Jurnal On Hand Surgery

THE JOURNAL OF
HAND
SURGERY
Journal of Hand Surgery:
The Journal of Hand Surgery (ISSN 0363-5023) is published 10 times
a year by Elsevier Inc., 360 Park Avenue South, New York, NY
10010-1710. Months of issue are January, February, March, April,
May-June, July-August, September, October, November and December. Business Office: 1600 John F. Kennedy Blvd., Ste. 1800, Philadelphia, PA 19103-2899. Editorial Office: 360 Park Avenue South,
New York, NY 10010-1710. Customer Service Office: 6277 Sea Harbor Drive, Orlando, FL 32887-4800. Periodicals postage paid at New
York, NY and additional mailing offices.
POSTMASTER:
Send address changes to The Journal of Hand Surgery, Elsevier Customer Service, 6277 Sea Harbor Drive, Orlando, FL 32887-4800.
Editorial:
Editorial correspondence for Scientific papers should be addressed
to: Paul R. Manske, MD, The Journal of Hand Surgery, Washington
University Medical Center, Department of Orthopaedic Surgery,
One Barnes Plaza, Suite 11300 WP, St Louis, MO 63110.
Editorial correspondence for Review/Technique papers should be
addressed to: Arnold-Peter C. Weiss, The Journal of Hand Surgery,
University Orthopedics, 2 Dudley St, Ste 200, Providence, RI 02905.
Subscriptions:
Correspondence regarding subscription or change of address
should be directed to The Journal of Hand Surgery, Elsevier,
Periodicals Department, 6277 Sea Harbor Dr, Orlando, FL
32887-4800 or Email elspcs@elsevier.com. Change of address notices, including both the old and new addresses of the subscriber
and the mailing label, should be sent at least one month in advance. Customer Service: (800) 654-2452; outside the United
States and Canada, (407) 345-4000.
Yearly subscription rates: United States and possessions: individuals, $249.00; institutions, $462.00; students and residents, $114.00;
single issues, $56.00. All other countries: individuals, $366.00; institutions, $589.00; students and residents, $184.00; single issues,
$59.00. For all areas outside the United States and possessions, there
is no additional charge for surface delivery. To receive in-training rate,
orders must be accompanied by name of affiliated institution, date of
term, and the signature of program/residency coordinator on institution letterhead. Orders will be billed at individual rate until proof of
status is received. Current prices are in effect for back volumes and
back issues. Single issues, both current and back, exist in limited
quantities and are offered for sale subject to availability. Back issues
sold in conjunction with a subscription are on a prorated basis. Checks
should be made payable to Elsevier and sent to The Journal of Hand
Surgery, Elsevier, Periodicals Department, PO Box 628239, Orlando,
FL 32862-8239.
Copyright:
Copyright 2007 by the American Society for Surgery of the Hand.
All rights reserved. No part of this publication may be reproduced or
transmitted in any form or by any means now or hereafter known,
electronic or mechanical, including photocopy, recording, or any

information storage and retrieval system, without permission in writing
from the Publisher. Printed in the United States of America. Permission
may be sought directly from Elseviers Global Rights Department in
Oxford, UK: phone 215-239-3804 or 44 (0)1865 843830, fax
44 (0)1865 853333, e-mail healthpermissions@elsevier.com.
Requests may also be completed online via the Elsevier homepage
(www.elsevier.com/permissions). The appearance of the code at
the bottom of the first page of an article in this journal indicates the
copyright owners consent that copies of the article may be made
for personal or internal use, or for the personal or internal use of
specific clients, for those registered with the Copyright Clearance
Center, Inc. (222 Rosewood Drive, Danvers, MA 01923; (508)
750-8400; www.copyright.com). This consent is given on the condition that the copier pay the stated per-copy fee for that article
through the Copyright Clearance Center, Inc. for copying beyond
that permitted by Sections 107 or 108 of the US Copyright Law.
This consent does not extend to other kinds of copying, such as
copying for general distribution, for advertising or promotional
purposes, for creating new collective works, or for resale. Absence
of the code indicates that the material may not be processed
through the Copyright Clearance Center, Inc.
Reprints:
For 100 or more copies of an article in this publication, please
contact the Commercial Reprints Department, Elsevier Science
Inc., 360 Park Avenue South, New York, New York 10010-1710.
Telephone (212) 633-3813; fax (212) 633-3820; e-mail,
reprints@elsevier.com. Reprints of single articles available online
may be obtained by purchasing Pay-Per-View access for $25 per
article on the journal web site, www.jhandsurg.org.
Advertising:
Advertising representative: Steve Everly, KI Lipton, Inc./MedMedia,
132 Welsh Road, Suite 120, Horsham, PA 19044. Telephone (267)
893-5686; fax: (267) 893-5682; e-mail: severly@ki-lipton.com
Disclaimer:
The ideas and opinions expressed in The Journal of Hand Surgery do
not necessarily reflect those of the Editors or the Publisher. Publication
of an advertisement or other product mention in The Journal of Hand
Surgery should not be construed as an endorsement of the product or
the manufacturers claims. Readers are encouraged to contact the
manufacturer with any questions about the features or limitations of
the products mentioned. The publisher does not assume any responsibility for any injury and/or damage to persons or property arising out
of or related to any use of the material contained in this periodical. The
reader is advised to check the appropriate medical literature and the
product information currently provided by the manufacturer of each
drug to be administered to verify the dosage, the method and duration
of administration, or contraindications. It is the responsibility of the
treating physician or other health care professionals, relying on independent experience and knowledge of the patient, to determine drug
dosages and the best treatment for the patient.
THE JOURNAL OF
HAND
SURGERY
Editors
Paul R. Manske St. Louis, Missouri
Arnold-Peter C. Weiss Providence, Rhode Island
Deputy Editors
Specialty Editors
Roy A. Meals Los Angeles, California
Mark E. Baratz Pittsburgh, Pennsylvania
Vincent R. Hentz Palo Alto, California
Richard A. Berger Rochester, Minnesota
Kevin C. Chung Ann Arbor, Michigan
Associate Editors: Scientific
Associate Editors: Review/Technique
Reid A. Abrams, San Diego, California

Marie A. Badalamente, Stony Brook, New York
Gregory I. Bain, Adelaide, Australia
Louis W. Catalano III, New York, New York
James Chang, Stanford, California
A. Lee Dellon, Baltimore, Maryland
Marc Garcia-Elias, Barcelona, Spain
Steven Glickel, New York, New York
Charles Goldfarb, St. Louis, Missouri
Brent Graham, Toronto, Canada
Richard S. Idler, Indianapolis, Indiana
Michelle A. James, Sacramento, California
Peter J.L. Jebson, Ann Arbor, Michigan
Wally Jones, Redmond, Washington
Jesse B. Jupiter, Boston, Massachusetts
Graham J.W. King, London, Ontario
Lewis B. Lane, Manhasset, New York
H. Relton McCarroll, Jr, San Francisco, California
Steven L. Moran, Rochester, Minnesota
Peter M. Murray, Jacksonville, Florida
Sanjiv H. Naidu, Hershey, Pennsylvania
Toshihiko Ogino, Yamagata, Japan
Rita Patterson, Galveston, Texas
William C. Pederson, San Antonio, Texas
Craig S. Phillips, Chicago, Illinois
Ghazi M. Rayan, Oklahoma City, Oklahoma
Alexander Y. Shin, Rochester, Minnesota
Walter H. Short, Syracuse, New York
Matthew M. Tomaino, Rochester, New York
Thomas Earl Trumble, Seattle, Washington
Scott W. Wolfe, New York, New York
Kimberly Amrami, Rochester, Minnesota

Peter Burge, Oxford, United Kingdom
Mark Cohen, Chicago, Illinois
Ragnar Haugstvedt, Oslo, Norway
Daniel Herren, Zurich, Switzerland
Scott H. Kozin, Philadelphia, Pennsylvania
Christophe Mathoulin, Paris, France
John A. McAuliffe, Fort Lauderdale, Florida
Marco Ritt, Amsterdam, The Netherlands
David Ruch, Winston-Salem, North Carolina
Kavi Sachar, Denver, Colorado
John Gray Seiler, Atlanta, Georgia
Ian Trail, Wigan, United Kingdom
Editorial Offices
The Journal of Hand Surgery (Dr. Manske)
Washington University Medical Center
Department of Orthopaedic Surgery
One Barnes Plaza, Suite 11300
St. Louis, MO 63110
Phone: (314) 747-2537
Fax: (314) 747-2801
The Journal of Hand Surgery (Dr. Weiss)
University Orthopedics
2 Dudley Street, Suite 200
Providence, RI 02905
Phone: (401) 457-1522
Fax: (401) 831-5874
British and European Volume

Editor
David Elliot, Chelmsford
Assistant Editors
Grey Giddins, Bath
Francisco del Pinal, Santander
Journal Committee Chairman
Tim Davis, Nottingham
Statistical Advisor
Stefan Sauerland, Cologne
Editorial Board
Michel Boeckstyns, Hellerup
Christian Dumontier, Paris
Angel Ferreres, Barcelona
Joel Engel, Ramat Gan
Jan-Ragnar Haugstvedt, Oslo
Carlos Heras-Palou, Derby
Jonathan Hobby, Basingstoke
Paul R. Manske, St. Louis
Ladislav Nagy, Zurich
Mark Pickford, East Grinstead
Philippe Saffar, Paris
David Shewring, Cardiff

Ian Trail, Wigan
Esther Vogelin, Bern
David Warwick, Southampton
Carlos Wigderowitz, Dundee
Andrzej Zyluk, Szczecin
Editorial Office
Journal of Hand Surgery
PO Box 7704
Oakham LE15 6ZH, UK
editor@journalofhandsurgery.com
THE JOURNAL OF
HAND
SURGERY
VOLUME 32A, NUMBER 7, SEPTEMBER 2007
Editorial
History at Our Fingertips
941
Paul R. Manske
History of Hand Surgery

A History of Surgery in the Instrument Tray: Eponymous
Tools Used in Hand Surgery
942
Clifton G. Meals and Roy A. Meals
Radius Fracture
Results of Dorsal or Volar Plate Fixation of AO Type C3
Distal Radius Fractures: A Retrospective Study
954
Susanne Rein, Hartmut Schikore, Wolfgang Schneiders, Michael Amlang,

and Hans Zwipp
The Risk of Adverse Outcomes in Extra-Articular Distal

Radius Fractures Is Increased With Malalignment in Patients
of All Ages but Mitigated in Older Patients
962
Ruby Grewal and Joy C. MacDermid
Biomechanical Evaluation of Locking Plate Radial Shaft

Fixation: Unicortical Locking Fixation Versus Mixed
Bicortical and Unicortical Fixation in a Sawbone Model
971
Jason W. Roberts, Steven I. Grindel, Brandon Rebholz, and Mei Wang
Distal Radius Osteotomy in the Elderly Patient Using

Angular Stable Implants and Norian Bone Cement
976
Santiago Lozano-Calderon, Michael Moore, Matthew Liebman, and

Jesse B. Jupiter
Rupture of the Flexor Pollicis Longus Tendon After Volar

Fixed-Angle Plating of a Distal Radius Fracture: A Case Report
984
Raymond A. Klug, Cyrus M. Press, and Mark H. Gonzalez
Brachial Plexus
Transfer of the Accessory Nerve to the Suprascapular Nerve
in Brachial Plexus Reconstruction
Jayme Augusto Bertelli and Marcos Flavio Ghizoni
989
Contents Continued
Clinical-Electromyography Correlation in Infants With

Obstetric Brachial Plexopathy
999
Carlos O. Heise, Mario G. Siqueira, Roberto S. Martins, and

Jose Luiz D. Gherpelli
Carpus
Anthropometry of the Human Scaphoid 1005
Andrew D. Heinzelmann, Graeme Archer, and Randy R. Bindra
Association Between Lunate Morphology and Carpal 1009

Collapse Patterns in Scaphoid Nonunions
Steven C. Haase, Richard A. Berger, and Alexander Y. Shin
Carpal Kinematics During Simulated Active and Passive 1013

Motion of the Wrist
Rita M. Patterson, Laura Williams, Clark R. Andersen, Shukuki Koh, and
Steven F. Viegas
Long-Term Follow-Up of an Undiagnosed Trans-Scaphoid 1020

Perilunate Dislocation Demonstrating Articular Remodeling
and Functional Adaptation
Elizabeth A. Bathala and Peter M. Murray
Congenital
Clinical Manifestations of Type IV Ulna Longitudinal Dysplasia 1024
Bassem El Hassan, Sam Biafora, and Terry Light
Objective Features and Aesthetic Outcome of Pollicized 1031

Digits Compared With Normal Thumbs
Charles A. Goldfarb, Valerie Deardorff, Ben Chia, Amy Meander, and
Paul R. Manske
Isolated Wedge Osteotomy of the Ulna for Mild Madelungs 1037

Deformity
Yann Glard, Andre Gay, Franck Launay, Didier Guinard, and Regis Legre
Surgical Treatment of the Pediatric Trigger Finger 1043

