The superficial system drains through the
cephalic vein; the deep system drains through the paired
venae comitantes, which are about 2.5 mm in diameter
Triceps brachii
medial head
Triceps brachii
long head
(8). Inoue and Fujino (6) reportedly transferred a lateral
arm flap based on an extended dissection of the cephalic
vein, without its interruption, while performing a conven-
tional microarterial anastomosis to a recipient artery in
the neck. This flap was used to resw:face a defect in the
temporal region. Nakayama et al. (13) described a similar
technique for a radial forearm flap in which the cephalic
vein was dissected to the level of the clavicle, with a sepa-
rate arterial anastomosis performed for the radial artery.
There are two sensory nerves that course through the
lateral IM septum. The nomenclature of these sensory
nerves in the literature is also confusing. The nerve that
supplies sensation to the skin of the lateral arm flap is the
PCNA, a branch of the radial nerve. The PCNA ramifies
into four to five fascicles within the subcutaneous tissue
(7).1llis nerve has also been referred to as the inferior
lateral brac::hial cutaneous nerve, the upper branch of the
posterior antebrachial cutaneous nerve (5), or the lower
lateral cutaneous nerve of the arm (15). The PCNF,
which runs through the septum en route to the forearm,
does not supply sensation to the lateral arm fiap.As noted
previously, this nerve can be used as a vascularized nerve
 LATERAL ARM 211

Brachioradialis m .

(PCNA)

Posterior cutaneous n. of the forearm

(PCNF)

FIGURE 13-4. The lateral arm flap has been elevated from an anterior approach to show the
PCNA ramifying in the subcutaneous tissue of the flap. This nerve can be traced proximally to
provide additional length for anastomosis to a suitable recipient nerve in the head and neck.
The PCNF. also a branch of the radial nerve, provides no sensation to the lateral arm flap but
may be used as a vascularized nerve graft However, if the lateral arm flap is designed to extend
distal to the elbow, then the PCNF may be considered for use to reinnervate the distal portion
of the flap. It is imperative to differentiate these two nerves to reinnervate the lateral arm flap
successfully. Interruption of the PCNF leads to an area of anesthesia in the lateral portion of the
proximal forearm distal to the lateral epicondyle.

graft. It too has been referred to by a variety of names,
including the posterior antebrachial cutaneous nerve (5).
Reportedly, the PCNF can be preserved to avoid the
sensory loss in the forearm, but it requires meticulous
dissection in the septUm to do so and may compromise
the vascular supply to the flap (Fig. 13-4). The associated
functional deficit over the lateral aspect of the forearm is
rarely an issue for the patient in the postoperative period.
Brandt and Khouri (1) descnbed a lateral arm-
proximal forearm flap that extended up to 12 em distal
to the lateral condyle. The vascular supply to the fore-
arm component was based on the rich vascular plexus
that was located over the posterior elbow and that was
fed by the PRCA. The primary sensory nerve supply
in this extended flap was the PCNF. In addition to an
alternative sensory nerve supply, the skin over the lateral
arm, distal to the lateral condyle, is usually thinner than
the skin in the lateral aspect of the upper arm.
ANATOMIC VARIATIONS
Unlike the radial forearm flap, there is no concern in
harvesting the lateral arm flap for the integrity of the col-
lateral circulation to the distal portion of the limb. The
profunda brac:hii may be interrupted without ischemic
sequelae. Most of the anatomic variations that have
been reported are related to duplication of the vascu-
lar pedicle within the septum. The incidence of double
profunda brachii arteries has been reported to be 4%
(7), 8% (12), and 12% (14) in di1Jerent studies. Moffett
et al. (12) recommended temporary occlusion of eac:h of
the arteries to determine whether one or both should be
revascularized. Sc:heker et al. (15) reported a single case
of duplication of the PRCA, which required two anerial
anastomoses to achieve total revascula:rization of the flap.
POTENTIAL PITFALLS
Postoperative radial nerve palsies have been reported and
were attributed to compressive dressings or tight wound
closure (8). Split-thickness skin grafts can be applied
when wound closure is difficult. A light compressive
dressing should be applied to avoid iatrogenic injuries.
Acknowledgments
The author thanks Dr. Michael Sullivan who contrib-
uted to this chapter in the first edition of this book.
212 CHAPTER 13

Lateral Arm Flap

FIGURE 13-5. The topographical anatomy of

the lateral arm flap is outlined. The key land-
marks are the V-sh aped point of insertion of the
deltoid into the humerus and the lateral epicon-
dyle. The dashed line represents the intersec-
tion of these two points. The point of insertion
of the deltoid may be best determined when the
patient is awake. The patient is placed in the
sitting position and asked to press their hand
against the hip.

FIGURE 13-&. The lateral 1M septum is located

approximately 1 em posterior to the line drawn
from the insertion of the deltoid and the lateral
epicondyle. The central axis of the flap design
is based on the IM septum. The territory of skin
may extend distal to the epicondyle and proxi-
mal to the deltoid.

FIGURE 13-7. Harvest of the lateral arm

flap may be performed either with or without
tourniquet control, but more commonly without.
The dissection begins with an anterior incision
through the skin and subcutaneous tissue down
to the brachioradialis and brachialis.
 LATERALARM 213

Lateral Arm Flap

FIGURE 138. The PCNF is identified in the soft

tissue of the flap as it courses distally to supply
sensation in the forearm. This nerve may be pre-
served by meticulous dissection but usually is
cut leaving an area of anesthesia in the forearm.

FIGURE 13-9. Dissection proceeds in a

subfascial plane toward the IM septum; at this
point, a number of septocutaneous perforators
(arrowheads) are identified coursing up into the
subcutaneous tissue.

FIGURE 13-10. Attention is then turned to the

posterior incision, which is carried through the
skin and subcutaneous tissue and the deep
fascia overlying the triceps muscle.
214 CHAPTER 13

Lateral Arm Flap

FIGURE 13-11. The posterior approach to

the IM septum is easier because, unlike the
brachioradialis, the triceps muscle does not
originate from the septum itself.

FIGURE 13-12. As the septum is approached,

the septocutaneous perforators are easily iden-
tified. These perforators lead to the PRCA.

FIGURE 13-11 Having identified the main vascu-

lar pedicle from the posterior approach, attention
is then tumed to finding the PRCA and PRCV
from the anterior approach. This is done by blunt
dissection along the fibers of the brachiorad ialis,
which must be separated from the septum.
 LATERAL ARM 215

Lateral Arm Flap

FIGURE 13-14. Continued dissection along

the anterior aspect of the IM septum leads to
the radial nerve (arrowhead). which is easily
identified because of its large caliber and its
course between the origins of the brachialis and
brachioradialis. The anterior branch of the radial
collateral artery travels with the radial nerve.

FIGURE 1315. Flap elevation proceeds distally

to proximally by sharplytranssecting the fascial
and vascular connections to the humerus.

FIGURE 1316. The continuation of the PRCA,

which anastomoses with the interosseous
recurrent artery. must be identified in the soft
tissue and ligated.
216 CHAPTER 13

Lateral Arm Flap

FIGURE 1317. After ligation ofthe distal por-

tion of PRCA, the dissection proceeds along the
depth of the IM septum. The PCNA and PCNF
are closely associated with the PRC vascular
pedicle.

FIGURE 1318. The neurovascular pedicle is

followed with the radial nerve toward the spiral
groove. Extreme care is taken not to injure the
radial nerve or its branches to the surrounding
musculature in the posterolateral arm.

FIGURE 13-19. Posterior dissection of the flap

reveals the PCNA !arrowhead), which can be
seen to ramify in the subcutaneous tissue of the
flap.
 LATERAL ARM 217

Lateral Arm Flap

FIGURE 1320. The neurovascular pedicle is

skeletonized as it passes in the spiral groove.
Blunt dissection and retraction may improve
visualization of its course.

FIGURE 13-21. Afew centimeters of attach-

ment of the deltoid to the humerus may also
be divided to improve visualization in the spiral
groove. The vascular pedicle is ligated at a
comfortable point in the spiral groove. Ideally,
the ligation of the pedicle is performed proximal
to the point where the venae comitantes merge
into a single vein. More proximal dissection of
the pedicle may be achieved by creating a tunnel
deep to the lateral head of the triceps and then
separating the long head from the lateral head.

FIGURE 1322. Closure is accomplished by

suturing the fascia of the brachia lis to the triceps.
A layered soft tissue closure is accomplished
routinely. A light pressure dressing is applied for
severs I days following the procedure.
218 CHAPTER 13
Lateral Arm Flap
FIGURE 13-23. The lateral arm flap has been
harvested. The neurovascular pedicle can be
lengthened by more proximal dissection and by
designing the flap more distally in the arm and
proximal forearm.
REFERENCES
1. Brandt K. Khouri R: The lateral arm/proximal forearm
flap. Ffast &consrr Surg 1993;92: 1137.
2. Culbertson J, Mutumer K.: The reverse lateral upper arm
flap for elbow coverage. Ann Plast Surg 1987; 18:62--68.
3. Daniel R, Terzis J, Schwarz G: Neurovascular free flaps: a
preliminary repon. Plmt Reconst1' Surz 1975;56:13-20.
4. Franklin J: The deltoid flap: anatomy and clinical appli-
cations. In: Buncke HJ, Furnas H, eds. Symposium on
Fmnrier.r in &comt.nlr:liw Mirmst.rrzery. Vol. 24. St. Louis:
MosbyYear Book; 1984.
5. Hollinshead WH: A~ for Surgeons. 3rd ed. Vol. 3.
Philadelphia: JB Lippincott; 1982.
6. Inoue T, Fujino T: An upper arm flap, pedicled on the
cephalic vein with arterial anastomosis for head and neck
reconstruction. Br J Plast Surg 1986;39:451-453.
7. Katsaros J, Schusterman M. Beppu M, Banis J, Acland R:
The lateral upper ann flap: anatomy and clinical applica-
tions. Ann Plast Surg 1984;12:489-:500.
8. Katsaros J, Tan E, Zoltie N, Barton M, Venugopalsrini-
vasan, Venkataram.akrisbnan: Further experience with
the lateral arm free flap. Pltur Recomt1' Surg 1991;87:
902-910.
9. Kuek L, Chuan T: The c:xte:nded lateral arm flap: a new
modification. J Reconm Mit:Tosurz 1991;7:167-173.
10. Martin D, Mondie JM, DeBiscop J, Schon H, Peri ~:
The osteocutaneous outer arm flap: a new concept m
microsurgical mandibular reconstructions. Rev SUmtatol
Chir Maxillofoe 1988;89:281-287.
11. Matloub H, Larson D, Kuhn J, Yousif J, Sanger J: lAteral
arm free ftap in oral cavity reconstruction: a functional
evaluation. Head Neck 1989;11:205-211.
12. Moffett T, Madison S, Derr J, Aclan.d R: An extended
approach for the vascular pedicle of the lateral arm free
flap. Plmt Recomt1' Surg 1992;89:259-267.
13. NakayamaY, Soeda S, Iino T: A radial forearm flap based
on an extended dissection of the cephalic vein. The lon-
gest venous pedicle? Br J Pfast Surg 1986;39:454-457.
14. Rivtt D, Buffet M, Martin D, et al.:The lateral arm flap:
an anatomic study. J &consrr Mit:To.surg 1987;3:121-132.
15. Scheker L, Kleinert H, Hanel D: Lateral arm compos-
ite tissue transfer to ipsilateral hand defects. J Hand Surz
[Am] 1987;12A:665--672.
16. Shenaq S: Pretmnsfer expansion of a sensate lateral arm
free flap. Ann Ffast Surg 1987;19:558--562.
17. Song R. SongY. Yu Y. SongY: The upper arm free flap.
Clin Plast Surg 1982;9:27-35.
18. Sullivan M, Carron W. Kuriloff D: Lateral arm free flap
in head and neclr: reconstruction. Atrlt Ot.Dlmyngol Head
Neck Surg 1992;118:1095-1101.
19. Urken MI.., Vickery C, Weinberg H, Biller HF: The neu-
rofasciocutaneous radial forearm flap in head and neck
reconstruction----a preliminary report. l...arynzoscope
1990;100:161-173.
20. Waterhouse N, Healy C: The versatility of the lateral arm
flap. Br J PlmtSurg 1990;43:398.
21. Yousif NJ, Warrm. R. Matloub H, Sanger J: The lateral
arm fascial free flap: its anatomy and use in reconstruc-
tion. Pitm Recomtf' Surg 1990;86:1138--1145.
 early one century ago, Monks (44) and Brown
N (11) separately reported cases in which the tem-
poroparietal fascia, based on the superficial temporal
vessels, was used for eyelid and auricular reconstruc-
tion. More recently, the temporoparietal fascial ftap
(TPFF) has become popular as a pedicled fl.ap for use
in periorbital (20,27,43) and auricular reconstruction
and as a free ftap for the management of a variety of
defects (5,22,30,55,58). During the last three decades,
the Bap has become a wluable tool in the reconstruc-
tion of a variety of extremity defects (32,35,62). As
experience with this ftap has increased, surgeons have
come to appreciate several features that make it particu-
larly useful in head and neck reconstruction (20,53). It
is ultrathin, highly vascular, and exhibits a significant
degree of fiexibility, allowing it to drape around grafts
and into cavities while, at the same time, maintaining its
durability. It is particularly resistant to infection when
transferred into infected or irradiated tissue beds. In
head and neck reconstruction, the fascia is most com-
monly transferred as a pedicled flap, but it may also be
used as a free ftap when the arc of rotation is not ade-
quate or for defects located at a distance from the donor
site. This Bap offers the advantage of a well-concealed
donor site in the haiJ:I..bearing scalp (20,48).
FLAP DESIGN AND UTILIZATION
The TPFF is based upon the superficial temporal artery
and vein. It may be transferred independently or in
combination with skin (9,10,16,20,21,26,33,48,49,
52,54) and calvarial bone (Fig. 14.-1) (18,24,41,42).
The key feature of the TPFF is its rich vascularity and
219
Z20 CHAPTER 14

pliability, making it especially useful in managing prob-
lem cavities (19,39) and for coverage of cartilage grafts
uaed in such challenging areas of head and neck recon-
struction as the auricle and the larynx (8-1 0,20,39).'Ibis
tissue may be used in the face, hand, Ql' lower extremity
when a skin graft is preferable to a bulky flap, but where
a suitable recipient bed is lacking (10,20,48,54).
The TPFF may be harvested with dimensions in the
range <>f 14 x 17 em without extensive scalp undermin-
ing. The thickness <>f the flap ranges &om 2 to 4 mm.
When necessary, a split-thickness skin graft is easily
applied to the fascia after its transfer. However, in the
oral cavity, even without a skin graft, it may provide a
watertight seal and a surface for remucosalization.
Brent and others (8-1 O) outlined in detail the use
of this Bap for auricular reconstruction, in conjunc-
tion with autogenous costal cartilage (7) and silastic
(47) framewQI'ks. The pliability of this flap makes it
ideally suited to cover a convoluted auricular frame-
WQI'k. In auricular reconstruction, the success of the

Frontal branch Galea aponeurotica

Temporal branch
of facial n.

Auriculotemporal n.

FIGURE 14-1. The TPFF is supplied by the superficial temporal artery and vein. The superficial
temporal vessels divide into the parietal and frontal branches at approximately the superior
limit of the helix. Prior to crossing the arch, the superficial temporal vessels usually give rise to
the middle temporal artery and vein, which supply the temporalis muscular fascia. There are
cutaneous branches that are given off to the root of the helix that allow transfer of a composite
graft from this region. The temporal branch of the facial nerve crosses the zygoma and is at risk
of injury during anterior dissection of the flap.
 TEMPOROPARIETAL FASCIA 221

surgery depends largely on robust but ultrathin soft
tissue coverage, which permits appreciation of the
three-dimensional details of the ear. It may also be
used in the acutely traumatized auricle as a method
of providing immediate auricular cartilage coverage
(23,37,61).
Acland et al. (2) identified the middle temporal
artery, which is a branch of the superficial temporal
artery and vein, as the primary vascular supply to the
temporalis muscular fascia (Fig. 14-2). This vascular
pattern. permits simultaneous tranafer of two separate
leaves of vascularized fascia. East et al. (27) used this
composite flap in a case of posttraumatic tracheoma-
lacia, in which a nasal septal cartilage graft was sand-
wiched between the two leaves of fascia and then inset
into the anterior tracheal wall.
The TPFF may also be used in the management of
radiated temporal bone and o:drital cavities (19,28,40,59).

Frontal and
parietal branches

Deep temporal arteries

Maxillary a.

Middle temporal a.
FIGURE 14-2. The muscular and fascial layers of the temporal fossa are shown. The
deepest plane is the temporalis muscle, supplied by the anterior and posterior deep temporal
arteries, arising from the internal maxillary artery. The temporal is muscular fascia is supplied
by the middle temporal artery, which arises from the superficial temporal artery below the
zygomatic arch. The terminal branches of the superficial temporal artery and vein supply the
temporoparietal fascia.
ZZ2 CHAPTER 14

The fiap has also been described in co:rre<:ting c<mtour
defects of the midface and orbital regi<m (42).
In addition to its traditional application as a fascial
Bap, the tissue may also he used to ttansfer overlying
scalp skin and hair during scalp and lip reconstruction
(34,38,39,45,46) (see Pip. 14-15 to 14-18). By extend-
ing the vascular pedicle with an interposed vein graft,
the fiap can be mobilized by using a V-Y technique to
close full-thickness defects of the scalp (32).
A series of reports descnbe reconstructions in which
the superficial temporal artery and vein were used to
transfer skin and cartilage from the root of the auricular
helix. This composite graft may he used to reconstruct
sizable defects of the nasal ala (49,50,57). Duplication
of the thin natural contour of the nasal ala with its car-
tilage covered by skin, both on the inside and outside, is
one of the most challenging aspects of nasal reconstruc-
tion. The root of the helix matches the ala perhaps better
than aiiY other tissue in the body and is well suited to
this particular defect.
McCarthy and Zido (42) descnbed the elevation of
temporoparietal fascia in c<mjunction with outer calvaJ.'I-
ial bone and documented the contnbution of the supet'l-
ficial temporal artery to the vascularity of this bone
(17,25,51). Experimental work by Antonyshyn et al. (3)
suggested that the vascularized calvarial bone transfers
are superior to standard calvarial bone grafts in terms
of early viability and new osteoid formation. The reli-
ability and l<mg-term results of vascularized calvarial
bone grafts placed into craniofacial defects have been
demonstrated in large clinical series (6,50). If bone is
to be transferred with the flap, a generous cutf of fascia
and pericranium must he preserved at the periphery of
the graft. The outer table of the skull is harvested as
a split cranial graft (20). The temporoparietal fascia is
fixed to the h<me with a suture to prevent shearing of the

Deep temporal a.

Middle temporal a.

Temporalis muscle fascia

Temporoparietal fascia

FIGURE 14-3. The layers of the scalp in the temporal fossa are shown, extending from the
calvaria I bone to the skin. The temporalis muscular fascia splits into two layers approximately
2 em above the zygomatic arch. These two fascial leaves are separated by fat, which provides
a natural plane of dissection. The temporalis muscular fascia is continuous with the masseter
muscular fascia below the arch; the temporoparietal fascia is continuous with the superficial
muscular aponeurotic system below the arch. The temporalis muscular fascia and the TPFF are
separated by a loose areolar plane, which also separates the pericranium from the galea in the
region cephalad to the superior temporal line. TPFF, temporoparietal fascial flap.
 TEMPOROPARIETAL FASCIA 223

delicate vessels that perforate the pericranium (53). By The superficial temporal artery and vein are moderate-
combining the bone graft with the temporoparietal fas- sized vessels that are most easily isolated approximately
cia, studies have documented that the surviving osseous 3 em superior to the root of the helix where they branch
mass is increased compared with that of cODVentional into frontal and parietal divisions (Fig. 14-1). These
nODVa.scularized calvarial grafts (14,15,25). However, branches anastomose freely with the supraorbital and
the clinical relevance of this higher surv.iving mass may supratrochlear vessels over the forehead (4). The flap is
not be clinically significant, as most surgeons do not most commonly based upon the parietal branch, with
have difficulty with volume loss following free calvarial its base centered over the middle third of the superior
bone grafting.
The vascularity of the TPFF extends to the midline
of the skull, and may be extended to this point if the flap
is to be used for intraoral reconstruction (31). Transfer
of the pedicled TPFF, with or without attached calva-
rial bone, will reach the malar, olbital, and mandibular
regions in most patients. Pedicle length is often inade-
quate when transposing the TPFF into the oral cavity.
Several maneuvers increase the arc of rotation, including
the temporary removal of the zygomatic arch or the prox- ~H'""''r-+--Temporalis
imal dissection of the pedicle below the tragus. Mobiliza- mullde fascia
tion of the vascular pedicle below the tragus places the
facial nerve at risk and requires identification of the nerve
TemporalIs
in the parotid gland. An incision can be made through mullde------+~~ ltTir--+T- Subcutaneous
the buccal sulcus to permit intraoral transfer of the flap. 1!S&~e

NEUROVASCULAR ANATOMY
Inconsistent nomenclature plagues the description
of the anatomic layers of the temporoparietal region 1/IJH~r---Temporoparietal
(Fig. 14-3) (1,36,60). The temporoparietal scalp con- fascia
sists offive distinct layers. The temporoparietal fascia lies (with superficial
temporal artery)
in the central position between two tis9Ue planes in the
area below the superior temporal line. It lies deep to the
skin and subcutaneous tissue to which it is firmly bound.
Superficial
The temporoparietal fascia must not be confused with temporal
the temporalis muscular fascia, which envelops the tem- fat pad ---r~rt:~
poralis muscle. The temporoparietal fascia is a superior
extension of the superficial musculoaponeurotic system,
both of which attach to the zygomatic arch. The tempo-
Zygomatic
ralls muscular fascia splits, and one lamella passes deep aroh--------~~~
to the arch to insert on the coronoid process of the man-
Deep
dible (43), while the other inserts on the lateral surface temporal
of the zygomatic arch, contiguous with the masseteric fat pad - - - -+T-T
fascia. Above the superior temporal line, the temporo-
parietal fascia becomes the galea aponeurotica. Below
the superior temporal line, the tissue planes deep to the +;++++---+-Masseter
muscle
temporoparietal fascia consist ofloose areolar tissue and
temporalis muscular fascia. Loose areolar tissue sepa-
11-l--++++------4--- Masseteric
rates the temporoparietal fascia from the muscular fascia fascia
of the temporalis, giving the scalp its natural mobility. Coronoid
When the scalp is moved, the temporoparietal fascia process
moves with it, but the muscular fascia and periosteum Parotid gland
remain stable. In the area above the temporal line, the FIGURE 14-4. A coronal section through the left temporal
temporalis muscular fascia and the periosteum converge region demonstrates the fascial and muscular anatomy. In
and continue cephalad as the pericranium overlying the addition, the relationship of the fat pads located above and
superior aspect of the cranium. (Fig. 14-4). below the zygomatic arch is demonstrated.
Z24 CHAPTER 14

auricular helix. The frontal branch is routinely ligated
approximately 3 to 4 em distal to its separation. from
the parietal branch. Dissection beyond this point risks
injury to the frontal branch of the facial nerve.
At its origin, the superficial temporal artery has
an average diameter of 1.89 mm and lies deep to, or
within, the parotid gland (55). In the first part of its
course, appr<Wmately 15 mm, it ascends behind the
ramus of the mand:&ble and then pierces the super-
ficial fascia 4 to 5 mm in front of the tragus (29). In
the second, or superficial, part of its course, it crosses
the posterior portion of the zygomatic process of the
temporal bone, lying anterior to the auriculotemporal
nerve and the superficial temporal vein where it can be
easily palpated. It may lie beneath the anterior auricular
muscle and may often have a tortuous course. The sig-
nificance of that tortuosity is that if this is "released," it
may increase the length of the pedicle of an island flap
by up to 1.5 em. The middle temporal branch of the
artery, which supplies the temporalis muscular fascia, is
given off in this region.. Because of this branching va5-
cular pattern, a two-layered fascial ftap can be raised on
a single vascular pedicle (Fig. 14-5) (23).
At a point 2 to 4 em above the zygomatic arch (range,
0 to 5 em), the superficial temporal artery divides into
two terminal branches: the frontal and the parietal.

Tempo<al;s muscle lascia~

Middle temporal
artery and vein

Temporoparietal fascia

Superficial temporal
artery and vein

FIGURE 14-5. If two layers of vascularized fascia are required for reconstruction, the TPFF may
be harvested along with the temporal is muscular fascia. The two separate nutrient vascular
systems allow two independent thin layers of tissue to be harvested, which makes this donor
site unique. Careful dissection of the middle temporal artery in the region of the zygomatic
arch is required to harvest these two fascial leaves as a single microvascular free flap. TPFF,
temporoparietal fascial flap.

A delayed division is commonly associated with a well-
developed zygomatico-orbital branch from the main
stem of the superficial temporal artery and occurs
in 80% of cases (51). The frontal branch is generally
slightly larger (1.2 mm in diameter) than the parietal
branch (1.1 mm in diameter) (22). The frontal branch
runs in a tortuous fashion, anterior and medial, supply-
ing all layers of the scalp, and it anastomoses with the
corresponding vessel of the opposite side and also with
the ipsilateral supraorbital and supratrochlear arteries.
The parietal branch passes superiorly toward the
vertex. In approximately 7.5% of patients, that branch
may divide into two branches, which travel roughly
parallel to each other. Its course lies within a 2-cm
strip centered on the auditory meatus and passing
upward to the vertex. Within this band, the artery usu-
ally traverses from the anterior to the posterior mar-
gins. The superficial temporal vein may be single or
duplicate. In most cases, the vein runs with the artery,
but slightly superficial to it. In 20% to 30% of cases,
however, the vein takes a divergent course above the
level of the root of the helix and may travel up to 3 em
posterior to the arterial pedicle (53,58). There are a
rich set of anastomoses of the terminal portions of the
parietal branch both with its opposite member, which
arises from the contralateral superficial temporal sys-
tem, and with the ipsilateral posterior auricular and
occipital arteries (22).
The superficial temporal system gives off several
branches to the skin of the face. The transverse facial
artery arises deep to the parotid gland and runs for-
ward over the masseter. In 35% of cases, the trans-
verse facial artery arises from the external carotid
artery directly. It runs forward and is accompanied by
branches of the facial nerve in the region between the
zygomatic arch and the parotid duct, often crossing the
duct. It supplies the parotid gland and duct, the mas-
seter muscle, and the skin. A large cutaneous branch is
consistently found at the point of intersection of a verti-
cal line drawn 2 em laterally to the lateral canthus with
a horizontal line through the alar base. The transverse
facial artery anastomoses superiorly with the lacrimal
and infraorbital arteries, anteriorly with the premas-
seteric and facial arteries, and deeply with the buccal
artery. This artery primarily supplies muscle, but it may
also make a significant contribution to the blood sup-
ply of the skin over the masseter and the parotid and,
to a lesser extent, to the skin of the inferior orbital and
nasolabial regions (22).
The superficial temporal artery gives rise to three
groups of auricular branches: (a) an inferior group sup-
plies the lobule and tragus; (b) two or three branches in
a superior group often form a common trunk and run
onto the upper part of the helix, its crura, and triangular
fossa (48); and (c) the superficial temporal or its parietal
branch gives off a small branch that runs down behind
TEMPOROPARIETAL FASCIA 225
the ear for a short distance, supplying the uppermost
part of the cranial surface of the ear, and anastomosing
with the posterior auricular artery (22).
The zygomatico-orbital artery may arise from the
superficial temporal artery, the middle temporal branch,
or the frontal branch. It runs along the upper border
of the zygomatic arch, in the fat pad between the deep
and superficial layers of the temporalis muscular fascia,
to the lateral aspect of the orbit (Fig. 14-4). It supplies
branches to the orbicularis oculi, anastomoses with
the lacrimal and palpebral branches of the ophthalmic
artery, and completes the perimbital ring with the infra-
and supraorbital vessels (22). Inadvertent entry into the
fat pad encased by the deep and superficial layers of
the temporalis muscular fascia places the zygomatico-
orbital artery at risk. Cautery to control bleeding in this
location can inadvertently damage the frontal branch of
the facial nerve.
There are various motor and sensory nerves that trav-
erse the temporoparietal donor site. The auriculotem-
poral nerve, a sensory branch of the trigeminal nerve,
lies posterior to the superficial temporal artery, within
the temporoparietal fascia, and supplies the regional
scalp skin. The frontal branch of the facial nerve courses
obliquely across the zygomatic arch and lies approxi-
mately 1.5 em lateral to the orbital rim. This nerve runs
within the temporoparietal fascia and represents the
anterior limit of flap elevation (see Figs. 14-6 to 14-14).
ANATOMIC VARIATIONS
Five distinct branching patterns of the superficial tem-
poral artery have been described (13,56), though in
general there is such high vascularity that these vari-
ations do not hold clinical significance. Variability
involves anastomotic connections with tributaries of the
occipital artery, and in cases where a large, posterior
segment of fascia is required, the course of the posterior
parietal branch can be traced with Doppler sonography
to ensure that the planned territory of the TPFF is well
vascularized.
POTENTIAL PITFALLS
The most common complication after the elevation of
the TPFF is secondary alopecia. This has been noted
around the incision site for up to 2 em and can be
avoided by meticulous preservation of hair follicles
on the undersurface of the scalp flap. Prior radiation
therapy or surgery in this area predisposes the overlying
scalp to ischemic injury after raising the flap, and care-
less surgical dissection can result in direct injury to the
hair follicles. Avoidance of cautery on the undersurface
of the skin flap greatly reduces the risk of alopecia.
226 CHAPTER 14
Anterior dissection of the TPFF is limited by the
frontal branch of the facial nerve, and failure to identify
this branch during the dissection of the flap will risk
its inadvertent injury. This may lead to brow paralysis.
When the injury is simple praxia, full recovery can be
expected, but transection injuries require microsurgical
repair.
PREOPERATIVE ASSESSMENT
Many factors influence the viability of the TPFF and
limit its use in the reconstruction of selected head and
neck defects. Preoperative radiation, neck surgery, or
external carotid embolization may affect the vascular
pedicle and are considered relative contraindications
to the elevation of this tissue. The most important pre-
operative test to determine the reliability of this flap
is Doppler auscultation, which should be performed
in the office before finalizing the surgical plan. This
ensures the presence of a viable arterial input. Preopera-
tive assessments of venous outflow are not critical when
the TPFF is utilized as a regional flap. This is because
venous egress from the TPFF occurs both through the
superficial temporal vein and the myriad of intercon-
nected venous tributaries; even in the absence of a
detectable vein, outflow through venules in the pedicle
is nearly always adequate. However, when free TPFF
transfer is contemplated, more formal venous assess-
ment is desirable through Doppler auscultation and/or
ultrasonographic color flow Doppler assessment.
POSTOPERATIVE WOUND CARE
After transfer of the TPFF, careful hemostasis is per-
formed with bipolar cautery. Special effort is made to
avoid thermal injury to the hair folJicles. A suction drain
is routinely used and should be placed in a superior
position to avoid inadvertent contact with the vascular
pedicle when a transposition flap has been performed.
The wound is closed in layers, and a bulky compressive
dressing is applied for 24 hours. The suction drain is
kept in place for 24 to 48 hours.
 TEMPOROPARIETAL FASCIA 227

Tempoparietal Fascial Flap

FIGURE 14-6. The topographical anatomy of the

TPFF is outlined. The approximate course of the
superficial temporal artery and vein has been
drawn, with division of the pedicle into anterior
and posterior branches at a point approximately
3 em above the tragus. The vein may run concur-
rentlywith the artery or may separate from the
artery and run 2 to 3 em posteriorly. The superior
temporal line is shown. This bony ridge begins
at the zygomatic process of the frontal bone and
curves upward and backward across the frontal
bone along the lateral margin of the forehead. It
passes over the parietal bone and ends by joining
the supramastoid crest The most cephalic origin
of the temporalis muscle is at the superior tem-
poral line. This is an important landmark,. where
the temporoparietal fascia becomes confluent
with the galea aponeurotica. At the superior tem-
poral line, the temporalis muscular fascia merges
with the periosteum that covers the calvarium.

FIGURE 147. A vertical incision extend-

ing from the root of the helix to the superior
temporal line is used in harvesting the tempo-
roparietal fascia. A V-extension atthe superior
aspect ofthe incision is used to gain full access
to the fascial layers in this region. Inferiorly, the
incision is placed in the preauricular crease
adjacent to the root of the helix. In most cases,
both the arterial and venous components of
the vascular pedicle are located anterior to the
initial incision; however, the dominant venous
system may travel a more posterior course, just
underneath the skin incision. Therefore, dissec-
tion must proceed cautiously until the vein is
identified. If the vein is found in an unconven-
tional location, flap harvest must be tailored to
capture its more posterior course.
Z28 CHAPTER 14

Tempoparietal Fascial Flap

FIGURE 14-8. The approximate course of the

frontal branch of the facial nerve has been
drawn as it courses from the main trunk, over
the zygoma, and toward the lateral aspect of the
forehead. The dissection must remain posterior
to this region to avoid injury to this nerve.

FIGURE 14--9. The dissectian begins by elevat-

ing anterior and posterior scalp flaps. Particular
attention is given to avoiding injury ta the hair
follicles as these flaps are elevated. The frontal
branch afthe superficial temporal artery is
Iig ated at the anteriar limit of the flap.
 TEMPOROPARIETAL FASCIA 229

Tempoparietal Fascial Flap

FIGURE 14-10. With the posterior, superior,

and anterior scalp flaps elevated in the plane
superficial to the temporoparietal fascia, the
deep dissection and elevation of the flap are
performed. This elevation is best initiated at
the superiortemporalline.ldentification of the
temporalis muscular fascia assures elevation
in the proper plane. The layer of loose areolar
tissue that separates the temporoparietal fascia
from the muscular fascia at this level permits a
straightforward dissection in a nonvascular plane.

FIGURE 14-11. The TPFF has been elevated to

the root of the helix. Meticulous dissection must
be performed in this region to avoid injury to the
vascular pedicle. If the temporalis muscular fas-
cia (arrowheads) is to be harvested, the plane
of dissection is developed along the surface of
the temporalis muscle. In addition, more caudal
dissection ofthe pedicle is required in order to
capture the middle temporal artery.

FIGURE 1412. The temporoparietal fascia is

elevated, and the superficial temporal artery
and vein (arrowheads) are isolated. The width
of the flap base is normally 2.0 to 2.5 em. As
demonstrated, the flap is flexible, thin, and
highly vascular. It exhibits exceptional drap-
ing characteristics.
Z30 CHAPTER 14

Tempoparietal Fascial Flap

FIGURE 14-13. When the temporoparietal

fascia is elevated and the superficial temporal
artery and vein are identified linked), the base
of the flap can be left extremely wide, even
capturing the occipital contributions (inked).
As demonstrated, the flap is flexible, thin, and
highly vascular, and it exhibits exceptional
draping characteristics. The elevation shown
would be useful for total auricular reconstruc-
tion to drape over a cartilage framework.

FIGURE 14-14. The superficial temporal artery

(diameter, 1.8 to 2.2 mm) and vein (diameter, 2.0
to 3.0 mm) are limited in length by the dangers
of extending the caudal dissection in the vicinity
of the main trunk of the fa ciaI nerve. Also shown
is the occipital pedicle Iarrow}, variably present
but critical to identify and preserve in the har-
vest of a large surface area pedicled TPFF.

FIGURE 14-15. In male patients with complex

upper lip defects, the hair-bearing TPFF can be
utilized to restore hair-bearing skin to the upper
lip. A template is made of the lip defect to assist
in flap design, as shown.
 TEMPOROPARIETAL FASCIA 231

Tempoparietal Fasciocutaneous
Fl

FIGURE 14-16. A 1.5-cm cutaneous pedicle is

centered over the superficial temporal vessels,
using Doppler auscultation.

FIGURE 14-17. The flap is elevated from the

true temporalis fascia in a superior to inferior
direction, working toward 1he zygomatic arch,
to maximize pedicle length. Care is taken to
capture 1he vascular pedicle, leaving a 1.5-cm
cutaneous base.

FIGURE 14-18. The hair-bearing TPFF is trans-

posed into the defect. This is followed, approxi-
mately 3 weeks, later, by division of the vascular
pedicle and inset of the lateral border of the
flap. During this stage, the pedicle is transected
and 1he proximal portion is either returned to its
original bed in the temporal scalp or discarded,
depending upon the healing of the donor site.
232 CHAPTER 14
REFERENCES
1. Abul-Hassan HS, von Drasek-Ascher G, Acland RD:
Surgical anatomy and blood supply of the fascial layers of
the temporal region. Plast Reconstr Surg 1986;77: 17.
2. Acland RD, Abul-Hassan H, Ulfe E: Superficial and
deep temporal fascia flap: an anatomic approach. Video-
tape presentation at the Annual Meeting of the Ameri-
can Society of Plastic and Reconstructive Surgeons, Las
Vegas, Nevada, October 12, 1983.
3. Antonyshyn 0, Colcleugh RG, Hurst LN, Anderson C:
The temporalis myo-osseous flap: an experimental study.
Plast Reconstr Surg 1986;77:406.
4. Antonyshyn 0, Gruss JS, Birt BD: Versatility of temporal
muscle and fascial flaps. Br J Plast Surg 1988;41:118.
5. Avelar JM, Psillakis JM: The use of galea flaps in cranio-
facial defonnities. Ann Prast Surg 1981;6:464.
6. Bite U, Jackson IT, Wanner Hw, Marsh RW: Vascularized
skull bone grafts in craniofacial surgery. Ann Prast Surg
1987;19:3.
7. Boudard PH, Benassayag CL, Alvarez FL, Portmann
0, Bebear JP, Portmann M: Temporoparietal fascial
flap in difficult total ear reconstruction. Rev Laryngol
1989;110:219.
8. Brent B: Auricular repair with autogenous rib cartilage
grafts: two decades of experience with 600 cases. Platt
Reconm Surg 1992;9:355.
9. Brent B, Byrd HS: Secondary ear reconstruction with
cartilage grafts covered by axial, random, and free
flaps of temporoparietal fascia. Plast Reconstr Surg
1983;72:141.
10. Brent B, Upton J, Acland RD, et al.: Experience with
the temporoparietal fascial free flap. Plast Reconstr Surg
1985;76:177- 188.
11. Brown WJ: Extraordinary case of horse bite: the exter-
nal ear completely bitten off and successfully replaced.
Lancet 1898;1:1533.
12. Byrd HS: The use of subcutaneous axial fascial flaps in
reconstruction of the head. Ann Plast Surg 1980;4: 191.
13. Busthianos NA: Etude anatomique sur les arteres tempo-
rales superficielles. Ann Anat Pathol (Paris) 1932;9:678.
14. Canalis RF: Further observation of the fate of pedi-
cle osseocutaneous grafts. Otolaryngol Head Neck Surg
1979;87:763.
15. Canalis RF, Saffouri M. Mirra J, Ward PH: The fate of
pedicle osseocutaneous grafts. Laryngoscope 1977;87:895.
16. Carstens MH, Greco RJ, Hurwitz DJ, Tolhurst DE: Clini-
cal applications of the subgaleal fascia. Plast Reconm Surg
1991;87:615.
17. Casanova R, Cavalcante D, Grotting JC, Vasconez LO,
Psillakis JM: Anatomic basis for vascularized outer-table
calvarial bone flaps. Plast Reconstr Surg 1986;78:300.
18. Cheney ML, Gliklich RE: The use of calavarial bone in
nasal reconstruction. Arch Otolaryngol Head Neck Surg
1995; 121 (6) :643-648.
19. Cheney ML, Megerian Ca, Brown MT, McKenna MJ,
Nadol JB: The use of the temporoparietal fascial flap
in temporal bone reconstruction. Am J Otol 1996; 17:
137-142.
20. Cheney ML, Varvares M. Nadol JB Jr:The temporopari-
etal fascial flap in the head and neck reconstruction. Arch
Otolaryngol Head Neck Surg 1993;119(6):618-623.
21. Chiarelli A, Baldelli A, Vincenzo AD, Martini G: Utili-
zation of the superficial temporoparietal fascia in recon-
structive plastic surgery. Ophthal Plast Reconstr Surg
1989;5:274.
22. Cormac GC, Lamberty BGH: The Arterial Anatomy of
Skin Flaps. New York: Churchill Livingstone; 1986.
23. Cotlar SW: Reconstruction of the burned ear using a
temporalis fascial flap. Platt Reconstr Surg 1983;71 :45.
24. Cutting CB, McCarthy JG: Comparison of residual osse-
ous mass between vascularized and non-vascularized
onlay bone transfers. Plast Reconstr Surg 1983;72:672.
25. Cutting CB, McCarthy JG, Berenstein A: Blood supply
of the upper craniofacial skeleton: the search for compos-
ite calvarial bone flaps. Platt Reconstr Surg 1984;74:603.
26. David SK, Cheney ML: An anatomic study of the tem-
poroparietal fascial flap. Arch Otolaryngol Head Neck Surg
1995;121: 1153-1156
27. East C, Grant H, Jones B: Tracheal reconstruction using
a composite microvascular temporoparietal fascia flap
and nasal septal graft. J Laryngol Oto/1992; 106:741.
28. East CA, Brongh MD, Grant HR: Mastoid obliteration
with the temporoparietal fascial flap. J Laryngol Otol
1991;105:417.
29. Esser JFS: Uber eine gestielte Ueberpflanzung eine sen-
recht angelegten Kells aus dem oberen Augenlid in das
g)eichsc::itige Unterlid oder umgekehrt. Klin Monatsbl
Augen}teilkd 1919;63:379.
30. Fox JW, Edgerton MT: The fan flap: an adjunct to ear
reconstruction. Plast Reconstr Surg 1976;58:663.
31. Godfrey PM: Sinus obliteration for chronic oro-antral fis-
tula: a case report. Br J Plast Surg 1993;46:341.
32. Hallock GG: The extended temporoparietal fascia "non
free" flap. Ann Plast Surg 1988;21 :65.
33. Hallock GG: Scalp fascia for the extremities. Contemp
Orthop 1990;21 :542.
34. Harii K, Ohmori K, Ohmori S: Hair transplantation with
free scalp flaps. Plast Reconm Surg 1974;53:410.
35. Hing DN, Buncke HJ, Alpert BS: Use of temporoparietal
free fascial flap in the upper extremity. Plast Reconstr Surg
1988;81:534.
36. Horowitz JH, Persing JA, Nichter IS, Morgan RF, Edg-
erton MT: Galeal-pericranial flaps in head and neck
reconstruction: anatomy and application. Am J Surg
1984;148:489.
37. Jenkins AM, Finucan T : Primary nonmicrosurgi-
cal reconstruction following ear avulsion using the
temporoparietal fascial island. Plast Reconstr Surg
1989;83: 148.

38. Kim JC, Hadlock T, Varwres MA, Cheney ML: Hair-
bearing temporoparietal fascial flap reconstruction
of upper lip and scalp defects. Arch Facial Plast Smg
2001;3:170-177.
39. Lyons GB, Milroy BC, Lendwy PG, Teston LM: Upper
lip reconstruction: use of the free superficial temporal
artery hair-bearing flap. BT J Plan Smg 1989;42:333.
40. Maillard FG, Gumener R. Montandon D: Correction of
depressed supratarsal sulcus by an arterial subcutaneous
composite flap. Plast Recomtr Surg 1984;74:362.
41. McCarthy JG, Cutting CB, ShawWW:Vascularized cal-
wrial flap. Clin Plast Smg 1987;14:37.
42. McCarthy JG, Zido BM: The spectrum of calvarial bone
grafting: introduction of the wscularized calvarial bone
graft. Plast Recomtr Smg 1984;74:10.
43. Mitz V, Peyronie M: The superficial musculoaponeu-
rotic system (SMAS) in the parotid and cheek area. Plast
Recomtr Surg 1976;58:80.
44. Monks GH: The restoration of a lower lid by a new
method. N Engl J Med 1898; 139:385.
45. Ohmori K: Free scalp flap. Plast Recomtr Surg 1980;65:42.
46. Ohmori K: Free scalp flap surgery. Ann Plast Smg
1980;5:17.
47. Ohmori S: Reconstruction of microtia using the Silastic
frame. Clin Plast Surg 1978;5:379.
48. Panje R. Morris MR: The temporoparietal fascial
flap in head and neck reconstruction. Ear Nose Throat
J 1991;70:311.
49. Parkhouse N, Evans D: Reconstruction of the ala of the
nose using a composite free flap from the pinna. Br J Plast
Surg 1985;38:306.
TEMPOROPARIETAL FASCIA 233
50. Pribaz )J, Falco N: Nasal reconstruction with auricular
microvascular transplant. Ann Plast Surg 1993;31:289.
51. Psillakis JM, Grotting JC, Casanow R, Cavalcante D,
Vasconez LO: Vascularized outer-table calwrial bone
flaps. Plart Recomtr Surg 1986;78:309.
52. Rieboung B. Mitz., Lass au JP: Artere temporale superfici-
elle. Ann Chir Plan Esthet 1975;20: 197.
53. Rose EH, Norris MS: The versatile temporoparietal fas-
cial flap: adaptability to a wriety of composite defects.
Plast Reconstr Surg 1990;85:224.
54. Smet HT: Fascial flaps. In: Tissue Tramfers in Recomtruc-
tifJe Smgery. Part 2. New York: Raven Press; 1989:51.
55. Smith RA: The free fascial scalp flap. Plast Reconstr Surg
1980;66:204.
56. Stock AI., Collins HP, Davidson TM: Anatomy of the
superficial temporal artery. Head Neck 1980;2:466.
57. Tanaka Y, Tojima S, Tsuijiguchi K, Fukea E, Ohmiya Y:
Microwscular reconstruction of the nose and ear defects
using composite auricular free flaps. Ann Plast Surg
1993;31:298.
58. Tegtmeier RE, Gooding RA: The use of a fascial flap in
ear reconstruction. Plan Recomtr Smg 1977;60:406.
59. Teichgraeber JF: Temporoparietal fascial flap in orbital
reconstruction. Laryngoscope 1993; 103:931.
60. Tolhurst DE, Carstins MH, Graco RJ, Hurwitz DJ:The sur-
gical anatomy of the scalp. Plast Reconstr Surg 1991;87:603.
61. Tmpin JM, Altman 01, Cruz G, Acjaver BM: Salwge of
the severely injured ear. Ann Plast Smg 1988;21:170.
62. UptonJ, Rogers C, Durham-Smith G, SwartzW: Clinical
applications of free temporoparietal fascial flaps in hand
reconstruction.J Hand Surg [Am] 1986;lla:475.
The anterolateral thigh (ALT) free flap was first
described by Song et al. in 1984 as a soft tissue flap that
is perfused by septocutmeous branches of the lateral
circumflex femoral artery (LCFA) (38). Subseq_uent
cadaver dissections (8,27,37,42,44) and an early clinical
case series (19) clarified that the majority of ALT flaps
are supplied by musculocutaneous perforators of the
LCFA that pierce the medial edge of the vastuslateralis
muscle (Fig. 15-1). It also became evident that while
most cutaneous perforators to the ALT Bap arise from
the descending branch of the LCFA, skin perforators
also arise from the transverse branch of the LCFA in
about 10% of cases. The ALT flap was popularized for
head and neck reconstruction in the 1990s by Koshima
et al. (20) and Kimata et al. (17). Thereafter, the ALT
Bap became a commonly used method for soft tissue
reconstruction of the head and neck in Asia. The largest
234
clinical series published to date included 1,284 cases,
including 911 cases of head and neck reconstruction,
performed betWeen 1991 and 2001 at the Chang Gung
Memorial Hospital inTaiwan (12). During the first deo-
ade of the 2 bt century, ALT flaps became increasingly
popular for head and neck reconstruction in North
America and Europe (23,26,43,46).
FLAP DESIGN AND UTILIZATION
A large amount of skin from the thigh is available for
transfer with the ALT flap. In a longitudinal direction,
the vascular territory extends from the level ofthe greater
trochanter of the femur superiorly to just above the level
of the patella inferiorly. The donor thigh is placed into
a neutral position without internal or extemal rotation,
 ANTEROLATERAL THIGH FREE FLAP 235

Septocutaneous
perforator

Musculocutaneous
perforator ---~;w.~

Vastus
lateralis nus11:1e -t:;wt~
LCFA
Vastus
intermedius
muscle

FIGURE 15-1. Cross-sectional anatomy of the thigh demonstrates the vascular blood supply to
the ALT flap. The LCFA arises from the proximal portion of the profunda femoris artery and gives
off cutaneous perforators to the skin paddle of the ALT flap in the mid-thigh region. About 10%
of perforators to the skin paddle of the flap are septocutaneous blood vessels that travel though
the intermuscular septum that separates the rectus femoris muscle from the vastus latera lis
muscle. About 90% of perforators to the skin paddle of the flap are musculocutaneous blood
vessels that travel through the medial edge of the vastus lateral is muscle.

and an elliptical skin paddle with a width that permits
primary closure of the donor site wound is designed.
The Bap is centered around the long axis of the flap
parallel to and a few centimeters lateral to a line that is
drawn from the anterior superior iliac spine (ASIS) to
the lateral border of the patella (Fig. 15-2). A territory
of up to 20 em in width and 30 em in length is avail-
able for transfer based upon cutaneous branches of the
LCFA. The width of the skin paddle that can be har-
vested while allowing for primary closure of the thigh
donor site defect varies according to patient stature,
body habitus, and skin laxity. The upper thigh usually
has more skin laxity than the distal thigh, which permits
the harvest of a wider Bap in the proximal thigh while
still achieving primary wound closure. The mu.imum
Bap width that allows for primary wound closure can be
estimated by pinching the skin of the thigh flap donor
site between the thumb and fingers. This width varies
from 8 to 10 em in most patients, although primary clo-
sure of defects as wide as 12 em has been reported (23).
'When wider skin paddles are needed, the donor site
defect can be repaired using a split-thickness skin graft
harvested from the anteromedial aspect of the ipsilateral
thigh or from the contralateral thigh. Repair of the thigh
wound using V-to-Y advancement flaps has also been
descnoed (15).
The role of the Doppler stethoscope in determin-
ing the location of cutaneous peiforators and, in turn,
using that information to design the ALT Bap skin pad-
dle position remains controversial. In this application, a
Doppler stethoscope is used to identify the location of
cutaneous perforators from the LCFA. LCFA cutane-
ous perforators are usually located within a 3 to 5 em
radius from the midpoint of the axis that is drawn from
the ASIS to the lateral border of the patella, which
denotes the position of the intermuscular septum that
separates the rectus femoris and vastus lateralis mus-
cles. Doppler-detected pulses, denoting the position of
peiforators, are most frequently identified 1 to 2 em
lateral to the intermuscular septum. Most authors
of clinical series on ALT flaps report that they use a
Doppler stethoscope to estimate the location of cutane-
ous perforators and to help position and design the flap
skin paddle. Lueg is a strong advocate of the Doppler
stethoscope technique, reporting that he centered ALT
flap skin paddles on DoppleJ.'I-dete<:ted pulses and made
no attempt to identify or dissect the cutaneous perfo-
rators during Bap harvest (23). However, there is con-
siderable evidence to doubt the accuracy and precision
of Doppler stethoscopes for localization of cutaneous
perforators from the LCFA. Yu and Adel examined the
efficacy of handheld Doppler stethoscopes for identifi-
cation of cutaneous peiforators in 100 patients under-
going harvest of ALT Baps (48). They noted that while
the sensitivity of two commercially available Doppler
stethoscopes was 91% to 100%, the specificity was only
0% to 55%. Cutaneous peiforators that were confirmed
by surgical exploration were present within 1 em of the
Doppler-detected pulse in 70% to 74% of cases. Based
on this experience, the authors abandoned use of a
Doppler stethoscope in planning the skin paddle design
in their next 20 cases and alternatively positioned the
Z3& CHAPTER 15

Anterior superior --~-

iliac spine

Perforator from
transverse branch
ofLCFA

Vastus lateralis
muscle -------4r-t

~-~r-+--+-+-- Rectus remoris

muscle

Perforator from
descending branch
ofLCFA

Lateral border of ----+-

patella

FIGURE 15-2. The skin paddle is centered on a line that is drawn from the AS IS to the lateral
border of the patella (green line). This line denotes the position of the intermuscular septum
that separates the rectus femoris muscle from the vastus latera lis muscle. Other muscles of the
anterior thigh include the sartorius and vastus medialis muscles. Cutaneous perforators of the
descending branch as well as the transverse branch of the lateral femoral circumflex artery are
demonstrated.

fiap relative to the location of the bo:ay landmarb of the
ASIS and patella. All flap transfers were successful with-
out attempting to localize the perforators by Doppler
stethoscope.
A variable amount of subcutaneous tissue is available
for transfer with the ALT free flap, depending upon
multiple variables that include body habitus, gender,
ethnicity, flap design and harvest technique. ALT flaps
offer an intermediate thickness of subcutaneous fat
when compared to other donor sites such as the radial
forearm and rectus abdominis flaps. Nakayama et al.
used ultrasonography to measure skin and subcutane-
ous thickness in 31 Asian patients undergoing evalua-
tion for head and neck reconstruction (30). The average
thickness of the flaps was 2.1 mm for radial forearm
flaps, 7.1 mm for ALT flaps, and 13.7 mm for rectus
abdominis flaps. ALT flap thickness correlated well to
body mass index (BMI) for males but not for females,

although their analysis of female patients suffered from
a small sample size. ALT flaps have a thicker subcutane-
ous layer in Western patients when compared to Asians,
and this observation is independent of BMI. Yu exam-
ined the characteristics of ALT flaps in 72 patients of
Western ethnicity (46). He found that ALT flaps were
consistently thicker in the proximal thigh and became
progressively thinner in the distal thigh. In that series
the mean proximal, mid, and distal ALT flap thickness
was 18.3, 15, and 12.5 mm, respectively. Mean flap
thickness in women (19. 9 mm in the midthigh) was sig-
nificantly thicker than mean flap thickness (12.9 mm)
in men. Yu found that flap thickness did indeed corre-
late well to BMI, with the correlation being stronger in
men than in women.
Variations in ALT flap harvest technique can also
greatly influence the volume of subcutaneous tissues
that are transferred. Suprafascial versus subfascial flap
dissection affects flap thickness. The suprafascial tech-
nique produces a thinner cutaneous ALT flap, while the
subfascial flap dissection produces a thicker fasciocu-
taneous ALT flap. With both flap harvest techniques,
the initial flap incision is made along the medial flap
skin paddle margin. With suprafascial flap dissection,
the subcutaneous tissues are elevated laterally in a plane
superficial to the investing fascia of the rectus femoris
muscle to the medial edge of the fascia lata, where cuta-
neous perforators from the LCFA are identified. The
cutaneous flap is then elevated while removing only a
small cuff of the deep fascia that is located immediately
adjacent to the perforators. With the subfascial flap dis-
section technique, the initial skin incision is carried
more deeply through the investing fascia of the rectus
femoris muscle. The rectus femoris muscle is retracted
medially, while the fasciocutaneous component of the
flap that includes the rectus femoris muscular fascia and
fascia lata is retracted laterally. Cutaneous perforators
to the skin component of the flap are then identified at
the point where they pierce the deep surface of the deep
fascia. Using this approach, a moderate or large amount
of deep fascia can be harvested with the flap. Since the
LFCA also provides the dominant blood supply to the
vastus lateralis muscle, a variable amount of vastus lat-
eralis muscle can be harvested with the flap to further
increase flap bulk. When designing a musculocutane-
ous flap that includes a vastus lateralis component to
increase flap bulk, it is important to anticipate that most
of the volume of the vastus lateralis muscle will be lost
secondary to denervation atrophy. When augmentation
is provided by the vastus lateralis muscle, overcorrection
of contour is recommended.
Primary ALT flap thinning has been reported as
a method to reduce the volume of subcutaneous fat
when the defect characteristics favor use of a thin
flap. With this technique, the subcutaneous fat of the
flap is excised, preserving a 3 to 4 mm thick layer of
ANTEROLATERAL THIGH FREE FLAP 237
subcutaneous fat to protect the subdermal vascular
plexus and preserving a 1 to 2 em wide cuff of subcuta-
neous tissues around the skin perforators. It is recom-
mended that flap thinning be done while the flap is still
attached to its vascular blood supply in the thigh, rather
than performing flap thinning during flap insetting in
the head and neck. This allows for improved visualiza-
tion of the cutaneous blood supply to the flap and for
clinical assessment of the flap skin vascularity after thin-
ning is completed. Kimura et al. reported outcomes in
31 patients who underwent primary thinning of ALT
free flaps (18). They concluded that cutaneous circula-
tion was reliable within a radius of 9 em of cutaneous
perforators after primary ALT flap thinning. However,
cautionary reports regarding primary ALT flap thin-
ning by Ross et al. should be noted (34). In their initial
report of four ALT flaps that were thinned primarily,
one flap failed completely and two flaps experienced
partial necrosis. In a follow-up cadaver dissection study,
the anatomy of cutaneous perforators from the LCFA
was examined by performing Indian ink and latex rub-
ber injections (3). These dissections demonstrated an
arterial plexus at the level of the deep fascia, with fur-
ther branches that traveled obliquely through the fat to
reach the subdermal plexus. The authors surmised that
primary flap thinning might disrupt the vascular supply
to the skin and lead to skin necrosis.
ALT flaps have been applied for reconstruction of a
wide variety of soft tissue defects in the head and neck.
The most common indication for ALT flap reconstruc-
tion of the head and neck is after resection of tumors
that arise in the oral cavity, with defects of the tongue,
buccal mucosa, palate, and lips accounting for more
than half of all cases described in a selection of case
series (23,26,36,43). Other common indications for
ALT flap transfer include reconstruction of the pharyn-
goesophageal segment and reconstruction of soft tissue
defects of the skull base, midface, and scalp.
A wide variety of defects of the tongue are amena-
ble to reconstruction using ALT free flaps. Thick, bulky
ALT flaps are useful for tongue reconstruction after total
or near-total glossectomy, while thin ALT free flaps are
useful for tongue reconstruction after hemiglossectomy.
Yu reported outcomes in 13 patients who underwent
ALT free flap tongue reconstruction after total or near-
total glossectomy (4 7). Among 11 patients with retained
larynges and adequate follow-up, all patients regained
intelligible speech, and 6 patients were able to main-
tain all of their nutrition solely through oral nutrition.
Sensory recovery and swallowing outcomes were signifi-
cantly improved among patients who underwent ALT
flap sensory reinnervation by anastomosis of the lateral
femoral cutaneous nerve to the lingual nerve. Farace
et al. examined functional outcomes in 20 patients
who underwent tongue reconstruction after hemiglos-
sectomy using either ALT free flaps or radial forearm
238 CHAPTER 15
free flaps (11). Speech and swallowing outcomes were
not significantly different between the two groups of
patients, indicating that the increased thickness of ALT
flaps compared to radial forearm flaps was of no func-
tional consequence in hemiglossectomy reconstruc-
tion. Agostini and Agostini described a method for oral
reconstruction using an adipofascial ALT flap (1). With
this technique, a standard fasciocutaneous ALT flap was
harvested, and then the skin and superficial subcutane-
ous tissues were excised, leaving a flap that consisted
of vascularized fascia lata covered by a variable amount
of subcutaneous fat that was used to reconstruct oral
cavity mucosal defects. Advantages of this technique
include the ability to precisely match the ALT flap's
volume to that of the defect while avoiding the risk of
skin paddle devascularization that is seen with primary
ALT flap subcutaneous thinning. The authors reported
that the vascularized fascia lata graft prevented scar
contraction of the reconstruction, and the remucosal-
ized adipofascial flap provided a hairless reconstruction
that mimicked the physical appearance and function of
native oral mucosa.
ALT flaps are also useful in the oral cavity for recon-
struction of the buccal mucosa and lips. Satisfactory
restoration of oral competence is achieved in patients
who undergo reconstruction of extensive lip defects
using ALT free flaps that contain vascularized fascia lata
that is used for static suspension of the reconstructed
lip position (21,45). Defects of the buccal mucosa
are effectively reconstructed using ALT flaps (7,31).
Through-and-though defects of the buccal mucosa can
be reconstructed using independent skin paddles for
replacement of buccal mucosa and cheek skin, either by
centering each skin paddle on independent cutaneous
perforators or by de-epithelializing the portion of the
flap that is located between the internal and external
skin paddles. Flap volume can be adjusted as needed,
with thin flaps used for isolated defects of the buccal
mucosa and thick flaps used for reconstruction of high
volume, through-and-through defects of buccal mucosa
and cheek skin.
ALT flaps provide an ample source of skin for
pharyngoesophageal reconstruction in patients who
undergo laryngopharyngectomy. Two methods of ALT
flap insetting are possible to achieve reconstruction of
the pharyngoesophageal segment, which involved use of
either a linear tubed flap or a spiral tubed flap (Figs.
15-17 to 15-29). Genden reported the results of ALT
flap pharyngoesophageal reconstruction in 12 patients
(13). A spiral tubed flap insetting was used in 11 of 12
patients. While one patient experienced perioperative
free flap failure and another patient developed stenosis
of the neopharyngeal segment, all long-term survivors
tolerated an unrestricted oral diet. Tracheoesophageal
speech was achieved in 10 of 11 patients who underwent
tracheoesophageal puncture. Yu and Robb reported the
outcome of ALT flap pharyngoesophageal reconstruc-
tion in 41 patients (49). A linear tubed flap insetting
was used in all patients. In his initial nine patients, a
small segment of flap skin was separated by de-epithe-
lialization and used for postoperative flap monitoring.
However, use of the flap monitoring skin paddle was
abandoned after this technique resulted in a high inci-
dence of fistulas (33%) and strictures (22%). In the fol-
lowing 32 cases without a flap monitoring skin paddle,
fistulas developed in 13% cases, while strictures devel-
oped in 9% cases, with all strictures occurring in patients
with circumferential defects. Successful tracheoesopha-
geal speech was achieved in three of three patients with
a preexisting tracheoesophageal puncture, four of four
patients with a secondary tracheoesophageal puncture,
and six of nine patients with a primary tracheoesopha-
geal puncture. Ultimately, 88% of patients were able
to consume a regular diet. Murray et al. reported the
outcome of ALT flap pharyngoesophageal reconstruc-
tion in 14 patients (29). Reconstructions were done
using flaps that were stented during the postoperative
period using salivary bypass tubes. There were no post-
operative salivary fistulas, and 14% of patients devel-
oped neopharyngeal stenosis. Seventy-nine percent of
patients ultimately achieved an oral diet, and all eight
patients who underwent tracheoesophageal punctures
achieved functional speech.
ALT flaps are becoming increasingly popular for
reconstruction of defects involving the skull base, scalp,
and midface (33). They offer many characteristics that
are well suited for reconstruction of defects in this region.
The long vascular pedicle allows for flap revasculariza-
tion using recipient blood vessels in the neck without
the need to use vein grafts (4,24), In addition, fascia lata
grafts are available for harvest to repair resected dura and
for creation of static slings when necessary. The subcuta-
neous bulk provided by ALT flaps allows for obliteration
of dead space and restoration offacial contour. Multiple
skin paddles can be designed based upon independent
cutaneous perforators or skin paddle de-epithelialization
to reconstruct the complex three-dimensional anatomy
of the midface and skull base (32,39).
NEUROVASCULAR ANATOMY
The blood supply to ALT flaps is derived from cutaneous
branches of the LCFA. The LCFA most often arises as a
proximal branch of the profunda femoris artery. Shortly
after its takeoff from the profunda femoris artery, the
LCFA divides into ascending, transverse, and descend-
ing branches (Fig. 15-3). There are from one to three
cutaneous perforators (average two perforators per flap)
having diameters of 0.5 to 1.5 mm that arise from the
descending branch of the LCFA and/or the transverse
branch of the LCFA. The length of the vascular pedicle
 ANTEROLATERAL THIGH FREE FLAP 239

can vary from 8 to 16 em, depending upon patient stat-
ure, the extent of proximal dissection of the LCFA that
is performed, and the location of the cutaneous perfo-
rators relative to their origin from the LCFA. Perfora-
tors located in the disw thigh provide a longer vascular
pedicle than those that arise in the proximal thigh. The
diameter ofthe proximal LCFA at the site used for micro-
vascular anastomosis is usually about 2.5 mm, while the
diameter of the proximal lateral circumBe:x femoral venae
comitantes is usually about 3.5 mm.
Sensory reinnervation of the skin paddle of the ALT
Bap can be performed using the lateral femoral cutane-
ous nerve. The lateral femoral cutaneous nerve enters
the proximal aspect of the ALT flap in a deep subcuta-
neous plane, immediately above the investing fascia of
the quadriceps muscles. It can be located along a line
that connects the ASIS and the superolateral patella.
Mter identifying the lateral femoral cutaneous nerve at
the proximal margin of the Bap skin paddle, it can be
dissected proximally through the subcutaneous tissues
of the thigh toward the ASIS to provide a sensory nerve
graft that is about 5 em in length (47).
The motor branch of the femoral nerve that innel'-
vates the vastus lateralis muscle runs in close proxim-
ity to the descending branch of the LCFA. This nerve
should be preserved if possible. Distal branches of this
nerve are commonly sacrificed when a musculocutane-
ous ALT flap that includes a significant portion of the
vastus lateralis muscle is harvested. When there is more
than one cutaneous perforator of the LCFA supply-
ing the skin paddle of an ALT Bap, some of the motor
branches to the vastus lateralis muscle may be intel.'l-
twined with the cutaneous perforators. The motor nerve
to the vastus lateralis muscle often runs medial to the

Ascending
LCFA branch of LCFA
LCFA
Transverse Transverse
branch of LCFA branch of LCFA
Profunda
Descending
femoris a
branch of LCFA femoris Descending
branch of LCFA

Rectus femoris
Rectus femoris
muscle
cutaneous muscle
perforators Musculo-
cutaneous
perforator
Vastus lateralis
muscle Vastus lateralis
muscle

A B
FIGURE 15-3. A:. Type I cutaneous perforators arise from the descending branch of1he LCFA and
occur in about 90% of patients. Approximately 90% of type I cutaneous perforators are musculocu-
taneous perforators that take a short intramuscular course through the medial aspect of the vas-
tus latera lis muscle. About 10% of type I cutaneous perforators are septocutaneous perforators
and travel through the intermuscular septum that separates the vastus lateralis and rectus femoris
(RF) muscles. B: Type II cutaneous perforators arise from the transverse branch of1he LCFA. Type
II perforators occur in about 5% to 10% of patients. Most type II perforators are musculocutaneous
perforators thattake a long intramuscular course 1hrough the vastus lateralis muscle. (Continued}
240 CHAPTER 15
//~\
Ascending
~..+--branch of LCFA
Profunda
femoris artery +--11"~~
Rectus femoris
muscle ---+-~;
cutaneous
perforator
c culocutaneous perforators from the transverse branch
FIGURE 15-3. (Continued} C: Type Ill cutaneous perfora-
tors arise directly from 1he profunda femoris artery. Type Ill
perforators occur in about 1% to 5% of patients. Most type
Ill perforators are musculocutaneous perforators thattake
an intramuscular course through the rectus femoris (Rf)
muscle. Most type Ill perforators are of very small caliber
and are unsuitable for microvascular anastomosis.
most proximal perforators and lateral to the most distal
perforators (5). In this instance, it is not possible to pre-
serve all of the perforators as well as the intertwined
motor nerves. It is the author's practice to divide the
intertwined motor nerve during flap elevation and then
perform a neurorrbaphy of the divided motor nerve
after flap harvest bas been completed.
ANATOMIC VARIATIONS
'Yu described a useful method for classification of the
common variants of the vascular anatomy of ALT flaps
(46) (Fig. 15-3). He noted that cutaneous perforators to
the ALT skin derive from three possible sites of origin.
'J:Ype I perforators accounted for 90% of perforators in
his series and originated from the descending branch of
the LCFA. Type n perforators arose from the transverse
branch of the LCFA and accounted for 4% of cutaneous
perforators. 'J:Ype m perforators originated directly
from the profunda femoris artery and accounted for 4%
of cutaneous perforators. Type m perforators were very
small in size (<1 mm in diameter) and were unsuitable
for microvascular anastomosis.
Numerous cadaver dissections and clinical case
series have documented that the cutaneous perforators
from the LCFA more frequently take a musculocuta-
neous course through the vastus lateralis muscle rather
than a septocutaneous course through the intermus-
cular septum that separates the vastus lateralis muscle
from the rectus femoris muscle (36). In the largest case
series to date that reported this data, Wei et al. reported
that 87.1% of 504 fasciocutaneous or cutaneous ALT
flaps were perfused by musculocutaneous perforators,
while 12.9% were pezfused by septocutaneous perfora-
tors (40). Dissection of type I and type n septocutane-
ous perforators is usually straightforward, as the entire
course of the perforators lie within the intermuscular
septum. Type I musculocutaneous perforators take a
relatively short (usually 2 to 3 em) medial-to-lateral
oriented intramuscular course from the descend-
ing branch of the LCFA through the medial edge of
the vastus lateralis muscle before they reach the thigh
skin. Because of their relatively short intramuscular
course, type I musculocutaneous perforators are usually
unroofed and dissected with relative ease. Type n mus-
of the LCFA take a relatively long (up to 10 em) supe-
rior-to-inferior oriented intramuscular course through
the medial aspect of the vastus lateralis muscle before
they reach the thigh skin. Because of their long intra-
muscular course, type n musculocutaneous perforators
are the most difficult to dissect.
POTENTIAL PITFALLS
The rectus femoris muscle receives its blood supply
from branches of the LCFA. There are usually one to
three branches that emerge from the proximal portion
of the LCFA and have a shon anteromedial course
before entering into the rectus femoris muscle. Com-
plete elimination of the LCFA blood supply to the rec-
tus femoris muscle places the rectus femoris muscle at
risk to suffer ischemic necrosis. When dissecting the
proximal segment of the LCFA, branches to the rectus
femoris muscle should ideally be preserved intact if
the LCFA vascular pedicle located distal to the rectus
femoris branches is of sufficient length and adequate
caliber for microvascular anastomosis to the recipient
vessels in the bead and neck. If multiple rectus femo-
ris branches are present, only the most proximal rec-
tus femoris branch needs to be preserved during ALT
flap harvest.

It is well established that the majority of cutaneous
perforators from the LCFA take a musculocutaneous
course through the medial edge of the vastus lateralis
muscle. When harvesting a cutaneous or fasciocutane-
ous perjoraror ALT flap that is perfused by musculocuta-
neous perforators, it is mandatory to identify and dissect
the musculocutaneous perforators. Controversy exists
regarding the need to dissect the intramuscular course
of these musculocutaneous perforators when harvesting
musculocutaneous ALT flaps that include the medial
aspect of the vastus lateralis muscle. In reporting a
series of 672 ALT flaps, Wei et al. transferred 95 mus-
culocutaneous ALT flaps that included a portion of the
vastus lateralis muscle (40). In these cases, Wei et al.
stated that no dissection of the cutaneous perforators
was required to harvest musculocutaneous ALT flaps.
The vastus lateralis muscle cuff technique for muscu-
locutaneous flap harvest was embraced by Lueg, who
reported 34 consecutive cases of ALT flap reconstruc-
tion of the head and neck. In his series, Lueg included
a cuff of the medial edge of the vastus lateralis mus-
cle in all cases, and no effort was made to identify or
dissect the cutaneous perforators (23). There were no
cases of complete flap necrosis and two cases of partial
flap necrosis, which the author attributed to excessive
flap de-epithelialization that disrupted the subdermal
vascular plexus while creating double skin paddled
flaps for through-and-through reconstructions. Mal-
hotra et al. performed cadaver dissections to determine
if the vastus lateralis muscle cuff technique was a safe
approach to harvesting musculocutaneous ALT flaps
(27). They determined that at least one cutaneous per-
forator was preserved intact in all of their 27 dissections
by harvesting a 2 em wide cuff of the medial edge of the
vastus lateralis muscle.
The risk of not performing an intramuscular dissec-
tion of the musculocutaneous perforators while har-
vesting musculocutaneous ALT flaps is evident when
faced with the situation in which type IT musculocuta-
neous perforators arise from the transverse branch of
the LCFA. The intramuscular course of musculocuta-
neous perforators through the vastus lateralis muscle is
not clear by observing their location or appearance as
they enter the skin paddle of the flap during the early
stages of flap harvest, and it cannot reliably be deter-
mined whether musculocutaneous perforators are type
I perforators or type II perforators unless their course
through the vastus lateralis muscle is defined by divid-
ing the overlying vastus lateralis muscle and following
the course of the perforators to their vessel of origin. In
a subsequent publication that described their preferred
technique for harvest of musculocutaneous flaps, Chana
and Wei stated: "it is prudent to determine the course
of the perforator to the source vessel by deroofing the
muscle fibres over the chosen perforator. The perfo-
rator may have a variable and tortuous intramuscular
ANTEROLATERAL THIGH FREE FLAP 241
course; this manoeuver avoids inadvertent injury to
the perforator during incision of the muscle, and also
establishes that the perforator supplies the skin paddle
of the musculocutaneous flap. Determining the course
of the perforator to the source vessel is important since
in 10% of cases the perforator may arise from the trans-
verse branch of the lateral circumflex artery, and enters
the muscle superiorly with a vertical course and is sus-
ceptible to damage during muscle incision at the upper
border of the flap" (6).
With increased clinical experience using both the
muscle cuff technique and the perforator unroofing
technique, the latter technique is now the author's pre-
ferred approach to harvest of musculocutaneous ALT
flaps. Unroofing of musculocutaneous perforators by
dividing the overlying vastus lateralis muscle is not usu-
ally difficult, since most of the muscular side branches
of the perforators arise from the lateral aspect of the
perforators, so that they are less prone to injury while
dividing the muscle that covers medial aspect of the per-
forators. We examined the efficacy of ultrasonic shears to
unroof and dissect musculocutaneous perforators from
the LCFA during harvest of ALT perforator flaps (2).
Successful dissection of intact perforators was achieved
in 27 out of 28 perforators arising from the descending
branch of the LCFA and 9 out of 9 perforators arising
from the transverse branch of the LCFA, for an overall
success rate of 97% for dissection of these musculocu-
taneous perforators. ALT flap viability was 100% with
this technique, with no instances of partial or complete
flap necrosis.
Anatomic variations in the blood supply to the skin
make ALT flap harvest, based upon the LCFA, not
possible in less than 5% of dissections (e.g., cases that
are found to have only type III perforators on explo-
ration). For this reason, it is the author's practice to
inform, consent, and prep and drape all patients to
undergo a bilateral thigh exploration, in case the ini-
tial dissection on one side reveals unfavorable anat-
omy. Another option in this situation is to harvest a
myofascial vastus lateralis flap based upon the LCFA
vascular pedicle. The vastus lateralis muscle can be
covered with a skin graft to achieve a very satisfactory
aesthetic outcome for external skin reconstruction
(25). Alternatively, vascularized vastus lateralis mus-
cle can be used for intraoral mucosal reconstruction,
with flap remucosalization occurring during the post-
operative period (41). When a skin paddle is needed
but the vascular anatomy of the LCFA proves to be
unfavorable for harvest of a skin paddle, then the dis-
section can be converted to harvest of an anterome-
dial thigh flap (35) or tensor fascia lata flap (9). Either
the anteromedial thigh flap or the tensor fascia lata
flap can be harvested while using the same initial skin
incision that is used to expose the LCFA during ALT
flap harvest.
242 CHAPTER 15
Long-term flap donor site and recipient site com-
plications appear to be rare after ALT flap reconstruc-
tion of the head and neck. Most case series describe
a very acceptable degree of donor site morbidity after
harvest of ALT free flaps, with the flap donor site
rarely resulting in problems that affect daily activities
(1,3,7,13,23,26,34). Kimata et al. noted increased
donor site morbidity in cases where wide flaps required
skin graft reconstruction of the donor site defect and
in cases where there was damage to the vastus lateralis
muscle (16). To the contrary, Lipa et al. found no cor-
relation between the occurrence of donor site morbidity
and the need to perform skin grafts or an intramuscular
dissection of the perforators through the vastus lateralis
muscle (22). However, persistent leg weakness was sig-
nificantly associated with extensive elevation and har-
vest of the leg fascia with the flap, as opposed to only a
very limited fascial sacrifice near the pedicle.
With regards to long-term recipient site wound com-
plications, thigh skin is a poor match of facial skin with
regard to skin color, thickness, texture, and hair growth
patterns (28). This problem is more pronounced in
patients who receive radiation therapy and in patients
of Western ethnicity as opposed to patients of Asian
ethnicity.
PREOPERATIVE ASSESSMENT
ALT flaps are contraindicated in patients with past thigh
surgery in whom cutaneous branches of the LFCA
may have been divided during elevation of the thigh
skin from the underlying quadriceps muscles. The util-
ity of the ALT flap in patients with peripheral vascular
disease remains controversial. As a branch of the pro-
funda femoris system, the LCFA is relatively spared by
peripheral vascular disease in patients who have devel-
oped vasoocclusive disease affecting the superficial
femoral system. Indeed, this recognition has led to the
use of the LCFA as an arterial graft source in patients
who require coronary artery bypass grafting (10). How-
ever, Hage and Woerdeman reported a case of a patient
with a history of intermittent claudication and no pal-
pable popliteal pulse who developed distal leg ischemia
that resulted in soft tissue necrosis of the toes and calf
after harvest of an ALT free flap (14). In this patient,
the descending branch of the LCFA likely provided a
major source of collateral blood flow to the distal lower
extremity in the face of severe peripheral vascular dis-
ease that resulted in occlusion of the superficial femoral
artery. The authors recommended preoperative angiog-
raphy before harvesting ALT flaps in patients who have
no palpable popliteal pulse.
POSTOPERATIVE CARE
When primary wound closure is achieved at the thigh
wound donor site, a subcutaneous closed suction drain
is placed and removed when the output is less than 30 cc
over a 24-hourperiod.Wounds that are closed by skin graft
are bolstered using a cotton and bismuth-petrolatum
gauze dressing for 5 to 7 days after surgery. Thereafter,
the skin graft recipient site wound is dressed daily with
a bismuth-petrolatum gauze dressing until the skin graft
site is matured and fully epithelialized. The patients are
allowed to weight bear on the leg of harvest as tolerated
by discomfort during the postoperative period.
 ANTEROLATERAL THIGH FREE FLAP 243

Anterolateral Thigh Flap

FIGURE 15-4. The donor thigh is positioned

in a neutral position without internal or exter-
nal rotation. The skin paddle of the ALTflap is
roughly centered on a line that is drawn from
the AS IS to the lateral border of the patella.
This line marks the position of the intermuscu-
lar septum that separates the vastus lateralis
muscle from the rectus femoris muscle. In this
illustration, cutaneous perforators of the LCFA
are shown to be arising from the descending
branch of the LCFA (type I perforators). This
is the most common anatomic variant and is
encountered in about 90% of dissections. The
most common perforators arise near the mid
point of the line that is drawn from the AS IS to
the patella and is denoted by the letter B in this
illustration. Most flaps will be supplied by one to
three perforators from the LCFA that are usually
found within a 5-cm radius of the point marked
by point B. In this illustration, the letters A and C
denote the upper and lower limits of that 5-cm
range of perforator location.