Donald S. Bae, Samir Sodha, and Peter M. Waters
Vascular
Replantation of Completely Amputated Thumbs With 1048
Venous Arterialization
Lijie Tian, Furong Tian, Feng Tian, Xiaochuan Li, Xianglu Ji, and Jiao Wei
Digital Ischemia Due to Essential Thrombocythemia: 1053

A Case Report
Anastasios Papadonikolakis, George D. Chloros, Beth P. Smith, and
L. Andrew Koman
Contents Continued
Aeromonas hydrophila Infection Causing Delayed Vascular 1058

Thrombosis in a Nearly Amputated Finger
Fernando A. Herrera, Karen Horton, Ahmed Suliman, and Gregory M. Buncke
Review Articles
Spontaneous Flexor Tendon Ruptures of the Hand: Case 1061
Series and Review of the Literature
Aaron J. Bois, Geoffrey Johnston, and Dale Classen
Immunologic Approaches to Composite Tissue Allograft 1072

Aure`le Taieb, Julio A. Clavijo-Alvarez, Giselle G. Hamad, and
W. P. Andrew Lee
Stability of the Distal Radioulna Joint: Biomechanics, 1086

Pathophysiology, Physical Diagnosis, and Restoration of
Function What We Have Learned In 25 Years
William B. Kleinman
Technique Article
Reconstructive Hand Surgery for Scleroderma Joint 1107
Contractures
Ananthila Anandacoomarasamy, Helen Englert, Nicholas Manolios, and
Stuart Kirkham
Instructional Course Lecture

Arthrodesis of the Interphalangeal Joints With Headless 1113
Compression Screws
Stephen J. Leibovic
Letters to the Editor

Locked Intramedullary Nailing of Metacarpal Fractures 1120
Secondary to Gunshot Wounds
Costochondral Autograft as a Salvage Procedure After 1121
Failed Trapeziectomy in Trapeziometacarpal Osteoarthritis
Book Review
Fractures of the Hand and Wrist 1122
Items
American Society for Surgery of the Hand 1123
International Federation of Societies for Surgery of the Hand 1139
Instructions to Authors
The Journal of Hand Surgery, British and European Volume,
Table of Contents, June 2007
A19
A41
This journal is indexed in Index Medicus/MEDLINE and Current Contents Clinical Medicine
THE JOURNAL OF
HAND
SURGERY
INSTRUCTIONS TO AUTHORS
General information: The Journal of Hand Surgery publishes
original, peer-reviewed articles related to the diagnosis, treatment,
and pathophysiology of diseases and conditions of the upper extremity; these include both clinical and basic science studies, along
with case reports. Special features include Clinical Perspective and
History of Hand Surgery articles, Comprehensive Review manuscripts, and Surgical Technique articles, which provide an overview of hand surgery, technical aspects of surgery, and current
controversial topics. Accepted articles will be published in the print
journal as well as on the online version. Supplemental materials
accepted with the manuscript that may not be published in the
print version due to space or medium constraints may appear in the
online version; this material will be cited in the article text and the
reader will be directed to the online journal. Authors lacking facility with English syntax should seek the appropriate editorial assistance prior to submitting their manuscript. Material requiring major
editorial work will be returned without review. The Journal uses
anonymous peer-review in evaluating manuscripts for publication.
Authors must submit new manuscripts electronically via the
Elsevier Editorial System (EES) at http://ees.elsevier.com/jhs. Send
all other correspondence relating to the editorial management of
The Journal of Hand Surgery to the appropriate editor:
Clinical, Basic Science, and Clinical Perspective articles:
Paul R. Manske, MD
Editor, The Journal of Hand Surgery
Washington University Medical Center
Department of Orthopaedic Surgery
One Barnes Plaza, Suite 11300 WP
St Louis, MO 63110
Phone: (314) 747-2537
Fax:
(314) 747-2801
e-mail: jhs@wudosis.wustl.edu
Review/Technique articles:
Arnold-Peter C. Weiss, MD
Editor, The Journal of Hand Surgery
University Orthopedics
2 Dudley Street, Suite 200
Providence, RI 02905
Phone: (401) 457-1522
Fax: (401) 831-5874
e-mail: arnold-peter_weiss@brown.edu
Editorial policies: Statements and opinions expressed in the
articles and communications are those of the author(s) and not
necessarily those of the Editor or Publisher, and the Editor and
Publisher disclaim any responsibility or liability for such material. Neither the Editor nor the Publisher guarantees, warrants, or
endorses any product or service advertised in this publication;
nor do they guarantee any claims made by the manufacturer of
such product or service.
Copyright: All material published in the Journal of Hand Surgery is
vested in the American Society for Surgery of the Hand. In accordance with the Copyright Act of 1976, the corresponding author of
each manuscript will be requested to complete a copyright assignment form on acceptance of the manuscript. When submitting a
paper the author(s) must make a full statement to the Editor about
all submissions and previous reports that might be regarded as prior
or duplicate publication of the same or very similar work. Copies of
such material should be included with the submitted paper to help
the Editor decide how to deal with the matter.
Financial interest: The Conflict of Interest Statement, to be completed by all listed authors, will be requested on acceptance of the
manuscript. This form requires disclosure from each author indicating that (a) no financial conflict of interest exists with any
commercial entity whose products are described, reviewed, evaluated or compared in the manuscript, except for that disclosed
under Acknowledgments or (b) a potential conflict of interest
exists with one or more commercial entities whose products are
described, reviewed, evaluated or compared in the manuscript.
The Editor, in his discretion, may disclose an actual or potential
conflict in a footnote to an article or elect not to publish the article
on the basis of such conflict. Otherwise, the information provided
will remain confidential.
Ethics: Do not use patients names, initials, or hospital numbers.
Articles emanating from a particular institution are assumed to be
submitted with the approval of the requisite authority.
Human subjects: Articles involving research conducted in human
subjects must include a statement in the Materials and Methods
section indicating approval by the institutional review board and
noting that informed consent, as well as any necessary HIPPA
consent, was obtained from each patient. For reports of research
using human subjects, provide assurance that (a) necessary and
appropriate consent was obtained from each patient and (b) the
study protocol conformed to the ethical guidelines of the 1975
Declaration of Helsinki as reflected in a prior approval by the
appropriate institutional review committee. Individual patients
should be referred to by number, not by initials.
Animal experimentation: Manuscripts reporting animal experiments must include a statement in the Materials and Methods
The Journal of Hand Surgery
A19
section that animal care complied with the guidelines of the

authors institution and the National Institutes of Health and any
national law on the care and use of laboratory animals.
Permissions: Lengthy direct quotations, tables, or illustrations taken
from copyrighted material must be accompanied by written permission for their use from the original copyright holder. The permission is presented as a text or table footnote or as an addition to
the figure legend and must provide complete information as to
source. Photographs of identifiable persons must be accompanied
by a signed release that indicates informed consent.
Units of measurement: Use Syste`me International (SI) measurements.
Statistics: Describe statistical methods with enough detail to
enable a knowledgeable reader with access to the original data
to verify the reported results. When possible, quantify findings
and present them with appropriate indicators of measurement
error or uncertainty (such as confidence intervals). Avoid sole
reliance on statistical hypothesis testing, such as the use of p values,
which fails to convey important quantitative information.
Abbreviations: If it is used more than three times in the abstract
and more than four times in the article, spell out an abbreviation
the first time it is used, followed by the abbreviation in parentheses. The abbreviation may be used subsequently. Spell out all
abbreviations at the beginning of a sentence.
Drug names: Use generic names. The proprietary name may be
mentioned in parentheses, along with the name and location of
the manufacturer.
Revised manuscripts: Revised manuscripts must be submitted according to the form of the original submission (ie, electronic or paper). Each
revised manuscript must be accompanied by a cover letter that addresses each of the reviewers comments. This letter should detail
point by point which changes have been made and should give
reasons for those recommended changes that have not been made.
The manuscript number must appear in the cover letter.
Organization of the manuscript: Failure to comply with these
requirements will delay consideration of the manuscript. The
manuscript may be returned to the author(s) for appropriate
modifications in format prior to review. Since the authors identities are withheld from the reviewers, authors names are to
appear only on the title page. Submit the manuscript as two
separate files, for blinding purposes: the title page as one file and
the manuscript, without author names, as another file. Type both
single-sided and double-spaced throughout (including the list of
references, tables and figure legends) formatted with 2.5 cm (1
inch) margins all around, with each line numbered in the margin. Arrange the manuscript as follows: title page, abstract, body
of the manuscript, references, figure legends, tables, figures. The
body of the manuscript should be organized as follows. For
peer-reviewed scientific studies, include introductory paragraphs, materials and methods, results, and discussion. For review, technique, symposia, and instructional course lecture
manuscripts, include historical background (introduction), specific subheadings for the main body of the text, and a summary.
Number all pages consecutively beginning with the title page.
Title page: The title should be a concise and informative description of the study.
Include the authors highest academic degrees, along with the
department and institution where the work was done. Designate
one author as the correspondent and supply his or her complete
mailing address, telephone number, fax number, and e-mail adA20
dress. If the name or address for reprint requests is different, so state.

Every person listed as an author should have materially participated
in the design, execution, and analysis of the study and should verify
the accuracy of the entire manuscript before its submission. Lesser
contributions may be noted in an acknowledgment section at the
end of the manuscript. Manuscripts should have no more than 6
authors; a greater number requires justification.
On the same page, provide a short title of 45 characters or fewer
to be used as a running head, and list in alphabetical order key
words for coding and indexing purposes.
Include disclaimers, if any, and a brief acknowledgment of grants
or other assistance.
Abstract: For peer-reviewed scientific studies, submit a structured abstract of approximately 300 words divided into 4 sections: Purpose, Methods, Results, and Conclusions. For review,
technique, symposia, and instructional course lecture manuscripts, submit a 1-paragraph brief description of the manuscript
contents. Do not include footnotes or references in the abstract.
Type the abstract on a separate page.
For clinical scientific studies, include the Type of Study and
the Level of Evidence (see table) at the end of the abstract.
Text: The text of the article should be typed double spaced and
divided into sections as outlined above (in Organization of the
manuscript). Articles may use section subheadings to clarify the
content. Do not repeat in the text all data in the tables or
illustrations; emphasize or summarize only important observations. Articles should not include a Conclusion section. Avoid
claiming priority of findings.
References: Authors are responsible for verifying the accuracy and
completeness of references. References should not be merely a
listing of the results of a computerized literature search but should
have been read by the author and deemed pertinent to the manuscript. Type references double-spaced on pages separate from the
text and number them consecutively by the order of their citation in
the text. Identify references in the text by arabic numerals within
parentheses. List all authors when 7 or fewer; when more, list the
first 6 and add et al. Use abbreviations of journal titles conforming
to Index Medicus. Include complete opening and closing page
numbers for each citation. For each reference from a journal
source, include on a separate page the reference citation directly
from the PubMed/Medline Web site. If a reference source is not
yet published but has been accepted for publication, include the
source in the reference list and submit the letter of acceptance
along with the manuscript.
Do not cite meeting abstracts, personal communications, or
unpublished material (including oral and poster presentations, correspondence club letters, and manuscripts not yet accepted for
publication) in the reference list. If critical to the manuscripts
message, this material may be identified in the text within parentheses. Please note the following examples of reference style.
Journal article
Saunders RA, Frederick HA, Hontas RB. The Sauve-Kapandji
procedure: a salvage operation for the distal radioulnar joint. J
Hand Surg 1991;16A:1125-1129.
Book
Taleisnik J. The wrist. New York: Churchill Livingstone, 1985:
25-32.
Chapter in edited book
Bowers WH. The distal radioulnar joint. In: Green DP, ed.
Operative hand surgery. 3rd ed. New York: Churchill Livingstone, 1993:973-1020.
Tables: Double-space table data with the table number and title
centered above the table and with explanatory notes below the
table. Each table should be on a separate page. Do not duplicate
material in tables with material in the text or figures. Overly lengthy
tables of pertinent data and appendices may be submitted for
publication as supplementary material in the online journal.
Figure legends: Number the figures with Arabic numerals in the
order mentioned in the text. Provide sufficient explanation to
render the figure intelligible without reference to the text. Define
all symbols and all abbreviations not yet spelled out in the text.
For any copyrighted material, indicate that permission has been
obtained (see Permissions, above). Figure legends should be
typed consecutively on a page separate from the body of the
manuscript.
Figures: Use professionally produced arrows or other markers
placed directly on the figure to identify important features. Do
not write on the illustrations. Crop figures as necessary to emphasize the subject material. Indicate the top of the photograph
so that the orientation is apparent. All figures and illustrations
should be oriented so the distal component (eg, fingertip) is at
the top. The authors names or names of institutions should not
appear anywhere on the figures. The figure title and caption
material appear in the legend, not on the figure. Figure numbers
must correspond with the order in which figures are presented in
the text. Do not include photographs or x-rays of normal findings. Line art should be submitted with no gradations of shading,
as they will not reproduce well. Use cross-hatching or patterns
where shading is necessary. Artwork and photographs submitted
in color will be reproduced in full color in the Journal at no
charge to the authors. If photographs of persons are used, either
the subjects must not be identifiable or the persons written
permission to use the photograph must accompany the manuscript. If a figure has been published, acknowledge the original
source and submit written permission from the original copyright holder to reproduce the material (see Permissions, above).
Special features:
Comprehensive Review Articles will be considered for publication
if they provide a broad-based and complete review and discussion
of a particular clinical topic related to hand and upper extremity
surgery. The manuscript should include a brief Abstract, Historical
Background, Discussion of Various Treatment Options, and Summary. Questions related to whether the Journal would be interested
in a particular topic may be directed to Dr. Weiss.
Instructional Course Lecture Articles will be considered for publication if they have been presented at a major hand surgery
meeting and provide a comprehensive overview of a particular
area and/or surgical technique related to hand and upper extremity
surgery. The article should be specifically written for the Journal. A
series of ICL articles which had been presented together are of
particular interest. Interested authors should contact Dr. Weiss if
further details are required. These articles should include a brief
Abstract, Introduction, Discussion and/or Surgical Treatment Options, and Summary.
Case Reports will be considered for publication if they contribute to an understanding of the diagnosis, treatment, and/or
pathogenesis of diseases or conditions related to hand surgery.
The manuscripts should include a brief Abstract, Introduction,
Case Report, and Discussion.
Book Reviewsbooks will be reviewed depending on their interest
and value to the reader. Two copies of the book to be reviewed
should be sent to the Editor. No books will be returned.
Letters to the Editors of sufficient interest will be considered for

publication. Letters should be typed double-spaced and limited to
500 words. All Letters to the Editors should be directed to Dr.
Manske. Letters related to previously published articles must be
submitted within 3 months of the articles publication; a copy of the
letter will be sent to the author(s) for an invited response. Letters are
subject to correct editorial policies.
Electronic illustration submission: Figures must be submitted in
electronic format. Please see instructions for submitting digital
art at http://ees.elsevier.com/jhs. Images should be provided in
EPS or TIF format. Graphics software such as Photoshop and
Illustrator, not presentation software such as PowerPoint, CorelDraw, or Harvard Graphics, should be used to create the art.
Color images must be CMYK, at least 300 DPI. Gray scale
images should be at least 300 DPI. Combinations of gray scale
and line art should be at least 1200 DPI. Line art (black and
white or color) should be at least 1200 DPI and accompanied by
a proof.
Authors are responsible for applying for permission for both
print and electronic rights for all borrowed materials and are
responsible for paying any fees related to the applications of
these permissions.
Video Clips: Journal of Hand Surgery invites authors to submit
video clips to be published on the Journals Web site. All video
clips will be subject to peer review.
Copyright of all video clips published on the Journals Web
site will be held by the American Society for Surgery of the
Hand. Each coauthor of a video clip must sign a form expressly
transferring copyright in the event that the video clip is published on the Journals Web site. Copies of this form may be
obtained from the Editorial Office. Peer review will not proceed
until signed copyright releases have been received by the Editorial Office.
Video clips must be no more than 1 minute in length and no
more than 5 MB in file size. Videos must be submitted in either
QuickTime or MPEG format. Authors who want their videos
accessible in a streaming format must also provide either a
single SureStream file or 3 uniquely named single-rate clips
(28.8, 56, T1) with a SMIL file to list the bandwidth choices.
Video clips must meet production quality standards to be published on the Web without modifications or editing by the
Editorial Office. The Journal can accept only video submissions
that meet the Journals formatting and image quality requirements. Authors will be notified if there are any problems with
submitted files and asked to resubmit modified files. Image
editing and correct formatting are the authors responsibility.
Authors may supply a still shot from the clip to be posted next to
the link to the video clip on the Web site; guidelines for figure
submission should be followed.
Video clips accepted for publication will be posted to the
Journals Web site, www.jhandsurg.org, in both nonstreaming
QuickTime or MPEG format for optimal image quality and in a
streaming video format for those who prefer faster downloading.
For detailed instructions on capturing, digitizing, saving, and
submitting videos by ftp, please see the Guidelines for Video
Submission available at www.jhandsurg.org.
Technical considerations for submitting manuscripts: Manuscripts must be submitted electronically at http://ees.elsevier.
com/jhs. All components of the manuscript must appear within
a single electronic file: references, figure legends, and tables
A21
must appear at the end of the manuscript. Please refrain from

using end notes as references, or using automatic list numbering
because these features are lost in conversion: simply type the
reference number in parentheses in the text and type the reference list. Formatting, such as Greek letters, italics, super- and
subscripts, may be used: the coding scheme for such elements
must be consistent throughout.
Business communications should be addressed to Elsevier Science, 360 Park Avenue South, New York, NY 10010. Attn:
Publisher, Journal of Hand Surgery, (212) 633-3958.
Reprints: A reprint order form is sent with page proofs to the
author. Reprints should be ordered prior to publication because those
ordered after publication are significantly more expensive. Questions
regarding reprints should be directed to the Publisher (212) 633-3958.
LEVELS OF EVIDENCE FOR PRIMARY RESEARCH QUESTION
Types of Clinical Studies

Therapeutic
Studies
Investigating the
Results of
Treatment
Level I
Level II
Prognostic Studies
Investigating the Effect
of a Patient
Characteristic on the
Outcome of Disease
Diagnostic
Studies
Investigating a
Diagnostic
Test
Economic and Decision

AnalysesDeveloping an
Economic or Decision
Model
High-quality randomized
controlled trial with
statistically significant
difference or no statistically
significant difference but
narrow confidence intervals
High-quality prospective
study4 (all patients were
enrolled at the same point in
their disease with 80%
follow-up of enrolled
patients)
Testing of previously developed

diagnostic criteria in series of
consecutive patients (with
universally applied reference
gold standard)
Sensible costs and

alternatives; values
obtained from many
studies; multiway
sensitivity analyses
Systematic review2 of Level-I

randomized controlled trials
(studies were homogenous)

studies

studies
Systematic review2 of
Level-I studies
Lesser-quality randomized
controlled trial (eg, 80%
follow-up, no blinding, or
improper randomization)
Retrospective6 study
Development of diagnostic
criteria on basis of consecutive
patients (with universally applied
reference gold standard)
Sensible costs and

alternatives; values
obtained from limited
studies; multiway
sensitivity analyses
Prospective4 comparative
study5
Untreated controls from a

randomized controlled trial
Systematic review2 of Level-II

studies
Level-II studies

studies or Level-I studies with
inconsistent results
Lesser-quality prospective
study (eg, patients enrolled at
different points in their
disease or 80% follow-up)
studies
Case-control study7
Level III
Case-control study7
Retrospective6 comparative
study5
Study of nonconsecutive patients Analyses based on limited

(without consistently applied
alternatives and costs;
reference gold standard)
poor estimates
Systematic review2 of Level-III
studies
Level-III studies
No sensitivity analyses
Level IV
Systematic review2 of Level-III

studies
Case series8
Case series
Case-control study
Level V
Expert opinion
Expert opinion
Expert opinion
Poor reference standard

Expert opinion
1. A complete assessment of the quality of individual studies requires critical appraisal of all aspects of the study design.
2. A combination of results from two or more prior studies.
3. Studies provided consistent results.
4. Study was started before the first patient enrolled.
5. Patients treated one way (eg, with cemented hip arthroplasty) compared with patients treated another way (eg, with cementless hip
arthroplasty) at the same institution.
6. Study was started after the first patient enrolled.
7. Patients identified for the study on the basis of their outcome (eg, failed total hip arthroplasty), called cases, are compared with those who
did not have the outcome (eg, had a successful total hip arthroplasty), called controls.
8. Patients treated one way with no comparison group of patients treated another way.
This chart was adapted from material published by the Centre for Evidence-Based Medicine, Oxford, UK. For more information, please see
www.cebm.net.
SUBMIT MANUSCRIPTS ONLINE AT HTTP://EES.ELSEVIER.COM/JHS

A22
History at Our Fingertips

As hand surgeons, we like to consider ourselves at the cutting edge of our clinical specialty.
It is also important, however, that we are rooted in the past. The names of our professional
ancestors are liberally sprinkled throughout our scientific writings and our conversations with
colleagues. Although we are well versed in proper anatomic nomenclature and precise scientific
terminology, we readily substitute the familiar names of those who described conditions we
treat, procedures we perform, and structures we dissect: Darrach resection, de Quervains
tendonitis, Bunnell suture, Colles fracture, Kaplans line, Guyons canal, Madelungs deformity, and so forth. This rich eponymous history is most apparent in the instruments
we use in the operating room (OP). They bear the names of those who came before us and serve
as daily reminders of our surgical heritage. In this issue, Clifton and Roy Meals present
historical profiles of these individuals. As the authors note, they represent an assortment of
practitioners from various surgical walks of life who have no established connections with
each other . . . except for their daily coming together in the OR instrument tray. Their diverse
backgrounds mirror the development of hand surgery itself, which originated from several
different surgical disciplines. We are indebted to the Meals for these fascinating narratives that
keep us in touch with our historical past.
Paul R. Manske, MD
Copyright 2007 by the American Society for Surgery of the Hand
doi:10.1016/j.jhsa.2007.06.001
941
A History of Surgery in the Instrument

Tray: Eponymous Tools Used
in Hand Surgery
Clifton G. Meals, BA, Roy A. Meals, MD
From the David Geffen School of Medicine at UCLA, Los Angeles, CA.
Hand surgery emerged as a specialty after World War II, and early hand surgeons borrowed
tools from established fields of the time. These tools remain in common use today, and many
are identified by the names of the men who created them. Because these men did not
specialize in surgery of the hand, their history remains obscure to modern surgeons who do.
We have investigated the history of eponymous instruments developed before 1945 and used
widely today in American hand surgery. Reflection on these eponymous instruments reveals
the rich and diverse history of hand surgery. (J Hand Surg 2007;32A:942953. Copyright
2007 by the American Society for Surgery of the Hand.)
Key words: Eponyms, hand surgery, history, instruments.
anavel, Tinel, Dupuytren, Kienbockthese

mens descriptions of signs and diseases are
central to the understanding of hand surgery.
Their original publications are readily identifiable
through the hand surgery literature, and their eponyms are lasting legacies of their observations and
contributions. When hand surgeons use such names
to describe findings and conditions, they pay respect
to and perpetuate the memory of these surgical pioneers.
The same is true in the operating room when one
asks for an Adson, a Kocher, or a Ragnell. The
difference here, however, is that many of the eponymous surgical instruments used by hand surgeons
today were devised before the specialty of hand
surgery came about, and thus awareness of these
innovators and their writings is dim at best.
Consider the era when these surgeons were working. Beginning with the demonstration of general
anesthesia in 1846, the following decades saw the
adoption of aseptic techniques, the production of
nickel-plated and then stainless steel instruments, the
discovery and application of x-rays, the advent of
blood transfusions, and the development of antibiotics. Surgery flourished. For the first time, surgery
exceeded its age-old role of letting blood, draining
pus, and performing amputations. Surgeons ranged
from scalp to toe devising surgical exposures and
complex operations along with the instruments to
942
perform them efficiently. As they became specialized

in neurosurgery, gynecology, otolaryngology, urology, plastic surgery, and orthopedic surgery, so did
their instruments.
The specialty of hand surgery emerged later, during and immediately after World War II, when the
efficiency of caring for limb-injured soldiers by a
single, regional specialist was recognized and effected. At that time, surgeons specializing in hand
surgery borrowed freely from the instrument trays of
established specialties and generally found tools suitable for their needs. Hand surgeons use many of
these tools today, and the eponyms of the instruments are common operating room parlance, yet
typically little is known of the innovators responsible for their development. Among them is a Nobel
laureate, a Pulitzer Prize winner, a fundamentalist
Bible thumper, a political dissident, and a mountaineer. Recognizing these pioneers pays respect to their
contributions, allows for reflection on the rich and
colorful history of surgery, and raises awareness that
hand surgery arose from multiple surgical disciplines.
Seminal Events
Liston
The oldest eponymous instrument in the modern
hand tray is the heavy bone shear designed by Robert
Liston (1794 1847)1 (Fig. 1). Liston was a Scottish
Meals and Meals / A History of Surgery in the Instrument Tray
Figure 1. Instruments related to seminal events in surgery.