FIGURE 15-5. The initial skin incision is made

along the medial aspect of the flap skin paddle.
In this case, the incision is carried through the
subcutaneous fat and through the investing
fascia ofthe rectus femoris muscle to harvest
a fasciocutaneous flap. The deep fascia and
skin paddle are reflected laterally off the rectus
femoris muscle until the intermuscular septum
between the rectus femoris and vastus lateralis
muscle, denoted by a yellow line of fat (arrow)
along the lateral aspect of the rectus femoris
muscle, is identified.
244 CHAPTER 15

Anterolateral Thigh Flap

FIGURE 15-6. A: At this point in the harvest,

the cutaneous perforators to the skin paddle of
the flap are identified immediately deep to the
fascia of the flap skin paddle. In this case, two
perforators are marked by red markers. B: Both
perforators are musculocutaneous perforators
that are traveling through the vastus latera lis
muscle, which is the muscle seen immediately
lateral to the intermuscular septum (arrowl. B
 ANTEROLATERAL THIGH FREE FLAP 245

Anterolateral Thigh Flap

FIGURE 15-7. A: The rectus femoris muscle is

retracted medially to expose the descending
branch of the LCFA (srrow), which runs in the
intermuscular septum and along the superficial
aspect of the vastus intermedius muscle. B: A
close up view of the LCFA is demonstrated. The
pedicle is traced proximally in the septum to
achieve greater length and diameter. B

FIGURE 15-8. At. this point, the course of

the musculocutaneous perforators is 1raced
by performing a distal to proximal perforator
dissection. The courses of the perforators are
unroofed by lifting the vastus lateral is muscle
that lies medial to the perforators.
246 CHAPTER 15

Anterolateral Thigh Flap

FIGURE 15-9. After lifting the vastus latera lis

muscle off the perforators, the vastus lateral is
muscle is divided and any small blood vessels
contained within the vastus lateralis muscle are
simultaneously sealed using ultrasonic shears.
Alternatively, hemostasis can be achieved
using bipolar electrocautery or hemaclips.

FIGURE 15-10. A: The portion of the vastus

lateralis muscle overlying bath musculocuta
neous perforators has been divided. B: This
has led to exposure of the full course of bath
perforators (black arrows), which are arising
from the descending branch of the LCFA (whfte
arrow). These are type /musculocutaneous
perforators, which is the most common ana-
tomic variant that is encountered during flap
harvest B
 ANTEROLATERAL THIGH FREE FLAP 247

Anterolateral Thigh Flap

FIGURE 15-11. ALT flap dissection (right leg)

in a different cadaver after unroofing of two
musculocutaneous perforators by division of
the overlying vastus lateralis muscle reveals
that the distal perforator is a type I perfora-
tor (black arrow) arising from the descending
branch LCFA. while the proximal perforator is a
type II perforator (white Bffowt arising from the
transverse branch of the LCFA. type II perfora-
tors account for 5% to 10% of the perforators
encountered during ALT flap harvest.
FIGURE 15-12. The lateral dissection is per-
formed after the medial dissection is completed
with successful dissection of the vascular
pedicle. In the instance where ALT flap elevation
is unsuccessful secondary to a lack of suit-
able cutaneous perforators from the LCFA, the
dissection can be converted to harvest of an
anteromedial thigh flap, a tensor fascia lata flap,
or the anterior thigh skin incision can be closed
in favor of performing a contralateral thigh ALT
flap harvest. After a favorable vascular pedicle
has been confirmed by the medial dissection, the
lateral skin incision is made to the fascia lata. A
variable amount of fascia lata can be harvested.
In this instance, a medial subcutaneous dissec-
tion is performed until the perforators are visual-
ized, and a small cuff offascialata is harvested
immediately adjacent to the perforators.

FIGURE 1513. The perforators are dis-

sected free from the vastus lateralis muscle.
A 5 to 10-mm cuff of vastus latera lis muscle
is harvested adjacent to the perforators to
harvest a perforatorfasciocutaneous flap. A
larger amount of vastus latera lis muscle can be
harvested in order to create a bulkier, muscu-
locutaneous flap. The vastus latera lis muscle
is divided and small muscular side branches of
the perforators are simultaneously sealed using
ultrasonic shears. Alternatively, bipolar electro-
cautery and hemoclips can be used to maintain
hemostasis during this dissection.
248 CHAPTER 15

Anterolateral Thigh Flap

FIGURE 1514. The motor nerve branches of

the femoral nerve (arrow) to the vastus lateralis
muscle is dissected and separated from the
vascular pedicle.

FIGURE 1515. Flap harvest of a fasciocutane-

ous perforator ALT flap that is perfused by two
type I musculocutaneous perforators has been
completed. The length of the vascular pedicle
usually varies from 8 to 16 em.

FIGURE 1516. View of the deep surface of

the harvested flap reveals a small cuff of deep
fascia that is preserved immediately adjacent to
the two cutaneous perforators.
 ANTEROLATERAL THIGH FREE FLAP 249

Inset of the Anterolateral Thigh Rap for Circumferential

PharyngoesophageaiReconshuction
----LDngitudinal orientation

FIGURE 1~11. A circumferential defect of the

pharyngoesophageal segment is shown after
tota IIaryngopha ryng ectomy with preservation
of the cervical esophagus. The black arrow
indicates the lumen of the oropharynx. The
blue arrow indicates the lumen of the cervical
esophagus.

FIGURE 15-18. The cervical esophagus is

spatulated by making a longitudinal linear inci-
sion for a length of about 2 em. This enlarges
the circumference and breaks up the linear
nature of the distal enteric anastomosis,
thereby reducing the risk of stricture forma-
tion at the anastomosis of the ALT flap to the
cervical esophagus.
250 CHAPTER 15

Inset of 1he Anterolateral Thigh Flap for Circumferential

PharvngoesophageaiReconsbuction
Longitudinal orientation

FIGURE 1519. Flap design is demonstrated in red ink for linear ALT flap tubing for pharyngoesophageal reconstruction. The
precise position of the flap along the axis (drawn from the AS IS to the lateral patella) is adjusted after flap harvest, based
upon 1he location of cutaneous perforators. The flap dimensions are adjusted according to the size of the pharyngoesopha
geal defect. With this design, the proximal flap wid1h equals 1he circumference of the proximal enteric anastomosis. Since
the circumference of the oropharynx is larger superiorly than it is inferiorly, 1his width can commonly vary from 8 to 14 em.
The distal flap width equals the circumference of the cervical esophagus before it is spatulated. This width is commonly
6 to 7 em. The leng1h of the flap is equal to the length of the pharyngoesophageal defect. The length of the triangular exten-
sion at1he distal end of the flap is equal to the length of the longitudinal incision that was made to spatulate the cervical
esophagus. An elliptical flap is harvested that encompasses the flap that will be used for pharyngoesophageal reconstruc-
tion. This design facilitate thigh wound closure and provides extra flap skin that can be used as an external skin paddle to
monitor postoperative flap perfusion. A disadvantage of this flap design arises when the width of the required flap exceeds
Sto 10 em. In this instance, primary closure of the thigh wound defect may not be possible, and skin graft reconstruction of
the thigh donor site may be necessary.

FIGURE 15-20. The flap is shown after de

epithelialization according to the dimension
described in Figure 15-19. The skin paddle is
centered on two skin perforators from the LCFA.
A small distal skin paddle (arrow) is included to
serve as an external flap monitor paddle.
 ANTEROLATERAL THIGH FREE FLAP Z51

Inset of the Anterolateral Thigh Rap for Circumferential

PharyngoesophageaiReconshuction
----LDngitudinal orientation
FIGURE 1~21. The flap is tubed longitudinally
ever a salivary bypass tube. In the author's
experience, stenting the reconstruction with a
size 10- or 12-mm salivary bypass tube reduces
the incidence of pestoperative salivary pha-
ryngocutaneous fistulas. One end of the red
rubber catheter is sutured te the praximal end
of the salivary bypass tube, while the ether
end is sutured to the caudal nasal septum. This
technique is applied to help prevent migration
of the salivary bypass tube and to facilitate its
removal at the bedside. The salivary bypass
tube is removed about 2 weeks after surgery
by cutting the nasal septum suture, grasping
the red rubber catheter in the oropharynx using
a tonsil clamp, and withdrawing the salivary
bypass tube through the mouth.

FIGURE 1~22. Longitudinal flap tubing is

completed.

FIGURE 1~23. The tubed flap is inset into the

pharyngoesophageal defect. The flap vascular
pedicle {blue arrow) is braught into the right
neck for anastomosis to cervical recipient blood
vessels. The flap monitoring skin paddle is inset
into the right neck suture line.
252 CHAPTER 15

Inset of 1he Anterolateral Thigh Flap for Circumferential

PharvngoesophageaiReconsbuction
~Djri'IJI orientation

FIGURE 15-24. Skin paddle design for a spiral

ALT flap tubing demonstrates a long, narrow
flap. The flap width is determined by thigh skin
laxity that allows for primary closure of the
thigh donor site wound and is frequently in
the range of 8 to 10 em. The length of the flap
should exceed the sum of the circumference
of the proximal and distal enteric anastomoses
and the linear length of the pharyngoesopha-
geal reconstruction. This is frequently in the
range of about 30 em.

FIGURE 15-25. The ALT flap is tubed using a

spiraling configuration. In this example, the skin
of the distal tip of the flap is sutured to the distal
lateral edge of the flap to create a distal flap
lumen circumference that matches the circum-
ference of the spatulated cervical esophagus.
 ANTEROLATERAL THIGH FREE FLAP 253

Inset of the Anterolateral Thigh Flap for Circumferential

Pharyngoesophageal Reconstruction
Spkmorie~6on ____________

FIGURE 15-2&. The skin of the proximal tip

of the flap is sutured to the proximal medial
edge of the flap to create a proximal flap lumen
circumference that matches the circumference
of the oropharyngeal lumen.

FIGURE 15-27. Spiral flap tubing continues by

suturing the medial flap skin edge to the lateral
flap skin edge. This view shows that the tubing
of the flap has almost been completed.
254 CHAPTER 15

Inset of 1he Anterolateral Thigh Flap for Circumferential

PharvngoesophageaiReconsbuction
~Djri'IJI orientation

FIGURE 15-28. Flap tubing has been

completed over a salivary bypass tube.

FIGURE 15-29. The spiral-tubed flap has been

inset into the pharyngoesophageal defect.
The vascular pedicle {arrow) is brought into
the right neck for anastomosis to right-sided
cervical recipient blood vessels.

REFERENCES
1. Agostini T, Agostini V: Further experience with adipo-
fascial ALT flap for oral cavity reconstruction. J Plast
Recomtr Aesthet Surg 2008;61: 1164-1169.
2. Aluned S, Sidell D, Blackwell KE, SercaiZ JA, Abemayor E,
NabiliV: The use of ultrasonic shears for the harvest ofperfo-
rator free flaps. Arch Facial Pfarr Surg. 2009;11 (5):343-346.
3. Alkureishi L\V, Shaw-Dunn J, Ross GL: Effects of thin-
ning the anterolateral thigh flap on the blood supply to
the skin. Br J Plast Surg 2003;56:40 1-408.
4. Amin A, Rifaat M, Civaatos F, Weed D, Abu-sedira M,
Bassiouny M: Free anterolateral high flap for reconstruc-
tion of major craniofacial defects. J Recomtr Microsurg
2006;22:97-104.
5. CaseyWJ, Rebecca AM, SmithAA, Craft RO, Hayden RE,
Buche! EW: Vastus lateralis motor nerve can adversely
affect anterolateral thigh flap harvest. Plast Recomtr Surg
2007;120:196-201.
6. Chana JS, Wei FC: A review of the advantages of the
anterolateral thigh flap in head and neck reconstruction.
Br J Plast Surg 2004;57:603--609.
7. Chen CM, Chen CH, Lsi CS, lin SD, Huang IY, Shieh
TY: Anterolateral thigh flaps for reconstruction of head
and neck defects. J Oral MaxiJJojac Surg 2005;63:948-952.
8. Choi SW, Park JY, Hur MS, Park HD, Hu KS, Kim HJ:
An anatomic assessment on perforators of the lateral
circumflex femoral artery for anterolateral thigh flap.
J Crani<Jjac Surg 2007;18:866-871.
9. Coskunfirat OK, Ozkan 0: Free tensor fascia lata perfO-
rator flap as a backup procedure for head and neck recon-
struction. Ann Plast Surg 2006;57: 159-163.
10. Fabbrocini M, Fattouch K, Carnporini G, et al.: The
descending branch of lateral femoral circumflex artery
in arterial CABG: early and midterm results. Ann Thorac
Surg 2003; 75: 1836-1841.
11. Farace F, Fois VE, Manconi A, et al.: Free anterolateral
thigh flap versus free forearm flap: functional results
in oral reconstruction. J Plan Reconstr Aesrlzet Surg
2007 ;60:583-587.
12. Gebebou TM, Wei FC, LinCH: Clinical experience of
1284 free anterolateral thigh flaps. Handckir Mikrockir
Plast Chir 2002;34:239-244.
13. Genden EM, Jacobson AS: The role of the anterolateral
thigh flap for pharyngoesophageal reconstruction. Arch
Otolaryngol Head Neck Surg 2005;131:796-799.
14. Hage D, Woerdeman IA: Lower limb necrosis after use of
the anterolateral thigh free flap: is preoperative angiogra-
phy indicated? Ann Plast Surg 2004;52:315-318.
15. Kimata Y, Uchiyama K, Ebihara S, Nakatsuka T, Harii
K: Anatomic variations and technical problems of the
anterolateral thigh flap. A report of74 cases. Plast Recon-
str Surg 1998;1 02:1517-1523.
16. Kimata Y, Uchiyama K, Ebihara S, et al.: Anterolateral
thigh flap donor-site complications and morbidity. Plast
Recomtr Surg 2000;106:584-589.
ANTEROLATERAL THIGH FREE FLAP 255
17. Kimata Y, Uchiyama K, Ebihara S, et al.: Versatility of
the free anterolateral thigh flap for reconstruction of
head and neck defects. Arch Otolaryngol Head Neck Surg
1997; 123:1325-1331.
18. Kimura N, Satoh K, Hasurni T, Ostuka T: Clinical
application of the free thin anterolateral thigh flap in
31 consecutive patients. Plast Recomtr Surg 2001;108:
1197-1208.
19. Koshima I, Fukuda H, Utunomiya R, Soeda S: The
anterolateral thigh flap: variations in its vascular pedicle.
Br J Plast Surg 1989;42:260-262.
20. Koshima I, Fukada H, Yamamoto H, Morigushi T,
Soeda S, Ohta S: Free anterolateral thigh flaps for recon-
struction of head and neck defects. Plast Recomtr Surg
1993;92:421-428.
21. Kuo YR, Jeng SF, Wei FC, Su CY, Chien CY: Functional
reconstruction of complex lip and cheek defects with free
composite anterolateral high flap and vascularized fascia.
Head Neck 2008;30: 1001-1006.
22. Lipa JE, Novak CB, Binhammer PA: Patient-reported
donor-site morbidity following anterolateral thigh free
flaps. J Recomtr Micorsurg 2005;21 :365-370.
23. Lueg EA: The anterolateral thigh flap: radial forearm's "big
brother" for c:xtensive soft tissue head and neck defects.
Arch Otolaryngol Head Neck Surg 2004;130;813-818.
24. Lutz BS: Aesthetic and functional advantages of the
anterolateral thigh flap in reconstruction of tumor-related
scalp defects. Microsurgery 2002;22:258-264.
25. Lutz BS: Beauty of skin-grafted muscle flaps in head and
neck reconstruction. Microsurgery 2006;26:177-181.
26. Makitie AA, Beasley NJ, Neligan PC, Lipa J, Gullane
PJ, Gilbert RWL: Head and neck reconstruction with
anterolateral thigh flap. Otolaryngol Head Neck Surg
2003; 129:547-555.
27. Malhotra K, Lian T, Chakradeo V: Vascular anatomy
of the anterolateral thigh flap. Laryngoscope 2008;118:
589-592.
28. Mureau MA, Posch NA, Meswis CA, Hofer SO: Antero-
lateral thigh flap reconstruction of large c:xternal facial
skin defects: a follow-up study on functional and aes-
thetic recipient- and donor-site outcome. Plast Recomtr
Surg 2005;115:1077-1086.
29. Murray DJ, Gilbert RW, Versely MJ, et al.: Functional
outcomes and donor site morbidity following circum-
ferential pharyngoesophageal reconstruction using an
anterolateral thigh flap and salivary bypass tube. Head
Neck 2007;29:147-154.
30. Nakayama B, Hyodo I, Hasegawa Y, et al.: Role of the
anterolateral thigh flap in head and neck reconstruction:
advantages of moderate skin and subcutaneous thickness.
J Recomtr Microsurg 2002;18:141-145.
31. Ozkan 0, Mardini S, Chen HC, Cigna E, Tang WR, Liu
YT: Repair of buccal defects with anterolateral thigh
flaps. Microsurgery 2006;26: 182-189.
32. Rodriguez-Vegas J, Angel PA, Manuela PR: Refining the
anterolateral thigh free flap in complex orbitomaxillary
reconstruction. Plast Reconstr Surg 2008; 121 :481-486.
256 CHAPTER 15
33. Rosenthal EL, King T, McGrew BM, Carrol W,
Magnuson JS, Wax MK: Evolution of a paradigm for free
tissue reconstruction of the lateral temporal bone. Head
Neck 2008;30:589-594.
34. Ross GL., Dunn R. Kirkpatrick J, et a.L: To thin or not to
thin: the use of the anterolateral thigh flap in the reconstruc-
tion of intraoral defects. BrJ Plast Surg 2003;56:409-413.
35. Schoeller T, Huemer GM, Shafighi M, Gurunluoglu R.
Wechselberger G, Piza-Katzer H: Free anteromedial
thigh flap: clinical application and review of the literature.
Microsurgery 2004;24:43--48.
36. Shieh SJ, Chiu HY, Yu JC, Pan SC, Tsai ST, Shen CL:
Free anterolateral thigh flap for reconstruction of head
and neck defects following cancer ablation. Plast Recomtr
Surg 2000; 105:2349-2357 _
37. Shimizu T, Fisher DR. Carmichael SW, Bite U: An ana-
tomic comparison of septocutaneous free flaps from the
thigh region. Ann Plast Surg 1997;38:604-610.
38. SongYG, Chen GZ, SongYL:The free thigh flap: a new
free flap concept based on the septocutaneous artery.
Br J Plast Surg 1984;37:149-159.
39. Wang X, Qiao Q, Liu Z, et al.: Free anterolateral thigh
adipofascial flap for hemifacial atrophy. Ann Plast Surg
2005;55:617-622.
40. Wei FC, Vivek Jain MC, Celik N, Chen HC, Chuang
DC, Un CH: Have we found an ideal soft-tissue flap? An
experience with 672 anterolateral thigh flaps. Plast Recon-
str Surg 2002; 109:2219-2226.
41. Wolff KD: Indications for the vastus lateralis flap in
oral and maxillofacial surgery. Br J Oral Maxi/lajac Surg
1998;36:358-364.
42. Wolff KD, Grundmann A: The free vastus lateralis flap:
an anatomic study with case reports. Plast Reconstr Surg
1992;89:469--47 5-
43. Wolff KD, Kesting M, Thurmuller P, Bockmann R.
Holzle F. The anterolateral thigh as a universal donor
site for soft tissue reconstruction in maxillofacial surgery.
J Craniomaxillojac Surg 2006;34:323--331.
44. Xu DC, Zhong SZ, Kong JM, et al.: Applied anat-
omy of the anterolateral thigh flap. Plast Recomrr Surg
1988;82:305-310.
45. Yildirim S, Gideroglu K, Aydogdu E, Avci G, Akan M,
Akkoz T: Composite anterolateral thigh-fascia lata flap:
a good alternative to radial forearm-palmaris longus
flap for total lower lip reconstruction. Plast Recomtr Surg
2006;117 :2033--2041.
46. Yu P: Characteristics of the anterolateral thigh flap in a
western population and its application in head and neck
reconstruction. Head Neck 2004;26:759-769.
47. Yu P: Reinnervated anterolateral thigh flap for tongue
reconstruction. Head Neck 2004;26: 1038-1044.
48. Yu P, Adel Y: Efficacy of the handheld Doppler in
preoperative identification of the cutaneous perfora-
tors in the anterolateral thigh flap. Plast Reconstr Surg
2006;118:928-933.
49. Yu P, Robb GL: Pharyngoesophageal reconstruction with
the anterolateral thigh flap: a clinical and functional out-
comes study. Plast Recomtr Surg 2005;116:1845-1855.P
INTRODUCTION
With the rise in popularity of the fibular osteocuta-
neous free flap, the lower leg has become one of the
most common donor sites for head and neck recon-
struction. A number of other lower leg donor sites have
been described, including the peroneal fiap, gastrocne-
mius flap, saphenous Bap, and posterior leg flap (17).
Although the other lower leg flaps have not gained
popularity in head and neck reconstruction, there has
been increasing interest in the use of the posterior tibial
artery (PTA) flap.
The PTA flap was initially descnoed by Hwang
et al. in 1985 who referred to it as the medial leg skin
Bap (5). This group reported on 17 cases with 100%
Bap survival rate. In this initial description, the flap was
used in all cases to cover post-traumatic tissue defects
of the extremities. Although Hwang et ai.a description
in 1985 was reported as a new medial leg skin fiap,
Zhang also described a similar flap in the OUnese lit-
erature in 1983 (19).
Based on the PTA, with the venous supply arising
from the paired <venae comitantes, this fiap is predomi-
nandy used as a fascial or fasciocutaneous flap. It can
also be elevated as a sensate flap with incoxporation of
the sensory nerve supply through the saphenous nerve.
Although there has been limited experience with this
flap in head and neck reconstruction, a number of
advantages have been proposed. These include large
flap dimension, relatively thin and Bez.ible tissue, and
long pedicle length with large-cahoer vessels. As it is a
lower leg donor site, the akin graft is less conspicuous
257
258 CHAPTER 1&
than that on the volar aspect of the upper extremity fol-
lowing harvest of a radial forearm flap. Also, there is lit-
tle anatomic variation in the vascular anatomy of this
area. Potential disadvantages include the placement of a
skin graft over the tibia in a vulnerable location that may
be at risk of noDhealing following trauma. In addition,
the sacrifice of one of the dominant blood supplies to
the foot may not be advisable in the event oflater devel-
opment of atherosclerosis that places the foot at risk of
ischemia. To date, these complications are theoretical
and have not been reported in the literature.
FLAP DESIGN AND UTILIZATION
The territory of the PTA free flap (Fig. 16-1) has been
investigated through a number of anatomic studies
(4,8,15,18). Although the maximum possible dimen-
sions of the skin paddle have not been specifically iden-
tified, it is felt that skin paddles of up to 19 X 13 em
may be safely raised on a single perforator (6). Chen
et al. reported the use of a 30 x 10 em PTA flap for
esophageal reconstruction (1). Venous insufficiency has
been described as a possible complication in the case of
large pedicled flaps based on the distal perforators of
the PTA. Ozdemir et al. described that three of eight
patients developed venous congestion postoperatively,
with one patient having partial flap loss (15). The flaps
with venous congestion ranged from 18 to 24 em in
maximum length. The authors felt that increased flap
size may have contributed to the venous congestion.
However, in this report, flaps were designed as distally
based, pedicled flaps, not free tissue transfers, and likely
do not reflect the vascular physiology when this flap is
used as a free tissue transfer. Venous insufficiency has
not been reported as a complication following transfer
of the PTA free flap.
Preoperative evaluation of the donor site is impor-
tant. The skin and subcutaneous tissue characteristics
are important to identify prior to utilizing this free flap
for complex head and neck reconstructions. Although
the area usually provides thin and pliable tissue, in
patients with a large body habitus, there may be signifi-
cant amounts of subcutaneous fat. This can make deli-
cate head and neck reconstructions challenging and
direct the reconstructive surgeon away from the tissue
of this region. When designing the flap, our experience
has been that the tissue from the anterior portion of
the flap over the tibia tends to be thinner and more pli-
able (Fig. 16-2). This tissue can be used preferentially
where thin pliable tissue is advantageous. Similarly,
the portion of the flap that extends posteriorly tends
to have more abundant subcutaneous tissue. This
area can be preferentially harvested in circumstances
where an increased volume of tissue is required, such
as for recontouring procedures. The predictable vascu-
lar anatomy and the presence of multiple subcutane-
ous perforators allow significant freedom with regard
to manipulation and contouring of the flap, as well as
the potential harvest of independent skin paddles. The
lower leg donor site facilitates elevation by a separate
team of surgeons.
Despite these advantages, there have been limited
descriptions of using this flap in head and neck recon-
struction. As stated above, one of the earliest descrip-
tions was its use for esophageal reconstruction as a large
tubed free flap (1). A recent report looking at the PTA
flap specifically for head and neck reconstruction was
published by Ng et al. (13). This paper reported on
11 patients having PTA free flap reconstruction with
100% flap survival. Nine of these flaps were used for
intraoral mucosal lining reconstructions. The remaining
two were used for external skin coverage. Flap dimen-
sions ranged from 6 x 8 em to 8 x 12 em. The authors felt
that the advantages of this flap were significant and that
it may, in many circumstances, replace the radial fore-
arm free flap for head and neck reconstruction. Another
recent report in the Chinese literature described the
"medial leg fascial cutaneous flap" for reconstruction
of base-of-tongue defects in four patients. They also
reported a 100% free flap survival rate and an accept-
able donor site morbidity.
Although proposed as an advantage of the PTA
free flap, the incorporation of the saphenous nerve
for sensation has not been specifically reported on in
head and neck reconstruction. We have successfully
incorporated the saphenous nerve with the PTA free
flap for sensate reconstruction of a hemiglossectomy
defect with good functional outcomes, and we have
previously shown the importance of reinnervation in
tongue reconstruction (14). Interestingly, in the initial
description of this flap by Hwang et al., the sural nerve
was incorrectly identified as the sensory nerve running
with the saphenous vein to supply this free flap. In fact,
the saphenous nerve is identified running with the long
saphenous vein. This nerve can be easily incorporated
into the flap design. Zhang et al. indicated that they
used this nerve for sensate reconstruction of hand and
foot defects (20).
NEUROVASCULAR ANATOMY
The lower leg is supplied primarily by the posterior tibial,
anterior tibial, peroneal, and sural arteries. The popliteal
artery, originating in the thigh, provides several branches
including the anterior and posterior tibial arteries.
The PTA arises at the distal border of the popliteus
muscle, between the tibia and the fibula, and descends
medially in the flexor compartment of the lower leg
 POSTERIOR TIBIAL ARTERY FREE FLAP 259

RGURE 16-1. The size and shape of1he posterior tibial artery flap vary with the defect. The axis of
the flap should be centered over1he course of the transverse intennuscular septum around zone
II of the lower leg as described by Wu et al. and described above. The flap may include most of1he
medial aspect of1he lower leg. There are regional differences in the 1hickness of1he subcutaneous
tissue. with the 1hinnestflaps harvested over the tibia and thicker flaps designed more posteriorly.
2&0 CHAPTER 16

-~r+-1-Gastrocnemius
muscle

\T'-'t - - t - + - - - Soleus
muscle

+ - l - H t - - - - Flexor
digitorum
longus
muscle

RGURE 1&-2. Medial view of lower leg, demonstrating the tibia, flexor digitorum longus muscle,
and soleus. The transverse intermuscular septum. through which the perforators pass, lies
between the soleus and the flexor digitorum longus muscle, as demonstrated in Figure 16-4A.
Palpating the posterior medial aspect of the tibia can identify the position of this septum.

Posterior
tibial artery - - - +-Ht-1!+-a
Saphenous
nerve-----+-++V
Long
saphenous
vein-----411-++H
FIGURE 16-3. The neurovascular supply of the medial
aspect of the lower leg is shown. The posterior tibial
artery (PTA) arises at the distal border of the popliteus,
between the tibia and the fibula, and descends medially in
the flexor compartment. The PTA passes under the fibrous
arch of the soleus muscle and runs distally between the
soleus and the flexor digitorum longus. Along its course,
the artery provides direct cutaneous branches that supply
the posterior tibial free flap. The saphenous nerve gener-
ally lies anterior to the long saphenous vein. Both of these
structures can be preserved or harvested with the flap, as
required.
POSTERIOR TIBIAL ARTERY FREE FLAP 261
(Fig. 16-3). The PTA passes under the fibrous arch of
the soleus muscle and I'Uil8 distally between the soleus
and the flexor digitorum longus muscles. Along its
course, the artery provides direct cutaneous branches
(Fig. 16-4B) that supply the posterior uoial free flap, as
well as muscular branches to supply adjacent muscles.
It is commonly accompanied by two venae comitantes,
referred to as the posterior tibial veins. It bifurcates
midway between the medial malleolus and the medial
tubercle of the calcaneus into the medial and lateral
plantar arteries, which supply the foot (3).
The PTA lies posterior to the tibialis posterior and
flexor digitorum longus muscles and behind the tibia and
the ankle joint. In the praximal por1ion ofthe lower leg, the
gastrocnemius, soleus, and deep t:nu:lSVei'Se fascia of the
leg are superficial to the PTA, whereas distally the vessel
becomes much more superficial and is covered only by the
skin and fascia.'IWo wnae comittmtes and the uDial nerve
accompany the PTA. The tibial nerve originates medially
but later crosses posterior and remains largely posterolat-
eral to the artery in the foot. Terminally, the artery is deep
to the flaor retinaculum and abductor hallucis.
The branches of the PTA include the peroneal and
circumflex fibular arteries as well as the nutrient artery
of the tibia, and the medial and lateral plantar arter-
ies. In addition, the PTA gives rise to the perforating
branches to the skin and periosteum and muscular
branches mentioned above. Both the circumflex fibular
artery and the nutrient artery of the tibia arise near the
takeoff of the peroneal artery from the PTA and as such
may be preserved during flap dissection. The circum-
flex fibular artery anastomoses with the lateral inferior
genicular artery, the medial genicular artery, and the
anterior tibial recurrent arteries. It supplies bone and
articular structures around the knee. The nutrient artery
of the tibia arises from the PI'A near its origin. It sup-
plies muscular branches to the soleus and deep flexors
of the lower leg. The communicating branch of the PTA
runs posterior and deep to the flez.or hallucis longus,
across the uoia, approximately 5 em above its distal end
to join a branch of the peroneal artery. The calcaneal
branches of the nutrient artery arise just proximal to the
terminal division of the PI'A and penetrate the flexor
retinaculum posterior to the tendocalcaneus muscle.
These branches anastomose with the medial malleolar
arteries and calcaneal branches of the peroneal artery.
The peroneal artery arises approximately 2.5 em
distal to the popliteus, high in the posterior compart-
ment. It descends between the tibialis posterior mus-
cle and the flexor hallucis longus to the tibiofibular
syndesmosis, where it subsequently divides into calca-
neal branches. This artery supplies a nutrient branch
to the fibula and a perforating branch that pierces the
interosseous membrane about 5 em proximal to the
262 CHAPTER 16