From left to right: Liston bone cutter, Lister scissors, Bard
Parker scalpel handle, Beaver scalpel handle.
surgeon, famous for his physicality in the operating

room. A colleague wrote, [Liston] would amputate
the thigh, single-handed, compress the artery with his
left hand, using no tourniquet, and do all the cutting
and sawing with his right.2 Liston was brusque with
his fellow surgeons, and it was said that he was a
teacher more by what he did than what he said.1
Likely a result of his uncompromising attitude, Liston engaged in no shortage of professional quarrels.
He was officially dismissed from the Royal Infirmary
for 5 years before returning to prominence.3
Standing over an operating table in 1846 (with a
young Joseph Lister in the audience), Liston announced, We are going to try a Yankee dodge today,
gentlemen, for making men insensible. The patient
was anesthetized with ether, as had been demonstrated by William Morton several weeks earlier in
Boston. The amputation took 28 seconds and was a
painless success. Liston remarked, This Yankee
dodge, gentlemen, beats Mesmerism hollow.3
Lister
Whereas Liston was confrontational and quick, Joseph Lister (18271912) was methodical and shy.4
His innovations spoke for themselves.
Lister introduced absorbable catgut sutures and the
bandage scissors that bear his name5 (Fig. 1). Lister
also held a unique opinion of the fact that simple
fractures healed while compound ones festered and
led to death.4 Unlike his contemporaries, he believed
that suppuration was the work of invisible and ubiquitous microorganisms. Lister based his conviction
on Pasteurs and Kochs germ theory of disease and
was undoubtedly encouraged by his fathers work as
a microscopist.5
Using carboxylic acid, Lister performed the first
successful antiseptic treatment of an open tibia frac-
943
ture on an 11-year-old girl in 1866. A year later he

wrote, Since the antiseptic treatment . . . wounds and
abscesses no longer poison the atmosphere . . . my
wards, though in other respects under precisely the
same conditions as before, have completely changed
their character, so that during the last nine months
not a single case of pyaemia, hospital gangrene or
erysipelas has occurred in them.5
Listers success actually marginalized him among
his British colleagues. His practices were strongly
opposed by senior physicians, and London medical
students found that if they subscribed to Listers
antiseptic practices, which became known as Listerism, they failed their exams. Lister received early
recognition outside of Britain, however, as empirical
proof mounted; and after years of patience, the doctor finally received the recognition he deserved.
Baron Lord Lister rests in Westminster Abby in the
company of only 2 other physicians, John Hunter and
Thomas Willis.4
Parker and Beaver
Antisepsis in mind, scalpels with ornately decorated
handles hewn from wood, tortoiseshell, and ivory
gave way in the late 19th century to instruments with
a single piece of metal forming both handle and
blade. These tools still dulled quickly, however, and
an operating room technician was required to sharpen
and sterilize instruments as they were used.
Surgeons turned to the disposable, double-edged
razor blades invented by King Gillette in 1905, either
placing them in specialized holders or gripping them
with hemostats. Doctors weighed this convenience
against the difficulty of working with straight-edged
blades. A physician voiced this frustration at home,
and by 1915, the physicians nephew, Morgan Parker
(18921976), had secured a patent for a 2-piece
scalpel, allowing spent blades to be quickly replaced6
(Fig. 1).
Later in the same year, Parker demonstrated his
invention at the American College of Surgeons
where he received encouragement to pursue largescale manufacture. Parker opened the telephone directory and contacted the first name listed under
medical suppliers. The Bard-Parker company was
born.7
Rudolph Beaver (1886 1968) led a division of
Gillette that tried to parlay the razor blade into a
successful surgical knife. He later formed his own
company and patented a variety of highly specialized
blades and handles8 11 (Fig. 1).
944
The Journal of Hand Surgery / Vol. 32A No. 7 September 2007
Instruments from Thoracic,

Abdominal, and Urologic Surgery
Halsted
After his appointment as chief of surgery at the
newly opened Johns Hopkins Hospital, William
Stewart Halsted (18521922) became a preeminent
figure in American surgery, establishing the surgical
resident system and training Harvey Cushing among
numerous other influential surgeons.12,13
His enthusiasm for Listerism was the logical adjunct to his meticulous surgical technique, delicate
handling of tissues, and rigorous hemostasis. Halsteds development of precision mosquito hemostats
likely contributed to his reputation as a bloodless
surgeon (Fig. 2). Halsted wrote:
The value of artery clamps is not likely to be overestimated. They determine methods and effect results impossible without them. They tranquilize the operator. In
a wound that is perfectly dry, and in tissues never
permitted to become even stained by blood, the operator
unperturbed may work for hours without fatigue. The
confidence gradually acquired from masterfulness in
controlling hemorrhage gives to the surgeon the calm
which is so essential for clear thinking and orderly
procedure at the operating table.13
Like Lister, Halsted immersed his instruments in

carbolic acid. Repeatedly retrieving the instruments
from this sterilizing solution created a rash on the
hands of his chief nurse (and future bride), and Halsted approached the Goodyear Rubber Company.
Together they developed a protective glove that did
not compromise the operators dexterity. Nurses and
surgeons alike began wearing rubber gloves, intending at first to protect themselves and realizing later
the benefit afforded to patients. In addition to rubber
gloves, Halsted introduced white surgical gowns, caps,
and masks to the American operating room.12,13
Figure 2. Instruments from thoracic, abdominal, and urologic surgery. From left to right: Weitlaner retractor, Kocher
hemostat, Halsted hemostat, Crile hemostat, Mayo scissors,
Allis forceps, Senn retractor.
Halsteds career was marked also by personal idiosyncrasies (he appears to have been obsessed with
the proper construction of his shoes), self-experimentation with cocaine, and subsequent periods of addiction. Despite his dependencies and quirks, Halsted
remains rightfully acknowledged as the father of
American surgery.12,13
Crile
During his internship at University Hospital in
Cleveland, George Washington Crile (1864 1943)
witnessed a colleague succumb to shock after bilateral high thigh amputations. The cold sweaty skin
and the pallor, the fading pulse, the high pulse rate,
the sunken eyes, and dilated pupils fixed themselves
in my memory. This event motivated Crile to become one of the first, true American physiologists.
He popularized blood pressure monitoring in the
operating room, performed blood transfusions before
there was an awareness of blood typing (adverse
reactions were likely masked by the recipients moribund condition), and, in frequent collaboration with
Harvey Cushing, contributed greatly to the understanding of physiologic duress.14
In addition to the Crile hemostats in use today,
Crile developed a pneumatic rubber suit to fight
hypotension in neurosurgery patients (Fig. 2). Modern fighter pilots wear G-suits descended from
Criles design.15
Crile was a prolific surgeon, once performing 32
thyroidectomies in a single day, adding to a career
total of roughly 25,000. Crile was renowned for his
academic productivity, publishing 24 books and
more than 400 other works. On the foundation of this
abundant work, Crile helped to found the American
College of Surgeons and the Cleveland Clinic.14
Halsted and Crile hemostats vary principally in
size. Two distinctly different ratcheting forceps feature specialized tips for secure grasping of tissues.
Allis
Oscar Huntington Allis (1826 1921) developed his
long-armed forceps as a means of manipulating intestinal ends during anastomosis. Allis described this
new means of closure: When near the end of the
approximation I have found toothed forceps, with
serrations on the edge, convenient for turning in the
mucous edges, adjusting the serous, and holding
them approximated until sutured . . . I do not care to
recommend these instruments simply as aids in intestinal work . . . the forceps with lateral serrations
are often most convenient as hemostats, or as search-
ers for tendons which are retracted in wounds of the

hands and feet16 (Fig. 2).
Kocher
In his textbook, Swiss surgeon Theodor Kocher
(18411917) described use of his toothed clamp.
The bleeding points, however, should not be immediately ligated for it is dangerous to expose ligature
material to the risk of contamination during the
whole course of the operation. The newer pattern of
artery forceps has the advantage that even when a
large number are necessary, they are conveniently
hung out of the way and do not interfere with the
surgeons movements. The variety we use is very
light, is easily applied and takes a firm grip in the
case of dense tissues17 (Fig. 2).
Kocher was an image of deliberation in a time
when surgeons were valued for their speed and flair.
This philosophy and Kochers strict adherence to
asepsis were responsible for his patients remarkably
low mortality rate, especially after thyroid surgery,
which Kocher helped pioneer.18 For this advance and
for his related experimental work, Kocher received
the Nobel Prize, the first of only 2 surgeons to be so
recognized.4
Mayo
The Mayo Clinic bears the name of the remarkable
family at its core. William Worral Mayo (1819
1911) moved to Rochester, Minnesota, in 1863 to
establish a medical practice.19 He believed his sons
should be handy, and raised them in close association with his own work. The older son, William
James Mayo (18611939), said of his upbringing,
We were reared in medicine as a farmer boy is
reared in farming.20 The younger son, Charles
Horace Mayo (18651939), became his fathers
anesthetist as a teenager.21 That both brothers became doctors was inevitable.
In 1883, a tornado swept through Rochester. The
Mayos and a local convent of nuns collaborated to
treat victims. Subsequently, the nuns began raising
money for a local hospital and asked the Mayos to be
the hospitals surgeons. The Mayo Clinic was born.22
Today, the Mayo Clinics international importance
and reputation for innovation are broadly recognized.
Although the Mayos added much to surgical technique and medical education, attribution of accomplishments to either brother individually is difficult.
A friend of the brothers was known to have said, I
believe if Dr. Will were elected president of the
United States he would accept the office in the name
of his brother and himself.23
945
We were unable to determine specifically which

brother developed the Mayo scissors (Fig. 2). Conversely, the Weitlaner (often bastardized as Wheatlander) self-retaining retractor is clearly attributed to
Austrian Franz Weitlaner, although we have been
unable to discover anything about this surgical innovator24 (Fig. 2).
Senn
Another wound retractor came from Nicholas Senn
(1844 1908), a PhD, a doctor of laws, and a physician. In his book, Senn depicts a number of instruments of his devising. His well-known retractor illustrates the section on rupture of the urethra25 (Fig. 2).
Senn worked in Milwaukee and later in Chicago,
was an early proponent of Listerism, and initially
advocated operating under a fog of carbolic acid
spray.26 Senn realized that aseptic technique demanded an evolution in surgical instruments, saying,
All attempts at ornamentation have been abandoned . . . . The modern surgical instruments are made
as plain and smooth as possible.27
Senn earned his reputation as an innovator with
hard work. His partner said, When [Senn] was in his
creative moods, he was quite unaware to whether he
had changed his clothing, eaten or sleptsuch was
the thrust of his hunger for work and for the satisfaction of his ambitions.28 Senn did much of his
experimenting in a private basement laboratory. His
tinkering reportedly coincided with a disappearance
of cats from the neighborhood.26
Senn was additionally an intrepid world traveler,
numbering Siberia, Tahiti, Africa, the Arctic, and
South America among his destinations. Senn died of
myocarditis, probably exacerbated by a Andean
climb to 16,000 feet.26,28
Instruments from Gynecology

Kelly
Howard Atwood Kelly (18531943), a contemporary
of Halsted at Johns Hopkins, established gynecology
as a specialty and is known for the widely used Kelly
clamp29 (Fig. 3). Kelly was a staunch advocate of
social reform, a fundamentalist Christian, and an avid
collector of reptiles.29 Kellys son, also a physician,
described the family library:
. . . its walls solidly lined with books and thronged with
curiosities from all quarters of the globe . . . never the
same place from one day to the next, its stacks of newly
arrived books on tables and chairs, its mysterious halfemptied packages of strange and foreign objects, and its
never-ending stream of visitors served to make of it a
living organism.29
946
Gelpi
Maurice Gelpi (18831939) worked in New Orleans
and served a term as president of the Orleans Parish
Medical Society.33 In the societys journal, he described his self-retaining perineal retractor:
Figure 3. Instruments from gynecology. From left to right:

Kelly clamp, Gelpi retractor.
Actual living organismsspecimens in Kellys

herpetological collectionwere allowed to slither
freely through the home.
Kelly aided women in all walks of life. He supported womens suffrage and worked to expose prejudice against women entering the medical profession.29,30 Outside the hospital, he rented a home
where prostitutes were invited to live, rent-free,
while they sought more seemly employment.29
In doing good, Kelly surely earned his share of
detractors. He worked with the Lords Day Alliance,
who patrolled Maryland on Sundays to discourage
citizens from alcohol, games of chance, sports, and
movies. Kellys most famous critic was the irreverent H.L. Mencken, and the two friends sparred
publicly.31 In 1932, Mencken delivered this shot:
He happens to be a man I have long known, and in
every respect save the theological, greatly respected.
But in that theological respect, it seems to me he is so
plainly a menace to the peace and dignity of this town
that what he believes should be made known to every
one, that the people may be alert to his aberrations and
keep a curb upon his public influence. If he had his way,
it must be obvious, life here would be almost impossible to civilized men.29
Anyone who realizes the tremendous number of surgical instruments in existence today can appreciate the
fact that it takes considerable moral courage to increase
that appalling number even by a single unit . . . . It is
only fair to state here that of those who tried the
instrument our own chief, Dr. Clark, was the one by
whom the instrument was commented upon with the
least enthusiasm. Yet, in spite of this, both Dr. Kostmayer and myself have managed at different times to
slip in the retractor in Dr. Clarks instruments and have
seen him use it apparently with perfect ease and success
on several occasions. Certainly the finished perineorrhaphy didnt seem to have suffered from the use of the
retractor . . . . simplicity was an argument in its favor . . . . The old instruments require more assistance
because they must be held in place and are often unsatisfactory because they may be pulled out of the
mucous membrane by a careless assistant34 (Fig. 3).
Instruments from Otolaryngology

Freer
Otto Tiger Freer (18571932) of Chicago developed
a series of elevators, flat knives with rounded, paddle-shaped . . . blades, to correct deflections of the
nasal septum (Fig. 4). His published descriptions
make frequent reference to dental instruments from
which some of his tools were derived. In discussion,
he explained that . . . many of them are little knives
whose blades have varying shapes and attachments to
their shanks to enable them to move in the proper
directions on the field of operation, which all must
Mencken could tease, but he could not detract

from the ability of a man who, through tireless will,
erected an enduring place in medicine for the care of
women. Kelly could also, it appears, return Menckens sarcasm. He once wrote to the editor of the
Baltimore Sun:
If Mr. Mencken would put a photospectroheliograph on
his ramshackle tergiversating cerebrum, I think he
would discover that he was something of a synentognathous physoclistous levirate leventine belone with
perissodactyl affinities . . . . This is my most gentle
response to his last innocuous blusterings.32
Figure 4. Instruments from otolaryngology. From top to bottom: Yankauer suction tip, Lempert rongeurs, Metzenbaum
scissors, Freer periosteal elevator.
reach by the same narrow passagethe nostril . . . .