~~......----Tibialis
anterior muscle

+-'~"~-llblalls
Flexor
posterior musde
digitorum
longus mus.---+fl~

Posterior
tibial artery,
vein and nerve

~~~----llblalls
anterior muscle

+"r-o\-r--llblalls
posterior musde

Flexor
digitorum
longusmus.

B
RGURE 16-4. A:. The cross sectional anatomy of the lower leg demonstrates the transverse
intennuscular septum lying between the flexor digitorum longus and soleus muscles. Unlike in many
cases of harvest of a fibular free flap, the perforators are true septocutaneous perforators and it
is not necessary to include a cuff of muscle to protect musculocutaneous perforators. B: In cross
sectional view. the flap has been elevated. The septum containing the perforating branches of the
PTA is identified, and the posterior tibial nerve has been left in the deep portion of the dissection.

lateral malleolus to enter the extensor compartment,
where it anastomoses with the anterior lateral malleo-
lar artery. It also supplies muscular branches to the
soleus, tibialis posterior, flexor hallucis longus, and
peronei muscles. The peroneal artery is a highly uti-
lized donor artery in the transfer of fibular free flaps in
head and neck reconstruction and is discussed in more
detail in Chapter 22.
The medial plantar artery and the much smaller
lateral plantar artery are the terminal branches of the
PTA. The medial plantar artery progresses forward
along the medial side of the foot, accompanied by the
medial plantar nerve. It branches from the PTA deep
to the abductor hallucis, running distally between this
muscle and the flexor digitorum brevis, supplying feed-
ing vessels to both. The artery terminates by joining the
digital branch of the deep plantar arch, which supplies
the medial border of the great toe. The lateral plantar
artery passes anterolaterally into the sole of the foot,
deep to the proximal end of the abductor hallucis mus-
cle. It then traverses obliquely across the sole between
the quadratus plantae and the flexor hallucis brevis and
curves medially to form the deep plantar arch, which
supplies the toes.
Venous drainage of the PTA flap is through its accom-
panying venae comitantes, the posterior tibial veins.
These veins receive branches from the gastrocnemius
and soleus muscles, including the venous plexus in the
soleus muscle, and contributions from the superficial
veins and the peroneal veins. The deep plantar venous
arch in the foot accompanies the arterial arch, giving
rise to the medial and lateral plantar veins. From the
deep plantar venous arch, the medial and lateral plantar
veins run backward close to the corresponding arter-
ies and, after communicating with the great and small
saphenous veins, unite behind the medial malleolus to
form the posterior tibial veins (3).
The PTA provides a robust supply of perforators in
the lower leg. Many perforator-based flaps have been
designed to avoid sacrifice of the PTA and its supply
to the muscles, nerves, and bones of the lower leg. The
reliability of the location of perforators has been a con-
cern in designing this type of flap. A study by Wu et al.
extensively examined the perforators of the PTA and its
clinical applications for pedicled flaps (18). They studied
the vascular anatomy of the posterior tibial vessels in 20
cadaveric legs and noted the number, size, and distribu-
tion of the direct cutaneous and direct muscle branches.
The lower leg was divided into four equal segments from
proximal to distal, with zone I being the most distal and
zone IV being the most proximal. The direct cutane-
ous branches were found to cluster primarily in zone II,
whereas the direct muscle branches to the soleus and
flexor digitorum longus muscles originate chiefly in
zones II and ill. As a result, the authors performed fas-
ciocutaneous and musculofasciocutaneous flaps based
POSTERIOR TIBIAL ARTERY FREE FLAP 263
on the posterior tibial perforator branches in 12 patients.
They achieved a 100% flap survival rate with no major
complications.
An investigation by Schaverien and Saint-Cyr further
illustrated the predictable presence of perforators from
the PTA, anterior tibial, and peroneal arteries when
they found that these vessels could be located in distinct
5-cm intervals within the intermuscular septum. The
authors mentioned that of the distribution of perfora-
tors of these three systems, the most clinically useful
were that of the PTA, stating that the distal perforators
could be used to cover defects of the heel, medial malle-
olus, Achilles tendon, and distal two-thirds of the tibia.
They suggested that this consistency in the distribution
of these nutrient vessels would allow for better design
and manipulation of pedicled perforator flaps for lower
leg reconstruction ( 16).
Hung et al. conducted an investigation of the anat-
omy of cutaneous perforators of the PTA in 20 limbs
of 10 cadavers and after localizing the majority of
the perforators to the middle two-quarters of the leg
(zones II and III) designed a free fasciocutaneous flap
based on one of these vessels. They named the flap
the posterior tibial perforator flap and transferred it in
six cases of hand and wrist defects (4). They reported
a 100% flap survival rate and very minor donor site
morbidity.
The saphenous nerve, arising from L2 to IA, is
the terminal sensory branch of the femoral nerve.
The saphenous nerve travels with the femoral artery
and vein through the adductor hiatus. After passing
through the hiatus, the saphenous nerve separates
from the vessels and travels through the fat between
the sartorius and vastus medialis. The saphenous nerve
provides sensation to the medial aspect of the lower
leg from the knee to the medial malleolus, and this ter-
ritory may extend as distal as the great toe in 20% of
the population (11). In the lower leg, it travels in close
proximity to the long saphenous vein. Saphenous neu-
ralgia describes the symptom complex of anesthesia,
hyperaesthesia, and pain in the area innervated by the
saphenous nerve and has been described after saphen-
ous vein harvesting for cardiac surgery procedures, but
the incidence following posterior tibial flap elevation is
unknown (12).
ANATOMIC VARIATION
Major variations in the popliteal artery branching pat-
terns are observed in up to 12% of individuals (6).
As the development of the arteries of the lower limb
results from the union of dorsal and ventral systems,
variations at many levels can be expected. Lippert
originally classified these variations into three types:
normal level of popliteal branching, high division of
264 CHAPTER 1&
the popliteal artery, and hypoplastic or aplastic artery
with altered distal supply (10). As stated earlier, nor-
mally the popliteal artery divides at the lower margin
of the popliteus muscle. This occurs in approximately
95% of individuals. In the remainder, the popliteal
artery often divides at or above the level of the knee
joint and in very rare cases, the artery branches at
a more inferior location in the lower leg. H the low
branching pattern is not identified prior to microvas-
cular flap harvesting, this variation can lead to the loss
of vascular supply to the foot.
One obvious peril of harvesting the PTA for recon-
struction of head and neck defects is the rare chance
that the artery will be absent. Lippert classified this
abnormality as a hypoplastic or aplastic PTA. This vari-
ation has been reported to occur in 3.8% to 5% of indi-
viduals and can result in altered vascular supply to the
lower leg in several forms (7). In cases where the PTA
is aplastic or hypoplastic, often the peroneal artery is
enlarged. It has many communicating branches and
gives rise to a distally based PTA and the plantar arter-
ies. One particular case study noted the absence of a
PTA, presence of a hypoplastic anterior tibial artery,
and compensatory marked hypertrophy of the peroneal
artery and supply of the dorsalis pedis, medial plantar,
and lateral plantar arteries (21). As such, in these situa-
tions, harvest of the peroneal artery for a microvascular
fibular flap would also compromise the vascularity of
the foot.
In a study of 1,000 femoral arteriograms by Kim
et al., it was found that in 0.8% of cases, the PTA
branches at or above the knee joint, also producing
a common trunk for the peroneal and anterior tibial
arteries (7). The incidence of other variations noted
by the authors was that of a hypoplastic-aplastic pos-
terior and anterior tibial artery, wherein (a) the dorsa-
lis pedis is a branch of the peroneal artery (1.6%) or
(b) both the posterior tibial and dorsalis pedis arter-
ies are branches of the peroneal artery (0.2%). Day
and Orme reported the incidence of this abnormality
to be in only I of 662 specimens examined (2). These
findings advocate for the use of preoperative vascular
imaging of the lower extremity to assess the vascularity
in the lower leg when considering a peroneal or pos-
terior tibial artery flap to avoid the catastrophic com-
plication of compromising the vascular supply to the
lower leg and foot.
Alternatively, vascular anomalies involving the
peroneal artery can compromise the blood supply
to the leg if the PTA is harvested without an aware-
ness of the abnormality. Abnormalities of the pero-
neal artery have been documented in the literature in
conjunction with anatomical variants of the popliteal
artery (7). In cases where the anterior tibial artery
is hypoplastic, often either the peroneal artery or the
PTA provides the main blood supply to the anterior
compartment of the lower leg. In the presence of such
an abnormality, the harvest of either of these arter-
ies could cause subsequent ischemic necrosis of their
supplied tissues, depending on the organization of the
anomalous vessels.
POTENTIAL PITFALLS
The greatest concern regarding the use of the PTA
flap is the requirement for sacrifice of one of the major
arterial supplies to the lower leg. Avoidance of lower
leg ischemic problems is fundamental to a successful
reconstruction. The approach to this potential problem
varies. In the initial description by Hwang et al., the
saphenous vein was used to restore the continuity of
the PTA (5). More recently, Ng et al. have suggested
that all patients have preoperative assessment of lower
leg vascularity by clinical and radiologic means (13).
They further suggested that all patients younger than
50 years of age undergo saphenous vein reconstruction
of the PTA to avoid future vascular insufficiency (13).
Li et al. reconstructed the PTA in half of their cases
with synthetic grafts but made no recommendations on
routine reconstruction of the vessels (9). A perforator-
based flap has been described to allow preservation of
the PTA (4). We evaluate all patients preoperatively
with computed tomography angiography and do not
routinely reconstruct the PTA. To date we have not had
any significant issues with lower leg ischemia. Further
study in this area is required prior to making defini-
tive recommendations. Long-term follow-up is criti-
cal in order to assess the impact of loss of one of the
major arterial supplies to the foot as the patient ages
and lower extremity atherosclerosis in the remaining
arteries worsens.
Donor site complications have not been well
described, but complications similar to other donor
sites may be predicted. Skin grafting directly over the
tibia may be of some concern; however, we have found
it possible to rotate the flexor digitorum longus over the
tibia to allow skin grafting to a muscular bed rather than
periosteum.
PREOPERATIVE MANAGEMENT
Careful clinical assessment of the leg should be per-
formed prior to selecting this donor site. Signs of
arterial or venous insufficiency should be carefully
noted and may require selection of another donor site.
Ng et al. advised that this flap is contraindicated in

patients with ischemic ulcers of the foot, claudication,
or foot or lower leg cold intolerance. Reports vary as
to the appropriate preoperative evaluation of lower
leg vasculature. Angiography and Doppler exam are
the two most common recommendations (8, 13). As
stated above, we routinely perform computed tomog-
raphy angiography to evaluate the vascular anatomy
of the legs. After tourniquet release, the foot should
be carefully assessed for vascular compromise and, if
necessary, revascularization can be performed. With
appropriate preoperative assessment, revascularization
should be rarely required.
POSTERIOR TIBIAL ARTERY FREE FLAP 265
POSTOPERATIVE CARE
A drain is placed in the deep compartment of the wound
underneath the muscle closure, and a skin graft is placed
over the defect (see "Flap Harvest Techniques" section).
A splint is applied postoperatively and left intact for 7
days to allow adherence of the skin graft. Mter 7 days
of no weight bearing, the patient is started on range-
of-motion exercises and weight bearing is increased to
full weight bearing by day 10. The patient is discharged
home with written instructions for an appropriate home
exercise program.
266 CHAPTER 1&

Posterior Tibial Artery Flap

FIGURE 16-5. The tnpographical anatomy is

outlined on the medial aspect of the leg. The
medial aspect of the tibia is marked in brown
between the posterior border of the medial
malleolus and the condyle of the tibia. This line
identifies the transverse intermuscular sep-
tum through which the perforators pass. The
approximate courses of the long saphenous
vein and saphennus nerve are shown as blue
and yellow lines, respectively.

FIGURE 16-6. The safid green line indicates

the anterior incision. The posterior aspect of
the incision is not performed until perforators
are identified following anterior dissection.

RGURE 16-7. The anterior incision is extended

to the level of the tibia. Periosteum is left on the
tibia. The flap is reflected posteriorly, taking
care to not extend the dissection beyond the
posterior border of the tibia.
 POSTERIOR TIBIAL ARTERY FREE FLAP 267

Posterior Tibial Artery Flap

FIGURE 16-8. A:. The long saphenous vein {blue

srrowt and saphenous nerve (yellows"ow) are
identified at the inferior portion of the dissection.
The long saphenous vein can be dissected free
of the flap and preserved intact or incorporated
with the flap. Alternatively, it can be preserved
and harvested later for posterior tibial artery
reconstruction. B: The saphenous nerve is found
in close proximity to the saphenous vein. This can
be left intact or incorporated with the flap to pro-
vide sensation. If the nerve is sacrificed, numb-
ness will occur in the distribution of the nerve in
the lower leg. This includes the medial aspect of
the lower leg and may extend to the great toe in
20% of patients undergoing flap transfer. B

FIGURE 16-9. Having elevated the flap to the

posterior border of the tibia, the flexor digitorum
longus muscle is identified. The fascia over this
muscle is divided at the posterior border of the
tibia.
268 CHAPTER 1&

Posterior Tibial Artery Flap

FIGURE 16-10. The fascia is reflected posteri-

orly off the flexor digitorum longus muscle, and
the septocutaneous perforators and the poste-
rior tibial artery pedicle are easily identified in
the septum between the flexor digitorum longus
and soleus muscles.

FIGURE 16-11. The anatomy ofthe posterior

tibial artery relations is shown. The red pin
identifies the posterior tibial artery pedicle. The
blue pin indicates the long saphenous vein.
The posterior tibial nerve lays posterior to the
pedicle and is indicated by the yellow pin. A
large septocutaneous perforator is indicated by
the blue arrow.

FIGURE 16-12. Having identified the location

of the perforators, a posterior incision may be
planned (green fine) depending on the size of
flap required. The posterior incision is extended
through the fa sci a of the soleus muscle, which
is reflected anteriorly, leading to the intermus-
cular septum and vascular pedicle.
 POSTERIOR TIBIAL ARTERY FREE FLAP 269

Posterior Tibial Artery Flap

FIGURE 1&-13. A:. The posterior tibial artery

is divided distal to the level of the cutaneous
perforators which are incorporated with the
flap. B: The pedicle is dissected from distal to
proximal with division of the muscular perfora-
tors (as shown). Care must be taken to protect
the posterior tibial nerve. B

FIGURE 16-14. If the long saphenous vein

(blue pin) and the saphenous nerve (yellow
pin) have been incorporated in the flap, they
must be identified and dissected atthe proxi-
mal aspect of the flap.
270 CHAPTER 1&

Posterior Tibial Artery Flap

FIGURE 16-15. The flap is dissected until

appropriate pedicle length has been attained.
If necessary, it can be dissected to the level of
the peroneal artery. The tourniquet if used, is
released. When ready for inset. the pedicle is
divided.

FIGURE 16-16. A-C: The incision is closed

primarily using deep and superficial sutures
at the superior aspect A drain is placed in the
deep compartment The flexor digitorum longus
(white srrawt, in many cases, can be rotated
over the tibia and closed to skin. The soleus
(yellow srrawt can be adva need to suture to the
posterior aspect of the flexor digitorum longus,
which provides a flat vascularized surface for
skin graft application. C

Posterior Tibial Artery Flap
FIGURE 1&-11. The posterior tibial artery free
flap is shown with the posterior tibial artery
pedicle (red pin), long saphenous vein (blue
pin), and saphenous nerve (yellow pin).
REFERENCES
1. Olen HC, TangYB, Noordhoff' MS: Posterior tibial artery
flap for reconstruction of the esophagus. Plast 1W:tm.m
Surg 1991;88(6):980-986.
2. Day CP, Orme R: Popliteal artery branching patterns-an
angiographic study. Clin Radicl2006;61 (8):696--699.
3. Gray H, et al.: Gray's A~: The Anatomical Bam of
Clinical Practice. 39th ed. Vol. 20. Edinburgh: Elsevier
O&urcl:lilll.ivingstone; 2005:1627.
4. Hung LK, Lao J, Ho PC: Free posterior tibial perfora-
tor flap: anatomy and a report of 6 cases. Microsurgery
1996;17{9):503-511.
5. Hwang WY, et al.: Medial leg skin flap: vascular anatomy
and clinical applications. Ann Pfasr Surg 1985;15(6):
489-491.
6. Kadir S: Atlas of Normal and Winant Angiographic Anat-
~.Vol. 11. Philadelphia: Saunder; 1991:529.
7. Kim D, Orron DE, Skillman U: Surgical significance of
popliteal arterial variants. A unified angiographic classifi-
cation. Ann Surg 1989;210{6):776--781.
8. Koshima I, et al.:The vasculature and clinical application
of the posterior tibial perforator-based flap. Pfasr 1W:tm.m
Surg 1992;90{4):643-649.
9. liYY, et al.: ~onstruction of limb defects with the free
posterior tibial artery wciocutaneous flap. Br J Plan Surg
1994;47{7):502-504.
10. lippert KM:Treatment of wounds of the popliteal artery.
AmJSurg 1949;77(1):114-116.
11. Meier G, Buttner J: Peripheral Regional Anesthesia: An
Atlas ofAnato'lt'ly and Techniques. NewYork: l'hieme; 2005.
12. Mountney J, Wllkinson GA: Saphenous neuralgia after
coronary artery bypass graf\iJJg. Eur J CardiotJufrtu: Surg
1999;16(4):440--443.
POSTERIOR TIBIAL ARTERY FREE FLAP 2:11
13. Ng RW, et al.: Free posterior tibial flap for head and
neck reconstruction after tumor expiration. lAryngoscope
2008;118(2):216--221.
14. O'Connell D, Reiger J, Dziegielewski PT, Tang JL,
Wolfaardt J, Harris ]R, Mlynarek A, Se:ikley H: Effect of
lingual and hypoglossal nerve reconstruction on swal-
lowing function on head and neck surgery: prospective
functional outcomes study. J Otolaryn.j:ol Head Heck Surg
2009;38(2):246--254.
15. Ozdemir R, et al.: Examination of the skin perfora-
tors of the posterior tibial artery on the leg and the
ankle region and their clinical use. Plan RJu:tmm Surg
2006;117(5):1619-1630.
16. Scllsverien M, Saint-Cyr M: Perforators of the lower
leg: analysis of perforator locations and clinical appli-
cation for pedicled perforator flaps. Plan Reoonstr Surg
2008;122(1):161-170.
17. Strauch B, Yu H-L: Atlas of Microoascular Surgery: Anat-
ono' and Operarive Techniques. 2nd ed. Vol. 8. New York:
Thieme; 2006:686.
18. Wu WC, et al.: The anatomic basis and clinical appli-
cations of flaps based on the posterior tibial vessels.
Br J Plast Surg 1993;46(6):470-479.
19. Zhang SC: Clinical application of medial skin flap
of leg-analysis of 9 cases. Zlwnghua Wai ~ Za Zhi
1983;21{12):743-745.
20. Zhang X, et al.: Posterior tibial artery-based multilobar
combined flap free transfer for repair of complex. soft tis-
sue defects. MicroSUYgery 2008;28{8):643-649.
21. Zwus A, Abdelwahab IF: A case report of anoma-
lous branching of the popliteal artery. Angiology 1986;
37(2):132-135.
INTRODUCTION
Fasciocutaneous flaps from the forearm are ideally
suited for reconstruction of head and neck defects. Ease
of harvest, remote location from the head and neck, pli-
ability of skin, minimal donor site morbidity, and abil-
ity to incorporate bone, muacle, tendon, and nerve for
composite resections are all advantages of the forearm
donor site. The ulnar forearm free flap (UFFF), as orig-
inally descnbed by Lovie et al. in 1984, is one of the
two forearm flaps commonly in use today (15). The first
transfer of this flap took place in New Zealand in 1982
shortly after the popularization of the radial forearm
free flap (RFFF) by the Chinese in 1981 (27).
The ulnar forearm flap has wide potential application
for reconstruction of head and neck defects, namely,
for cutaneous and soft-tissue defects, intraoral mucosal
defects, and glossal, pharyngeal, tracheal, and esophageal
defects. It has also been used for a variety of other recon-
structions including upper extremity, hand, and urologic
defects, which are outside the focus of this chapter.
Its advantages as compared to the RFFF include (a)
less donor site morbidity from flexor tendon exposure,
(b) better cosmesis (the wound is less conspicuous on
the ulnar/volar aspect of the forearm), and (c) rela-
tive hairlessness of the flap. It is easily harvested using
a two-team approach. The surgical time, reliability of
the vascular pedicle, and size of skin paddle do not dif-
fer significantly when compared to the RFFF (15,25).
Additionally, in a comparative study, the ulnar pad-
dle was found to be thinner than the skin of the radial
forearm (25). Although not studied through detailed
functional outcomes studies, this may confer additional
advantage in the reconstruction of some defects.
272
A commonly cited disadvantage to the UFFF is the
length of the pedicle. The ulnar pedicle is taken distal
to the takeoff of the common interosseous artery, and
thus the maximal length of the ulnar pedicle (1 0 em) is
shorter than that of the radial pedicle by 4 to 5 em. This
may be of concern in reconstructing defects that require
a longer pedicle length.
NEUROVASCULAR ANATOMY
The blood supply to the lower arm and hand is supplied
by the brachial artery, which divides into the ulnar artery
and the radial artery (Fig. 17-lA,B) at the level of the
antecubital fossa. After its takeofffrom the brachial artery,
the ulnar artery descends approximately 1 em distal to the
flexor crease of the elbow and passes medially to the ulnar
border of the forearm, midway between the elbow and
wrist. The artery then crosses the transverse carpal liga-
ment on the radial side of the pisiform bone, terminating
in the superficial palmar arch in the hand.11lroughout its
course, the ulnar artery is situated deeply in the forearm,
covered in the proximal half of the forearm by the prona-
tor teres, fl.ez.or carpi radialis, palmaris longus, and flexor
digitorum superficialis. Distally, the artery is bordered
medially by the flexor carpi ulnaris (FCU) and laterally
by the flexor digitorum superficialis and lies superficial
to the flexor digitorum profundus (Fig. 17-2A,B). The
ulnar artery is accompanied throughout its course in the
forearm by one or two 'Ve'IU1e comitantes (9). The ulnar
bone is supplied through multiple metaphyseal nutti-
ent foramina that transmit branches of the radial, ulnar,
anterior, and posterior interosseous arteries. Usually one,
but occasionally two, major nuttient diaphyseal foramina
 ULNAR FOREARM FREE FLAP 2:13

Palmartslongus Flexor carpi

A ulnarls

B Ulnar artery
FIGURE 17-1. A:. The superficial muscles of 1he forearm are illustrated. The septocutaneous
perforators for the ulnar forearm free flap pass through the intermuscular septum between 1he
flexor carpi ulnaris and flexor digitorum superficialis muscles. B: The radial and ulnar arteries are
shown after their branching from the brachial artery. The ulnar artery descends approximately
1 em distal to the flexor crease of the elbow, passes medially to 1he ulnar border of the forearm,
midway between the elbow and wrist, and terminates in 1he superficial palmar arch in the hand.

are located on the anterior surface of the bone, directed
proximally toward the elbow. A network of small fascia-
periosteal and mWiculoperiosteal branches given offfrom
the compartmental vessels reaches the bone via septal
and muscular attachments (9).
}Wit distal to the radial tuberosity, the ulnar artery
gives off the common interosseous artery laterally,
which subsequently separates into anterior and pos-
terior branches (Fig. 17-3). The anterior interosseous
artery descends on the anterior aspect of the interos-
seous membrane with the anterior interosseous branch
of the median nerve. The posterior interosseous artery
passes dorsally between the extensor carpi ulnaris and
the extensor digitorum and accompanies a deep branch
of the radial nerve before distally anastomosing with the
terminal component of the anterior interosseous artery
in the hand. The posterior interosseous artery has been
used to supply a fasciocutaneous flap for the recon-
struction of hand defects (5,20).
Distal to the elbow joint, the ulnar artery gives
rise to the anterior ulnar recurrent artery, which runs
proz.imally in the arm, betWeen the brachialis and pro-
nator teres. The posterior ulnar recurrent artery arises
distal to the anterior ulnar recurrent artery and runs
superior and posterior to the medial epicondyle, where
it lies deep to the tendon of the FCU. The dorsal ulnar
artery (also known as the dorsal carpal branch of the
ulnar artery) arises 2 to 5 em before the pisiform bone
and runs dorsally and distally under the FCU muscle,
to which it provides both proximal and distal branches.
The proximal branch extends as far as the medial epi-
condyle of the humerus. It forms the pedicle on which
the dorsal ulnar artery flap is based (2). The distal
branch supplies the pisiform bone. The middle branch
of the dorsal ulnar artery supplies the skin and divides
into two smaller arterial branches that pierce the fascia.
The ascending branch of the artery runs between the
ulna and FCU, supplying the skin of the medial border
of the distal forearm. The descending branch accompa-
nies the dorsal sensory branch of the ulnar nerve that
supplies the skin over the ulnar metacarpals and the
ulnar hypothenar region (11).
274 CHAPTER 17

Flexor--~
diprum
superficialis

A Flexor digitorum
profunclls

Ulnar
Flexor dlgltorum
S(.lperflclalls muscle

Meclal
antebrachial
cutaneous
nerve
Basilie
vein

Flexor
carpi
ulnaris

Flexor
dlgltorum
profundus
B
FIGURE 112. A: The cross-sectional anatomy of the ulnar forearm flap identifies the location
of the ulnar pedicle. The pedicle is bounded by the flexor carpi ulnaris medially, the flexor digito-
rum superficial is laterally, and the flexor digitorum profundus on the deep surface. B: Across-
sectional view demonstrates the surrounding anatomic structures of the distal forearm after
elevation of the ulnar flap. The ulnar nerve is left intact in the forearm. The median antebrachial
cutaneous nerve has been incorporated and may be used to perform a sensate reconstruction.
 ULNAR FOREARM FREE FLAP 2:15

,~--+/ Anterior ulnar

recurAlnt artery

Posterior ~~+--+-Common Inter-

Interosseous osseous artery
branch - - - + - --lli'""""''HI
Deep radial
M~-----+~~~H

/J-1--_.,.-+1-l...._ Posterior
interosseous
artery
Anterior
Interosseous
branch -----+-~t+--H Extensor
clgitorum - - - + - HI'Ii-
oM-~-- Extensor
carpi
Interosseous radialis
~mb~----~1---r w

A B
FIGURE 11-3. The proximal branches of the ulnar artery are shown. Just distal to the radial
tuberosity, the ulnar artery gives off the common interosseous artery laterally, which subse-
quently separates into anterior and posterior branches. A:. The anterior interosseous artery
descends on the anterior aspect of the interosseous membrane with the anterior interosseous
branch of the median nerve. B: The posterior interosseous artery passes dorsally between the
extensor carpi ulnaris and the extensor digitorum and accompanies a deep branch of the radial
nerve before distally anastomosing with the terminal component of anterior interosseous artery
in the hand. The posterior interosseous artery has been used to supply a distally based fascio-
cutaneous flap for the reconstruction of hand defects.
276 CHAPTER 17

Perforators to the skin arise &om both the ulnar and
posterior ulnar recUITent arteriea (26). The skin ovet'-
lying the FCU is supplied by both musculocutaneous
and septocutaneous perforators &om the ulnar artery.
Septocutaneous perforators &om the ulnar artery sup-
ply the skin extending from the cubital fossa to the wrist
and from the medial third of the anterior aspect of the
forearm to the medial quarter of the posterior sutface
of the forearm (9) (Fig. 17-4A,B). Lovie et al. descnbed
the position of the perforators as being 3 to 4 em from
the takeoff of the common interosseous branch, the
most dominant of which was constant in location (15).
Yii and Niranjan described the perforator anatomy of
the ulnar forearm pedicle based on their experience
with the reconstruction of 13 donor defects resulting
from radial forearm flap harvesting. They reported the
consistent presence of one or two perforators located 8
to 10 em proximal to the pisiform bone (31).
A 2008 study by Shen confirmed these findings and
added that there are 6 to 7 septocutaneous perforators
that arise from the radial border of the FCU (26). In
their anatomic description, there were two proximal
septocutaneous perforators and three distal fasciocuta-
neous perforators that were captured when harvesting
a standard ulnar artery fasciocutaneous flap. Notably,
septocutaneous perforators in a UFFF were found to be
evenly spaced from the wrist to the medial epicondyle,
whereas perforators associated with the RFFF are typi-
cally concentrated in the distal third of the forearm (26).
Additionally, there are two pro.ximal and distal muscu-
locutaneous perforators arising from the FCU muscle,
along with a dominant musculocutaneous perforator
that is found midway betWeen the medial epicondyle
and the wrist. If the FCU is required for reconstruction,
attention should be paid to preserving these perforators.
A cadaveric study and subsequent dynamic studies
including Doppler sonography in 22 individuals eluci-
dated the issue of vascular dominance in the forearm
(10). Theae studies demonstrated that the ulnar artery
is dominant at the elbow, but after giving off collat-
eral branches, the radial artery becomes the dominant
artery in the distal forearm. They concluded that the
radial artery is the major blood supply to the hand and
stated that there is no anatomic basis for selecting the

RGURE 174. A. B: The size and shape of the ulnar flap is designed to suit the defect. Flap
dimensions of 10 x 22 em can be safely harvested. The flap should be centered over the ulnar
artery and basi lie vein. Septocutaneous perforators from the ulnar artery supply the skin
extending from the cubital fossa to the wrist and from the medial third of the anterior aspect of
the forearm to the medial quarter of the posterior surface of the forearm.
 ULNAR FOREARM FREE FLAP 2'17

radial artery to the ulnar artery in invasive maneuvers
(i.e., catheterization).
Vascularity of the hand relies on the deep palmar
arch, which is the distal extension of the radial artery,
and the superficial palmar arch, which is derived from
the ulnar artery. Sacrifice of either artery for use in a
fa.sciocutaneow flap requires that there be a communi-
cation between the deep and superficial palmar arches
to maintain an adequate circulation to all portions of
the hand (30).
From the hand, venous arches accompany the super-
ficial and deep arterial palmar arches. Common pal-
mar digital veins drain into the superficial arch; palmar
metacarpal veins join the deep arch. The deep veins of
the forearm are responsible for draining the deep and
superficial palmar venous arches. Most of the blood
from the upper limb is returned by the superficial
venous system, and thus, the deep veins or venae comi-
tantes are relatively small (Fig. 17-5). The radial wnae
comitantes receive blood from the deep dorsal veins of
the hand. The ulnar venae comitantes drain the deep pal-
mar venous arch and connect with superficial veins near
the wrist. AB both the radial and ulnar venae comitantes
ascend in the forearm, they receive the anterior and pos-
terior interosseous wnae comitantes, after which a large
branch connects the deep system to the medial cubital
vein (1).
The reestablishment of venous drainage of forearm
fasciocutaneous flaps is usually performed by the deep
radial or ulnar venae comitantes and/or by the superficial
venous system, which forms multiple anastomoses along
the forearm (Fig. 17-5). Hence, the basilic vein as well as
the t.1e11ae comitantes drain the ulnar forearm Bap. Con-
sideration of the different branching patterns of the deep
and superficial venous systems as well as the variability
of the size of the vessels and the identified course of the
subcutaneous veins, all play a role in the decision to anas-
tomose to either a superficial vein or the venae comitantes.
The ulnar nerve enters the forearm after passing
posterior to the medial epicondyle of the humerus and
gives offmotor branches to the head of the FCU. It then
descends inferiorly between the FCU and the fiexor
digitorum profundus, providing branches to the ulnar
(medial) part of the muscle that sends tendons to the
third and fourth digits (17). The ulnar nerve runs on
the medial aspect of the ulnar artery and lateral to the
FCU tendon. The nerve descends medially on the fiexor
digitorum profundus, covered proximally by the FCU,
while the distal half, covered by skin and fascia, is lat-
eral to this muscle. The proximity of the ulnar nerve to
the ulnar pedicle can make dissection of this Bap more
complicated and precariow.
Approximately 5 em proximal to the wrist, the ulnar
nerve provides a dorsal branch, which passes distally
into the hand on the lateral side of the pisiform bone,
anterior to the Bexar retinaculum and posteromedial to
the ulnar artery. The nerve passes behind the superficial
portion of the Bexar retinaculum with the ulnar artery
and divides into superficial and deep terminal branches
in the hand.

Cephallcv.

Baslllcv.
FIGURE 115. The venous and sensory supply to the forearm is shown. The reestablishment of
venous drainage of the ulnar flap can be safely performed by anastomosis of 1he ulnar venae
comitantes and/or by the superficial venous system, which forms multiple anastomoses along
the forearm. Hence, the basilic vein as well as the venae comitsntes drain the ulnar forearm
flap. Consideration of the different branching patterns of the deep and superficial venous
systems, 1he variability of the size of the vessels, and the identified course of the subcutane-
ous veins all play a role in the decision to anastomose to either a superficial vein or1he venae
comitsntes.
278 CHAPTER 17

ANATOMIC VARIATIONS or index finger and a lack of communication between

Several vaacular anomalies must be considered before
harvest:il:lg any forearm 1lap, as the potential hazard
aists for compromised blood supply to the forearm and
hand. Theae anomalies include the hypoplastic or aplas-
tic radial artery, the incomplete superficial palmar arch
with no communication to the deep palmar arch, and
the superficial ulnar artery. When harvesting a UFFF
in patients with such vascular anomalies, it is the sur-
geon's responsibility to identify and avoid potentially
catastrophic ischemia of the soft tissues of the hand.
The hypoplastic or aplastic radial artery is found in
approximately 0.1% of the population and can be asso-
ciated with Down's syndrome (12,19). In patients with
this anomaly, the radial artery is markedly reduced in
caliber and usually terminates proximal to the wrist.
The ulnar artery is often enlarged as a result and sup-
plies both the deep and superficial palmar arches. Alter-
natively, the m edian artery may be enlarged and persist
with a codominant ulnar artery. Obviously, an ulnar
microvascular 1lap is not feasible in the absence of a
normal radial arte.r y.
The blood supply to the hand can also be compro-
mised by the combination of an incomplete super1icia1
palmar arch that does not send branches to the thumb
the deep and superficial palmar arches. Coleman and
Anson found that in 265 cadaveric specimens, a com-
plete superficial arch was present in only 77.3% of
cases; the coexistence of an incomplete superficial arch
and a lack of communication between deep and supel'-
ficial palmar arches was found in 12% of anatomic di.-
sections (4). Thus, one can conclude that 12% of the
population who have these coexisting arterial anomalies
would n ot be candidates for harvest of a fasciocutaneoua
flap from the forearm. After reviewing 92 angiographic
studies of the hand, Varro et al. found that the superficial
palmar arch was incomplete in 66.3% and the deep arch
was incomplete in 9 .8% of banda (29). In a study by
Ozkus et al., the superficial palmar arterial systems of 80
cadaver hands were dissected. Although 78 of the speci-
mens (97 .5%) were shown to possess a superficial arch,
17.5% of the specimens revealed a superficial arch that
was formed by the ulnar artery alone without commu-
nication to the deep arch (18) . These studies underscore
the importance of adequate preoperative evaluation, in
the form of careful palpation of the forearm, Allen's test,
and/or ultrasound Doppler evaluation, as 10% to 20%
of forearms may have a vascular anomaly that could
compromise blood supply to the hand after harvest of
the radial or ulnar artery (Fig. 17-6).

Normal Hypoplastic radial artery Incomplete superficial ard'land

no connection
FIGURE 11-6.. Anatomic variations of the ulnar and radial systems are demonstrated. The
hypoplastic or aplastic radial artery is found in approximately 0.1% of the population and can
be associated with Dawn's synd ram e (15, 16). In patients with this anomaly, the radial artery is
marked ly reduced in ca liber and usually terminates proximal to the wrist. The ulnar artery is
aften enlarged as a result and supplies both the deep and superficial palmar arches. Alterna-
tively, the med ian artery may be enlarged and persist with a codominant ulnar artery. Obviously,
an ulnar microvascular flap is not feasible in the absence of a normal radial artery. The blood
supply to the hand can also be compromised by the combination of an incomplete superficial
palmar arch that does nat send branches ta the thumb or index finger and a lack af communica-
tion between the deep and superficial palmar arches.

Another anatomic variant, the superficial ulnar
artery, was first documented by McCormack et al. ( 16).
This anomaly occurs in 0.7% to 9.38% of cases (32).
The superficial ulnar artery can originate from the bra-
chial or, less commonly, the axillary artery (6). It passes
medial to the biceps and superficial to the pronator
teres and flexor muscles, lying deep or superficial to the
fascia, finally establishing a normal superficial palmar
arch in the hand with adjunct supply from the super-
ficial branch of the radial artery. The palmaris longus
muscle is often described as being absent in cases where
a superficial ulnar artery is found (32).
During dissection of fasciocutaneous forearm flaps
based on the ulnar artery, Sieg et al. revealed a super-
ficial ulnar artery (SUA) in 4 of 107 forearms: 3.3% in
women and 3. 9% in men. All four of the flaps harvested
based on the SUA survived. Postopemtive angiography
allowed for the detection of a bilateral vascular anomaly
in one of the four cases, noting additionally that the
SUA divided about 10 em proximal to the elbow joint
and failed to terminate in a superficial palmar arch. The
authors concluded from their study that (a) the SUA
is typically not identified preoperatively; (b) in radial
forearm flaps, sacrifice of arteries that cross the fore-
arm superficially should be avoided; and (c) the SUA
is found to be an easy and safe alternative to the radial
forearm flap (24). Other authors suggest that the SUA
can be identified preoperatively with careful palpation
of the forearm as well as a directed Allen's test but agree
that the finding of an SUA should not preclude the per-
formance of a fasciocutaneous flap harvest (32).
Other rare anomalies of the ulnar artery have also
been described. Tcacencu discovered a large branch of
the ulnar artery in the carpal tunnel, which crossed the
median nerve anteriorly and terminated without pro-
ducing a superficial palmar arch (28). Another study by
LeGeyt and Ghobadi noted the position of the ulnar
nerve and artery overlying the carpal canal, leaving
Guyon's canal empty and posing risk of damage to the
artery and paralysis to the nerve (13). Durgun et al.
reported on a case of multiple and bilateral arterial vari-
ations of the radial and ulnar arteries and with associ-
ated abnormal metacarpal vascularity (7). The authors
suggest that a documented abnormality in one area of
the upper limb should prompt investigation for other
vascular anomalies.
FLAP DESIGN AND UTILIZATION
Design of a UFFF is similar to the design of an RFFF.
The flap is usually harvested with the skin paddle cen-
tered over the axis of the ulnar artery. The skin pad-
dle can measure as much as 10 x 22 em according to
some series (22,23). One of the distinct advantages of
the UFFF is the relatively thin nature of the adipofascial
ULNAR FOREARM FREE FLAP 279
component on the dorsal ulnar aspect of the forearm
when compared to the RFFF. Similar to the RFFF, the
UFFF can be utilized for the reconstruction of a myriad
of defects in the head and neck, which include oral cav-
ity (tongue, floor of mouth), oropharyngeal (base of
tongue, soft palate, lateraVposterior pharyngeal wall),
pharyngoesophageal defects, and neck skin or other
soft-tissue defects. Additionally, the UFFF can be used
reliably with an osseous fibula or iliac crest flap in cases
that require double flaps (8).
The first large series involving the ulnar forearm free
flap was published in 1995 by Salibian et al. who reported
on oropharyngeal reconstruction involving the base of
tongue in patients after primary extirpation or after
neoadjuvant radiotherapy and/or chemotherapy failure
(22). Ten patients underwent reconstruction with an
ulnar flap after having 30% to 100% of the tongue base
resected. The base of tongue was resected to the level of
the hyoid, sacrificing the hypoglossal nerve and the lin-
gual artery, as well as various other nearby oropharyn-
geal subsites. In all the patients, both superior laryngeal
nerves and one hypoglossal nerve were left intact. The
authors used a large flap measuring 8 x 20 em, three
of which were neurotized using the medial antebrachial
cutaneous nerve of the forearm. The proximal portion
of the flap was used to reconstruct the soft palate and
pharynx, while the remaining distal portion was used in
a "jellyroll" fashion to reconstruct the tongue base. The
authors advocated suturing the rolled portion of the flap
in layers to the cut edge of the tongue to minimize con-
tractures and grooving along the flap-tongue interface.
Additionally, the authors suggested doubling the rolled
flap volume compared to the tongue base defect to allow
for flap shrinkage during healing (22). The authors con-
cluded that the UFFF has several properties that make
it well suited for oropharyngeal reconstruction involving
the base of tongue. First, the proximal portion of the
flap confers more soft-tissue bulk as compared to the
RFFF, which allows one to overcorrect the volume of
the resected tongue base and helps to approximate the
palate and pharyngeal wall during swallowing. Second,
the distal skin paddle is thin and pliable, which makes it
optimal for resurfacing the tonsillar fossa and postero-
lateral pharyngeal wall, as well as for reconstructing the
soft palate (22).
The ulnar forearm flap has also been demonstrated
to be effective for reconstruction of the pharyngoesoph-
agus after total laryngopharyngectomy. In a 1998 study,
Li et al. performed 20 ulnar forearm flaps for partial
to near total circumferential pharyngoesophagectomy
defects. The flaps avemged 9 x 22 em and were designed
in a trapezoidal fashion, 7 em wide distally and 10 em
wide proximally, centered on the ulnar pedicle. The
authors included the basilic vein as their venous pedi-
cle and followed it proximal to the antecubital fossa to
obtain another 10 to 15 em of length. The region of the
280 CHAPTER 17
flap along the suture line was de-epithelialized, and the
adjacent vascularized fascia was used to provide a sec-
ond layer of closure. After reconstruction of the pharyn-
goesophageal segment, the remaining distal portion of
the flap was de-epithelialized and was used to obliterate
dead space and/or to provide coverage for the cervical
vessels. A small cutaneous paddle was exteriorized as a
skin monitor in all cases. Out of 20 flaps, 19 were suc-
cessful. The one failure was early in the authors' expe-
rience and involved a flap that could not be salvaged
due to venous thrombosis, despite a successful primary
transfer. Fistulae occurred in three cases (15%), includ-
ing the one failure. Stricture occurred in one patient.
Swallowing function was restored in all but one patient
who had undergone a previous mandibulectomy and
tongue base resection. However, speech was restored
with tracheoesophageal prostheses in only two patients,
with the rest communicating with the aid of an elec-
trolarycx. The authors concluded that the ulnar fore-
arm flap offered the benefits of abundant vascularized
fascia with a reliable pedicle and minimal donor site
morbidity, making it well suited for pharyngoesopha-
geal reconstruction {14).
The UFFF has been described for use in oral cav-
ity reconstruction as well. In a series of 13 patients, a
large 10 x 20 em ulnar forearm flap was transferred for
reconstructing a hemiglossectomy defect. The authors
chose to inset the flap in a spiral fashion, using the dis-
tal two-thirds of the flap for reconstruction of the oral
tongue and the proximal one-thirds to recreate the floor
of mouth along a separate suture line. The rationale
was to design the ulnar forearm flap in such a way as to
maximize bulk at the root of the tongue, with a gradual
reduction in volume toward the front. Replacing the
resected mylohyoid muscle with a deepithelialized por-
tion of the proximal flap and suturing this to the ante-
rior mandible also addressed the potential for caudal
flap displacement.
Christie reported on the use of ulnar flaps over a
7-year period. Of 56 ulnar flaps, 38 were used for recon-
struction of head and neck cancer defects. Of these,
31 were used to reconstruct intraoral defects.The major-
ity of these flaps {32) included a segment of the ulna,
along with a fasciocutaneous component. Six fractures
occurred in total. One fracture was managed intraop-
eratively. The remainder were managed with immobi-
lization in the postoperative period. The authors added
that in addition to bone, the palmaris longus tendon
and/or the FCU may also be harvested depending on
the reconstructive requirements. The authors suggested
that in order to include a length of hemiulna, either the
FCU should be included or a cuff of FCU should be
dissected in order to preserve the periosteal perforators
to the bone. Additionally, it is suggested that limited
subperiosteal stripping of extensor compartment mus-
cles may help to provide an unobstructed longitudinal
view for the osteotomy. Up to 16 em of the ulna can be
harvested, noting that the bone is especially thin in the
middle and distal portions.
Wax et al. reported on the use of the ulnar flap for
oral cavity and oropharyngeal defects as well as for neck
skin and soft tissue for the lateral skull. Their indica-
tions for the use of the ulnar forearm flap included a
failed Allen's test, need for a less hairy part of the fore-
arm, and surgical preference. The authors stressed the
decreased donor site morbidity and improved cosme-
sis of the ulnar flap as compared to the RFFF (30).
Rodriguez et al. have also used the UFFF for closure of
cutaneous defects involving the cheek, nasal ala, fore-
head, lips, and most notably, eyelid (21). The superficial
temporal artery and vein were the recipient vessels in
the majority of cases. The authors commented on the
reliability of the UFFF and stated that its use does not
result in functional, motor, sensory, or vascular compli-
cations, as once thought.
POTENTIAL PITFALLS
Without a doubt, the most devastating complication
of UFFF harvest is hand ischemia. In all reports of
UFFF, none of the authors described the development
of hand ischemia postoperatively. However, flap har-
vest had to be abandoned twice in one series secondary
to vascular variations that were recognized intraopera-
tively. In the first case, there was no identifiable sep-
tocutaneous perforator, and in the second case, 50% of
the flap showed signs of ischemia (3). Other series have
demonstrated that a finding of SUA is not a contrain-
dication to the UFFF harvest, although its presence
along with other vascular aberrations should be identi-
fied preoperatively (24,32). Indeed, prior to dividing
the UFFF pedicle, the blood supply to the hand should
be tested while clamped, ensuring that the hand has
adequate blood flow.
Donor site complications include hematoma, minor
skin graft losses, and pathologic ulnar fracture. The inci-
dence of hematoma is roughly 3% to 5% after evaluat-
ing all series (30). The reported rate of flexor tendon
exposure after UFFF varies from 0% to 14% across all
studies (30). Rodriguez et al. underscored the impor-
tance of suprafascial dissection in order to reduce ten-
don exposure and subsequent donor site complications
(21). The only series to report on the incidence of path-
ologic fracture in the remaining length of the hemiulna
stated a 29% rate. To minimize this risk, it is suggested
that in patients in whom a portion of the hemiulna is
also harvested, the arm should be immobilized for at
least 6 weeks with an above-the-elbow cast (3). Experi-
ence with the ulnar osteocutaneous flap is not nearly as
robust as the experience with the radial osteocutaneous,
which is described in detail in Chapter 23.

One of the criticisms of the UFFF is the proximity of
the ulnar nerve to the vascular pedicle and the potential
for injury to the ulnar nerve and resulting paresthesia.
In one of the first series on the UFFF, the incidence
of transient ulnar nerve paresthesia was 32% (3). It is
notable, however, that all cases of paresthesia were mild
and resolved within 2 weeks. In two series of UFFF
transfers, there were no ulnar nerve injuries or pares-
thesias noted (21,30). H the ulnar nerve is exposed, as
a result ofFCU harvest, some authors suggest that the
nerve should be buried by suturing the flexor digitorum
superficialis over the nerve to cover it in order to prevent
direct apposition of the skin gmft to the ulnar nerve (3).
PREOPERATIVE MANAGEMENT
Harvest of the UFFF is usually performed on the
nondominant forearm. Preopemtive management
includes careful assessment with Allen's test to assess
the adequacy of flow to the hand through the mdial
artery. Additionally, a careful clinical exam should be
ULNAR FOREARM FREE FLAP 281
performed, looking for scars or vascular anomalies that
may preclude the use of the UFFF. Some authors have
advocated the use of Doppler flow imaging for preop-
erative arterial evaluation, especially in the case of upper
limb vascular abnormalities (30). We do not study the
vasculature of the arm for routine preoperative evalua-
tion, unless preopemtive clinical exam indicates a pos-
sible vascular abnormality.
POSTOPERATIVE CARE
Postoperative care involves placement of a skin gmft in
cases where defects cannot be closed primarily. Wax et
al. advocate the use of a circumferential "purse string"
stitch to reduce the size of the defect prior to skin gmft-
ing (30). Volar splints are typically placed after skin
grafting and removed on postopemtive day 10. H an
osteocutaneous UFFF has been harvested and internal
fixation is not applied, it is advised that an above- the-
elbow cast be placed for 6 weeks to reduce the risk of
pathologic fracture following a UFFF (3).
Z82 CHAPTER 17

Ulnar Forearm Free Flap

FIGURE 11-7. The design of the ulnarfore-

arm flap begins by identifying the path of the
subcutaneous veins, including 1he basilic vein
and the ulnar artery. The course of the basili c
vein and the ulnar artery have been drawn on
the forearm in blue and red, respectively. The
approximate course of the medial antebrachial
cutaneous nerve of the forearm has been
outlined in yellow. Tendons of the PT, FDS, and
FCU are shown in brown. PT, palmaris longus;
FDS, flexor digitorum superficialis; FCU, flexor
carpi ulnaris.

FIGURE 17-8. A: A rectangular ulnar forearm

flap has been outlined in green on the distal
forearm including the superior aspect of the
incision that provides access to the proximal
course of the ulnar artery and basilic vein. The
flap is centered over the course of the ulnar
artery. B: Flap design may include the basilic
vein. However, small ulnar flaps may be based
only on the ulnar artery and paired venae
comitantes. The dissection begins after exsan-
guination of the forearm and application of the
tourniquet to 250 mm Hg. B
 ULNAR FOREARM FREE FLAP 283

Ulnar Forearm Free Flap

FIGURE 17-9. Circumferential incisions to the

level of the dermis are then made around the
skin paddle, including the proximal extension
to the antecubital fossa.

FIGURE 1710. Medial and lateral skin flaps

are elevated along the proximal incision line.
These flaps may be elevated along the supra-
fascial muscular plane, preserving the basilic
vein (blue pin) and median antebrachial cuta-
neous nerve (yellow pin), if a neurotized thin
skin flap is desired. Alternatively, subdermal
flaps may be elevated in order to incorporate
proximal fat and fascia from the upper forearm,
if proximal fat around the pedicle is desired for
reconstructive purposes.

FIGURE 1111. Dissection is initially performed

from the radial aspect of the flap. The flap is
elevated off the flexor muscles. The flap eleva-
tion may be either subfascial or suprafascial.
As the dissection is carried from radial to ulnar,
the flexor carpi radialis, palmaris and flexor
digitorum superficialis muscles will be encoun-
tered. Careful dissection on the medial aspect
of the flexor digitorum superficialis allows
retraction of this muscle and identification of
the full course of the ulnar artery and ulnar
nerve (red pin).
Z84 CHAPTER 17

Ulnar Forearm Free Flap

FIGURE 11-12. Dissection can now be per-

formed along the ulnar aspect of the flap. The
flap is elevated to the level of the flexor carpi
ulnaris. Careful dissection is performed on this
muscle, allowing itto be retracted medially
to identify the ulnar artery and intermuscular
septum. The flexor carpi ulnaris is freed along
its entire length to provide exposure to the
proximal portion of the ulnar artery.

FIGURE 11-13. If the basilic vein has been

incorporated into the free flap, it is ligated at
its distal portion on the ulnar side of the flap.

FIGURE 11-14. The ulnar artery is now ligated

under direct visualization at its distal aspect.
Care is taken to preserve the ulnar nerve.
 ULNAR FOREARM FREE FLAP 285

Ulnar Forearm Free Flap

FIGURE 17-15. The free flap is then elevated

from distal to proximal, with careful dissection
and ligation of the muscular perforators and
preservation of the septocutaneous perfora-
tors and ulnar nerve.

FIGURE 17-16. A large muscular perforator

(yellow arrow) and smaller septocutaneous
perforator are shown from the radial side of
the dissection.

FIGURE 17-17. The flap is elevated proximally

to the level of the interosseous artery. (arrow)
near to where the ulnar artery passes deep to
the median nerve (upper yellow pin). At this
level. the artery is carefully separated from the
paired vense comitsntes. Often, the two vense
comitsntes will connect to form a single larger
vena comitantes. The tourniquet is released.
and a check is made for adequate vascular
supply to the free flap through dermal bleed-
ing. Hemostasis in the flap in situ in the arm is
performed.
Z8& CHAPTER 17

Ulnar Forearm Free Flap

FIGURE 1118. The flap is harvested by divid-

ing the ulnar pedicle, basilic vein, and median
antebrachial nerve. The median nerve (upper
yellow pin) and ulnar nerve (lower yellow pin)
are carefully preserved.

FIGURE 1119. The ulnar forearm flap is

shown with the ulnar artery Ired pin) and basilic
vein (blue pin), and median antebrachial cuta-
neous nerve identified (yellow pin).

REFERENCES The experience of 100 anatomical dissections and 102

1. Arnstein, PM, Lewis JS: Free ulnar artery forearm flap: a
modification. Br J Plan Surg 2002;55(4):356--357.
2. Becker C, Gilbert A; The ulnar flap. Ht:mdchit Mikrothtr
Plast Chir 1988;20(4):180-183.
3. Christie DR, Duncan GM, Glasson DW: The ulnar
artery free flap: the first 7 years. Plast Recomtr Surg
1994;93(3) :547-551.
4. Coleman SS, Anson BJ: Arterial patterns in the hand
based upon a study of 650 specimens. Surg Gynecol Obsrer
1961;113:409-424.
5. Costa H, Pinto A. Zenha H: Th.e posterior interosse-
ous flap-a prime technique in hand reconstruction.
clinical cases. J Plast Rec<mstr Aesthet Surg 2007;60(7):
740-747.
6. Dcvansh MS: Superficial ulnar artery flap. Piau Rectmm
Surg 1996;97(2):420-426.
7. Durgun B, Yiiccl AH, Kizilkanat ED, Dere F: Multiple
arterial variation of the human upper limb. Surg RJldiol
Anat 2002;24(2):125-128.
8. Gabr EM. Kobayashi MR. Salibian AH, Armstrong WE,
Sun.dine ~ Calvert Jw. Evans GR: Role of ulnar forearm
free flap in oroman.dlbular reconstruction. ~
2004;24(4):285-288.
9. Standring S: Gray'r Anatm!U': The Anaromical Bam of
Clinical Praaice. 39th ed. London, UK: Churchill Living-
stone; 2004.

10. Haerle M, Hiifner HM, Dietz K, Schaller HE, Brunelli F:
Vascular dominance in the forearm. Plast Reconstr Surg
2003;111(6):1891-1898.
11. lgnatiadis lA, Mavrogenis AF, Avram AM, Georgescu AV,
Perez ML, Gerostathopoulos NE, Soucacos PN: Treat-
ment of complex hand trawna using the distal ulnar
and radial artery perforator-based flaps. Injury 2008;39
(suppl3):S116--S124.
12. Kadir S: Atlas of Normal and Mlriant Angiographic Anat-
omy. Philadelphia: WE Saunders Company; 1991.
13. LeGeyt MT, Ghobadi F: Aberrant position of the ulnar
nerve and artery overlying the carpal canal. Am J OrdJop
1998;27(6) :449-450.
14. U KK. Salibian AH, Allison GR, Krugman ME, Arm-
strong W, Wong B, Kelly T: Pharyngoesophageal recon-
struction with the ulnar forearm flap. Arch Orolaryngol
Head Neck Surg 1998;124(10):1146-1151.
15. Lovie MJ, Duncan GM, Glasson DW: The ulnar artery
forearm free flap. Br J Plast Surg 1984;37(4):486-492.
16. McCormack lJ, Cauldwell Ew, Anson BJ: Brachial and
antebrachial arterial patterns: a study of 750 extremities.
Surg Gynecol Obstet1953;96(1):43-54.
17. Moore KL, Dalley AF: Clinically Oriented Anatono~. 4th
ed. Baltimore, MD: UppincottWilliams &Wilkins; 1999.
18. Ozkus K, Pe~telmaci T, Soyluoglu AI, Akkin SM, Ozkus
HI: Variations of the superficial palmar arch. Folia Mor-
pho/ (W&rsz) 1998;57(3):251-255.
19. Porter CJ, Mellow CG: Anatomically aberrant forearm
arteries: an absent radial artery with co-dominant median
and ulnar arteries. Br J Plast Surg 2001;54(8):727-728.
20. Purl V, Mahe:ndru S, Rana R: Posterior interosseous artery
flap, fasciosubcutaneous pedicle technique: a study of 25
cases.JPlasz&comzrAesthetSurg2007;60(12):1331-1337.
21. Rodriguez ED, Mithani SK, Bluebond-Langner R,
Manson PN: Hand evaluation following ulnar forearm
perforator flap harvest: a prospective study. Plast Reconstr
Surg 2007;120(6):1598-1601.
22. Salibian AH, Allison GR, Krugman ME, Strelzow VV,
Brugman JJ, Rappaport I, McMicken BL, Etchepare TL:
ULNAR FOREARM FREE FLAP 287
Reconstruction of the base of the tongue with the micro-
vascular ulnar forearm flap: a functional assessment. Plast
Reconszr Surg 1995;96(5):1081-1089; discussion 1090-
1091.
23. Salibian AH, Allison GR, Armstrong WB, Krugman ME,
Strelzow VV, KellyT, Brugman JJ, Hoerauf P, McMicken
BL: Functional hemitongue reconstruction with the
microvascular ulnar forearm flap. Plast Reconstr Surg
1999; 104 (3) :654-660.
24. Sieg P, Jacobsen HC, Hakim SG, Hennes D: Superficial
ulnar artery: curse or blessing in harvesting fasciocutane-
ous forearm flaps. Head Neck 2006;28(5):447-452.
25. Sieg P, Bie:rwolf S: Ulnar versus radial forearm flap in
head and neck reconstruction: an experimental and clini-
cal study. Head Neck 2001;23(11):967-971.
26. Shen S, Pang J, Seneviratne S, Ashton MW, Corlett
RJ, Taylor GI: A comparative anatomical study of bra-
chioradialis and flexor carpi ulnaris muscles: implica-
tions for total tongue reconstruction. Plast &conszr Surg
2008;121 (3):816-829.
27. Song R, Gao Y, SongY, YU Y, SongY: The forearm flap.
Clin Plast Surg 1982;9(1):21-26.
28. Tcacencu 1: A rare hwnan variation: a major branch of
the ulnar artery found in the carpal tunnel. Surg Radiol
Anat 2001;23(5):359-360.
29. Varro J, Horvath L, Varga G: Anatomy of the hand arter-
ies based on angiographic studies. Magy Traumatol Orthop
Helyreallito Seb 1978;21(2):127-134.
30. Wax MK, Rosenthal EL, Winslow CP, Bascom DA,
Andersen PE: The ulnar fasciocutaneous free flap in head
and neck reconstruction. Laryngoscope 2002;112(12):
2155-2160.
31. Yii Nw. Niranjan NS: Fascial flaps based on perforators
for reconstruction of defects in the distal forearm. Br J
Plast Surg 1999;52(7):534-540.
32. Yildirim M, Kopuz C, Yildiz Z: Report of a rare human
variation: the superficial ulnar artery arising from the
axillary artery. Okajimas Folia Anaz Jpn 1999;76(4):
187-191.
"T""'he subscapular system of flaps is unique among
~ all awilable donor sites for free tissue transfer
because of the diversity of tissue types, the potential
surface area of tissue that can be transfe!Ted, and the
mobility of the various flaps relative to each other, and
in particular, to the bone. In addition, this donor site is
more easily camouflaged than most other sites due to
its location. One of the earliest free flaps that was har-
vested from the axillary region was the lateral thoracic
flap, which is based on the lateral thoracic artery or the
accessory lateral thoracic artery (2,4,7). The variabil-
ity in the vascular anatomy to this flap, in addition to
the emergence of the subscapular system of flaps, soon
relegated the lateral thoracic flap to one of historical
interest only.
288
The branching pattern of the subscapular artery and
vein permits the transfer of the following flaps on a sin-
gle pedicle (Fig. 18-1):
1. Scapular fasciocutaneous flap.
2. Parascapular fasciocutaneous flap.
3. Scapular-parascapular osteofasciocutaneous flap (e.g.,
angular branch osteowtaneous flap with two pedicles).
4. Latissimus dorsi muscle flap.
S. Latissimus dorsi musculocutaneous flap.
6. Latissimus dorsi rib osteomusculocutaneous flap.
7. Serratus anterior muscle flap.
8. Serratus anterior musculocutaneous flap.
9. Serratus anterior musculocutaneous-no flap.
 SUBSCAPULARSYSTEM 289

Axillary a.

Subscapular a.

Circumflex scapular

Descending
cutaneous Thoracodorsal a.
branch

Angular branch

Periosteal

Branch to serratus anterior m.

Vertical branch to
Transverse branch tn ----7~ latissimus dorsi m.
latissimus dorsi m.

FIGURE 18-1. The multiple branches of the subscapular artery are the key to understand-
ing the range of flaps with independent pedicles that can be harvested with this system. The
subscapular artery divides into the circumflex scapular and the thoracodorsal arteries. The
former supplies the periosteum of the lateral scapular border and the scapular and parascapu-
lar fasciocutaneous flaps. The thoracodorsal artery gives off the angular branch to the tip of the
scapular bone and the muscular branch to the serratus anterior. It terminates in the transverse
and vertical branches that supply the latissimus dorsi muscle.
290 CHAPTER 18
Although the necessity to transfer more than one flap
to the head and neck is rare, there are occasions when
this can be very advantageous and help to overcome
some of the most challenging reconstructive situations.
Harii et al. (6,8) reported the combination of a latissi-
mus dorsi musculocutaneous flap with a serratus ante-
rior musculocutaneous flap to provide inner and outer
linings for through-and-through defects of the cheek.
In 1984, Batchelor and Tully {3) reported on the resur-
facing of a total scalp defect with a single cutaneous flap
that included the territories of the scapular, parascapu-
lar, latissimus dorsi, and lateral thoracic flaps. In their
patient, the thoracodorsal and subscapular arteries had
a separate origin and required two arterial anastomoses.
Richards (1 0) used the combination of the latissimus
dorsi and the serratus anterior muscles to resurface a
scalp and calvarial defect. A vascularized segment of the
6th rib was included with the serratus muscle to recon-
struct the superior and lateral orbital rims.
In a large series using the scapular osteocutaneous
flap for head and neck reconstruction, Swartz et al. {11)
included the latissimus dorsi musculocutaneous flap for
reconstruction of a composite defect of the oral cavity.
The symphysis was restored with the lateral scapular bor-
der, and the floor of mouth was closed with the scapular
skin paddle. The latissimus dorsi flap provided coverage
of the external cutaneous defect. Similarly, Granick et al.
(5) transferred the latissimus dorsi-scapular osteocutane-
ous flap to close a composite defect, but skin grafted the
latissimus dorsi muscle because of the bulk of its cutane-
ous island. These authors also reported that the use of this
composite flap with an innervated latissimus dorsi mus-
cle improves lower lip competence. The muscle was sus-
pended from the 01bicularis oris on both sides, and the
thoracodorsal nerve was anastomosed to the lower divi-
sion of the facial nerve. The overlying skin of the latissi-
mus dorsi was used to reconstruct the cutaneous defect of
the mentum. The restoration of dynamic activity and the
degree of oral competence could not be assessed, because
the patient had an anoxic cerebral injury followiDg surgery.
However, the concept of using the dynamic potential
of the latissimus dorsi muscle remains an intriguing one.
This is particularly true when reconstructing composite
defects of the cheek in which the mimetic muscles are
removed but the proximal portion of the facial nerve
remains intact. In this situation, a portion of the latis-
simus dorsi muscle can be inset to restore upward and
lateral movement to the corner of the mouth by anas-
tomosing the thoracodorsal nerve to the facial nerve.
Complex reconstructions that include the mandible,
inner and outer linings, and facial reanimation can be
achieved by the transfer of a variety of flaps based on the
subscapular pedicle (1). I have found the combination
of the scapular osteocutaneous flap with the latissimus
dorsi muscle to be particularly beneficial in patients
who undergo salvage surgery after radiation therapy. In
this population of patients, the placement of a sheet of
healthy, well-vascularized muscle in the neck over the
carotid artery and jugular vein provides a valuable layer
of protection in the event of either cervical skin break-
down or intraluminal breakdown with the development
of a salivary fistula.
Composite defects of the midface often cause very
complex problems because of the three-dimensional
nature of the reconstruction. Replacement of the palate
with vascularized bone must often be accompanied by
restoring oral lining, nasal lining, and cutaneous defects
of the cheek. The freedom to move the multiple soft tis-
sue flaps of the subscapular system relative to the bone
has been extremely beneficial (1,9).
The latissimus dorsi, its musculocutaneous flap, the
scapular-parascapular fasciocutaneous and osteofascio-
cutaneous flaps, as well as the serratus anterior flap with
the subjacent rib, are some of the more commonly used
components of the subscapular system of flaps.
Acknowledgments
The author acknowledges Michael J. Sullivan, MD, for
his contributions to this chapter in the first edition of
this atlas.
REFERENCES
1. Aviv JE, Urken ML, Vickery C, Weinberg H,
Buchbinder D, Biller HF: The combined latissimus dorsi-
scapular free flap in head and neck reconstruction. Arch
Otolaryngol Head Neck Surg 1991;117:1242.
2. BakerS: Free lateral thoracic flap in head and neck recon-
struction.Arch Otolaryngol Head Neck Surg 1981; 107:409.
3. Batchelor A, Tully 1..: A multiple territory free tissue
transfer for reconstruction of a large scalp defect. Br J
Plast Surg 1984;37:76.
4. Baudet J, Guimberteau JC, Nascimento E: Successful
clinical transfer of two free thoracodorsal axillary flaps.
Plast Reconstr Surg 1976;58:680.
5. Granick MS, Newton ED, Hanna DC: Scapular free flap
for repair of massive lower facial composite defects. Head
Neck Surg 1986;8:436-441.
6. Harii K. Ono I, Ebihara S: Closure of total cheek defects
with two combined myocutaneous free flaps. Arch
Otolaryngol Head Neck Surg 1982; 108:303.
7. Harii K, Torii S, Sekiguchi J:The free lateral thoracic flap.
Plast Reconsr.r Surg 1978;62:212.
8. Harii K, Yamada A, Ishihara K. Miki Y, Itoh M: A free
transfer of both latissimus dorsi and serratus anterior

flaps with thoracodorsal vessel anastomosis. Plart Reconm
Surg 1982;70:720.
9. Jones N, Hardesty R, SwartzW, Ramasastry S, Heckler F,
Newton E: Extensive and complex defects of the scalp,
middle third of the face, and palate: the role of microsur-
gical reconstruction. Plarl Reconstr Surg 1988;82:937.
SUBSCAPULARSYSTEM 291
10. Richards M: Free composite reconstruction of a complex
craniofacial defect. Aust N Z J Surg 1987;57:129.
1 1. Swartz W, Banis J, Newton D, Ramasastry S, Jones N,
Acland R: The osteocutaneous scapular flap for
mandibular and maxillary reconstruction. Plart Reconstr
Surg 1986;77:530.
 aijo (20) should be credited with being the first to
S recognize the potential for transfer of a vascular-
ized cutaneous free fiap based on the circumfiex scapu-
lar artery (CSA) and circumBex scapular vein (CSV).
Dye-injection studies of the CSA led to staining of
the skin overlying the scapula. Saijo hypothesized that
separate uial pattern flaps based on the CSA and the
thoracodorsal artery and vein could be successfully har-
vested as free fiaps. dos Santos (10,11) made significant
contributions to our understanding of this donor site
through a large series of cadaver dissections and per-
formed the first clinical ttansfer of a free scapular flap
292
along with Gilbert and Teot (13). The next major mile-
stone in the evolution of this donor site was reported by
Teot et al. (25) in1981 when a series of cadaver dis-
sections demonstrated the area of vascularized bone
that could be harvested from the lateral scapular border
based on the circumfiex scapular pedicle. In addition,
they described the various cross-sectional shapes and
dimensions ofbones that were present at diffen:nt points
along the lateral border from the glenohumeral joint to
the tip. The technical aspects of harvesting a composite
flap from this region were described along with three
successful clinical cases of free scapular osteocutaneous
 SCAPULAR AND PARASCAPULAR FASCIDCUTANEDUS AND OSTEOFASCIOCUTANEOUS AND SUBSCAPULAR MEGA FLAP
flap transfers that included one case of orbital floor and
one case of mandibular reconstruction. The use of this
composite flap for head and neck reconstruction was
popularized by the work of Swartz et al. (23) in the lat-
ter part of the 1980s.
In 1982, Nassif et al. (18) reported a longitudinally
oriented skin paddle referred to as the parascapular flap.
This fasciocutaneous flap was based on the descend-
ing branches of the CSA and CSV. An additional major
contribution to the clinical usefulness of this donor site
was made by Coleman and Sultan (6) in 1991 who
described an alternative vascular supply to the tip of the
lateral scapular border through the angular branch of
the thoracodorsal artery and vein.
FLAP DESIGN AND UTILIZATION
The unique features that make the scapular system of
flaps so useful for head and neck reconstruction are as
follows:
1. The long length and large caliber of the vascular
pedicle.
2. The abundant surface area of relatively thin skin
that can be transferred.
3. The separation of the soft tissue and bone flaps,
which provides the most freedom for three-dimen-
sional insetting compared to any of the available
composite free flaps.
4. The ability to combine the latissimus dorsi and the
serratus anterior muscles, along with overlying skin,
and adjacent segments of rib.
The dimensions of the territory supplied by the cir-
cumflex scapular pedicle have grown since the earliest
description. Based on the results of the dye-injection
studies, dos Santos (10,11) placed the following limita-
tions on the skin flap supplied by the CSA: 10 em in the
vertical dimension, 13 em in the horizontal dimension,
no further cephalad than the scapular spine, no further
caudal than 3 em above the inferior scapular tip, no fur-
ther medial than 2 em from the vertebral column, and
no further lateral than the posterior axillary line. Other
authors placed similar or even more stringent restric-
tions (2,14,29). Urbaniak et al. (29) proposed the "rule
of twos" to define the skin territory. They advised that
the upper limit should be 2 em inferior to the scapu-
lar spine and the inferior limit should be 2 em superior
to the tip of the scapula. They also limited the medial
extent to a point 2 em lateral to the vertebral processes.
In 1982, Nassif et al. (18) introduced the parascapu-
lar flap, which is based on the descending branch of the
CSA. This vertical skin paddle, oriented over the lateral
scapular border, markedly expanded the dimensions
of the cutaneous territory of the CSA (4). Successful
293
parascapular flaps have been reported with lengths up
to 25 em (5). An anatomic study of the dorsal thoracic
fascia by Kim et al. (16) provided justification for refer-
ring to the scapular and parascapular flaps as fasciocuta-
neous flaps. Their study also demonstrated rich vascular
communications between branches of the CSA and the
musculocutaneous perforators of the trapezius and latis-
simus dorsi, suggesting that much of the skin overlying
the latissimus dorsi muscle could be transferred based
on the CSA by including the dorsal thoracic fascia. Kim
et al. also showed that the CSA runs within the layers
of the posterior thoracic fascia, which could therefore be
transferred by itself as a thin vascularized tissue layer. De-
epithelialized scapular and parascapular flaps have been
used extensively for soft tissue augmentation to restore
the facial contour in a wide range of disorders, including
hemifacial microsomia; atrophy; and deformities caused
by radiation, trauma, and ablative cancer surgery (28).
The anatomic features of the dorsal thoracic fascia and
the transverse and descending branches of the CSA pro-
vide the basis for designing multiple skin paddles when
needed. Scapular and parascapular skin flaps may be
simultaneously transferred based on a single CSA. Alter-
natively, the dorsal thoracic fascia allows the transfer of
multiple skin paddles that are separated by a considera-
ble distance and connected only by this fascial layer (26).
The medial extent of the scapular skin flap has been
traditionally limited by the midline of the back. This
concept was challenged byThoma and Hiddle (27) who
reported their experience in five patients who required
"extended free scapular flaps;' which crossed the mid-
line and measured up to 39 em in length. The basis for
their work was a report by Batchelor and Bardsley (3)
of a scapula flap with two pedicles that spanned the
entire width of the back and was based on the anas-
tomosis of both CSAs. However, after the release of
the clamps on the anastomosis of the first CSA and
CSV and prior to performing the second set of anas-
tomoses, the entire skin paddle was noted to be well
vascularized. The reliability of the extended scapular
flap can be examined by invoking the angiosome con-
cept ofTaylor and Palmer (24). The vascular anatomy
of the back can be separated into longitudinally divided
zones based on different source arteries. The regions
over each scapula represent the primary angiosomes
of the CSA. On either side of the midline, there are
two trapezius angiosomes, each supplied by the trans-
verse cervical artery and vein. Hence, as the transverse
dimensions of a scapular flap are extended, it is neces-
sary to cross three successive angiosomes (Fig. 19-1).
Taylor and Palmer suggested that the source artery of
a single angiosome can reliably capture the territory of
an adjacent angiosome, but that the angiosome once
removed (i.e., zone III) is less predictable. The dynam-
ics of the flow across angiosomes through the connect-
ing choke vessels can be altered by prior ligation of the
294 CHAPTER 19

RGURE 19-1. The vascular zones of the scapular system have been labeled I through IV. When
harvesting a transverse scapular flap based on the left CSA, the skin paddle extends from the pri-
mary angiosome (I) into the angiosome of the transverse cervical artery and vein (II). As this trans-
verse flap crosses the midline, the third angiosome in the series, that of the contralateral trapezius
flap, is entered (Ill). Finally, the fourth angiosome in the series is the one primarily supplied by the
contralateral CSA (IV). By performing a radical neck dissection on the left side and interrupting
the transverse cervical artery, the skin overlying zone II may be partially or totally delayed so as to
make it, and the skin of zone Ill, more reliably captured in the transfer of the left scapular flap.
 SCAPULAR AND PARASCAPULAR FASCIOCUTANEOUS AND OSTEOFASCIOCUTANEOUS AND SUBSCAPULAR MEGA FLAP 295

source artery that supplies the adjacent territory. This
maneuver essentially produces a delay phenomenon
by opening up the choke vessels and achieving a more
favorable hemodynamic situation by which to capture
more distant angiosomes. It is tempting to speculate
that such a delay phenomenon may be achieved when
harvesting an extended scapular flap on the side of the
back on which a radical neck dissection was previously
performed. Prior interruption of the transverse cervical
vessels should allow the scapular flap to be more reli-
ably extended across the midline.
Regardless of the design or dimensions of the distal
portions of the scapular-parascapular flaps, it is critical
that the base of the flap is centered over the infraspinatus
fossa, which is the dorsal enent of the triangular space.
That space may be found by palpation of the muscular
hiatus along the lateral scapular border or by Doppler
sonographic localization of the CSA as it emerges from
its origin in the uilla. The infraspinatus fossa has been
roughly localized to a point either halfway (14) or two
fifths of the way (18) along the lateral scapular border
when measuring from the spine to the tip (Fig. 19-2).

Circumflex scapular a.

Infraspinatus m.

Triangular
space

Thoracodorsal a.

Triceps brachii m.
long head

- - - - - - - Serratus anterior m.

Latissimus dorsi m.

FIGURE 19-Z. The muscles that make up the posterior axillary and scapular region are critical to
the understanding of the subscapular system of flaps. The CSA and CSV traverse the triangular
space before reaching the infraspinatus fossa. The triangular space is bounded by the teres major,
teres minor, and the long head of the triceps muscles. The teres minor originates from the upper
two thirds of the lateral scapular border and inserts into the greater tuberosity of the humerus.
Its action is opposite to that of the teres major, which arises from the inferior aspect of the lateral
scapular border and inserts into the bicipital groove of the humerus. The terminal branches of the
CSA anastomose with the suprascapular and transverse cervical vessels.
29& CHAPTER 19

The successful transfer of vascularized hone from the

lateral scapular border dramatically expanded the versa-
tility ofthe subscapular donor site and the range of appli-
cations to the head and neck. The length of bone that
can be harvested ranges from 10 to 14 em, depending on
the sex and size of the patient (Figs. 19-3 and 19-4). It is
limited in its cephalad extent by the glenohumeral joint,
which must be protected. The variations in thickness of
the bone allow it to be used for different purposes in the
head and neck. The thin blade of the midportion of the
scapula is useful for reconstructing the orbital floor and
the palate. The greatest application of this composite
flap is in restoring the bone and soft tissue components
of oromand:ibular defects. The periosteal blood supply
derived from the CSA permits osteotomies to be made
to contour the bone to the shape of the mandible. The
vascularity to distal segments is maintained as long as
the periosteum and a cuff of muscle are preserved when
providing exposure to perform the osteotomy (23).
The separation of the scapular and parascapular skin
flaps from the bone provides a unique capacity to
restore the complex three-dimensional defects of the
FIGURE 19-4. A lateral view of the scapular border reveals
head and neck. Extensive experience with the compos-
that the bone is fairly straight Contouring of this bone to fit
ite scapular flap has demonstrated ita utility in recon-
the shape of bone defects in the maxillomandibular skeleton
structing the oral cavity following trauma and ablative
requires ostectomies to be performed while preserving the
surgery (21,22).
nutrient periosteal layer.

Swartz et al. (23) descnoed an extension of the bone

harvest along the medial aspect of the scapular tip to
provide an additional 3 to 4 em of bone. However, the
blood supply to distal portions of the scapular bone was
somewhat suspect, especially after the creation of an
osteotomy. This was evident by the findings on postop-
erative hone scans, the occurrence of nonunions, and the
necessity to perform sequestrectomies. In 1991, Cole-
man and Sultan (6) reported their experimental and
clinical findings using a separate vascular supply to the
caudal portion of the lateral scapular border based on
the angular branch of the thoracodorsal artery and vein
(Fig. 19-5). They credited Deraemaecher et al. (9) for
the initial discovery and report of this branch in 1988.
The identification of a separate blood supply to the tip
of the scapula is important for a number of reasons.
1. Osteotomies can be made in the lateral scapula with-
out concern about devascularization ofthe distal bone
because of the preservation of the angular branch.
RGURE 19-3. The lateral scapular border. extending 2. Two separate segments of bone with their own
from the glenohumeral joint to the scapular tip, may be vascular supply can he used to re<:onstruct bone
transferred as a vascularized bone flap based on the CSA defects that are separated in space.
and the CSV. The thick bicortical bone of the lateral border 3. The tip of the scapula can be ttansferred alone
becomes markedly thinner in the midsection of the blade without requiring the intervening bone of the lateral
of the scapula. Approximately 10 to 14 em in length can be scapula.The tip provides a reasonable replication of
harvested from the lateral border. The bone cuts may be the shape of the hard palate for restoring bone to
extended to include the inferomedial border. that region of the oral cavity.
 SCAPULAR AND PARASCAPULAR FASCIOCUTANEOUS AND OSTEOFASCIOCUTANEOUS AND SUBSCAPULAR MEGA FLAP 297

4. Maximum separation betWeen the soft and hard
tissues can be achieved by tran.sfenin.g a scapular
or paraacapular Bap based on the CSA and a seg--
ment of bone supplied by the angular branch. These
two components may be separated by as much as
15 em. This contrasts with the 2.5 em that separates
the skin and bone segments when transfe:r:ri:ng both
on the CSA (Fig. 19-6).
Coleman and Sultan (6) reported the successful trans-
fer of segments of the scapula measuring up to 8 em
in length. They used the osteocutaneous fiap with
two pedicles to advance the symphysis in a patient with
a hypoplastic mandible by performi:ng bilateral mandib-
ular osteotomies and inserting two vascularized bone
segments to maintain the advanced symphysis. They
also reported using the scapular tip to reconstruct the
orbital floor. We have had experience in using the scapu-
lar tip supplied by the angular branch to reconstruct
partial and total hard palate defects.
Thoma et al. (26) transferred the medial scapular
border, based on the CSA's supply of the dorsal thomcic
fascia. The vascularity of this segment ofbone is entirely

Axillary a. Circumflex
scapular a.
Subscapular a.

Thoracodorsal a.

~-~~T-~~~-!.-- Periosteal
blood supply

Angular branch
Branch to
latissimus dorsi m.

serratus anterior
Latissimus dorsi m. ~----

FIGURE 19-5. With the overlying muscles removed, the blood supply to the scapular bone is
shown arising from periosteal feeders of the CSA that supplies the upper portions ofthe lateral
scapula and the angular branch that supplies the periosteum of the caudal scapular border.
298 CHAPTER 19

A. B.

Parascapular flap

Latissimus dorsi
musculocutaneous flap

Circumflex
scapular a.

dorsi m.

D.
RGURE 19-6. A variety of different flaps can be simultaneously harvested on 1he subscapular
artery and vein. A: The bone ofthe lateral scapular border can be transferred with a small cuff
of muscle for pure bony reconstructions. The caudal tip can be used to reconstruct the angle of
the mandible. Alternatively, the thin central portion of the bone can be used as a shelf to replace
the hard palate. B: The most common varieties of subscapular soft tissue flaps are shown. These
flaps can be harvested separately or in concert In addition, the serratus anterior muscle and
musculocutaneous flaps can be included with 1his vascular axis. C: The bone of the lateral scapu-
lar border can be divided on two separate vascular pedicles: the circumflex scapular and the
angular. A scapular-parascapular cutaneous flap has been reflected to show the periosteal vas-
cular supply. D: The multiple different soft tissue and osseous components 1hat can be harvested
on 1he subscapular system are shown. Intramuscular bifurcation of1he thoracodorsal artery
allows splitting of the latissimus dorsi muscle. Both 1he latissimus dorsi and the serratus anterior
muscles can be transferred with 1heir nerve supply for restoration of dynamic motor activity.
 SCAPULAR AND PARASCAPULAR FASCIDCUTANEDUS AND OSTEOFASCIOCUTANEOUS AND SUBSCAPULAR MEGA FLAP
dependent on preserving the fascial attachments to it.
The rationale for designing this composite flap is that
it lengthens the vascular pedicle to the bone and avoids
disruption of the muscular attachments to the lateral
scapula. The disadvantages of this flap are that the bone
of the medial scapular border is thinner than that of the
lateral border, the relationship of the bone to the under-
surface of the skin must be maintained to preserve the
fascial blood supply, and the tolerance of this bone to
contouring osteotomies is uncertain.
NEUROVASCULAR ANATOMY
The parent vessels of the scapular flap are the subscapu-
lar artery and vein, which arise from the third part of the
axillary artery and vein (Fig. 19-6). However, depend-
ing on the length of the vascular leash that is required,
the CSA and CSV may also be used, thereby, preserving
the vascular supply to the latissimus dorsi through the
thoracodorsal vessels. The CSA runs through the trian-
gular space where it supplies muscular branches to the
teres major and minor and the periosteal branches to
the lateral border of the scapula. The CSA terminates in
the transverse and descending cutaneous branches that
supply the scapular and parascapular fasciocutaneous
flaps. In its course, the CSA is accompanied by paired
venae comitanteJ. These two veins are usually different
sizes, with the larger having a diameter in the range of
2.5 to 4.0 mm. In the majority of cases, the two venae
comitan~s join with the thoracodorsal vein. In approxi-
mately 10% of cases, the CSV enters the axillary vein
separate from the thoracodorsal vein. The average diam-
eter of the CSA at its origin from the subscapular artery
is 4 mm (range, 2 to 6 mm). At its origin from the axil-
lary artery, the subscapular artery has an average diam-
eter of 6 mm (range, 4 to 8 mm) (19). dos Santos (11)
reported the diameter of the CSA to be slightly smaller
(average, 2.8 mm).
As noted previously, the vascular pedicle length var-
ies depending on the extent of proximal dissection. If
only the cutaneous branch of the CSA is used, then a
pedicle length of 4 to 6 em is obtained. When the CSA
is harvested at its takeoff from the subscapular vessels,
then the fasciocutaneous flaps have a pedicle length of
7 to 10 em. A maximum pedicle length in the range of
11 to 14 em is obtained by transecting the subscapular
vessels at their junction with the axillary artery and vein
(18).
The unique anatomy of the subscapular vascular
system provides the opportunity to lengthen the artery
by a considerable amount by utilizing reverse flow
through the thoracodorsal artery following ligation
of the subscapular artery. In this technique the distal
end of the thoracodorsal artery is anastomosed to the
recipient artery in the neck and retrograde flow in the
299
thoracodorsal artery provides antegrade flow in the
CSA. Due to the presence of the valves in the thora-
codorsal vein, the opportunity to lengthen the venous
pedicle cannot be safely accomplished through a similar
strategy of reverse flow.
The CSA and CSV run in the fascial septum between
the teres major and minor, and then they divide into the
transverse and descending branches, which run in the
fascial layers and send perforators to the overlying skin
and subcutaneous tissue (7). As noted previously, this
fascial plexus spreads out over the adjoining muscles
and communicates with the musculocutaneous perfora-
tors of the latissimus dorsi and trapezius muscles (16).
Coleman and Sultan (6) reported that the angular
branch to the tip of the scapula was present in 100% of
the cadaver dissections and clinical cases. The angular
branch arose from the thoracodorsal artery just proxi-
mal to the serratus anterior branch in 58% of cases. In
the remaining 42%, it arose from the crossing branch
of the thoracodorsal artery to the serratus anterior. In
its course toward the scapular tip, the angular branch
supplies small feeders to the subscapularis and the ser-
ratus anterior muscles prior to its terminal arborization,
which supplies the periosteum at a point about 3 em
cephalad to the inferior scapular border. Differentiation
of the two patterns of origin of the angular branch is not
critical in the flap harvest. Following identification of
the CSA, the angular branch is easily isolated by open-
ing the plane between the teres major and the latissimus
dorsi. The teres major is then transected, leaving a small
cuff attached to the scapula. The course of the angular
branch can then be readily traced (22).
The standard posterior approach to the subscapular
pedicle involves working through the triangular space
or with the added exposure afforded by transecting the
teres major. An alternative route to the proximal portion
of the pedicle involves a counter incision in the axilla
that permits a direct visualization of the thoracodor-
sal, angular, and subscapular vessels. In addition, this
maneuver allows the flap to be delivered into the axilla
without interrupting the vascular supply during closure
of the donor site (12).
The cutaneous nerve supply to the scapular region
is derived from the dorsal rami of the spinal nerves.
Following an extensive review of the literature, there
were no reported cases of a successful sensate scapula
or parascapular flap, although Upton et al. (28) noted
two unsuccessful cases following the anastomosis of the
dorsal rami.
ANATOMIC VARIATIONS
There have been no reports in which the CSA has not
been identified in the triangular space. However there is
some reported variability in the course of the descending
300 CHAPTER 19
branch of the CSA that supplies the parascapular flap.
In 7 of 30 dissections, the descending branch assumed
a course that was deep to the teres major and ascended
to the fascial layer to supply the skin by running in the
plane between the teres major and latissimus dorsi mus-
cle. Upton et al. (28) reported on two clinical cases
involving this variant and detached the teres major to
maintain continuity of the vascular supply to the paras-
capular skin. Because of the rich vascularity to the dor-
sal thoracic fascia, it is unclear that such a maneuver
and the integrity of the descending branch are critical
to successful parascapular flap transfer.
In a series of 100 cadaver dissections, Rowsell et al.
{19) reponed that the subscapular artery arose from the
axillary artery in 97% of cases. In 81%, it arose from the
third part, and in 13%, it was a branch of the second
part. In 3%, the subscapular artery arose from the first
part, and in 3%, it was absent. In the latter group, the
CSA was a direct branch of the axillary artery. Hitzrot
(15) divided the branching pattern of the axillary artery
into seven different types based on 4 7 cadaver dissec-
tions. He did not report any cases in which the thora-
codorsal artery and CSA had a separate origin from the
axillary artery. However, in two cases, a large subscapu-
lar artery provided the origin for the acromiothoracic
trunk, in addition to the usual branches. DeGaris and
Swartley (8) performed a much more extensive study of
512 axillary artery dissections and divided the branch-
ing patterns into 23 different groups. They noted a
common trunk of the subscapular and thoracoacromial
arteries in 4.5% of the dissections. A separate origin
for the thoracodorsal artery and CSA was rare, noted
in 0.8% of cases. In a third anatomic series of 50 dissec-
tions, Bartlett et al. (1) reponed this variant as occur-
ring with an incidence of 4%. The presence of double
CSAs was noted in 8% of cases.
Variations in the venous anatomy are much more
common. Separate origins for the CSA and the thora-
codorsal vein from the axillary vein were reported to
occur in 12% of dissections (1).
POTENTIAL PITFALLS
There are a variety of complications that have been
reponed in harvesting flaps based on the subscapular
vascular system. There is potential morbidity to the bra-
chial plexus as a result of arm position during flap har-
vest. Similarly, uncertainty regarding the integrity of the
vascular supply in a patient who had undergone prior
axillary node dissection would contraindicate against
the use of this donor site.
The bone stock of the lateral scapular border is
fairly limited when considering placement of endosteal
implants for dental rehabilitation. The most favorable
quantity of bone is obtained from the caudal aspect of
the lateral scapula, and thus, this must be incorporated
into the strategy of graft orientation in mandibular
reconstruction (17). Particular caution is advised when
making the cephalad bone cuts along the lateral scapu-
lar border to be certain to stay 1 em below the glenoid
fossa to avoid injury to the joint space.
A variety of muscles in the axilla may be disrupted
in the process of harvesting an osteocutaneous scapular
flap. Perhaps the most significant of these is the teres
major, which is usually detached in whole or in pan
from its origin to the scapula. In addition, the tech-
nique of flap harvest most often leads to denervation
and devascularization of this muscle. The teres major
is an internal rotator, extensor, and adductor of the
arm. There is some uncertainty as to whether shoulder
function is affected by reattaching the teres major by
placing large horizontal mattress sutures through the
muscle overlying the dorsal aspect of the scapular bone.
Although reattachment would help to anchor the scap-
ula and prevent winging, a scarred, denervated, fibrotic
muscle may actually limit the range of motion of the
arm. If there is concern for the vascularity of the teres
major at the end of the procedure, it should be excised
rather than risk a wound infection as a result of muscle
necrosis (6).
The aesthetic appearance of the donor site is usually
related to the amount of skin that is removed. Widened
scars are not uncommon when large cutaneous flaps are
required. A skin graft placed on the back to close this
donor site is less favorable and should be avoided by
judicious planning. Pretransfer expansion of the scapu-
lar region was reponed for unusual circumstances (28).
POSTOPERATIVE CARE
The shoulder disability following the harvest of flaps
from the scapular region is related to the nature of the
tissue that is removed. The use of the scapular and par-
ascapular flaps alone is unlikely to produce significant
morbidity. Postoperative rehabilitation is indicated in
patients who undergo osteocutaneous flap harvest. Fol-
lowing 3 to 4 days of immobilizing the arm against the
trunk, active and passive range-of-motion exercises are
begun. A program for strengthening the muscles of the
shoulder girdle should be instituted within 2 to 3 weeks
after surgery and monitored by a physical therapist on
an outpatient basis.
Acknowledgments
The authors acknowledge the contributions of
Dr. Michael J. Sullivan to the writing of this chapter in
the first edition of this book.
 SCAPULAR AND PARASCAPULAR FASCIOCUTANEOUS AND OSTEOFASCIOCUTANEOUS AND SUBSCAPULAR MEGA FLAP 301

Scapular Osteocutaneous Flap

FIGURE 19-7. The topographic anatomy is

outlined on the upper /stetBI bsck. The medial
and lateral borders of the scapula are outlined
in black. The muscular triangle located medial
to the lateral border of the scapula is composed
ofthe teres major, teres minor, and the long
head of the triceps. This triangle can be identi-
fied by palpation or Doppler ultrasonography
ofthe CSA. The approximate courses of the
transverse and descending branches of the
CSA are drawn.
FIGURE 19-8. The entire arm and shoulder are
prepared into the operative field to allow for
shoulder mobilization during the dissection to
improve visualization of the vessels in the axilla.
Patients are positioned on their sides with an
angulation of approximately 45.ln most cases,
this position accommodates both the ablative
and reconstructive teams. An axillary roll must
be placed under the contralateral axilla. A
transverse scapular flap has been drawn and
will be harvested in this dissection; however,
over the years, the author has come to prefer
the design of a parascapularflap in order to
facilitate the harvest of the lateral scapular
border, which requires transection of the teres
major muscle from the lateral border. In addi-
tion, the harvest of a parascapular flap requires
less turning of the patient than that required in
the harvest of a transverse scapular flap.

FIGURE 19-9. The dissection proceeds in a

medial to lateral direction. The initial incisions
are made through the skin and subcutaneous
tissue to the deep fascia overlying the rhomboid
and infraspinatus musculature.
302 CHAPTER 19

Scapular Osteocutaneous Flap

FIGURE 1~10. The teres major is an important

landmark. Careful sharp and blunt dissection
along the upper border afthis muscle advances
the dissection into the muscular triangle. As
soon as the circumflex pedicle is identified, the
teres major muscle is detached from the lateral
scapular border while leaving a small cuff of
muscle on the bone. Performing this maneuver
early in the procedure affords the surgeon with
maximum exposure of the course af the circum-
flex pedicle.

FIGURE 19-11. The circumflex scapular ves-

sels are easily palpated as they course onto
the undersurface of the flap. Branches from the
pedicle to the teres major must be ligated.

FIGURE 19-12. The superior incision has been

mad e. The flap is elevated off the deltoid and
the teres minor. As the dissection praceeds
along the inferior border of the teres minor,
the CSA and CSVare again visualized and the
upper limit of the muscular triangle is defined.
 SCAPULAR AND PARASCAPULAR FASCIOCUTANEOUS AND OSTEOFASCIOCUTANEOUS AND SUBSCAPULAR MEGA FLAP 303

Scapular Osteocutaneous Flap

FIGURE 1913. As the scapular flap is elevated

on its pedicle, care is taken to preserve a cuff of
soft tissue attachment of the cutaneous paddle
to the fascia of the infraspinatus along the lat-
eral scapular border. The CSA and CSV (srrow)
are easily visualized on the undersurface of the
scapular flap.

Circumflex
scapular a.

Teres minor m.
retracted

--- ,..,_,...~"Mi+-~*--~~.1,.- perforators to

bone ligated

Infraspinatus m.

FIGURE 19-14. The anatomy of the muscular triangle reveals the thoracodorsal artery continu-
ing in its caudal course and the CSA supplying the scapular skin paddle. This illustration shows
the periosteal perforators transected, which must be done if a fasciocutaneous flap, without
bone, is harvested.
304 CHAPTER 19

Scapular Osteocutaneous Flap Thoracodoreal a.

FIGURE 19-15. When the bane af the latera I

scapula is to be harvested, the periosteal feed-
ers must be preserved. This blood supply to the
bone is never skeletonized to the extent shown
in this illustration. Perforators to bone Circumflex scapular a.

FIGURE 19-16. The harvest of a composite

osteocutaneous flap requires transection of
the teres major (large arrow) from the lateral
scapular border. A cuff of this muscle is left
attached to the bone to protect the periosteal
blood supply. The inferior border of the teres
major must be separated from the latissimus
dorsi Ismail arrows).

FIGURE 19-17. Following transection of the

teres major, the thoracodorsal(srrow}, angular,
and proximal circumflex scapular vessels can
be visualized.
 SCAPULAR AND PARASCAPULAR FASCIOCUTANEOUS AND OSTEOFASCIOCUTANEOUS AND SUBSCAPULAR MEGA FLAP 305

Scapular Osteocutaneous Flap

FIGURE 1918. With the cut edge of the

teres major and the long head of the triceps
retracted, 1he vascular anatomy of the axilla
is well visualized. The descending course of
the thoracodorsal system is easily seen (large
arrow). The primary neurovascular pedicle to
the teres major (small arrow) must be tran-
sected to allow fur1her dissection in the axilla.
This maneuver will permit complete visualiza-
tion of the vascular pedicle up to the axillary
vessels. It will also permit identification and
harvest of the angular branch to the scapular
tip if that is required in the harvest.

FIGURE 1919. The thoracodorsal pedicle must

be ligated (srrowt and transected to extend
the dissection proximally to the subscapular
vessels. The surgeon would not take this step
in 1he dissection if (a) the latissimus dorsi or
the angular branch supply to 1he scapula is to
be included in the dissection or (b) the CSA and
the CSV were of sufficient caliber and length for
anastomosis to the recipient vessels. Following
division of the thoracodorsal pedicle, the dis-
section can proceed to the subscapular artery
and vein.

FIGURE 1920. Access to the bone to make

the osteotomies is achieved by dividing the
infraspinatus muscle in a longitudinal direction
(srrows), leaving a 2- to 3-cm cuff attached to
the lateral border. Several muscular branches
to this muscle are encountered in the same
transverse plane as 1he triangular space, which
must be ligated and divided.
306 CHAPTER 19

Scapular Osteocutaneous Flap

Teres minor m.
(cut)

Perforators to bone

Infraspinatus m.
(cut)

Teres major m.
FIGURE 19-21. The dotted line shows the osteotomy that is made to harvest bone from the
lateral scapular border. Care must be taken when making the superior transverse cut so that the
glenohumeral joint is not violated.

FIGURE 19-22. This view is taken from the

vantage point of the iliac crest looking cephalad
toward the scapula following the osteotomy.
The subscapularis must be divided, leaving a
small cuff of muscle attached to the bone.
 SCAPULAR AND PARASCAPULAR FASCIOCUTANEOUS AND OSTEOFASCIOCUTANEOUS AND SUBSCAPULAR MEGA FLAP 307

Scapular Osteocutaneous Flap

FIGURE 1923. In the cephalad portion of the

lateral scapular dissection, the CSA and CSV
(srrowt must be directly visualized while making
the final releasing cuts of the subscapularis.

FIGURE 1924. With the osteocutaneous flap

completely isolated, except for its nutrient sup-
ply, the remainder of the dissection along the
subscapular vessels can be performed. After
this is completed, the vascular pedicle (srrowt
is transected. Particular attention must be taken
in 1he final stages of the dissection to ensure
thatthe ligation of1he subscapular artery and
vein does not compromise 1he caliber of the
axillary artery and vein. This is particularly true
of the axillary vein, which can be compromised
by pulling on that vessel in the final stages prior
to ligation of1he subscapular vein. Care is taken
during the course of this dissection to avoid
injury to the thoracodorsal nerve.

FIGURE 1925. Closure of the donor site is

accomplished by suturing the cut end of 1he
teres major to the muscles attached to 1he
lateral scapular border.
308 CHAPTER 19

Scapular Osteocutaneous Flap

FIGURE 1t-Z6. Wide undermining of the skin is

required to bring the skin edges together for a
tension-free closure. A suction drain is usually
placed through a separate opening in the skin
along the anterior axillary line.

FIGURE 1t-Z1. The osteocutaneous scapu-

lar flap has been harvested. The length ofthe
circumflex scapular and subscapular pedicle is
noted as is the freedom of movement of the skin
relative to the bone. However, it is important to
state that the degree of isolation of the circum-
flex branch to the skin paddle that is shown in
this dissection is much more than what is usu-
ally performed in a routine harvest.
 SCAPULAR AND PARASCAPULAR FASCIOCUTANEOUS AND OSTEOFASCIOCUTANEOUS AND SUBSCAPULAR MEGA FLAP 309

Scapular-Latissimus Dorsi Mega Flap

FIGURE 19-28. A:. The start of the harvest of a

combined scapular osteocutaneous flap with a
latissimus dorsi muscle or musculocutaneous A
flap involves identification of the relevant land-
marks. The segment of bone from the lateral
border of the scapula is indicated by the green
srrows. The intermuscular triangle is palpated
and marked in order to center the scapular or
parascapularflap to be harvested (blue srrow).
B: The boundaries of the parascapular skin
paddle to be harvested are indicated by the
blue arrows, while the approximate anterior
and superior border of the latissimus dorsi
muscle are indicated by the yellowsrrows. The
author prefers to harvest a parascapular flap
when harvesting bone either with or without the
latissimus dorsi because the elevation of that
flap provides immediate exposure to the teres
major muscle, which must be transected from
the lateral border of the scapula to proceed
with the harvest B
310 CHAPTER 19

Scapular-Latissimus Dorsi Mega Flap

FIGURE 19-29. The skin flaps have been ele-

vated over the anterior border of the latissimus
dorsi muscle (blue arrow) and over the supericr
transverse border of that muscle (green arrow).

FIGURE 19-30. The latissimus dorsi muscle

has been elevated off of the chest wall, and
the caudal portion is transected with a stapling
device that applies staples on both sides of the
muscle cut.

FIGURE 19-31. The posterior cut has been

made in the latissimus dorsi muscle Iwhite
arrow). The teres major muscle has been dis-
sected, so that its lower border (black arrowl
and upper border (yellow arrow) are well
defined. Early transaction of the teres major
muscle from the lateral border of the scapula
is critic aI to provide exposure to dissect the vas-
cular pedicle into the axilla {dotted line).
 SCAPULAR AND PARASCAPULAR FASCIOCUTANEOUS AND OSTEOFASCIOCUTANEOUS AND SUBSCAPULAR MEGA FLAP 311

Scapular-Latissimus Dorsi Mega Flap

FIGURE 19-32. Elevation of the latissimus dorsi

muscle allows identification ofthe vascular
hilum where 1he thoracodorsal artery and vein
enter the undersurface of the muscle (yellow
arrow). In the course of this dissection, the
branches to the serratus muscle must be identi
tied and transected. The para scapular skin
paddle has been elevated to the upper border
of the teres major muscle where the circumflex
scapular artery and vein are identified (white
arrow).

FIGURE 19-33. The skin paddle {blue arrow)

has been completely mobilized with isolation of
the circumflex vessels {yellow arrowt and 1he
thoracodorsal vessels {black arrow).

FIGURE 19-34. The muscles overlying the

posterior aspect of 1he scapula have been
separated {black arrowt in order to gain access
for performing the osteotomies to free up the
lateral scapular bone segment.
312 CHAPTER 19

Scapular-Latissimus Dorsi Mega Flap

FIGURE 19-35. The bone cuts have been made

(oreen arrow). The latissimus dorsi muscle has
been mobilized on the thoracodorsal pedicle
(blue srrow). The angular branch to the scapu-
lar tip (blsck arrow) must be identified and
preserved in cases where the added vascular-
ity to the lower border of the scapular border
is required or desired. The angular branch may
arise directly from the thoracodorsal pedicle
or from a branch to serratus anterior. This is
particularly true in cases where multiple con-
touring osteotomies ofthe lateral scapular bone
are required. The cut margin of the teres major
muscle (yellow arrow) has been retracted and
provides the critical exposure for optimal mobi-
lization of the subsea pular vessels.

FIGURE 19-36. The composite flap has been

harvested with the latera I border of the scapula
and the separate soft tissue flaps, the para-
scapular flap, and the latissimus dorsi muscle
flap. The neurovascular bundle is demonstrated
with the subscapular artery (red pin), subscapu-
lar vein (blue pin), and the thoracodorsal nerve
(yellow pin}. The angular branch to the lower
border of the lateral border of the scapula has
been preserved (yellow arrow).
 SCAPULAR AND PARASCAPULAR FASCIOCUTANEOUS AND OSTEOFASCIOCUTANEOUS AND SUBSCAPULAR MEGA FLAP 313

The Scapular Tip Based on the Angular Branch and the Parascapular Flap

FIGURE 19-31. A:. Harvest of the composite

scapular tip and parascapularfasciocutaneous
flap is commenced with marking the important
anatomic landmarks on the skin. The lateral bor-
der of the scapular bone (black arrow) and the
approximate course of the circumflex scapular
artery and vein emerging from the intermus-
cular triangle are identified by palpation of the
bone and placement of the index finger into
the axilla in order to feel the boundaries of the
intermuscular triangle (yellow arrow).
B: The para scapular flap is outlined on the skin
in green, with the superior component of that
flap overlying the intermuscular triangle in order
to capture the nutrient blood supply to the skin. B

FIGURE 19-38. The tip of the parascapularflap

is incised and elevated up to the upper border
of the teres major muscle. Early in the dissec-
tion, the diastasis between the lower border of
the teres major muscle and the upper border of
the latissimus dorsi muscle is identified (yellow
arrow).
314 CHAPTER 19

The Scapular lip Based on the Angular Branch and the Parascapular Flap

FIGURE 19-39. The "vascular highway.. (green

arrow) into the axilla is established by dissect-
ing along the upper border of the teres major
muscle and then the plane deep to the teres is
established in order to transect those muscle
fibers with a small cuff left attached to the
lateral border of the scapula (dotted line).

FIGURE 19-40. A stapling device is placed

across the teres major in order to transect the
muscle and achieve hemostasis.

FIGURE 19-41. The teres major muscle has

been cut (yellow arrow) and will be retracted
upward in order to gain exposure for further
dissection of the circumflex scapular, thora-
codorsal and angular vessels.
 SCAPULAR AND PARASCAPULAR FASCIOCUTANEOUS AND OSTEOFASCIOCUTANEOUS AND SUBSCAPULAR MEGA FLAP 315

The Scapular Tip Based on the Angular Branch and the Parascapular Flap

FIGURE 19-42. The teres major muscle has

been reflected in the axilla and the neurovas-
cular bundle to that muscle has been isolated
(white arrow). The distal portion of the thora-
codorsal vessels (green srrowt has been identi-
fied at the point just proximal to its entry into
the hilum of the latissimus dorsi muscle.

FIGURE 19-43. The full course of the thora-

codorsal vessels (yellow srrowt is identified
and the branch to the serratus anterior muscle
(green srrowt is demonstrated. The angular
branch is elevated by a forceps.

FIGURE 19-44. The cephalad portion of the

skin paddle has been elevated off the deltoid
and teres minor muscles (white arrow) in order
to visualize the upper portion of the intermus-
cular triangle and to further isolate the circum-
flex scapular vessels. The angular branch is
dissected from its origin from the thoracodorsal
artery (green srrowt. although in some patients
that branch may originate from the branch to
the serratus anterior muscle.
316 CHAPTER 19

The Scapular lip Based on the Angular Branch and the Parascapular Flap

FIGURE 19-45. The anatomy of the subscapu-

lar system is clearly demonstrated from the
subscapular vessels Iyellow a"ow) to the
serratus anterior (green arrow) and the angular
branches (black arrow).

FIGURE 194. Isolation of the parascapular

flap on the circumflex scapular artery and vein
(yellow arrow) requires that the fine branches
to the periosteum of the lateral scapular border
(black arrow) must be ligated and transacted.
This is a very delicate dissection and requires
great care to prevent both bleeding and injury to
the fasciocutaneous branches supplying the skin.

FIGURE 19-47. A clase-up view of the vascular

anatomy of the periostea I branches {yellow
arrow) from the circumflex scapular artery
and vein is demonstrated. After the takeoff of
the periosteal feeders, the distal portion ofthe
circumflex vessels to the skin is shown (black
arrow).
 SCAPULAR AND PARASCAPULAR FASCIOCUTANEOUS AND OSTEOFASCIOCUTANEOUS AND SUBSCAPULAR MEGA FLAP 317

The Scapular Tip Based on the Angular Branch and the Parascapular Flap

FIGURE 19-48. The detailed vascular anatomy

in 1he axilla is demonstrated in close-up view
wi1h the subscapular artery (red arrow) and the
subscapular vein (blue arrow) as well as the
circumflex vessels (green arrow), which are
shown at the top of this photo because of the
skin being positioned superiorly. The proximal
branches of the thoracodorsal vessels (blsck
arrow) and the nerve (white arro'IA are shown.

FIGURE 19-49. The portion of1he scapular

bone to be harvested is demonstrated. The
dotted line indicates the point where the oste-
otomy is to be performed. The angular branch is
isolated and elevated by a clamp.

FIGURE 19-50. The osteotomy (yellow arrow)

has been performed and remaining soft tissue
attachments must be cut in order to pedicle the
bone flap on 1he angular branch.
318 CHAPTER 19

The Scapular lip Based on the Angular Branch and the Parascapular Flap

FIGURE 19-51. A: Following the osteotomy and

mobilization ofthe bone lblue amJw), the distal
portion of the thoracadorsa I vessels (green
arrow) and the branch to the serratus anterior
muscle (yellow srrow) must be ligated and
transected. B: Ligation and transection of the
branch to the serratus anterior. C: Demonstra-
tion of the ligation and transection of the distal
thora codorsal vessels prior to entry into the
latissimus darsi muscle. c
 SCAPULAR AND PARASCAPULAR FASCIOCUTANEOUS AND OSTEOFASCIOCUTANEOUS AND SUBSCAPULAR MEGA FLAP 319

The Scapular Tip Based on the Angular Branch and the Parascapular Flap

FIGURE 19-52. The scapular tip and 1he

parascapular fasciocutaneous flap have been
harvested. The branch to the serratus is dem-
onstrated along wi1h the subscapular artery
(red pin), subscapular vein (blue pin) and the
thoracodorsal nerve (yellow pin).
320 CHAPTER 19

Inset of Scapular Osteocutaneous Flap-Latissimus Dorsi Flap

for Mandibular Reconstruction

FIGURE 19-53. The lateral border of the

scapula has been harvested with the parascap-
ular skin and the latissimus dorsi muscle based
on the subscapular artery and vein. The angular
branch (yellow arrow) has been harvested
to augment the blood supply to the tip of the
lateral border ofthe scapula. The hemimandible
has been harvested to show the extent of the
segmental defect.

FIGURE 19-54. The scapular bone has been

inset into the mandibular defect and fixed into
position. The skin paddle will be used tn line
the oral cavity (yellow arrow) and the latissi-
mus dorsi muscle will be placed in the neck to
provide coverage. The subscapular pedicle is
shown extending inferiorly from the neoman-
dible (blue arrow).
 SCAPULAR AND PARASCAPULAR FASCIOCUTANEOUS AND OSTEOFASCIOCUTANEOUS AND SUBSCAPULAR MEGA FLAP 321

Inset of Scapular Osteocutaneous Flap-Latissimus Dorsi Flap

for Mandibular Reconstruction

FIGURE 19-55. The skin paddle has been

placed over the neomandible for lining of the
oral cavity.

FIGURE 19-56. The latissimus dorsi (green

srrow) provides a valuable coverageforthe
vessels in the neck especially for clinical sce-
narios where the patient has been radiated and
the surgeon is concerned about the risk of both
cervical skin breakdown as well as salivary
fistula. Coverage of the carotid artery and the
microvascular pedicle is extremely helpful in
these situations.
322 CHAPTER 19

Reconstruction of the Hemipalatal Shelf with a Half of the Scapular Tip

FIGURE 19-51. The scapular tip can be divided

in half and used to reconstruct one half ofthe
palatal shelf. The length of the vascular pedicle
(white arrawt using the angular branch is
demonstrated. This pedicle is longer than the
conventional sea pula r flap with harvest of the
lateral scapular border and use of the circum-
flex scapular vessels.

FIGURE 19-51. A view of the palataI recon-

struction from below provides a glimpse of the
bony reconstruction of the hemipalate_ The skin
(white arrawt can be transposed into the oral
cavity to provide coverage of the neopa late.
 SCAPULAR AND PARASCAPULAR FASCIOCUTANEOUS AND OSTEOFASCIOCUTANEOUS AND SUBSCAPULAR MEGA FLAP 323

Reconsh'uction of the Total Palatal Defect with the Scapular Tip Osteocutaneous Flap
Based on the Angular Artery and Vein

FIGURE 19-59. A total palatal defect is shown.

The superstructure of the maxillae remains
intact. These defects are very difficultto reha-
bilitate with a prosthetic obturator due to the
inability to achieve retention and stability. The
reconstruction of this defect with soft tissue
alone does not provide support for the upper lip,
and there is no chance for restoring dentition.

FIGURE 19-al. The similarity in shape and size

of the total palate and the tip of the scapula are
shown. This similarity can be utilized to recon-
struct the entire palatal shelf with bone from
the scapular tip. This type of reconstruction is
better served by transfer of a fibular flap due to
the better quality of bone from that donor site. It
is rare thatthe scapular tip will accommodate
dental implants, and therefore transfer of this
composite flap provides bone and soft tissue for
support of the upper lip and separation of the
mouth from the sinonasal cavity.
324 CHAPTER 19

Reconstruction of the Total Palatal Defect with the Scapular Tip Osteocutaneous Flap
Based on the Angular Artery and Vein

FIGURE 19-&1. The scapular tip has been

placed into the defect and fixed with plates to
the bodies of the zygoma. The bone fills this
defect well and the vascular pedicle is long
enough to be placed into the 11eck under the
cheek skin {yellow arrow!.

FIGURE 19-&2. The parascapular skin paddle

can be transposed into the mouth for resurfac-
ing ofthe oral side of the neopalate.
 SCAPULAR AND PARASCAPULAR FASCIOCUTANEOUS AND OSTEOFASCIOCUTANEOUS AND SUBSCAPULAR MEGA FLAP
REFERENCES
1. Bartlett SP, May Jw.Yaremchuk MJ: The latissimus dorsi
muscle. A fresh cadaver study of the primary neurovascu-
lar pedicle. Plast Reconstr Surg 1981;67:631.
2. Barwick W, Goodkind D, Serafin D: The free scapular
flap. Plast Reconm Surg 1982;69: 779.
3. Batchelor A, Bardsley A: The hi-scapular flap. Br J Plast
Surg 1987;40:510.
4 . Chandrasekhar B, Lorant J, Terz J: Parascapular free
flaps for head and neck reconstruction. Am J Surg
1990;160:450.
5. Chiu D, Sherman J, Edgerton B: Coverage of the calvarium
with a freeparascapular flap. Ann PlasrSurg 1984;12:60.
6. Coleman J, Sultan M: The bipedicled osteocutaneous
scapula flap: a new subscapular system free flap. Plast
Reconstr Surg 1991;87:682.
7. Cormack G, Lamberty B: The anatomical vascular basis
of the axillary fasciocutaneous pedicled flap. Br J Plast
Surg 1983;36:425.
8. DeGaris C, Swartley W: The axillary artery in white and
Negro stocks. Am J Anat 1928;41 :353.
9. Deraemaecher R, Thienen CV, Lejour M, Dor P: The
serratus anterior-scapular free flaps: a new osteomuscular
unit for reconstruction after radical head and neck sur-
gery (abstract). In: Proceedings of the Second Interna-
tional Conference on Head and Neck Cancer. 1988.
10. dos Santos lF: Retalho escapular: urn novo retalho livre
microcirurgico. Rev Bras Cir 1980;70:133.
11. dos Santos LF: The vascular anatomy and dissection of
the free scapular flap. Plast Reconstr Surg 1984;73:599.
12. Gabhos F, Tross R, Salomon J: Scapular free flap dissec-
tion made easier. Plast Reconstr Surg 1985;75:115.
13. Gilbert A, Teot L: The free scapular flap. Plast Reccmstr
Surg 1982;69:601.
14. Hamiton S, MorrisonW: The scapular free flap. BrJ Plast
Surg 1982;35:2.
15. Hitzrot J: A composite study of axillary artery in man.
Johns Hopkins Bull1901;12:136.
16. Kim P, Gottlieb J, Harris G, Nagle D, Lewis V: The dor-
sal thoracic fascia: anatomic significance with clinical
applications in reconstructive microsurgery. Plast Reccmstr
Surg 1987;79:72.
325
17. Moscoso J, Keller J, Genden E, et al.: Vascularized bone
flaps in oromandibular reconstruction: a comparative
anatomic study of bone stock from various donor sites
to assess suitability for enosseous dental implants. Arch
Otolaryngol Head Neck Surg 1994;120:36.
18. Nassif TM,Vidal L, Bovet JL, Baudet J: The parascapular
flap: a new cutaneous microsurgical free flap. Plast Recon-
str Surg 1982;69:591.
19. Rowsell A, Davies M, Eisenberg N, Taylor GI: The
anatomy of the subscapular thoracodorsal arterial sys-
tem: study of 100 cadaver dissections. Br J Plast Surg
1984;37:374.
20. Saijo M: The vascular territories of the dorsal trunk: a
reappraisal for potential flap donor sites. Br J Plast Surg
1978;31:200.
21. Sullivan M, Baker S, Crompton R, Smith-Wheelock
M: Free scapular osteocutaneous flap fur mandibu-
lar reconstruction. Arch Otolaryngol Head Neck Surg
1989;115:1534.
22. Sullivan MJ, Carroll WR, Baker SR: The cutaneous sea~
ular free flap in head and neck reconstruction. Arch Oto-
laryngol Head Neck Surg 1990;116:600.
23. Swartz W, Banis J, Newton D, Ramasastry S, Jones N,
Acland R: The osteocutaneous scapular flap for man-
dibular and maxillary reconstruction. Plast Reconstr Surg
1986;77:530.
24. Thylor GI, Palmer J:The vascular territories (angiosomes)
of the body: experimental study and clinical applications.
Br J Plasz Surg 1987;40:113.
25. Teot L, Bosse JP, Moufarrege R, Papillon J, Beuregard
G: The scapular crest pedicled bone graft. lnt J Microsurg
1981;3:257.
26. Thoma A, Archibald I, Payk I, Young J: The free medial
scapular osteofasciocutaneous flap for head and neck
reconstruction. Br J Plast Surg 1991;44:477.
27. Thoma A, Riddle S:The extended free scapular flap. Br J
Plast Surg 1990;43:712.
28. Upton J, Albin R, Mulliken J, Murray J:The use of scapu-
lar and parascapular flaps for cheek reconstruction. Plast
Reconstr Surg 1992;90 :959.
29. Urbaniak J, Komar A, Goldman R, Armstrong N,
Nunley J: The vascularized cutaneous scapular flap. Plast
Reconstr Surg 1982;69 :772.
 erfused by the terminal branches of the thoracodQl'-
P sal artery, the latissimus dorsi and serratus anterior
Baps have a well-established role in reconstructive sur-
gery. The latissimus dorsi Bap was the first musculocu-
taneous Bap descnoed in the medicalliterature.1ltnsini
(79) reported this technique for chest wall reconstruc-
tion following radical mastectomy in 1896. There fol-
lowed a number of other publications that repQl'ted
using the latissimus dorsi musculocutaneous flap for
primary reconstruction of the postmastectomy defect
and the prevention oflymphedema in the ipsilateral arm
(13,30). This technique remained buried in the medi-
cal literature until the 1970s when it was resurrected
by Olivari (56,57) for chest wall reconstruction. More
extensive series by Bostwick et al. (7) and Maxwell et al.
(50) demonstrated the safety and the reliability of this
reconstructive technique.
326
Quillen et al. (62) are credited with being the first to
use the pedicled latissimus dorsi musculocutaneous flap
for head and neck reconstruction in 1978. Subsequent
reports by Quillen (61) and Barton et al. (5) established
the latissimus dQl'Si musculocutaneous flap as a front-
line reconstructive technique for head and neck defects.
In 1979, Watson et al. (87) reported the first success-
ful microvascular transfer of a free latissimus flap. The
length and cahoer of the neurovascular pedicle, the ease
of dissection, the large surface area, and the minimal
donQl' site morbidity are the majQl' factors that explain
the popularity of this donor site for the transfer of tissue
as a pedicled or free Bap to the head and neck region.
The first English language clinical description of
serratus anteriQl' flaps was reported by 'Illbyanagi and
Tsukie (77) in 1982, who reported two successful cases
of lower extremity reconstruction using the serratus

anterior myofascial or musculocutaneous free flaps.
That same year, Harii et al. (25) reported two cases of
head and neck reconstruction with the combined ser-
ratus anterior-latissimus dorsi musculocutaneous free
flaps that were used for reconstruction of through-
and-through defects of the buccal mucosa and cheek
skin. In addition to the applications for head and neck
reconstruction discussed in the next section of this
chapter, the serratus anterior flap was applied to the
reconstruction of the upper extremities (42), the lower
extremities (54), and the trunk (1) during the 1980s.
The latissimus dorsi muscle is a broad fiat muscle
that covers a large portion of the lower back. It arises
from the spinous processes of the lower six thoracic
vertebrae and from the thoracolumbar fascia, which
attaches to the lumbar and sacral vertebrae (Fig. 20-1).
Laterally, it arises from the fascia that is attached to the
iliac crest. The latissimus dorsi muscle also arises from
the lower four ribs, where its fibers coalesce with those
of the external oblique muscle. In its upper medial por-
tion, the latissimus dorsi muscle is overlapped by the
caudal fibers of the trapezius muscle. As it passes later-
ally and cephalad, the free edge of the latissimus dorsi
muscle overlies the tip of the scapula, from which it may
also take a minor origin. The converging fibers of the
latissimus dorsi muscle spiral around the caudal border
of the teres major muscle to insert on the medial sur-
face of the humerus. Along with the teres major mus-
cle, the tendon of the latissimus dorsi muscle forms the
posterior axillary fold and these two muscles insert in
close proximity along the intertubercular groove of the
humerus.
The major actions of the latissimus dorsi muscle are
to adduct, inwardly rotate, and extend the arm. These
actions are most easily visualized in the completion of
the arm motion performed in the free-style swimming
stroke. Through its attachments to the caudal tip of the
scapula, the latissimus dorsi muscle also stabilizes the
scapula and prevents its superior and lateral displace-
ment. This action is most easily visualized in the climb-
ing motion when the arms are raised above the head
and the trunk is pulled upward and forward. Following
transfer of a latissimus dorsi flap, its major muscular
actions are said to be well compensated by the other
muscles that act across the glenohumeral joint. How-
ever, progressive morbidity to the shoulder occurs when
sacrifice of the latissimus dorsi muscle is combined
with loss of either the trapezius muscle, following radi-
cal neck dissection, or the pectoralis major muscle. This
issue is discussed in detail later in this chapter.
The serratus anterior muscle is located on the lateral
thoracic wall (Fig. 20-2). Muscular slips of the serratus
anterior muscle arise from the anterolateral surface of
the first 8 to 10 ribs and insert into the anterior aspect
of the scapula along the entire length of the medial bor-
der of the scapula. Based upon its blood supply and
LATISSIMUS DORSI AND SERRATUS ANTERIOR 327
common applications in reconstructive surgery, the ser-
ratus anterior muscle can be divided into a superior seg-
ment and an inferior segment. The upper portion of the
serratus anterior muscle consists of five slips of muscle
that are perfused by the lateral thoracic artery, which
is a branch of the proximal portion of the second part
of the axillary artery. The lower portion of the serratus
anterior muscle consists of three to five slips of muscle
that are perfused by one or more branches of the thora-
codorsal anery. Through its attachments to the anterior
surface of the medial border of the scapula, the serratus
anterior muscle is largely responsible for protraction of
the scapula, or pulling the scapula forward against the
rib cage. Loss of serratus anterior muscle function will
therefore result in winging of the scapula and difficulty
in anterior projection and raising of the arm. To avoid
this potential donor site morbidity, harvest of serratus
anterior flaps is usually limited to the inferior portion of
the serratus anterior muscle, thereby preserving func-
tion in the upper portion of the serratus anterior muscle.
FLAP DESIGN AND UTILIZATION
Latissimus Dorsi Flap
The latissimus dorsi flap may be used as either a free or
pedicled flap. As a pedicled flap, one can reach virtu-
ally any site on the head and neck by passing the flap
through the axilla between the pectoralis minor and
major muscles. The arc of rotation may be improved by
exteriorizing the pedicle, which allows the flap to reach
most areas of the scalp (48). Excision of the intervening
muscle is performed following a delay of several weeks.
The arc of rotation of the pedicled latissimus dorsi flap
is also enhanced by several maneuvers. Transection of
the arterial and venous branches to the serratus anterior
muscle prevents tethering of the thoracodorsal pedi-
cle. Friedrich et al. (17) advised that additional length
could also be achieved through division of the circum-
flex scapular branch. However, others caution that
preservation of this branch helps to prevent kinking of
the pedicle as it traverses the axilla from posteriorly to
anteriorly (5,48). Barton et al. (5) also speculated that
the intact circumflex scapular vessels provided collateral
flow to the latissimus dorsi. Transection of the tendon
of the latissimus dorsi muscle provides a considerable
amount of additional freedom and markedly improves
the arc of rotation. Designing the skin paddle over the
caudal portion of the muscle also helps to improve the
flap's reach. However, the density of the perforators in
this region is diminished, and therefore, the blood sup-
ply to the skin is more tenuous. Quillen (61) reported a
delay procedure, to enhance the reliability of the distal
muscle and skin that involved a staged division of the
intercostal perforators that run from the chest wall to
328 CHAPTER 20

FIGURE 20-1. The latissimus dorsi muscle covers the entire lower back from the iliac crest to
the lower border of the scapula. It arises from the lower six thoracic vertebrae, the thoracolum-
bar fascia, and the iliac crest. Its free upper border overlaps the inferior angle of the scapula.lt
forms the posterior axillary fold and then inserts on the medial surface of the humerus.
 LATISSIMUS DORSI AND SERRATUS ANTERIOR 329

~:--~+-+f-- Serratus
anterior mus.

FIGURE 20-2. The serratus anterior muscle arises from the first 8to 10 ribs in the region of the
anterior axillary line and inserts into the anterior aspect of the medial border of the scapula
330 CHAPTER ZO
the deep surface of the muscle approximately 1 week
prior to flap transfer.
The dimensions of the total area covered by the latis-
simus dorsi muscle are in the range of 25 to 40 em. The
maximum size of the skin paddle that can be trans-
ferred is limited by the patient's body habitus and the
surgeon's ability to achieve primary closure. Although
less satisfactory, donor-site closure with a skin graft has
been successfully accomplished (57). The reliability of
skin paddles designed over different regions of the ter-
ritory of the latissimus dorsi muscle varies and is dis-
cussed later in this chapter. Transfer of the latissimus
dorsi muscle with a skin graft provides an alternative
solution to the problem of covering massive defects of
the head and neck while still achieving primary closure.
This technique has been used extensively in the recon-
struction of the scalp (20).
The latissimus dorsi muscle is one of the thinnest
muscles in the body. Denervation atrophy produces an
even thinner flap. Primary "thinning" of this muscle has
been reported in the reconstruction of the forehead.
The location of the vascular pedicle on the deep sur-
face of the muscle permits the superficial muscle layers
to be removed prior to coverage with a skin graft (66).
Hayashi and Maruyama (28) described the "reduced"
latissimus dorsi musculocutaneous flap in which a
proximal skin paddle overlying the muscle is transferred
with a "reduced" distal fasciocutaneous unit. The blood
supply to the fasciocutaneous segments, which meas-
ured up to 12cm in length, was derived from the fas-
cial, subcutaneous, and subdermal vascular plexuses.
The authors also transferred cutaneous extension flaps
from the region anterior to the border of the muscle.
The concept of a reduced flap is an outgrowth of the
observation reported by Barton et al. (5) that, although
skin islands overlying the distal muscle were unreliable,
this territory could be successfully transferred when
designed as an extension of a superior skin island. This
phenomenon is presumably the result of the capture of
the more plentiful musculocutaneous perforators over-
lying the proximal muscle.
The skin overlying the thoracolumbar aponeurosis is
notoriously unreliable. However, this fascial sheath may
be used in head and neck reconstruction for a number of
purposes. Smith et al. (75) described resurfacing a com-
posite defect of the scalp and skull in which the thora-
columbar fascia was used to patch a dural defect and the
musculocutaneous unit was used to replace the scalp.
Harii et al. (23) introduced a further modification
of this donor site in an effort to increase the surface
area and the reach of the transferred skin territory. They
described a combined pedicled musculocutaneous and
microvascular free flap in which the latissimus dorsi
flap was combined with a conventional groin flap. The
blood supply to the upper portion of this skin was from
the thoracodorsal vessels; the groin skin was perfused
by either the superficial inferior epigastric vessels or the
superficial circumflex iliac vessels. When rotated as a
superiorly based flap, the thoracodorsal vessels were left
attached, and the blood supply to the groin flap compo-
nent was anastomosed to recipient vessels in the head
and neck. Alternatively, the inferiorly based compound
flap could be left attached to the vessels supplying the
groin flap and the thoracodorsal pedicle, anastomosed
to recipient vessels in the lower extremity (Fig. 20-3).
Complex defects of the head and neck region often
require two epithelial surfaces to repair the inner
mucosal lining and the overlying skin. This can be
achieved by folding the latissimus dorsi skin paddle
and de-epithelializing the intervening bridge of skin or
dividing the subcutaneous tissue to the muscle layer
(48). This, however, may compromise the vascularity to
the tip of the flap. An alternative solution to this prob-
lem was described byTobin et al. (82,83) who designed
two separate musculocutaneous units based on the
transverse and the longitudinal intramuscular branches
of the thoracodorsal vascular system. The feasibility of
splitting the latissimus dorsi was investigated in dogs
and then subsequently applied to humans (Fig. 20-4).
Another desirable feature of a donor site for head and
neck reconstruction is the ability to incorporate vascu-
larized bone and, therefore, design a composite flap that
would permit restoration of the calvarium or the maxil-
lomandibular skeleton. Investigative studies performed
by Schlenker et al. (69) in dogs suggested that a poste-
rior segment of the rib could be vascularized through
the latissimus dorsi muscle. The anatomic basis for this
composite flap is the perforating branches of the poste-
rior intercostal artery, which traverse the 5th through
the lOth intercostal spaces along the posterior axillary
line. Microopaque injections of the subscapular vessels
revealed retrograde filling of the posterior intercostal
perforators. The blood supply to the bone was then
hypothesized to occur through the nutrient medullary
branch of the posterior intercostal artery. Tetracycline-
labeling studies were performed in the dog model and
confirmed the bone's vascularity. However, injection
studies in fresh human cadavers performed by Friedrich
et al. (17) led to the conclusion that the latissimus dorsi
branch of the thoracodorsal system did not provide
a blood supply to the rib. It is unclear from the latter
report whether the authors investigated the possibility of
capturing the intercostal vascular system through perfo-
rating branches, as described by Schlenker et al. (69).
Despite these conflicting views, there are two reports
in the literature of successful transfer of the latissimus
dorsi-rib composite flap for oromandibular reconstruc-
tion. Schmidt and Robson (70) reported three success-
ful cases and described the transfer of the 7th, 8th, or
9th rib, depending on which intercostal space had the
dominant perforator. They transferred this compos-
ite flap as a microvascular procedure, in contrast to

FIGURE Z0-3. Harii et al. (15) described the combined
latissimus dorsi and groin flaps transferred in series with
one vascularized as a free flap and the other as a pedicled
flap. For use in the head and neck, the thoracodorsal ves-
sels remained intact the vessels supplying the groin flap
were interrupted and anastomosed to recipient vessels in
the neck.
LATISSIMUS DORSI AND SERRATUS ANTERIOR 331
the report of Maruyama et al. (46) who performed a
transfer of a pedicled composite flap. Primary wound
healing, radiographic evidence of bone union, and early
postoperative hone scans were used as evidence of the
viability of the osseous component.
Increased concern for ma:ximizing the functional
recovery in head and neck reconsttuction has height-
ened the desirability of restoring sensation to the oral
cavity and pharym: through the uae of sensate free flaps.
There are two reports of restoring sensation to the latis-
simus dorsi musculocutaneous fl.ap. Dahh and Co.nklin
(1 0) described the sensory supply to the skin of the lower
back as arising from two sources: the cutaneoua branches
of the d()l'Sa) rami and the posterior branches of the lat-
eral cutaneous branch of the intercostal nerve. They
reported the successful restoration of sensation follow-
ing reconstruction of the foot. The posterior branches
of the lateral cutaneous nerves were anastomosed to the
medial plantar branch of the posterior dbial nerve. In
addition, the dorsal rami were anastomosed to the ante-
rior uoial nerve. The detection of pressure and pain were
reported at 6 months. Gordon et al. (20) described an
alternative route to provide sensory nerve ingrowth into
the latissimus dorsi ftap. They anastomosed the thoraco-
dorsal nerve to a recipient sensory nerve and noted that
five of seven patients were able to distinguish sharp from
dull stimuli. The mechanism for the recovery of sensa-
tion in these patients was uncertain, not only in light of
using a motor nerve hut also because the latissimus dorsi
muscle ftaps were covered with skin grafts.
The latissimus dorsi has been uaed to restore dynamic
activity by either preserving the integrity of the thoraco-
dorsal nerve or anastomosing it to an appropriate recipi-
ent motor nerve. Zancolli and Mitre (93) reported the
restoration of elbow Be:zion in patients with poliomyeli-
tis or traumatic loss of such function by transferring the
latissimua dorsi muscle so that it crossed the elbow joint.
Dynamic facial reanimation was reported by the
transfer of the latissimus dorsi muscle to the para-
lyzed face in a two-stage procedure (44). The first stage
involves placement of a cross facial nerve graft that is
anastomosed to buccal branches on the noni:avolved
side. By eliciting theTinel sign, the advancing nerve fib-
ers can be followed until they reach the mid portion of
the upper lip, and then a free muscle ftap transferred
and both revascularized and reneurotized. Various dif-
ferent muscles have been used for this purpose, includ-
ing the gracilis muscle (24), the extensor digitorum
brevis muscle (52), the pectoralis minor muscle (81),
and the serratus anterior muscle (90). The advantages
of the latissimus dorsi for this procedure are the length
and caliber of the neurovascular pedicle and its divi-
sion into two segments, which allows the transfer of two
separate muscle units. One segment has been used for
reanimation of the mouth and the other for the lower
eyelid. Mackinnon and Dellon (44) outlined a number
332 CHAPTER 20

FIGURE 20-4. Two skin paddles may be harvested overlying the latissimus dorsi based on the
transverse and the vertical branches of the thoracodorsal artery. By dissecting on the und ersur-
fa ce of the latissimus dorsi, the surgeon can identify the division of the thora codorsal vessels
and, therefore, completely separate these two musculocutaneous paddles.

of subtle changes in this procedure that include mark-
ing of the resting muscle length at the donor site and
then precisely reestablishing that resting tension dur-
ing insetting of the flap into the face. In addition, they
described a distal dissection of the thoracodorsal nerve
so that nerve fibers that extend beyond the muscle seg-
ment can be reinserted into the muscle to help increase
the dynamic activity.
The latissimus dorsi musculocutaneous flap has
also been used for dynamic reconstruction of compos-
ite cheek defects that include the loss of the mimetic
muscles of the midface. In this situation, a one-stage
procedure can be performed whereby the musculocu-
taneous unit is transferred and revascularized and the
thoracodorsal nerve is sutured to the ipsilateral facial
nerve. Dynamic activity and electromyographic record-
ings confirmed the muscle's contraction (45).
The other area in which a dynamic reconstruction is
desirable is in restoring function to the mouth follow-
ing significant or total glossectomy. The complexity of
the tongue's musculature, which is composed of both
extrinsic and intrinsic muscle fibers, has posed a daunt-
ing problem for surgeons over the last several decades.
Although it is tempting to try to duplicate the success of
free muscle transfer in facial paralysis, the motions of the
tongue are far more complex. In considering this prob-
lem, it is difficult to decide which of the almost limitless
changes in position and shape of the native tongue to try
to restore with a free, reneurotized muscle flap composed
of unidirectional fibers. Haughey (26) and Haughey and
Fredrickson (27) described the use of the reinnervated
latissimus dorsi musculocutaneous flap for total tongue
reconstruction. The muscle fibers were oriented trans-
versely to the long axis of the mouth and sutured to the
pterygoid, constrictor, and masseter muscles and ten-
dons, depending on which were available. The thora-
codorsal nerves were anastomosed to the stumps of the
hypoglossal nerve. In this reconstruction, the upward
movement of the flap that was observed by the authors
was caused by either the contraction of the latissimus
dorsi or the muscles to which the flap was sutured.
The pedicled latissimus dorsi musculocutaneous flap
has been utilized for a number of reconstructive prob-
lems, including repair of mucosal defects of the pharynx
and oral cavity and cutaneous defects of the neck and
face (5,48,61). Watson and Lendrum (88) reported a
one-stage, tubed latissimus dorsi musculocutaneous flap
for circumferential pharyngoesophageal reconstruction.
Watson et al. (89) described several technical considera-
tions in using the latissimus dorsi flap for pharyngeal
defects. The authors advised transferring a large seg-
ment of muscle around the circumference of the skin
paddle that is to be tubed. The muscle was sutured to
the surrounding tissues to provide a second-layer seal.
The large surface area of this donor site has made it
particularly suitable for resurfacing very large defects of
LATISSIMUS DORSI AND SERRATUS ANTERIOR 333
the face and neck (36,71). Composite defects involv-
ing both mucosa and skin can be readily restored by
the methods described previously. However, a word of
caution is needed regarding the weight of this flap, espe-
cially when it is folded on itself for coverage of through-
and-through defects. Secondary debulking procedures
are often needed (86). The latissimus dorsi flap easily
reaches the midline of the lower neck and, therefore,
is an excellent alternative donor site to the pectoralis
major for defects resulting from resections for stomal
recurrent cancer (16).
The free latissimus dorsi musculocutaneous flap has
been used for a variety of midface defects that require
bulk and freedom to place epithelial surfaces in a num-
ber of different three-dimensional planes. Baker (2)
reported the suitability of this flap for extensive orbito-
maxillary defects. This flap has also been used for clo-
sure of palatal and midfacial cutaneous defects. Shestak
et al. (74) described using two skin paddles that were
produced by de-epithelializing an intervening strip of
skin that permitted their placement in two different
planes oriented at 90 degrees to each other. Several
reports in the literature demonstrate the efficacy of the
latissimus dorsi flap for highly complex and extensive
defects of the midface and skull base (22,34).
Scalp defects located in the temporal or occipital
regions are readily accessible by a pedicled latissimus
dorsi flap. However, when the area to be resurfaced
enlarges and/or extends to the vertex, then a free flap
is often required. The latissimus dorsi is an excellent
choice for this problem because of the surface area of
either skin or muscle that is available, the length of the
nutrient pedicle (which easily extends to recipient ves-
sels in the neck), and the thinness of the tissue (which
matches that of the remaining scalp). The number of
reports using this flap for this purpose is testimony to
these distinguishing characteristics (3,15,59,64,75, 76).
Pennington et al. (59) noted their preference for resur-
facing the convex contour of the skull with vascularized
muscle and a skin graft because of the intrinsic ability
of the muscle to stretch and therefore to avoid dog ears,
which often occur with skin flaps. In addition, vascu-
larized muscle flaps eliminate the concern for primary
donor site closure when large skin paddles are harvested.
As noted previously, the thoracolumbar fascia has been
used to repair dural defects (75). The efficacy of using
the latissimus dorsi flap to achieve a stable wound fol-
lowing debridement of scalp and calvarium for osteora-
dionecrosis has been reported by Robson et al. (64) in a
series of six cases. In one patient, these authors reported
the transfer of a latissimus dorsi osteomusculocutane-
ous flap containing a vascularized segment of the 4th rib
to replace a portion of the calvarial defect. The merits of
restoring structural support along with soft tissue cover-
age vary with the size and location of the skull defect.
Although protection of the brain and improved cosmesis
334 CHAPTER ZO
are the most frequent indications for cranioplasty, some
authors espouse the merits of restoring the calvarium to
prevent or correct a variety of symptoms including pain,
headache, dizziness, and posttraumatic seizures (76). In
addition to vascularized bone, this problem has been
solved by using split rib grafts, synthetic cranioplasty,
and titanium plates (62, 75,76).
Serratus Anterior Flap
Like the latissimus dorsi flap, the serratus anterior flap
can be used as either a pedicled flap or as a free flap.
Inoue et al. (3I) reponed a series of II patients who
underwent head and neck reconstruction using pedicled
serratus anterior musculocutaneous flaps. The authors
advocated tunneling the flap into the head and neck
using a subclavicular path in order to provide the short-
est possible route for passage of the vascular pedicle
and to minimize the risk of compression of the vascular
pedicle if it were draped over the clavicle. Using this
technique, the arc of rotation of the pedicled serratus
anterior flap reached to the temporal scalp. However,
Roswell et al. (65) reported unfavorable outcomes after
pedicled, as opposed to free, serratus anterior muscle
flaps used for reconstruction of hemifacial microsomia.
Two of the three pedicled flaps failed, and better out-
comes were seen in patients who underwent serratus
anterior free flaps, leading the authors to recommend
against the use of pedicled serratus anterior flaps for
head and neck reconstruction. Likewise, Richards et al.
(63) reponed a series of seven cases of oromandibu-
lar reconstruction using pedicled or free serratus ante-
rior-rib flaps. One case of pedicled flap reconstruction
required an urgent revision surgery for conversion to
a free flap to relieve compression of the flap,s vascu-
lar pedicle, and an increased incidence of postoperative
complications was noted in the cases of pedicled flap
reconstructions. They recommended that free flaps be
performed rather than pedicled flaps when microsurgi-
cal facilities are available in order to improve the relia-
bility of the reconstruction and to shorten the operating
time required to tunnel the flap from the axilla to the
head and neck. Whitney et al. (90) established the reli-
ability of free serratus anterior flaps in I990, reporting a
99% success rate in IOO consecutive cases.
A variety of tissue components are available for har-
vest based on the serratus anterior branch of the thora-
codorsal artery. As the serratus anterior branch of the
thoracodorsal artery enters into the superficial aspect
of the serratus anterior muscle, a very thin fascial flap
that incorporates the superficial investing fascia of the
serratus anterior muscle can be harvested while spar-
ing the serratus anterior muscle (85). Alternatively, a
myofascial flap that incorporates the lower slips of the
serratus anterior muscle with its investing fascia pro-
vides a thin muscle flap that has an average thickness
of about 7mm (40). A musculocutaneous flap can be
harvested that incorporates a skin paddle that is limited
superiorly by the inframammary crease, posteriorly by a
point that is about 2 em posterior to the anterior border
of the latissimus dorsi muscle, inferiorly by the 9th rib,
and anteriorly by a point that is about 2 em medial to
the midclavicular line. The skin paddle may be extended
anteriorly beyond the anterior border of the serratus
anterior muscle by including a fasciocutaneous exten-
sion that includes the investing fascia of the external
oblique muscle (60). A serratus anterior-rib osteomus-
cular flap can be harvested that includes up to three seg-
ments of vascularized rib (73). Finally, serratus anterior
flaps may be combined with a multitude of other flaps
that are perfused by the subscapular vascular system as
a component of a subscapular system "megaflap."
There are three major applications of serratus ante-
rior flaps for the reconstruction of defects in the head
and neck: (a) serratus anterior fascial, myofascial, and
musculocutaneous flaps are useful for soft tissue recon-
struction, including reconstruction of defects involving
the scalp and skull base, cheek and midface, oral cav-
ity, neck, and pharyngoesophageal segment; (b) rein-
nervated serratus anterior myofascial flaps are useful for
facial reanimation; and (c) serratus anterior-rib osteo-
muscular flaps are useful sources of vascularized bone
reconstruction of the mandible, midface, and calvarium.
The most commonly reponed head and neck appli-
cations of soft tissue serratus anterior free flaps have
been for facial reanimation and restoration of facial
contour (90). The ultrathin serratus anterior fascial flap
is useful for cases of oral cavity and midface reconstruc-
tion, where it can be used to wrap bone grafts, restore
deficient intranasal lining, and restore oral mucosa
while avoiding problems related to excessive flap thick-
ness and hair growth that are sometimes seen with skin
flap reconstruction of these areas (85). Serratus anterior
myofascial flaps that include the lower slips of the ser-
ratus anterior muscle provide vascularized muscle with
average dimensions of I8 x 9cm (9). When harvested
in combination with latissimus dorsi myofascial flaps,
combined latissimus dorsi-serratus anterior myofascial
flaps provide a very large surface area of muscle that is
perfused by a single vascular pedicle and is well suited
for resurfacing of extensive soft tissue defects of the
scalp (18).
Skin flap dimensions as large as 220 cm2 have
been reported in clinical case series of serratus ante-
rior musculocutaneous flaps (3I). Pittet et al. (60)
reported 27 cases of serratus anterior musculocuta-
neous free flaps used for facial skin reconstruction.
Skin paddle size ranged from 35 to I58cm2 There
was one case (4%) of total flap necrosis and two cases
(7%) of partial flap necrosis, and the aesthetic results
of reconstruction were classified to be satisfactory to
good in most cases.

Use of serratus anterior myofascial flaps as a func-
tional muscle transfer was popularized by Whitney et al.
(90), who performed 20 cases of serratus anterior free
flap facial reanimation. Serratus anterior myofascial
flaps offer several theoretical advantages over other free
muscle flaps used for facial reanimation. Cadaver dis-
sections have demonstrated that up to five different ser-
ratus anterior slips can be harvested that are separately
innervated by individual branches of the long tho-
racic nerve (19). Furthermore, each serratus anterior
slip contains a loose areolar plane between them that
allows separation of each slip into a superficial subslip
and a deep subslip ( 40). Theoretically, this anatomic
feature allows for insetting of a reinnervated serratus
anterior myofascial flap with up to 10 different force
vectors, thereby potentially more closely reproducing
the individual actions of the 16 facial mimetic mus-
cles. Furthermore, serratus anterior muscle slips have a
muscle thickness and strength of contraction that more
closely approximates those of facial mimetic muscula-
ture when compared to the more widely utilized gracilis
muscle (39). Relatively few studies have analyzed the
long-term results of facial reanimation using serratus
anterior myofascial free flaps. Leonetti et al. reported
that three patients achieved Hause-Brackman grade
3 facial function after undergoing a wide-field parot-
idectomy with facial nerve resection, followed by sural
nerve cable graft reconstruction of the upper division
of the facial nerve, and creation of an oral commissure
dynamic sling using a reinnervated serratus anterior
myofascial free flap (38). Yla-Kotola et al. (92) com-
pared the long-term results of facial reanimation in
patients with long-standing facial paralysis using cross
facial nerve grafts to reinnervate gracilis free flaps
(n = 11), latissimus dorsi free flaps (n = 10), or serra-
tus anterior free flaps (n = 6). The authors noted that
the poorest long-term results were observed in patients
who underwent serratus anterior free flaps. However,
the authors also noted that all of the serratus anterior
free flaps were performed during the early period of
their experience with facial reanimation surgery, so
that the poorer results seen in the serratus anterior flap
patients may be a reflection of a surgical learning curve
rather than the type of muscle used.
The serratus anterior branch of the thoracodorsal
artery provides a reliable periosteal blood supply to
the ribs from which the lower serratus anterior muscle
slips arise. Angiography shows that the 6th through the
9th ribs are most consistently supplied through a rich
anastomotic network arising from the serratus branch
of the thoracodorsal artery and the intercostal arteries
that supply the periosteum of the ribs. The segments of
rib harvested with the composite serratus anterior-rib
flaps should be centered in the region of the anterior
axillary line. In that location, the rib's vascular supply
is most robust through both the periosteal attachments
LATISSIMUS DORSI AND SERRATUS ANTERIOR 335
of the serratus anterior muscle to the rib and the blood
supply provided by the serratus anterior branch of the
thoracodorsal system (29).
Serratus anterior-rib osteomusclular flaps have
been used for the reconstruction of defects of the scalp
and underlying calvarium (78,84). In this technique,
up to three vascularized rib grafts can be harvested
for reconstruction of cranial bone defects (73). When
two rib grafts are removed, many authors recommend
removal of alternating ribs to reduce the risk of post-
operative chest wall instability. For harvest of three rib
bone grafts, Lin et al. advocated harvest of alternating
and consecutive ribs (e.g., ribs 5, 6, and 8), but in this
instance, the authors also recommended that chest wall
reconstruction using Marlex mesh be performed in the
area of consecutive rib harvest (41).
Vascularized rib offers distinct advantages and dis-
advantages for mandible reconstruction. As a membra-
nous bone, rib has a very thin cortex and is therefore
ill-suited for placement of dental implants after rib graft
reconstruction of the mandible. As a result of this limita-
tion, rib grafts are better suited for reconstruction of the
posterior-lateral mandible than the anterior mandible.
However, rib grafts can provide a very long bone graft,
making them useful for reconstruction of long segmen-
tal defects of the mandible in patients in whom fibula
flap reconstruction is precluded by the presence of
peripheral vascular disease. In cases that include resec-
tion of the mandibular condyle, condylar reconstruction
can be accomplished using a vascularized costochon-
dral rib graft that contains bone for mandible recon-
struction and costochondral cartilage for reconstruction
of the temporomandibular joint. Furthermore, the
natural curvature of rib approximates that of the man-
dibular body, which often allows for reconstruction of
hemimandibular defects without creation of contouring
osteotomies. For defects that include the anterior arch
of the mandible, a narrow strip of rib inner cortex is
excised at the midline of the mandibular reconstruction,
and the outer cortex is green stick fractured in order
to create the curvature of the anterior mandible
(see Figs. 20-39 and 20-40).
The first clinical case of the osteomuscular ser-
ratus anterior-rib flap for mandible reconstruction
was described in 1985 by Richards et al. ( 63). They
reported total flap survival in four of seven pedicled and
free serratus anterior-rib flaps that were used for man-
dible reconstruction. Penfold et al. (58) reported the
outcome of five serratus anterior-rib free flaps and two
serratus anterior-rib pedicled flaps used for mandible
reconstruction. One of two pedicled flaps failed com-
pletely. Among the five serratus anterior-rib free flaps,
one flap experienced partial bone graft necrosis, while
four of five free serratus anterior-rib flaps survived com-
pletely. Ioannides et al. (32,33) reported two series of
patients who underwent serratus anterior-rib flaps for
336 CHAPTER ZO
mandibular reconstruction. Monocortical defects of the
mandible resulting from debridement of osteoradione-
crosis was a common indication, and in this situation a
monocortical rib graft containing only the outer cortex
of the rib was fashioned for use as an onlay vascular-
ized bone graft. In the eight patients who underwent
serratus anterior-rib flap reconstruction after debride-
ment of osteoradionecrosis using this technique, 25%
experienced postoperative pathologic fractures that
required additional free flap reconstruction. These frac-
tures occurred outside the region of the rib bone graft.
The rib bone grafts appeared to be well incorporated,
indicating that the cause of the pathologic fracture was
probably related to progression of the osteoradione-
crosis rather than to failure of the bone graft. N etscher
et al. (55) reported no complications in four patients
who underwent reconstruction of segmental defects
of the mandible using serratus anterior-rib flaps. Two
patients who underwent condylar reconstruction using
vascularized costochondral rib grafts developed good
mandible function with no pain on mastication. In the
largest series of mandible reconstructions using serratus
anterior-rib flaps published to date, Kim and Blackwell
(35) reported no cases of flap failure, no cases of rib
graft resorption, and one case of bone graft nonunion
in 28 patients who were reconstructed in this fashion.
NEUROVASCULAR ANATOMY
I noted that some important arterial branches stem from
a prominent branch of the subscapular artery and extend
to the pedicle of my flap. It is this branch that is called
the scapular circumflex, and of this branch, the lower one
is the most involved in the feeding of the flap. The arte-
rial branch makes a path toward the surface between the
teres major and teres minor and the branch reaches in part
to the latissimus dorsi muscle and to the skin. It is also
important to note that the latissimus dorsi muscle receives
branches straight from the subscapular artery. From this
observation, it is understood that in order to ensure the
vitality of the flap, it is necessary to include at least the latis-
simus dorsi muscle: by this method, we not only preserve
the flow of blood to the skin that stems from the scapular
circumflex, but also that which arrives there by way of the
latissimus dorsi muscle. Tansini (1906)
The degree ofTansini's understanding of the blood
supply to the skin of this musculocutaneous flap is truly
astounding and boggles the minds of students of medical
history who ponder the tremendous time delay between
Tansini's discovery and the widespread enthusiasm with
which this donor site is presently embraced ( 49). Although
Tansini's description wrongly placed too great an empha-
sis on the role of the circumflex scapular artery, it is evi-
dent that he understood the axial pattern blood supply to
the latissimus dorsi muscle and the necessity to include
that blood supply to ensure the viability of the skin.
According to the classification system of Mathes and
Nahai (47), the latissimus dorsi and serratus anterior
muscles are type 5 muscles, with one dominant vascu-
lar pedicle and a series of smaller segmental pedicles.
The dominant pedicle is composed of the thoracodor-
sal artery and vein, which are the terminal branches of
the subscapular artery and vein. There are numerous
musculocutaneous perforators over the entire latis-
simus dorsi, with a particular preponderance located
along the anterior border. In addition, there are seg-
mental paraspinal perforators in the medial portion of
the muscle (Fig. 20-5).
Taylor and Palmer (80) described a system of angi-
osomes to help explain the vascular basis for reconstruc-
tive flaps. They divided the body into three-dimensional
territories that are each supplied by a named source
artery and vein and connected to each other by a sys-
tem of choke vessels. The latissimus dorsi flap is divided
into three angiosomes: the proximal portion of the mus-
cle, which is supplied by the thoracodorsal pedicle; the
mid and medial portion, which is supplied by the pos-
terior intercostals; and the most caudal portion, which
is nourished by the lumbar arteries (Fig. 20-6). The
intercostal system of perforators are sizable vessels that
range from l to 1.5 mm in diameter and have been used
to support posteromedially based, or "reversed," muscle
and musculocutaneous flaps (8). However, the thora-
codorsal pedicle is the dominant supply for pedicled or
free flap transfers to the head and neck.
The angiosome concept allows us to predict the
reliability of muscle and skin transferred from differ-
ent regions overlying the latissimus dorsi muscle. Tay-
lor and Palmer (80) hypothesized that muscle and skin
can be reliably harvested from an adjacent angiosome
by crossing one system of choke vessels. However, the
vascularity diminishes when the "tertiary" territory is
harvested by crossing the second set of connecting ves-
sels.1bis has been the experience encountered with the
latissimus dorsi flap, which is unreliable in its caudal
and medial segments.
The thoracodorsal artery and vein arise from the
subscapular vessels, which are branches of the third
portion of the axillary artery and vein. The thoracodor-
sal vessels run in a cephalocaudal direction through the
fatty tissue of the axilla prior to entering the hilum of
the latissimus dorsi muscle. In their course, they give
off branches to a variety of muscles, including the sub-
scapularis, teres major, and serratus anterior muscles.
In addition, they supply a consistent angular branch
to the tip of the scapula. The serratus anterior branch
of the thoracodorsal artery arises just proximal to the
point where the thoracodorsal vessels enter the hilum of
the latissimus dorsi muscle (Fig. 20-7).
In the majority of cases, there is a single branch to
the serratus anterior muscle (54%), with double (44%)
and triple branches (2%) encountered less frequently
 LATISSIMUS DORSI AND SERRATUS ANTERIOR 337

Subclavian a.

Circumflex scapular a .

Thoracodorsal n.

intercostal arteries

FIGURE 20-5. The primary blood supply to the latissimus dorsi is from the thoracodorsal
artery and vein that arise from the subscapular axis. Secondary blood supply arises from the
para spinous perforators. The thoracodorsal pedicle consistently divides into transverse and
longitudinal branches. The thoracodorsal pedicle can usually be mobilized to achieve a length
of 10to 12cm. The thoracodorsal artery has a diameter of 1.5 to 4.0mm prior to branching. The
thoracodorsal nerve supplies motor innervation to the muscle and also divides into transverse
and longitudinal branches.
338 CHAPTER 20

FIGURE 20-6. The latissimus dorsi may be divided into three primary angiosomes. By doing
so, we can better understand the reliability of the different skin paddles overlying this muscle.
The upper territory is primarily supplied by the source artery, the thoracodorsal artery (1). The
second vascular territory (II) of the latissimus dorsi is supplied by the intercostal perforators that
enter both medially in the paraspinal region, as well as laterally. The lower portion of the muscle
is supplied by lumbar vessels (Ill). The skin paddles overlying the third angiosome are less
reliable when based entirely on the thoracodorsal system. It is important to realize this factor
when using the latissimus dorsi musculocutaneous flap as a pedicled flap. Improving the arc of
rotation is often done by placing the skin paddle over the more caudal extent of the muscle but,
therefore, incurring a greater risk of skin ischemia.
 LATISSIMUS DORSI AND SERRATUS ANTERIOR 339

./

Circumflex
scapular artury --+~~---,;ali:--~:a~-.t4