Their simple form and cheapness make up for their
number.35
Yankauer
Sydney Yankauer (18721932) designed an anesthesia mask on which ether or chloroform could be
dripped.36 The device became standard issue in operating rooms across the globe, although now entirely obsolete. He held a patent for a scalpel with a
detachable blade, which was displaced by Parkers
1915 revelation.37 Conversely, his tonsil suction tip,
originally made of steel, remains an operating room
staple today in both metallic and plastic forms (Fig. 4).
Yankauer contributed greatly to the specialization of
bronchoscopy and devised also a pharyngeal speculum, an electrode for fulgurating the larynx, and a
radium needle for the esophagoscope.38
Lempert
As a youth in Czarist Russia, Julius Lempert (1890
1968) was pursued by police after participating in an
antigovernment demonstration. The experience concerned Lemperts father, and the family soon emigrated to the United States.39 If Lemperts young life
was shaped by a rebellious act, his professional life
was no different.40
In New York, Lempert developed a 1-stage fenestration operation for the treatment of otosclerotic
hearing loss.39 To this end, he developed special
tools, including a small single-action rongeur, the
prototypes of which Lempert rejected in great number41 (Fig. 4). His perfectionism and the obstreperous
defense he mounted around his controversial procedure so perturbed the elders of his profession that he
was barred from the American Otologic Society.42
Lemperts operation, nonetheless, earned him wide
recognition. Lempert is said to have completed more
than 3,000 fenestrations while simultaneously teaching his technique to a worldwide audience of surgeons.42
Perhaps Lempert became uncomfortable as his detractors dwindled. He took on the American government, failing to properly pay the tax on his sizeable
income.42 When stapectomy superseded his fenestration technique in popularity, Lempert refused to
budge. To his death, Lempert resisted the inversion
of the status quo he had personally established. Although this quirk of Dr. Lemperts was lamented by
some of his students, it is possible that precisely by
rejecting stapectomy, Lempert did much to improve
it.42 Faced with a formidable opponent, the doctors
947
that championed this new technique had to be at their

best.
Metzenbaum
Myron Firth Metzenbaum (1876 1944) practiced
otolaryngology in Cleveland, and he used the scissors he invented to remove inflamed tonsils (Fig. 4).
Metzenbaum owed much to his famous mentor,
George Crile. When Crile solicited freshman medical
students to live in the hospital and assist in the wards,
Metzenbaum volunteered. The two bonded, Metzenbaum boarded in the hospital and eventually accepted
an invitation to join Crile in practice. Metzenbaum
later became interested in facial reconstructive surgery and was a founding member of the American
Board of Plastic Surgery.
Metzenbaum organized the Cleveland ambulance
service, which was probably the first service of its
kind in the United States. He also developed a dropether anesthesia technique after experiencing difficulty sedating hard-drinking Cleveland factory workers. This refinement persisted in American medicine
for 50 years.43
Instruments from Ophthalmology

Stevens
George Thomas Stevens (18331921), a veteran of
the Civil War, was a professor of physiology and
diseases of the eye in New York.44 In 1889, he
described his technique for correcting strabismus.
Stevens recommended an intratendinous shortening
and tenorrhaphy of the underactive rectus muscle
using scissors of his design, which are depicted and
described in his article: Their cutting properties are
perfect, they are strong enough to overcome every
resistance, while their points are so very slender that
they work beneath the conjunctiva through an extremely small opening (Fig. 5). Previous techniques
for correcting strabismus involved advancing the attachment of the rectus muscle on the eyeball.
Stevens2 observation on those procedures were insightful and remain pertinent. It is probable that
nearly every surgeon who has had any considerable
experience in these operations has learned that each
of the methods which have from time to time been in
vogue promises much better than it fulfills, and that
failure in producing any exact results is the rule
rather than the exception.45
Barraquer
After he observed a leech lift a small stone from the
bottom of an aquarium, Ignacio Barraquer (1884
948
Figure 5. Instruments from ophthalmology. From lower left

to upper right: Barraquer needle holder (locking version),
Stevens scissors, Castroviejo needle holder (nonlocking
version).
1965) wondered if he could, in similar fashion, remove a human cataract with suction. The technique
he developed was a resounding success. Dr. Barraquers numerous other innovations, including the
needle holder used today in microsurgery, live
onas does the Barraquer family, which, for more
than 100 years, has contributed greatly to medicine
and health care (Fig. 5).
Barraquers father inducted him into the Spanish
fraternity of ophthalmology, the elder Barraquer having
laid much of the groundwork in this field himself.
Ignacios 2 sons and several grandchildren have continued the family tradition in ophthalmology. Ignacio
Barraquer established an eye clinic in Barcelona, a rich
source of continuing education, and a hospital where
nearly half of all beds were reserved for indigent patients. Barraquer designed a small zoo at the center of
his institute where he indulged his love of animals. Dr.
Barraquer donated his own eyes to science.46
Castroviejo
Ramon Castroviejo (1904 1987) also grew up in
Spain but spent most of his professional years in New
York. He tenaciously perfected the keratoplasty,
turning a dangerous and rarely attempted procedure into one that now affords sight to countless
patients. Castroviejos success had much to do with
his enthusiasm for the microscope and his willingness to develop smaller instruments.47 Whereas others persisted in the use of clamps with inferior results, Castroviejo took advantage of newly developed
atraumatic suture needles and used locking micro
needle holders (nearly identical to those designed by
Barraquer except for the shape of the handle) to
suture transplanted tissue to the cornea48 (Fig. 5). He
developed roughly 60 instruments in his career. Castroviejo said of the creative process, Maybe when
you are operating, your subconscious speculates on
what you should do and with which instruments. But
it is when your are driving the car that you suddenly
realize and discover what you must do the next
time. He laughed about his passion for innovation,
When I die, the epitaph on my tomb should be
wake me at 8 a.m.47
Castroviejo once arrived at a Chicago conference
where he learned that many of the attendees intended
to dismiss his claims. Castroviejo hired a rail coach
and delivered dozens of his New York postoperative
patients to Chicago. He supplied each patient with a
pen light so that detractors were treated to lucid
justification for Castroviejos keratoplasty.48 When
he died, Castroviejo also donated his eyes.47
Instruments from Neurosurgery

Cushing
Harvey Williams Cushing (1869 1939) is among the
most famous neurosurgeons of all time. He trained
under Halsted and served at Johns Hopkins before
moving to Harvard. His fame is attributable to his
lust for hard work, his pioneering use of electrocautery, and his progressive understanding of human
physiology. Before Cushing, neurosurgery was probably as threatening to patients as the tumors they
developed. Cushing did much to change this.
Cushing traveled and collaborated greatly. Early in
his career, he developed, with Ernest Codman, the
ether chart so that surgeons and anesthesiologists
could cooperate in the monitoring of operative patients. Trips through Europe inspired him to turn
exclusively to the brain and particularly the pituitary
gland. His work with physicist William Bovie was
instrumental in developing a cautery tool, so important in neurosurgery, that would both sever and seal
bleeding vessels. Cushing introduced and popularized use of the mercury sphygmomanometer in
America.15 He was also the first to use a pneumatic
tourniquet in surgery, first on the upper limb and then
placing it around the head and inflating it with a
bicycle pump to quell scalp bleeding during craniotomies.49 The pneumatic tourniquet sees daily use in
extremity surgery, as does the Cushing vein retractor
(Fig. 6).
Cushing was gentle with his patients and a fierce
taskmaster among his co-workers. He frequently demanded that his operating room staff join him in
surgery on Saturday and Sunday mornings. Offhours, Cushing was a tireless author and talented
949
instruments, creating the so-called Adson-Brown pickups (Fig. 7). During his career at Washington University in St. Louis, Brown coined the expression
split-thickness skin graft as he revolutionized burn
management and coverage of large wounds.53 Attribution is not clear, however, especially in light of
Alfred Adsons close association with George Elgie
Brown (18851935), who worked in the Department
of Medicine at Mayo and shared Adsons interest in
the effects of sympathectomy on blood flow in a
number of disease states.54,55
Figure 6. Instruments from neurosurgery. From top to bottom: Cushing retractor, Adson forceps, Frazier suction tip.
illustrator. In 1926, he won the Pulitzer Prize for his

biography of his next-door Baltimore neighbor, Sir
William Osler.50
Frazier
In the early 1900s, Cushing and Charles Frazier
(1870 1936) were considered to be the only 2 American doctors with extensive neurosurgical practices.
At the University of Pennsylvania (where Alfred
Adson was enrolled), Frazier was known both as a
charitable tutor and a fearsome pedagogue. A Frazier Club sprang up at Penn welcoming those students who had been ignominiously dismissed from
Pop Fraziers operating room. In fits of insomnia,
Frazier was known to call his interns at 3 in the
morning and demand that they pre-op. His legacy
includes the suction tip that bears his name51 (Fig. 6).
Adson
Early in his career at the Mayo Clinic, Alfred Washington Adson (18871951) removed a brain tumor
while three elite European military surgeons looked
on. One later reported seeing a high school boy
perform neurosurgery. Thirty years old at the time,
Adsons appearance belied his ambition. The interest
these esteemed men showed for this particular case
inspired Adson, and from this point, he focused increasingly on surgery of the brain and nerves. The
pick-ups with small biting teeth attributed to him
were well adapted to precise lifting and removing of
neural tissue52 (Fig. 6).
Webster
The Webster needle holder has, at times, been attributed to different individuals (Fig. 7). It is sometimes
credited to a pioneer plastic surgeon, Jerome Pierce
Webster (1888 1974). He was long-time chief of
plastic surgery at Columbia and may have been complicit in the foggy history of this instrument.56
George Van OLinda Webster (19111988), no apparent relation, trained under Jerome Webster. The
younger Webster then served in the U.S. Navy during
World War II and was a founding member of the
American Society for Surgery of the Hand as well as
its 13th president.57 In an article published by
George Webster, he describes a small needle holder
with absolutely flat jaws and slightly rounded edges
that furnish a secure, noncutting grip even with the
finest silk suture.58 It is said that, as a resident,
George Webster had the idea to grind off the teeth of
a needle holder. Jerome Webster told him that doing
so would be grounds for dismissal. In later years, the
elder Webster apparently declined to rectify erroneous attribution of the ground-down instrument to
himself (E. Zook, MD, personal communication,
2006).
Instruments from Plastic Surgery

Brown
James Barrett Brown (1899 1971) probably added
two rows of multiple, fine teeth to one of Adsons
Figure 7. Instruments from plastic surgery. From lower left to

upper right: Ragnell retractor, Webster needle holder, AdsonBrown forceps.
950
Ragnell
Allan Ragnell (19021982) trained in plastic surgery
in England before returning to Sweden. In 1944, he
became chief of plastic surgery at Serafimerlasarettet
in Stockholm, the first department of its kind in
Scandinavia. His influence spread, and he later directed a larger plastic surgery and burn department at
the Karolinska Institute. At the time of his death, 7 of
8 Swedish plastic surgery centers were chaired by
either first- or second-generation descendents of Ragnells training.59 His elegant, double-ended retractor
has further memorialized his name (Fig. 7).
Instruments from Orthopedic Surgery