~~~~l:"t--t--+--- Serratus anterior

branch of
lhoracodo1'881 811ery

FIGURE 20-7. The primary blocd supply to the lower portion of the serratus anterior muscle
arises from one to three branches of the thora codorsaI artery. The serratus anterior muscle is
innervated by the long thoracic nerve, which travels on the superficial aspect of the serratus
anterior muscle in close proximity to the point where the serratus branch of the thoracodo rsal
artery enters into the superficia I fascia of the serratus anterior muscle.
340 CHAPTER ZO
(4). The serratus anterior branch of the thoracodorsal
artery travels inferiorly and anteriorly in an oblique
course before entering into the serratus anterior muscle.
The average length of the vascular pedicle from the sub-
scapular artery to the serratus anterior muscle is 9.2cm
(19). After entering into the serratus anterior muscle,
the serratus anterior branch of the thoracodorsal artery
gives off a common "slip" branch to each of the lower
four or five slips of the serratus anterior muscle.
The vascular anatomy of the thoracodorsal vessels has
been extensively studied. The average diameter of the
artery at its origin is 2.7mm (nmge, 1.5 to 4.0mm); the
diameter of the vein is 3.4mm (range, 1.5 to 4.5mm);
the average length of the thoracodorsal pedicle is 9.3cm
(range, 6.0 to 16.5cm). In the majority of cadaver dis-
sections (92%), the subscapular artery and vein arose in
close proximity from the axillary vessels, and the thoraco-
dorsal vessels traversed the axilla together. In the remain-
ing cases, the artery and vein arose at a distance from each
other and ran separate courses until joining as far caudal
as the takeoff of the branches to the serratus anterior.
At the neurovascular hilum, the thoracodorsal vein
is situated lateral to the thoracodorsal artery, and the
thoracodorsal nerve runs in between. In 99% of cases,
there was a single neurovascular hilum; in the remain-
ing I%, there were two distinct hila identified (83). In
86% of cases, the latissimus dorsi branch of the thora-
codorsal vessels was found to bifurcate into transverse
and longitudinal branches. The transverse branch usu-
ally paralleled the free upper border of the latissimus
dorsi muscle, separated from the edge by an average
of 3.5 em. The longitudinal branch was usually slightly
smaller and ran a cephalocaudal course toward the iliac
crest at a distance of 2.1 em from the lateral edge of
the latissimus dorsi muscle. Dye-injection studies of
each of these branches produced staining of the entire
latissimus dorsi muscle and the overlying skin, which
reflected the rich anastomoses between the two major
branches. The intramuscular branching pattern of the
thoracodorsal nerve parallels the vessels and, therefore,
provides the anatomic basis for harvesting two separate
vascularized neuromuscular units. It also provides the
opportunity to preserve a functional muscle and, there-
fore, to reduce donor-site morbidity (83).
As the latissimus dorsi muscle is interposed between
the serratus anterior branch of the thoracodorsal artery
and the chest wall skin, the serratus anterior branch of
the thoracodorsal artery gives off no direct cutaneous
branches to the skin of the lateral chest wall. However,
serratus anterior musculocutaneous flaps can be trans-
ferred with a skin paddle that is perfused by a rich col-
lateral network that exists between the serratus anterior
branch of the thoracodorsal artery and cutaneous per-
forators from intercostals vessels that pierce the serratus
anterior muscle as they course to the chest wall skin. A
dye injection study indicated that an average cutaneous
area of 144cm2 can be transferred based upon the
serratus anterior branch of the thoracodorsal artery (53).
One of the most appealing characteristics of the latis-
simus dorsi-serratus anterior donor site is the length
of the vascular pedicle. There is considerable varia-
tion in the combined length of the thoracodorsal and
subscapular vessels, which ranged from 7.6 to 14.4em
(average, 9.7 em). Differences in length appeared to be
more related to the size of the patient rather than to a
variation in the point of entry into the muscle (17). In
two separate cadaver dissection series, there was little
atherosclerosis noted (4,17).
The thoracodorsal nerve supplies the motor innerva-
tion to the latissimus dorsi muscle. It arises from the
posterior cord of the brachial plexus. It enters the axilla
from behind the axillary vessels and then descends with
the thoracodorsal artery and vein to the neurovascular
hilum. The thoracodorsal nerve usually crosses the axil-
lary vessels approximately 3 em proximal to the sub-
scapular artery and vein. The length of nerve that can be
harvested ranges from 8.5 to 19.0cm (average, 12.3cm).
Bifurcation of the thoracodorsal nerve was noted in 85%
of dissections (4). Earlier in this chapter, the sensory
supply to the skin of the back and trunk was discussed.
The long thoracic nerve runs on the superficial aspect of
the serratus anterior muscle, which it innervates. Motor
fibers arising from the C5 and C6 nerve roots innervate
the upper portion of the serratus anterior muscle, while
the lower portion ofthe serratus anterior muscle is inner-
vated by the C7 nerve root (6). The long thoracic nerve
runs in close proximity to the serratus anterior branch
of the thoracodorsal artery beginning at the point where
the serratus anterior branch of the thoracodorsal artery
enters into the superficial fascia of the serratus anterior
muscle, usually at the junction of muscle slips 5 and 6.
From this point inferiorly, the long thoracic nerve con-
sistently runs deep to the serratus anterior branch of the
thoracodorsal artery. For purposes offacial reanimation,
the long thoracic nerve gives off a separate branch to
each serratus anterior muscle slip. Nerve fascicles trave-
ling to each serratus anterior muscle slip can be sepa-
rated from each other by incising the epineurium of the
long thoracic nerve, producing individual nerve grafts to
each slip that have a length of approximately 2 em. The
total length of the long thoracic nerve from the junction
of muscle slips 5 and 6 to the apex of the axilla is about
13 em. The average diameter of the long thoracic nerve
is 2.0mm at the apex of the axilla and l.6mm at the
junction of muscle slips 5 and 6 (19).
ANATOMIC VARIATIONS
There are few significant anatomic variations of the
latissimus dorsi or serratus anterior muscles themselves.
As noted previously, the subscapular artery and vein

arise in close proximity to each other in the majority of
dissections. In those cases in which their origins are sep-
arated, the subscapular artery arises proximally in the
axilla, by an average of 4.2 em. The surgeon should be
aware of this anatomic variant to avoid confusion when
harvesting this flap ( 4). On occasion the thoracodor-
sal artery arises as a separate branch from the axillary
artery. In a series of 5I2 cadaver dissections, DeGaris
and Swartley {11) reponed that this anomaly occurs
with an incidence of0.8%.
The thoracodorsal artery gives rise to at least one
serratus anterior branch in approximately 99% of cases
(9I). In the remaining I% of cases, the serratus ante-
rior branch of the thoracodorsal artery is absent, and
the lower portion of the serratus anterior muscle may
derive its blood supply from a variety of sources includ-
ing a branch of the subscapular artery, a branch of the
axillary artery, a branch of the I st intercostal artery, or
a branch of the lateral thoracic artery (2I).
POTENTIAL PITFALLS
The latissimus dorsi flap is a safe and reliable recon-
structive option, as evidenced by the abundance of
reports in the literature. However, marginal flap necro-
sis is not uncommon and most likely results from
improper design over the more distal aspects of the
muscle (I5). This is usually caused by pushing the limits
of the maximum size of this flap or the placement of the
skin paddle closer to the iliac crest to enhance the arc
of rotation. This problem can be avoided by transferring
the latissimus dorsi as a free flap.
The route of passage of the latissimus dorsi and
serratus anterior flaps as pedicled flaps is through the
axilla and over the anterior chest, between the pectoralis
major and minor muscles. This transfer must be done
with great caution. The tremendously long course that
the vessels traverse places them at risk of occlusion with
changes in arm position. Immobilization of the arm in
a flexed position across the chest has been advocated
{48). If the pectoralis major has not previously been
used as a reconstructive flap, it is imperative to avoid
injury to the thoracoacromial vessels when creating the
tunnel. In addition to being cautious about injury to the
long thoracic nerve and paralysis of the serratus anterior
with winging of the scapula, Baker (2) warned against
injury to the medial brachial cutaneous nerve that sup-
plies sensation to the medial aspect of the upper arm.
Failure to create an adequate tunnel may lead to the
compression of the vascular pedicle between the pecto-
ralis major and the clavicle. Watson et al. (89) reported
one case of total flap necrosis caused by this problem.
An even greater concern is the reponed injuries to
the brachial plexus that have occurred as a result of posi-
tioning the arm during flap harvest. Quillen {6I) noted
a single case of temporary radial nerve weakness, and
LATISSIMUS DORSI AND SERRATUS ANTERIOR 341
Barton et al. (5) reported four cases of temporary weak-
ness and sensory changes in the upper extremity. The
mechanism of injury to the brachial plexus was studied
by Logan and Black ( 43) who described an additional
case of brachial plexus injury with a permanent deficit
noted at 7 months after latissimus dorsi flap harvest.
These authors described impingement of the brachial
plexus by the clavicle on extreme elevation of the arm.
They advised placing a pad between the shoulder and
neck to help prevent the occurrence of this injury.
Sacrifice of the nerve supply of a muscle carries an
intrinsic m01bidity caused by the loss of its functional
activity. Although the deficits resulting from the loss of
the latissimus dorsi muscle or the serratus anterior muscle
have been examined, the combined loss of several muscles
that act across the glenohumeral joint has not. It is fairly
common for patients with head and neck cancer to lose
the action of the trapezius following radical neck dissec-
tion and the pectoralis major for reconstructive purposes.
Harvest of a scapular composite flap may additionally lead
to disruption or denervation of the teres major, subscapu-
laris, and infraspinatus muscles. The detrimental effects of
radiation therapy on the brachial plexus may add to shoul-
der and upper extremity morbidity. Laitung and Peck
(37) compared shoulder function in a group of I9 male
patients who underwent latissimus dorsi transfer with that
of a group of matched controls. The range of shoulder
motion was normal in I3 of I9 patients who underwent
the flap procedure. The disability score diminished with
greater time from surgery. Scar contracture had a detri-
mental effect on the disability score as well as the range of
motion. These authors found that latissimus dorsi muscle
sacrifice caused few occupational problems and did not
interfere with normal sporting activities. These findings
differ from those reported by Russell et al. {68) who noted
a change in occupation, household activities, and sport-
ing activities attributable to the loss of the latissimus dorsi
muscle in a group of 23 patients. They found a global
weakness of virtually all the muscles tested surrounding
the operated shoulder. In a small group of patients with
an ipsilateral loss of pectoralis major function as a result
of either Poland,s syndrome or mastectomy, the authors
noted an even greater degree of shoulder weakness follow-
ing latissimus dorsi muscle transfer.
Loss of the latissimus dorsi muscle has been asso-
ciated with additional donor site problems in children,
which were related to the development of scoliosis and
underdevelopment of the paravertebral muscles. Serra et
al. {72) described a technique of reneurotizing the resid-
ual muscle to preserve latissimus dorsi function. In this
procedure, the thoracodorsal nerve is transected at the
point where it enters the muscle. Following the transfer
of a portion of the muscle, the cut ends of the nerve
are reintroduced into the remaining muscle. Follow-up
electromyographic studies revealed reinnervation of the
muscle. There was no mention, however, of whether the
development of the feared sequelae was avoided.
342 CHAPTER ZO
The potential morbidity of transferring the latissimus
dorsi in pediatric patients may differ from that in adults.
However, with the majority of free flaps transferred to
the head and neck occurring in an older population of
patients with cancer, the functional shoulder deficits
must be studied in a systematic fashion to account for
the variables noted.
Derby et al. examined donor site morbidity in 34
cases of serratus anterior free tissue transfer using
chart review and a follow-up questionnaire. Mild wing-
ing of the scapula was noted in three of nine patients
with documented postoperative physical examinations,
although only one patient was symptomatic. The num-
ber of muscle slips harvested did not significantly affect
the prevalence of winging of the scapula. No patients
attributed any work disability to their donor site, and
many participated in athletic activities that required
upper extremity strength and dexterity (12). Dumont
et al. (14) performed a prospective objective analysis
of shoulder strength in seven patients who underwent
a serratus anterior myofascial flap harvest. They found
no significant difference when comparing preoperative
shoulder strength with shoulder strength that was tested
3 months after serratus anterior flap harvest.
PREOPERATIVE ASSESSMENT
The anterior border of the latissimus dorsi can be read-
ily palpated by instructing patients to press their hands
against their hips while seated. Preoperative localiza-
tion of the thoracodorsal vessels is rarely necessary
because of the consistency of anatomy. In patients who
have had a prior axillary lymph node dissection, the sit-
uation is more complex because of the likelihood that
the thoracodorsal vessels have been ligated. Collateral
flow to the muscle is derived through the rich anas-
tomotic network around the scapula, with retrograde
flow in the circumflex scapular artery and the branches
to the serratus anterior (51). Although coverage of the
chest wall may be feasible under these circumstances,
it is unlikely that the flap could be safely harvested for
use in the head and neck unless it is transferred as a
free flap. Preoperative angiography is advocated in this
setting (67).
Acknowledgments
The authors thank Dr. Michael Sullivan for his contri-
butions to this chapter in the first edition of this book.
 LATISSIMUS DORSI AND SERRATUS ANTERIOR 343

Latissimus Dorsi Myocutaneous Flap

FIGURE 8. The topographical anatomy of

the latissimus dorsi flap has been outlined on
the left flank. The important landmarks include
the midpoint of the axilla, the iliac crest,
and the scapular tip. A dsshed line is drawn
between the midpoint of the axilla and a point
midway between the anterior superior iliac
spine and the posterior superior iliac spine on
the iliac crest. This line represents the anterior
border of the latissimus dorsi. Eight to 10 em
below the midpoint of the axilla, along this line,
the thoracodorsal artery and vein enter the
undersurface of the latissimus dorsi, and the
vessels divide into a horizontal branch, which
runs a few centimeters below the scapula tip,
and a more vertically directed branch, which
runs 3to 4cm posterior to the anterior edge of
the muscle.

FIGURE 20-9. The outline for a cutaneous

paddle has been drawn based on the vertically
oriented branch. The cutaneous paddle is usu-
ally drawn as a fusiform shape to assist in clo
sure. The cutaneous paddle has been designed
with a few centimeters of random skin anterior
to the leading edge of the latissimus dorsi.

FIGURE 20.10. The initial incision is made at

the midpoint of the axilla and runs along the
dashed line superiorly and the anterior edge of
the cutaneous paddle inferiorly. Through this
incision, the anterior leading edge of the latis-
simus muscle is easily identified.
344 CHAPTERZU

Latissimus Dorsi Myocutaneous Flap

FIGURE 20-11. The thoracodorsal pedicle runs

through the adipose tissue of the axilla, provid-
ing branches to the regional muscles and the
angular branch to the scapula before enter-
ing the hilum of the latissimus dorsi muscle.
The branches to the serra'b.Js anterior (double
arrows) are often the first vascular struc'b.Jres
to be encountered. These branches lie on the
superficial surface of the serra'b.Js anterior (large
arrow) and can be used to find the thoracodor-
sal vessels from which they arise.

FIGURE 20-12. With the anterior edge (arrows)

of the latissimus dorsi retracted posteriorly, a
su'b.Jre has been placed around the branches to
the serratus anterior.

FIGURE 13. In a dissection in which much

of the adipose tissue has be en removed, the
anatomy of the thoracodorsa I system, consist-
ing of artery, vein, and nerve, can be more
clearly visualized.
 LATISSIMUS DORSI AND SERRATUS ANTERIOR 345

Latissimus Dorsi Myocutaneous Flap

FIGURE 20-14. Division of the serratus anterior

branch allows near-complete mobilization of
the thoracodorsal pedicle (white arrows).

FIGURE 20-15. An incision is then made cir-

cumferentially around the posteromedial portion
of the skin paddle. This incision is made to the
level of the fascia overlying the muscle. At. this
juncture, the dissection may proceed by harvest-
ing only a limited portion ofthe latissimus dorsi
underlying the skin paddle or the entire muscle.

FIGURE 20-16. Elevation of the back skin off

the muscle provides exposure of the latissimus
dorsi to the posterior midline.
346 CHAPTERZU

Latissimus Dorsi Myocutaneous Flap

Skin paddle
FIGURE 20-17. The latissimus dorsi is mobilized by blunt and sharp dissection off the chest wall
and the external oblique and serratus anterior muscles. The muscle and aponeurotic attach-
ments to the iliac crest, the vertebrae, and the ribs are sharply transacted. Feeders entering the
deep surface ofthe muscle during the inferior and medial dissection must be ligated and divided.
 LATISSIMUS DORSI AND SERRATUS ANTERIOR 347

Latissimus Dorsi Myocutaneous Flap

FIGURE 2018. As the dissection proceeds

distally to proximally, the vascular hilum is iden-
tified. Careful dissection along the thoracodor-
sal neurovascular pedicle toward the axilla
requires division of muscular branches and the
angular branch.

FIGURE 20-19. Complete mobilization of the

latissimus dorsi requires transection ofthe ten-
dinous insertion to the humerus (dotted white
line). This maneuver is performed carefully
while protecting the vascular pedicle.

FIGURE 2020. The tendon (arrowt has been

cut leaving the muscle attached by the neuro-
vascular pedicle. A suture is placed around the
circumflex scapular vessels that can be ligated
to further mobilize the thoracodorsal pedicle.
348 CHAPTERZU

Latissimus Dorsi Myocutaneous Flap

FIGURE 20-21. Passage of the pedicled flap

requires preparation of a tunnel between the
pectoralis major and minor. The lateral edge of
these muscles is identified in the anterior axilla.

FIGURE 20-22. An incision parallel and inferior

to the clavicle is required to complete the
tunnel. The pectoralis major attachments to the
clavicle are incised.

FIGURE 20-23. The thoracoacromial pedicle is

identified and preserved.
 LATISSIMUS DORSI AND SERRATUS ANTERIOR 349

Latissimus Dorsi Myocutaneous Flap

FIGURE 20-24. A generous tunnel must be

created to allow passage of the latissimus dorsi
flap. A good guide to an adequate passage is
one that accommodates at least four of the
surgeon's fingers.

FIGURE 20-25. Under direct vision, the latis-

simus dorsi flap is passed through the tunnel
while being certain not to twist the pedicle or
to cause shearing forces between the skin and
the muscle.

FIGURE 2026. The latissimus dorsi flap has

been 1ransferred to the temporal region without
tension on the vascular leash. This important
point can only be assured by directly visual-
izing the pedicle as it travels through the axilla
and over the chest wall. Once the safety of the
pedicle is determined, the surgeon must be
certain that there is no further tension placed
on the skin paddle during flap inset.
350 CHAPTERZU

Latissimus Dorsi Myocutaneous Flap

FIGURE 2D-Z7. The importance of transacting

the tendon of the latissimus muscle to maximize
the arc of rotation is illustrated by the posi-
tion of the flap shown here compared with the
previous figure in which the tendon had not yet
been cut.

FIGURE 20-ZI. Harvest of the free latissimus

dorsi musculocutaneous flap reveals the long
neurovascular pedicle and the tremendous
surface area that can be covered.

FIGURE 20-29. Primary closure of the donor

site can usually be accomplished by wide
undermining to produce a linear scar.
 LATISSIMUS DORSI AND SERRATUS ANTERIOR 351

Latissimus Dorsi-Serratus Anterior-Rib Osteomyocutaneous Flap

for Oromandibular Reconstruction

FIGURE 20-30. Patient positioning and flap

design is demonstrated for a right-sided latis-
simus dorsi-serratus anterior-rib osteomyocu-
taneous flap for oromandibular reconstruction.
The flap is harvested with the patient in a supine
position, which allows for simultaneous dissec-
tion in the head and neck and axilla without intra-
operative patient repositioning. The dimensions
of the skin paddle are varied according to the
soft tissue defect. The precise position of1he rib
graft is detennined by surgical exploration that
reveals 1he location of1he branch(es) of 1he tho-
racodorsal artery to the seratus anterior muscle.
Most frequently 1his is 1he 6th or 1he 7th rib.

FIGURE 2D-31. The anterior skin incision is

deepened onto the muscles of 1he lateral chest
wall. Longitudinally oriented muscle fibers are
those of the latissimus dorsi (LD) muscle, while
horizontal and obliquely oriented muscle fibers
are from the serratus anterior muscle (SA).

FIGURE 20-32. The anterior border of the lati-

simus dorsi muscle (yellow srrows) is elevated
off the serratus anterior muscle, revealing
the vascular pedicles. In this case, there are
two serratus anterior branches (SAB) of 1he
thoracodorsal artery (TOA). The proximal ser-
ratus anterior branch is ligated, and the flap is
perfused by the distal serratus anterior branch.
This allows for harvest of an inferior slip of
the serratus anterior muscle, while preserving
function in the superior slips of the muscle to
prevent winging of the scapula.
352 CHAPTER 20

Latissimus Dorsi-Serratus Anterior-Rib Osteomyocutaneous Flap

for Oromandibular Reconstruction

FIGURE 20-33. A single slip of serratus

anterior muscle that is perfused by the distal
serratus anterior branch has been isolated
by deepening the plane of dissection onto 1he
intercostal muscles located above and below
the selected serratus anterior muscle slip. This
requires transection of the long thoracic nerve
just above the point where the SAB enters into
the selected serratus anterior muscle slip. The
rib bone graft is centered at1he anterior axillary
line, where the periosteal attachments and
blood supply of the serratus anterior muscle to
the underlying rib are most robust.

FIGURE 20-34. The pleura is exposed above

and below the identified rib graft by dividing the
intercostals muscles. A periosteal elavator is
used to dissect the posterior surface of the rib
graft free from the underlying pleura, and then
the rib is cut anteriorly and posteriorly.
A variable length of rib can be harvested,
depending on the length of the mandibular
defect. If the mandibular defect includes resec-
tion ofthe condyle, then a portion of costo-
chondral cartilage is included atthe anterior
aspect of the rib graft for reconstruction of the
temporomandibular joint.
 LATISSIMUS DORSI AND SERRATUS ANTERIOR 353

Latissimus Dorsi-Serratus Anterior-Rib Osteomyocutaneous Flap

for Oromandibular Reconstruction

FIGURE 20-35. Elevation of the serratus

anterior-rib component of the flap has been
completed. The remaining dissection is the
same as that described for the elevation of a
latissimus dorsi myocutaneous free flap (see
Figures 2D-15 through 20-25 for full details).

FIGURE 20-36. The posterior skin incision is

made onto the latissimus dorsi muscle. Flap
elevation is completed by transection ofthe
attachments of the latissimus dorsi muscle to the
thoracolumbar fascia inferiorly, the spine pos-
teriorly, and the humerus superiorly. To achieve
maximal pedicle length and caliber, the thora-
codorsal vascular pedicle is dissected proxi-
mally through the axilla up to the take off of tile
subscapular vessels from the axillary vessels,
ligating the side branches including the angular
branch and the circumflex scapular branch.

FIGURE 20-31. The harvested flap demon-

strates the long length of the vascular pedicle
and the great arc of rotation between the
independent soft tissue and bone components
of the flap.
354 CHAPTER 20

Latissimus Dorsi-Serratus Anterior-Rib Osteomyocutaneous Flap

for Oromandibular Reconstruction

FIGURE 20-38. The flap is used for reconstruc-

tion of a defect of the anterior and right lateral
mandible and floor of mouth. The curvature of
the rib is appropriate for reconstruction of the
right hemimandible, but a single osteotomy will
be made in the midline of the reconstructed
mandible to produce the curvature of the
reconstructed anterior mandible.

FIGURE 20-39. A 10-mm-wide strip of posterior

rib cortex is excised atth e site of the mid line
osteotomy.

FIGURE 20-40. The anterior cortex of the rib is

green stick fractured to produce the curvature
of the reconstructed anterior mandible.
 LATISSIMUS DORSI AND SERRATUS ANTERIOR 355

Latissimus Dorsi-Serratus Anterior-Rib Osteomyocutaneous Flap

Ofor Oromand ibular Reconstruction

FIGURE ZD-41. The rib bone graft is fixed to

the mandibular reconstruction plate.

FIGURE 20-42. The floor of mouth is recon-

structed using the latissimus dorsi myocutane-
ous component of the flap. The flap vascular
pedicle is positioned for anastomosis to recipi-
ent blood vessels in the right neck (yellow
arrow).

REFERENCES 4. Bartlett SP, May Jw.Yarernchuk MJ: The latissimus dorsi

1. Arnold PG, Pairolero PC, Waldorf JC: The serratus ante-
rior muscle: intmthoracic and extmthoracic utilization.
Pfast Reconst:r Surg 1984;73:240-248.
2. Baker S: Closure of large orbital maxillary defects with
free latissimus dorsi myocutan.eous flaps. Head Neck
1984;6:828.
3. Barrow D, Nahai F, Fleischer A: Use of free latissimus
dorsi musculocutaneous flaps in various neurosurgical
disorders. J Neurosurg 1983;58:252.
muscle. A fresh cadaver study of the primary neurovascu-
lar pedicle. Pfast Reconsw Surg 1981;67:631.
5. Barton F, Spicer T, Byrd H: Head and ned:: reconstruc-
tion with the latissimus dorsi myocutaneous flap: ana-
tomic observations and report of 60 cases. Pfast Reconst:r
Surg 1983;71:199.
6. Bertelli JA, Ghizoni MF: Long thoracic nerve: anatomy and
functional assessment. J. &me Joint Swg 2005;87:993-998.
7. Bostwick J, Nahai F, Wallace J, Vasconez L: Sixty latis-
simus dorsi flaps. Pfast Reconst:r SUf'g 1979;83:31.
356 CHAPTER ZO
8. Bostwick J, Schiflan M, Nahai F, Jurkiewicz M: The
"reverse" latissimus dorsi muscle and musculocutaneous
flap: anatomical and clinical considerations. Plasz &conslT
Surg 1980;65:395.
9. Cuadros CL, Driscoll CL, Rothkopf DM: The anatomy
of the lower serratus anterior muscle: a fresh cadaver
study. Plan Reconm Surg 1995;95:93-97.
10. Dabb R, Conklin W: A sensory innervated latissimus
dorsi musculocutaneous free flap: case report. J Microsurg
1981;2:289-293.
11. DeGaris C, Swartley W: The axillary artery in white and
Negro stocks. Am J Anat 1928;4:353.
12. Derby W, Bartlett SP, Low DW: Serratus anterior free-
tissue transfer: harvest-related morbidity in 34 consecu-
tive cases and review of the literature.J ReconslT Microsurg
1997;13:397-403.
13. D'este S: La teclmique de !'amputation de la mamelle
pour carcinome mammaire. Rev Chir 1912;45:164.
14. Dumont CE, Domenghini C, Kessler J: Donor site mor-
bidity after serratus anterior free muscular flap: a pro-
spective clinical study. Ann Plast Surg 2004;52: 198-198.
15. Earley MJ, Green M, Milling M: A critical appraisal of
the use of free flaps in primary reconstruction of com-
bined scalp and calvarial cancer defects. Br J Plasz Surg
1990;43:283-289.
16. Fredrickson JM,JahnAF, Bryce DP: Leiomyosarcoma of
the cervical trachea. Report of a case with reconstruction
using a latissimus dorsi island flap. Ann Owl Rltino/Laryn-
gol1919;88:463-466.
17. Friedrich W, Berbenhold C, lierse W: Vascularization of
the myocutaneous latissimus dorsi flap. Acta Anal (Basel)
1988;131:97.
18. Furnas H, lineaweaverWC,Alpert BS, Huneke HJ: Scalp
reconstruction by microvascular free tissue transfer. Ann
Plast Surg 1990;24:431-444.
19. Godat D, Sanger J, Lifchez S, et al.: Detailed neurovascu-
lar anatomy of the serratus anterior muscle: implications
for a functional muscle flap with multiple independent
forse vectors. Plast Reconm Surg 2004;114:21-29.
20. Gordon L, Bunche H, Alpert B: Free latissimus dorsi
muscle flap with split-thickness skin graft cover: a report
of 16 cases. Plast &conm Surg 1982;70:173.
21. Halim AS, Wan Z: Anomalous arterial supply to the
muscles in a combined latissimus dorsi and serratus
anterior flap. ClinAnat 2004;17:358-359.
22. Hardesty R, Jones N, Swartz N, et al.: Microsurgery
for macrodefects: microvascular free-tissue transfer
for massive defects of the head and neck. Am J Surg
1987;154:399.
23. Harii K, Iwaya T, Kawaguchi N: Combination myocuta-
neous flap and microvascular free flap. Plan ReconslT Surg
1981;68:700.
24. Harii K. Ohmori K. Toru S: Free gracilis muscle
transplantation with rnicroneurovascular anastomoses
for treatment of facial paralysis. Plasz &conslT Surg
1976;57:133.
25. Harii K. Ono I, Ebihara S: Closure of total cheek defects
with two combined myocutaneous flaps. Arch Otolaryngol
1982;108:303-307.
26. Haughey B: Tongue reconstruction: concepts and prac-
tice. Laryngoscope 1993; 103:1132.
27. Haughey B, Fredrickson J: The latissimus dorsi donor
site. Current use in head and neck reconstruction. Arch
Owlaryngol Head Neck Surg 1991; 117: 1129.
28. Hayashi A, Maruyama Y: The "reduced" latissimus dorsi
musculocutaneous flap. Plast ReconslT Surg 1989;84:290.
29. Hui KC, Zhang F, Lineaweaver WC, Moon W, Huneke
GM, Huneke HJ: Serratus anterior-rib composite flap:
anatomic studies and clinical application to hand recon-
struction. Ann Plast Surg 1999;42:132-136.
30. Hutchins E: A method for the prevention of elephantiasis.
Surg Gynecal Obstet 1939;69:795.
31. Inoue T, Ueda K. Hatoko M, Harashina T: The pedicled
extended serratus anterior myocutaneous flap for head and
neck reconstruction. Br J Plast Surg 1991;44:259-265.
32. Ioannides C, Fossion E, Boeckx W: Serratus anterior
muscle in composite head and neck flaps. Head Neck
1992;14:177-182.
33. Ioannides C, Fossion E, BoeckxW, Hermans B,Jacobs D:
Surgical management of the osteoradionecrotic manbible
with free vascularized composite flaps. J Craniomaxillofac
Surg 1994;22:330-334.
34. Jones N, Hardesty R, Swarty W, Ramasastry S, Heck-
ler F, Nauton E: Extensive and complex defects of the
scalp, middle third of the face, and palate: the role of
microsurgical reconstruction. Plast ReconrlT Surg 1988;
82:937.
35. Kim PD, Blackwell KE: Latissimus-serratus-rib free flap
for oromandibular and maxillary reconstruction. Arch
Otolaryngol Head Neck Surg 2007; 13:791-795.
36. Krishna B, Green M: Extended role of latissimus dorsi
myocutaneous flap in reconstruction of the neck. Br
J Plart Surg 1980;33:233.
37. l.aitung J, Peck F: Shoulder function following loss of the
latissimus dorsi muscle. Br J Plart Surg 1985;38:375.
38. Leonetti JP, Zender CA, Vandevender D, Marzo SJ:
Long-term results of microvascular free-tissue trans-
fer reanimation of the paralyzed face: three cases. ENT
J 2008;87:226-233.
39. Lifchez SD, Gasparri M, Sanger JR, et al.: Serratus ante-
rior in vivo contractile force study. Plast &conslT Surg
2005;116:839-845.
40. Lifchez S, Sanger J, Godat D, Recinos R, LoGiudice
JA, Yan JG: The serratus anterioe subslip: anatomy and
implications for facial and hand reanimation. Plast Recon-
slT Surg 2004; 114:1068-1076.
41. Lin CH, Wei FC, Levin LS, et al.: Free composite ser-
ratus anterior and rib flaps for tibial bone and soft-tissue
defect. Plast ReconslT Surg 1997;99: 1656-1665.
42. Logan SE, Alpert BS, Huneke HJ: Free serratus anterior
muscle transplantation for hand reconstruction. Br J
Plast Surg 1988;41 :639-643.

43. Logan A, Black M: Injury to the brachial plexus resulting
from shoulder positioning during latissimus dorsi flap
pedicle dissection. Br J Plast Surg 1985;38:380.
44. Mackinnon S, Dellon L: Technical considerations of
the latissimus dorsi muscle flap: a segmentally inner-
vated muscle transfer for facial reanimation. Microsurgery
1988;9:36--45.
45. Maruyama Y, Nakajima H, Fossati E, Fujino T: Free
latissimus dorsi myocutaneous flaps in the dynamic
reconstruction of cheek defects: a preliminary report.
J Microsurg 1979;1:231.
46. Maruyama Y, Urita Y, Ohnishi K: Rib-latissimus dorsi
osteomyocutaneous flap in reconstruction ofa mandibular
defect. Br J Plast Surg 1985;38:234.
47. Mathes S, Nahai F: Clinical Applications for Muscle and
Musculocutaneous Flaps. Toronto: CV Mosby; 1982.
48. Maves M, Panje W. Shagets F: Extended latissimus dorsi
myocutaneous flap reconstruction of major head and
neck defects. Otolaryngol Head Neck Surg 1984;92:551.
49. Maxwell G: Iginio Tansini and the origin of the latissi-
mus dorsi musculocutaneous flap. Plast Reconstr Surg
1980;65:686.
50. Maxwell G, Manson P, Hoopes J: Experience with thir-
teen latissimus dorsi myocutaneous free flaps. Plast
Reconstr Surg 1979;64: 1.
51. Maxwell G, McGibbon B, Hoopes J: Vascular consid-
erations in the use of a latissimus dorsi myocutaneous
flap after a mastectomy with an axillary dissection. Plast
Reconstr Surg 1979;64:771.
52. Mayou BR, Watson J, Harrison D, Wynn-Parry C: Free
microvascular and microneural transfer of the extensor
digitorum brevis muscle for the treatment of unilateral
facial palsy. Br J Plast Surg 1981;34:362.
53. Mijatovic D, Bulic K. Dzepina I, Unusic J: The supply
of blood in the skin territory above the lower part of the
seratus anterior muscle. CoU. Antrrpo/2006;3:543-54 7.
54. Moscona RA, Ullmann Y, Hirshowitz B: Free composite
serratus anterior muscle-rib flap for reconstruction of the
severely damaged fuot. Ann Plast Surg 1988;20: 167-172.
55. Netscher D, Allford EL, Wigoda P, Cohen V: Free com-
posite myo-osseous flap with seratus anterior and rib:
indications in head and neck reconstruction. Head Neck
1998;20: 106-112.
56. OlivariN:Thelatissimus flap.BrJ PlastSurg 1976;29:126.
57. Olivari N: Use of thirty latissimus dorsi flaps. Plast Recon-
str Surg 1979;64:654.
58. Penfold CN, Davies HT, Cole RP, Evans BT, Hobby JA:
Combined latissimus dorsi-serratus anterior/rib compos-
ite free flap in mandible reconstruction. Inz J Oral Maxil-
kJfac Surg 1992;21:92-96.
59. Pennington D, Stern H, LeeK: Free flap reconstruction
of large defects of the scalp and calvarium. Plast Reconstr
Surg 1989;83:655.
60. Pittet B, Mahajan AI.., Alizadeh N, Schllausraff KU, Fasel
J, Montandon D: The free serratus anterior flap and its
cutaneous component for reconstruction of the face: a
LATISSIMUS DORSI AND SERRATUS ANTERIOR 357
series of 27 cases. Plast Reconstr Surg 2006;117:1277-
1288.
61. Quillen C: Latissimus dorsi myocutaneous flaps in
head and neck reconstruction. Plast Reconstr Surg 1979;
63:664.
62. Quillen C, Shearin J, Georgiade N: Use of the latissimus
dorsi myocutaneous island flap fur reconstruction in the
head and neck area. Plast Reconstr Surg 1978;62: 113.
63. Richards MA, Poole MD, Godfrey AM: The serratus
anterior/rib composite flap in mandibular reconstruction.
Br J Plast Surg 1985;38:466-477.
64. Robson MC, Zachary LS, Schmidt DR, FaibisoffB, Hek-
matpanah J: Reconstruction of large cranial defects in the
presence of heave radiation damage and injection utiliz-
ing tissue transferred by microvascular anastomoses. Plast
Reconstr Surg 1989;83:438-442.
65. Roswell AR, Davies DM, Eisenberg N, Taylor GI: The
anatomy of the subscapular-thoracodorsal arterial sys-
tem: study of 100 cadaver dissections. Br J Plast Surg
1984;37:574-576.
66. Rowsell A, Godfrey A, Richards M: The thinned latis-
simus dorsi free flap: a case report. Br J Plast Surg
1986;39:210.
67. Rubinstein ZJ, Shafir R, Tsur H: The value of angiog-
raphy prior to use of the latissimus dorsi myocutaneous
flap. Plast Reconstr Surg 1979;63:374.
68. Russell R, Pribaz J, Zook E, Leighton W. Eriksson E,
Smith C: Functional evaluation of the latissimus dorsi
donor site. Plast Reconstr Surg 1986;78:336.
69. Schlenker J, Indresano A, Raine T, Meredith S, Rob-
son M: A new flap in the dog containing a vascularized
rib graft-the latissimus dorsi myoosteocutaneous flap.
J Surg Res 1980;29:172-183.
70. Schmidt D, Robson M: One-stage composite reconstruc-
tion using the latissimus myoosteocutaneous free flap.
AmJ Surg 1982;144:470.
71. Schuller D: Latissimus dorsi myocutaneous flap for mas-
sive facial defects. Arch Owlaryngo/1982;108:414.
72. Serra J, Samayoa V, Valiente E, Kloehn G: Neurotization
of the remaining latissimus dorsi muscle following muscle
flap transplant. J Reconstr Microsurg 1988;4:415.
73. Serra MP, Longhi P, Carminati M, Righi B, Robotti E:
Microsurgical reconstruction using a combined flap
composed of serratus anterior myo-osseous flap and
latissimus dorsi myocutaneous flap. J Plast Reconstr Surg
2007;60:1158-1161.
74. Shestak K. Schusterman M, Jones N, Johnson J: Imme-
diate microvascular reconstruction of combined palatal
and midfacial defects using soft tissue only. Microsurgery
1988;9:128.
75. Smith P, Morgan B, Crockard H: Immediate total scalp
and skull reconstruction. Microsurgery 1983;4:23.
76. Stueber K, Saloman M, SpenceR: The combined use
of the latissimus dorsi musculocutaneous free flap and
split-rib grafts for cranial vault reconstruction. Ann Plast
Surg 1985;15:155.
358 CHAPTER ZO
77. Takayanagi S, Tsukie T: Free serratus anterior muscle
and myocutaneous flaps. Ann Plast Surg 1982;8:277-283.
78. Tanaka Y, Miki K, Akamatsu J, Tsukazaki, Inomoto T:
Reconstruction of an extensive scalp defect using split latis-
simus dorsiflap in combination with the serratus anterior
muscul(H)SSeous flap. Br J Plast Surg 1998;51 :250-254.
79. Tansini I: Spora il mio nuovo processo di amputazione
della mammaella per cancre. Rijorma Med (Palermo, Nap-
oli) 1896;12:3.
80. Taylor GI, Palmer J:The vascular territories {angiosomes)
of the body; experimental study and clinical applications.
Br J Plast Surg 1987;40: 113.
81. Terzis J, Manktelow R: Pectoralis minor: a new concept in
facial reanimation. Plast Surg Forum 1982;5.
82. Tobin G, Moberg A, DuBou R, Weiner I.., Bland K: The
split latissimus dorsi myocutaneous flap. Ann Plan Surg
1981;7:272-280.
83. Tobin GR, Schusterman M, Peterson GH, Nichols G,
Bland KI: The intramuscular neurovascular anatomy of
the latissimus dorsi muscle: the basis for splitting the flap.
PlastReconstr Surg 1981;67:637-641.
84. Ueda K, Harashina T, Inoue T, Tanaka I, Harada T:
Microsurgical scalp and skull reconstruction using
a serratus anterior myo-osseus flap. J Craniojac Surg
1992;3:207-212.
85. Ugurlu K, Sacak B, Huthut I, Karsidag S, Sakiz D,
Bas L: Reconstructing wide palatomaxillary defects
using free flaps combining bare serratus anterior muscle
fascia and scapular bone.J OralMaxillojac Surg 2007;65:
621-629.
86. Watson JS: The use of the latissimus dorsi island flap for
intra-oral reconstruction. Br J Plan Surg 1982;35:408.
87. Watson JS, Craig R, Orton C: The free latissimus dorsi
myocutaneous flap. Plan Reconstr Surg 1979;64:299.
88. Watson JS, Lendrum J: One stage pharyngeal reconstruc-
tion using a compound latissimus dorsi island flap. Br
J Plast Surg 1979;64:654.
89. Watson JS, Robertson GA, Lendrum J, Stranc MF,
Pohl MJ: Pharyngeal reconstruction using the latissimus
dorsi myocutaneous flap. Br J Plast Surg 1982;35:401-
407.
90. WhitneyT, Buncke H, Alpert B, Buncke G, Lineaweaver
W:The serratus anterior free muscle flap: experience with
100 consecutive cases. Plast Reconstr Surg 1990;86:481.
91. Yii N, Cronin K: Vascular anatomy of the serratus ante-
rior muscle. Plast Reconstr Surg 2005; 116:680-682.
92. Yla-Kotola TM, Kauhanen SC, Asko-Seljavaara SL:
Facial reanimation by transplantation of a micronneu-
rovascular muscle: long-term follow up. J Plan Reconstr
Surg Hand Surg 2004;38:272-276.
93. Zancolli E, Mitre H: Latissimus dorsi transfer to restore
elbow flexion. J Bone Joint Surg [Am] 1973;55A: 1265.
~e search for sources of expendable bone in other
~ parts of the skeleton for re<:onstruction of mazillo-
mandibular defe<:ts has continued for several decades. Of
any donor site, the ilium has arguably been placed under
the greatest usault for this pw:pose (18) (Figs. 21-1 to
21-3). It has been used as a source of nonvascularized
bone grafts, corticocancellous chips, and vascularized
bone through application of microvascular techniques.
The amount of bone stock, and the ease of harvesting
with a separate sw:gical team, make it an attractive donor
site. Manchester (6) was one of the first to report on the
similarity in shape and curvature of the anterior ilium
to the native hemimmdible. The era of microvascular
surgery and the use of vascularized bone-containing free
flaps brought multiple attemptS to transfer portions of the
ilium based on a variety of different vascular pedicles.The
superlicial cirCUIIlfh:x iliac artery (SCIA) is the vascular
supply to the groin flap, which was one ofthe first free flaps
to be reponed (12).Although the SCIA provides excellent
vascularity to the overlying skin, it has a variable anatomy
and a marginal nutrient ftow to the bone (11). The ilium
was also transferred as a compound flap with the tensor
fasciae latae, which was based on the ascending branch
of the lateral circum1le:z: femoral artery ( 1). The superior
deep branch of the superior gluteal artery has also been
used as the nutrient pedicle for iliac bone transfers (3).
359
360 CHAPTER 21
FIGURE 21-1. The pelvic girdle is formed by 1he paired
coxal or innominate bones, which articulate with the sacrum.
Each coxal bone is composed of three parts: the ilium, the
ischium, and the pubis. The ilium provides a large amount of
bone 1hat can be 1ransferred for reconstructive purposes to
1he maxillomandibular skeleton. The natural curvature of the
iliac crest provides an excellent match to 1he contour of the
hemimandible. The area of vascularized bone that can be
harvested based on 1he DCIA extends from the AS IS (srrow)
to the posterior superior iliac spine (PSIS) (double srrows).
FIGURE 21-2. The medial aspect of the right ilium is composed
largely of the iliac fossa, which is filled anteriorly by the iliacus
muscle. The transversus abdominis, internal oblique, and exter-
nal oblique muscles insert on 1he inner aspect of 1he crest
FIGURE 21-3. A variety of muscles insert on 1he lateral
aspect of the ilium (right ilium shown here). Moving anteriorly
to posteriorly, the sartorius, tensor fasciae latae, external
oblique, internal oblique, and latissimus dorsi muscles insert
along the crest However, 1he gluteus minim us, medius, and
maximus muscles occupy 1he majority of the surface area
of the lateral aspect of 1he ilium. The distance between the
crest and the acetabulum provides a significant amount of
bone height (double srrowt in addition to 1he length that can
be harvested extending from 1he ASIS to the PSIS.
However, it was not until 1979, when two separate
reports by Taylor et al. (16) in Australia and by Sanders
and Mayou (13) in England identified the deep circum-
flex iliac artery (DCIA) and deep circumfiex iliac vein
(DCIV) as the most reliable and most favorable vascular
pedicle for transfer of the ilium. Dye-injection studies by
Taylor et al. (16) revealed both an endosteal and a peri-
osteal supply to the entire iliac bone, extending &om the
anterosuperior iliac spine (ASIS) to the sacroiliac joint.
The DCIA was also found to supply the skin overlying
the ilium through an array of perforators that traverse
the three muscle layers of the abdominal wall (13).
In 1984, experimental work by Ramasastry et al.
(8) identified the ascending branch of the DCIA as the
primary blood supply to the internal oblique muscle.
It therefore became possible to transfer two separate
soft tissue fiaps, the skin and the internal oblique mus-
cle, with the iliac bone based on the DCIA and DCiv.
Ramasastry et al. reported the use of this composite
free fiap for extremity reconstrUction. In the latter
part of the 1980s, we became increasingly dissatis-
fied with the skin paddle of the iliac osteocutaneous
 ILIAC CREST OSTEOCUTANEOUS AND OSTEOMUSCULOCUTANEOUS 361

flap because of its hulk, its limited maneuverability
relative to the bone, and its tenuous blood supply. In
1989, my colleagues and I reported the adaptation of
the iliac osteomusculocutaneous-intemal oblique flap
to oromandihular reconstruction (19,20). The major
advantage that the inclusion of the internal oblique
muscle brought to this composite flap was its increased
maneuverability and decreased bulk, which enhanced
its utility in intraoral reconstruction (Fig. 21-4).While
body habitus has an impact on the thickness of the
overlying skin, the caliber of the internal oblique mus-
cle is not affected.

FIGURE 21-4. The internal oblique muscle arises from the thoracolumbar fascia, the iliac crest
and the lateral aspect of the inguinal ligament. It inserts into the linea semilunaris of the rectus
sheath and the 10th rib through the 12th rib. Harvest of the internal oblique muscle involves inci-
sions along the linea semilunaris, the costal margin, and the anterior axillary line (dotted line).
362 CHAPTER 21
Additional soft tissue flaps have been transferred
with the iliac bone. The anterolateral thigh flap has been
incorporated into this flap, but a second set of anas-
tomoses is required to vascularize the skin (5) . Alter-
natively, an iliac crest musculoperitoneal flap was used,
with the blood supply to the layer ofperitoneum derived
from the ascending branch of the DCIA (4). However,
both of these flaps have received little attention since
their initial reports.
FLAP DESIGN AND UTILIZATION
Although the DCIA can be used to transfer the skin of the
groin and the internal oblique muscle, these soft tissue
flaps are rarely used independent ofthe iliac bone because
of the large number of alternative soft tissue donor sites
from which to choose. The design of the skin paddle usu-
ally assumes a fusiform shape to facilitate donor-site clo-
sure. There is little variability in the skin flap aside from
the size and position on the abdominal wall. The size of
the skin paddle must be large enough to incorporate a
sufficient number of musculocutaneous perforators. Its
maximum size is not clearly defined, but primary closure
of the abdominal wall usually determines the upper limits
of the skin that can be harvested. The relationship of the
skin to the bone is somewhat fixed by the array of perfo-
rators that exit the external oblique just cephalad to the
ilium. The skin flap achieves greater mobility relative to
the bone when designed in a more cephalad location on
the abdominal wall (21), provided that its inferior portion
incorporates the zone of the cutaneous perforators (17).
The internal oblique muscle is usually harvested in
its entirety, pedicled inferiorly on its attachment to the
inner table of the iliac crest. In 80% of cases, there is a
single dominant ascending branch that allows the sur-
geon to isolate the entire muscle solely on its nutrient
vascular supply (Fig. 21-5).
The rich vascularity of the ilium allows great flexibil-
ity in the size and shape of the segment of bone that is
harvested. The iliac bone is composed of a thick layer of
cancellous bone that is sandwiched between two layers of
cortical bone. The amount of bone, based on the cross-
sectional surface area, was determined to be greater than
the fibula, scapula, and radius (7).There is a wide range of
bone orientations that can alter the position of the flap's
pedicle, depending on the location ofthe recipient vessels.
The natural curvature of the ilium must be accounted
for in designing and planning the bone to be harvested.
Height may be added to the bone by extending the oste-
otomies deeper into the body of the ilium. According to
the principles established by Manchester (6), the ASIS
can be used to fashion the angle of the neomandible, and
the ramus and condyle may be formed by extending the
bone cuts to the anteroinferior iliac spine (Fig. 21-6).
The iliac bone may be further contoured to match the
curvature of the symphysis by making osteotomies. These
osteotomies result in wedge-shaped openings in the bone
that must be packed with corticocancellous bone chips.
Alternatively, closing ostectomies can be performed by
removing wedge-shaped segments of the ilium, to col-
lapse the bone and match virtually any defect's shape.
The vascularity to the distal segments is preserved by
maintaining the integrity of the inner periosteum and the
DCIA-DCIV pedicle, which lies in close proximity to the
inner table of bone (Fig. 21-7). Finally, unicortical seg-
ments of vascularized ilium may be harvested by making
a sagittal cut through the crest. By preserving the outer
cortical layer of the ilium, the attachments of the upper
thigh muscles remain undisturbed, which presumably
helps to reduce postoperative morbidity, including pain,
and to preserve normal hip contour (Fig. 21-8). How-
ever, I have found it advantageous to harvest bicortical
segments of ilium to achieve a more functional mandibu-
lar and palatomaxillary reconstruction that can receive
endosteal osteointegrated dental implants.
The most common applications of the iliac composite
flap in head and neck surgery is the reconstruction of seg-
mental mandibular and palatomaxillary complex defects.
We have applied this composite flap to midface defects,
especially those that extend to involve the suprastructure
of the maxilla, where the vertical height of the bone can
be restored with a block of bone from the ilium. Up to 16
em of bone length can be harvested, which is suitable for
the restoration of most ablative defects of the lower jaw.
Subtotal or total mandibular reconstruction requires the
use of the fibula. There is a tremendous amount of versa-
tility in the use of the soft tissue components of the iliac
composite flap. The osteocutaneous flap may be placed
with the crest as the neoridge and the skin situated on
top of the crest in its normal anatomic relationship to the
bone (Fig. 21-9). Alternatively, the crest may be placed
at the inferior aspect of the neomandible, leaving the cut
surface of the cancellous bone to form the new alveolar
ridge. In the latter orientation, the skin must be tunneled
on the lingual or the buccal aspect of the neomandible.
In virtually all cases, the bone is oriented with the DCIA
and DCIV on the lingual aspect of the neomandible so
that rigid fixation plates can be safely applied to the buc-
cal cortex. In virtually all cases in which the skin of the
osteocutaneous flap is placed in the oral cavity, second-
ary procedures are required to debulk the subcutaneous
tissue or replace it with a skin graft.
The tripartite osteomusculocutaneous flap provides
an added dimension for oromandibular reconstruction.
The mobility of the internal oblique muscle, by virtue of
its axial pattern blood supply in 80% of cases, allows the
surgeon the freedom to position it in three dimensions
relative to the bone (Fig. 21-10). In most cases, the inter-
nal oblique is placed intraorally and wrapped around
the neomandible or transposed posteriorly to resurface a
portion ofthe pharynx. A split-thickness skin graft placed
 ILIAC CREST OSTEOCUTANEOUS AND OSTEOMUSCULOCUTANEOUS 363

Ascending Ascending
branch branches

DCIA DCIA

65% 20%

Ascending
branch

DCIA

15%
FIGURE 2.15. Two separate series of cadaver dissections revealed almost identical incidence
rates for the three different anatomic patterns of the ascending branch {9,17). In 65% of cases,
the ascending branch originates from the DCIA within 1 em medial to the AS IS. In 15% of cases,
the ascending branch originated in a more medial location, i.e., 2 to 4 em from the ASIS.In the
final group, constituting 20% of dissections, there was no single dominant ascending branch.
Instead, the internal oblique muscle was supplied by a series of smaller branches arising at
various points along the course of1he DCIA.
364 CHAPTER 21

RGURE 21-6. Different shapes of the iliac bone may be harvested, depending on which portion of
the mandible must be reconstructed and where 1he recipient vessels are located. The variations
become even more numerous when we consider1hat the bone may be oriented with 1he crest
either along the neoridge or along the inferior border of1he new mandible. There are1hree different
bone designs 1hat have been demonstrated in 1hese illustrations 1hat reflect a number of different
options 1hat are available for reconstructing the hemimandible. A: Ahemimandible designed from
the ipsilateral hip provides a neomandible of appropriate curvature and the pedicle exiting at the
angle. The height of the neomandible is obtained by extending 1he height of bone down to include
the anterior inferior iliac spine {AilS). B: By selecting a flap from the contralateral side of the pelvis,
the hemimandible can be restored with the pedicle exiting near the midline. In 1his instance, the
vertical height of the neomandible is harvested by extending the depth of1he osteotomy more pos-
teriorly along the iliac bone. C; When the defect is limited to the ramus and condyle, the design can
be changed so that the crest is used to achieve appropriate mandibular height.
 ILIAC CREST OSTEOCUTANEOUS AND OSTEOMUSCULOCUTANEOUS 365

7
I r
'- ...... {
.._
'I
I

''

\ I
\
,.. _,
I
,
'- / -, I
I ' \
I
I

r
I
I ,(
"'-...<-:
''

c
FIGURE 21-7. Several different orientations of1he iliac bone may be harvested, based on the
demands of the mandibular defect. A:. When symphyseal defects are present, a straight piece
of iliac bone is harvested 1hat requires osteotomies to achieve a gentle curvature that matches
the anterior mandible. B: This same principle applies when extending the defect from the ramus
or condyle across the midline, in which case the length of bone to be harvested extends more
posteriorly. The ramus, condyle, and body of the mandible rarely require significant contouring
because of1he similarity in shape of the pelvis to 1he hemimandible. However, distal osteoto-
mies are needed to achieve a gentle curvature of the central mandible. C: The iliac bone may
also be placed as a platform in a horizontal position to reconstruct the symphysis and the
tongue in patients who have undergone composite resections 1hat involve the anterior mandible
and significant or total glossectomies. Placement of the bone in a horizontal position allows the
skin to be supported against the effects of gravity while duplicating the gentle curvature of the
symphysis. This is achieved by removing a V-shaped ostectomy of the ilium, which allows the
bone to be collapsed and,1herefore, produces 1he desired curvature.
36& CHAPTER 21

FIGURE 21-8. The inner table of the iliac bone may be harvested by splitting the cortex in the
sagittal orientation and transferring only the inner cortex of the bone with the muscle cuffs that
protect the DCIA. This can then be used to reconstruct the mandible as shown. In the process of
insetting the bone, the cancellous portion is oriented on the buccal aspect of the mandible. The
advantage of this bone design is that it preserves the attachments of the upper thigh muscles,
which facilitates the rehabilitation of the hip in the postoperative period.

over the muscle serves as a primary vestibuloplasty to
restore the anatomy of the gingivobuccal sulci and to
maintain the mobility of the tongue (14).The denervated
internal oblique muacle undergoes atrophy and provides
a well-vascularized, thin, immobile tissue layer over the
neomandible. In situations in which the mucosal defect
is limited to the gingiva, then the internal oblique mus-
cle can be left bare to remucosalize and, therefore, avoid
placement of a skill graft. This flap is ideal for the recon-
struction of composite defects that involve mucosa, bone,
and external skin. The skin of this osteomusculocutane-
ous flap is well positioned to resurface cutaneous defe<:ts
of the neck and lower face. As it hangs from the lower
border of the neomandible, the skin is in a favorable posi-
tion to avoid torsion or tension on the nutrient perfo-
rators (Fig. 21-11). When the cutaneous defect extends
above the level of the lateral commissure, then the use of
the iliac skin paddle becomes unfavorable because of the
detrimental dfect on its vascular supply. An alternative
solution to this complex defect is to use the subscapular
system of flaps, which has muimum mobility of the soft
tissue components relative to the bone (20).
The reconstruction of the tongue of a patient who
requires a total or subtotal glossectomy is challenging.
 ILIAC CREST OSTEOCUTANEOUS AND OSTEOMUSCULOCUTANEOUS 367

-
I .. I

FIGURE 21-9. The iliac osteocutaneous flap may be inset with the crest forming 1he ridge of
the neomandible. This orientation places the skin in its normal anatomic position relative to the
bone and, therefore, avoids any torsion on 1he delicate musculocutaneous perforators. This flap
configuration is suitable for use in patients who have very thin subcutaneous tissue overlying
the inguinal region. The selection of the appropriate donor hip is determined by1he location of
the recipient vessels. When this osteocutaneous flap is harvested from the contralateral hip, the
vessels exit at the angle of the neomandible. When the ipsilateral hip is utilized, the pedicle exits
nearer the midline, in closer proximity to recipient vessels in the contralateral neck.

The problem is even greater when the anterior man-
dible is also resected. The combination of a total glos-
sectomy defect with a lateral mmdibulectomy can be
managed well with a soft tissue flap and a rigid mandib-
ular reconsttuction plate to maintain the contour of the
lateral aspect of the lower jaw. However, when the bo:ay
defect involves the region of the symphysis, a vascula.r-
ized bone graft should be utilized. Salibian et al. (11,12)
introduced a unique design for the iliac osteocutaneous
flap in which the bone was placed in the floor of the
mouth in a horizontal position to act as a bony shelf
to support the overlying soft tissue flap. The skin flap
of the osteocutaneous flap was transposed into the oral
cavity to reconstruct the tongue. As originally reported,
both the DCIA and the SCIA were revascularized to
ensure the blood supply to the skin. In our experience, it
has only been necessary to anastomose the DCIA when
using this flap design. This may be the result, in part, of
the trapezoidal bone that was described by Salibian et
al. As an alternative, we have harvested a longer piece
of the ilium, which is then contoured to the shape of
the anterior arch and placed as a horizontal platform
by removing a triangular wedge of bone that comes to
a point at the crest (Fig. 21-7C). The adwntage of this
modification is that a longer course of the DCIA and
DCIV is harvested, which ensures a greater number
of perforators. Long-term follow-up of patients who
have undergone this type of reconsttuction has shown
excellent maintenance of the height of the neotongue
because of the support provided by the bo:ay platform.
The iliac crest has also been used for the reconstruc-
tion of hard palate defects followiD,g which the insertion
of endosteal implants has allowed functional dental res-
toration comparable to that of the lower jaw (8). Our
approach to palatomazillary defect reconstruction has
evolved over the last decade with the iliac crest-internal
oblique muscle flap utilized, especially when the vt:rtical
dimensions of the maxilla, orbital rim, and orbital floor
must be reatored (Fig. 21-56). Infrastructure palatoma:DJ.-
lary de~cts are very readily reconstructed with the fibular
free flap. The inset of the iliac crest-internal oblique into
the midface is demonstrated in Figures 21-57 through
21-63. Finally, the iliac crest-internal oblique flap has
been utilized in the restoration of skull base defects (10).
368 CHAPTER 21

Transversus

.:.---c..:::.,..,..-;~ -.~---------External
otilique m.

Iliac bone
RGURE 2110. The tripartite iliac internal oblique osteomusculocutaneous flap is shown in its nor-
mal anatomic position following harvest from 1he left side of the pelvis. Cuffs of the iliacus, transver-
sus abdominis, and external oblique are harvested with the internal oblique muscle to protect the
perforators to the skin. When this flap is inset into the oral cavity for oromandibular reconstruction,
it is usually turned 180 degrees so 1hat1he crest forms the inferior border of 1he neomandible.

FIGURE 21-11. The combined iliac crest-internal oblique flap transfers a skin paddle and 1he
broad sheet of 1he internal oblique muscle. The selection of 1he appropriate hip is determined
by the availability of the recipient vessels in the neck. When the recipient artery and vein are
located in the contralateral neck, then harvest of 1he flap from the contralateral hip places the
vessels at a more favorable location closer to 1he midline. Alternatively, when the recipient ves-
sels are located in the ipsilateral neck, this requires a flap to be harvested from the ipsilateral
hip, which allows orientation of the vessels atthe angle of the mandible.
 ILIAC CREST OSTEOCUTANEOUS AND OSTEOMUSCULOCUTANEOUS
NEUROVASCULAR ANATOMY
The DCIA arises from the lateral aspect of the exter-
nal iliac artery approximately 1 to 2 em cephalad to the
inguinal ligament. The deep inferior epigastric artery
arises just opposite the DCIA on the medial aspect of
the external iliac artery. The DCIA courses toward the
ASIS enclosed within a fascial envelope that consists of
the fasciae of the transversalis and iliacus muscle. In its
more lateral curvilinear course, it travels along the inner
table of the ilium in the groove formed by the junction
of the transversus abdominis and the iliacus muscles.
That groove is located from 0.4 to 2.2 em inferior to
the inner lip of the iliac crest (2). That range of distance
between the DCINDCIV and the top of the iliac crest
is very important because it determines the minimum
height of bone that is required in order to ensure cap-
ture of the nutrient vascular supply to the flap. Dur-
ing its course, it gives off the ascending branch, which
supplies the internal oblique muscle and the periosteal
and endosteal perforators to the ilium. It supplies the
overlying skin through a series of branches that traverse
the three layers of the abdominal wall. The DCIA ter-
minates as a dominant skin feeder approximately 9 to
10 em posterior to the ASIS (13) (Fig. 21-12).
A series of cadaveric dissections revealed the diam-
eter of the DCIA to be 2 to 3 mm; the DCIV ranged
from 3 to 5 mm (2). The length of the "free" portion of
the DCIA, which extends from the ASIS to its junction
with the external iliac artery, is between 5 and 7 em (2).
The DCIV may be several centimeters longer because
it usually runs a longitudinal, cephalad course prior
to entering the external iliac vein. The DCIV is usu-
ally composed of two paired venae comitantes, which
merge at a variable point before entry into the exter-
nal iliac vein. The DCIV receives a consistent ascend-
ing branch proximal to its junction with the external
iliac vein. This branch must be transected to achieve the
maximum length of the DCIV pedicle. This branch also
marks the point where the DCIV makes its turn from
a transverse orientation to a more cephalad one. The
DCIV may pass either superficial or deep to the exter-
nal iliac artery and, therefore, diverges from the DCIA
in its medial portion.
There are two important factors that the surgeon
must keep in mind to ensure preservation of the blood
supply to the skin. The first is related to the flap design,
which must capture the three to nine perforators that
exit the external oblique muscle in a zone that extends
approximately 9 em posterior to the ASIS and about
2.5 em medial to the crest. This zone of perforators can
be readily incorporated by designing a skin paddle cen-
tered along an axis drawn from the ASIS to the inferior
border of the scapula. The second major technical fac-
tor is the preservation of the obligarory cuff of external
oblique, internal oblique, and transversus abdominis
369
muscle layers through which these perforators traverse.
Although the perforators can be identified on the under-
surface of the skin, I usually maintain a healthy distance
from the inner table of the ilium to avoid these vessels. It
is this arrangement of very small cutaneous perforators,
located in this zone, that makes it imperative to main-
tain the relationship of the skin to the bone to avoid
twisting or stretching these vessels.
The internal oblique muscle is a thin broad muscle
that lies between the transversus abdominis and the
external oblique muscles on the anterior abdominal
wall. It originates from the thoracolumbar fascia, the
iliac crest, and the inguinal ligament. It inserts into the
1Oth through the 12th ribs and the rectus sheath (Fig.
21-4). The ascending branch of the DCIA provides the
dominant blood supply to the internal oblique. How-
ever, there are contributions from branches of the deep
inferior epigastric artery and branches of the lower tho-
racic and lumbar arteries that run in the same neurovas-
cular plane between the transversus abdominis and the
internal oblique. The ascending branch is a large vessel
measuring 1 to 2 mm in diameter that arises from the
DCIA and courses through the transversus abdominis
to reach the undersurface of the internal oblique. It is
important to reiterate that the ascending branch does
not contribute to the cutaneous circulation. In two inde-
pendent cadaver dissections, the anatomy of the ascend-
ing branch was categorized into three distinct groups
(8,16). In approximately 65% of cases, the ascending
branch arose from the DCIA within 1 em of the ASIS.
In 15% of dissections, the ascending branch had a more
medial takeoff between 2 and 4 em from the ASIS. In
the remaining 20% of cases, there were multiple smaller
branches that entered the muscle lateral to the ASIS.
The important conclusion from these statistics is that
approximately 80% of internal oblique muscles are sup-
plied by a single dominant vessel arising medial to the
ASIS. The surgeon therefore has the freedom to treat the
internal oblique muscle as a flap with an axial pattern
blood supply in these patients (7,13). In the remain-
ing patients, the muscle must remain attached to the
inner lip of the iliac crest to preserve the smaller, more
numerous feeders from the DCIA.
The internal oblique muscle has a segmental nerve
supply from the lower thoracic (T8 to T12), iliohy-
pogastric (L1), and ilioinguinal nerves (L1). It is there-
fore not a frontline muscle to reinnervate and use in a
dynamic reconstruction. The lateral femoral cutaneous
nerve exits from the pelvis, running a course medial to
the ASIS, either superficially or deep to the DCIA and
DCIV. This nerve may be preserved through meticulous
dissection. However, I often harvest a portion of this
nerve as a free nerve graft for bridging nerve defects in
the head and neck, such as the inferior alveolar nerve.
Finally, the femoral nerve runs in a deeper plane lat-
eral to the external iliac artery and vein. Although rarely
370 CHAPTER 21

Internal oblique m.
(reflected)

DCIA

Femoral
and artery

RGURE 2112. The DCIA and DCIV originate from the external iliac vessels a few centime-
ters above the inguinal ligament. Three separate approaches to the DCIA and DCIV have been
described. The first is a caudal approach that begins below the inguinal ligament by dissecting
proximally along the plane of the femoral vessels underneath the inguinal ligament. The sec-
ond approach is transinguinal, through the posterior wall of the inguinal canal. Finally, the third
approach utilizes the ascending branch as a guide to dissect in a retrograde fashion to its takeoff
from the DCIA. The DCIA and DCIV run a relatively straight course from the external iliac artery and
vein to the ASIS. From the AS IS, they continue laterally in a curvilinear fashion in the groove formed
by the junction of the iliacus and the transversus abdominis.ln its course, the DCIA gives off the
ascending branch, which runs through the transversus abdominis to supply the internal oblique
muscle. It also supplies the iliac bone through endosteal and periosteal feeders. Musculocutane-
ous perforators are given off in an array along the inner table ofthe iliac crest. These perforators
must traverse the three muscle layers of the abdominal wall before supplying the skin. The relation-
ship of the femoral nerve to the external iliac artery and vein should be noted. The lateral femoral
cutaneous nerve crosses medial to the AS IS, either superficial or deep to the DCIA-DCIV pedicle.
 ILIAC CREST OSTEOCUTANEOUS AND OSTEOMUSCULOCUTANEOUS
identified during the course of the dissection, its posi-
tion should be noted to be certain that injury is avoided
during the closure. To date, there have been no reports
of restoring sensation to the skin paddle of the iliac crest
composite flap.
ANATOMIC VARIATIONS
Aside from the variable position and number of
branches of the ascending branch, the major anatomic
variations of this donor site are few and infrequent. In
the multiple series of cadaver dissections noted earlier
and my own experience with more than 300 iliac crest
free flaps, there have been no cases of an absent DCIA
or DCN. However, in a small number of dissections,
Taylor and Daniel (15) found that the DCIA passed
through the transversus abdominis medial to the ASIS,
assuming a more superficial position. It is imperative
that this anomalous course does not cause the surgeon
to mistake the DCIA for the ascending branch.
In my experience, the ascending branch has a sepa-
rate takeoff from the external iliac artery in about 5%
of cases. Included in this group are the cases in which
the ascending branch arises from the DCIA within 1 to
2 mm of the external iliac artery. In such cases, these
two vessels must be treated as separate pedicles. I have
also encountered the rare anomaly of duplication of the
DCIA. In these cases, I placed a temporary microvas-
cular clamp on each artery to determine their relative
dominance before deciding which branch to anasto-
mose in the neck. When the results of this test were
inconclusive, I performed a double arterial anastomosis.
Duplication of the DCIV has been encountered in only
a small number of cases. However, when the two venae
comitantes join within 1 to 2 mm from the external iliac
vein, the ligation of that vessel often creates the neces-
sity to perform two separate venous anastomoses.
POTENTIAL PITFALLS
There are several potential problems that may be
encountered in using composite flaps from the ilium.
Perhaps the greatest concern is related to the integrity
of the abdominal wall. The presence of a weakness or
hernia prior to surgery may suggest that an alternative
donor site be used or that the surgeon should be alerted
to the need for additional measures in donor-site clo-
sure. On occasions where the transversus abdominus
muscle is very thin, following harvest of the internal
oblique muscle, I have utilized a synthetic mesh to aug-
ment the closure of the abdominal wall. The details of
371
the inset of that mesh are illustrated in Figures 21-38
and 21-39. As in all aspects of reconstructive microsur-
gery, meticulous technique and attention to detail are
critical factors in producing a successful outcome. In
addition to frank hernia formation, the harvest of the
internal oblique muscle causes denervation of the rec-
tus abdominis through interruption of its motor nerve
supply, which runs in the neurovascular plane between
the internal oblique and the transversus abdominis.
These nerves are readily identified and can be preserved
through careful dissection. In my experience, incisional
hernias are rare as are patient complaints related to
abdominal wall bulging. Attention must also be paid to
vital neighboring structures, such as the femoral nerve
and the intraperitoneal contents, to prevent injury
during harvest or closure.
Although atherosclerotic involvement of the external
iliac artery is fairly common, I have not encountered
similar problems in the DCIA. There have been no
cases in my experience in which this flap was abandoned
intraoperatively because of atherosclerosis. There was
one case in my experience in which the DCIV was too
small for transfer of this flap, which was then abandoned
and an alternative donor site used. The blood supply to
the soft tissue components of this composite flap may
be compromised by a number of factors. Prior surgery,
such as an appendectomy, may lead to direct damage
of the ascending branch of the internal oblique muscle.
The blood supply to the skin may be at risk as a result of
(a) failure of flap design to incorporate a sufficient num-
ber of perforators, (b) failure to preserve an adequate
muscle cuff of the three abdominal wall muscles, and (c)
altering the relationship of the skin paddle to the bone,
causing tension or kinking of these perforators.
POSTOPERATIVE CARE
Postoperative care of the donor site following harvest
of the iliac crest osteocutaneous or osteomusculocu-
taneous flaps involves progressive mobilization, which
begins on the third or fourth postoperative day. In the
early postoperative period, the knee that is ipsilateral to
the harvest is elevated on a pillow in order to alleviate
the tension on the abdominal wall closure. In addition,
enteral nutrition is delayed until the patient recovers
bowel function as a result of the usual postoperative
ileus. Assisted ambulation usually begins by the seventh
postoperative day along with passive and active range-
of-motion exercises. Progressive assisted ambulation is
carried out during the second postoperative week. Stair
climbing does not usually begin until the third week
after surgery.
312 CHAPTER Z1

Iliac Crest-Internal Oblique Flap

FIGURE 21-13. Harvest of the iliac crest-

internal oblique osteomusculocutaneous flap
is accomplished by modification of the original
technique that was described for harvesting the
iliac crest osteocutaneous flap. The approach
to the vascular pedicle that I use when harvest-
ing the internal oblique muscle serves as the
basis for the approach to the vascular pedicle
when the osteocutaneous flap is harvested.
I begin with a dissection of the internal oblique
muscle.

FIGURE 21-14. The dissection begins by incis-

ing along the cephalad limit of the skin paddle,
which is marked by drawing a fusiform skin
island. This skin island is usually centered along
an axis that runs from the anterior superior iliac
spine to the inferior border of the scapula. It is
essential to capture the dominant musculocuta-
neous perforators that exit through a zone that
extends from approximately 2.5 em medial and
cephalad to the crest and runs approximately
9 em posterolatera lly from the anterior superior
iliac spine. The incision extends through the
skin and subcutaneous tissue to expose the
external oblique muscle muscle and aponeu-
rosis. The skin may be elevated off the external
oblique muscle until approximately 2 to 2.5 em
from the medial aspect of the crest.
 ILIAC CREST OSTEOCUTANEOUS AND OSTEOMUSCULOCUTANEOUS 373

Iliac Crest-Internal Oblique Flap

FIGURE 2115. By careful dissection in this

plane between the subcutaneous tissue and
the external oblique muscle, the surgeon can
usually identify dominant musculocutaneous
perforators, as shown in this dissection. These
perforators must be preserved to maintain the
vascularity to the skin. A dotted line has been
drawn in the direction of the fibers of the exter-
nal oblique muscle, approximately 2.5 to 3 em
above the iliac crest and marks the incision 1hat
is made through the external oblique muscle
and fascia.

FIGURE 2116. The external oblique muscle

has been incised throughout the full extent of
the wound, as previously noted. This cuff of
the external oblique muscle, which measures
approximately 2.5 em in width, should not be
elevated in a caudal direction off the internal
oblique muscle to avoid interfering with the
perforators that come through the three muscle
layers of the abdominal wall.
374 CHAPTER Z1

Iliac Crest-Internal Oblique Flap

FIGURE 21-17. The entire internal oblique

muscle is exposed by elevating the external
oblique muscle up to the level ofthe costal mar-
gin by blunt and sharp dissection. This plane of
dissection is relatively avascular.

FIGURE 21-11. Complete exposure of the

internal oblique muscle has been obtained.
The 12th rib can be easily palpated. A dotted
line has been pia ced along the margins of the
internal oblique. Incisions are made around the
perimeter of the internal oblique muscle prior
to elevating it off the transversus abdominis
muscle. The inferior attachments of the muscle
to the crest are left undisturbed.

FIGURE 21-19. The plane of dissection

between the internal oblique and the transver-
sus abdominis is most easily identified in the
region just caudal to the 12th rib. In this location,
the division between the two muscle layers is
the most well defined. A change in the directian
of the muscle fibers signifies the correct inter-
muscular plane. Meticulous dissectian in this
plane is required ta elevate the internal oblique
muscle completely from cephalad to caudad.
 ILIAC CREST OSTEOCUTANEOUS AND OSTEOMUSCULOCUTANEOUS 375

Iliac Crest-Internal Oblique Flap

FIGURE 21-20. Elevation of the internal oblique

muscle has been completed. In the process
of elevating this muscle, the integrity of the
transversus abdominis muscle and aponeurosis
must be maintained. In addition, meticulous
dissection is necessary to identify and maintain
the ascending branch (lsrge srrowhesd) of the
DCIA, which is the nutrient blood supply to the
internal oblique muscle. Segmental neurovas-
cular bundles traverse the abdomen laterally
to medially between the internal oblique and
the transversus abdominis (small arrowheads)
muscles. The cephalad neurovascular bundles
can be maintained. However, in the more
caudal locations, the terminal branches of
the intercostal neurovascular bundles inter-
twine with the ascending branch and must be
transsected. There are often interconnecting
branches between the ascending branch and
the segmental branches that must be carefully
coagulated using bipolar cautery. Unipolar
cautery should be avoided in this location.

FIGURE 21-21. A close-up view of the under-

surface of the internal oblique muscle reveals
the ascending branch (blsck srrowt of the
DCIA. As the dissection is followed inferiorly
and medially, multiple tributaries converge to
the main trunk of the ascending branch, which
then travels through the transversus abdominis
muscle layer (blue srrowt to join with the DCIA
and DCIV.
376 CHAPTER Z1

Iliac Crest-Internal Oblique Flap

FIGURE 2122. The junction of the ascending

branch with the DCIA occurs most often in the
region medial to the ASIS. The DCIA and DC IV
(small arrowheads) are then traced to their
junction with the external iliac artery and vein
(large arrowhead). The preperitoneal fat must
be retracted medially and cephalad to gain
exposure of the iliacus.

FIGURE 21-23. The lateral portion of the trans-

versus abdominis muscle is then transsected,
leaving a 2 em cuff attached to the inner table
of the iliac crest. The exposure afforded by this
maneuver provides access to the iliacus and
to the lateral femoral cutaneous nerve (arrow-
heacft. The DCIA and DC IV run in a fibrous tunnel
created by convergence of the fasciae of the
transversalis and the iliacus. The more lateral
course of the DCIA and DCIV runs in the groove
between the iliacus and the transversus abdami-
nis muscles. Although the pulse ofthe DCIA can
be palpated in this groove, there is no reason to
expose the vascular pedicle lateral to the AS IS.
 ILIAC CREST OSTEOCUTANEOUS AND OSTEOMUSCULOCUTANEOUS 317

Iliac Crest-Internal Oblique Flap

FIGURE 2124. The iliacus is transsected to

expose the inner table of the ilium. A 2-cm cuff
of iliacus (arrowheads) is left attached to the
inner table as protection for the DCIA and DCIV.

FIGURE 2125. The lateral dissection now

begins by incising along the inferolateral border
of the skin paddle to the level of the tensor
fasciae latae and the tendon of the gluteus
medius.
378 CHAPTER Z1

Iliac Crest-Internal Oblique Flap

FIGURE 21-26. Sharp dissection along the

entire lateral border of the iliac crest provides
exposure to the outer table of the ilium to
perform the osteotomies.

FIGURE 2121. The dissection is viewed from

the vantage point of the midabdomen, looking
directly at the ASIS. The internal oblique muscle
has been completely elevated, and the skin
paddle sits in its normal anatomic position atop
the iliac crest The lateral thigh muscles have
been transsected to provide exposure of the
outer table of the ilium.
 ILIAC CREST OSTEOCUTANEOUS AND OSTEOMUSCULOCUTANEOUS 379

Iliac Crest-Internal Oblique Flap

FIGURE 21-28. The OCIA and OCIV (arrow-

hesds) must be dissected free of the surround-
ing soft tissues in the region between their
takeoff from the external iliac artery and vein
to the ASIS. Meticulous dissection around the
AS IS determines whether the lateral femoral
cutaneous nerve runs a course superficial or
deep to the DCIA and DC IV. Careful transsection
of the iliopsoas and sartorius muscles along the
medial aspect of the ilium completes the soft
tissue dissection, all the while protecting 1he
DCIA and DCIV.
380 CHAPTER Z1

Iliac Crest-Internal Oblique Flap

Gluteus medius m.

Tensor fascia lata m.

us femoris m.

Sartorius

FIGURE 21-29. The dotted line marks the cuts that are made to gain exposure of the iliac bone.
A 2 to 3-cm cuff of iliacus is left attached to the ilium to protect the DCIA. Medial exposure
of the ilium is made by cutting the iliopsoas muscle with the DCIA and DCIV completely mobi-
lized and protected in that region. The gluteus medius, sartorius, and tensor fasciae latae are
removed flush with the outer table of the ilium. Alternatively, if a unicortical bone graft is to be
harvested, then the upper thigh muscles are left attached at this point in the dissection, and
bone cuts are made sagittally through the crest and transversely through the inner table.
 ILIAC CREST OSTEOCUTANEOUS AND OSTEOMUSCULOCUTANEOUS 381

Iliac Crest-Internal Oblique Flap

FIGURE 21-30. While protecting the abdominal

contents by using deep retractors, osteotomies
(srrowhesds) are made from the lateral
exposure. The protection of the DCIA and DCIV
must be maintained while making the medial
bone cuts.

FIGURE 21-31. The osteotomy has been

completed and the bone segment that is to be
included in this composite flap is now freed.
382 CHAPTER Z1

Iliac Crest-Internal Oblique Flap

FIGURE 21-32. From a medial vantage point,

the composite flap has been rotated
180 degrees while still attached to the external
iliac artery and vein. Harvesting of this flap is
completed by a double ligation of the DCIA and
a single ligation of the DCIV.

FIGURE 21-33. The completely harvested

internal oblique iliac crest osteomusculocuta-
neous flap with its vascular pedicle is shown
from a lateral view. A bicortical segment of
bone with the upper thigh muscles completely
freed from the lateral table of the crest is
demonstrated.
 ILIAC CREST OSTEOCUTANEOUS AND OSTEOMUSCULOCUTANEOUS 383

Iliac Crest-Internal Oblique Flap

FIGURE 21-34. Examination of the medial

aspect of this composite flap reveals the under-
surface of the internal oblique muscle with the
ascending branch (smsllsrrowhesd). The cuff
of the transversus abdominis is demonstrated
by the double small arrowheads. The cuff of the
iliacus is demonstrated by the lsrge arrowhead.
The cuff of the external oblique muscle and
aponeurosis, which lies superficial to the inter-
nal oblique muscle, is not visualized.

FIGURE 21-35. The use of the internal oblique

muscle as coverage of the bone is demon-
strated by wrapping the sheath of muscle over
the cut margin of the iliac bone. Using this
technique, the ascending branch runs on the
undersurface of the muscle adjacent to the
bone.
384 CHAPTER Z1

Iliac Crest-Internal Oblique Flap

FIGURE 21-36. In the majority of cases. a

single dominant ascending branch provides
an axial blood supply to the internal oblique
muscle. Further mobilization of the internal
oblique muscle may be achieved by making
a back cut parallel to the bone, from laterally
to medially, while preserving a cuff of muscle
around the ascending branch. It is imperative to
maintain a 2- to 2.5-cm cuff of internal oblique
muscle adjacentto the iliac bone while making
this back cut to protect the integrity of the mus-
culocutaneous perforators that are traversing
the three layers of the abdominal wall.

FIGURE 21-37. Closure of the abdominal wall

must be completed in a meticulous fashion to
prevent a ventral hernia. After copious irriga-
tion and hemostasis, the first layer of closure
is achieved by approximating the transversus
abdominis muscle to the iliacus muscle (yellow
arrows). To reinforce this layer of the closure,
drill holes may be placed along the cut margin
of the iliac bone through which sutures are
placed that traverse the iliacus and the trans-
versus abdominis. This technique is designed to
enhance the purchase of the sutures above that
that can be achieved with muscle-to-muscle
approximation. It is essential to note the posi-
tion of the femoral nerve, which lies laterally to
the femoral artery. Deep sutures in this region
must be avoided to prevent injury to that nerve.
 ILIAC CREST OSTEOCUTANEOUS AND OSTEOMUSCULOCUTANEOUS 385