Lambotte
Albin Lambotte (1866 1955), a Belgian, was as
much a mechanic as he was a surgeon. His workshop
aptitude allowed him to make his own surgical instruments, including the eponymous osteotomes
and many of the bone clamps in contemporary
use60 (Fig. 8). Lambotte also crafted fine violins.61
He coined the word osteosynthesis to describe
bone healing and greatly advanced the techniques of
both internal and external fracture fixation. His 1913
book describes the intramedullary nailing of scaphoid fractures and the pinning of Bennetts fractures.5,62 Lambotte fostered mechanical ability
among his students at the University of Antwerp
insisting that they saw straight, drill, and thread precisely.60 Lambotte himself was a model of precision,
operating with white gloves and using a meticulous
Figure 8. Instruments from orthopedic surgery. From upper

left to lower right: Bennett retractor, Cobb periosteal elevator,
Key periosteal elevator, Lambotte osteotome, Hohmann retractor, Verbrugge clamp.
no touch technique that he believed protected his

patients from infection.5
Verbrugge
Jean Verbrugge (1896 1964) was an esteemed student of Lambotte and his eventual academic successor. Verbrugge learned the science of fracture fixation from its progenitor and applied it with success,
especially in children.63 The Verbrugge bone clamp
is a contemporary symbol of his seminal work (Fig. 8).
In marked distinction to domineering surgeons of his
time, Verbrugge was known for his friendliness. He
was heard to address a clumsy operative assistant
saying, I do not think that I would have set about it
that way.64
Key
Medical school for John Albert Key (1890 1955)
was interrupted by financial hardship. After a stint as
an anatomy instructor, he returned to his studies and
volunteered in World War I France as a studentintern. While in Europe, Key received notice of his
medical school graduation by cablegram. On return
to the United States, he taught applied physiology at
Harvard and then orthopedics at the University of
Maryland before settling in St. Louis where he became research director at the Shriners Hospital.
Later, Key became chief of orthopedics at Washington University. Key is remembered not only for his
periosteal elevator but also for his generous support
of students and trainees of limited means65 (Fig. 8).
Hohmann
Georg Hohmann (1880 1970) started and finished
his career in Munich with intervening years in Frankfurt. He founded an early rehabilitation center for
veterans disabled in World War I and summarized
his vast experience gained from the treatment of war
injuries in his thesis on pseudarthroses and flail
joints. He described transferring the pectoralis major
to restore elbow flexion as early as 1917 and devised
popular techniques for treating tennis elbow, bunion,
and hammer toe.66,67 The bone retractor bearing his
name is an additional memorial (Fig. 8).
Bennett
A bone retractor similar to Hohmanns carries the
name of Irish surgeon Edward Hallaran Bennett
(18371907) who is perhaps best remembered for the
eponymous thumb metacarpal fracture (Fig. 8). Bennett contributed greatly to the Irish body of medical
knowledge and became president of the Royal College of Surgeons of Ireland in 1884. Bennett ex-
panded the Trinity College collection of fracture and

dislocation specimens and was a supreme authority
on fracture care through the end of the 19th century.68 With his interest in bones and joints, it was
natural that Bennett embraced the nascent science of
radiology. He was also an early and outspoken proponent of Listerism and aseptic technique.69 Upon
his retirement, 200 former students gathered to show
their appreciation. He returned their gift as a fund
from which Trinity College began awarding outstanding students.68
Cobb
John Robert Cobbs (19031967) obituary in the
Journal of Bone and Joint Surgery reflects, He
loved carpenters tools and surgical instruments of
which he had great knowledge and mastery. He
arrived at orthopedics after serving on a merchant
steamer and paying his own way through Brown
University, where he earned a degree in English
literature. Cobb joined a dynamic staff at the Hospital for the Ruptured and Crippled in New York and
was charged with developing a scoliosis clinic.
Through strict adherence to evidence-based decision
making, Cobb made valuable contributions to the
understanding and treatment of scoliosis. He divided
scolioses by etiology and severity and taught that
only those idiopathic curves with proven progression
required surgery. All his patients, over whom he
watched as a father over his own children, must
have benefited from his careful and conservative
care.70 His periosteal elevators find current use in
hand surgery principally for exposing the pelvis during bone grafting procedures (Fig. 8).
Meyerding
Another spinal surgeon, Henry Meyerding (1884
1969), spent his career at the Mayo Clinic and wrote
widely on bone tumors, fractures, Volkmanns contracture, back pain, and spondylolisthesis.71,72 Several flat-bladed retractors bear his name, including a
finger retractor, presumably named for the closed
loop at its nonworking end that is easily retained by
a single hooked finger. Meyerding was a master of
clinical diagnosis and stressed its importance in this
conversation with his son in 1960:
Maybe it was a good thing you didnt become a physician after all . . . . Medical colleges are turning out
nothing but technicians now. All they know is what dial
to turn and what pill to prescribe. And do you know
with all this technology theyre always harping about,
theyve had at their disposal all along a single instrument that contained all the information about what was
951
going on in their patients bodies, right down to the

atomic level: the patients brains. But most of these
young fellows couldnt interview a patient to find out if
it was raining, let alone find out what was ailing him.72
Discussion
The lives of these 33 men span nearly 2 centuries
from Listons birth in 1794 to G.V. Websters death
in 1988. Collectively, the work of these men and
their contemporaries forms the basis of modern surgery. Many of these physicians will remain wellknown for generations to come. Other names, such as
Gelpi and Weitlaner, might be inclined to fade were
it not for the usefulness and ubiquity of their inventions.
Of the inventors described, patents were obtained
only by Parker and Beaver, neither of whom was a
doctor. The surgeon innovators apparently shared
their instruments altruistically, as they shared their
techniques and experiences. A search of the United
States Patent and Trademark Office, however, reveals a multitude of patented surgical instruments, all
more complex and task-specific than the ones described by us. Simplicity and wide applicability has
fostered the endurance of these legally unprotected
tools.
Every surgical specialty is represented in the hand
surgery instrument tray. These eponymous tools memorialize men exclusively, evidence of a now waning tradition in surgery. We were unable to attribute
the Edna towel clamp, which may have been named
for a female operating room nurse (T. Fischer, MD,
personal communication, 2006). Other unsolved
mysteries include the Backhaus towel clamp and the
Hiess (or Heiss) self-retaining retractor (sometimes
described as a mastoid retractor).
Undoubtedly, we have excluded many readers
favorite eponymous instruments, both those honoring
their mentors and others of particular personal usefulness. Before accurate attribution becomes impossible, we encourage others to document the origins of
such tools, including all that have been developed
since 1945. In doing so, we maintain a vital connection with the great heritage of surgical craft.
Received for publication March 16, 2007; accepted in revised form May
8, 2007.
No benefits in any form have been received or will be received from
a commercial party related directly or indirectly to the subject of this
article.
Corresponding author: Roy A. Meals, MD, 100 UCLA Medical Plaza,
#305, Los Angeles, CA 90024; e-mail: rmeals@ucla.edu.
0363-5023/07/32A07-0001$32.00/0
doi:10.1016/j.jhsa.2007.05.007
952
References
1. Bailey H, Bishop WJ. Notable names in medicine and surgery. 2nd ed. London: H.K. Lewis & Co. Ltd., 1946:5153.
2. Mc Dowell F. Commentary by the editor. Plast Reconstr
Surg 1971;48:168 169.
3. ODowd MJ, Philipp EE. The history of obstetrics and
gynecology. New York, London: The Parthenon Publishing
Group, 1994:83 84.
4. Firkin BG, Whitworth JA. Dictionary of medical eponyms.
1st ed. New York, London: The Parthenon Publishing
Group, 1987:242243.
5. LeVay D. The history of orthopaedics. Carnforth, UK: Parthenon, 1990:105108.
6. Parker M. Detachable blade knife. United States Patent
1,139,769. Issued May 18, 1915. Available at www.uspto.
gov.
7. Kirkup J. Who made what? Bard-Parker scalpels. J Med
Biogr 2005;13:206.
8. Beaver R. Detachable blade surgical knife. United States
Patent 1,915,244. Issued June 20, 1933. Available at www.
uspto.gov.
9. Beaver R. Cutting instrument. United States Patent
2,242,936. Issued May 20, 1941. Available at www.uspto.
gov.
10. Beaver R. Knife. United States Patent 1,813,723. Issued July
7, 1931. Available at www.uspto.gov.
11. Ochsner J. The surgical knife. Bull Am Coll Surg 1999;84:
2737.
12. Jacek B. Man behind the enstrument (2) Halsted, William
Stewart (18521922). Br J Theatre Nursing 1994;4:25.
13. Crowe S. Halsted of Johns Hopkins. The man and his men.
Springfield, IL: Charles C. Thomas, 1957:4756, 181.
14. Hermann RE. George Washington Crile (1864 1943).
J Med Biogr 1994;2:78 83.
15. Crile G, ed. Grace Crile. George Crile. An autobiography.
Philidelphia: Lippincott, 1947:146.
16. Yap LH, Ahmad T. Allis forceps: notes on the inventor. Br J
Plast Surg 2001;54:561.
17. Kocher T. Textbook of operative surgery. 3rd English ed.
from 5th German ed. New York: Macmillan, 1911:3334.
18. Kazi RA, Theodor E. Kocher (18411917): Nobel surgeon
of the last century. J Postgrad Med 2003;49:371372.
19. Spinner RJ, Al-Rodhan NRF, Piepgras DG. 100 years of
neurological surgery at the Mayo Clinic. Neurosurgery
2001;49:438 445.
20. Clapesattle H. The Mayo brothers. Minnesota Med 1954;37:
779.
21. Carson C. Charles Horace Mayo (18651939). Invest Urol
1981;18:313314.
22. Jacek B. Man behind the enstrument (11). Br J Theatre
Nursing 1995;4:25.
23. Clapesattle H. The doctors Mayo. 2nd ed. Minneapolis:
University of Minnesota Press, 1954:667 696.
24. Weitlaner F. Ein automatischer Wundspreizer [An automatic
wound spreader]. Wiener klinische Rundschau 1905;19:
114 115.
25. Senn N. Practical surgery for the general practitioner.
Philadelphia: WB Saunders, 1901:306.
26. Natvig P. Nicholas Senn of Milwaukee and Chicago, his
contributions to plastic surgery. Plast Reconstr Surg 1978;
61:167176.
27. Edmonson JM. American surgical instruments: an illustrated

history of their manufacture and a directory of instrument
makers to 1900. San Francisco: Norman Publishing, 1997:
319 320.
28. Hamilton S. The great master of abdominal sugery: Nicholas
Senn, MD, 1844 1908. Wisc Med J 1992;91:245249.
29. Speert H. Memorable medical mentors: III. Howard Atwood
Kelly (1858 1943). Obstet Gynecol Surv 2004;59:227233.
30. Allen PM. Howard Atwood Kelly (1858 1943): his life and
his enduring legacy. South Med J 1991;84:361368.
31. Karp LE. Doctors afield: the fundamentalist and the
iconoclast: H.A Kelly and H.L. Mencken. N Engl J Med
1975;292:297299.
32. Davis AW. Dr. Kelly of Hopkins: surgeon, scientist, Christian. Baltimore: The Johns Hopkins Press, 1959:194 195.
33. OPMS. OPMS Presidents 1978 2004. Available at www.
opms.org/about/presidents.htm.
34. Gelpi MJ. A new self-retaining perineal retractor. N Orl Med
Surg J 1913;66:182183.
35. Freer OT. The correction of deflections of the nasal septum
with a minimum of traumatism. JAMA 1902;38:636 642.
36. M.A.G. In memoriam: Dr. Sidney Yankauer. Laryngoscope
1932;42:818 820.
37. Yankauer S. Scalpel. United States Patent 445,237. Issued
January 27, 1891. Available at www.uspto.gov.
38. Anonymous. Sidney Yankauer. Ann Otol Rhinol Laryngol
1932;41:1278 1281.
39. Cawthorne T. Julius Lempert, a personal appreciation. Arch
Otolaryngol 1969;90:680 686.
40. Senturia BH. Julius Lempert. Arch Otolaryngol 1969;90:
702.
41. House HP. Memories. Arch Otolaryngol 1969;90:694 696.
42. Hoople GD. Personal recollections of Julius Lempert. Arch
Otolaryngol 1969;90:690 693.
43. Metson R. Myron F. Metzenbaum, MD: innovative surgeon,
caring physician. Otolaryngol Head Neck Surg 1994;110:
477 481.
44. Anonymous. Deaths, George Thomas Stevens. JAMA 1921;
76:465.
45. Stevens GT. Tendon resection and tendon contraction for
shortening the recti muscles. N Y Med J 1889;49:345348.
46. Greene AK. Ignacio Barraquer (1884 1965) and the Barraquer family of ophtahlmologists. Can J Ophthalmol 2001;
36:5 6.
47. Murube J. Ramon Castroviejo centenary: a life dedicated to
corneal transplantation. Surv Ophthalmol 2005;50:215225.
48. Polack FM. Ramon Castroviejo 1904 1987. Cornea 2000;
19:593 602.
49. Cushing H. Pneumatic tourniquets: with special reference to
their use in craniotomies. Med News 1904;84:577580.
50. Thompson EH. Harvey Cushing, surgeon, author, artist.
New York: Schuman, 1950:114.
51. Boockvar JA, Virella A, Kotapka M, Flamm ES, Grady MS.
Neurosurgery at the University of Pennsylvania Medical
Center. Neurosurgery 2000;46:12231228.
52. MacCarty CS. Alfred Washington Adson. Surg Neurol
1977;7:5354.
53. Cannon B. James Barrett Brown. Ann Plastic Surg 1981;7:
79 89.
54. Shepherd J, Vanhoutte P. Local modulation of adrenergic
neurotransmission. Circulation 1981;64:655 666.