Synthetic Mesh Closure of the Abdominal Wall

.P-:-1-"f--+- - - Exemal
oblique
reflected

~~~~-------Tre~w~s
Linea
abdomlnls
semilunaris --;.--~

Internal
oblique cut - -+-----__,;;.......;=11

Pre-peritoneal
firt---------T----~~~--~~~~~

CUtASIS--~t---,.;...;.~~~~~

/
FIGURE 21-38. A synthetic mesh is often used to augment the abdominal wall closure, espe-
cially when the internal oblique muscle is harvested. This mesh is placed between the trans-
versus abdominus muscle layers and the external oblique layers and is attached to the bone
below and to the cut margins of the internal oblique muscle along the abdominal wall above.
The defect in the abdominal wall is illustrated in this drawing. The reader should note that in an
actual flap harvest, the external oblique is reflected and left attached along the linea semiluna-
ris medially.
386 CHAPTER Z1

Synthetic Mesh Closure of the Abdominal Wall

Transversus
abdomlnls ----+~:::+;

FIGURE 21-39. A: The suturing of the synthetic

mesh begins with a series of drill holes placed
along the cut edge of the iliac bone. The per-
manent sutures are placed through the holes
and then through the iliacus muscle. The stitch Iliacus mus. - +t"'"r
is completed by incorporating the transversus
abdominus and the synthetic mesh. All sutures
are placed while protecting the peritoneal
contents with a ribbon retractor. All sutures are
placed and then tied. The mesh is sutured along
the entire length of the defect in the abdominal
wall (Continued). A
 ILIAC CREST OSTEOCUTANEOUS AND OSTEOMUSCULOCUTANEOUS 387

Synthetic Mesh Closure of the Abdominal Wall

~-+---- Syn1hellc
mesh
Internal
oblique cut--T-- -.......::;;;......,M

"+--+~~o+--+- Gluteus
medius

B
FIGURE 21-39. {Continued} B: Once the mesh is secured to the iliac bone, it is trimmed and
then sutured to the cut edges of the internal oblique muscle along the medial, superior, and
posterior margins, as shown. The external oblique muscle closure is then performed as shown
in Figures 21-39 and 21-40.
388 CHAPTER Z1

Synthetic Mesh Closure of the Abdominal Wall

Gluteus
muscle--..;.;

c
RGURE 21-39. (Continued) C; Upon completion of this layer of repair, the external oblique
muscle is then sutured to the upper thigh musculature, followed by repair of the subcutaneous
tissues and the skin.

FIGURE 21-40. The next layer of closure

approximates the external oblique muscle and
aponeurosis to the tensor fasciae latae and the
tendon of the gluteus medius.
 ILIAC CREST OSTEOCUTANEOUS AND OSTEOMUSCULOCUTANEOUS 389

Closure of the Abdominal Wall

FIGURE 21-41. The second muscle layer

closure has been completed.

FIGURE 21-42. Completion of the closure is

achieved by a layered approximation of the
subcutaneous tissue and the skin.
390 CHAPTER Z1

Iliac Crest Osteocutaneous Flap

FIGURE 21-43. The harvest of the iliac crest

osteocutaneous flap is perfarmed by using
atechnique similar to that described for the
osteomusculocutaneous flap. The topographi-
cal anatomy of the left hip is drawn at the outset
of this dissection.

FIGURE 21-44. The cephalad rna rg in of the

skin paddle has been incised to the level of the
external oblique muscle and fascia.
 ILIAC CREST OSTEOCUTANEOUS AND OSTEOMUSCULOCUTANEOUS 391

Iliac Crest Osteocutaneous Flap

FIGURE 21-45. The external oblique muscle

and aponeurosis have been transsected,
leaving a cuff of external oblique (arrowhead),
approximately 2 em in width, attached to the
inner table of the ilium.

FIGURE 21-46. An incision through the internal

oblique muscle, leaving 2 em attached to
the ilium, provides exposure of the ascend-
ing branch, which runs on its undersurface.
ligatures are shown on the ascending branch,
which is then transsected.
392 CHAPTER Z1

Iliac Crest Osteocutaneous Flap

FIGURE 21-47. A medial view of the dissec-

tion reveals the ligated ascending branch as
it courses through the transversus abdominis
muscle.

FIGURE 21-48. The ascending branch

has been followed through the transversus
abdominis, exposing the DCIA and DCIVto
their junction with the external iliac vessels.
The transversus abdominis has been incised
laterally to expose the preperitoneal fat and the
iliacus. The remainder of this dissection follows
that of the description for the osteomusculo-
cutaneous flap. The closure of this defect is
modified as a result of the preservation of the
internal oblique muscle, which is incorporated
with the transversus abdominis for the inner
layer of the abdominal wall closure, which is
approximated to the iliacus muscle and the
iliac bone as noted above. A synthetic mesh is
rarely used for a abdominal wall closure in this
situation.
 ILIAC CREST OSTEOCUTANEOUS AND OSTEOMUSCULOCUTANEOUS 393

Inset of Right Iliac Crest-Internal Oblique Composite Flap to Reconstruct

a Right Oromandibular Defect

FIGURE 21-49. A right hemimandibulectomy

defect has been reconstructed with a right iliac
crest-internal oblique osteomyocutaneous free
flap with one opening osteotomy to contour the
bone to the defect (white srrow). The DCIAJ
DCIV (yellow arrow) exitthe flap from the inner
posterior aspect in position to anastomose to
recipient vessels in the ipsilateral neck. The
skin paddle (green arrow) is attached through
its mesentery to the caudal portion of the flap
and is in position to suture to a cutaneous
defect in the neck or lower cheek. It may simply
serve as an external monitor to assess the
circulation to the flap.

FIGURE 21-50. The natural curvature of the

ipsilateral iliac bone matches the natural cur-
vature of the hemimandible such that additional
contouring in addition to the single osteotomy is
not necessary.
394 CHAPTER Z1

Inset of Right Iliac Crest-Internal Oblique Composite Flap to Reconstruct

a Ri Oromandibular Defect

FIGURE 2151. The opening osteotomy is filled

with corticocancello11s bone harvested from the
iliac donor site (yellow srrowl.

FIGURE 21-52. The internal oblique muscle

has been transposed through the floor of mouth
defect and brought over the bone and plate
(yeHowa"ow).lhe maneuverabilityofthe
internal oblique muscle affords a far greater
degree of versatility to this composite flap
than can be achieved with the iliac crest
osteocutaneous flap.
 ILIAC CREST OSTEOCUTANEOUS AND OSTEOMUSCULOCUTANEOUS 395

Inset of Left Iliac Crest-Internal Oblique Musculocutaneous Flap to Reconstruct

an Anterolateral Mandibular Defect

FIGURE 21-53. A. B, C: Three views of a

composite defect that has been created in the
right anterolateral aspect of the mandible and
floor of mouth. The bone defect extends from
the right angle to the left paramedian region
(Continued). B
396 CHAPTER Z1

Inset of Left Iliac Crest-Internal Oblique Musculocutaneous Flap to Reconstruct

an Anterolateral Mandibular Defect

FIGURE 21-53. (Conunued }. c

FIGURE 2154. A left iliac crest-internal

oblique osteomuculocutaneous flap has been
harvested to reconstruct this defect. Two verti-
cal osteotomies (yellow arrows) have been
outlined to contour the bone to the anterolateral
defect.
 ILIAC CREST OSTEOCUTANEOUS AND OSTEOMUSCULOCUTANEOUS 397

Inset of Left Iliac Crest-Internal Oblique Musculocutaneous Flap to Reconstruct

an Anterolateral Mandibular Defect

FIGURE 21-55. The iliac bone has been con-

toured to the bony defect with two osteotomies
(yellow arrows) that achieve bone-to-bone
contact between the flap and the native man-
dible. The OCIAIDCIV (white arrow) exit from
the anterior aspect of the flap in proximity to
recipient vessels in the left neck.
398 CHAPTER Z1

Reconstruction of a Right Total Maxillectomy Defect with a Right Iliac

Crest-Internal Oblique Flap

FIGURE 21-56. A, B, C: A defect in the right

palatomaxillary complex has been created that
includes the inferior orbital rim and the orbital
floor. A coronoidectomy is performed in order
to prevent impingement of the mandible on the
posterior aspect of the maxillary reconstruction
and limit the excursion of the mandible leading
to trismus. The goals in restoring this defect
are to 11 J reestablish the maxillary alveolus.l2l
reestablish the contour of the inferior orbita I
rim, (3) reconstruct the orbital floor for support
of the globe, (4) achieve separation of the
oral cavity from the sinonasal cavities, and (5)
restore the anteroposterior projection of the
midface region. B
 ILIAC CREST OSTEOCUTANEOUS AND OSTEOMUSCULOCUTANEOUS 399

ReconstJuction of a Right Total Maxillectomy Defect with a Right Iliac

Crest-Internal Obli ue FlaP-

FIGURE 21-56. (Continued). c

FIGURE 21-57. A right iliac crest-internal
oblique osteomuscularflap has been harvested.
The bone has been harvested from the region of
the anterior superior iliac spine (yellow srrowt
to a point posteriorly along the crest (white
arrow). The vertical dimension of the bone pro-
vides sufficient height to extend from the level
ofthe alveolus to the zygomatic arch and the
lateral orbital rim. The internal oblique muscle
has been harvested based on the ascending
branch (blsck srrowt. The DCIA (red pin) and
the DCIV (blue pin) are shown. The additional
length of the DCIV relative to the length of the
DCIA is achieved by harvesting the full extent
of the DCIV as it diverges from the DCIA and
extends cephalad for several centimeters
before entering the external iliac vein.
400 CHAPTER Z1

Reconstruction of a Right Total Maxillectomy Defect with a Right Iliac

Crest-Internal Oblique Flap

FIGURE 2151. The iliac bone has been

contoured and fixed into position. The natural
curvature of the inferior orbital rim has been
established by cutting the top of the iliac bone.
In addition, the natural bony contour of the
pyriform aperture is reestablished in order to
avoid narrowing the nasal airway. Bone to bone
contact between the bone graft and the zygoma
and anterior alveolus has been achieved in
order to promote a stable union. One opening
osteotomy has been created to contour the
shape of the bone to the anterior face of the
maxilla. Rigid fixation has been achieved to the
zygomatic arch, lateral orbital rim, the nasal
bone, and the anterior alveolus.

FIGURE 2159. A corticocancellous bone

graft (yellow arrow) has been placed into the
opening osteotomy site in order to promote a
bone union at the midpoint in the flap.
 ILIAC CREST OSTEOCUTANEOUS AND OSTEOMUSCULOCUTANEOUS 401

ReconstJuction of a Right Total Maxillectomy Defect with a Right Iliac

Crest-Internal Oblique Fla

FIGURE 21-60. A:. An orbital floor plate (yellow

srrow) (Synthes, Inc., 1302 Wrights lane East,
West Chester, Pennsylvania 19380.) has been
contoured and fixed to the anterior aspect of
the iliac bone in order to provide support for the
globe. B: The plate extends posteriorly to bridge
the defect in the orbital floor and care must be
taken to prevent either pressure or impinge-
ment on the globe by the plate. The plate should
ideally rest upon the surface of the remaining
posterior segment ofthe orbital floor. B
402 CHAPTER Z1

Reconstruction of a Right Total Maxillectomy Defect with a Right Iliac

Crest-Internal Oblique Flap

FIGURE 21-61. The defect in the palate is

restored by passage ofthe internal oblique
muscle through the defect into the maxillary
cavity (blue/yellow arrow) ta fill that cavity and
restore the lateral wall of the nose.

FIGURE 21-&2. The muscle has been trans-

posed thraugh the defect in the palate and
sutured to the palatal mucoperiosteum. The
vascular pedicle is brought through a soft tis-
sue tunnel either on the lateral aspect of the
mandible or medial ta the mandible in close
proximity to recipient vessels in the subman-
dibular triangle. The DCIA and DCIA may not
reach to the submandibular triangle and vein
grafts may be required to achieve a tension-
free anastomoses.

FIGURE 21-63. The bulky appea ranee of the

muscle that is used to reline the palate changes
dramatically as denervation atrophy occurs and
the natural shape and contour ofthe palatal
arch are restored. Muscle ta mucosal contact
(red arrows) is established in order to prevent
the development of an oronasal or oraantral
fistula. lateral muscle to buccal mucosal con-
tact must be ensured in order to prevent egress
of either sinus or oral secretions through the
tunnel into the neck and thereby threaten the
microvascular pedicle (black arrow!.
 ILIAC CREST OSTEOCUTANEOUS AND OSTEOMUSCULOCUTANEOUS
REFERENCES
1. Baker S: Reconstruction of mandibular defects with the
revascularized free tensor fascia lata osteomyocutaneous
flap. Arch OtolaryngolHead Neck Surg 1981;107:404.
2. Fredrickson JM, Man SC, Hayden RE: Revascularized
iliac bone graft for mandibular reconstruction. Acta Oto-
laryngol (Stockh) 1985;99:214.
3. Gong-Kang H, Hu R, Miao H, Yin Z, Lan T, Pan G:
Microvascular free transfer of iliac bone based on the
deep superior branches of the superior gluteal vessel.
Plast Reconst:r Surg 1985;75:69.
4. Karener H, Helborn B, Redner H: The osteomusculocu-
taneous musculoperitoneal groin flap in head and neck
reconstruction. J Recomtr Microsurg 1989;5:31.
5. Koshina I, Fakuda H, Soeda S: Free combined antero-
lateral thigh flap and wscularized iliac bone graft with
double vascular pedicle. J Recomt:r Microsurg 1989;
5:55.
6. Manchester W: hnmediate reconstruction of the man-
dible and temporomandibular joint. Br J Plast Surg
1965;18:291.
7. Moscoso J, Keller J, Genden E, et al.: Vascularized bone
flaps in oromandibular reconstruction: a comparative
anatomic study of bone stock from various donor sites
to assess suitability for enosseous dental implants. Arch
Otolaryngol Head Neck Surg 1884;120:36.
8. Ramasastry SS, Granick MS, Futrell J: Clinical anat-
omy of the internal oblique muscle. J Recomt:r Microsurg
1986;2:117.
9. Ramasastry SS, Tucker ]B, SwartzWM, Hurwitz DJ:The
internal oblique muscle flap: an anatomic and clinical
study. Plast Recmutr Surg 1984;73:721.
10. Riediger D: Restoration ofmasticatory function by micro-
surgically revascularized iliac crest bone grafts using
enosseous implants. Plast Reconm Surg 1988;81:861.
403
11. Salibian A, Rappaport I, Allison G: Functional oroman-
dibular reconstruction with the microvascular composite
groin flap. Plast Recomtr Surg 1985;76:819.
12. Salibian A, Rappaport I, Furnas D, Achauer B: Micro-
vascular reconstruction of the mandible. Am J Surg
1980;140:499.
13. Sanders R, Mayou B: A new vascularized bone graft
transferred by microvascular anastomosis as a free flap.
Br J Surg 1979;66:787.
14. Shenaq SM: Reconstruction of complex cranial and
craniofacial defects utilizing iliac crest-internal oblique
microsurgical free flap. Microsurgery 1988;9: 154.
15. Thylor GI, Daniel RK:The free flap: composite tissue tran&-
fer by vascular anastomosis. Aust N Z J Surg 1973;43:1.
16. Thylor GI,Townsend P, Corlett R: Superiority of the deep
circumflex iliac vessels as the supply for free groin flaps:
experimental work. Plast Recomr:r Surg 1979;64:595--604.
17. Taylor GI, Watson N: One-stage repair of compound leg
defects with free, revascularized flaps of groin skin and
iliac bone. Plast R econst:r Surg 1978;61 :494-506.
18. Urken ML: Composite free flaps in oromandibular
reconstruction. Review of the literature. Arch Otolaryngol
Head Neck Surg 1991;118:724-732.
19. Urken ML, Vickery C, Weinberg H, Buchbinder D, Biller
HF: The internal oblique-iliac crest osseomyocutaneous
microvascular free flap in head and neck reconstruction.
J Recomt:r Microsurg 1989;5:203.
20. Urken ML, Vickery C, Weinberg H, Buchbinder D, Law-
son W, Biller HF: The internal oblique-iliac crest osseo-
myocutaneous free flap in oromandibular reconstruction:
report of 20 cases. Arch Otolaryngol Head Neck Surg
1989;115:339.
21. Urken ML, Weinberg H, Vickery C, Buchbinder D, Law-
son W, Biller HF. The internal oblique-iliac crest free flap
in composite defects of the oral cavity involving bone,
skin and mucosa. Laryngoscope 1991;101:257.
 aylor et al. (29) were the first to report the success-
T ful transfer of the vascularized fibular bone flap for
reconstruction of an open fracture of the lower extrem-
ity in 1975. A posterior approach to harvesting this flap
was described. Gilbert (7) introduced a much simpler
lateral approach to fibular flap harvest in 1979, and this
technique is universally applied today. Olen andYan (4)
are credited with being the first to report the tmD.sfer of
a fibular osteocutaneous flap in 1983. In the same year,
Yoshimura et al. (34) described the transfer of the fibu-
lar bone with a "buoy" flap ofskin that served as a moni-
tor of the circulation. The primary purpose for using the
fibular donor site was to reconsttuct long-bone defects
of the extremities until Hidalgo (14) adapted this flap
to the restoration of segmental mandibulectomy defects
in 1989. A number of detailed anatomic dissections
404
have helped to clarify the circulation from the peroneal
artery to the skin (2-4,23,27,32). However, the variabil-
ity in the vascular supply to the skin occasionally limits
the use of this flap in oromand:ibular reconsttuction in
which defects of the mucosa and the skin are commonly
encountered.The ease of harvest and the distance of the
donor site from the head and neck region have enhanced
the popularity of this composite flap.
The fibula is a long, thin, non-weight bearing bone of
the lower extremity (Fig. 22-1). It has a tubular shape,
with thick cortical bone around the entire circumference
that renders it one of the strongest bones available for
transfer (28). Approximately 22 to 25 em of bone can
be harvested while preserving 6 to 7 em of bone both
distally and pro.ximally to maintain the integrity of the
knee and ankle joints.The path of the common peroneal

FIGURE 221. AIaten~ I view of the fibula reveals the head at the proximal end (srrowt and the
lateral malleolus distally. This long narrow bone is triangular on cross section. It forms a tube of
dense cortical bone that surrounds the central medullary cavity and can supply up to 25 em of
vascularized bone in an average adult.
nerve, which wraps around the neck of the fibula, also
limits proximal dissection and bone harvest. However,
the ability to tra:asfer up to 25 em of bone makes this
donor site unique in terms of the ability to restore total
or subtotal defects of the mandible (8).
FLAP DESIGN AND UTILIZATION
The fibula can be tra.Il.sfeiTed aa a free osseous or as a
free osteocutaneous fiap. The skin of the lateral aspect of
the calf is supplied by either septocutaneous or muscu-
locutaneous perforators arising from the peroneal artery
and vein. Variations in the location of these perforators
occur along the posterior crural septUm. It is therefore
advisable to design a longer skin paddle that accounts for
these variations (Fig. 22-2). Fleming et al. (6) described
a division of the skin flap for internal and enernal lin-
ings when multiple skin perforators were visualized.
The width of the akin paddle is limited by the ability
to achieve primary closure, although a skin graft can be
effectively applied to the donor defect when needed.
The vascular pedicle to the fibular fiap, although
consistent in the location and caliber of the vessels
lumen, is often limited in length by the bifurcation of
the posterior tibial artery (11). Additional length can
be obtained on the vascular leaah by harvesting a more
distal segment of bone and skin while discarding the
more pr<Wmal fibula. A subperiosteal dissection of the
soft tissue surrounding the proximal bone that is to be
removed effectively preserves the blood supply to the
distal fiap. Hidalgo (16) reponed obtaining vascular
pedicles as long as 12 em by using this technique. In
addition, Wei et al. (31) designed a proximal skin pad-
dle combined with a distal bone segment to separate
and increase the flexibility of the two components of
this composite fiap. The vascular supply of the skin was
preserved by performing a subperiosteal dissection and
discarding the proximal bone segment.
The peroneal artery and vein run along the entire
length of the fibula without a significant change in the
FIBULAR OSTEOCUTANEOUS 405
cahoer of the vessel. This arrangement permits the use
of the fibula as a "flow-through" flap to supply a seo-
ond free fiap anastomosed to the free ends of the dis-
tal artery and vein. Wei et al. (31) transferred two free
flaps supplied by one set of donor vessels by using this
technique.
The primary application of the fibular donor site in
the head and neck has been for the reconsttuction of
segmental defects of the mandJole. The strength of the
cortical bone effectively withstands the powerful forces
of mastication. There are certain situations in which the
characteristics of the fibula are particularly conducive
to a successful reconstruction. As noted previously,
the length of bone that can be harvested makes it the
only donor site that is suitable for the reconstruction of
subtotal or total mandJoular defects. In the secondary
reconstruction of mandJOular defects that involve the
ramus and condyle, it is often possible to create only
a narrow tunnel for the placement of the graft into the
glenoid fossa. The creation of a larger tunnel is lim-
ited by scaning and concerns about injury to the facial
nerve. The narrow fibula is much easier to pass through
such a tunnel than a bulkier segment of bone, such as
the ilium.
The long straight fibular bone must be contoured
to match the shape of the mandible by creating wedge-
shaped closing osteotomies. Hidalgo (16) described a
series of techniques that allow more aesthetic contoUl'-
ing of the fibula to mimic the shape of the native man-
dible. Multiple osteotomies can be performed without
compromising distal bone circulation if the periosteum
is not significantly traumatized. Hidalgo advised against
stripping the periosteum when performing an osteot-
omy by cutting the bone and the periosteum together.
Jones et al. (19) showed that the fibular bone could be
osteotomized and folded on itself to produce a "double-
barreled" vascularized bone graft (see Figs. 22-41 to
22-43). The vascularity to the distal segment was main-
tained through the periosteum and surrounding cuff of
muscles. The double-barreled fibular fiap was applied
to the reconstruction of segmental defects of the femur.
40& CHAPTER 22

Common peroneal n.

Head of fibula

Lateral sural
cutaneous n.

Gastrocnemius m.

Soleus m.

Peroneus longus m.

Peroneus brevis m.

FIGURE !22. The posterior crural septum is located between the posterior muscles, the gas-
trocnemius and soleus. and the anterior muscles. the peroneus longus, and brevis. The approxi
mate position of the septum can be gauged by drawing a line between the head of the fibula and
the lateral malleolus. The lateral sural cutaneous nerve arises from the common peroneal nerve
and supplies sensation to the skin of the fibular osteocutaneous flap.

It has also been applied to the reconstruction of seg-
mental mandibular defects in the dentate mandible in
an effort to match the height of the native mandible and
reduce the crown to root ratio in dental rehabilitation.
This technique was advanced even further by Sad-
ove and Powell (24) who removed a 3-cm segment
of the fibula, in addition to performing osteotomies,
to contour the bone to reconstruct the mandible and
the maxilla in a patient with a severe posttraumatic
deformity. The segment of bone was removed in a
subperiosteal plane, which permitted the proximal
and distal segments to be rotated and placed into two
separate three-dimensional planes. The author has
utilized this technique in a single patient who devel-
oped severe trismus at approximately 6 months after
surgery. Follow-up imaging demonstrated calcification
and actual bone formation along the intervening seg-
ment of periosteum located between the mandible and
maxilla. Resolution of the trismus required removal of
the ectopic bone.
The ability to harvest two segments of bone can be
very useful in select circumstances where two vascu-
larized bone segments that are separated in space are
required. This technique has been successfully applied
to the reconstruction of bilateral segments of the man-
dibular bodies affected by osteoradionecrosis (18). By
removing the intervening central segment of bone, the
proximal and distal bone segments could be placed into
the bilateral defects in a tension-free manner.
The fibula is a triangular-shaped bone that has one
surface where the pedicle runs, a second surface that
cannot be violated due to the path of the cutaneous per-
forators arising from the peroneal artery and vein, and
the third surface that is most commonly used for rigid
fixation. There are two options in the orientation of the
fibular bone that relate to the position of the skin relative
to the bone and the point of exit from the bone of the
peroneal vessels. When reconstructing the mandible, the
bone should be oriented so that the vascular pedicle of
the fibula is located on the lingual surface of the neoman-
dible, and depending on which leg is selected, the skin
can be placed either along the superior or the inferior
borders. H the skin is located along the inferior border,
it can be delivered into the oral cavity by transposing it
through the floor of mouth or over the buccal surface. In
the latter orientation, the mesentery to the skin, through
which the perforators run, can also be used to provide
coverage of the fixation hardware. In addition, the cuff
of the flexor hallucis longus provides augmentation in
the submandibular region (12). If the skin and bone are
oriented such that the skin is situated on the top of the
neomandible then there is no need for transposition of
the skin paddle and therefore no added tension is placed
on the delicate perforators to the skin. Knowledge of the
relationships between the skin, the bone, and the exit
point of the proximal vascular pedicle will allow the sur-
FIBULAR OSTEOCUTANEOUS 407
geon to select the appropriate leg for harvest based on
the defect requirements and the location of the recipi-
ent vessels. When preoperative imaging demonstrates
that only one fibula is available for harvest, due to either
atherosclerosis or a vascular anomaly, then the surgeon
can anticipate the orientation of the donor vessels and
the skin based on the information outlined above and
demonstrated in Figures 22-21 through 22-25.
Functional mandibular reconstruction involves the
successful placement of a lower dental prosthesis to
restore mastication. In patients who have residual teeth
in the native mandible, a tissue-home partial prosthe-
sis can be effectively used. Such a prosthesis gains sta-
bility and retention through clasps that attach to the
remaining teeth. However, when the patient is edentu-
lous or there are no remaining teeth of sufficient qual-
ity to allow the use of clasps, then successful dental
rehabilitation requires the use of endosteal implants
that function as tooth root analogues. The thick cor-
tical bone of the fibula accepts implants well (22,35).
However, the smaller dimensions of the bone require
the use of a shorter dental implant than can be used
in the ilium. Long-term follow-up is required to deter-
mine whether marginal bone loss around the implant
will lead to extrusion because, with a shorter implant,
there is a greater percentage of implant exposure for the
same amount of bone loss. The alternative option of a
diJuble-barreled fibula flap has been applied in order to
increase the height of bone into which longer implants
can be placed (19) (Figs. 22-41 through 22-43). This
approach can be applied in a strategic fashion to aug-
ment the neomandibular height in those portions of the
mandible into which implants will be placed.
The era of the application of free tissue transfer to
palatomaxillary defects has led to the use of the fibu-
lar osteocutaneous flap to this reconstructive challenge.
The fibular bone contours well to horizontal defects
of the infrastructure of the maxilla and the skin can be
used to resurface the central palate. The bone is used
to restore the maxillary alveolus and provides an excel-
lent structure for the placement of implants. The cen-
tral palate can be effectively restored with the soft tissue
components of this flap. Hemi- or total palatal defects
can be reconstructed with this technique. Unlike in the
mandible, the height of the bone is less of a problem. In
the mandible, the bone is usually set along the inferior
border in order to restore the jaw line, and as a result,
the height of the fibular bone determines the height of
the neomandible. The fibula matches the height of the
atrophic mandible but is significantly deficient relative
to the native dentate mandible. In the maxilla, the fib-
ula matches up well with the contour of the maxillary
alveolus and therefore implants are well positioned for
dental restoration. We have even used this flap for alveo-
lar augmentation in edentulous patients with atrophic
maxillae.
408 CHAPTER D

The introduction of the sensate fibular osteocuta-

neous Bap by Hayden and O"Leary (12) added a new Common peroneal Popliteal a.

dimension to this donor site. The lateral sural cutaneous

nerve can be harvested and anastomosed to a suitable Anterior
tibial a.
recipient nerve to restore sensation to the skin compo-
nent. These authors also deacnbed the harvest of the
peroneal communicating branch for use as a waculSJ:I-
ized nerve graft to bridge the gap in the inferior alveo- Posterior
lar to mental nerve for the restoration of sensation in tibial a.
cutaneous n.
the lower lip. Sadove et al. (25) used the sensate fibu-
lar osteocutaneous flap to achieve a one-stage penile
reconstruction. The lateral sural cutaneous nerve was Interosseous
anastomosed to an appropriate recipient nerve, which membrane

reportedly restored sensation to the neophallus.

Medullary

NEUROVASCULAR ANATOMY
The peroneal artery and vein provide the primary blood
supply to the fibular osteocutaneous Bap. The pop-
liteal artery is classically described as bifurcating into Tibia
the anterior and posterior tibial arteries, and the lat-
ter vessel subsequendy gives rise to the peroneal artery
(Figs. 22-3 and 22-4). The peroneal artery and its two
venae comitantes descend in the lower leg between the
Bexor hallucis longus and the tibialis posterior. A discus-
sion of the vascular supply to the fibular Bap, of neces-
sity, must include a description of the various patterns
of vascular supply to the foot. Knowledge of these vari-
ations and a preoperative evaluation to determine their
presence is imperative to avoid ischemic complications.
In addition to supplying the nutrient artery ofthe fib-
ula and musculoperiosteal vessels, the peroneal artery
and vein also give rise to fasciocutaneous perforators
that run in the posterior crural septum to supply the
RGURE !2-3. Posterior view left leg. The popliteal artery
skin. There has been a considerable amount of interest
usually divides into the anterior and posterior tibial arteries.
and research into the location, size, course, and reliabil-
The peroneal artery arises from the posterior tibial artery
ity of the blood vessels supplying the skin. This is a par-
within 2 to 3 em of the bifurcation of the popliteal artery. The
ticularly important issue when dealing with composite
peroneal artery provides an endosteal vascular supply to
defects of the head and neck in which mucosal and cuta-
the fibula through the nutrient medullary artery as well as
neous defects often re(}.uire soft tissue flaps in addition
numerous periosteal feeders. Proximal and distal segments
to the bone that is used to reconstruct the mand:&ble and
of the fibula atthe ankle and knee must be preserved
maxiJJa. In his initial description of the application of the
(dashed line). The common peroneal nerve winds around
fibular free flap to mandibular reconstruction, Hidalgo
the neck of the fibula where it is most susceptible to injury.
(14) reported 12 cases. Among these 12 patients,
4 had intraoral mucosal defects, and 1 had an external
cutaneous defect. Only one of the five skin paddles sur- It is worthwhile to review and summarize the anatomic
vived entirely, and one developed a 30% necrosis. The studies that have been reported on the blood supply to
remaining three skin paddles were excised intraopera- the fibular skin paddle. The blood vessels that enter the
tively because of ischemia. Hidalgo's conclusion from skin in the vicinity of the posterior crural septum have
this experience was that the fibular donor site should been classified in a variety of different ways. In 1983,
not be used for reconstructing mand:&bular defects with Yoshimura et al. (33) described three different types of
large intraoral mucosal defects. arteries: type A vessels pass through the peroneus longus
This rather pessimistic view of the fibular skin pad- in the more proximal portion of the leg; type B vessels
dle is not shared by others who have successfully used run between the soleus and peroneus muscles, giving off
this composite Bap in head and neck reconstruction. muscular branches prior to supplying the skin; type C

Superficial
peroneal n.
Anterior
tibial a.
Interosseous
membrane
ng
branch of
peroneal a.
FIGURE 224. Anterior view left leg. The anterior tibial
vessels pass above the upper edge of the interosseous
membrane to reach the anterior compartment of the leg. The
common peroneal nerve divides into the superficial and deep
peroneal nerves. The latter descends in the leg in conjunc-
tion with the anterior tibial artery. The distal portion of the
peroneal artery passes through a gap in the interosseous
membrane to reach the anterior compartment where it joins
with the lateral malleolar branch of the anterior tibial artery
forming a plexus around the heel and the lateral malleolus.
branches run a course similar to that of type B but with-
out the muscular branches. Beppu et al. (1) adopted
the same classification system in their anatomic study
reported in 1992. Wei et al. (32) described two types
of septocutmeous perforators: those that traverse the
FIBULAR OSTEOCUTANEOUS 409
posterior crural septum throughout their entire course
and those that travel through either the flexor hallucis
longus, tibialis posterior, or soleus before entering the
septum and supplying the skin. Wei et al. referred to the
latter branches as musculocutaneous vessels. However,
Shusterman et al. (27) classified the vessels that have
a muscular course before traveling through the septum
to the skin as septomuscular perforators to distinguish
them from the musculocutaneous branches that exit the
surface of the muscle before ramifying in the subcutane-
ous layer. It is important to recognize the course of the
septomuscular branches to include a protective cut! of
muscle around the bone (Fig. 22-5).
A number of cadaveric dissection studies have
addressed the quantity and the geography of these
different vessels to help in Bap planning and harvest.
Carriquiry et al. (3) reported that the diameter of the
perforators ranged from 0.4 to 1.3 mm, with the largest
perforators located at either end ofthe fibula. The proxi-
mal large perforator crosses in close proximity to the
common peroneal nerve and has been referred to as the
"circumBex peroneal" artery. However, this vessel com-
monly arises from the posterior tibial artery instead of
the peroneal. Chen andYan (4) observed that there were
four to five cutmeous branches that e:zited the soleus
muscle in a segmental distribution along the length of
the calf. Yoshimura et al. (34) reported an average of
4.8 cutmeous branches from the peroneal artery, each
with a minimum diameter of 0.3 mm. The posterior tib-
ial artery contnbuted 5.7% of the cutmeous branches
to the lateral calf skin; the popliteal artery gave rise to
3.5% and the anterior tibial artery to only 0.4%. The
cutaneous branches arising from the arteries other than
the peroneal were primarily noted in the proximal pol'-
tion of the leg.Yoshimura et al. divided the length of the
fibula into 10 zones and quantified the perforators aris-
ing in each zone. The greatest density of peroneal cuta-
neous branches was concentrated in the zone that was
eight-tenths of the way along the fibula. The authors
also noted that 71% of the cutaneous perforators were
musculocutaneous and 29% were septocutaneous.
Beppu et al. (1) similarly studied the distribution of
cutaneous branches to the lateral calf and found that
one perforator was consistendy located at the midpoint
between the head of the fibula and the lateral malleolus.
In 21 of 23 dissections, a cutmeous branch was identi-
fied within 2 em of that point. The proximal third of the
calf was inconsistendy supplied by the peroneal artery.
In 5 of 23 dissections, there were no branches from the
peroneal artery to the pro:zimal calf skin. Beppu et al.
reported that at least one septocutaneous vessel was
noted in the middle third of the calfin all 23 dissections.
Wei et al. (32) reported their findings in 35 dissections
in which they found four to seven cutmeous branches
arising from the peroneal artery, they also noted one
to four septocutmeous arteries, except in one leg, in
410 CHAPTERD

Peroneal a. and v.

Musculocutaneous perforator

Extensor digitorum longus m.

Flexor hallucis longus

and soleus m. cuff
Extensor hallucis longus

Peroneus longus and

brevis muscles

Tibialis posterior m. Deep peroneal n. ,

anterior tibial a. and v.
Tibialis posterior m.

'A'~~t-- Fiexor digitorum longus m.

~-- Tibial n.

Flexor hallucis longus

FIGURE 22-5. The cross-sectional anatomy of the lower leg reveals the path of the septocutaneous
perforators, which may run entirely through the septum or, in part, through the flexor hallucis longus
before entering the lateral aspect of the septum. A cuff of the flexor hallucis longus is harvested to
protect these perforators. The musculocutaneous perforators (inse~ run their entire course through
the posterior muscle group, necessitating a cuff of both the soleus and flexor hallucis long us.

which there were none. By contrast, Shustel'DUUl et al.
(27) reported that 20% of the 80 cadaver leg dissections
showed no septocutaneous perforators and 6.25% also
showed no muscular or septomuscular vessels. The sep-
tal vessels tended to be located more distal in the leg in
contrast to the muscular or septomuscular branches. It
was unclear from this report whether there were any dis-
se<:tions that showed a total absence of both septal and
muscular perforators; however, one can surmise that
there is an absolute skin paddle reliability in the nmge
of 93% to 94%.
Despite the conttoversy regarding the skin paddle,
Wei et al. (31) reported a 100% success rate with the
skin paddle of the fibular osteoseptocutaneous flap in 80
cases of extremity reconsttuction and 27 cases of man-
dibular reconstruction. In the latter group, there was
1 case of total flap failure, but no partial or total skin loss
among the remaining 26 cases. The authors centered
the skin paddle at the junction of the middle and lower
third of the fibula. They warned that the posterior crural
septum had to be included with the flap and that excess
traction during harvest or closure was detrimental to

the skin's blood supply. Despite their success with the
skin paddle, Wei et al. used a second soft tissue free flap
in 14 of their 26 mandibular reconstructions. In a dis-
cussion of this article, Hidalgo (16) reported that, from
his experience and that of others, the skin paddle of the
osteoseptocutaneous flap was reliable in only 91.5% of
cases overall. He also stated that, in his experience, the
skin island often survives even in the absence of any vis-
ible perforators in the septum. Despite this claim, most
surgeons use the approach described by Fleming et al.
(6) to harvest the skin paddle of the fibular flap from
an anterior direction in a subfascial plane. A long flap
is usually designed to allow for variations in the loca-
tion of perforators that occur along the length of the
septum. H no septal perforators are visualized, then
musculocutaneous branches must be identified to sup-
ply the skin. The absence of musculocutaneous vessels
that can be traced back to the peroneal system usually
indicates the need for an alternative soft tissue flap (30).
A cuff of flexor hallucis longus and soleus should be
included, even in those cases in which a septal branch is
identified, because of the possibility that it may diverge
from the septum to run through the muscle as it courses
toward the peroneal vessels (Fig. 22-5). In his discus-
sion of the article by Shusterman et al. (27), Hidalgo
(15) questioned the validity of harvesting a large cuff of
soleus muscle. In his experience, he noted that many of
the large muscular perforators did not take their origins
from the peroneal vessels and that large cuffs of soleus
muscle are often devascularized and must be excised.
Dye-injection studies of the peroneal artery stained a
skin territory that averaged 9. 9 em in width and 21.4 em
in length. Fleming et al. (6) successfully split the skin
paddle by incising to, but not through, the level of the
fascia. They also advised mapping cutaneous perfora-
tors preoperatively with Doppler ultrasonography. Dual
skin paddles with freedom of movement between them
can be safely harvested without de-epithelialization,
when more than one perforator is identified coursing to
each cutaneous segment.
As noted previously, the sensory supply to the skin
of the lateral calf is derived from the lateral sural cuta-
neous nerve (Fig. 22-2). This branch arises from the
common peroneal nerve within or above the popliteal
fossa. However, this nerve is variable, as described by
Huelke (17), who reported it to be absent in 22% of
cases. Kosinski (20) reported only a 1.7% incidence of
this nerve being absent but also noted that in an addi-
tional9.4% of cases, there were no cutaneous branches
before the lateral sural nerve joined the medial sural
cutaneous nerve. The peroneal communicating nerve is
a second superficial nerve that traverses the territory of
the fibular flap. It also arises from the common peroneal
nerve and joins with the medial sural cutaneous nerve to
form the sural nerve. That junction may occur at virtu-
ally any location from the popliteal fossa to the lateral
FIBULAR OSTEOCUTANEOUS 411
malleolus. This nerve was absent in 50% of dissections
reported by Kosinski (20) and 20% of those reported by
Huelke (17). As noted above, Hayden and O'Leary (12)
described using the peroneal communicating branch as
a vascularized nerve graft.
ANATOMIC VARIATIONS
The fibula is the most common long bone in the body to
be absent or markedly diminished in size to the extent
that it is replaced by a fibrous band. The anomaly of an
absent fibula is usually associated with a shortened leg
and an abnormal tibia, which is bowed forward, findings
that would be readily apparent on preoperative radio-
graphs of the donor limb.
Variations in the arterial supply to the foot are the
greatest concern in harvesting the peroneal artery to
supply a fibular flap. According to Senior (26), there
have been no reports of the absence of the peroneal
vessels, and the same holds true for the anterior tibial
artery. The latter, however, may be significantly dimin-
ished in size. In roughly 10% to 20% of cases, either the
anterior tibial or the posterior tibial artery may become
attenuated during their course in the lower leg. When
this happens, a communicating branch from the pero-
neal artery supplies the missing or diminutive vessel's
territory in the distal extremity. It is evident that, under
such circumstances, the sacrifice of the peroneal artery
would result in ischemia of the foot.
POTENTIAL PITFALLS
The most serious consequence of fibula flap transfer is
the lack of collateral circulation to the foot, leading to
ischemia following interruption of the peroneal artery.
Preoperative assessment helps to avoid this potential
problem that may be due to atherosclerosis or a con-
genital abnormality noted above. Atherosclerosis of
the peroneal artery may predict a difficult flap transfer,
while atherosclerosis of the anterior and/or posterior
tibial arteries may predict ischemic complications to
the foot.
There are a variety of donor-site complaints that have
been reported including cold intolerance and edema.
Functional deficits include weakness in dorsiflexion of
the great toe, related either to injury to branches of the
peroneal nerve or scarring of the muscles, in particu-
lar, the flexor hallucis longus (9). Patients have reported
pain and weakness on ambulation for several months
after surgery. Muscle weakness is believed to be caused
by the disruption of the muscular origins that attach to
the fibula and the interosseous membrane. Detailed gait
analysis revealed abnormalities in stride, joint angles,
and "ground reaction forces," which were thought to
412 CHAPTER 22
be attributable to muscle weakness and altered load
transmission (21).
Injury to the common peroneal nerve may occur as a
result of traction or errant dissection, leaving the patient
with an equinovarus deformity and anesthesia along the
anterior and lateral sides of the leg and dorsum of the
foot. This complication can be avoided by identifying
this nerve early in the dissection. A 6- to 7-cm segment
of bone should be left attached to the knee as an addi-
tional protection. A similar length of bone should be
preserved distally to maintain the integrity of the ankle
joint. Hematomas were reported in 25% of 20 patients
in whom the fibula was used for long-bone reconstruc-
tion. This complication was attributed, in part, to ooz-
ing from the exposed medullary surfaces of the bone
ends (5).
There continues to be some uncertainty in the blood
supply to the skin of the fibular osteocutaneous flap.
The surgeon cannot be assured of the skin's vascularity
until after harvest. Because of the loss of the skin paddle
that has been reported in 5% to 10% of cases, a backup
plan should be formulated prior to surgery. The patient
should be informed of the potential need for a second
soft tissue flap. This is especially true in those patients
with defects that require significant soft tissue recon-
struction. An alternative flap, such as the scapula, or a
separate soft tissue flap combined with the fibula should
be anticipated.
The width of the fibular skin paddle is usually lim-
ited by the ability to achieve primary wound closure.
Excess tension must be avoided to prevent a compart-
ment syndrome. When there is any question as to the
need for a skin graft, the surgeon should not hesitate
to place one onto the cutaneous defect on the lateral
calf. It may be removed by serial excision. Alterna-
tively, a tissue expander may be placed primarily
through the defect to expand the dorsal skin of the
calf. Secondary closure with a linear scar may then be
achieved (I 0).
PREOPERATIVE ASSESSMENT
The possible absence or diminished size of the anterior
or posterior tibial arteries and the prevalence of athero-
sclerosis in the lower extremities require that a preop-
erative evaluation be performed prior to fibula transfer.
Angiography is the conventional method to perform
such an evaluation. Magnetic resonance angiography
is our method of choice and has also been shown to
provide at least comparable if not improved defini-
tion of the vascular anatomy of the lower extremities.
This noninvasive technique provides valuable informa-
tion that guides the selection of the appropriate leg to
use for flap harvest or whether to select an alternative
donor site (13).
Acknowledgments
The authors would like to acknowledge the contributions
of Dr. Michael Sullivan to the writing of this chapter in
the first edition of this book.
 FIBULAR OSTEOCUTANEOUS 413

Fibular Osteocutaneous Flap

FIGURE 22-6. The topographical anatomy is

outlined on the lateral aspect of the left leg. The
landmarks to the intermuscular septum are the
fibular head superiorly and the lateral epicon-
dyle of the ankle inferiorly. A dashed line con-
necting these points identifies the location of
the intermuscular septum. The peroneal nerve,
which passes 1to 2 em below the fibular head,
is marked in red.

FIGURE 22-7. A: The cutaneous flap is

designed with a fusiform shape centered A
on the intermuscular septum. The dominant
septocutaneous perforators are usually located
more distally in the leg, and therefore, the skin
paddles are usually centered over the junction
of the middle and distal thirds. B: Over time,
the author (MLU) has adopted the strategy of
designing a long anterior curvilinear incision
(red line) that allows exposure to the intermus-
cular septal perforators along the entire length
of the available bone. The posterior incision can
then be modified (dashed green lines) based on
the actual location of the perforator(s) and the
dimensions of the soft tissue defect. B

FIGURE 22-8. The dissection is performed with

a thigh tourniquet inflated to 350 mm Hg. The
lower leg is exsanguinated with a compressive
bandage prior to inflation of the tourniquet.
The skin paddle is incised anteriorly through
the skin and subcutaneous tissue. The fascia
overlying the peroneus longus and brevis is also
incised, and the dissection proceeds from ante-
rior to posterior in a subfascial plane toward
the intermuscular septum.
414 CHAPTER ZZ

Sensate Fibular Osteocutaneous Flap

.it!+~+.--+- Lateral sural

cutaneous n.

Reflected ---~~11:;:
skin paddle

FIGURE 229. When harvesting a

sensate fibular osteocutaneous flap, the
technique is slightly altered. The lateral
sural cutaneous nerve is found by tracing
the common peroneal nerve in a proximal
direction into the popliteal fossa. The lat-
eral sural cutaneous nerve arises at any
point along the course of that dissection.
Once identified, the lateral sural nerve
can be traced caudally to the cutane-
ous paddle, which has been reflected
anteriorly. The peroneal communicating
branch also arises from the common
peroneal nerve in this location and travels
through the soft tissue ofthe flap but does
not supply sensation to the flap skin. The
peroneal communicating branch mav be
included in the flap and used as a vascu-
larized nerve graft.
 FIBULAR OSTEOCUTANEOUS 415

Fibular Osteocutaneous Flap

FIGURE 2210. The skin paddle has been

reflected to reveal the posterior crural septum.
The peroneus longus is retracted. A septocu-
taneous perforator is identified coursing out to
the skin (srrow).

FIGURE 2211. Dissection along the anterior

aspect of the fibula requires elevation of the
peroneus longus, peroneus brevis, and exten-
sor hallucis longus. The deep peroneal nerve,
anterior tibial artery, and anterior tibial vein are
identified in the anterior compartment.

FIGURE 22-12. Further dissection along the

medial aspect of the fibula reveals the interos-
seous membrane (arrows).
416 CHAPTER ZZ

Fibular Osteocutaneous Flap

Peroneus longus m.

Common peroneal n.

Gastrocnemius m.

FIGURE 2213. At this juncture in the dissection, bone cuts are made in the proximal and distal
fibula. A segment of fibula must be preserved both proximally and distally, as shown. Distrac-
tion of the fibula is needed to proceed with the remainder of the dissection. This can only be
accomplished by transection of the interosseous membrane that releases the firm attachments
between the tibia and the fibula.

FIGURE 22-14. The posterior incision around

the skin paddle is made through the fascia over-
lying the gastrocnemius and soleus. In this par-
ticular dissection, a septocutaneous perforator
(arrow) has been identified, and therefore, a
septocutaneous, rather than a musculocuta ne-
ous flap, will be harvested. The gastrocnemius
and soleus muscles are transacted longibJdinally
with a cuff of muscle left attached to the bone.
 FIBULAR OSTEOCUTANEOUS 417

Fibular Osteocutaneous Flap

FIGURE 22-15. The distal portion of the pero-

neal artery and vein are identified after distrac-
tion of the fibula. The distal pedicle is ligated
and transsected.

FIGURE 22-16. After cutting the interosseous

membrane, the chevron-oriented muscle fibers
of the tibialis posterior are visualized and trans-
sected (white dashed line) to follow the peroneal
artery and vein proximally in the calf.

FIGURE 22-11. The entire length of the pero-

neal vascular system (srrowt has been dis-
sected to the bifurcation of the posterior tibial
vessels. The surgeon should be alerted to the
medial course ofthe proximal peroneal vessels
that run obliquely in the dissection, toward the
junction with the posterior tibial artery, due to
lateral distraction of the bone.
418 CHAPTER ZZ

Fibular Osteocutaneous Flap

FIGURE 22-18. Prior to ligation of the pedicle, the flexor hallucis longus is transected, leaving a
cuff attached to the composite flap.

FIGURE 22-19. The fibular osteocutaneous

flap has been harvested with a cuff of flexor
hallucis longus and tibialis posterior. The vas-
cular pedicle consisting of the peroneal artery
and two venae comitantes can be lengthened
by performing a subperiosteal dissection and
removing the proximal bone but preserving the
intervening periosteum and muscle cuffs.

FIGURE 22-20. Primary closure of the leg

has been accomplished. If necessary, a skin
graft can be applied to this defect. Following
closure, a posterior splint is fashioned for the
lower leg and foot. Application of a compres
sive dressing should be done with care to
ensure protection of the pressure points over
the medial and lateral aspects ofthe ankle as
well as the dorsum of the foot.
 FIBULAR OSTEOCUTANEOUS 419

Orientation of the Fibular Osteocutaneous Flap Relative to the Laterality of the Leg of
Harvest and the Position of the Cutaneous Paddle and the Donor Vascular Pedicle

FIGURE 2221. Provided that preoperative

imaging demonstrates suitable peroneal ves-
sels and suitable collateral flow in both legs.
the surgeon has a choice of which fibular flap
to harvest. The decision as to which donor
site to use is determined by the location of the
recipient vessels in 1he neck and the desir
ability of placing the attachment of the skin
mesentery on 1he top or the bottom of the
neomandible. Right and left fibular osteocu-
taneous flaps are shown with the skin paddle
and 1he peroneal vessels.

FIGURE 22-22.. Figures 22-22 to 22-25 provide

information for the surgeon as to the orienta-
tion of the skin, donor vessels, and the plating
surface of the fibular bone for each of the four
possible orientations of1he right and left fibular
flaps. A similar defect is shown with solid lines
denoting the position of the osteotomies for
removal of the body and coronoid process of
the mandible. In this figure, the contralateral
right fibular flap is demonstrated wi1h the plate
applied to the appropriate surface and 1he skin
is shown positioned at the inferior aspect of the
neomandible while the donor vessels are exiting
from the anterior aspect of 1he reconstruction.
420 CHAPTERZZ

Orientation of the Fibular Osteocutaneous Flap Relative to the Laterality of the Leg of
Harvest and the Position of the Cutaneous Paddle and the Donor Vascular Pedicle

FIGURE 22-23. lfthe right fibular flap is turned

180 degrees, the pedicle exits from the posterior
aspect of the neomandible and the skin is now
positioned on the top of the bone for repair of a
mucosal defect.

FIGURE 2224. Harvest of an ipsilateral left

fibular flap orients the vessels toward the pos-
terior aspect of the neomandible with the skin
positioned in the neck to repair a cutaneous
defect or transposed over the mandible to
repair a mucosal defect

FIGURE 2225. lfthe left fibular flap is turned

by 180 degrees, then the donor vessels exit
at the anterior aspect of the reconstructed
mandibular body and the skin paddle is situated
on the superior surface for reconstruction of a
mucosal defect.
 FIBULAR OSTEOCUTANEOUS 421

Contour and Rigid Fixation of the Fibular Free Flap to Reconstruct a Right
Hemimandibular Defect

FIGURE 22-26. A. B: Mandibular reconstruc-

tion with a fibular free flap is required most
often after segmental resection due to a muco-
sal carcinoma invading the mandible. Exposure
to the mandible is created by a lip-split {as
shown) or a visor incision. B
422 CHAPTERZZ

Contour and Rigid Fixation of the Fibular Free Flap to Reconstruct a Right
Hemimandibular Defect

FIGURE 22-27. Cheek flap is reflected, expos

ing the segment of the mandible to be resected
and reconstructed.

FIGURE 22-28. A. B: If the tumor does not

penetrate through the buccal cortex, a 2-mm
locking reconstruction plate (LRP) is applied
to the rna nd ible prior to resection. A 2-mm LRP
template is applied to the mandible to facilitate
contouring of the fixation plate to match the
mandible's native shape. B
 FIBULAR OSTEOCUTANEOUS 423

Contour and Rigid Fixation of the Fibular Free Flap to Reconstruct a Right
Hemimandibular Defect

FIGURE 2229. A. B: A2-mm LRP is contoured

to the template shape and applied to the man-
dible. A 1.5-mm drill is used to create three or
four screw holes on either side of the defect. B
424 CHAPTERZZ

Contour and Rigid Fixation of the Fibular Free Flap to Reconstruct a Right
Hemimandibular Defect

FIGURE 22.-30. The LRP maintains condylar

position and the shape of the native mandible.
Maxillomandibular fixation is therefore unnec-
essary. The plate is then removed, the tumor
is resected, and the plate is reapplied to the
remaining mandible using the holes that were
drilled prior to tumor resection.

FIGURE 22.-31. A: After the left fibula is

harvested, wedge-shaped ostectomies are
then made in the fibula to create the proper
contour to match the contour of the plate and
native mandible prior to revascularising the
flap. The soft tissue inset is completed as well.
B: The segments of the fibula are maintained
in position with wire fixation (yellow arrows)
prior to rigid fixation achieved through the plate
and screws. This temporary measure helps to
maintain some internal stability of the bone seg-
ments, which facilitates the contouring process. 6
 FIBULAR OSTEOCUTANEOUS 425

Contour and Rigid Fixation of the Fibular Free Flap to Reconstruct a Right
Hemimandibular Defect

FIGURE 22-32. A. B: The fibula is fixated to

the 2-mm LRP with monocortical screws. One
or two screws can be placed in each segment
depending on how much internal rigidity is
achieved through bone-to-bone contact With
the use of the contralateral fibula. 1he pedicle
(yellow arrowt exits from the anterior aspect
of the flap, and the skin mesentery is attached
to the undersurface of the neomandible. The
skin is ideally positioned for restoration of a
cutaneous defect in the neck or cheek. The ori-
entation of the composite flap could have been
flipped by 180 degrees to place the pedicle at
the posterior aspect of the neomandible and
the skin would have been positioned on the
top of the neomandible rather than requiring
transposition through either a buccal or lingual
route into the oral cavity. B
426 CHAPTERZZ

Contour and Rigid Fixation of the Fibular Free Flap to Reconstruct a Right
Hemimandibular Defect

FIGURE ZZ-33. A: An ipsilateral right fibular

flap has been harvested and contoured and
fixated to the mandibular defect. The pedicle
exits from the mandible at the posterior aspect
of the reconstructed segment while the skin is
oriented for repair of a cutaneous defect in the
neck or cheek. B: Provided that there is suf-
ficient laxity of the mesentery, the skin paddle
can be transposed over the bone into the floor
of mouth through either a buccal route (yellow
arrow) or a floor of mouth route. B
 FIBULAR OSTEOCUTANEOUS 427

Contour and Rigid Fixation of the Fibular Free Flap to Reconstruct a Right
Hemimandibular Defect

FIGURE 22-33. (Continued) C: If that

ipsilateral fibular flap is rotated 180 deg re es,
then the pedicle exits the anterior aspect of the
neomandible and the skin is situated on the top
of the bone graft and positioned relative to the
neomandible for repair of a mucosal defect. c
428 CHAPTERZZ

Management of the Resected Mandibular Condyle

FIGURE 22.-34. A. B: The mandibular condyle

and subcondylar region can be preserved in
most instances of segmental mandibulecto-
mies. Rigid fixation to this segment requires that
at least two to three screws are placed in the
subcondylar ramus in order to achieve stability. B
 FIBULAR OSTEOCUTANEOUS 429

Management of the Resected Mandibular Condyle

FIGURE 22-35. A:. A prosthetic condyle

has been fashioned to the end of the three
dimensional reconstruction plate. The contour
of the prosthesis is shown relative to that of
the native condyle. The prosthesis is shown in
position in the glenoid fossa both in a lateral (B)
and anterior (C) projection. B
430 CHAPTERZZ

Management of the Resected Mandibular Condyle

FIGURE ZZ-35. (Continued) c

FIGURE ZZ-36. The end of a fibular bone flap is

shewn next to the native condyle.

FIGURE ZZ-37. Contouring of the end ofthe

fibular bone is performed to allow the bone to fit
into the fossa. Two drill holes are placed into the
end of the neocondyle for fixation of the condyle
into the fossa. A soft tissue cap (usually muscle)
is placed aver the end of the bone if the ca rti-
lage has been removed from the joint space.
Twa drill holes are placed inta the lateral lip af
the fossa in order to pass permanent sutures or
a wire to prevent neocondylar migration.
 FIBULAR OSTEOCUTANEOUS 431

Management of the Resected Mandibular Condyle

B
FIGURE 22-38. A:. A contralateral right fibular
osteocutaneous flap has been harvested that
places the vascular pedicle atthe anterior
aspect of the flap and the skin is attached to the
inferior surface of the bone. After placing the
patient in maxillomandibularfixation, a 2.4-mm
LAP is cantilevered off the remaining mandible
to create the neomandibular shape extending
to just inferior to the glenoid fossa. With the
neocondyle positioned in the fossa, the first
contouring ostectomy is marked in a V-shaped
fashion. B: A segment of bone is removed for
contouring of the bone atthe level of the neo-
angle. C: The angle of the hemimandible has
been fashioned with a closing ostectomy. c
432 CHAPTERZZ

Management of the Resected Mandibular Condyle

FIGURE 22-39 A: A second ostectomy has

been performed to restore the body of the
mandible (blue srrowj. B: The shape of the
neomandible has been established and IC) it is
then fixed to the defect with the precontoured
reconstruction plate. C
 FIBULAR OSTEOCUTANEOUS 433

Management of the Resected Mandibular Condyle

FIGURE 22-40. A:. The neocondyte has been

positioned in the fossa (blue arrow) and two
permanent sutures are shown placed between
the lateral lip of the fossa and the neocondyle.
B: The skin paddle is rotated over the neoman-
dible to restore a mucosal defect. The pedicle
exits the neomandible anteriorly in this situation
in order to avoid having the pedicle make an
acute turn at the level of the temporomandibu-
lar joint were it to exit from the posterior aspect
of the reconstruction. B
434 CHAPTERZZ

Contouring and Inset of a Right Double Barreled Fibular Flap to Reconstruct a Right
Hemimandibulectomy Defect

FIGURE 22.-41. In select circumstances,

when reconstructing a tooth-bearing segment
in a dentate mandible, and in particular when
the goal is to restore the patient with dentures
fixed to dental implants, the discrepancy in
the crown-to-rootdistance can be overcome
by increasing the height of the neomandible
by doubi ing the bone segment with a double-
barreled configuration. The two proximal seg-
ments (yellow arrows) are contoured to the
defect in the mandible and rigidly fixed to the
lower border of the mandible. The distal seg-
ment Iwhite arrow) is redundant and would
otherwise be discarded. The top surface of
the distal segment has been removed so that
when it is placed atop, the other segment
there will be broad bone-to-bone contact. The
intervening segment of bone (blue arrow) is
removed while preserving the periosteum to
provide the laxity to move the distal segment.

FIGURE 22.-42. A: Rigid fixation of the

proximal two segments has been achieved,
and the distal segment (yellow arrow) is
positioned to be folded back over the anterior
portion of the reconstructed neomandible. A
 FIBULAR OSTEOCUTANEOUS 435

Contouring and Inset of a Right Double Barreled Fibular Flap to Reconstruct a Right
Hemimandibulectomy Defect

FIGURE 22-42. {Continued) B: With the distal

segment in position (blue arrow), it is fixated
with a 1.5-mm titanium miniplate. B
R6URE 22-43.. The soft tissue is rotated over
the plate and inset to the floor of mouth tissues.
In this situation,1he skin must be positioned on
the undersurface of the neomandible so that
it does not interfere with the desire to achieve
bone-to-bone contact in the double-barreled ..
configuration. The vascular pedicle (white
arrowt exits at the level of 1he neo-angle for
anastomosis to vessels in the ipsilateral neck.
The surgeon must realize thatthe use of bone
from the distal end of the flap for 1his purpose,
has implications for a more limited length of
the peroneal vessels that can be achieved for
anastomosis atthe proximal end of 1he flap. It is
therefore imperative that 1he recipient vessels
are proximate to 1he reconstruction to avoid the
need for vein grafts.
436 CHAPTERZZ

Contouring and Fixation of a Fibular Osteocutaneous Flap for Reconstruction of a Right

Infrastructure Maxillectomy Defect

FIGURE ZZ-44. A: Reconstruction of the

infrastructure maxillectomy defect with the
fibular osteocutaneous free flap is more
straightforward with regard to flap orienta-
tion than when recanstrucung the mandible.
The fibula ipsilateral to the defect is used to
allow the peroneal vessels to drape into the
neck, the skin paddle is oriented to recreate
the palate, and the lateral aspect of the fibula
faces buccally to aIIow plating away from the
periosteal blood supply. A complete or modified
Weber-ferguson type incision is generally rec-
ommended to allow adequate access for tumor
extirpation and flap reconstruction. B: The
cheek flap is elevated over the zygomatic body
to achieve exposure for resection and contour-
ing of the fibula. B
 FIBULAR OSTEOCUTANEOUS 437

Contouring and Fixation of a Fibular Osteocutaneous Flap for Reconstruction of a Right

Infrastructure Maxillectomy Defect

FIGURE 22-45. The palatal incision is outlined

for maxillary resection with a hemipalatectomy.

FIGURE 22-46. A:. The infrastructure maxillec-

tomy defectto be reconstructed is created. A
438 CHAPTERZZ

Contouring and Fixation of a Fibular Osteocutaneous Flap for Reconstruction of a Right

Infrastructure Maxillectomy Defect

FIGURE 2246. (Continued) B: The hemipalatal

specimen is shown. B

FIGURE 22.-47. A: The reconstructed maxillary

segment is in close proximity to the anterior
aspect of the ramus of the mandible. To avoid
impingementthat would affect the nonnal
excursion of the mandible, the coronoid pro
cess is removed. A
 FIBULAR OSTEOCUTANEOUS 439

Contouring and Fixation of a Fibular Osteocutaneous Flap for Reconstruction of a Right

Infrastructure Maxillectomy Defect

FIGURE 22-47. (Continued) B: The defect and

the coronoidectomy are shown. B

FIGURE 22-48. An ipsilateral rightfibularflap

has been harvested with the vessels exiting
at the posterior aspect of the neomaxilla and
the skin attached to the inferior surface of the
alveolar ridge.
440 CHAPTERZZ

Contouring and Fixation of a Fibular Osteocutaneous Flap for Reconstruction of a Right

Infrastructure Maxillectomy Defect

FIGURE ZZ-49. A: The fibular bone is contoured

to fit the defect in the maxillary alveolus. The
position for the ostectomy is chosen based on
the location of the perforator to the skin paddle
and is balanced by the desire to maximize the
length of the vascular pedicle for delivery into B
the neck. Creating the ostectomy more distally
in the fibula will lengthen the vascular pedicle
and will require that the proximal bone is dis-
carded. Ensuring thatthe cutaneous perforator
is preserved is vital to ensuring that a viable
skin paddle is available for closure of the palatal
defect. B: An ostectomy has been performed
in order to create the natural contour of the
maxillary alveolus. C: The neoalveolus has been
rigidly fixed and the tremendous length ofthe
perc neal artery and vein fer transpcsition intc
the neck is readily evident (green arrow). c
 FIBULAR OSTEOCUTANEOUS 441

Contouring and Fixation of a Fibular Osteocutaneous Flap for Reconstruction of a Right

Infrastructure Maxillectomy Defect

FIGURE 22-50. A:. Rigid fixation of the fibula in

the defect has been achieved. Miniplates are
used to achieve bone-to-bone contact ante-
riorly to the remaining alveolus in the midline
and posteriorly at the level ofthe zygomatic
body. B: The plane of the zygoma is different
from the plane of the alveolus, and the fixation
plate (blue arrowt must be contoured in order to
accommodate that discrepancy. B
442 CHAPTERZZ

Contouring and Fixation of a Fibular Osteocutaneous Flap for Reconstruction of a Right

Infrastructure Maxillectomy Defect

FIGURE 2251. The two segments of the fibula

are shown {yellow arrows). The position and
contour of the fixation plate at the IeveI of the
zygomatic body is shown {blue arrow).

FIGURE 2252. The skin paddle is sutured

around the extent of the remaining defect in the
palate. Achieving a water-tight sealatthe level
of the tunnel for the vascular pedicle is critical
to prevent egress of saliva along the tract cre-
ated through the cheek.
 FIBULAR OSTEOCUTANEOUS 443

Contouring and Fixation of a Fibular Osteocutaneous Flap for Reconstruction of a Right

Infrastructure Maxillectomy Defect

FIGURE 22-53. The vascular pedicle can be

transferred through the cheek in a subcutane-
ous plane (yellow arrow) or alternatively can be
delivered through a submucosal plane and exit
into the neck along the body of the mandible.

FIGURE 22-54. An alternate pathway to deliver

the vessels is to create a tunnel deep to the
mandible (yellow arrow) and avoid potential
injury to the facial nerve. This path is the short-
est route to recipient vessels in the subman-
dibular triangle.
444 CHAPTER 22
REFERENCES
1. Beppu M, Hand D, Johnston G, Carmo J, Tsai T: The
osteocutaneous fibula flap: an anatomic study. J &conslT
Microsurg 1992;8:215-223.
2. Carr A, MacDonald D, Waterhouse A: The blood supply
of the osteocutaneous free fibular graft. J Bone Joint Surg
[Br] 1988;70B:319--321.
3. Carriquiry C, Costa A, Vasconez I..: An anatomic study of
the septocutaneous vessds of the leg. Plast Reconst:r Surg
1985;76:354-361.
4. Chen z, Yan W: The study and clinical application of the
osteocutaneous flap of fibula. Microsurgery 1983;4: 11-16.
5. Coghlan B, Townsend P:The morbidity of the free vascu-
larized fibula flap. Br J Plast Surg 1993;46:466.
6. Fleming A, Brough M. Evans N, et al.: Mandibular
reconstruction using vascularized fibula. Br J Plast Surg
1990;43:403-409.
7. Gilbert A: Vascularized transfer of fibula shaft. Int
J Microsurg 1979;1:100.
8. Gilbert R, Dovion D: Near total mandibular reconstruc-
tion: the free vascularized fibular transfer. ()per Tech Oto-
laryngol Head Neck Surg 1993;4:145.
9. Goodacre TE,Walker CJ, Jawad AS, Jackson AM, Brough
MD: Donor site morbidity following osteocutaneous free
fibula transfer. Br J Plast Surg 1990;43:410-412.
10. Hallock G: Refinement of the fibular osteocutaneous
flap using tissue expansion. J &conslT Microsurg 1989;5:
317-322.
11. Harrison DH: The osteocutaneous free fibular graft.
J Bone Joint Surg [Br] 1986;68B:804-807.
12. Hayden R, O'Leary M: A neurosensory fibula flap: ana-
tomical description and clinical applications. Presented
at the 94th Annual Meeting of the American Laryngo-
logical, Rhinological and Otological Society Meeting in
Hyatt Regency Waikoloa, Big Island of Hawaii, Hawaii,
May 8, 1991.
13. Hayden RE, Carpenter J, Thaler E: Magnetic reso-
nance angiography: noninvasive evaluation of poten-
tial free flaps. Presented at the Annual Meeting of the
American Academy of Facial Plastic and Reconstruc-
tive Surgery in Minneapolis, Minnesota, October 1,
1993.
14. Hidalgo D: Fibula free flap: a new method of mandible
reconstruction. Plast Reconst:r Surg 1989;84:71.
15. Hidalgo D: Discussion of "The osteocutaneous free
fibula flap: Is the skin paddle reliable?" by Schusterman
et al. Plast &conslT Surg 1992;90:797.
16. Hidalgo D: Discussion of "Fibula osteoseptocutaneous
flap for reconstruction of composite mandibular defects"
by Wei F, Seah C, Tsai Y, Liu S, Tseu M. Plast &const:r
Surg 1994;93:305.
17. Huelke DF: The origin of the peroneal communicating
nerve in adult man.AnatRecord 1958;131:81.
18. Jacobson AS, Buchbinder D, Urken MI..: Reconstruction
of bilateral osteoradionecrosis of the mandible using a sin-
gle fibular free flap. Laryngoscope 2010;120(2):273-275.
19. Jones N, SwartzW, Mears D,Jupiter J, Grossman A: The
"double-barrd" free vascularized fibular bone graft. Plast
Reconst:r Surg 1988;81:378.
20. Kosinski C:The course, mutual rdations and distribution
of the cutaneous nerves of the metazonal region of leg
and foot. J Anat 1926;60:274.
21. Lee EH, Goh JC, Helm R, Pho RW: Donor site morbid-
ity following resection of the fibula. J Bone Joint Surg [Br]
1990;72:129-131.
22. Lyberg T, Olstad 0: The vascularized fibular flap for
mandibular reconstruction. J Craniomaxillojac Surg
1991;19:113.
23. Onishi K, MaruyamaY, IwahiraY: Cutaneous and fascial
vasculature of the leg: an anatomic study offasciocutane-
ous vessds.J Reconst:r Microsurg 1986;2:181.
24. Sadove R, Powdl L: Simultaneous maxillary and man-
dibular reconstruction with one free osteocutaneous flap.
Plast ReconslT Surg 1993;92:141.
25. Sadove R, Sengenzer M. McRoberts J, Wdls M: One
stage total penile reconstruction with a free sensate osteo-
cutaneous fibula flap. Plast &const:r Surg 1993;92: 1314.
26. Senior HD: An interpretation of the recorded arterial
anomalies of the human legandfoot.JAnat 1919;53:130.
27. Shusterman MA, Reece GP, Miller MJ, Harris S: The
osteocutaneous free fibula flap: is the skin paddle rdi-
able? Plast Reconst:r Surg 1992;90:787-793.
28. Serra A, Paloma V. Mesa F, Ballesteros A: The vascular-
ized fibula graft in mandibular reconstruction. J Oral
Maxillofac Surg 1991;49:244-250.
29. Taylor GI, Miller DH, Ham FJ: The free vascularized
bone graft: a clinical extension of microvascular tech-
niques. Plast Reconst:r Surg 1975;55:533.
30. Von Twisk R, Pavlov P, Sonneveld J: Reconstruction of
bone and soft tissue defects with free fibula transfer. Ann
Plast Surg 1988;21:555-558.
31. Wei F, Seah C, TsaiY, Liu S, Tsai M: Fibula osteosepto-
cutaneous flap for reconstruction of composite mandibu-
lar defects. Plast &const:r Surg 1994;93:294.
32. Wei FC, Chen HC, Chuang CC, Noordho:ff MS: Fibu-
lar osteoseptocutaneous flap: anatomic study and clinical
application. Plast Reconst:r Surg 1986;78: 191-199.
33. Yoshimura M, Shimada T, Hosokawa M: The vascula-
ture of the peroneal tissue transfer. Plast &const:r Surg
1990;85:917-921.
34. Yoshimura M, Shimamura K, IwaiY, Yamauchi S, Ueno
T: Free vascularized fibular transplant. J Bone Joint Surg
[Am] 1983;65A:1295-1301.
35. Zlotolow I, Huryn J, Piro J, Lenchewski E, Hidalgo D:
Osseointegrated implants and functional prosthetic reha-
bilitation in microvascular fibular free flap reconstructed
mandibles.AmJ Surg 1992;165:677-681.
~e radial forearm flap was first developed at the
~ Shen:yang Military General Hospital and a large
clinical series showed the flap could be used very suc-
cessfully to reconstruct head and neck defects (16). In
1983, Soutar et al. (17) noted the benefits of this flap,
with or without bone, in the reconstruction of intraoral
defects. They espoused the merits of this donor site for
intraoral reconstruction due to the dependable vascular-
ity, pliability, thickness, and relatively hairless quality of
the skin as well as the availability of vascularized bone for
mandibular reconstruction. Furthermore, the reliability
of the vascular pedicle and the cutaneous nerve inner-
vation are well-documented strengths of this technique.
While the transfer of skin and fascia from the radial
forearm donor site are covered in detail in Chapter 12,
this chapter reviews the details of flap harvest of the
osteocutaneous radial forearm flap.
FLAP DESIGN AND UTILIZATION
Prior to improved donor-site management, the
osteocutaneous radial forearm free flap (OCRFFF)
had not achieved widespread popularity because of
concerns about bone quantity, bone quality, and
donor-site morbidity (1,2,6,8,9,11,18,19). The
445
446 CHAPTER 23
+---l- Pronator teres
Insertion
..,..,_--+-- Brachioradialis
Insertion
FIGURE 231. Harvest of a segment of the radius is limited
to the region between the insertions of the pronator teres
and the brachioradialis muscles. This length of bone is
roughly 10to 12cm and the harvested bone cannot be any
more than 40% of the circumference of the radius.
radius bone is critical to the function of the wrist and
hand, restricting its usefulness as a bone donor site.
The length of the bone is limited to 10 to 12 em by
the insertions of the pronator teres and the brachio-
radialis muscles (Fig. 23-1). Also, to maintain the
structural integrity of the radius bone, only 40% of
the circumference of the bone should be harvested
making the useful bone stock less than the fibula,
scapula, or iliac crest (19).
Early published series of patients undergoing
OCRFFF harvest demonstrated serious donor-site mor-
bidity in the form of radius bone fracture (2,6,1 0,18,20).
Ranging from 8% to 67%, the combined incidence
of donor radius fracture in these early series averaged
24%. Often, the donor arms in these series were pro-
phylactically immobilized postoperatively. Fractures
were treated with a variety of methods including pro-
longed immobilization and external fixators. Although
most fractures in these series healed after treatment,
some reportedly experienced delayed union or nonun-
ion, requiring additional treatment with external fixa-
tion and more complex reconstructions involving both
vascularized and nonvascularized bone grafts. Loss of
hand function after these fractures was due to radial
shortening and distal radioulnar subluxation and dis-
location. Furthermore, substantial impairment in pro-
nation and supination was found, as well as a 50%
decrease in grasp and pinch strength, in association
with the pathologic fracture of the donor site (15).
Unfortunately, morbidity related to the potential for a
pathologic fracture following ostectomy of the radius
justifiably caused enthusiasm for the OCRFFF to wane.
Investigators then examined ways to minimize the
potential complication of the radial bone fracture,
resulting in several different recommendations. A Limit
on bone resection to 40% or less of the cross-sectional
area of the radius was advocated (11,19). To decrease
the stress risers, or a point in the bone that concentrates
the forces of stress, switching to the "keel-boat,. modi-
fication of the bony harvest will theoretically reduce
the possibility of radial bone fractures (15,20). This
method bevels the ends of the bony cut to gradually rise
to the level of the native radius. Casting the arm for 6
to 8 weeks was a popular method to allow for increased
bone deposition and increased strength of the residual
bone. Others have advocated prophylactic bone grafting
and an external fixator (1,6). However, little science was
applied in formulating these practices and it was dif-
ficult to determine if these methods had any impact on
the incidence of radius fracture.
1\vo studies were then published to quantify the
weakening of the radius that occurs following graft
harvest. Swanson et al. (19) noted that formalin-fixed
ostectomized human radii were only 24% as strong as
intact radii. They were separated into uncut, square-cut,
and bevel-cut groups. There was a trend of improved
strength but no significant difference in beveled cuts
over square cuts. Further, in a study of donor-site mor-
bidity in sheep tibiae that were used as a model of the
ostectomized radius, Meland et al. found that even with
instituting these precautions, the overall strength of
the osteomized bone in torsion was decreased by more
than 70% (11). This finding led them to recommend
abandonment of the use of OCRFFFs as a source of
vascularized bone. Use of this flap for head and neck
reconstruction essentially ceased after publication of
these studies and in the face of more widespread use of
composite fiaps harvested from the iliac crest and fibu-
lar donor sites.
In an effort to reduce this donol.'-site morbidity and
therefore safely make use of the superior soft tissue c:har-
acteristics of the volar forearm, biomechanical studies
demonstrated gready improved strength when a recon-
struction plate was prophylactically applied to the donor

radius after graft harvest. Bowers et al. (3) compared
&esb.-frozen radial bones that had a 50% thickness, and
an 8-cm. defect created with square cut ends. They then
used a 3.5-mm stainless steel dynamic compression
plate to bridge the defect on half of the cut radii. They
found the resected radius to be 82% weaker than the
intact radius to bending and 76% weaker to torsion. In
addition, the radii with the ostectomy were 62% weaker
than those that were ostectomized and plated. Being a
cadaveric study, the issue of stress shielding and limited
devascularization of the bone could not be addressed.
This provided a strong argument for the routine practice
of prophylactically plating the radial bone after partial
thickness removal, and is now done in e:very case (5,14).
The OCRFFF has the advmtage of tremendous soft
tissue versatility. A long vascular pedicle, large amounts
of thin pliable skin, and reliable sensory innervation make
the soft tissue components of the RFFF ideal for the
reconstruction of the tongue, fl.oor of the mouth, alveo-
lar ridge, pharyngeal wall, and soft palate (17,20,21,22).
The available radius bone is of adequate length for many
limited mandibular defects. Although the bone graft is
usually not of sufficient stock to reliably accept dental
implants, this is less of an issue for a defect of limited
length. The ideal situation for this flap is for lateral man-
dibular defects when most of the native mandible remains
to support either dental implants or a tissue-borne pros-
thesis (21). In this setting, the bone of the OCRFFF is
more than adequate for many reconstructive problems
and provides superior soft tissue for other aspects of the
reconstruction. It has also been successfuDy used in the
management of mandibular osteoradionecrosis (13).
The other main indication for the use of this fl.ap is
in the reconstruction of limited maxillectomy defects,
especially those located in the premaxilla (4,7 ,21). The
thin radial bone that is harvested with this flap closely
approximates the thin bony stock of the premaxilla.
The soft tissue from the forearm also conforms well to
the palate and separates the oral and nasal cavities reli-
ably. If implants are to be used for the upper denture,
we have placed them in the native maxilla lateral to
the radial bone. Usually, we do not place implants in
the upper arch because dentures can be formed to fit
the slightly altered anatomy of the new anterior maxil-
lary arch, with stabilization of the prosthesis coming
from the normal lateral alveolar ridges. If a larger area
than the premuilla is involved, then we tend to choose
either the fibula or iliac crest flaps, because dental
implants become necessary to hold dentures in place.
NEUROVASCULAR ANATOMY
The radial forearm free fl.ap is based on the radial
artery, which branches from the brachial artery in the
antecubital fossa. Ita venous drainage is provided by the
OSTEOCUTANEOUS RADIAL FOREARM FREE FLAP 447
Brac:t'lioradialis ~A--T-Pronator tervs
muscle--~~ muscle
~~f-rt--Fiexor carpi
radialis
Flexor pollicis
longus muscle ------l+~l
FIGURE 23-2. The lateral intermuscular septum is located
between the flexor carpi radialis and the brachioradialis
muscles. The blood supply to the radius is derived through
the insertions of the septum into that bone. The flexor polli-
cus longus muscle must be divided along the medial aspect
of the radius, in order to obtain access for the ostectomy to
be performed.
paired <venae comitans, which run with the radial artery,
and the superficial venous system, composed of the
cephalic and basilic veim. Sensation to the flap is pro-
vided by the lateral and medial antebrachial cutaneous
nerves. The soft tissue portion of this flap and its neu-
rovascular components are described in greater detail
in Chapter 12.
The lateral intermuscular septum, which contains
perforators from the radial artery supplying the peri-
osteum of the radius, passes between the flexor carpi
radialis and the brachioradialis muscles. The relation-
ship of the radial artery, the intermuscular septum,
and the bone must be preserved in order to preserve
the vascularity of the bone segment to be harvested
(Fig. 23-2). The lateral intermuscular septum also
allows for a degree of freedom when rotating the skin
paddle relative to the bone segment, which is supe-
rior to other bony flaps such as the fibula and iliac
crest. The length of bone that can be harvested is up
to 12 em, but is limited by the insertions of the pro-
nator teres muscle pro:zimally and the brachioradialis
muscle distally.
448 CHAPTER 23
Brachloradlalls
mu~e--------~~~~~WI
Superficial branch
of radial nerve - - - -F---rl
FIGURE 23-3. Retraction of the brachioradialis muscle
provides exposure to the lateral aspect of the radius for the
bone cuts to be made. The superficial branch of the radial
nerve lies on the undersurface of that muscle and must be
identified and protected. The nerve becomes more superfi-
cial in the distal forearm prior to entry into the dorsum of the
hand where it supplies sensation to the thumb and first digit.
ANATOMIC VARIATIONS
The variations of the blood supply to the flap and the
hand are detailed in Olapter 12. There are no specific
variations to the bony portion ofthis flap except the diam-
eter and length of the radius in each individual patient.
POTENTIAL PITFALLS
The major risk associated with this Bap is radius fracture.
limiting the harvest to 40% to 50% of the circumfer-
ence of the radius, as well as mating a locking recon-
struction plate over the osteotomized radius, minimizes
this risk. The combined studies from two institutions,
analyzing over 100 patients who underwent OCRFFF
harvest and prophylactic plating, reponed only two
instances of postoperative radius fracture (12,21,22).
ator
musde
FIGURE 23-4. The composite radius osteocutaneous flap
is shown with the segment of bone attached to the inter-
muscular septum through which perforators run to supply
the periosteum. In addition, the septocutaneous perforators
to the skin run through the more superficial aspect of the
septum. The relationship of the skin and bone must not be
altered, to a significant extent, in order to preventtorsion of
those perforators.
In both cases, the fracture occurred distal to the recon-
struction plate after a patient fall.
Plate exposure in the arm is another potential prob-
lem. The skin paddle should be designed with an ulnar
bias such that there is more skin remaining on the radial
aspect of the defect for advancement and closure over
the bone (Figs. 23-3 and 23-4). The flexor digitorum
superficialis muscle can also be detached from the
radius bone during graft harvest. It can then be reposi-
tioned over the top of the flexor carpi radialis tendon to
improve closure over the defect and enhance healing of
a skin graft over the Be:z.or carpi radialis tendon.
PREOPERATIVE CONSIDERATIONS
The same considerations are made for the OCRFFF as
the fasciocutaneous flap and are detailed in Chapter 12.
Most importantly, an Allens test is performed to
confirm adequate circulation to the hand via the ulnar

artery and to assist in the selection of the donor arm.
Ideally, the nondominant arm is utilized.
POSTOPERATIVE CONSIDERATIONS
As is the case for the fasciocutaneous radial forearm
free flap (Chapter 12), the arm is placed in a rigid
ulnar gutter splint after application of the skin graft for
OSTEOCUTANEOUS RADIAL FOREARM FREE FLAP 449
approximately 7 days. During this time, the forearm is
elevated and vascularity of the hand is monitored to be
certain that the bandage does not cause compression
of the circulation as a result of postoperative edema.
The splint is then removed and the wound inspected.
The patient is encouraged to resume full activity of the
wrist and fingers, including weight bearing and range
of motion exercises. The donor arm is protected with a
soft dressing until the wound and skin graft has healed.
450 CHAPTER23

Radial Foreann Osteocutaneous Flap

FIGURE 23-5. The radial artery and cephalic

vein are marked, along with the proposed skin
paddle. The soft tissue component is mea-
sured and the flap outline is drawn on the volar
aspect of the forearm. The skin paddle is posi-
tioned at least 2em proxima I to the wrist crease
and with an ulnar bias. This ensures adequate
skin coverage of the internal fixation plate.

FIGURE 23-6. The toumiquet is inflated to

250mm Hg and the soft tissue component of the
flap is raised in the standard fashion, with care
taken to preserve the lateral intermuscular sep-
tum. The skin paddle is raised with the deep fas-
cia to the level of the brachioradialis and flexor
carpi radialis tendons. After medial and lateral
elevation of the soft tissue, the fascia along
the brachioradialis muscle Iarrow! is incised to
expose the flexor hallicus longus muscle.

FIGURE 23-7. Careful dissection medial to

the cephalic vein and radial to the vascular
pedicle preserves the full Iength of the super-
ficial branches ofthe radial nerve. The radial
artery and the vena comitans are ligated,
and the pedicle and the fascia are raised
for approximately 1to 2cm. The bone avail-
able for harvest lies between the insertion
of the pronator teres and the insertion of the
tendon of the brachioradialis muscle. Lateral
retraction of the brachioradialis tendon fully
exposes the lateral aspect of the radius.
 OSTEOCUTANEOUS RADIAL FOREARM FREE FLAP 451

Radial Foreann Osteocutaneous Flap

FIGURE 23-8. The maximum amount of bone

to be harvested is outlined in blue. The proximal
limit is the pronator teres muscle (blue srrowt
and the distal limit is 2 to 4cm from the wrist
joint, near the insertion of the brachioradialis
tendon (yellow Bfrowt.
FIGURE 23-9. The flexor digitorum super-
ficial is is released from the radius, allowing
visualization of the flexor pollicis longus. The
flexor pollicus longus muscle belly is divided
over the midline of the volar surface of the
radius with electrocautery or scalpel until the
necessary length of radial bone is exposed.
The required length of bone is marked. The
distal cut must be made at least 2.5 em from
the radial styloid to allow later fixation of the
radius. Proximally. the bone can be cut even
beyond the pronator teres insertion; however.
the pronator teres tendon will require rein-
sertion. A saggital saw with a fine blade is
used to make a longitudinal cut in the radius
through its mid portion at a thickness of
40% to 50% of its diameter. Beveled cuts are
made proximally and distally.

FIGURE 2310. The osteotomy is completed

through the cortex on the ulnar side. The
periosteum is then incised dorsally. complet-
ing the bone graft harvest.
452 CHAPTER23

Radial Foreann Osteocutaneous Flap

FIGURE 2311. The elevatian cantinues in the

usual fashian from distal to proximal maintain-
ing a cuff of flexar hallicus longus muscle
attached to the bone. The deep vascular
pedicle and the cephalic vein are skeletonized
up to the antecubital fossa.
FIGURE 23-12. The lateral antebrachial
cutaneous nerve can then be easily separated
fram the cephalic vein for approximately 8 to
10 em. The tourniquet is deflated and the flap
is then left to perfuse on its pedicle, which
is divided proximally just before transfer of
the flap to the recipient site. Prior to transfer,
meticulous hemostasis should be performed in
order to minimize bleeding following revascu-
larization in the head and neck. The harvested
flap is shown with the combined skin flap and
a portion of the radius bone.

FIGURE 23-13. Prophylactic fixation of the

radius is begun by fully exposing the dorsal
aspect ofthe radius proximally and dis-
tally. Distally, the radial wrist extensors are
retracted. Proximally, the supinator is visual-
ized. Great care is taken to protectthe poste-
rior interosseus nerve, which passes through
the supinatar but is not usually adjacent to the
area of fixation.
 OSTEOCUTANEOUS RADIAL FOREARM FREE FLAP 453

Radial Foreann Osteocutaneous Flap

FIGURE 23-14. An appropriately sized 2.4-mm

locking reconstruction plate (Synthes CMF,
Paoli, PA) is positioned over the radius and
bentto the contour of the bone with three
holes on either side of the osteotomized radius
segment.

FIGURE 23-15. Bicortical screws are placed

proximally and distally to the osteotomized
segment to fixate the plate to the radius.
Screws are placed only in the holes on either
side ofthe harvested radius segment.

FIGURE 23-16. The residual muscle is

sutured to the brachioradialis tendon with
30 resorb able suture to fully cover the plate.
Closure then continues with a split thickness
skin graft as is done for a fasciocutaneous
forearm flap.
454 CHAPTER Z3
REFERENCES
1. Bardsley AF, Soutar DS, Elliot D, Batchelor AG: Reduc-
ing morbidity in the radial forearm flap donor sit. Plan
Reconstr Surg 1990;86:287-294.
2. Boorman JG, Brown JA, Sykes PJ: Morbidity in the fore-
arm flap donor arm. Br J Plast Surg 1987;40:207-212.
3. Bowers K.W, Edmonds JL., Girod DA, et al.: Osteocu-
taneous radial forearm free flaps. J Bone Joint Surg Am
2000;82:694-704.
4. Cordeiro PG, Bacilious N, Schantz S, et al.: The radial
forearm osteocutaneous sandwich free flap for recon-
struction of the bilateral subtotal maxillectomy defect.
Ann Plast Surg 1998;40:397-402.
5. Edmonds JL., Bowers MW, Toby B, Jayaraman G, Girod
DA: Torsional strength of the radius after osteofasciocu-
taneous free flap harvest with and without primary bone
plating. Otol Head and Neck Surg 2000;123:400-408.
6. Fenton OM, Roberts JO: Improving the donor site of the
radial forearm flap. British J Plast Surg 1985;38:504-505.
7. Foster RD, Anthony JP, Singer MI, et al.: Microsurgical
reconstruction ofthemidface.ArchSurg 1996;131:960-966.
8. Inglefield CJ, Koble PS: Fracture ofthe radial forearm osteo-
cutaneous donor site. Ann Plast Surg 1994;33:638-643.
9. Juretic M, Car M, Zambelli M: The radial forearm free
flap: our experience in solving donor site problems.
J Cranio-Ma:xillo-Fac Surg 1992;20: 184-186.
10. Mathews RN, Fatha F, Davies DM, et al.: Experience
with the radial forearm flap in 14 cases. Scand J Plan
Reconm Surg 1984;18:303-310.
11. Meland NB, Maki S, Chao EY, Rademarker B: The
radial forearm flap: a biomechanical study of donor-
site morbidity utilizing sheep tibia. Plan Reconstr Surg
1992;90:763-773.
12. Militsakh ON, Werle A, Mohyuddin N, et al.: Compari-
son of radial forearm to fibula and scapula osteocutane-