55. Nerve Congress. Time. Monday, Auguest 12, 1935. Available at
www.time.com/time/magazine/article/0,9171,771796,00.html.
56. Tanzer RC. Obituary: Jerome Pierce Webster, M.D. Plast
Reconstr Surg 1975;55:728 731.
57. Gurdin MM. Obituary: George Van OLinda Webster, M.D.
19111988. Plast Reconstr Surg 1989;84:860 861.
58. Webster GV. A new needle holder for plastic surgery. U S
Naval Med Bull 1948;41:1434 1439.
59. Keh BJ. In memoriam: Allan Ragnell. Scand J Plast Reconstr Surg 1983;17:161.
60. Anonymous. Albin Lambotte 1866 1955. J Bone Joint Surg
1956;38A:464 466.
62. Lambotte A. Chirurgie: operatoire des fractures. Paris: Masson and Cie, 1913:350.
953
64. E.V.E. Jean Verbrugge, 1896 1964. J Bone Joint Surg 1964;
46A:924 926.
65. Abbot LC. John Albert Key 1890 1955. J Bone Joint Surg
1956;38A:453 463.
67. Guentz E. Georg Hohman zum 70 Geburtstag. Archiv fuer
orthopaedische und unfall Chirurgie 1949 1951;44:133
134.
68. Anonymous. In memoriam: Edward Hallaran Bennett, M.D.,
M.Ch.. Dublin J Med Sci 1907;124:78 80.
69. Anonymous. Obituary: Edward Hallaran Bennett, M.D.,
FRCSI. Br Med J 1907;1:15751578.
70. P.D.W. John Robert Cobb 19031967. J Bone Joint Surg
1968;50A:1074 1076.
71. Social Security Death Index Search Results. Available at
http://ssdi.rootsweb.com/cgi-bin/ssdi.cgi.
72. Wright IP. Who was Meyerding? Spine 2003;28:733735.
Results of Dorsal or Volar Plate

Fixation of AO Type C3 Distal
Radius Fractures: A Retrospective Study
Susanne Rein, MD, Hartmut Schikore, MD, Wolfgang Schneiders, MD,
Michael Amlang, MD, Hans Zwipp, PhD
From the Carl Gustav Carus University Hospital of Dresden, Department of Trauma and Reconstructive
Surgery, Dresden, Germany.
Purpose: The aim of this study was to define the outcome after dorsal or volar plating of
Association for Osteosynthesis (AO) type C3 distal radius fractures based on the fracture
morphology.
Methods: Twenty-nine patients with AO type C3 distal radius fractures were surgically
managed between 1996 and 2005. Group 1 (n 15) had volar plating. Group 2 (n 14) had
dorsal plating. Outcomes were evaluated at an average of 22 months after surgery. Statistical
analysis was performed using the Wilcoxon test and chi-square test.
Results: No significant differences were seen for the scores of Gartland and Werley, Castaing,
Stewart I and II, Green and OBrien, and Disability of the Arm, Shoulder and Hand between
the 2 groups. The visual and verbal pain analog scales did not show significant differences
between the 2 groups. Radiology analysis showed significant difference in comparison with
the contralateral side in terms of dorsopalmar inclination (3 3) and distal radioulnar joint
angle (98 8) for the patients in group 1, whereas there were no significant differences in
group 2. The development of radiographic post-traumatic arthritis was significant in both
groups. Significant functional differences were seen for flexion (45 15) and hand span
(20 cm 2) in group 1 as well as for extension (37 19), flexion (42 12), and radial
deviation (16 10) in group 2. We found more complications after dorsal plate osteosynthesis than after volar plate osteosynthesis.
Conclusions: This study shows satisfactory functional and subjective outcome results in both
groups. Group 1 had non-significant better functional results than group 2, whereas both
groups showed good to very good radiology results. (J Hand Surg 2007;32A:954 961.
Copyright 2007 by the American Society for Surgery of the Hand.)
Type of study/level of evidence: Therapeutic III.
Key words: DASH score, distal radius fracture, dorsal plate osteosynthesis, volar plate
osteosynthesis.
ractures of the distal radius are very common,

with an incidence ranging between 10% and
25% of all fractures.1 Anatomic reduction with
stable fixation is the treatment of choice for displaced, unstable fractures of the distal part of the
radius.2 The goals of treatment are to restore the
articular surface congruency and to restore the radial
height, radial inclination, and palmar tilt. The ideal
fixation method is one that can maintain a satisfactory reduction and allow early motion to avoid joint
stiffness and disuse atrophy.3 The dorsal approach is
954
usually favored in patients with excessive dorsal

comminution or displacement that are not amenable
to indirect reduction via a volar approach.1 The dorsal approach, however, is controversial because of
extensor tendon complications.2,4,5 Recently, the
treatment of volar fixation of dorsally unstable distal radius fractures with a fixed-angle plate was introduced to avoid extensor tendon complications of
dorsal plates.6,7 Functional outcome is related to the
quality of reduction.8 Failure to achieve proper orientation and correct length of the distal radius rela-
Rein et al / Fixation of AO Type C3 Distal Radius Fractures
955
Figure 1. Posteroanterior (A) and lateral (B) radiographs of a patient with a fracture of the distal part of the radius. Preoperative
CT scanning (C) shows AO type C3 distal radius fracture with volar fragmentation but without dorsal comminution. Postoperative
posteroanterior (D) and lateral (E) radiographs demonstrate nearly anatomic fixation of the fracture with a T-plate.
tive to the carpus and ulna can result in pain and

instability.8 The dorsal approach clearly affords excellent articular exposure, consistently allowing anatomic reductions of the joint surface.2
Thus far, few studies compared the volar and dorsal plate position in the surgical treatment of fractures of the distal radius including different kinds of
distal radius fractures according to the Association
for Osteosynthesis (AO) classification.1,4,9,10 Most
clinical studies only reported the outcome of either
volar or dorsal plate osteosynthesis.2,5,8 Therefore,
exact comparison of the outcome of intra-articular
comminuted fractures between volar and dorsal plate
osteosynthesis is not possible as, for example, type A
fractures are extra-articular. Letsch et al compared
the outcome of dorsal versus volar plate position of
fractures of the distal radius.1 This study revealed
better anatomic reduction and clinical outcome after
dorsal plate osteosynthesis.1 In contrast, the study of
Ruch and Papadonikolakis showed a higher rate of
volar collapse and worse functional results after dorsal plate osteosynthesis in comparison with volar
plate osteosynthesis.11
Because of this contradictory literature, the purpose of the current study was to evaluate the subjective, functional, and radiology outcome after dorsal
or volar plating of AO type C3 distal radius fractures
based on the fragment morphology of the fracture.
Materials and Methods

Twenty-nine patients with AO type C3 distal radius
fractures were surgically plated between 1996 and
2005 either from the volar approach (n 15) in
group 1 or the dorsal approach (n 14) in group 2.
Further details of the 2 groups are described in
Table 1. All patients were treated in the following
standardized algorithm: Our primary treatment of
AO type C3 distal radius fractures is the initial closed

reduction and, if necessary, stabilization of this fracture with an external fixation with or without ancillary K-wires on the day of injury. Preoperative computed tomography (CT) scanning of the wrist is
Table 1. The Different Parameters of the 2

Groups
Age at time of injury (years)

Age at time of follow-up evaluation
(years)
Time difference between injury and
follow-up evaluation
Gender (no.)
Women
Men
Handedness (no.)
Right-handed
Left-handed
Dominant hand
Nondominant hand
Cause of injury (no.)
Work-related
Accident
Leisure
Additional external fixation (no.)
Time of external fixation (weeks)
Time injury-definitive
osteosynthesis (days)
Type of plate (no.)
T-plate
Pi-plate
Locking plate
Immobilization (days)
Hardware removal (cases)
Systematic (cases)
Time plate osteosynthesis to
hardware removal (months)
Group 1
(n 15)
Group 2
(n 14)
54 14
45 12
59 15
50 13
4.4 3
4.6 3
8
7
4
10
7
8
7
8
10
4
8
6
1
1
13
7
.2 6
5
2
7
9
22
78
85
12
0
3
21 7
7
All
7
5
2
24 14
7
5
7 10
13 6
956
Figure 2. Posteroanterior (A) and lateral (B) radiographs of a patient with a dorsally displaced fracture of the distal part of the
radius. After closed reduction and temporary external fixation, the CT scanning (C and D) shows a dorsally displaced fragment,
which cannot be reduced anatomically via a volar approach. Dorsal open reduction and internal fixation with a Pi-plate was
performed. Postoperative posteroanterior (E) and lateral (F) radiographs demonstrate nearly anatomic fixation of the fracture with
a Pi-plate.
performed to analyze the fracture morphology exactly. Six to 8 days later, the definitive osteosynthesis
is performed. In our practice, the volar approach is
favored (Fig. 1). Dorsal plate osteosynthesis is only
used in cases of fractures with dorsal comminution in
combination with dorsal fracture dislocations, which
cannot be reduced anatomically and fixed securely
from a volar approach. Figure 2C shows a persistently dorsally displaced fragment after closed reduction and stabilization with an external fixation. All
other fracture morphologies of AO type C3 distal
radius fractures are plated from a volar approach. In this
study, the T-plate, Pi-plate, and locking plate were used
(all plates: Synthes, Umkirch, Germany). Postoperatively, the patients wore a splint for 3 weeks. Finger
mobilization exercises were started from the first
postoperative day. Wrist mobilization began from the
14th postoperative day. Removal of the plate depends on radiologically verified bony consolidation
of the fracture. Outcomes were evaluated at an average of 22 months (range, 19 years) after surgery by
making use of the Disability of the Arm, Shoulder
and Hand (DASH) questionnaire, visual and verbal
pain analog scales, Gartland and Werley, Castaing,
Stewart, et al as well as the Green and OBrien
scoring system.1216
Follow-up evaluation was performed by 1 of the
5 authors. Wrist range of motion was measured
with a standard goniometer and grip strength with
a dynamometer (Collin, Schreiber GmbH, Fridingen, Germany). X-ray analysis included determination of the radial height, radial inclination, articular step-off, ulnar variance, dorsopalmar
inclination, post-traumatic arthritis, and distal radioulnar joint angle in comparison with the contralateral side.1719 The post-traumatic arthritis
was analyzed using the classification of Knirk and

Jupiter, whereby grade 0 indicated no arthritis;
grade 1, joint space narrowing; grade 2, clear joint
space narrowing and grade 3, no joint space with
existence of pseudocysts and osteophytes.19 The
incidence of radiographic post-traumatic arthritis
will probably increase with time. Statistical analysis was performed by using the Wilcoxon test and
the chi-square test with a level of significance
taken at p .05.
Results
Scores
Based on the Gartland and Werley score, groups 1
and 2 had mean scores of 9 5 and 10 7 points,
respectively. For the Castaing score, groups 1 and 2
had scores of 7 4 and 9 5 points, respectively.
For the Stewart I score, groups 1 and 2 had mean
scores of 0.93 1 and 0.86 1 points, respectively,
and for the Stewart II score, groups 1 and 2 had mean
scores of 9 6 and 10 7 points, respectively. For
the Green and OBrien score, groups 1 and 2 had
scores of 72 14 or 60 23 points, respectively
(Fig. 3). The average DASH score was 14 14
points in group 1 and 17 15 points in group 2. The
visual pain analog scale was 1 point at rest in both
groups, 2 2 and 3 2 points at exertion, and
2 2 and 2 2 points after exertion for groups 1
and 2, respectively (Fig. 4). The verbal pain analog
scale was 0.2 0.4 and 0.4 0.5 points at rest, 1 1
and 2 1 points at exertion, and 1 1 and 1 1
points after exertion for groups 1 and 2, respectively
(Fig. 5). No statistically significant differences were
seen in all of the scores between the 2 groups.
Figure 3. Outcome by functional scores is reported, whereas

no statistically significant differences were seen between the
2 groups. G&W, Gartland and Werley score; G&B, Green
and OBrien score.
Radiology Results
Radiology analysis showed significant differences in
comparison with the contralateral side in terms of
dorsopalmar inclination (3 3, p .01) and distal
radioulnar joint angle (98 8, p .04) for group 1
(Table 2). No statistically significant difference in
radiology data was revealed for group 2. The development of post-traumatic arthritis was significant in
both groups (p .008 in group 1, p .025 in group
2). The radiographic post-traumatic arthritis showed
that nobody in group 1 and 3 patients in group 2 had
grade 0, 3 patients in group 1 and 4 patients in group
2 had grade 1, 9 patients in group 1 and 3 patients in
group 2 had grade 2, and 3 patients in group 1 and 4
patients in group 2 had grade 3.
957
Figure 5. The results of the verbal pain analog scale, whereas