ous free flaps for oromandibular reconstruction. Arch
Otolaryngol Head Neck Surg 2005;131(7):571-575.
13. Militsakh ON, Wallace DI, Kriet JD, Tsue TT, Girod DA:
The role of the osteocutaneous radial forearm free flap in
the treatment of mandibular osteoradionecrosis. Otolarn-
gol Head Neck Surg 2005;133:80-83.
14. Nunez VA, Pike J, Avery C, et al.: Prophylactic plating
of the donor site of osteocutaneous radial forearm flaps.
Br J Oral-MaxilkJ-Fac Surg 1997;37:210-212.
15. Smith AA. Bowen CV. RabeczakT, Boyd JB: Donor site
deficit of the osteocutaneous radial forearm flap. Ann
Plast Surg 1994;32:372-376.
16. Song R. Gao Y, SongY, Yu Y, SongY: The forearm flap.
Clin Plast Surg 1982;9:21-26.
17. Soutar DS, Scheker NS, Tanner NS, McGregor IA: The
radial forearm flap: a versatile method for intra-oral
reconstruction. Brizish J Plast Surg 1983;36: 1--8.
18. Soutar DS, Widdowson WP: Immediate reconstruction
of the mandible using a vascularized segment of radius.
Head Neck Surg 1986;8:232-246.
19. Swanson E, Boyd JB, Mulholland RS:The radial forearm
flap: a biomechanical study of the osteotomized radius.
Plast Reconm Surg 1990;85:267-272.
20. Vaughan ED: The radial forearm free flap in orofacial
reconstruction: personal experience in 120 consecutive
cases. J Cranio-MaxilkrFac Surg 1990;18:2-7.
21. Villaret DB, Futran NA: The indications and outcomes in
the use of osteocutaneous radial forearm free flap. Head
Neck 2003;25:475-481.
22. Werle AH, Girod DA, Tsue TT, et al.: Osteocutaneous
radial forearm ree flap: its use without significant donor
site morbidity. Otolaryngol Head Neck Surg 2000;123:
711-717.
"C"ree jejunal autografta (FJAs) have a unique place in
r the history ofmicrovascular surgery because this was
the first tissue to be tnll:lSplanted in humans. Seidenberg
et al. (50) conducted a number of canine experiments
involving FJA transfers to the head and neck to replace
the pharyngoesophagus. In 1959, they reported their
experience in one patient who underwent a pharyngoe-
sophagectomy for reCUITeilt cancer. The patient sur-
vived for 5 days until a cerebrovascular accident caused
his death. However, the findings at autopsy revealed a
viable jejunal transplant. This procedure was performed
without the benefit of a microscope. The arterial anas-
tomosis was sutured, and the venous anastomosis was
performed by using a tantalum ring prosthesis (28).
Although 1959 marks the beginning of the era of
human free tissue tnll:lSfer, the concept of moving tissue
around the body and reestablishing its circulation was
first introduced in the writings of Ale:zis Carrel (9).
In 1907, he reported his experimental work in an ani-
mal model with organ transfers, including the success-
ful autotransplantation of a segment of jejunum to the
neck. He described the resumption of peristaltic activity
after completing the microvascular anastomoses.
The first successful transfer of an FJA in which a
patient was able to resume swallowing was described
by Roberts and Douglas (46) in 1961 .Mter the intro-
duction of the operating microscope for performing
microvascular anastomoses, this technology was rapidly
applied, both clinically and experimentally. In 1966,
Green and Som (17) performed a number of animal
experiments involving the transfer ofFJAs with the help
of the microscope. In addition, they introduced the
455
456 CHAPTER 24
concept of the split jejunal patch graft, which was cre-
ated by incising the jejunum along its antemesenteric
border to change it from a mucosa-lined tube to a flat
mucosal patch. The split jejunal autograft has since been
used to reconstruct various noncircumferential defects
of the upper aerodigestive tract.
In addition to being the first reported free tissue
transfer, the FJA has perhaps been written about the
most of any reconstructive free flap in the head and
neck. Various issues related to the technique, periop-
erative management, and postoperative function are
addressed in this chapter.
FLAP DESIGN AND UTILIZATION
Prior to its use in humans as a free flap, the jejunum
was introduced as a pedicled intestinal transfer to
reconstruct the thoracic esophagus. Wiillskin (59)
reported this technique in 1904. Pedicled segments of
jejunum were transferred by cutting the first two vas-
cular arcades and basing the segment on the vascular
supply through the third and fourth arcade. However,
ischemic necrosis of the most cephalad portion of the
jejunum was common, and this technique was restricted
to esophageal defects that did not extend above the infe-
rior pulmonary vein (18). The concept of "turbocharg-
ing" the pedicled jejunum was introduced by Chang
et al. {10) in 1985. To circumvent the problem of
ischemia in the most cephalad portion of the pedicled
jejunum, they described the reestablishment of flow
through the first or second arcade by anastomoses to
recipient vessels in the neck. This partially attached and
partially free segment of jejunum provided an effective
conduit for the replacement of the entire thoracic and
cervical esophagus.
Currently, the jejunum is primarily transferred as a
microvascular autograft. It is most commonly used as
a mucosal tube or a mucosal patch, depending on the
configuration of the defect. A reliable and functional
reconstruction of pharyngoesophageal defects has been
pursued by head and neck surgeons for many decades.
Pedicled cervical flaps were popularized by Wookey
{60) in the early 1940s, but this technique was criticized
because of the necessity for multistage procedures. The
use of a skin graft over a stent was also advocated but
was soon superseded by pedicled regional flaps, the most
important of which was the tubed, deltopectoral flap,
which was introduced for pharyngoesophageal recon-
struction by Bakamjian {1) in 1965 (Figs. 5-7 to 5-9).
Bakamjian's technique was a significant advance. How-
ever, it was still limited by the necessity for a two-staged
procedure. Other regional flaps, most importantly the
pectoralis major musculocutaneous flap, were applied
to this problem by fashioning them into an epithelial
conduit. However, these flaps were often too bulky to
form into a tube and, therefore, were unreliable for
circumferential defects. The problem of bulk was cor-
rected in part by the use of free cutaneous flaps, which
could be more readily formed into a tube in a one-stage
procedure (22).
The transfer of pedicled visceral flaps, such as the
stomach and colon, has been used extensively to replace
the thoracic and cervical esophagus. However, both of
these mucosal flaps require extensive abdominal and
thoracic dissections, which are fraught with complica-
tions and are not warranted when the defect is con-
fined to the cervical region. The gastric pull-up (GPU)
procedure is also problematic because of its limited
reach when used to resurface defects that extend more
cephalad into the oropharynx. When a GPU is used in
heavily radiated tissues, its weight and the effect of the
gravitational pull on it have also led to problems with
wound healing. The colonic interposition is useful for
thoracic esophageal replacement, but its limited reach
has restricted its use when the defect extends into the
cervical region. The bacterial flora of the colon can also
lead to infectious problems in the abdomen and chest.
The FJA was therefore introduced and popularized
as a solution for reconstructing circumferential defects
that were limited to the neck (3,34). Vascularized seg-
ments of ileum have also been used (21). Unrestricted
by a vascular pedicle based in the abdomen, the FJA
can easily be used for defects that extend more cephalad
into the oropharynx and nasopharynx. Partially split
and partially circumferential segments of jejunum can
readily satisfy the need for a circumferential mucosal
tube to replace the pharyngoesophagus and a flat piece
of mucosa to replace the posterior pharyngeal wall all
the way up to the base of the skull (Fig. 24-1). This
partially tube-shaped and partially split design can also
be used for reconstructing the cervical esophagus and
hypopharynx when the larynx is not involved, the func-
tion of which can be preserved.
The caliber of the lumen of the FJA matches the
esophagus fairly well in most individuals. However, in
a pharyngoesophageal reconstruction, the pharyngeal
opening may be considerably larger. The cephalad por-
tion of the FJA can be opened along its antemesenteric
border to achieve a caliber that is more suitable for
anastomosis. When the defect extends to the oral cav-
ity following glossectomy and requires reconstruction
of the entire floor of the mouth along the inner table
of the mandible, the necessity to enlarge the lumen is
even greater. Jones et al. {26) proposed a design for the
FJA that involved folding the "split jejunum" on itself
to effectively double the size of the lumen. The ability
to achieve a tension-free closure of the oral cavity seem-
ingly offsets the considerably longer suture line when
the FJA is sutured to itself.
When considering the three major options for
one-stage reconstruction of cervical esophageal or

FIGURE 24-1. A segment of jejunum can be partially or completely divided in a longitudinal
fashion along its antemesenteric border. By so doing, we can create either a completely flat
mucosal patch or a partially tube-shaped and partially flat segment of mucosa.
pharyngoesophageal defects, there are a number of
points to be made. The GPU and colonic interposition
are the only methods that allow resection and replace-
ment of the thoracic esophagus. However, this is done
at the expense of significant abdominal and thoracic
dissections. In addition, the limited cephalad reach has
been noted. Swallowing after a GPU is also problematic
becauae of early satiety and reflux of food into the oro-
pharym. and <n'8l cavity.The use of tube-shaped cutane-
ous free flaps offers tremendous fi=ibility in design and
little potential morbidity compared with that of a lapa-
rotomy. However, when the inferior cervical esophageal
margin is very low, leaving a pro.zimal thoracic esopha-
geal stump that is tucked into the close confines behind
the tracheal remnant, it is often difficult to perform an
anastomosis betWeen the tube-shaped skin fiap and the
esophagus. This can be done effectively by using an FJA
with an enteric stapling device that produces a safe and
reliable anastomosis without the requisite exposure that
is needed for the placement of circumferential sutures.
However, the surgeon must be mindful of the potential
complications that can occur with this approach. Jeju-
noesophageal anastomoses that are located in the upper
mediastinum carry an additional risk if an anastomotic
leak occurs. The recognition of such a leak and the abil-
ity to manage it effectively are compromised. Finally, the
FREE JEJUNAL AUTOGRAFT 457
occurrence of a stricture in this location is also far more
difficult to manage than one that is located in the neck
(44). In our experience, the "short proximal esophagus"
represents a clear indication for the harvest of an FJA
and provides an alternative solution to the GPU when
the thoracic esophagus itself is not involved by cancer.
However, as noted above, the surgeon must be mindful
of the potential complications of this technique but, in
the author's opinion, it represents a safer approach than
a GPU when the thoracic esophagus is not involved in
the disease process.
The FJA has also been successfully uaed for recon-
structing the pbaryngoesophagus in benign conditions,
such as strictures caused by prior surgery, radiation
therapy, or gastroesophageal rdiux and fistulas that
have not responded to conservative treatment. This may
be accomplished with either a patch graft or a circum-
ferential tube, depending on the status of the native
mucosa (7,23,25,37,43).
Reconstruction of mucosal defects of the oral cavity
have evolved tremendously during the 1960s and 1970s.
Early in the 1960s, the primary emphasis was placed
on restoring a watertight seal and avoiding a multistage
procedure through the creation of an orostoma. With
the introduction of new techniques, primarily the safe
and reliable transfer of free flaps, the emphasis changed
458 CHAPTER 24
so that restoration of function became the primary
consideration. With this in mind, the quality of the tissue
that is used to replace the oral mucosal lining has taken
on paramount importance. Thin pliable tissue is required
when a partial glossectomy is performed and the mainte-
nance of tongue mobility is desired. This can be readily
accomplished with thin cutaneous free flaps, such as the
radial forearm, lateral arm, and anterolateral thigh flaps,
with the added advantage of also restoring sensation to
that tissue. However, the introduction of cutaneous flaps
into the oral cavity is not an exact replica of the native
oral lining. The rationale of using a split FJA in this set-
ting is that it transfers a moist mucosal lining (8). This
may be particularly advantageous in patients with severe
xerostomia caused by prior radiotherapy. Experimental
studies in animals showed that the FJA conforms well to
the three-dimensional contour of the oral cavity, and the
serosa adheres to the denuded or partially cut mandible
(52). Another application for jejunum in the oral cavity
was reported by Black et al. (6) who transferred an FJA
to reconstruct a total palatal defect. The mucosal sur-
face was used on the oral side, and the serosal surface
was placed on the nasal side of the defect. This provided
a functional separation of the oral and nasal cavities,
which permitted the patient to resume oral nutrition.
The FJA has also been used in experimental settings
to investigate its potential use as a replacement for the
trachea in patients who have developed symptomatic
tracheal stenosis. Jones et al. (27) introduced the idea
of using an FJA for this purpose and reported their suc-
cessful replacement of one half of the circumference of
the trachea over a length of four rings using a vascular-
ized mucosal patch. Letang et al. (31) replaced entire
segments of the cervical trachea in dogs with a jejunal
tube. A silicone tube was placed as an intraluminal stent
for 2 weeks. The authors reported favorable results in 12
dogs that were followed for a maximum of 60 days after
surgery. From this investigation, it was concluded that
the FJA was a promising substitute for the trachea and
that the potential problem of excess secretions in the
airway was not realized. Constantino et al. (13) reported
similar findings in a group of eight dogs that underwent
circumferential replacement of the trachea. Although
mucus production was not problematic, the authors
noted a period of intermittent peristalsis in the FJA,
which caused airway obstruction. This occurred despite
placement of both a temporary intraluminal stent and
a permanent rigid mesh tube on the outside of the FJA
to which it was sutured. They raised concerns about the
reliance on the FJA to attach to the surrounding tissues
in the neck by fibrosis to achieve a stable airway when a
rigid framework was not added. Somatostatin was used
to reduce the peristaltic activity of the FJA.
As an extension of this application of the FJA, Zeis-
mannetal. (61) described theuseofasegmentofjejunum
to produce speech following laryngopharyngectomy.
A portion of the bowel was used to replace the gullet,
and a segment was sutured to the end of the trachea
to create a shunt for air to pass into the neopharynx.
The midpoint of the loop of jejunum was sutured to
the floor of mouth so that the tracheal limb rose to a
higher level than the pharyngojejunal anastomosis. By
so doing, aspiration was prevented. In three patients
who were treated in this manner, fluent speech was
achieved in two, and none of the patients had signifi-
cant aspiration.
NEUROVASCULAR ANATOMY
The small intestine extends from the pylorus to the ile-
ocecal valve and is divided into three segments: the duo-
denum, the jejunum, and the ileum. The duodenum and
the jejunum are demarcated by the ligament of Treitz.
The jejunum and ileum are attached to the posterior
abdominal wall by the mesentery and are arranged in
a series of loops. This fan-shaped mesentery contains
branches of the superior mesenteric artery and vein,
together with the autonomic nerves, lymph nodes, lym-
phatic channels, and a variable amount offat (Fig. 24-2).
The superior mesenteric artery is the second ventral
unpaired branch of the abdominal aorta. The jejunal
arteries (range, 12 to 15) arise from the left side of the
superior mesenteric artery. They run nearly parallel to
one another, with each vessel dividing into two arteries
that unite with the adjacent branches to form a series of
convex arches within the mesentery. A second and third
series of arches may arise, particularly in the more distal
segment. Each successive arch has a smaller caliber as
the wall of the jejunum is approached. Eventually, small
straight vessels (vasa recta) arise, which pass alternately
to one side or the other of the small intestine. The veins
that supply the jejunum and ileum accompany the arter-
ies and drain into the superior mesenteric vein. There
are often two venlU comiranres for each mesenteric artery.
PREOPERATIVE ASSESSMENT
A thorough history and physical examination must be
performed during the initial evaluation to determine
whether there are any contraindications to performing an
FJA. The two major factors in the recipient site that argue
against FJA transfer are the absence of suitable recipi-
ent vessels and the extension of disease into the thoracic
esophagus (4). The presence of either of these problems
represents an absolute indication for the use of a GPU
or colonic interposition. However, with the expansion of
recipient vessel possibilities through the use of the inter-
nal mammary vessels, there are now very few patients in
whom a free tissue transfer cannot be performed, even in
the previously operated and irradiated neck.
 FREE JEJUNAL AUTOGRAFT 459

FIGURE 24-Z. The transition from duodenum to jejunum occurs at the ligament ofTreitz. The
jejunum and ileum are arranged in a series of loops. The vascular supply to the small bowel runs
in the mesentery along with the lymphatics. A segment of jejunum to be used as an FJA can be
harvested from virtually any location.
460 CHAPTER 24
There are a number of donor-site factors that should
point the sw:geon toward an alternative method of
reconstruction. The presence of ascites is considered
a contraindication to harvesting an PIA. Likewise,
chronic intestinal diseases, such as Crohn's disease,
should alter a sw:geon's plan to harvest a jejunal seg-
ment. Relative contraindications to performing an PIA
are a history of extellSive prior abdominal surgery or
intraperitoneal sepsis, both of which predispose to the
development of adhesions. Patients with compromised
pulmonary reserve are at increased risk of postoperative
complications when a laparotomy is performed, and this
factor should be considered when an alternative form of
reconstruction is available. The development of safe lap-
aroscopic techniques for the harvest of PIAs will proba-
bly gready reduce the momidity of an open laparotomy.
SURGICAL TECHNIQUE
As with all large head and neck procedures in which
the upper aerodigestive tract is violated, periopera-
tive anu"b:iotics are administered. However, because
of the violation of the lower intestinal tract, additional
antimicrobial precautions, such as the use of a bowel
preparation, have been discussed but remain controver-
sial. As early as 1959, experimental studies by Lillehei
et al. (32) did not demonstrate any increase in survival
in dogs that had undergone bowel sterilization prior to
clamping of the superior mesenteric artery for a 5-hour
period. Additional experiments in dogs by McGill
et al. (35) in 1979 showed no decrease in the extent of
mucosalinjurywhenintraluminalantisepsiswasinstituted
before rendering a segment of jejunum ischemic for
FIGURE 24-3. A segment of proximal jejunum is selected for the graft. Transillumination is help
ful to visualize the vascular arcade and to ensure that a feeding artery and vein of sufficient size
for anastomosis are present. The view shown is from the patient's head.
90 minutes. Although preoperative bowel preparation has
been advocated by a few authors, most do not cOilSider it
essential for free jejunal transfers (12,15,16).
A two-team approach is usuaDy instituted when using
an PIA to recOilStruct a head and neck defe<:t. The general
sw:gical service that harvests the jejunal segment must be
attuned to the needs of the microsurgeon, and therefore,
preoperative and intraoperative communication is essen-
tial for a successful outcome. The same holds true when
the reconstructive team differs from the ablative team
and the fate of potential recipient vessels is at stake.
The operation begins with an upper midline laparot-
omy and identification of the ligament ofTreitz. There
is some controversy in the literature regarding the ideal
location of the jejunum that is to be harvested. Some
authors propose harvesting from just beyond the liga-
ment ofTreitz, while others advocate segments that are
100 to 150an. distal to this point (36,41,53). Virtually
every location between these extremes has also been
proposed (3,30,42,53,58). We prefer to use a segment
of proximal jejunum. Transillumination is helpful to
highlight the vascular arcades (Fig. 24-3), which can
facilitate selection of a suitable segment of jejunum
(12,14). The particular segment must be supplied by a
single vascular arcade with nutrient vessels of sufficient
size for microsurgical trallSfer.
The size of the PIA is determined by the dimellSions
of the defect. Once the appropriate segment has been
chosen, the mesentry is divided proximally and distally,
carefully preserving the feeding vessels. An additional
window can be made in the mesentery to mark the
junction between the PIA and a short proximal seg-
ment that will serve as an exteriorized monitor for graft
viability (discussed below), as shown in Figures 24-4
 FREE JEJUNAL AUTOGRAFT 461

_.,.......,._ Pharyngoesophageal
segment

a;.;_,-.,_- - External monitor

segment

FIGURE 24-4. A small segment of jejunum can be transferred with the FJA to serve as a moni-
tor of the circulation. The vascular supply to the monitor is derived from the primary mesenteric
vessels that supply the main portion of the jejunum used in 1he reconstruction.
462 CHAPTER 24
FIGURE 24-5. The mesentery is divided along the proximal
and distal ends of the chosen segment, being careful to
preserve the vascular arcade and the feeding vessels. An
additional window is made in the mesentery to mark the
separation between the jejunal graft {yellow arrowt and a
short proximal segment (blue arrow) that will be used as an
exteriorized monitor for graft viability. The view shown is
from the patient's head.
and 24-5. The jejunum is divided with a stapling device
(Fig. 24-6A,B). It is imperative that the orientation of
the FJA be marked with a suture so that isoperistal-
tic reconstruction of the pharyngoesophagus can be
achieved. The FIA is left attached to its nutrient sup-
ply in the abdomen until the head and neck defect is
prepared. At this point, using Ioupe magnification, the
vessels are carefully prepared by isolating the artery
and vein to be used for the microvascular anastomosis
(Fig. 24-6C). Once the recipient site and vessels are
prepared, the pedicle is divided close to the origin of
the donor artery and vein (Fig. 24-7). Particular atten-
tion must be given to handling the mesenteric vein,
which is fragile and easily injured. Intestinal continuity
is restored with a hand-sewn two-layered end-to-end
jejunojejunostomy. The mesenteric trap must be closed
to prevent the risk of internal hernia.
Insetting of the FJA into a pharyngoesophageal
defect (Fig. 24-8) is performed by careful distal
anastomosis of the jejunal segment to the proximal
esophageal lumen (Fig. 24-9). In most instances, the
caliber of the jejunum matches favorably with that of
the esophagus. The surgeon should anticipate that the
FJA will lengthen after it is revascularized. In addi-
tion, the mismatch of the FJA and the pharynx must
be accounted for, before discarding any of the proxi-
mal portion of the jejunal segment. The caliber of the
pharynx is usually much greater than the jejunum,
and the performance of a tension-free anastomosis is
facilitated by making a cut along the antemesenteric
border in order to widen the proximal opening (Figs.
24-10 and 24-11). Mter this maneuver is performed,
the back wall of the proximal repair is performed fol-
lowed by placement of a salivary bypass tube (Figs.
24-12 and 24-13). The repair is completed by sutur-
ing the anterior wall of the pharyngojejunal anasto-
mosis (Fig. 24-14).
There has been considerable controversy in the litera-
ture regarding the necessity for improving the FIA tolel.'l-
ance to ischemia through the use of pharmacotherapy or
hypothermia. Fisher et al. (15) advocated using a hypel.'l-
osmolar perfusate infused into the mesenteric artery to
stabilize the endothelium and prolong the ischemic tol-
erance. Heparinized saline has been instilled through
the FIA to asanguinate the vascular system (41,58).
However, McKee and Peters (36) reported no diffel.'l-
ence in outcome between a group of FJA transfers in
which a heparin perfusate was used and a second group
in which there was no perfusion.
There is general agreement that the ischemic tolel.'l-
ance ofan FJA can be increased by instituting hypothel.'l-
mia during the transfer period prior to reestablishing the
flap's circulation. Lillehei et al. (32) demonstrated that
FJAs that were cooled to 5C could survive an ischemic
period of 5 hours. McGill et al. (35) reported a signifi-
cant improvement in the extent of mural injury follow-
ing 90 minutes of ischemia when an FIA was cooled to
between 6C and 12C. Elaborate methods to achieve
cooling have been proposed, such as a system of glass
rods placed in the lumen through which water at 10C
was continuously instilled (53).
In most situations in which ischemia time is kept to
a minimum, pharmacotherapy and hypothermia are not
believed to be ne<:essary. Ullehei et al. (32) and Mullens
and Pezacki (39) reported successful FJA transfers in an
animal model following 2 hours of euthermic ischemia.
Reuther et al. (45) also reported that some therapeutic
value was achieved by a 2-hour period of warm ischemia,
leading to a decrease in the amount of mucus produo-
tion from the FJA. There are two situations in which
ischemic tolerance may be an issue and warrant the con-
sideration of instituting hypothermia during the ischemic
period. Walkinshaw et al. (57) reported a lesser tolerance
to ischemia for distal jejunal segments compared with
those harvested from the region close to the ligament of
Treitz.When prior abdominal surgery limits the selection
of jejunum to this region, then the surgeon should have
a heightened concern about the length of the ischemic
period. The other circumstance in which improving
ischemic tolerance should be entertained is during salvage
procedures for a failing FJA. Under conditions in which
the exact length of secondary bowcl ischemia cannot be
determined, it may be helpful to use a pe:rfusate and hypo-
thermia to enhance the chances for successful salvage.
As noted previously, the circulation to the FIA should
not be interrupted until the head and neck defect is

prepared and the reCJplent vessels are isolated and
examined under the microscope. Insetting of the FJA
should be performed before the microvascular anasto-
moses for a variety of reasons. The length and geometry
of the vascular pedicle can be more precisely gauged
when the enteric anastomoses have been completed,
thereby establishing the exact position of the FJA (55).
The enteric anastomoses, in particular that of the esoph-
agus to the jejunum, may be technically difficult to per-
form, especially when the proximal esophageal stump is
low in the thoracic inlet. This anastomosis is made even
more difficult when performed on a revascularized bowel
because of the bleeding end, the postischemic engorge-
ment, and the limited mobility of the FJA for fear of
disrupting the microvascular anastomoses. As noted
previously, the size mismatch between the jejunum and
the pharynx, following laryngopharyngoesophagectomy,
often requires opening of the antemesenteric border of
the FJA (Fig. 24-2). The FJA should be inset in an isop-
eristaltic direction when it is used as a tubular conduit.
In addition, it is important to have the FJA under a slight
amount of stretch during insetting of the ischemic bowel
to account for the elongation that occurs after revascular-
ization and to avoid the problem of a redundant conduit.
The use of an enteric stapling device to perform the
jejunoesophageal anastomosis was mentioned earlier.
This technique is reserved for those cases in which cir-
cumferential sutures cannot be placed or are placed only
with difficulty and compromised precision. Schuster-
man et al. ( 49) reported a 33% incidence of stenosis for
stapled anastomoses compared with a 17% incidence
for anastomoses that were sewn. Green and Som (17)
cautioned against the use of a running suture, which
led to a higher incidence of stenosis in their experience.
The use of the stapler for the proximal pharyngojeju-
nal repair is rarely feasible because of the mismatch in
luminal size. Following inset of the FJA and comple-
tion of the revascularization, any redundant mesentery
should be used to cover the enteric anastomoses as long
as it does not lead to a distortion ofthe vascular pedicle.
Mter the reconstruction is completed, the wounds
are closed and drained with either passive or suction
drains. Gastric decompression is usually performed with
a nasogastric tube. Once the postoperative ileus resolves,
typically around the third or fourth postoperative day,
the nasogastric tube can be removed and enteral feeding
initiated with a jejunostomy tube placed at the time of
the initial laparotomy. Oral nutrition is usually delayed
for a week in nonirradiated patients and up to 2 weeks
when the neck was previously irradiated.
POSTOPERATIVE MANAGEMENT
Aside from the routine postoperative care following a major
head and neck procedure, the problem of monitoring
FREE JEJUNAL AUTOGRAFT 463
the circulation to the FJA must be addressed. Different
techniques have been proposed. However, few satisfy the
criteria for the ideal monitor, which include (a) continu-
ous, (b) rapid determination of either arterial or venous
compromise, (c) noninvasive, (d) reliable, (e) objective,
and (f) easy to perform (56). Some authors have reponed
monitoring the circulation with implantable devices, such
as probes for Doppler sonography, which register the flow
of blood across the anastomoses. Temperature probes
have also been used to monitor changes in the temper-
ature recorded from the intraluminal mucosal surface
(49). The problem with all implantable devices is that any
changes that occur raise the question of whether the cir-
culation has changed or the probe has become dislodged.
Direct graft observation can easily be performed
when the defect extends into the oral cavity or orophar-
ynx, where the mucosa can be observed to assess the
color, bleeding to needle stick, and peristalsis to manual
stimulation. The problem arises when the FJA is not
accessible unless a laryngoscope or a fiberoptic nasopha-
ryngoscope is passed (53,54). Both of these techniques
are problematic because of limited exposure, patient
discomfort, and the fact that they are not continuous
and must be performed by experienced personnel.
The most reliable method of FJA monitoring is to
observe a portion of the flap directly. This was initially
done by exteriorizing a portion of the serosal surface
through a "window" in the skin that was either covered
with Silastic or a skin graft (24). The most direct method,
however, was introduced by Katsaros et al. (29) in 1985.
They described exteriorizing a small segment of jejunum
that was completely isolated from the "primary" segment
of jejunum, except for its vascular connections to the
main vascular pedicle through a common mesentery (Fig.
24-8). We adopted this technique and found it to be effec-
tive. To avoid potential soilage of the neck, the two ends of
the exteriorized jejunal segment are stapled, and then the
serosa is sutured to the skin opening in the neck through
which it is passed. A small opening into this blind loop
must be created to allow an egress for enteric secretions.
An ostomy bag may be placed on the monitor to help
prevent skin breakdown and infection. The exteriorized
jejunum can be excised under local anesthesia and the
mesentery ligated 5 to 7 days after surgery (Fig. 24-9).
A barium swallow is usually performed 7 to 12 days
after surgery, depending on whether prior radiation ther-
apy was administered to the neck. The jejunostomy tube
can be removed when the patient's oral intake is suffi-
cient. In those patients who are to undergo postopera-
tive adjuvant radiotherapy, we routinely leave the feeding
tube in place until after that treatment is completed.
It is well established that postoperative adjuvant radi-
otherapy can be administered without a significant risk
of complications. Several studies have demonstrated
that doses of 6,000 cGy produce few, if any, adverse
sequelae (5, 15,54). McCaffrey and Fisher (33) reponed
464 CHAPTER 24
an increased incidence of graft failure in a small number
of FJAs placed into previously irradiated defects. How-
ever, the group of patients that were reconstructed with
FJAs and treated with postoperative radiotherapy had
an equal incidence of complications compared with a
similar group of FJA recipients who were not treated
with radiotherapy. Mustoe et al. ( 40) confirmed the
tolerance of FJAs to radiation in a dog model. Despite
a submucosal inflammatory response and flattening of
jejunal folds, they did not find an increased incidence
of mucosal slough or stricture formation. Gullane et al.
(19) described a functional benefit derived from post-
operative radiotherapy, which resulted from decreased
mucus production and decreased intrinsic motility. The
latter finding was believed to produce improved degluti-
tion by producing a more passive conduit.
POTENTIAL PITFALLS
The outcome of FJA transfers to the head and neck
depends in part on the defect to which these are
applied. Most series report experience with FJAs used
for circumferential pharyngoesophageal defects. Early
outcome measures for this type of defect include the
success of flap transfer, avoidance of salivary fistulas,
and avoidance of stricture formation. The long-term
measures of success include the functional parameters
of swallowing and speech.
Transfer of an FJA is a relatively safe and reliable
method of reconstruction. Shangold et al. (51) com-
piled a meta-analysis of 633 reported cases in the
literature in which there was a 4.4% incidence of peri-
operative death. Successful FJA transfer was reponed in
91.1% of cases. Fistulas occurred in approximately 18%
of cases in which FJAs were used to reconstruct the
pharyngoesophagus. In those patients who developed
salivary fistulas, most occurred at the proximal anasto-
mosis between the jejunum and pharynx. In two-thirds
of cases, the fistulas resolved without additional surgery.
The most important functional parameter following
the reestablishment of continuity between the pharynx
and the esophagus is the resumption of oral intake.
Swallowing can be assessed on many different levels, of
which the most basic is the ability of the patient to sat-
isfy all nutritional requirements by oral intake to main-
tain normal body weight. However, swallowing may be
more critically assessed based on the type of food that
the patient is able to eat and also the time that it takes
for the completion of an average meal. Both of these lat-
ter factors may have a significant impact on a patient's
ability or willingness to eat meals with family members
and to dine in public restaurants. Detailed data are
not available in most published articles on FJAs. How-
ever, in the review by Shangold et al. (51), 81.8% of
402 patients were able to maintain their nutrition
entirely by oral intake. Most series reported some degree
of dysphagia in all patients in whom the reconstruction
was done with an FJA. However, it is important to rec-
ognize that the extent of the defect in a pharyngoesoph-
ageal reconstruction may vary considerably. Patients
who have also lost portions of their tongues would be
expected to experience greater swallowing problems,
regardless of the method used to restore the pharyngoe-
sophageal segment (48).
The actual function of the FJA during the act of
swallowing is somewhat controversial. Although some
investigators maintain that the FJA plays a role in pro-
pelling a food bolus into the esophagus by coordinated
peristalsis, there is considerable evidence to the con-
trary (27,38,46). Manometric and electrical tests dem-
onstrated a variable amount of contractile activity in the
FJA but did not show that the contractions occurred
in a coordinated fashion (30,46). Nakamura et al. (41)
stated that the motility of an FJA diminishes over time.
The diminution in peristaltic activity leading to the
development of a more passive conduit may help to
explain why swallowing tends to improve with time fol-
lowing FJA reconstruction. In an attempt to eliminate
the factor of early FJA contraction, Harashina et al. (20)
described a technique whereby an FJA was harvested
at a length that was twice the size of the defect. The
FJA was opened on its antemesenteric border and then
sutured to itself to achieve twice the diameter of the fab-
ricated gullet. This not only doubled the lumen but also
disrupted the muscular ring of the FJA. Swallowing was
thought to be improved by using this technique.
The reports on alaryDgeal speech rehabilitation in
patients who have undergone FJA reconstructions are
limited. This may be a reflection of the difficulties that
these patients encounter in the restoration of this function.
Batitis et al. (2), howevt:r, reponed that all six patients in
their series were able to achievt: a voice using a duckbill
prosthesis placed through a tracheojejunal shunt. There are
additional reports ofpatients with FJA reconstructions who
were able to gain functional neoesophageal speech (51).
There is a range of abdominal complications that
have been reponed following the harvest of a seg-
ment of jejunum. They are reported to occur with a
frequency of 5.8%. The most common complications
include abdominal wound dehiscence, bowel obstruc-
tion, gastrointestinal hemorrhage, G-tube leakage, and
prolonged ileus (51).
The FJA is unique among the free flaps used in head
and neck reconstruction because of its serosal surface,
which is believed to be the reason why the FJA does
not undergo neovascularization and remains dependent
on its vascular pedicle. In most free tissue transfers, the
flap is usually no longer dependent on the microvascu-
lar pedicle by the third week after surgery. The process
and degree of neovascularization depend on the quality
of the vascularity of the recipient bed, which depends

on the amount of scarring that results from prior sur-
gery and radiation. Late graft necrosis, occurring up to
18 months after surgery, has been reported as a result of
transection of the vascular pedicle ( 4 7,54).
Various other potential problems have been reported.
Redundancy of the pharyngoesophageal reconstruc-
tion has been described, which is caused by an under-
estimation of the true length of the FJA when inset in
its ischemic state (11). Unlike the rugae of the gastric
mucosa, which disappear over time, the plicae circula-
ris of the jejunum tend to persist. These mucosal folds
often trap food particles and lead to halitosis (37).
The transfer of FJAs is a highly reliable and safe
method of restoring defects of the upper aerodigestive
tract with vascularized mucosa. However, the success
FREE JEJUNAL AUTOGRAFT 465
of this procedure depends on the successful completion
of five anastomoses: two microvascular anastomoses
to restore circulation and three enteric anastomoses to
restore continuity of the gullet and the small intestine.
The failure of any one of these anastomoses may result
in a significant complication requiring further surgery,
prolonged hospitalization, and untoward long-term
sequelae that may have a severe detrimental effect on
the patient,s function and quality of life.
Acknowledgments
The authors would like to acknowledge the contribu-
tions of Dr. Michael Sullivan to the writing of this chap-
ter in the first edition of this book.
466 CHAPTERZ4

Harvest of Free Jejunal Autograft

FIGURE 24-6. A: The jejunum is divided with a

stapler at the three sites where the mesentery
has been cleared (A). B: The two segments of
the FJA have been isolated but remain per-
fused, while further isolation of the mesenteric
artery and vein is performed. C: With Ioupe or
microscope rna gnification, the donor vessels
are meticulously dissected. C
 FREE JEJUNAL AUTOGRAFT 467

Harvest of Free Jejunal Autograft

FIGURE 24-7. The mesenteric vessels sup-

plying the vascular arcade to the graft are
only divided after 1he recipient site has been
prepared. The artery and vein have been marked
with red snd blue pins, respectively. Note the
silk suture (srrow)1hat was placed earlier
during 1he dissection to mark the proximal end
of1he graft in order to ensure an isoperistaltic
reconstruction. The vascular supply to the moni-
tor is derived from 1he primary mesenteric ves-
sels that supply 1he main portion of the jejunum
used in the reconstruction.
468 CHAPTERZ4

Reconstruction of a Circumferential Pharyngoesophageal Defect

FIGURE 24-8. Atotal laryngopharyngectomy

defect has been created with the tracheal
lumen, the proximal esophagus, and the phar-
ynx indicated in the dissection.

FIGURE 24-9. A FJA has been transferred to

the defect and the distal anastomosis to the
esophagus completed. The vascular pedicle
and the monitor segment have been oriented to
the right side of the repair. The monitor seg-
ment can be exteriorized either at the proximal
or distal end of the FJA while ensuring that the
isoperistaltic orientation of the FJA is main-
tained.
 FREE JEJUNAL AUTOGRAFT 469

Reconsh'uction of a Circumferential Pharyngoesophageal Defect

FIGURE 2410. The antemesenteric border of

the FJA is cut in order to widen 1he lumen of the
FJA for anastomosis to 1he pharynx.
470 CHAPTER Z4

Reconstruction of a Circumferential Pharyngoesophageal Defect

Monitoring
segment

FIGURE 2411. The configuration of the FJA is shown after transection of the antemesenteric
border to create a larger caliber lumen of the proximal portion of the bowel.
 FREE JEJUNAL AUTOGRAFT 471

Reconsh'uction of a Circumferential Pharyngoesophageal Defect

FIGURE 24-12. The posterior wall of the

anastomosis is repaired prior to completing the
anterior wall.

FIGURE 24-13. In this instance, a salivary

bypass tube has been placed in order to provide
further protection of the enteric anastomotic
suture lines.
412 CHAPTER Z4

Reconstruction of a Circumferential Pharyngoesophageal Defect

FIGURE 24-14. The anterior wall of the enteric

repair is completed by subJring the anterior wall
of the FJA to the base of tongue. The monitor
segment (yellow arrow) and the donor vessels
Iwhite arrow} are shown. The monitar segment
is brought out through either an existing suture
line or through a separate incision where the
serosa is sutured to the skin margins.

REFERENCES 5. Biel MA, Maisel RH: Postoperative radiation-essociated

1. BakamjianVY: A two-stage method for pharyngoesopha-
geal reconstruction with a primary pectoml skin flap.
Plan Reeomtr Surg 1965;36: 173.
2. Batitis H, Stallinga JO, Ban J: A reliable method for moni-
toring the microva&cular patency of free jejunal t:ran.sfers
in reconstructing the pharynx and cervical esophagus.
Plan Reeomtr Surg 1989;83:896-898.
3. Berger A. TlZian C, Hausam.en J, Schulz-Coulon H,
Loblcin D: Free jejunal graft for reconstruction of oral,
oropharyngeal, and pharyngoesophag defectli.J Reconstr
Miao~ 1984;1:83.
4. Biel MA, Maisel RH: Free jejunal autograft recon-
struction of the pbaryngoesophagus: review of a
10-year experience. Otblaryngol Head Neck Surg 1984;
97:369.
changes in free jejunal autograft&. Arch Ot.oltnyngol Head
Neck Surg 1992;118:1037.
6. Black P, Bevin G, Arnold P: One-stage palate recon-
&truction with a free ncovasculari2cd jejunal graft. Pltut
Reeomtr Surg 1971;47:316.
7. Brain RHF: The place for jejunal transplantation in the
treatment of simple strictures of the esophagus. Ann R
Coil Smg 1967;40:100.
8. Buckspan GS, Newton ED, Franklin JD, Lynch )B: Split
jejunal free-tissue transfer in oml pharyngoesophageal
reconstruction. Plast Reconsw Surg 1986;77:717-728.
9. Carrel A: The surgery of blood vcs&el&, etc. Johm Hopkins
Hosp Bull1907;190: 18.
10. ChangT,WangW. Huang 0: One-stage rcconatruction of
esophageal defect by free transfer of jejunum: treatment
and complications. Ann Plan Surg 1985;51 :492.

11. Coleman ll III, Searles JM Jr, Hester TR. et al.: Ten
years experience with free jejunal autograft. Am J Surg
1987;154:394-398.
12. Colemanll III, Tan KC, Searles JM, HesterTR. Nahai F:
Jejunal free autograft: analysis of complications and their
resolution: Plast Recunst:r Surg 1989;84:589--595.
13. Constantino P, Nuss D, Snyderman C, et al.: Experimen-
tal tracheal replacement using a revascularized jejunal
autograft with an implantable macron mesh tube. Ann
Otol Rhinol Laryngol 1992; 10 1:807.
14. Deane IM, Gilbert DA, Schecter GL, Baker JW: Free
jejunal transfer for the reconstruction of pharyngeal and
cervical esophageal defects. Ann Plast Surg 1987; 19:499.
15. Fisher SR. Cole TB, Meyers WC, Seigler HF: Pharyngo-
esophageal reconstruction using free jejunal intaposition
grafts.Arch OtolaryngolHeadNeckSurg 1985;111:747-752.
16. Flynn MB, Acland R: Free intestinal autograft& for
reconstruction following pharyngolaryngoesophagec-
tomy. Surg Gynecol Obstet 1979;749:858.
17. Green GE, Som ML: Free grafting and revascularization
of intestine. I. Replacement of the cervical esophagus.
Surgery 1966;60: 10 12.
18. Grimes OF: Surgical reconstruction of the diseased
esophagus. Part I: Intetposition of the jejunum. Surgery
1967;61:325.
19. Gullane P, Havas T, Patterson A, Todd T, Boyd B: Pharyn-
geal reconstruction: current controversies. J Otolaryngol
1987;16:169--173.
20. Harashina T, Inoue T, Andoh T, Sugimoto C, Fujino T:
Reconstruction of cervical oesophagus with free double-
folded intestinal graft. Br J Plast Surg 1985;38:483--487.
21. Harashina T, Kakegawa T, Imai T, Suguro Y: Secondary
reconstruction of oesophagus with free revascularized
ileal transfer. Br J Plast Surg 1981;34:17.
22. Harii K, Ebihara S, Ono I, Saito H, Terui S, Takato T:
Pharyngoesophageal reconstruction using a fabricated
forearm free flap. Plast Reconst:r Surg 1985;75:463--476.
23. Hester TR. McConnel F, Nahai F, Cunningham SJ,
Jurkiewicz MJ: Pharyngoesophageal stricture and fistula.
Ann Surg 1984;199:762-769.
24. Hester TR. McConnel FM, Nahai F, Jurkiewicz MJ,
Brown RG: Reconstruction of cervical esophagus, hypo-
pharynx and oral cavity using free jejunal transfer. Am J
Surg 1980;140:487-491.
25. Jones B, Gustavson E: Free jejunal transfer for recon-
struction of the cervical oesophagus in children. A report
of two cases. Br J Plast Surg 1983;36: 162.
26. Jones NF, Eadie P, Myers E: Double lumen-free jejunal
transfer for reconstruction ofthe entire floor ofmouth, phar-
ynx. and cervical esophagus. Br J Pfast Surg 1991;44:44.
27. Jones R. Morgan R, Marcella K. Mills S, Kron I: Tracheal
reconstruction with autogenous jejunal microsurgical
transfer. Ann Thorac Surg 1986;41 :636.
28. Jurkiewicz MJ: Vascularized intestinal graft for recon-
struction of the cervical esophagus and pharynx. Plast
Recomt:r Surg 1965;36:509.
FREE JEJUNAL AUTOGRAFT 473
29. Katsaros J, Banis JC, Acland RD, Tan E: Monitoring free
vascularized jejunum grafts. Br J Plast Surg 1985;38:22.
30. Kerlin P, McCafferty GJ, Robinson DW, Theile D: Func-
tion of a free jejunal "conduit" graft in the cervical esoph-
agus. Gastroenterology 1986;90: 1956-1963.
31. Letang E, Sanchez-Uoret J, Gimferrer J, Ramirez J,
Vicens A: Experimental reconstruction of the canine tra-
chea with a free vascularized bowel graft. Ann Thorac Surg
1990;49:955.
32. I...illehei RC, Goon B, Miller FA: The physiological
response of the small bowel of the dog to ischemia includ-
ing prolonged in-vitro preservation of the bowel with suc-
cessful replacement and survival. Ann Surg 1959; 150:543.
33. McCaffrey TV, Fisher J: Effects of radiotherapy on the
outcome of pharyngeal reconstruction using free jejunal
transfer. Ann Otol Rhinol Laryngo/1987;96:22.
34. McDonough ll, Gluckman JL; Microvascular
reconstruction of the pharyngoesophagus with free
jejunal graft. MICrosurgery 1988;9: 116.
35. McGill CW, Taylor BH, Flynn MB, Acland RD, Flint
LM: Effects of cooling and intraluminally administered
antiseptics on surgically induced ischemia of the intestine
in dogs. Surg Gynecol Obstet 1979;149:377-379.
36. McKee DM, Peters CR: Reconstruction of the hypo-
pharynx and cervical esophagus with microvascular jeju-
nal transplant. Clin Plast Surg 1978;5:305.
37. McLear P, Hayden R. Muntz H, Fredrickson J: Free flap
reconstruction of recalcitrant hypopharynges.] stricture.
AmJ Otolaryngo/1991;12:76.
38. Meyers WC, Seigler HF, Hanks JB, et al.: Postoperative
function of "free" jejunal transplants for replacement of
the cervical esophagus. Ann Surg 1980;192:439-450.
39. Mullens JE, Pezacki EJ: Reconstruction of the cervi-
cal esophagus by revascularized autograft& of intestine:
an experimental study using the Inokuchi stapler and
reporting the use of revascularized intestine to construct
an artificial human larynx. Int Surg 1971;55:157.
40. Mustoe T, Fried M, Horowitz Z, Botnick l.E, Strome M:
Reconstruction de l'oesophage cervical par transplant
libre d'anse jejunale. Etude experimentale chez la chien.
Ann Otolaryngol Chir Ceuicojac (Paris) 1986; 102:227-233.
41. Nakamura T, Inokuchi K. Sugimuchi K: Use of revascu-
larized jejunum as a free graft for cervical esophagus. Jpn
J Surg 1975;5:92.
42. Nozaki M, Huang TT, Hayashi M, Endo M, Hirayama
T: Reconstruction of the pharyngoesophagus following
pharyngoesophagectomy and irradiation therapy: Plast
Recunst:r Surg 1985;76:386-394.
43. Pash AR, Putnam T: Jejunal interposition for recur-
rent gastroesophageal reflux in children. Am J Surg
1985; 150:248.
44. Peters CR. McKee DM, Berry BE: Pharyngoesophageal
reconstruction with revascularized jejunal transplant. Am
J Surg 1971;121:675.
45. Reuther JF, Steinaw H, Wagner R: Reconstruction oflarge
defects in the oropharynx with a revascularized intestinal
474 CHAPTER 24
graft: an experimental and clinical report. Plan Reconstr
Surg 1984;73:345.
46. Roberts RE, Douglas FM: Replacement of the cervical
esophagus and hypopharynx by a rewscularized free jeju-
nal autograft: report of a case successfully treated. N Engl
J Med 1961;264:342.
4 7 _ Salamoun W, Swartz WM, Johnson JT, et al.: Free jejunal
transfer for reconstruction of the laryngopharynx. Otolar-
yngol Head Neck Surg 1987;96:149-150.
48. Schecter GL, Baker ]w, Gilbert DA: Functional evalu-
ation of pharyngoesophageal reconstruction techniques.
Arch OtolaryngolHeadNeck Surg 1987;112:40.
49. Schusterman MA, Shestak K, de Vries EJ, et al.: Recon-
struction of the cervical esophagus: free jejunal transfer
versus gastric pull-up. Plast Reconst.r Surg 1990;85: 16-21.
50. Seidenberg B, Rosmak SS, Hurwittes, et al.: Immediate
reconstruction of the cervical esophagus by a rewscular-
ized isolated jejunal segment. Ann Surg 1959;149:162.
51. Shangold LM, Urken MI.., Lawson W: Jejunal transplan-
tation for pharyngoesophageal reconstruction. Otolaryn-
gol Clin NorthAm 1991;24:1321.
52. Sheen R, Mitchell M, Macleod A, O'Brien B: Intraoral
mucosal reconstruction with microvascular free jejunal auto-
grafts: an experimental study. Br J Plast Surg 1988; 41 :521.
53. Shumrick DA, Savoury LW: Recent advances in laryn-
gopharyngoesophageal reconstruction. Acta Otolaryngol
Suppl (Stockh) 1988;458:190.
54. Theile DE, Robinson DW, McCafferty GJ: Pharyngolar-
yngectomy reconstruction by rewscularized free jejunal
graft. Aust N Z J Surg 1986;56:849.
55. Urken MI.., Vickery C, Weinberg H, Buckbinder D, Biller
HF: Geometry of the vascular pedicle in free tissue trans-
fers to the head and neck. Arch Otolaryngol Head Neck
Surg 1989;115:954-960.
56. Urken MI.., Weinberg H, Vickery C, Buckbinder D, Biller
HF: Free flap design in head and neck reconstruction to
achieve an external segment fur monitoring. Arch Otolar-
yngolHeadNeck Surg 1989;115:1447-1453.
57. Walkinshaw M, Downey D, Gottlieb JR: Ischemic injury
to enteric free flaps: an experimental study in the dog.
Plast Reconst.r Surg 1988;81:939.
58. Wang ID, Sun YE, Chen Y: Free jejunal grafts for recon-
struction of the pharynx and cervical esophagus. Ann
Otol Rhinol Laryngo/1986;95:348.
59. Wiillskin L: Ueber antethorakale Oesophago-
jejunostomie und Operationen nach gleichem Prinzipo.
Dtsch MedWochenschr 1904;31:734.
60. Wookey H: Surgical treatment of carcinoma of the
pharynx and upper esophagus. Surg Gynecol Obstel
1942;75:499.
61. Zeismann M, Boyd B, Manktelow R, Rosen 1: Speak-
ing jejunum after laryngopharyngectomy with neo-
glottic and neopharyngeal reconstruction. Am J Surg
1989;158:321.
~e function ofthe omentum in the abdomen has cap-
~ tured the imagination of physicians and philosophers
for centuries. Morison (23) referred to the greater omen-
tum as the "abdominal policeman."Wilkie (37) reported
other early opinions as to the function of the omentum,
notiD,g that Aristode ascribed a proteaive role in prevent-
ing cold from reaching the viscera. He also reported that
Verhagen viewed the omentum as a protector of the vis-
cera against sudden jars and friction. Hansen believed
that the omentum helped to pull the stomach downward
when it was full and to assist the downward excursion of
the diaphragm in respiration. At times, the omentum has
been given an almost mystical intelligence and motility
that allowed it to seek areas of injury and disease.
Many of these qualities have been refuted through
animal experiments that have shown that the omentum
does not possess a higher intelligence or a separate
motility. It travels about the abdomen by way of intesti-
nal peristalsis and the movement of the diaphragm. The
omentum has the ability to form adhesions that help
to wall off inflammatory processes and toxic substances
within the abdomen. The omentum is also endowed
with a rich vascular and lymphatic netWork that allows
it to absorb large quantities offtuid (30).
Buncke (5) is credited with being the first to inve5-
tigate the potential transfer of the greater omentum in
an experimental setting. McLean and Buncke (20) were
the first to use the greater omentum as a free flap clini-
cally when they reconstructed a large scalp defect with
omentum that was covered with a skin graft. nus was
soon followed by a report by Harii and Ohmori (15)
in 1973 where the greater omentum was used in two
415
476 CHAPTER Z5
cases. They successfully transferred the omental flap to
resurface the scalp in one patient and the forehead in
another. In addition, these authors reported the use of
the gastroepiploic pedicle as recipient vessels to supply a
free flap used for chest wall reconstruction. Kiricuta and
Goldstein (19) described other uses of the omentum in
reconstructive surgery.
For many years, the stomach has been viewed as a
reservoir of mucosa to reconstruct the upper alimen-
tary tract. A variety of different procedures, includ-
ing the gastric pull-up and the reversed gastric tube,
have been used in certain situations to reconstruct the
thoracic and cervical esophagus (16,39). Seidenberg
et al. (32) reported the first successful transfer of tissue
from a distant site to the head and neck in humans, using
microvascular techniques. They used a segment of jeju-
num to reconstruct the pharyngoesophagus following
a laryngopharyngectomy. In 1961, Hiebert and Cum-
mings (17) reported the first successful transfer of a seg-
ment of the gastric antrum to reconstruct the cervical
esophagus and pharynx. They described the microvas-
cular aspect of this procedure as "Lilliputian surgery"
and relegated the technique to those situations in which
a gastric pull-up or colon interposition could not be
performed.
Interest in the stomach as a potential donor site
waned in the 1960s and 1970s, in large part due to the
enthusiasm of surgeons and researchers for the experi-
mental and clinical work using free jejunal transfers.
Renewed interest in the use of the free gastric mucosal
flap began with the experimental work ofPapachristou's
group. In 1977, these investigators described the use of
a pedicled gastric island flap that was based on the left
gastroepiploic vessels, which was passed through a sub-
cutaneous tunnel along with the greater omentum to
reach the head and neck (28). In a subsequent article
published in 1979, Papachristou et al. (29) reported
their experimental work in canines with free gastric
mucosal flaps to resurface portions of the pharynx and
cervical esophagus. These flaps were based on the left
gastroepiploic artery and vein. Although this technique
was successful in the 16 animals that were able to eat
I week following surgery, the experience in I animal that
died was instructive. In that particular case, a flap was
obtained from the acid-secreting part of the stomach.
The animal died from a perforated ulcer in the esopha-
geal wall opposite the flap. From this experience, the
authors cautioned against the harvest of antral mucosal
flaps from acid-secreting portions of the stomach.
The first successful transfer of a composite flap of
greater omentum and stomach was reported by Baudet
(3) in 1979.This flap was used for the secondary closure
of a pharyngostome. The right gastroepiploic pedicle
was selected along with a more proximal portion of the
greater curvature of the stomach for the harvest of the
mucosa. This provided a vascular pedicle that was 30 em
in length and allowed the microvascular anastomoses to
be performed to the subscapular artery and vein in the
axilla. The greater omentum was draped over the ante-
rior neck and then covered with a split-thickness skin
graft. The patient resumed an oral diet on the 17th day
after surgery. In addition, pH values of 5 were obtained
from the reconstructed region.
The gastro-omental free flap was subsequently
popularized by Panje et al. (25) in 1987 when they
reported its successful transfer in five of seven patients
with defects of the oral cavity and pharynx. The right
gastroepiploic vessels were used as the vascular pedicle,
and the authors described the successful application
of a stapling device for flap harvest and simultaneous
closure of the stomach. The authors delineated the
attributes of the gastric mucosa, which are that it is not
hair-bearing tissue, pliable, and readily fashioned to
restore complex three-dimensional defects of the upper
aerodigestive tract. The smooth mucosal surface of the
stomach was considered to be a better replacement for
the digestive tract than skin or jejunum, both of which
tend to trap food. The persistence of circular folds of
the jejunum is less desirable than the rugae of the stom-
ach that tend to flatten within a few days after transfer.
As with free jejunal grafts, the early excessive mucus
production of the gastric mucosa may be problematic.
The quantity of mucus usually subsides over time but
is sufficient to provide relief of xerostomia in the pos-
tradiation patient. Panje et al. (25) also noted the use of
the greater omentum to provide coverage of the carotid
artery, augmentation of soft tissue, and a bed for split-
thickness skin grafting. They reported a 50% atro-
phy in the volume of the omental flap during the first
6 months. The extensive lymphatic channels within the
omentum led to speculation that it might be advanta-
geous in reducing facial edema following radical neck
dissection and in forming an immunologic barrier to
the spread of cancer.
Moran et al. (22) investigated a number of questions
related to the gastro-omental flap in an animal model
in which a patch of gastric mucosa was transferred
to the floor of mouth, and the omentum was placed
in the neck following lymphadenectomy. A pH probe
recorded measurements in the range of 6.4 to 6.9 from
the oral secretions up to 5 months following surgery.
The histologic evaluation of the gastric mucosa revealed
atrophy and fibrosis, which may partially explain the
diminished acid production. In addition, interstitial
lymphoscintigraphy was performed following injection
of technetium-99m antimony sulfide colloid into the
gastric mucosal flap. The postinjection images showed
uptake in two to three nodes in the omentum, which
was interpreted as preliminary evidence that the omen-
tum may provide drainage to help decrease edema
and may provide possible oncologic benefits through
immunosurveillance.

FLAP DESIGN AND UTILIZATION
The greater omentum is a double layer of peritoneum
that hangs like a sheet from its major attachments to
the greater curvature of the stomach and the transverse
colon (Fig. 25-1). The blood supply to this structure
arises from the right and left gastroepiploic vessels,
which run in the cephalad edge of the omentum where
it attaches to the stomach. Numerous epiploic branches
are given off from these primary vessels, which descend
toward the free edge of the greater omentum. The inter-
connecting arcade of vessels within the omentum per-
mits dividing and lengthening the omental apron while
still maintaining its viability (1,7). The omentum has
been used as a pedicled flap in a variety of different
reconstructive tasks, including the head and neck, tho-
rax, and perineum (11-13,19).
Considerable variation exists in the length and width
of the greater omentum of the virgin abdomen. In an
extensive study of 200 cadavers and 100 laparotomies,
Das (7) reported a rough correlation between the size of
the omentum and the patient's height and weight. How-
ever, he warned that the ability to predict the length and
width of the omentum in any individual situation based
on these parameters is limited. Prior abdominal surgery
and peritonitis both cause significant scarring and con-
traction of the omentum, which usually makes it unsuit-
able for use in reconstructive procedures.
The application of the greater omentum to a vari-
ety of different reconstructive problems is a direct out-
growth of our understanding of its range of function
within the abdomen. In 1911, Rubin (30) dispelled
the belief that the omentum was capable of independ-
ent motility, chemotaxis, and intelligence. Rubin also
refuted the ability of the omentum to repair visceral
perforations spontaneously or to restore vascularity to
nonviable tissues. Attributes of this unusual organ that
continue to be upheld are its ability to form adhesions
to tissues that are ischemic, inflamed, or contaminated,
so that they become excluded from the remainder of the
abdomen. It achieves this through fibroblast and cap-
illary ingrowth. Although unable to revascularize and
restore viability to necrotic tissue, the omentum is capa-
ble of surrounding and encapsulating such tissues. The
omentum also appears to have the ability to promote
hemostasis when applied to raw surfaces through the
activation ofprothrombin and the conversion of fibrino-
gen to fibrin. Although unable to restore vitality to non-
vital tissues, the omentum appears to be able to provide
neovascularization to ischemic tissues through capil-
lary ingrowth. Yonekawa and Yasargil (38) investigated
the ability of the omentum to revascularize an ischemic
brain when applied to its denuded surface. Tbrough
animal studies, these authors found that the ingrowth
of vessels after the removal of the arachnoid layer
provided protection against cerebral infarction when
FREE OMENTUM AND GASTRO-OMENTUM 477
the middle cerebral artery was ligated. Vineberg (35)
reported his experience using omentum to revascular-
ize the ischemic heart. His technique entailed removal
of the epicardium and revascularization of the myocar-
dium by implantation of the internal mammary artery
into the myocardium of the left ventricle. In addition, a
free nonvascularized omental graft was placed over the
myocardium to serve as a conduit for the ingrowth of
mediastinal vessels. Vineberg's study was purported to
demonstrate that the omentum was a critical avenue for
vessel ingrowth and the prevention of extensive myocar-
dial infarction.
The final quality that has been attributed to the
omentum is the absorption of fluid through its extensive
lymphatic channels. The omentum reportedly absorbs
one-third of the fluid that is removed from the perito-
neal cavity. This absorptive function and rich lymphatic
network led to speculation that the omentum may be
valuable in relieving chronic lymphedema of the extrem-
ities. Harii (14) reported no reduction in swelling when
omental transfers were applied clinically to this prob-
lem. There has been continued speculation about the
potential role of the omentum in relieving facial edema
following radical neck dissection (25) .
The omentum has been used for protection in a vari-
ety of surgical procedures, such as to wrap a prosthetic
vascular graft following radical resection of an inguinal
tumor (11). Goldsmith et al. (13) transferred pedi-
cled omentum through an opening in the diaphragm
to reinforce an anastomotic suture line in the thoracic
esophagus. Goldsmith and Beattie (12) also described
a technique of transferring omentum through a sub-
cutaneous tunnel to protect the carotid artery against
salivary contamination or exposure following the break-
down of cervical skin flaps. As an extension of this tech-
nique, Freeman et al. (10) reported favorable results
using omentum to protect the carotid artery following a
gastric pull-up or colon interposition for the reconstruc-
tion of pharyngolaryngoesophagectomy defects.
Since the initial publication by McLean and Buncke
(20) on the use of omentum to cover a large scalp
defect, there have been a number of additional reports
describing similar uses for the omentum (15,18). Brown
et al. ( 4) reported their experience in reconstructing an
extensive midface defect by using a free omental flap to
cover a nonvascularized framework composed of split
rib grafts. Panje et al. (27) expanded the range of defects
in the head and neck reconstructed with an omental
free flap to include a stomal recurrence and osteoradi-
onecrosis of the mandible and maxilla.
The greater omentum has been used in treating osteo-
myelitis of the extremities and the head and neck (2).
Sanger et al. (31) reported a case of chronic osteomyeli-
tis of the skull that was successfully treated by aggressive
debridement followed by coverage with a vascularized
omental transfer and long-term antibiotics. The authors
418 CHAPTER 2S

Gastroduodenal a. Left
gastroepiploic a.

Right
gastroepiploic a.

Accessory
omental a. omental a.

Right
omental a.

FIGURE 25-1. The blood supply to the greater curvature of the stomach and to the greater omentum is derived from the right
and left gastroepiploic artery and vein. Both of these vessels most commonly arise from the celiac axis. The four major omental
arteries are shown, but the patterns of branching and anastomoses within the omentum are variable.

favored the omentum to achieve coverage because of its
pliability, its predictable vascularity, and its ability to pro-
vide increased vascularity to the bone for improved anti-
biotic delivery. Moran and Panje (21) reported the use
of free omentum to achieve coverage following seques-
trectomy of the mandible affected by osteoradionecrosis.
The pliability and rich vascularity of the omentum are
important attributes for this clinical application.
The final application of the free omental fiap in
head and neck reconstruction is to augment soft tis-
sue defects to restore facial contour. Wallace et al. (36)
were the first to descn'be the use of this flap to treat
hemifacial microsomia. They noted that the omentum
is well suited for this problem because of the ability
to compartmentalize the fl.ap and the natural feel that
omentum creates in the augmented side of the face.
Upton et al. (33) reported on a more extensive series
using omental free flaps to correct the facial deform-
ity of hemifacial atrophy and microsomia. Emphasis
was placed on creating three subcutaneous pockets
and compartmentalizing the omentum in these pockets
to avoid migration caused by gravity. In addition, the
three segments of omentum were sutured to the fascial
layer in the temporal and midface regions.
Das et al. (8) studied the long-term dfects of using
nonvasc:ularized omental grafts to achieve soft tissue
augmentation. In a rabbit model, it was found that the
peripheral zones of the omentum survived through

neovascularization but the central zones underwent
nea:osis, whic:h was proportional to the size of the graft.
Three months after surgery, some of the 1aJ:ger Baps
weighed only 37% of their original weight. Howm::r, with
lODger periods, the wei,ght ofthe o:II'Imtal gndts inaased in
proportion to the increase in the body weight of the animal.
As noted previously, Hiebert and Cummings (17)
performed the first successful gastric mucosal free
flap in the early 1960s. They used a segment of gastric
mucosa that was shaped into a tube for replacement
of a circumferential defect of the pharyngoesophagus.
The rich vascularity of the stomach permits a large
segment of mucosa to be transferred to complete the
reconstruction of virtually any mucosal defect in the
head and neck. Baudet (3) described the transfer of
a composite gaatro-omental free flap (Fig. 25-2). The
RGURE 25-2. A gastroomental flap has been harvested,
and the defect in the stomach has been closed. This can
be effectively performed with a stapling device. The right
gastroepiploic artery is a more favorable choice for supply-
ing this flap. It is imperative nut to harvest gastric mucosa
in the vicinity of the pylorus to avoid causing a gastric ouUet
obstruction. A longer pedicle may be achieved by harvesting
the mucosal flap at a greater distance from the pylorus.
A portion of or the entire greater omentum may be harvested.
FREE OMENTUM AND GASTROOMENTUM 479
omentum may be used for virtually all the indications
outlined above. In addition, a piece of the omentum
may be exteriorized in a suture line to serve as a moni-
tor of the vascularity of the entire flap (34). Calteux
et al. (6) described the use of the gastro-omental flap
to reconstruct the oral cavity following total glosseo-
tomy. The omentum was wrapped around the man-
dible and also placed beneath the mucosal graft to
provide bulk. These authors reported that a segment
of mucosa, measuring 15 X 10cm, may be harvested
from the greater curvature and body of the stomach.
Six weeks after surgery, acid secretion from the trans-
planted gastric mucosa was measured after pentagas-
trin stimulation and found to be 25 times less than in
the normal stomach. The pH of the oral secretions was
7. They reported no intraoral or esophageal ulceration.
Panje and Moran (26) reported that the secretions from
gastric mucosal grafts were alkaline and attributed this
to the histologic finding of progressive atrophy of the
gastric glands. An added factor determining the nature
of the secretions from the transplanted gastric mucosa
is the fact that it is also denervated and is no longer
under vagal control. Following truncal vagotomy, there
is an average 85% reduction in basal acid secretion and
a 50% reduction in muimal acid output. The dener-
vated parietal cell has a markedly reduced sensitivity
to circulating gastrin, which is probably also true of the
parietal cells in the transplanted mucosa.
There are two major factors that determine the posi-
tion along the greater curvature for mucosal harvest. The
first is concern for narrowing the region near the pylorus,
which could lead to gastric outlet obstruction. The sec-
ond consideration is related to the length of the vascu-
lar pedicle. Increased length of the right gastroepiploic
artery and vein can be obtained by harvesting mucosa
from the more proximal portion of the greater curvature.
NEUROVASCULAR ANATOMY
The stomach has a rich vascular supply making it pos-
sible to perform such procedures as the gastric pull-up.
The dominant blood supply to the greater curvature
is derived from the right and left gastroepiploic artery
and vein. Both of these arteries are terminal branches
of the celiac axis. The right gastroepiploic artery arises
from the gastroduodenal artery, which is a branch of the
common hepatic artery. There is considerable variabil-
ity in the branching pattern of the celiac uis with the
gastroduodenal artery arising from a variety of different
sources, including the superior mesenteric artery or
directly from the aorta. The anterior and posterior
superior pancreaticoduodenal arteries are usually given
off prior to the point at which the right gastroepiploic
artery assumes a course along the greater curvature.
The left gastroepiploic artery is a branch of the splenic
artery that arises at a variable distance proximal to the
480 CHAPTER Z5
hilum of the spleen (Fig. 25-1). Both arterial systems
give rise to a series of corporeal branches that supply
the stomach. The right and left gastroepiploic systems
anastomose with each other to a variable extent and on
a variety of different anatomic levels. These anastomo-
ses occur along the greater curvature, in the submucosal
layer of the stomach through the gastric branches, and
in the omentum by the epiploic branches (9).
The right gastroepiploic artery is usually dominant
when compared with the left in terms of its size and
the distance that it courses along the greater curvature.
The diameter of the right gastroepiploic artery ranges
from 1.5 to 3.0mm; the left gastroepiploic artery has
a diameter of 1.2 to 2.9mm. The pattern of branching
of the main omental arteries is variable. However, there
are usually right, left, and middle omental arteries with
an accessory omental artery arising proximally from the
right gastroepiploic artery (Fig. 25-1). The branching
pattern of the omental vessels has been divided into five
different types. The presence of these patterns in a given
patient is important from a surgical perspective if there is
a need to lengthen or divide the greater omentum (1,33).
SURGICAL TECHNIQUE
After administering proper perioperative antibiotics, an
upper midline abdominal incision is made. The omentum
is separated from the transverse colon, being careful to
identify and preserve the gastroepiploic vessels during this
dissection. The stomach and its attached omentum are
then delivered into the wound, and the location and length
of the greater curvature to be used for the reconstruction
are selected (Fig. 25-3). Care must be taken to stay proxi-
mal enough relative to the pylorus to avoid narrowing the
gastric outlet when the graft is harvt:sted. Complete mobi-
lization of the stomach requires freeing any attachments
of the posterior gastric wall in the lesser sac (Fig. 25-4).
The gastric portion of the graft is harvested using multiple
firings of a stapling device (Fig. 25-5). The omental flap
to be included with the graft is isolated by serially divid-
ing and ligating the omentum vertically along the medial
and lateral boundaries of the gastric portion of the graft.
Prior to dividing the distal end of the gastric flap, the right
gastroepiploic artery and vein are identified and the addi-
tional length on these vessels is preserved for the anasto-
mosis. The vessels are divided once the recipient site in
the head and neck has been prepared (Fig. 25-6).
When used to reconstruct a pharyngoesophageal
defect following a laryngopharyngectomy (Fig. 25-7),
a tubed portion of the greater curvature is harvested
that is of sufficient length to bridge the distance
between the pharynx and the esophageal opening. The
length of greater curvature should be slightly longer
than anticipated because a portion of the antemes-
enteric border should be opened in order to increase
the caliber of the lumen. The distal repair may be
performed with sutures or an endostapling device
(Fig. 25-8). The proximal anastomosis is sutured,
starting with the posterior wall and then complet-
ing the repair with the anterior wall suture line (Fig.
25-9). The greater omentum is then transposed over
the enteric anastomoses as well as the microvascular
pedicle. In addition, vital mediastinal structures and
the entire trachea opening can be draped with this
thin pliable tissue (Fig. 25-1 0).
POTENTIAL PITFALLS
Microvascular surgery has greatly expanded the available
donor sites and the range of tissues from which to choose
for the reconstruction of a given defect in the head and
neck. The desirability of restoring mucosal defects of the
upper aerodigestive tract with mucosal flaps from the
gastrointestinal tract is self-evident. However, the neces-
sity for a laparotomy should not be taken lightly. The
surgeon must ask the critical question of whether the
tissue that is to be harvested from the abdominal cavity
offers a distinct functional or aesthetic advantage that
justifies the risks of the abdominal procedure.
A wide range of intra-abdominal complications may
occur following gastro-omental flap harvt:st including
gastric leak with peritonitis and intra-abdominal abscess
formation. Upton et al. (33) advised the placement of
sutures between the greater curvature of the stomach
and the transverse colon to help avoid the formation of a
volvulus. Gastric outlet obstruction is another potential
problem that may occur if the mucosal flap is too large
and harvested too close to the pylorus. A careful his-
tory and physical examination should be obtained pre-
operatively to avoid doing a laparotomy in patients who
have had prior abdominal surgery or infection. A history
of gastric outlet obstruction or peptic ulcer disease is a
contraindication to using a gastro-omental flap.
Excess mucus production should be anticipated dur-
ing the postoperative period, and protection of the airway
is essential in patients with an intact larynx. The extent to
which the volume of the transferred omentum will dimin-
ish is somewhat controversial. Ohtsuka and Shioya (24)
reported no loss of volume following an extended follow-
up. However, Panje and Moran (26) noted omental atro-
phy in the range of 20% to 50% within the first 3 months.
Finally, it is evident that the harvest of a gastro-
omental free flap eliminates the possibility of perform-
ing a gastric pull-up. The surgeon must be mindful of
this fact when deciding to use this donor site to harvest
either a free gastric patch or tube.
Acknowledgments
The authors would like to acknowledge the contribu-
tions of Dr. Mack Cheney to the writing of this chapter
in the first edition of this book.
 FREE OMENTUM AND GASTROOMENTUM 481

Tubed Gastro-Omental Free Flap

FIGURE25-3. After delivering the stomach and

omentum into the wound, the gastric portion of the
flap is marked along the greater curvature. The
distal end of the graft should be several centime
ters proximal to the pylorus (blue arro'll! in order
to avoid narrowing the gastric outlet The size of
the graft depends on the defect to be filled, but
is generally Sto 14cm long and 3 to 4cm wide
(dashed white line). The gastroepiploic vessels
(short arrows) are seen running parallel to, and
approximately 1em away from, the greater curve
of the stomach.

FIGURE 25-4. Mobilization of the omentum

requires that it be freed from attachments to
the transverse colon (TC). The posterior wall
of the stomach also needs to be freed from its
attachments in the lesser sac. In this figure, the
stomach (St) is flipped cephalad. The gastroepi
ploic vessels (arrows) are clearly seen.
482 CHAPTERZ5

Tubed Gastro-Omental Free Flap

FIGURE 25-5. A: A stapling device is used to

divide the stomach, beginning at the proximal
end of the greater curvature and (B) continuing
toward the distal, pyloric end {arrow marks the
pylorus). The dashed white line indicates the
final line of stapling to complete the mucosal
harvest B
 FREE OMENTUM AND GASTROOMENTUM 483

Tubed Gastro-Omental Free Flap

FIGURE 25-6. Prior to dividing the distal end

of the graft, the right gastroepiploic vessels are
identified and additional length is preserved for
the anastomosis. The gastro-omental graft is
shown, with the right gastroepiploic artery and
vein marked with red and blue pins, respec-
tively. The vascular pedicle can be lengthened
by harvesting a mucosal flap at a distance
higher up along the greater curvature of the
stomach. In addition, safe proximal dissection
of the right gastroepiploic pedicle can be per
formed in order to achieve greater length.
484 CHAPTERZ5

Reconstruction of Circumferential Pharyngoesophageal Defect

FIGURE 25-7. A total laryngopharyngectomy

defect has been created to demonstrate the
inset of this flap for reconstruction of the pha-
ryngoesophageal segment.
 FREE OMENTUM AND GASTROOMENTUM 485

Reconsh'uction of Circumferential
~nelrvnaoe~sOJllh&lleal Defect

FIGURE 25-8. A B: The distal anastomosis of

the gastric mucosa to the esophagus has been
performed and the posterior wall of the proxi
mal enteric anastomosis has been completed.
A salivary bypass tube has been inserted. The
donor vessels are positioned laterally in the
neck for anastomosis (yellow arrow).
486 CHAPTERZ5

Reconstruction of Circumferential Pharyngoesophageal Defect

FIGURE 25-9. A. B: The anterior wall of the

proximal anastomosis has been completed
by suturing the gastric mucosa to the base of
tongue. B
 FREE OMENTUM AND GASTROOMENTUM 487

Reconsh'uction of Circumferential Pharyngoesophageal Defect

FIGURE 2510. The greater omentum is

draped over the entire surgical field including
the enteric suture lines and the microvascular
anastomoses. In addition, the mediastinal con
tents can be protected by providing coverage
with the omentum. Wrapping the trachea with
omentum beneath the skin to tracheal suture
line to formalize the stoma helps to protect
against a dehiscence, which can have devas-
tating consequences.

REFERENCES 6. Calteux N, Hamoir M, van den Eeckb.eut J, Vanioijck R:

1. Alday E, Goldsmith H: Surgical technique for omen-
tal lengthening based on arterial anatomy. Surg Gynecol
Obstet 1971;5:103-107.
2. Azuma H, Kondo T, Mikami M, Harii K: Treatment of
chronic osteomyelitis by ttansplantation of autogenous
omentum with microvascular anastomosis. Aaa Onhop
Scand 1976;47:271.
3. Baudet J: Reconstruction of 1he pharyngeal wall by free
transfer of the greater omentum and stomach. Im J
Microsurg 1979;1:53.
4. Brown R, Nahai F. Silverton J: The ome11tum in facial
r:onatruction. Br J Pkut Surg 1978;31 :58-62.
10. Freeman], Brondbo K, Osborne M, et al.: Greater omell-
5. Buncke HJ: Early experimental ome11tal tramplantation
by microvascular anastomosis. In: 7tansaaicns of the Sixth
l7JUTnarional Ctm.grus of.1'fastk and Reeorutnu:riue Surgery.
Paris: Masson; 1976:58.
Ret:onstruction of 1he floor of the mouth after total glos-
seao:my by free transfer of a gastro-om.e11tal flap. Head
Neck 1988;10:512-516.
7. Das S: The size of the human omentum and methods
of lengthening it for transplantation. Br J Platt Surg
1976;29;170-174.
8. Das S, Cragun J, Wheeler E, Goshgarian G, Miller T:
Free grafijng of the omentum for soft tissue augmenta-
tion: a preliminary study. Plast ReCO'IU1:f Surg 1981;68:
556-560.
9. El-Eisbi H, Ayoub S, Abd-el-Khalek:, M: The arte-
rial supply of the human stomach. Acta Anat 1973;86:
565-580.
tum used for carotid cover after pharyngolaryngoesopha-
gectomy and gastric "pull-up" or colonic "swing." Arch
Otolaryncol Head Neck Surg 1982;108:685-687.
488 CHAPTER Z5
11. Goldsmith H, Beattie E: Protection of vascular prosthe--
ses following radical inguinal excisions. Surg Clin North
Am 1969;49:413-419.
12. Goldsmith H, Beattie E: Carotid artery protection
by pedicled omental wrapping. Surg Gynecol Obsrn
1970;130:57-60.
13. Goldsmith H, Kiely A, Randall H: Protection of
intrathoracic esophageal anastomoses by omentum.
Surgery 1968;63:464--466.
14. Harii K: Clinical application offree omental transfer. Clin
Plast Surg 1978;5:273-281.
15. Harii K, Ohmori S: Use of the gastroepiploic vessels as
recipient or donor vessels in the free transfer of composite
flaps by microvascular anastomoses. Plast Recomtr Surg
1973;52:541-548.
16. Heimlich H: Use of a gastric tube to replace esophagus
as performed by Dr. Dan Gauriliu of Bucharest. Surgery
1957;42:693-699.
17. Hiebert CA, Cummings GO Jr.: Successful replacement
of the cervical esophagus by transplantation and revas-
cularization of a free graft of gastric antrum. Ann Surg
1961;154:103-106.
18. Ikuta Y: Autotransplant of omentum to cover large denu-
dation of the scalp. PlastRecomtr Surg 1975;55:490--493.
19. Kiricuta I, Goldstein A: The repair of extensive vesi-
covaginal fistulas with pedicled omentum: a review of
27 cases.J Uro/1972;108:724-727.
20. McLean D, Buncke H: Autotransplant of omentum to a
large scalp defect, with microsurgical revascularization.
Plast Recomtr Surg 1972;49:268-274.
21. Moran W, Panje W: The free greater omental flap for
treatment of mandibular osteoradionecrosis. Arch
Otolaryngol Head Neck Surg 1987; 13:425-427.
22. MoranW. Soriano A, Little A, Montag A, Ryan J, PanjeW:
Free gastro-omental flap for head and neck reconstruc-
tion: assessment in an animal model. Am J Otolaryngol
1989;10:55.
23. Morison R: Functions of the omentum. BMJ 1906;1:3.
24. Ohtsuka H, Shioya N: The fate of free omental transfers.
Br J Plast Surg 1985;38:478-482.
25. Panje WR. Little AG, Moran WJ, FeigU.Son M, Scher N:
Immediate free gastro-omental flap reconstruction of the
mouth and throat. Ann Om/Rhino/ Laryngol1987;95: 15--21.
26. Panje W, Moran W: Gastric mucosal and omental grafts.
In: Baker S, ed. Microsurgical Recomtruction of the Head
and Neck. NewYork: Churchill Livingstone; 1989:261.
27. Panje W, Pitcock J, Vargish T: Free omental flap recon-
struction of complicated head and neck wounds.
Otolaryngol Head Neck Surg 1989;100:588-593.
28. Papachristou DN, Fortner J: Experimental use of a gas-
tric flap on an omental pedicle to close defects in the tra-
chea, pharynx or cervical esophagus. Plast Reconstr Surg
1977;59:382-385.
29. Papachristou DN, Trichillis E, Forner JG: Experimental
use of free gastric flaps for the repair of pharyngoesopha-
geal defects. Plast Reconstr Surg 1979;64:336-339.
30. Rubin IC: The functions of the great omentum. A path-
ological and experimental study. Surg Gynecol Obstet
1911;12:117-131.
31. Sanger J, Maiman D, Matloub H, Benzel E, Gingrass R:
Management of chronic osteomyelitis of the skull using vas-
cularized omental transfer. Surg Neurol1982;18:267-270.
32. Seidenberg B, Rosenak S, Hurwitt E, Som M: Immediate
reconstruction of the cervical esophagus by a revascular-
ized isolated jejunal segment. Ann Surg 1959;149:162.
33. Upton J, Mulliken J, Hicks P, MUiray J: Restoration of
facial contour using free vascularized omental transfer.
Plast Reconstr Surg 1980;66:560.
34. Urken MI.., Weinberg H, Vickery C, Buchbinder D, Biller
HF: Free flap design in head and neck reconstruction to
achieve an external monitoring segment. Arch Otolaryngol
Head Neck Surg 1989; 115:1447.
35. Vineberg A: Revascularization of the right and left coro-
nary arterial systems: internal mammary artery implan-
tation, epicardiectomy, and free omental graft operation.
Am J Gardio/1967;19:344-353.
36. Wallace J, Schneider W, Brown R, Nahai F: Reconstruc-
tion ofhemifacial atrophy with a free flap of omentum. Br
J Plast Surg 1979;32: 15-18.
37. Wilkie D: Some functions and surgical uses of the
omentum. BMJ 1911;2: 1103.
38. Yonekawa Y, Yasargil M: Brain vascularization by
transplanted omentum: a possible treatment of cerebral
ischemia. Neurosurgery 1977;1:256-259.
39. Yudin S:The surgical construction of 80 cases of artificial
esophagus. Surg Gynecol Obstet 1944;78:561-583.
 ver the last several decades, we have gained a bet-
O ter understmding of the microanatomy of periph-
eral nerves, and the factors that influence their recovery
of function. This greater understanding, coupled with
advances in microsurgery, has resulted in more predict-
able functional results after nerve repair. The use of
autogenous nerve grafts for the repair of cranial nerve
defects following ablative surgery in the head and neck
has become a well-recognized technique (1-5). Tradi-
tionally, the great auricular and the sural nerves have
been the primary sources of donor autografta. However,
more recently, a branch of the median nerve, the medial
antebrachial cutmeous nerve (MACN), has been intro-
duced as an option for the repair of cranial nerve gaps
(6--1 0). The sensory distribution of this nerve includes
the anterior antecubital fossa and the ventral forearm
(Fig. 26-1). The donor site and anatomic features of the
nerve offer a number of advantages for the reconstruc-
tion of peripheral nerve gaps.
FLAP DESIGN AND UTILIZATION
The MACN provides 20 to 25 em of donor length, and
its cross-sectional diameter (1.5 to 2.0mm) matches
the diameter of most of the cranial nerves commonly
requiring reconstruction after ablative head and neck
surgery (V, vn:, X, XI, and XD). In addition, the proxi-
mal MACN, as it exits &om the medial cord of the bra-
chial plexus, is similar in diameter to the facial nerve
as it runs through the fallopian canal, making it use-
ful for grafting within the temporal hone. More distally
491
492 CHAPTER 2&

- - -- -+-Medial cord,
brachial plexus

;....--- +---Medial antebrachial

cutaneous n.

Posterior branch

FIGURE 2&-1. The MACN arises primarily from the medial cord of the brachial plexus with
contributions from the ventral rami of C8 and the first thoracic nerve. As it enters the arm, it lies
superficial to the brachial artery and runs in close proximity to the basilic vein. At the elbow, it
divides into posterior and anterior branches that supply sensation to the ulnar aspect of both
the flexor and extensor surfaces of the forearm.

it tapers, and exhibits a branching pattern that closely
resembles the peripheral branching pattern of the dis-
tal facial nerve. For this reason, the nerve is well suited
to facial nerve reconstruction from within the temporal
bone out to the peripheral branches.
The main donor site incision is well concealed on
the medial aspect of the arm, and the sensory deficit
is normally limited to a 6 X 6 cm2 area of the forearm,
which predictably diminishes in size over time. The fact
that the donor area can be accessed by a second surgi-
cal team at the time of surgical ablation expedites the
procedure.
NEUROVASCULAR ANATOMY
The MACN arises from the medial cord of the bra-
chial plexus, adjacent to the ulnar nerve, and carries
fibers from the eighth cervical and first thoracic nerves
(11-13). Proximally in the arm, it lies medial to the axil-
lary artery and, more distally, it lies anterior and medial
to the brachial artery. At the junction of the middle and
lower thirds of the arm, it pierces the brachial fascia
medially and lies in close association with the basilic
vein. It is at this point that it divides into anterior and
posterior (ulnar) branches (11).
The anterior branch of the MACN may pass superfi-
cial or deep to the medial cubital vein and then divides
into several branches that supply the anteromedial sur-
face of the forearm, extending to the wrist. The smaller
ulnar, or posterior branch, passes posterior to the medial
epicondyle of the humerus. It then branches to supply
the skin on the posteromedial aspect of the forearm.
ANATOMIC VARIATIONS
Masear et al. (7) performed 50 cadaveric dissections of
the MACN and found that the nerve arose from the
medial cord in 78% of specimens and from the lower
trunk in 22% of specimens. In 54% of the dissections,
there was a common origin of the MACN and the
medial brachial cutaneous nerve (T1) from either the
medial cord or the lower trunk. When this was the case,
the two nerves divided, on average, 6 em distal to the
common origin. Twenty-six nerves had a second medial
cutaneous nerve branch from the MACN (14).
Masear et al. (7) found that the distal nerve divided
into anterior and posterior branches an average of
14.5 em proximal to the medial humeral epicondyle.
The branches traveled together with the basilic vein
until the posterior branch turned in an ulnar direction.
MEDIAL ANTEBRACHIAL CUTANEOUS NERVE GRAFT 493
The anterior branch divided into two to five branches,
usually in a zone between 6 em proximal and 5 em distal
to the elbow. In 35% of specimens, there were articulat-
ing branches to the elbow (14).
The sensory distribution of the nerve includes
the skin of the distal arm, the anteromedial aspect of
the antecubital fossa, the posterior olecranon, and the
medial aspect of the forearm from the midline ventrally
to the midline dorsally.
POTENTIAL PITFALLS
During dissection of the MACN, the major concern is
inadvertent injury to the median nerve in the midpor-
tion of the arm (15). This can be avoided by using the
basilic vein as a landmark for the identification of the
MACN. Additionally, extensive proximal dissection of
the nerve may result in brachial plexus injury.
Inadvertent vascular injury is uncommon; however,
the basilic vein and the brachial artery are closely asso-
ciated with the course of the MACN. Therefore, these
structures must be identified and avoided during the
harvest of this graft.
PREOPERATIVE ASSESSMENT
The preoperative assessment for MACN transfer should
include a careful inspection of the upper arm. Patients
who have extensive subcutaneous tissue in this area may
be less favorable candidates for the use of this donor
nerve because of the difficulty in harvesting. In addition
to this, the upper arm and axilla should be examined
for any scars; previous surgery or trauma in this area is
a relative contraindication for the use of this nerve as a
donor graft. The patient should be educated preopera-
tively in regard to the sensory field deficit that will result
from the use of this nerve graft, and the risk of neuroma
formation. The probability of neuroma formation can
be significantly reduced by burying the proximal stump
of the nerve into an adjacent muscle belly (biceps or
triceps muscle).
POSTOPERATIVE WOUND CARE
After harvesting of the nerve graft, the subcutaneous tis-
sues are closed in two layers. The wound is not routinely
drained; however, a compressive dressing is applied over
the donor site and is left in place for 48 hours. Elevation
of the donor arm minimizes extremity edema.
494 CHAPTERZI

Medial Antebrachial Cutaneous

Nerve

FIGURE 2&-Z. An important topographical

landmark for dissection and identification of the
MACN is the medial epicondyle of the humerus.
In addition, the fascial plane separating the
musculi biceps brachii and triceps brachii
should be palpated and marked. The use of a
proximal tourniquet helps to engorge the basilic
vein, facilitating its identification.
FIGURE 2&-3. To expose the nerve completely,
a longitudinal incisior1 is made from the midarm
to the midforearm, just medial to the midsagittal
plane of the extremity. The dissection is begun
superiorly through the subcutaneous tissue
until the basilic vein (twa smslf arrows) is iden-
tified. This vein pierces the brachial fascia and
becomes more superficial as it descer~ds in the
arm. The nerve runs in close proximity to this
vein, and after identification of the basilic vein
is achieved, the MACN (farge arrow) is easily
identified and dissected, between the muscular
fascia ofthe biceps and triceps muscles. The
median nerve (three small arrows) lies just
abave the basilic vein and must be avaided.
With the proximal nerve identified, the more
distal branches can be traced until adequate
length is obtained. The anterior branch of the
MACN exhibits a consistent branching pat-
tern in the area of the antecubital fossa and
normally displays three to five large caliber
terminal branches.
 MEDIAL ANTEBRACHIAL CUTANEOUS NERVE GRAFT 495

Medial Antebrachial Cutaneous

Nerve

FIGURE 2&-4. With the proximal dissection of

the nerve complete, the division of the MACN is
identified, and the distal branches are followed
to obtain adequate length.

FIGURE 2&-5. The caliber, length, and branch

ing pattem of the MACN are consistent and
can be used to repair gaps from the main trunk
of the facial nerve as high as the second genu,
and as far distal as the midface region where
four or five distal branches can be reinnervated.
496 CHAPTER Z&
REFERENCES
1. Dellon AL, Mackinnon SE: Injury to the medial ante-
brachial cutaneous nerve during cubital tunnel surgery.
J Hand Surg [Br] 1985;10b:33.
2. Izzo KL, Aravabhuni S, Jafri A, Sobel E, Demopoulos
JT: Medial and lateral antebrachial cutaneous nerves:
standardization of technique, reliability and age effect on
healthy subjects. Arch Phys Med Rehahil1985;66:592.
3 . Kimura I, Ayyar DR: Sensory nerve conduction study in
the medial antebrachial cutaneous nerve. To'Mku J Exp
Med 1984; 142:461.
4 . Louie G, Mackinnon SE, Dellon AI..., Patterson GA,
Hunter DA: Medial antebrachial cutaneouslateral femo-
ral cutaneous neurotization in restoration of sensation to
pressure-bearing areas in a paraplegic: a four year follow-
up. Ann Plasr Surg 1987;19:572.
5. Mackinnon SE, Dellon AI..: Surgery of the PeTipheTal
Nerve. New Yolk: Thieme; 1988.
6. Mackinnon SE, Dellon AL, Patterson GA, Gruss JS:
Medial antebrachial cutaneous-lateral femoral cutaneous
neurotization to provide sensation to pressure-bearing
areas in paraplegic patients. Ann Plast Surg 1985; 14:541.
7. Masear VR, Meyer RD, Pichora DR: Surgical anatomy
of the medial antebrachial cutaneous nerve. J Hand Surg
1989;14a:267.
8. Millesi H: Microsurgery of peripheral nerves. Hand
1973;5:157.
9. Millesi H: Interfascicular nerve repair and secondary
repair with nerve grafts. In: Jewett DL, McCarroll
HR, eds. Nerve Regeneration and Repair: Its Clinical and
Expen'mentalBasis. St. Louis: CV Mosby; 1980:299.
10. Millesi H: Nerve grafting. Clin Plast Surg 1984;11:105.
11. Millesi H: Technique of peripheral nerve repair. In:
Tubiana R, ed. The Hand. Vol. 3. Philadelphia: WE Saun-
ders; 1988:557.
12. Nunley JA, Ugino MR, Goldner RD, Regan N, Urbaniak
JR: Use of the anterior branch of the cutaneous nerve as
a graft for the repair of defects of the digital nerve. J Bone
Joint Surg [Am] 1989;71A:.563.
13. Race CM, Saldana MJ: Anatomic course of the medial
cutaneous nerves of the arm.J Hand Surg 1991;16a:48.
14. Sunderland S: Nerws and Ner ve Injuries. 2nd ed. Edin-
burgh: Churchill Iivingstone; 1978.
1.5. Woodworth RT: Essenziah of Human An~. 6th ed.
Oxford: Oxford University Press; 1978.
""rbe sural nerve is a commonly used donor nerve for
~ the management of a variety of peripheral nerve
repairs (1-5,9,12). Ita advantages include its favorable
diameter match for cranial nerve repair, unsurpassed
graft length (40cm), and ease ofharvest (Fig. 27-1). In
addition, the sensory deficit produced following its sac-
rifice causes little morbidity. Given its location in the
lower extremity, two surgical teams may work simulta-
neously at the donor and recipient sites, thereby expe-
diting the surgical procedure.
GRAFT DESIGN AND UTILIZATION
1b.ree basic methods for harvesting the sural nerve
have been described (7,11). One longitudinal incision
along the course of the nerve from the lateral malleolus
to the popliteal fossa is the most straightforward and
atraumatic to the nerve (Figs. 27-2 and 27-3). A series
of staintep incisions have also been advocated, but they
increase harvest time and possibly neural trauma. In
addition, nerve and tendon sttippers have been used to
isolate and harvest the nerve from the lower leg. The
third technique involves a minimally invasive approach,
utilizing a "sttippe~' device and two small incisions, or
an endoscopic harvesting kit and a single incision (6).
When using a nerve sttipper, the sural nerve is iden-
tified through a small incision posterior to the lateral
malleolus and is dissected proximally with the snip-
ping instrument (8) (Fig. 27-4). Proximal division can
be accomplished by placing gentle longitudinal traction
on the graft and then utilizing the cutting edge of the
instrument to sever the nerve. This tec:hnique is only
appropriate when a simple nonbranching nerve graft is
497
498 CHAPTER 'l1

Medial sural
cutaneous n. ---,~--..:....! '~..-..-- Lateral sural
cutaneous n.

Peroneal
communicating n.

Small
saphenous v. ----..,i--\J.~I

Achilles
tendon
----i--

RGUREZl-1. The sural nerve supplies sensation to the lateral surface of the lower leg and to the
lateral and dorsal aspects of the foot The sural nerve is most commonly formed by the union of two
branches: the medial sural cutaneous nerve and a branch from the latera I sural cutaneous nerve,
which is referred to as the peroneal communicating branch. The junction of the peroneal commu-
nicating branch with the medial sural cutaneous nerve is variable and may occur at any location in
the calf. The medial sural cutaneous nerve is a branch of the tibial nerve and is the major contribu-
tor to the sural nerve in most cases. The medial sural cutaneous nerve and the proximal portion of
the sural nerve run deep to the deep fascia of the calf between the two heads of the gastrocne
mius muscle. In its distal course in the calf, it lies in a superficial plane over the muscular fascia.
The point at which the sural nerve pierces the fascia is variable, but it is usually located in the
midcalf. The sural nerve supplies sensory branches to the skin of the lower lateral calf, the lateral
aspectofthe calcaneus, and then the dorsal and lateral surfaces ofthefoot.ltterminates as the
lateral dorsal cutaneous nerve of the foot, which also supplies the lateral aspect of the fifth toe.

required because supplementary branches of the sural
nerve cannot be preserved when using this instrument
(Fig. 27-5). The sural nerve bifurcates in a reliable fash-
ion along the lateral aspect of the foot into two branches
that can be captured and utilized as needed in the
reconstruction (Fig. 27-6).
The widest applications for the sural nerve in the
head and neck are for cable grafting of the facial nerve
after ablative surgery, and for cross-face nerve grafting
for facial reanimation (Fig. 27-7). In cross-face nerve
grafting, the sural nerve is usually inset in a reversed
orientation so that the distal end from the lower leg is
sutured to the proximal branch of the nonparalyzed
side of the face. The rationale for orienting the nerve in
this fashion is that regenerating axons will not be lost
through side branches as would occur if the nerve graft
were placed in an antegrade fashion. As such, the maxi-
mum number of regenerating axons reaches the desired
location. By using a Tinel sign, the progress of nerve
growth can be monitored. At the appropriate time, usu-
ally up to 1 year after the initial surgery, a free muscle
flap is transferred to the face and neurotized to the end
of the cross-face sural nerve graft.
NEUROVASCULAR ANATOMY
The sural nerve is formed by the union of the medial
sural cutaneous nerve and a single communicating fasci-
cle of the lateral sural cutaneous branch of the peroneal
nerve (2,3). The dominant contributor, the medial sural
cutaneous nerve, arises from the tibial nerve in the pop-
liteal fossa between the superior heads of the gastrocne-
mius muscle. The nerve runs deep to the muscular fascia
for a variable distance down the posterior calf and then
pierces this fascia to lie in close association, but deep to
the short saphenous vein, at the lateral malleolus. The
nerve and vein run in a lateral compartment between
the lateral malleolus and the tendon of the calcaneus.
At this point, the nerve divides into several branches
that pass around the malleolus distally and supply the
skin of the posterior and lateral aspect of the ankle and
the lateral surface of the foot (8). The nerve is devoid
of major branches until it divides into two dependable
branches on the lateral aspect of the foot (Fig. 27-6). As
the nerve courses proximally over the lateral head of the
gastrocnemius muscle, it can be traced in a superficial
plane over the muscular fascia if additional nerve graft
length is required.
ANATOMIC VARIATIONS
Ortiguela et al. (10) described the course of the sural
nerve from the distal thigh to the ankle in 20 lower limbs.
SURAL NERVE GRAFT 499
The branches forming the sural nerve were identified
and measured for length and caliber. In all limbs, both a
medial sural cutaneous nerve and sural nerve were pre-
sent. A lateral sural cutaneous nerve was identified in 19
of 20 limbs. In 16 of 20 limbs, a peroneal communicat-
ing branch from the lateral sural cutaneous nerve con-
tributed to the sural nerve. In 80% of cases, the sural
nerve was formed by a union of the medial sural cuta-
neous nerve and the peroneal communicating branch.
In the other 20% of cases, the sural nerve originated
from the medial sural cutaneous nerve only. In 94% of
cases, the peroneal communicating branch originated
from the lateral sural cutaneous nerve. In addition, they
also noted that the peroneal communicating branch was
larger in caliber than the medial sural cutaneous nerve
and, when present, was a useful source of nerve graft
material.
POTENTIAL PITFALLS
Painful neuromas, although described in the literature,
have been uncommon when the distal stump of the
nerve is managed as described (Fig. 27-8). The disad-
vantage of this nerve graft site is that when an open pro-
cedure is performed, the incision must be placed over
the lower aspect of the leg. This makes it less desirable
in female patients because the resulting scar tends to
widen and is difficult to conceal.
PREOPERATIVE ASSESSMENT
Preoperative assessment involves careful inspection
of the cutaneous surface of the leg. Any evidence of
ischemic changes in the skin should be noted prior to
surgery. Patients who exhibit ischemic compromise
of the leg, severe peripheral vascular disease, or lower
extremity edema are poor candidates for the harvesting
of this nerve graft.
In addition, any evidence of previous lower extremity
surgery that may have resulted in scarring or compro-
mise of the general condition of the nerve should be
noted and an alternative donor site selected.
POSTOPERATIVE WOUND CARE
The wound is not routinely drained but is wrapped
with a gentle compressive dressing. Low-dose heparin
administration may be indicated in patients who have
an increased risk of deep venous thrombosis, and early
ambulation is encouraged in all patients. In diabetic
patients, early evidence of cellulitis should be treated
aggressively with intravenous antibiotic therapy.
500 CHAPTERD

Sural Nerve Graft

FIGURE Zl-Z. The key topographical landmark

for locating the sural nerve is the lateral mal-
leolus. The initial incision (black line) is planned
posterior to this bony landmark in an attempt to
identify the short saphenous vein as it courses
behind the malleolus into the lateral aspect of
the foot.

FIGURE Zl-3. The dissection of the nerve is

begun by identifying the short saphenous vein
(arrows), which lies lateral to the nerve. The
dissection is continued proximally over the
lateral head of the gastrocnemius muscle to
obtain an adequate length of the nerve graft.
The distal dissection may be extended onto
the lateral aspect of the foot where the nerve
branches. This branching pattern of the nerve
graft may be incorporated into the design of the
repair, depending on the requirements of the
nerve defect.

FIGURE Zl-4. The sural nerve can be identi-

fied and harvested endoscopically through a
horizontal incision made posterosuperior to the
lateral malleolus (left ankle shown, posterior
border of lateral malleolus is marked).
 SURAL NERVE GRAFT 501

Sural Nerve Graft

FIGURE 27-5. Endoscopic tunnel showing

sural nerve at the 6 o'clock position.

FIGURE 27-6. Utilizing the endoscopic sural

nerve harvesting technique, up to 30 em of
nerve can be removed through a single small
incision.

FIGURE 'Zl-1. The sural nerve graft is use-

ful in repairing long nerve gaps that extend
from the temporal bone to the peripheral facial
nerve. The sural nerve graft usually has only
two primary nerve branches, limiting its use
in situations in which a more complicated
branching pattern must be reconstructed.
502 CHAPTER D

Sural Nerve Graft

FIGURE Zl-1. In an effort to prevent neuroma

formation, the proximal sural nerve stump may
be placed within the body of the gastrocnemius
muscle. After a pocket has been made in the
muscle, the distal end of the nerve is placed
within the gastrocnemius muscle and stabilized
with a permanent suture. The soft tissue of the
donor site is closed in layers.

REFERENCES 6. Hadlock TA, Cheney ML: Single-incision endoscopic

1. Doi K, Kuwata N, Kawakami F, Tamaru K, Kawai S:
'I'he free vascularized sural nerve graft. Micronagery
1984;5:175.
2. Doi K, Knwata N, Sakai X. Tamaru K, Kawai S: A reli-
able technique of free vascularized sural nerve grafting
and preliminary results of clinical applications. J Hsad
Swg 1987;12a:677.
3. Doi K, 1\maru K, Sakai X. Knwat:a N, K.urafujiY, Kawai
S: A comparison of vascularized and conventional sural
nerve grafts. J Hand Surg 1992;17a:670.
4. Gilbert A: Vascularized sural nerve graft. Clin Pfast Swg
1984;11:73.
5. GuYD, Wu MM, Zb.eng YL, U HR, Xu YN: Arterialized
vmousfreesuralnervegrafting.AmePfastSwg1985;15:332.
sunl nerve harvest for cross face nerve grafting. J Reeomtr
Miaonwg 2008;24:519-523.
7. Hankin FM, Jaeger SH, Beddings A: Autogenous
sunl nerve grafts: a h.arveating technique. ~
1955;8:1160.
8. Hill HI.,Vacone2 LO, Jurkiewicz MJ: Method fur obtain-
ing a sural nerve graft. Plan Ream~tr SUTg 1986; 1:1 77.
9. May M: Th8FIJI:ialNsrw. NewYork:'Ibieme; 1986.
10. Ortiguela ME, Wood MB, Cahill DR: Anatomy of the
sunl nerve complex. J Hand Swg 1987;12a:1119.
11. Rin.dell X. Telaranta T: A new &traumatic and simple
method of taking sural nerve grafts. A:rm ChiT Gytucol
1984;73:40.
12. Ruben LR: 1M .Pamlyud Faa. St. l..ouia: Mosby Year
Book; 1991.
""rbe success of free tissue transfers to the head and
~ neck depends on many factors. On the most basic
level, the temtory of tissue that is harvested must be
supplied by the donor anery and vein. The design and/
or the dimensions of a flap may exceed that temtory,
leading to areas of regional ischemia within an otherwise
well-vascularized ftap.
The delivery of blood into and out ofthe fl.ap depends
on meticulous harvesting of the nutrient pedicle, care-
ful preparation of the recipient artery and vein in the
head and neck, and technically perfect microvascular
anastomoses. Careful attention must also be paid to the
geometry of the vascular pedicle to prevent tension and/
or kinking caused by the mobility of the head and neck
(11). In this respect, free fl.ap transfers to the head and
neck differ markedly &om the free flaps used in extrem-
ity reconstruction in which immobilization is readily
achieved. In addition, it is abS<llutely essential to pre-
vent infection in the region of the microvascular pedicle,
which is far less tolerant than a regional ftap of expo-
sure to a bacterial insult. Meticulous technique must be
utilized in the insetting process to help avoid a salivary
fistula (12). The orientation of a ftap is also important
to ensure that the most well-vascularized portion of the
tissue is used to complete a watertight seal of the gul-
let. Careful attention to proper drainage of the neck
helps to ensure that accumulated blood, serous ftuid,
or chyle does not lead to secondary infection. Finally, in
505
506 CHAPTER Z8
heavily irradiated wounds, in which the problem of poor
wound healing places the patient at greater risk for the
formation of salivary fistulas, additional measures may
be helpful. We routinely create a control pharyngostome
in patients who undergo carotid artery replacement in
which the pharynx is violated. In such patients, the most
important life-threatening problem is protection of the
bypass graft from salivary contamination. Likewise, the
microvascular pedicle should also be protected in a sim-
ilar fashion when there is a high likelihood of a salivary
fistula. In selected patients, we advocate coverage of the
pedicle by well-vascularized tissue, which may include
the transfer of an additional segment of healthy tissue to
cover the nutrient pedicle in the neck. This may involve
the incorporation of a segment of the latissimus dorsi
muscle with a scapular free flap. On numerous occa-
sions, I have used the pectoralis major muscle for cov-
erage in the neck. We also described a modification of
the radial forearm flap that included a segment offascia
and subcutaneous tissue to protect the microvascular
pedicle in the event of a salivary leak (9).
Head and neck surgeons are often faced with the
necessity to perform Jalvage surgery in patients who have
undergone prior surgery and radiation. These clinical
situations often lead to the necessity to find recipient
vessels in the vessel-depleted neck. Advanced recipient
vessel selection often involves identifying suitable ves-
sels that are outside the area of prior surgery and radia-
tion, which may require dissection and harvest ofvessels
located below the level of the clavicles. The techniques
for harvest of those vessels are demonstrated in this
chapter.
GENERAL CONSIDERATIONS
Recipient vessel selection is one of the most critical
steps in ensuring a successful outcome in microvascular
surgery of the head and neck. The plethora of vessels in
this region provides a wide array of choices. However,
atherosclerosis or prior radiation or surgery may greatly
diminish those options. Careful intraoperative selection
of the recipient vessels facilitates the process of revascu-
larization and reduces the period of ischemia.
It is my practice to select and isolate the recipient
vessels prior to flap harvest. When possible, I try to
have multiple arteries and veins from which to choose.
I also routinely do a second venous anastomosis for
insurance when an additional donor vein is available. In
virtually all situations, I complete the majority of flap
insetting prior to beginning the anastomoses. This is
vitally important for a number of reasons as follows:
(a) the insetting of the flap into defects of the pharynx
and oral cavity must be accomplished with maximum
exposure and flap maneuverability, which may be lim-
ited when the surgeon is concerned about disruption
of the completed anastomoses; (b) the insetting is also
facilitated by working with an ischemic flap, which
frees the surgeon from the troublesome bleeding and
engorgement that occur after revascularization; and (c)
the position of the donor vessels becomes fixed after
insetting, which eliminates the guesswork of setting the
tension on the vascular pedicle.
Despite the meticulous planning, it is often difficult
to be completely certain about where the donor ves-
sels will lie after insetting. The availability of multiple
recipient arteries and veins leaves open many options
to allow for variations in the positions of donor vessels.
The preparation of recipient vessels prior to flap harvest
affords the surgeon the freedom and the confidence to
spend the time during the ischemic period to inset the
flap in a meticulous fashion.
There are a variety of factors, aside from availability,
that must be considered when selecting the recipient
vessels. The location of the defect has a great impact
on the decision-making process. Reconstruction of the
skull base is much different from reconstruction of the
oral cavity, with regard to recipient vessel proximity. The
position of the segmental defect in the mandible also
has an impact on recipient vessel selection. Reconstruc-
tion of the angle of the mandible in a patient with a
high-riding carotid bifurcation may make it impossible
to perform a technically perfect microvascular anasto-
mosis to a branch of the external carotid artery. This is
not the case when the segmental defect is limited to the
region anterior to the midbody.
The particular flap that is selected has intrinsic
restrictions in regard to the caliber and length of the
donor vessels. Although insufficient length of the donor
vessels is the more common problem, the surgeon may
be limited in how much the donor vessels can be short-
ened to achieve a more favorable geometry. This is par-
ticularly true in a flap such as the iliac crest, in which
the venae comitames are often unsuitable for anastomo-
sis until they have joined to produce a vein of sufficient
caliber. The presence of a prior ipsilateral radical neck
dissection severely limits the availability of recipient ves-
sels. Advanced age and prior radiation therapy may lead
to atherosclerosis, which also limits their availability.
Finally, direct tumor extension from the primary tumor
or a regional metastasis may also limit the surgeon's
options.
An axiom of microvascular surgery is that the best
microsurgical technique can be applied when the
surgeon is comfortable and has maximum visualiza-
tion of the microscopic field. In free tissue transfers
of the head and neck, this factor should be taken into
account when selecting recipient vessels. The head posi-
tion must also be accounted for prior to trimming the
donor and recipient vessels. Ablative procedures in
the head and neck are frequently performed with the
neck extended by placing a roll of towels under the
 RECIPIENT VESSEL SELECTION IN FREE TISSUE TRANSFER TO THE HEAD AND NECK
shoulders. Performing the microvascular anastomoses
with the neck in the extended position often exaggerates
the distance between the reconstructive defect and the
recipient vessels. In most situations, particularly when
the vein is running in the long access of the neck, we
advocate removal of the shoulder roll to achieve a more
normal postoperative relationship. The surgeon must
also be cognizant of the impact that head movement
will have on the tension and redundancy of the pedicle
in the postoperative period. This should be established
by direct observation of the neck while the wound is still
open so that postoperative parameters for head position
can be established in the operating room.
RECIPIENT ARTERY SELECTION
The two major sources of recipient arteries in the
neck are the branches of the external carotid artery
and the branches of the thyrocervical trunk. Because
of their availability and proximity to most head and
neck defects, the lower branches of the external
carotid artery are often most suitable. The superior
thyroid artery, owing to its more caudal position, is
usually the most accessible branch. The superior thy-
roid artery must be dissected to its takeoff from the
carotid to permit adequate mobilization. The caliber
of this vessel is usually suitable for several centimeters
until it bifurcates. When cephalad branches are inac-
cessible and the superior thyroid artery is a poor
size match, end-to-side anastomosis to the external
carotid artery can be used. It is imperative that the
arteriotomy is performed at least 2 to 3 em cephalad
to the carotid bifurcation to help reduce the risk of
temporary occlusion and the potential morbidity
should hemorrhage from the anastomosis occur.
The availability of the external carotid artery or its
branches may be limited by age, radiation-induced
atherosclerosis, and by direct tumor invasion by the
primary or a nodal metastasis.
Knowledge of the anatomy of the lingual artery
can be put to good use by microvascular surgeons in
search of a suitable recipient artery. Lesser's mangle is
an obscure anatomic detail that can be used to identify
the distal portion of the lingual artery (Fig. 28-1) (6).
This triangle is defined by the posterior and anterior
bellies of the digrastic, and the crossing of the hypo-
glossal nerve (Fig. 28-2). When the mylohyoid muscle is
dissected within this small triangle, the lingual artery is
encountered and can be mobilized for a significant dis-
tance while making certain not to injure the hypoglossal
nerve through traction or sharp dissection (Figs. 28-3
and 28-4). This artery is prone to spasm but is usually a
suitable caliber in this location. This prevents the neces-
sity for dissection more posteriorly at the takeoff from
the external carotid artery.
507
The concern for using radiated recipient vessels in
the neck is more of historical interest as an issue that
delayed the use of free flaps in previously irradiated
patients. In my experience, this factor does not have a
significant impact on the success of free flap transfers.
Mulholland et al. (7) compared the success rate of free
flap transfers in irradiated patients to that in nonirra-
diated patients. The failure rate of 3.5% in the irradi-
ated group was not significantly different from the 2. 9%
reported in the nonirradiated group. They found that
only two factors had an impact on free flap failure: post-
operative infection in the recipient bed and the length of
time between radiation and the free flap surgery. Both
the reason for and the significance of the latter finding
are unclear.
The branches of the thyrocervical trunk, and in par-
ticular, the transverse cervical artery (TCA), can usu-
ally be preserved following neck dissection and, in many
ways, serve as a better recipient artery. The TCA can be
traced for a significant distance along its course under-
neath the trapezius muscle. The caliber of the lumen
usually remains adequate for microvascular anastomo-
sis, despite giving off branches in the posterior triangle.
Distal ligation and transsection of this vessel allows it
to be transposed to a position in the midportion of the
neck where it can be readily used as a recipient artery.
In our experience, this vessel is far less prone to athero-
sclerosis than are branches of the external carotid, but
it is more susceptible to vasospasm. In addition, when
performing microsurgery in the previously irradiated
patient, the TCA is usually outside the area of the most
intense radiation. The positioning of the end of this
artery in the midportion of the neck permits a com-
fortable microsurgical procedure with excellent expo-
sure. An additional benefit of using the TCA is that the
full length of the donor artery can be utilized without
trimming back to a smaller caliber vessel. The proxim-
ity of the external carotid artery branches to the recipi-
ent defect often forces the surgeon to trim the donor
artery to prevent redundancy of the pedicle. Use of the
TCA allows the full length and the greatest caliber of
the donor artery to be used. Trimming of the TCA, to
prevent redundancy, takes it closer to its source and,
therefore, usually leads to a larger lumen.
The availability oftheTCA is rarely limited because
of involvement by cancer. Rather, the two most
common reasons why it may not be available are liga-
tion during a neck dissection and an anatomic varia-
tion in its course. Cadaver dissections have shown that
as many as 20% of TCAs arise directly from the sub-
clavian artery and may run a circuitous course through
the brachial plexus, which severely limits its suitability
to be transposed. H the TCA is not identified in its nor-
mal position in the inferomedial aspect of the neck, the
surgeon should explore the area lateral to the brachial
plexus.
508 CHAPTER Z8
Neceuity u the mother of invention ... and the need for
recipient vessels in the vessel-depleted neck has led to
the discovery of new options that were not previously
utilized in head and neck reconstruction. We reported
the use of the thoracoacromial artery cephalic vein
(TAC) system of vessels for use as recipient vessels in
head and neck microsurgery (5) (Fig. 28-5). By utilizing
an infraclavicular site for recipient vessels, the surgeon
can be confident that these vessels have not suffered the
ill effects of radiation therapy and should be unaffected
by prior neck surgery. The use of the thoracoacromial
system eliminates the potential for harvest of a pecto-
ralis major flap, which is a major disadvantage of this
vessel selection (Figs. 28-6 to 28-11). The cephalic vein
is a very favorable choice in most individuals and offers
the distinct advantage of being able to reach to almost
any location in the head and neck, simply by dissecting
further down the arm in order to lengthen the vessel
that is transposed over the clavicle (Figs. 28-12 and
28-13). Prior placement of indwelling venous catheters
for antibiotic or chemotherapy usage will sometimes
lead to venous thrombosis and thereby make this vein
not usable.
Another infraclavicular recipient artery and vein is
the internal mammary system that has long been uti-
lized by plastic surgeons in free tissue transfers for
breast reconstruction and by cardiothoracic surgeons in
coronary artery surgery (l 0). I began to utilize this sys-
tem for vessel-depleted necks and found it to be a highly
reliable set of vessels both for salvage surgery and in
patients with ischemic flaps requiring revascularization.
The harvest of these vessels is somewhat tedious and
there is a risk of violation of the pleural cavity leading
to a pneumothorax (Figs. 28-14 to 28-22). However,
the caliber of the vessels is outstanding, albeit there is a
distance from most defects, other than those located in
the neck such as in pharyngoesophageal reconstruction.
The length of these vessels can be extended by more
distal dissection, which yields a more favorable cephalad
position following transposition over the clavicle.
RECIPIENT VEIN SELECTION
There are three primary recipient veins in the neck
that should be used in most free tissue transfers. The
external jugular and the transverse cervical veins can
usually be saved in a radical neck dissection. The
internal jugular vein serves as an excellent outflow for
a free flap when preserved following modified neck
dissection.
It is often tempting to use the anterior jugular veins.
However, because the majority of patients undergoing
free flap reconstruction will require a tracheotomy, the
caudal portions of these veins are at risk. Iatrogenic
injury to the anterior jugular veins while performing a
tracheotomy, or the peristomal inflammation that invar-
iably occurs, makes the anterior jugular veins a poor
choice.
As noted in the discussion above, the cephalic veins
may be used as a source for vein grafts or as recipient
veins. My decision to use the cephalic vein is made only
when the other options are not available and when there
is a good recipient artery in the ipsilateral neck. H the
recipient artery is located in the contralateral neck, then
it is our preference to run the venous pedicle to that
side as well. We have most often used the cephalic vein
in situations of flap salvage when a new recipient vein
is necessary. The technique for harvesting this vein is
demonstrated in Figures 28-6 to 28-11, and the extent
to which this vein can be dissected and mobilized is
shown in Figures 28-12 and 28-13. The deltopectoral
groove is exposed through elevation of the distal end of
a deltopectoral flap, which serves as a delay procedure
for that flap should it be needed. Particular attention
must be paid to the fulcrum point of the transposed
cephalic vein, which lies just below the clavicle where it
enters the subclavian vein. The prevention of kinking or
compression in this region is vital to the success of this
procedure. On occasion, we have created a small depres-
sion in the superficial surface of the clavicle to help pre-
vent movement and compression of the cephalic vein. In
addition, the placement of a small fixation plate that is
bent in a convex configuration can be fashioned to pro-
vide protection against compression over the clavicle.
In secondary reconstructions, the paucity of recipient
veins makes it tempting to consider the use of retrograde
venous drainage. For example, the distal end of the
superior thyroid vein may be used to provide drainage
through the thyroid gland, or the distal end of the super-
ficial temporal vein may be used to achieve drainage over
the top of the scalp (7). The use of a long vein graft to the
opposite neck or a transposed cephalic vein is considered
a far better option than reliance on retrograde flow.
SPECIAL CONSIDERATIONS
Under certain situations, it is necessary to use interposi-
tion vein grafts to achieve a tension-free vascular pedi-
cle that will accommodate the full range of motion of
the head and neck (2). In most reconstructions, careful
selection of recipient vessels, the type of free flap, and
the orientation of the flap during insetting help avoid
the need for vein grafts.
Skull base reconstruction presents the most common
situation requiring interposition vein grafts because
of the distance between the defect and recipient neck
vessels. In some patients, the superficial temporal artery
and vein may be utilized. However, in my experience,
the superficial temporal vein, in particular, is often
unsuitable as a reliable recipient vessel.
 RECIPIENT VESSEL SELECTION IN FREE TISSUE TRANSFER TO THE HEAD AND NECK
Vein grafts are occasionally required in head and
neck microsurgery. The potential need for them should
be anticipated in "problem" reconstructions in which
recipient vessels are likely to be lacking. The surgeon
should anticipate the necessity for a vein graft when
embarking on an emergent procedure to salvage a fail-
ing flap. Vein grafts from virtually any location can be
harvested. The cephalic vein is often readily available in
the surgical field. However, I would advise against the
use of that vein to harvest a vein graft for fear that it
would potentially eliminate a recipient vein that may be
needed should an emergent situation arise. As a result,
I have most often resorted to the greater or lesser saphe-
nous veins. Lower extremity vein grafts can be easily
harvested under tourniquet control. After harvest, the
proximal end should be temporarily occluded with a
microvascular clamp or a tie while a heparinized solu-
tion is perfused into the distal end. This maneuver helps
to identify any side branches that require ligation. Gen-
tle perfusion pressure also helps to dilate the vein graft
so that it can be inset without underestimating its length.
It is imperative that the surgeon selects a vein graft that
has suitable dimensions relative to the donor and recipi-
ent vessels so as to avoid introducing a size mismatch
that may promote turbulence and thrombosis.
There have been reports of creating arteriovenous
shunts prior to flap transfer in the lower extremity to
help ensure the patency of at least two anastomoses.
Delayed free tissue transfer was performed several days
later. When vein grafts are placed at the time of the free
flap transfer, the microsurgeon has the choice of ini-
tially performing the anastomoses of the vein graft to
the recipient or to the donor vessels. The arguments in
favor of creating a temporary arteriovenous shunt to the
recipient vessels are that it reduces the warm ischemia
time and permits the proximal anastomoses to be tested
(3,4,8). However,Acland (1) warned that this technique
predisposes to the formation of a "high-flow thrombus"
at the arterial end of the graft. Alternatively, a blind vein
graft loop can be created by anastomosing the vein to
the flap vessels at the outset. This set of anastomoses
can be comfortably performed on a side table under
"laboratory-like" conditions. It is imperative that the
509
vein loop is oriented properly to account for the direc-
tional valves. Following flap inset, the vein loop can be
cut at an appropriate location for anastomoses to the
recipient artery and vein.
The use of vein grafts is a necessary part of head and
neck microsurgery. The anticipated requirement to per-
form them should not make the surgeon decide not to
use a free tissue transfer. Careful planning and meticu-
lous technique should allow the use of vein grafts with-
out an unfavorable impact on the rate of success.
The tunneling of the vascular pedicle from the central
skull base to the neck can be done either through a sub-
cutaneous tunnel under the central cheek or through a
subcutaneous tunnel created anterior to the auricle. The
first approach, although more direct, is done blind and
introduces uncertainty in regard to the exact orientation
of the vascular pedicle in the tunnel. This may be particu-
larly problematic when vein grafts are used and the set of
anastomoses between the vein graft and the flap vessels is
buried. There is no way to assess this set of anastomoses
after flow has been established, following completion of
the anastomoses of the vein grafts to the recipient vessels
in the neck. This may be problematic if there is poor flow
in the flap or if oozing from the tunnel is observed.
A preferable approach to the tunneling of the ves-
sels is to elevate a subcutaneous cheek flap through a
preauricular incision. Although a less direct course,
the exposure afforded by this approach eliminates the
problems outlined earlier. In either situation, the route
of the vascular pedicle must be drained with a passive
drainage system.
Careful attention to recipient vessel selection, pedi-
cle geometry, and the minute details of microvascular
surgery helps to eliminate some of the guesswork of this
technique and to avoid some of the potential compli-
cations that may occur. Although every surgeon's skills
must progress along an experience curve, adherence
to the principles and suggestions outlined earlier will
hasten the arrival of success rates approaching 100%.
In addition, there are now far fewer instances in which
the option of performing life-saving or quality of life-
enhancing surgery is not possible due to the inability to
prepare appropriate recipient vessels.
510 CHAPTER Zl

Harvest of the Lingual Artery in Lessers Triangle

FIGURE ZB-1. The skin and subcutaneous

tissue have been removed to demonstrate the
muscular and neural anatomy of the subman-
dibular triangle. Lesser's triangle is bounded by
the junction of the anterior and posterior bellies
ofthe digastric muscles and the hypoglossal
nerve. The lower posterior border of the mylo-
hyoid muscle crosses this triangle and actually
forms the anterior border.

FIGURE 28-2. A close up view of the anatomy

of this triangle demonstrates the key soft tissue
relationships. PBD, posterior belly of digastric
muscle; ABO, anterior belly of digastric muscle;
HN, hypoglossal nerve; MM, mylohyoid muscle.
 RECIPIENT VESSEL SELECTION IN FREETISSUETRANSFERTO THE HEAD AND NECK 511

Harvest of the Lingual Artery in Lessers Triangle

FIGURE 28-3. Division of the hyoglossus

muscle in the depth of Lesser's triangle, caudal
to the hypoglossal nerve leads to exposure of
the lingual artery that can be further dissected
(srrowt.

FIGURE 28-4. The lingual artery has been

dissected for a considerable distance and
mobilized for use as a recipient artery. Despite
its more peripheral location and tendency to
vasospasm, it is usually very suitable for anas-
tomosis atthis location.
512 CHAPTER Zl

Harvest of the TAC System of Recipient Vessels

Thoracoacromlal
branch of sub-
clavian artery

FIGURE 21-5. The anatomy of the TAC system Cephalic vein -"'"""+'~!ft
of recipient vessels is shown. The cephalic vein
runs in the deltopectoral groove and the arterial
branches of the thoracoacramial system are
accessed bytransecting the attachments of the
lateral aspect of the pectoralis major muscle to
the clavicle.

FIGURE 28-&. The topographic anatomy of the

thoracoacromial cephalic system is identified
with the clavicle shown in brown, the expected
location ofthe thoracoacromial system in red,
and the cephalic vein in the deltopectoral
groove in blue. The incision for access is identi-
fied in green and is designed to preserve and
stage the tip of the deltopectoral flap.
 RECIPIENT VESSEL SELECTION IN FREETISSUETRANSFERTO THE HEAD AND NECK 513

Harvest of the TAC System of Recipient Vessels

FIGURE 28-7. The distal end of the deltopec-

toral flap is elevated in a subfascial plane over
the deltoid and the pectoralis major muscles.

FIGURE 28-8. The flap has been elevated.

The deltopectoral groove containing the
cephalic vein is identified with the blue pin.
An incision is planned (black line) 1em below
the clavicle through the lateral aspect of the
clavicular insertion ofthe pectoralis major
muscle.

FIGURE 28-9. The cephalic vein has been

dissected out of the deltopectoral groove. The
lateral aspect of the pectoralis major muscle
is divided to provide access to the thoracoac-
romial vascular system.
514 CHAPTER Zl

Harvest of the TAC System of Recipient Vessels

FIGURE 28-10. The thoracoacromial artery

(red arrow with white borden has been dis-
sected to the level of the subclavian artery (red
arrow with yellow borden. The cephalic vein
(blue srrow with white borden is shown enter-
ing into the subclavian vein lblue arrow with
white borden. Usually, one branch of the thora-
coacromial artery, often the pectoral branch, is
identified and traced back to the common trunk
to facilitate rapid identification.

FIGURE 28-11. A: The thoracoacromial artery

and cephalic vein (blue arrow) have been
transposed superiorly in position for anastomo-
sis to donor vessels. B: The cephalic vein and
thoracoacromial artery vessels are shown with
blue and red pins, respectively. The cephalic
vein can be dissected distally into the arm
and can harvested in lengths that can reach
virtually anywhere in the head and neck area.
The thoracoacromial system has a more limited
reach, and flaps with longer pedicles or vein
grafts may need to be utilized. B
 RECIPIENT VESSEL SELECTION IN FREETISSUETRANSFERTO THE HEAD AND NECK 515

Harvest of the TAC System of Recipient Vessels

FIGURE 28-12. The cephalic vein may be used

either as a vein graft or as a recipient vein,
usually in situations in which a previous radical
neck dissection has been performed. Exposure
of the cephalic vein can be accomplished in
a number of different ways. The cephalic vein
may be traced distally in the arm as far as
necessary. The length should be measured
using a free suture with the fulcrum located
just inferior to the clavicle to be certain that an
adequate amount of length has been obtained.
It is imperative 1hat the pa1h of the vein into
the concavity of the supraclavicular fossa be
accounted for when determining a suitable
length. Distal exposure of1he cephalic vein may
be accomplished through a linear incision or
through serial transverse stair-step incisions.
It is always wise to harvest a greater length
than is needed to ensure that a tension-free
anastomosis can be achieved and to ensure
that the vein is not tented over 1he clavicle but
rather assumes a gentle course in the concavity
of the supraclavicular fossa.

FIGURE 28-13. Ligation and transsection of1he

cephalic vein is then performed. The cephalic
vein is transposed over the clavicle into position
for anastomosis to the donor vein. It is essential
that particular attention is paid to the turn of the
cephalic vein over the clavicle. This is the most
critical point at which kinking or compression
may occur, leading to venous obstruction.
516 CHAPTER Zl

Harvest of the Internal Mammery Artery and Vein

FIGURE Zl-14. The topographic anatomy of

the parasternal regian is marked an the chest
wall with the approximate course of the internal
mammary artery and vein.

FIGURE 28-15. A longitudinal incision is made

parallel to the sternum with exposure of the
second and third costal cartilages.

FIGURE 28-16. A subperichondrial dissection

is performed in order to isolate the cartilage af
the 2nd or 3rd rib that is remaved.
 RECIPIENT VESSEL SELECTION IN FREETISSUETRANSFERTO THE HEAD AND NECK 517

Harvest of the Internal Mammery Artery and Vein

FIGURE 28-11. Once the cartilage has been

removed, the posterior perichondrium is
then incised in order to expose the internal
mammary artery.

FIGURE 28-18. After isolation of the artery and

meticulous ligation of its branches, the vein is
dissected and its branching system is identi-
fied and ligated by gently "'rolling'"' that vessel
and isolating each of the numerous branching
vessels. The vein is always located medial to
the artery and additional cartilage may need
to be removed in order to gain access for
dissection of that vessel.

FIGURE 28-19. Proximal and distal dissec-

tion is performed below the second costal
margin and the fourth costal margin. Attention
must be paid to avoid injury to the underlying
pleura.
518 CHAPTER Zl

Harvest of the Internal Mammery Artery and Vein

FIGURE 28-20. The internal mammary vein

has been elevated {blue arrows) and dissected
distally in order to provide added length for
transposition over the clavicle. The vein should
be dissected caudally as long as the caliber
remains favorable as it will enhance the arc of
rotation for anastomosis to donor vessels.

FIGURE 21-21. The internal mammary artery

(red arrowj has been dissected distally and is
ready for transposition. Additional cartilage
(white arrawt has been removed from the 4th
rib in order to gain access for this more caudal
dissection and delivery of the artery.
 RECIPIENT VESSEL SELECTION IN FREETISSUETRANSFERTO THE HEAD AND NECK 519

Harvest of the Internal Mammery Artery and Vein

FIGURE 28-22. A:. Both the artery and

vein have been transposed in a cephalad
direction and are in position for microsurgi-
cal anastomosis. B: The caliber of the internal
mammary artery is very favorable for most
donor vessels of free flaps used in head and
neck reconstruction. B
520 CHAPTER 28
REFERENCES
1. Acland RD: Refinements in lower extremity free flap
surgery. Clin Piast Surg 1990;17:733--744.
2. Biemer E: Vein grafts in microvascular surgery. Br J Plast
Surg 1977;30:197-199.
3. Gronga T, Yetman R: Temporary arteriovenous shunt
prior to free myo-osseous flap transfer. Microsurgery
1987;8:2-4.
4. Hallock G: The interposition arteriovenous loop revis-
ited.J Reconstr Microrurg 1988;4:155-159.
5. Harris JR. Lueg E, Urken ML: The thoracoacrornial/
cephalic vascular system for microvascular anastomoses
in the vessel-depleted neck. Arch Otolaryngol Head Neck
Surg 2002;128(3):319-323.
6. Hollinshead WH: Anatomy for Surgeons: The Head and
Neck. VoL L Philadelphia: Hluper and Row Publishers;
1982:375.
7. Mulliolland S, Boyd J, McCabe S, et al.: Recipient vessels
in head and neck microsurgery: radiation effect and vessel
access. Plasr Reconsrr Surg 1993;92:628.
8. Threlfall G, Little J, Cummerie J: Free flap transfer-
preliminary establishment of an arteriovenous fistula: a
case report. Aust N Z J Surg 1982;52: 182- 184.
9. Urken ML, Futran N, Moscoso J, Biller HF: A modified
design of the buried radial forearm free flap for use in
oral cavity and pharyngeal reconstruction. Arch 0/olaryn-
gol Head Neck Surg 1994;120:1233-1239.
10. Urken ML, Higgins KM, Lee B, Vickery C. Internal
mammary artery and vein: recipient vessels for free tissue
transfer to the head and neck in the vessel-depleted neck.
Head Neck 2006;28(9):797-801.
11. Urken ML, Vickery C, Weinberg H, Buchbinder D, Biller
HF: Geometry of the vascular pedicle in free tissue trans-
fers to the head and neck. Arch Otolaryngol Head Neck
Surg 1989;115:954-960.
12. Urken ML, Weinberg H, Buchbinder D , et al.: Microvas-
cular free flaps in head and neck reconstruction: report of
200 cases and review of complications. Arch Otolaryngol
Head Neck Surg 1994;120:633--640.

Note: Page numbers in "iuUics" indicates figures.
A deaign and utilization
Abdominal wall de:fccts
iliac crest osteomusculocutaneous
tlap. 38~388
rectus abdominis free tlap, 150
Acromial artery, 16
Adductor artery
gracilis,164, 165
thigh,166
Allen's test, 184-186, 280,281
Angiosome c:onc:ept
deltopec:toral flap, 82-84
latisaimua dorsi flap, 336, 338
lower uapezius island
musculocutaneous flap. 33--35
pectoralis major musculocutaneous
flap, 16-17
rectus abdominis flap, 148
aubmcntal ialand flap, 107
superior trapezius tlap, 31-33,32
Anterior deep temporal artery, 49, 50
Anterior inferior iliac apine (AilS), 364
Anterior interosseous artery
radial foreann Sap, 181
ulnar forearm free tlap, 273
Anterior jugular vein, 508
Anterior scalp flap
design and utilization, 90,91
harveating tec:bnique
blood supply, 97
tlap elevation, 96
full-thickness skin graft, 95
lateral incision., 95
acalp tranafcr, 97
akin tranafcr, 94
vascular supply, 94
historical perspectives, 89-90
neurovascular anatomy, 91,93
pitfalls, 93
poatopcrative wound care, 93
Anterior aupe:rior iliac apine (ASIS),
235,399
Anterior tibial artery
fibular osteocutaneous tlap, 409, 409
poaterior tibial artery free flap, 264
Anterior ulnar recum:nt artery, 273
Anterolateral thigh (ALT) free tlap
anatomic variations, 240
anatomy, 234, 235
blood supply, 238-239
cutaneous perforator location.,
Doppler stethoscope, 235--236
oral cavitY rec:onsttuc:tion, 238
pharyngoesophageal
rc:consttuction, 238
primary ALT Sap 1hinning, 237
sa:asory recovery and swallowing
outcomes, 237-238
skin and subcutaneous thickness.
236-237
akin paddle:, 234-235, 236
skull base, acalp, and midfac:e
defects, 238
suprafasc:ial fJS. subfascial flap
dissection, 237
tongue rec:onsttuction, 237
harvest tc:dmiques, 243-248
hiatorical perapc:ctivdl, 234
inset techniques. 249--254
neurovascular anatomy, 238-240,
239-240
pitfall&, 24~242
poatopcrative care, 242
preoperative assessment, 242
Arch of Douglas, 149
Arcuate line, 149
Arteriapro~dabrachli,208,209
Arteriovcnoua ahunt, 509
Artery ac:lc:ction, 507-508
Auriculotemporal nerve, 224, 225
Autogenous nerve graft, 491
B (08~),146,148,152
Bakamjian flap, 76
Bipedicle trapezius Sap, 27
Body mass index,. anterolateral thigh
free Sap, 236-237
Brachial artery, 178, 180, 182,208,210
Brachial plCXW~, 341
Brachioradialia, 176, 178, 181-183,
188,190
Brachioradialis muscle rettaction,
448,448
Brow paralysis, 226
c omentum free tlap, 484--487
Calvarial bone graft, 127-128
Cephalicvein,176,177,180-183,
187-189,191,193-195,508
Index
Chcvron-oricntc:d muscle fibers, 417
Circumferential pharyngoesophageal
reconsttuction,180
Circumtla peroneal artery, 409
Circumtla scapular anery, 292-303
Circumflex scapular vein, 292, 293,
295, 296, 299
Clavipectoral fascia, 16:, 17
Common peroneal nerve, 408,
412,414
Compoaite radius osteocutanc:ous flap,
448,448
Coronoidectomy, 439
Corticocancellous bone graft, 400
Cranial nerve injury, temporalis muscle
flap,48
Croas-face nerve: graft, 163-164,
499,501
D
Deep circumflex iliac artery, 379-381
Dc:c:p c:ircu:mfla iliac vein., 360,
369-371,375,376,38~382,
392,397
Deep inferior epipstric artery (DIEA),
141, 144-148, 150, 152, 155,
157,159
Dc:c:p inferior c:pigBstric vein (DIEV),
141, 145-147, 150, 152, 155,
157, 159
Deep palmar arc:b., UFFF, 277,
278,279
Dc:c:p superior c:pigaatric artery
anatomic variatioJUI, 150
arborization,146
Deep superior epigastric vein (DSEV),
146,147
anatomic variatioJUI, 150
arborization, 146
Defects
circumferential pharyngoesophageal
reconstruction
anterolateral thigh flap, 249--254
frc:e jejunal autograft, 464, 468-472
frc:e omentum and pstro-
latissimus dorsi musculocutaneous
flap, 333
radial forearm flap, 198-200
521
522 INDEX
Defects (Continued)
hand defect
posterior tibial artery free flap, 263
ulnar foreann flap, 273, 275
hemimandibular defect
fibular osteocutaneous flap,
421-427, 434-435
iliac crest-internal oblique
osteomyocutaneous free flap, 393
serratus anterior-rib
osteomusclular flaps, 335
mandibular/oromandibular defects
fibular osteocutaneous flap,
404-406
iliac composite flap, 362
iliac crest-internal oblique
composite flap, 393-394
iliac crest-internal oblique
musculocutaneous flap, 395-397
latissimus dorsi-serratus flap,
351-355
osteocutaneous radial forearm free
flap,477
scapular osteocutaneous
flap-latissimus dorsi flap,
320-321
serratus anterior-rib
osteomyocutaneous flap,
351-355
midfacial defects
greater omentum flap, 4 77
latissimus dorsi musculocutaneous
flap,333
paramedian forehead flaps, 123
temporalis flap, 48, 53
temporoparietal falp, 222
mucosal defects
deltopectoral flap, 76
free jejunal autografts, 457--458
greater omentum flap, 480
pectoralis major flap, 11
submental island flap, 102
palatomaxillary defects
fibular osteocutaneous, 407
iliac composite flap, 362
osteocutaneous radial forearm free
flap,447
pharyngoesophageal defects
free jejunal autografts, 456, 462,
464,468-472
greater omentum flap, 4 77, 480
radial forearm flap, 196-197
ulnar foreann flap, 279
pharynx and oral cavity defects
anterolateral thigh flap, 237-238
deltopectoral flap, 85
free jejunal autografts, 463
gastro-omental free flap, 476
lateral arm flap, 208
latissimus dorsi musculocutaneous
flap,333
radial forearm flap, 178
recipient vessel selection, 506
sternocleidomastoid
musculocutaneous flap, 64
submental island flap, 102
ulnar forearm flap, 280
post-traumatic tissue defect,
posterior tibial artery
free flap, 257
resected mandibular condyle, fibular
osteocutaneous flap, 428-433
scalp defect
greater omentum flap, 475,477
latissimus dorsi musculocutaneous
flap, 333
temporoparietal falp, 222
segmental mandible defect
recipient vessel selection, 506
serratus anterior-rib
osteomusclular flaps, 335-336
soft tissue defect
anterolateral thigh flap, 237
free omental flap, 476
serratus anterior flaps, 334, 335
total glossectomy defect, pectoralis
major flap, 12
total maxillectomy defect, iliac crest-
internal oblique flap, 398-402
total palatal defect, scapular tip
osteocutaneous flap, 323-324
tracheal defect, sternocleidomastoid
musculocutaneous flap, 67, 68
Deltoid artery, 15, 16
Deltoid flap, 206-208
Deltopectoral flap
design and utilization
arc of rotation, 77-78, 80
distal portion fenestration, 77-78
hemifacial defect, 80
innervated dcltopectoral flap, 79, 81
intraoral reconstruction, 79, 80, 82
island deltopectoral flap, 77, 78
microvascular free flap, 79
pharyngoesophageal segment
reconstruction, 80, 82
recipient site transfer, 77
vertically split, 77, 79
harvesting technique
cutaneous branch isolation, 86
donar site closure, 87
incisions, 86
historical perspectives, 76
length, 77
neurovascular anatomy
angiosome concept, 83-84
blood supply, 82, 84
internal mammary artery, 82
internal mammary vein, 82
sensory nerve supply, 84
pitfalls, 84--85
preservation, 8
Donor site complications
anterior scalp flap, 90
palatalisland flap, 131
pectoralis major musculocutaneous
flap, 18-19
posterior scalp flap, 90
posterior tibial artery flap, 264
sternocleidomastoid flap, 71
Donor vascular pedicle, 41 9--420
Doppler stethoscope, 235-236
Dorsal scapular vein (DSV), 35, 43, 44
Double-barreled fibula flap, 405, 407,
434-435
Down's syndrome, 278
H
Endoscopic sural nerve harvesting
technique, 501
Epigastric island flap, 141
Extended deep inferior epigastric flap
(see Rectus abdominis flap)
External carotid artery, 507
External iliac artery, 371
External jugular vein, 508
External oblique muscle, 373,383, 391
F
Facial artery
platysma muscle, 70
submental island flap, 107-108
temporoparietal fascia, 225
Facial paralysis
gracilis muscle flap, 174
lateral island trapezius flap, 33
latissimus dorsi flap, 333
serratus anterior flap, 335
temporalis muscle flap, 47
Facial reanimation
gracilis muscle flap, 164-165, 171
lateral island trapezius flap, 33
latissimus dorsi flap, 331
pectoralis major musculocutaneous
flap, 13
sural nerve, 499, 501
temporalis muscle flap, 48
Fan-shaped mesentery, 458
Fibular osteocutaneous flap
anatomic variations, 411
anatomy, 404-405, 405
design and utilization
cutaneous perforators, 407
double-barreled fibular flap, 405,
407
functional mandibular
reconstruction, 407
lateral sural cutaneous
nerve,408
mandible segmental defect
reconstruction, 405, 407
palatomaxillary defect, 407
peroneal artery and vein, 405
posterior crural septum, 405, 406
vascular pedicle, 405
harvest techniques
anterior and medial dissection, 415

bone cuts, proximal and distal
fibula, 416
cutaneous flap design and
incision, 413
flexor hallucis longus
transection, 418
lateral sural cutaneous nerve, 414
peroneal communicating
branch, 414
peroneal vascular system
dissection, 417
posterior incision, 416
primary closure, 418
skin paddle, 415
historical perspectives, 404
inset techniques
cutaneous paddle and donor
vascular pedicle, 419--420
resected mandibular condyle,
428--433
right hernimandibular defect,
421--427
right hemimandibulectomy defect,
434--435
right infrastructure maxillectomy
defect, 436-443
neurovascular anatomy
cutaneous perforator, 409
dye-injection study, 411
peroneal artery and vein, 408, 408
sensory supply, 411
septocutaneous perforators,
409--410,410
skin paddle, 410--411
pitfalls, 411--412
preoperative assessment, 412
Flap harvest technique
anterior scalp flap
blood supply, 97
flap elevation, 96
full-thickness skin graft, 95
lateral incision, 95
scalp transfer, 97
skin transfer, 94
vascular supply, 94
anterolateral thigh free flap,
243-248
deltopectoral flap
cutaneous branch isolation, 86
donar site closure, 87
incisions, 86
fibular osteocutaneous flap
anterior and medial dissection, 415
bone cuts, proximal and distal
fibula, 416
cutaneous flap design and
incision, 413
flexor hallucis longus
transection, 418
lateral sural cutaneous nerve, 414
peroneal communicating
branch, 414
peroneal vascular system
dissection, 417
posterior incision, 416
primary closure, 418
skin paddle, 415
free jejunal autografts, 466-467
free omentum and gastro-omentum
free flap, 481-483
gracilis muscle flap
anterior obturator nerve,169-170
initial incision, 168
left gracilis muscle flap, 168
neurovascular pedicle, 168
vascular pedicle dissection, 169
wound closure, 170
iliac crest osteocutaneous flap
cephalad margin incision, 390
defect closure, 392
external oblique muscle and
aponeurosis, 391
internal oblique muscle incision, 391
iliac crest osteomusculocutaneous flap
abdominal wall closure, 384, 389
ascending branch, blood supply, 384
DCIA and DCIV dissection, 379
external oblique muscle incision, 373
iliac bone exposure, 380
iliacus transsection, 377
internal oblique muscle
elevation, 375
internal oblique muscle
exposure, 374
lateral dissection, 377-378
medial aspect, 383
medial bone cuts, 381
meticulous dissection, 374
osteotomy, 381
skin incision, 372
synthetic mesh closure, abdominal
wall, 385-388
transversus abdominis muscle
transsection, 376
vascular pedicle, 372, 382
lateral ann flap
anteriorapproach,214-215
deltoid muscle insertion, 212
flap elevation, 215
lateral IM septum location, 212
meticulous dissection, 213
neurovascular pedicle, 216, 217
posterior incision, 213-214
PRCA ligation, 215-216
septocutaneous perforators, 214
soft tissue closure, 217-218
spiral groove visualization, 217
lateral island trapezius flap
distal TCA and TCV ligation, 40
donor site selection, 39
flap transposition, 41
neck dissection, 39
scapula position, 41
skin paddle incision, 40
INDEX 523
latissimus dorsi flap
adipose tissue removal, 344
arc of rotation maximization, 350
back skin elevation, 345
blunt and sharp dissection, 346
cutaneous paddle, 343
donor site closure, 350
initial incision, 343
pedicled flap passage, 348-349
posterior retraction, 344
tendinous insertion, 347
thoracoacrornial pedicle, 348
thoracodorsal pedicle mobilization,
344,345
osteocutaneous radial forearm free
flap
bicortical screws, 453
flexor pollicis longus, 451, 452
lateral antebrachial cutaneous
nerve separation, 452
locking reconstruction plate, 453
prophylactic fixation, 452
radial artery and cephalic vein
marking, 450
residual muscle suture, 453
palatal island flap
contralateral pedicle isolation and
transsection, 134-135
flap rotation, 136
mucoperiosteal flap elevation,133
tonsillar fossa and retromolar
trigone defect closure, 136
paramedian forehead flap
calvarial bone graft, 127-128
radial forearm tissue, 125
structural elements, 126
superior and inferior flap
elevation, 126
supratrochlear vessels, 124
trilayered approach, 126
parascapular fasciocutaneous flap
anatomic landmarks, 313
angular branch elevation, 315
axilla, 317
flap incision and elevation, 313
osteotomy, 317-318
parascapular flap isolation, 316
periosteal branches, 316
scapular tip and flap harvest, 319
stapling device placement, 314
teres major muscle retraction,
314-315
vascular highway, 314
pectoralis major musculocutaneous
flap
circumferential incision, 21
donor site closure, 24
hemostasis, muscle
transsection, 22
intercostal muscle dissection, 21
lateral border indentification, 20
medial attachments, 21
524 INDEX
Flap harvest technique (Continued)
pectoralis flap transfer, 23
skin paddle marking, 20
thoracoacromial artery
visualization, 22
twnnelcreation,23
vascular pedicle and transsected
muscle fibers, 23
posterior scalp flap
arc of rotation, 99
incisions, 98
posterior neck defect, 100
scalp elevation, 98-99
posterior tibial artery flap
anterior incision, 266
fascia, 267-268
flap dissection, 269-270
incision closure, 270-271
long saphenous vein and
nerve,267
possterior tibial artery, 268, 269
posterior incision, 268
radial forearm flap
beavertail modification, 192-195
donor site closure,191
proximal radial artery and venae
exposure, 190
radial artery dissection, 189
radial artery ligation and division,
188
sensory nerve markings, 187
skin flap elevation, 188, 189
ulnar/radial dissection, 188, 189
rectus abdominis flap
anterior rectus sheath incision,
152-154
cutaneous perforators, 158
deep inferior epigastric perforator
flap harvest, 158
donor defect closure, 156-157
flap extension, 152
lateral dissection, 154
Iineasanilunari~153
medial dissection, 154
perforator flap harvest, 159
segmental nerve supply, 155
scapular osteofasciocutaneous
flap
blood supply, 304
circumflex scapular vessel
palpation, 302
donor site closure, 307-308
lateral scapular dissection, 307
latissimus dorsi muscle, 309-312
medial to lateral dissection, 301
osteotomy, 305-306
scapular flap elevation, 303
scapular skin paddle, 303
shoulder mobilization, 301
superior incision, 302
teres major muscle dissection, 302
teres major transection, 304-305
thoracodorsal pedicle ligation and
transection, 305
sternocleidomastoid flap
caudal and middle blood supply
transsection, 74
incision and caudal aspect, 72
inferior and middle blood
supply, 73
random caudal extension, 73
superior thyroid pedicle
preservation, 73
submental island flap
anterior belly of digastric muscle
insertion, 115
defect and neck closure, 119
facial artery and vein
mobilization, 118
facial artery and vein position
identification, 111
facial artery and vien isolation, 114
flap elevation, 115-116
inferior incision, 112
ipsilateral anterior belly, 111
marginal mandibular branch
identification, 111
skin closure, 120
subplatysmal dissection, 113
superior incision, 113
vascular pedicle identification, 117
vertical and horizontal
dimension, 112
temporalis muscle flap
deep muscular fascia, 53
facial expression changes, 52
facial reanimation, 55
midfacial mimetic musculature, 52
neo-origin after transfer, 57
scalp incision, 53
temporalis muscle origin, 57
temporalis transfer, temporalis
tendon, 56
tendon dissection, 57
zygomatic arch, 54
temporoparietal fascial flap
anterior and posterior scalp flap
elevation, 228-229
complex upper lip defects, 230
frontal branch course, 228
superficial temporal artery and
vein, 227, 230
vertical incision, 227
tempoparietal fasciocutaneous flap,
231
trapezius muscle
lateral island trapezius flap, 33
lower trapezius island
musculocutaneous flap, 42-45
superior trapezius flap, 37-38
ulnar forearm free flap
circumferential incisions, 283
distal to proximal skin flap
elevation, 285
flap harvest, 286
medial and lateral skin flap
elevation, 283
radial and medial dissection, 283
subcutaneous vein identification, 282
superior incision, 282
ulnar artery ligation, 284
ulnar aspect, 284
Flap inset technique
anterolateral thigh free flap, 249-254
fibular osteocutaneous flap
cutaneous paddle and donor
vascular pedicle, 419-420
resected mandibular condyle,
428-433
right hemimandibular defect,
421-427
right hemimandibulectomy defect,
434-435
right infrastructure maxillectomy
defect, 436-443
free jejunal autografts, 468-472
free omentum and gastro-omentum
free flap, 484-487
gracilis muscle flap
bilateral gracilis muscle transfer, 174
cross facial nerve graft
anastomosis, 172
facial reanimation, 171
masseteric motor nerve isolation, 173
masseteric nerve reinnervation,
174
iliac crest osteocutaneous flap
right oromandibular defect,
393-394
right total maxillectomy defect,
398-402
iliac crest osteomusculocutaneous
flap,395-397
latissimus dorsi flap
anterior and right lateral
mandible, 354
anterior skin incision, 351
distal serratus anterior branch, 352
patient positioning and flap
design, 351
pleura exposure, rib graft
identification, 352
posterior rib cortex excision, 354
posterior skin incision, 353
rib bone graft fixing, 355
serratus anterior-rib component
elevation, 353
vascular pedicle and arc of
rotation, 353
palatal island flap
flap rotation, implants, 138
flap suture, 137
oroantral and oronasal defects
reconstruction, 137
paramedian forehead flap, 124-125,
129

radial forearm flap
circumferential
pharyngoesophageal
reconstnletion, 198-200
mediastinal trachea lengthening,
201-203
pharyngoesophageal
reconstnletion, 196-197
scapular osteofasciocutaneous flap
angular branch harvest, 320
carotid artery coverage, 321
scapular bone, 320
skin paddle, 321
serratus anterior flap
anterior and right lateral mandible,
354
anterior skin incision, 351
distal serratus anterior branch, 352
patient positioning and flap design,
351
pleura exposure, rib graft
identification, 352
posterior rib cortex excision, 354
posterior skin incision, 353
rib bone graft fixing, 355
serratus anterior-rib component
elevation, 353
vascular pedicle and arc of
rotation, 353
tempoparietal fasciocutaneous flap,
231
temporoparietal fascial flap, 221
Flexor carpi ulnaris (FCU), 272, 273,
276,277
Flexor digitorum superficialis muscle,
448
Flexor hallucis longus, 418
Forearm
anatomic variations, 183-184
anatomy, 178, 179
Free gastro-omental flap
design and utilization, 477--479
harvest techniques
defect closure, 487
distal anastomosis, 485
gastroepiploic pedicle, 483
proximal anastomosis, 486
stapling device, 482
total laryngopharyngectomy defect,
484
historical perspectives, 475--476
neurovascular anatomy, 4 79--480
pitfalls, 480
surgical technique, 480
Free jejunal autografts (FJAs)
flap design and utilization
circumferential defect
reconstruction, 456, 457
gastric pull-up procedure, 456
oral cavity mucosal defect
reconstnletion,457-458
pedicled jejunum, 456
pharyngoesophageal defect
reconstruction, 456-457
trachea replacement, 458
harvest technique, 466-467
historical perspectives, 455--456
inset technique, 468-472
neurovascular anatomy, 458, 459
pitfalls, 464-465
postoperative management, 463--464
preoperative assessment, 458, 460
surgical technique
enteric anastomoses, 463
enteric stapling device, 463
gastric decompression, 463
ischemic tolerance, 462
jejunal graft separation and
viability, 462
ligament ofTreitz identification, 460
perioperative antibiotics, 460
preoperative bowel
preparation, 460
transillumination, 460, 460
Free omental flap
design and utilization, 477--479
harvest techniques
defect closure, 487
distal anastomosis, 485
gastroepiploic pedicle, 483
proximal anastomosis, 486
stapling device, 482
total laryngopharyngectomy
defect, 484
neurovascular anatomy, 4 79--480
pitfalls, 480
surgical technique, 480
G design and utilization, 362-368
Gastric pull-up (GPU), 456, 457
Gemini flap, 8
Gillies up-and-down flap, 90
Glenohumeral joint, 292, 296
Gracilis muscle flap, 162-174
anatomic variations, 166
design and utilization
cross-face nerve grafting,
163-164
facial reanimation, 164-165
gracilis muscle segment
transfer, 164
masseter muscle, 164-165
neural plasticity, 165
obturator nerve, branching pattern,
164, 165
transverse skin paddle design, 165
vascular supply, 162, 164
flap harvest technique
anterior obturator nerve,169-170
initial incision, 168
left gracilis muscle flap, 168
neurovascular pedicle, 168
vascular pedicle dissection, 169
wound closure, 170
INDEX 525
historical perspectives, 162
inset technique
bilateral gracilis muscle transfer, 174
cross facial nerve graft
anastomosis, 172
facial reanimation, 171
masseteric motor nerve
isolation, 173
masseteric nerve reinnervation, 174
neurovascular anatomy
adductor artery, 165
blood supply, 166
minor vascular pedicle, 165
motor input, 166
pitfalls, 167
surgical approach, 162, 163
Greater auricular nerve, 49 1
Greater palatine artery, 131,
131-132, 132
Greater palatine canal, 131, 132
Greater palatine foramen, 131
Greater palatine nerve, 131
Greater palatine vein, 132
Groin flap, 330, 331
H
Harvest technique (see Flap harvest
technique)
I
Iliac bone
inner table, 366
orientations, 365
shapes, 364
Iliac crest osteocutaneous flap
anatomic variations, 371
harvest technique
cephalad margin incision, 390
defect closure, 392
external oblique muscle and
aponeurosis, 391
internal oblique muscle
incision, 391
historical pc:rpectives, 359
inset technique
right oromandibular defect, 393-394
right total maxillectomy defect,
398-402
neurovascular anatomy, 369--371
pitfalls, 371
postoperative care, 371
Iliac crest osteomusculocutaneous flap
anatomic variations, 371
design and utilization, 362-368
harvest techniques
abdominal wall closure, 384, 389
ascending branch, blood supply, 384
DCIA and DCIV dissection, 379
external oblique muscle
incision, 373
iliac bone exposure, 380
52& INDEX
iliac crest osteomusculocutaneous flap
(Continued)
iliacus transsection, 377
internal oblique muscle
elevation, 375
internal oblique muscle
exposure, 374
lateral dissection, 377-378
medial aspect, 383
medial bone cuts, 381
osteotomy, 381
skin incision, 372
synthetic mesh closure, abdominal
wall, 385-388
transversus abdominis muscle
transsection, 376
vascular pedicle, 372, 382
historical perpectives, 359
inset techniques, 395-397
neurovascular anatomy, 369-371
pitfalls, 371
postoperative care, 371
Infrastructure maxillectomy defect,
436-443
Inset teclmiques (see Flap inset
technique)
Internal jugular vein, 508
Internal mammary artery, deltopectoral
flap, 77,82
Internal mammary vein, deltopectoral
flap, 82
Internal oblique muscle, 361,
369,384
Ischemia, 411
J donor site selection, 39
Janus flap, 9, 10
Jejunum, 455--456 (see also Free jejunal
autografts (FJAs))
L outcome, 35-36
Laryngectomy
pectoralis major musculocutaneous
flap, 13
radial forearm flap, 178
Laryngopharyngectomy
anterolateral thigh free flap,
238
deltopectoral flap, 80
free jejunal autograft, 458
free omentum and gastro-omentum
flap,484-485
Lateral antebrachial cutaneous nerve,
183,187,189,191,195
Lateral arm flap
anatomic variations, 211
design and utilization
donor site closure, 207-208
dye-injection studies, 207
fascial flap, 208
fasciocutaneous flap, 206, 208
osteocutaneous flap, 208
posterior cutaneous nerve of the
arm, 208
pretransfer tissue expansion, 208
harvesting technique
anterior approach, 214-215
deltoid muscle insertion, 212
flap elevation, 215
lateral IM septum location, 212
meticulous dissection, 213
neurovascular pedicle, 216, 217
posterior incision, 213-214
PRCA ligation, 215-216
septocutaneous perforators, 214
soft tissue closure, 217-218
spiral groove visualization, 217
historical perspectives, 206-207
neurovascular anatomy
profunda brachii artery, 208,
20~210, 211
sensory nerves, 210-211,211
superficial and deep venous
system, 210
vascular supply, 211
pitfalls, 211
topographical anatomy, 207, 212
Lateral arm-proximal forearm
flap, 211
Lateral circumflex femoral artery
(LCFA), 237-243, 245-250
Lateral intermuscular septum, 447,447
Lateral island trapezius flap
applications, 35
facial reanimation, 33
harvesting technique
distal TCA and TCV ligation, 40
flap transposition, 41
neck dissection, 39
scapula position, 41
skin paddle incision, 40
pitfalls, 36
Lateral pectoral nerve, pectoralis
major flap, 16
Lateral plantar artery, 263
Lateral sural cutaneous nerve, 408,414
Lateral thoracic artery, pectoralis
major flap, 15, 16
Latissimus dorsi flap
anatomic variations, 340-341
anatomy, 327, 328
design and utilization
arc of rotation, 327
arterial and venous branch
transection, 327
cheek defects, 333
composite flap, 330--331
dynamic facial reanimation, 331,
333
groin flap transfer, 330, 331
midface defects, 333
perforators, 327, 330
pharyngoesophageal
reconstruction, 333
primary thinning, 330
reduced flap, 330
scalp defects, 333-334
sensory supply, 331
skin paddle, 330
total glossectomy, 333
harvest technique
adipose tissue removal, 344
arc of rotation maximization, 350
back skin elevation, 345
blunt and sharp dissection, 346
cutaneous paddle, 343
donor site closure, 350
initial incision, 343
pedicled flap passage, 348-349
tendinous insertion, 347
thoracoacromial pedicle, 348
thoracodorsal pedicle mobilization,
344,345
historical perspectives, 326
inset techniques, oromandibular
reconstruction
anterior and right lateral mandible,
354
anterior skin incision, 351
distal serratus anterior branch, 352
patient positioning and flap design,
351
pleura exposure, rib graft
identification, 352
posterior rib cortex excision, 354
posterior skin incision, 353
rib bone graft fixing, 355
serratus anterior-rib component
elevation, 353
vascular pedicle and arc of
rotation, 353
mechanism of actions, 327
neurovascular anatomy
angiosome concept, 336, 338
blood supply, 336,337
thoracodorsal nerve, 340
vascular pedicle length, 340
pitfalls, 341-342
preoperative assessment, 342
Lesser palatine artery, 131, 132
Lesser palatine foramen, 131, 132
linea alba, 149
linea semilunaris, 149
lingual artery, 507
Locking reconstruction plate (LRP),
422--424
Lower trapezius island
musculocutaneous flap
angiosome concept, 33-35
arc of rotation, 35
flap harvest technique
donor site closure, 45
DSA and DSV identification, 43
DSA and DSV preservation, 44-45

muscle anaclunents, 43
skin paddle incision, 42
outcome, 36
skin paddle, 33
M locking reconstruction plate, 453
Mandibulectomy
iliac crest osteocutaneous and
osteomusculocutaneous flap, 367
pectoralis major musculocutaneous
flap, 12-13
Masseter muscle, 164-165
Mastectomy, 341
Mastoid-occiput-based shoulder flap, 27
Medial antebrachial cutaneous nerve
graft
anatomic variations, 493
design and utilization, 491,493
flap harvest technique, 494-495
neurovascular anatomy, 493
pitfalls, 493
postoperative wound care, 493
preoperative assessment, 493
sensory distribution, 491, 492
Medial arm flap, 206
Medial pectoral nerve, pectoralis major
flap, 15, 16
Medial plantar artery, 263
Medial sural cutaneous nerve, 498, 499
Medially based deltopectoral flap,
76,77
Median cubital vein, 182
Middle temporal artery, 220, 221,
224,229
Mucosal carcinoma, 421
Musculocutaneous perforator, 241, 276
N tonsillar fossa and retromolar
Neuroma, 499
Neurovascular bundles, 375
0 flap rotation, implants, 138
Omentum
blood supply, 477, 478
lymphatic channels, 476,477
neurovascular anatomy, 4 79-480
Ostectomy
donor-site morbidity, OCRFFF,
446--447
resected mandibular condyle, 431--432
right infrastructure maxillectomy
defect, 440
Osteocutaneous radial forearm free flap
(OCRFFF), 178
advantage, 447
anatomic variations, 448
design and utilization
bonelength,446,446
donor-site morbidity, radius bone
fracture, 446--447
limited maxillectomy defects
reconstruction,447
soft tissue versatility, 447
harvest techniques
bicortical screws, 453
flexor pollicis longus, 451, 452
lateral antebrachial cutaneous
nerveseparation,452
prophylactic fixation, 452
radial artery and cephalic vein
marking, 450
residual muscle suture, 453
historical perspectives, 445
neurovascular anatomy, 447, 447
pitfalls, 448
postoperative consideration, 449
preoperative considerations, 448--449
Osteotomy
donor-site problem, radial forearm
flap, 185
iliac crest-internal oblique fla~ 381, 394
oromandibular reconstruction,
latissimus dorsi, 354
parascapular flap, 307, 318
radial forearm osteocutaneous
flap, 451
scapular osteocutaneous flap, 306
temporalis muscle tendon, 56
ulnar forearm free flap, 280
p harvesting technique, 301-324
Palatal island flap
design and utilization, 131
harvest techniques
contralateral pedicle isolation and
transsection, 134-135
flap rotation, 136
mucoperiosteal flap elevation, 133
trigone defect closure, 136
historical perspectives, 130-131
inset techniques
flap suture, 137
oroantral and oronasal defects
reconstruction, 137
neurovascular anatomy, 131, 131-132,
132
pitfalls, 13 2
postoperative care, 132
Paramedian forehead flap
anatomic variations, 125
design and utilization
nasal, orbital, or midfacial defi:ct, 123
pedicle, incision extension, 122-123
radial forearm free flap, 123
harvest techniques
calvarial bone graft, 127-128
radial forearm tissue, 125
structural elements, 126
superior and inferior flap
elevation, 126
supratrochlear vessels, 124
trilayered approach, 126
INDEX 527
historical perspectives, 122
inset technique, 124-125, 129
neurovascular anatomy, 125
pitfalls, 125, 129
postoperative care, 129
Parascapular fasciocutaneous flap
anatomic variations, 299-300
design and utilization, 293-299
harvesting technique
anatomic landmarks, 313
angular branch elevation, 315
axilla, 317
flap incision and elevation, 313
osteotomy, 317-318
parascapular flap isolation, 316
periosteal branches, 316
scapular tip and flap harvest, 319
stapling device placement, 314
teres major muscle retraction,
314-315
vascular highway, 314
historical perpectives, 292-293
neurovascular anatomy, 299
pitfalls, 300
postoperative care, 300
Parascapular osteofasciocutaneous flap
anatomic variations, 299-300
design and utilization, 293-299
historical perspectives, 292-293
neurovascular anatomy, 299
pitfalls, 300
postoperative care, 300
Parotidectomy
rectus abdominis flap, 148
sternocleidomastoid flap, 64
Pectoralis major musculocutaneous
flap
advantages, 5-6
anatomic variations, 18
anatomy, 4, 4
design and utilization
anterior mandibulectomy, 12-13
arc of rotation, improving
methods, 6
excessive flap bulk, 6-7, 8
facial reanimation, 13
laryngectomy, 13
neck, muscular pedicle, flap
modifications, 10--11
pharyngoesophageal
reconstruction, 13
stomal recurrent cancer, 13, 14
tracheal resection, 13
two epithelial surface, 7-9, 9, 10
fascia, 5
harvest techniques
donor site closure, 24
hemostasis, muscle transsection, 22
intercostal muscle dissection, 21
lateral border indentification, 20
medial attachments, 21
528 INDEX
Pectoralis major musculocutaneous flap
(Continued)
pectoralis flap transfer, 23
skin paddle marking, 20
thoracoacromial artery
visualization, 22
nwrunelcreation,23
vascular pedicle and transsected
muscle fibers, 23
historical perpectives, 3-4
mechanism of action, 5
neurovascular anatomy
angiosome concept, 16-17
clavipectoral fascia, 17, 18
internal mammary perforators, 17
lateral thoracic artery, 13-14
nerve supply, 17-18
primary vascular supply, 13, 15
superior thoracic artery, 16
thoracoacromial artery, 13, 14, 16
pitfalls
donor site selection, 18, 19
flap necrosis, 18
inflamammary skin paddle, 19, 19
pedicle compression, 18
pulmonary complications, 18--19
skin paddle, 5, 5
Peroneal artery, 405,417
Peroneal nerve, 412
Peroneal vascular system, 41 7
Peroneal vein, 405, 41 7
Pinch test, submental island flap, 103
Poland's syndrome, 18,341
Popliteal artery, 408
Posterior auricular artery,
sternocleidomastoid flap, 69
Posterior crural septum, 406, 415
Posterior cutaneous nerve of arm
(PCNA), 208, 210,211, 216
Posterior cutaneous nerve of forearm
(P~,208,210,211,213,216
Posterior scalp flap
design and utilization, 90
harvesting technique
arc of rotation, 99
incisions, 98
posterior neck defect, 100
scalp elevation, 98-99
historical perspectives, 89-90
neurovascular anatomy, 91, 93
pitfalls, 93
postoperative wound care, 93
Posterior tibial artery (PTA) flap
anatomic variation
hypoplastic/aplastic PTA, 264
popliteal artery branching patterns,
263-264
vascular anomalies, 264
design and utilization, 258, 259, 260
harvest techniques
anterior incision, 266
fascia, 267-268
flap dissection, 269-2 70
incision closure, 270-271
long saphenous vein and nerve,
267
posterior tibial artery, 268, 269
posterior incision, 268
historical perspectives, 257
neurovascular anatomy
lower leg, 261, 261, 262
medial and lateral plantar artery,
263
perforator-based flaps, 263
peroneal and circumflex fibular
artery, 261, 263
saphenous nerve, 263
venae comitantes, 261
venousdr~nage,263
pitfalls, 264
postoperative care, 265
preoperative management, 264-265
skin graft placement, 257-258
Posterior ulnar recurrent artery, UFFF,
273
Postlaryngectomy voice rehabilitation,
180
Primary ALT flap thinning, 237
R lingual artery, lessers triangle,
Radial artery,176-190
Radial forearm flap
anatomic variations, 183-184
design and utilization
beavertail modification, 180
brachioradialis, 178
circumferential
pharyngoesophageal
reconstruction, 180
fascial free flap, 177
flap size and shape, 176, 177
head and neck defects,178, 180
hypopharynx, 180, 181
intraoral reconstruction, 177
larynx and pharynx. 180
palatal reconstruction, 178
pharyngeal defect, 177-178
postlaryngectomy voice
rehabilitation, 180
residual tongue mobility, 177
sensate flaps, 177-178
skin thickness, 180
harvesting technique
beavertail modification, 192-195
donor site closure, 191
proximal radial artery and venae
exposure, 190
radial artery dissection, 189
radial artery ligation and division,
188
sensory nerve markings, 187
skin flap elevation,188, 189
ulnar/radial dissection, 188, 189
historical perspectives, 176
inset techniques
circumferential
pharyngoesophageal
reconstruction, 198-200
mediastinal trachea lengthening,
201-203
pharyngoesophageal
reconstruction, 196-197
neurovascular anatomy
arterial systems, 181
blood supply, 180-181
cephalic vein, 182
radial recurrent artery, 182
sensory nerve supply,183, 183
venous supply,181-182
pitfalls, 184-185
postoperative care, 186
preoperative management, 185-186
Radial nerve, 183
Recipient vessel selection
artery selection, 507-508
caliber and length, 506
defect location, 506
flap insetting, 506
harvest techniques
internal mammery artery and vein,
516--519
510-511
TAC system, 512-515
interposition vein grafts, 508-509
meticulous planning, 506
microvascular surgery, 506-507
pedicle geometry, 509
vascular pedicle tunneling, 509
vein selection, 508
Rectus abdominis flap
anatomic variations, 150
anatomy, 141-143
design and utilization
defect size and volume, 143
donor site, TRAM flap, 143, 145
motor-sensory nerves, 143
periumbilical perforators, 143, 144
scalp reconstruction, 146
skull base reconstruction, 145-146
split-thickness skin graft, 143
thinner flap harvest, 145
three-cavity defect, 146
total glossectomy, 145
harvesting technique
anterior rectus sheath incision,
152-154
blunt dissection, 155
cutaneous perforators, 158
deep inferior epigastric perforator
flap harvest, 158
donor defect closure,156-157
flap extension, 152
lateral dissection, 154
linea semilunaris, 153
medial dissection, 154

meticulous dissection, 153
perforator flap harvest, 159
segmental nerve supply, 155
historical perspectives, 141, 150
neurovascular anatomy
angiosome concept, 148
inferior pedicle, 146, 148
nerve supply, 148
vascular supply, 146, 147
venous supply, 146
pitfalls, 150-151
postoperative care, 151
Rectus fascia, 149-150
Regional lymphatic metastases,
sternocleidomastoid flap,
59-60
Reverse flow modification, submental
island flap, 102
Rib, pectoralis major, 9, 11
s anterior-rib osteomusclular flaps,
Saphenous nerve, 263
Saphenous neuralgia, 263
Scalp
dermis, 90
epidennis, 90
neurovascular anatomy, 91-93
vascular supply, 89-90, 91-92
Scapular bone, blood supply, 296
Scapular fasciocutaneous flap
anatomic variations, 299-300
flap design and utilization, 293-299
harvesting technique, 323-324
historical aspects, 292-293
neurovascular anatomy, 299
pitfalls, 300
postoperative care, 300
Scapular osteofasciocutaneous flap
anatomic variations, 299-300
flap design and utilization, 293-299
harvesting technique
blood supply, 304
circumflex: scapular vessel
palpation, 302
donor site closure, 307-308
lateral scapular dissection, 307
latissimus dorsi muscle, 309-312
medial to lateral dissection, 301
osteotomy, 30~306
scapular flap elevation, 303
scapular skin paddle, 303
shoulder mobilization, 301
superior incision, 302
teres major muscle dissection, 302
teres major transection, 304-305
thoracodorsal pedicle ligation and
transection, 305
historical aspects, 292-293
inset techniques, mandibular
reconstruction
angular branch harvest, 320
carotid artery coverage, 321
scapular bone, 320
skin paddle, 321
neurovascular anatomy, 299
pitfalls, 300
postoperative care, 300
utilization, 293--299
Scapular tip, 313-319
Secondary alopecia, 225
Semicircular line, 149
Sensate flaps,177-178
Septocutaneous perforators
fibular osteocutaneous flap, 409, 410
lateral arm flap, 207,213, 214
posterior tibial artery flap, 268
lnnar forearm free flap, 273, 276
Serratus anterior flap
anatomic variations, 341
anatomy, 327,329
design and utilization
anterior myofascial flaps, 335
335-336
facial reanimation, 334-335
free flap, 334
head and neck reconstruction, 334
myofascial flap, 334--335
pedicled flap, 334
thoracodorsal artery, 324
vascularized rib, 335
historical perspectives, 326-327
inset techniques, oromandibular
reconstruction
anterior and right lateral mandible,
354
anterior skin incision, 351
distal serratus anterior branch, 352
patient positioning and flap design,
351
pleura exposure, rib graft
identification, 352
posterior rib cortex excision, 354
posterior skin incision, 353
rib bone graft fixing, 355
serratus anterior-rib component
elevation, 353
vascular pedicle and arc of
rotation, 353
neurovascular anatomy
blood supply, 336, 339, 340
thoracodorsal nerve, 340
vascular pedicle length, 340
pitfalls, 341, 342
Spinal accessory nerve,
sternocleidomastoid flap, 31
Split musculocutaneous flap,
sternocleidomastoid flap, 62
Sternocleidomastoid (SCM) flap
anatomy, 69-71
design and utilization
donor site contour deformity, 62
facial reanimation, 64
mandibular reconstruction, 65, 65
INDEX 529
musculoclavicular flap, 65-66
musculocutaneous unit transfer,
60,62
myoperiosteal flap, 66, 66
oral and pharyngeal mucosal
defects, 64
parotidectomy, 64
trachealrepak,66,67-68
vascularized bone transfer, 64--65
harvesting techniques
caudal and middle blood supply
transsection, 74
incision and caudal aspect, 72
inferior and middle blood supply, 73
random caudal ex:tension, 73
superior thyroid pedicle
preservation, 73
historical perspectives, 59
neurovascular anatomy
platysma muscle, 70, 70
posterior auricular artery, 69
superior thyroid artery, 64, 69,71
vascular supply, 69, 69-70
pitfalls, 71
Sternum, pectoralis major flap, 10, 12
Stomach
blood supply, 478
neurovascular anatomy, 479--480
Subfascial flap dissection, anterolateral
thigh free flap, 237
Submandibular triangle, submental
island flap, 109
Submental island flap
anatomic variations, 109
angiosome concept, 107
design and utilization
arc of rotation, 103, 107
cheek reconstruction, 103
facial artery, 107-108
flap dimension, 103
hak-bearing skin, upper lip and
chin reconstruction, 103
oral commissure reconstruction, 103
pinch test, 103
prefabrication and multiple flap
combination, 103
reverse flow strategy, 104--107,107
skin laxity, 103
subcutaneous flap, 103
vascular supply, 103
vascularized bone harvest, 103
venous anatomy, neck, 104, 105
flap harvest techniques
anterior belly of digastric muscle
insertion, 115
defect and neck closure, 119
facial artery and vein mobilization,
118
facial artery and vein position
identification, 111
facial artery and vien isolation, 114
flap elevation, 115-116
530 INDEX
Submental island flap (Continued)
inferior incision, 112
ipsilateral anterior belly, 111
marginal mandibular branch
identification, 111
skin closure, 120
subplatysmal dissection, 113
superior incision, 113
vascular pedicle identification, 117
vertical and horizontal dimension,
112
historical perspectives, 102
neurovascular anatomy
anterior belly of the digastric
muscle, 109
mylohyoid muscle, 109
perforating vessels, 107-108
submandibular gland, 108-1 09
submandibular triangle, 109
submental artery, 107-109, 108
vascular information, 109
pitfalls, 109-110
postoperative care, 11 0
preoperative management, 110
Subscapular artery, multiple branches,
288--289
Subscapular flap system
latissimus dorsi muscle flap, 290
osteocutaneous flap, 290
subscapular artery and vein,
branching pattern, 288, 289
total scalp defect, 290
Subscapular mega flap
anatomic variations, 299-300
design and utilization, 293-299
harvesting technique, 311-312
historical perspectives, 292-293
neurovascular anatomy, 299
pitfalls, 300
postoperative care, 300
Superficial circumflex iliac artery, 359
Superficial femoral artery, gracilis
muscle, 165
Superficial inferior epigastric artery
(SIEA),148
Superficial palmar arch, UFFF, 272,
277,278,279
Superficial temporal artery,
temporoparietal fascial flap, 227,
230
Superficial temporal vein,
temporoparietal fascial flap, 227,
230
Superficial ulnar artery, 277, 279
Superior thoracic artery, pectoralis
major flap, 16
Superior thyroid artery,
sternocleidomastoid flap, 64,
69,71
Superior trapezius flap
angiosome concept, 31, 33
applications, 33
harvesting technique, 37-38
outcome, 33
pitfalls, 36
Supratrochlear artery, 123
Sural nerve graft
advantages, 497
anatomic variations, 499
anatomy, 496
design and utilization, 497, 499
flap harvest technique
endoscopic technique, 501
horizontal incision, 500
nerve dissection, 500
nerve gaps repairing, 501
proximal sural nerve stump
placement, 502
topographical landmarks, 500
neurovascular anatomy, 499
pitfalls, 499
postoperative wound care, 499
preoperative assessment, 499
T surgical approach, 163
Temporal fossa, 221, 222
Temporalis muscle flap
design and utilization, 48-49
harvest techniques
deep muscular fascia, 53
facial expression changes, 52
facial reanimation, 55
midfacial mimetic musculature, 52
neo-origin after transfer, 57
scalp incision, 53
temporalis muscle origin, 57
temporalis transfer, temporalis
tendon, 56
tendon dissection, 57
zygomatic arch, 54
neurovascular anatomy
muscular fascia, 50-51
vascular supply, 49-50, 50
pitfalls, 51
postoperative wound care, 51
preoperative assessment, 51
Temporalis muscular fascia, 220,
221-225,227,229
Temporoparietal fascial flap, 219-231
anatomic variations, 225
design and utilization
auricular reconstruction, 220-221
muscular and fascial layers, 221,
221
nasal reconstruction, 222
scalp and lip reconstruction, 222
skin and cartilage transfer, 222
temporoparietal fascia, 222-223
vascularity, 223
vascularized calvarial bone
transfer, 222-223
harvesting technique
anterior and posterior scalp flap
elevation, 228-229
complex upper lip defects, 230
frontal branch course, 228
hair-bearingTPFF transposition,
231
superficial temporal artery and
vein, 227, 230
vertical incision, 227
historical perspectives, 219
inset technique, 221, 231
neurovascular anatomy
motor and sensory nerves, 225
superficial temporal artery and
vein, 223-225
vascularized fascia, 224, 224
zygomatico-orbital artery, 225
pitfalls, 225-226
postoperative wound care, 226
preoperative assessment, 226
Thigh
adductor artery, 166
muscular anatomy, 162, 163
sensory supply, 166
Thoracoacromial artery, pectoralis
major flap, 16
Thoracoacromial pedicle, 348
Thoracodorsal artery, 335, 337, 341
Thoracodorsal nerve, 340
Thoracodorsal neurovascular pedicle,
347
Thoracodorsal pedicle, 334, 335
Thoracodorsal vein, 336,337, 340
Thoracolumbar fascia, 333
Thyrocervical trunk,
sternocleidomastoid flap,
59,69
Tracheal resection, pectoralis
major, 13
Transient ulnar nerve paresthesia, 281
Transverse cervical artery (TCA), 27,
29,30-41,45,507
Transverse cervical vein (TCV), 27, 28,
31, 33, 35-40,45
Transverse cervical veins, 508
Transverse rectus abdominis
musculocutaneous (TRAM)
flap,143,145,146,148
Transversus abdominis muscle, 376,
392
Trapezius muscle flap
historical perspectives, 27
lateral island trapezius flap, 33,39-41
lower trapezius island
musculocutaneous flap
angiosome concept, 33, 34, 35
arc of rotation, 35
DSA pedicle, 35
harvesting techniques, 42-45
muscle anatomy, 28, 29
neurovascular anatomy, 28
accessory nerve, 31
blood supply, 28

TCA and DSA, 28, 30, 31
venous anatomy, 31
pitfalls, 35-36
superior trapezius flap
angiosomes, 31, 32, 33
blood supply, 31
harvesting techniques, 37-38
radical neck dissection, 33
total necrosis, 31
trapezius osteomusculocutaneous
flap,35
Trapezius osteomusculocutaneous
flap,35
Tripartite iliac internal oblique
osteomusculocutaneous
flap,368
Tubed gastro-omental free flap,
481--483
u neurovascular anatomy
Ulnar artery, 180, 183, 184, 185
Ulnar forearm free flap (UFFF)
anatomic variations, 278, 278-279
cross-sectional anatomy, 272, 274
design and utilization
advantages, 279
head and neck cancer defects, 280
oral cavity reconstruction, 280
oropharyngeal reconstruction, 279
pharyngoesophagus
reconstruction, 279-280
skin paddle, 279
flap harvest techniques
circumferential incisions, 283
distal to proximal skin flap
elevation, 285
flap harvest, 286
medial and lateral skin flap
elevation, 283
radial and medial dissection, 283
subcutaneous vein identification,
282
superior incision, 282
ulnar artery ligation, 284
ulnar aspect, 284
historical perspectives, 272
blood supply, 272-273, 273, 274
deep and superficial palmar
arch,277
posterior interosseous artery, 274,
275
posterior ulnar recurrent
artery, 273
INDEX 531
proximal and distal
musculocutaneous
perforator, 276
septocutaneous perforators,
276
superficial muscles, 272, 273
ulnar nerve, 277
vascular dominance, 276-277
venous and sensory supply,
277,277
pitfalls, 280--281
postoperative care, 281
preoperative management, 281
proximal branches, 273,275
vs. radial forearm free flap, 272
Unipedicle lateral island trapezius
flap, 28
v
Vascularized bone
pectoralis major, 9-12
submental island flap, 103
Vascularized muscle flap, 333
Vein grafts, 509
z
Zygomatico-orbital artery, 225