0 no pain, 1 little pain, 2 median pain, 3 strong
pain, and 4 excruciating pain. There was no statistically
signficant difference between the 2 groups.
Functional Results
The right hand was injured in 7 cases in group 1 and
10 cases in group 2. When compared with the contralateral side, significant differences in the range of
motion were seen in flexion (45 15, p .03) and
hand span (20 cm 2, p .05) in group 1. In group
2, significant differences were seen in extension
(37 19, p .003), flexion (42 12, p .03),
and radial deviation (16 10, p .005) (Table 3).
No statistically significant differences were observed
in all parameters of the hand function between the 2
groups (Fig. 6).
Complications of both groups are reported in Figure 7. Four patients in group 1 and 2 patients in group
2 felt that hand function was similar to preinjury
level. Five patients in each of the groups felt that
their wrists were adversely sensitive to weather
changes.
Discussion
Figure 4. Results of the visual pain analog scale are shown,

whereas a value of 0 indicates pain-free and a value of 10
means excruciating pain. There was no statistically significant difference between the 2 groups.
The current study shows results similar to those

reported by other authors, who reported that most of
comminuted intra-articular AO type C fractures can
be managed successfully with open reduction and
internal fixation with plates.1 The position of the
plate is, however, controversial when discussed in
literature. It often is inserted as a buttress plate (ie, on
the side to where the distal fragment has been displaced).3 Some authors favor the volar fixation with
a fixed-angle plate for dorsally displaced distal radius
fractures to prevent extensor tendon complications.6,7
One reason for the variety of implants used was
the proliferation of plates available between 1996
and 2005. In this study, not all plates were removed
routinely at 1 year. Two patients in group 2 with
958
Table 2. The Mean of the Radiology Results
Group 1
Injured
Contralateral
Group 2
Injured
Contralateral
Dorsopalmar
Inclination ()
Radial Inclination ()
Radial Height (mm)
Ulnar Variance (mm)
Distal Radioulnar
Joint Angle ()
3 3*
85
22 6
22 4
14 3
12 2
23
11
98 8*
91 10
66
96
22 5
24 3
12 2
13 3
13
05
99 8
101 11
*Indicates statistically significant differences in comparison with the contralateral hand.
extensor tendon irritation had their plates removed

early. In these 2 cases, hardware removal was done 6
months after plate osteosynthesis. Both patients from
group 2 reported improvement in their symptoms
after hardware removal. To minimize extensor tendon irritations, we also recommend systematic plate
removal 4 to 6 months after dorsal plate osteosynthesis.20 Some patients had senile osteoporosis or
high surgical comorbidity, and in such cases the
implants were left in situ. There were no cases of
refracture after hardware removal.
A recent study reported unimpeded extensor tendon function after dorsal plating with a low-profile
plate in 59 patients, which could be because of the
thinness of the plate.21 In the literature, extensor
tendon rupture and wound infection have been described as complications after a dorsal approach.2224
We have not, however, seen either extensor tendon
rupture or wound infection after dorsal plate osteosynthesis in our cases. Only 2 of 14 patients had
extensor tendon irritations in group 2. Complex regional pain syndrome type I occurred in 1 case in the
current study. This particular patient was referred to
a pain service for management.
Median nerve compression was seen in 2 patients.
The symptoms disappeared after reduction in 1 case,
and carpal tunnel release was done during definitive
plate osteosynthesis in the other patient. Median
nerve compression as a postoperative complication
was noted in a polytraumatized patient, and carpal
tunnel release was performed 1 month after definitive
osteosynthesis.
Reasons for poor functional long-term outcome
could be related to secondary fragment displacement
and articular step-off. Intra-articular step-off over 2
mm causes a post-traumatic arthritis in 91% of all
cases.19 We had only 2 patients with a dorsal plate
osteosynthesis who had a 2-mm intra-articular stepoff at the time of the follow-up evaluation. We found
signs of post-traumatic arthritis in all cases in group
1 and 78% of cases in group 2, whereas 20% (group
1) and 28% (group 2) had grade 3 of the classification

of Knirk and Jupiter. In accordance with other authors, we found more complications after dorsal plate
osteosynthesis than after volar plate osteosynthesis.11
Therefore, we recommend a regular close follow-up
evaluation after dorsal plate osteosynthesis.
A reliable comparison with results from the literature
is difficult because of the lack of standardized and
reproducible methods for their evaluation. The large
number of different scoring systems illustrates this discrepancy.1315 Therefore, we analyzed our data with the
widely accepted scores of Gartland and Werley, Stewart, et al and Green and OBrien. We also used the
Castaing score, which is common in French-speaking
countries. The DASH questionnaire is a standardized
instrument that measures patients own perspective of
their upper-extremity disabilities.12 Felderhoff et al reported an average DASH score of 23 points involving
213 patients after dorsal and/or volar plate osteosynthesis of distal radius fractures.10 The mean DASH score
differs from 11 to 18 points after combined dorsal
and volar plate osteosynthesis in the literature.25,26 Rozental et al reported a mean DASH score of 15 points
after dorsal plate osteosynthesis of 28 patients.2 These
results are similar to ours. We had an average DASH
score of 14 14 points in group 1 and 17 15 points
in group 2. In contrast with that, a previous study
revealed a mean DASH score of 8 points after dorsal
Pi-plate osteosynthesis of AO type C3 distal radius
fractures in 29 patients.8 Our results of visual and verbal
pain analog scales indicate pain at exertion in both
groups, correlating with the results of other authors.10
In contrast with other authors, we have not found significant differences either in the Gartland and Werley
score or in the Stewart I score between the two
groups.1,11
Extension was diminished to 79% and 64% and
flexion to 82% and 79% in comparison with the
contralateral side in groups 1 and 2, respectively. The
average grip strength was 90% of the contralateral
side in group 1, which is more than the 74% to 77%
21 2
21 2
22 14
26 13
*Indicates statistically significant differences in comparison with the contralateral hand.
76 31
88 8
23 9
26 7
16 10*
23 10
42 12*
53 15
37 19*
57 15
959
Figure 6. The average percentage of the different hand functions in comparison with the contralateral side. *Indicates
statistically significant differences in comparison with the
contralateral hand. No statistical differences were noted between the 2 groups.
87 8
84 16
20 2*
21 2
18 14
20 13*
45 15*
55 18
Group 1
Injured
Contralateral
Group 2
Injured
Contralateral
50 25
63 23
18 10
22 9
21 8
26 9
79 22
73 35
76 26
83 19
Hand Span (cm)

Extension ()
Table 3. Functional Results
Flexion ()
Radial
Deviation ()
Ulnar
Deviation ()
Pronation ()
Supination ()
Grip Strength (kg)
after volar plate osteosynthesis reported in the literature.27,28 Grip strength of 70% to 85% of the contralateral side is reported after combined dorsal and
volar plate osteosynthesis.25,26 The average grip
strength was 85% of the contralateral side in group 2
in our study, which is in contrast with another study,
where the grip strength was 56% of the opposite side
after dorsal plate osteosynthesis.5
In the current study, 2 patients in group 1 and 3
patients in group 2 were polytraumatized, which could
be a reason for prolonged immobilization or delayed
secondary treatment. Patients with AO type C3 distal
radius fractures not only require surgical therapy with
the goal of anatomic reduction and stable osteosynthe-
Figure 7. Complications in both groups are demonstrated.

CRPS, complex regional pain syndrome type I.
960
sis but also an intensive early functional postoperative

physiotherapy, which could enable a better recovery of
the mobility in the wrist joint.
In principle, the volar plate osteosynthesis is sufficient for most of the fracture types, because the
dorsal comminution alone is not an indication for
dorsal plate osteosynthesis in our practice, which is
in contrast with other authors.1 Fractures with a single dorsally comminuted fragment can be stabilized
with a locking plate from a volar approach.28 Also,
displacement of fractures with a metaphyseal dorsally comminuted fragment can be minimized with
volar locking plates.28 There are, however, types of
fractures that cannot be treated from a volar approach. Indications for a dorsal approach are fractures with dorsal comminution in combination with
dorsal fracture dislocations, which cannot be reduced
anatomically and securely fixed from a volar approach (Fig. 2). As this study indicated, the subjective, functional, and radiology outcome is comparable between dorsal and volar plate osteosynthesis.
This study shows satisfactory functional and subjective results in both groups. Group 1 had non-significant better functional results than group 2 (Fig. 6),
whereas both groups showed good to very good x-ray
results. We have seen more complications after dorsal plate osteosynthesis. Therefore, we recommend a
careful follow-up evaluation and plate removal 4 to 6
months after dorsal plate osteosynthesis to minimize
complications. In addition, newer and smaller implants may decrease the rates of extensor irritations
after dorsal plate osteosynthesis.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
The authors thank the following individuals for their contributions to this
article: Thomas Albrecht, Carl Gustav Carus University of Dresden,
Department of Trauma and Reconstructive Surgery (Dresden, Germany)
and Martin Rein (Perth, Australia).
Received for publication May 11, 2006; accepted in revised form May
8, 2007.
article.
Corresponding author: Susanne Rein, MD, Carl Gustav Carus University of Dresden, Department of Trauma and Reconstructive Surgery,
Fetscherstr. 74, 01307 Dresden, Germany; e-mail: susanne.rein@web.de.
0363-5023/07/32A07-0002$32.00/0
doi:10.1016/j.jhsa.2007.05.008
References
1. Letsch R, Infanger M, Schmidt J, Kock HJ. Surgical treatment of fractures of the distal radius with plates: a comparison of palmar and dorsal plate position. Arch Orthop
Trauma Surg 2003;123:333339.
2. Rozental TD, Beredjiklian PK, Bozentka DJ. Functional
outcome and complications following two types of dorsal
15.
16.
17.
18.
19.
20.
21.
plating for unstable fractures of the distal part of the radius.

J Bone Joint Surg Am 2003;85A:1956 1960.
Leung F, Zhu L, Ho H, Lu WW, Chow SP. Palmar plate
fixation of AO type C2 fracture of distal radius using a
locking compression platea biomechanical study in a cadaveric model. J Hand Surg 2003;28B:263266.
Sanchez T, Jakubietz M, Jakubietz R, Mayer J, Beutel FK,
Grunert J. Complications after Pi Plate osteosynthesis. Plast
Reconstr Surg 2005;116(1):153158.
Ring D, Jupiter JB, Brennwald J, Buchler U, Hastings H II.
Prospective multicenter trial of a plate for dorsal fixation of
distal radius fractures. J Hand Surg 1997;22A:777784.
Orbay JL. The treatment of unstable distal radius fractures
with volar fixation. Hand Surg 2000;5(2):103112.
Orbay JL, Fernandez DL. Volar fixation for dorsally displaced fractures of the distal radius: a preliminary report.
J Hand Surg 2002;27A:205215.
Krukhaug Y, Hove LM. Experience with the AO Pi-plate for
displaced intra-articular fractures of the distal radius. Scand
J Plast Reconstr Surg Hand Surg 2004;38:293296.
Zettl RP, Ruchholtz S, Taeger G, Obertacke U, Nast-Kolb D.
Postoperative Morbiditt der operativ behandelten distalen Radiusextensionsfraktur. Eine Vergleichsstudie zwischen dorsaler
und volarer Plattenlage. Unfallchirurg 2001;104(8):710 715.
Felderhoff J, Wiemer P, Dronsella J, Weber U. Operative
Versorgung der distalen, instabilen Radiusfraktur mit der
dorsalen/palmaren Absttzplatte. Eine retrospektive Studie
unter Bercksichtigung des DASH-Score. Orthopde 1999;
28:853 863.
Ruch DS, Papadonikolakis A. Volar versus dorsal plating in
the management of intra-articular distal radius fractures.
J Hand Surg 2006;31A:9 16.
Germann G, Wind G, Harth A. Der DASH-Fragebogen-Ein
neues Instrument zur Beurteilung von Behandlungsergebnissen an der oberen Extremitt. Handchir Mikrochir Plast Chir
1999;31:149 152.
Gartland JJ Jr, Werley CW. Evaluation of healed Colles
fractures. J Bone Joint Surg 1951;33A:895907.
Castaing J. Les fractures rcents de lextrmit infrieure du
radius chez ladulte. Rev Chir Orthop Reparatrice Appar
Mot 1964;50:581 696.
Stewart HD, Innes AR, Burke FD. Functional cast-bracing
for Colles fractures. A comparison between cast-bracing
and conventional plaster casts. J Bone Joint Surg 1984;66B:
749 53.
Green DP, OBrien ET. Open reduction of carpal dislocations:
indications and operative techniques. J Hand Surg 1978;3A:
250 265.
Oestern HJ. Distale Radiusfrakturen. Teil I. Grundlagen und
konservative Therapie. Chirurg 1999;70:1180 1192.
Frstner H. Das distale Radio-Ulnar-Gelenk (DRU). Mor berlegung und chirurgisch-orthopdische
phologische U
Konsequenzen. Unfallchirurg 1987;90:512517.
Knirk JL, Jupiter JB. Intra-articular fractures of the distal
end of the radius in young adults. J Bone Joint Surg 1986;
68A:647 659.
Keller M, Steiger R. Osteosynthetische Versorgung distaler
Radiusextensionsfrakturen bei Frauen ber 60 Jahren mit der
dorsalen Radiusplatte (Pi-Platte). Handchir Mikrochir Plast
Chir 2006;38(2):82 89.
Simic PM, Robison J, Gardner MJ, Gelberman RH, Weiland
22.
23.
24.
25.
AJ, Boyer MI. Treatment of distal radius fractures with a

low-profile dorsal plating system: an outcomes assessment.
J Hand Surg 2006;31A:382386.
Cooney WP III, Dobyns JH, Linscheid RL. Complications of
Colles fractures. J Bone Joint Surg 1980;62A:613 619.
Oestern HJ. Distale Radiusfrakturen. Teil II. Operative
Therapien. Chirurg 1999;70:13811394.
Herron M, Faraj A, Craigen MA. Dorsal plating for displaced intra-articular fractures of the distal radius. Injury
2003;34:497502.
Espen D. Kombinierter palmarer und dorsaler Zugang fr
komplexe Frakturen des distalen Radius. Handchir Mikrochir Plast Chir 2003;35:2230.
961
26. Beyermann K, Prommersberger KJ. Die gleichzeitige Versorgung mehrfragmentrer distaler Radiusfrakturen von
einem palmaren und dorsalen Zugang. Handchir Mikrochir
Plast Chir 2000;32:404 410.
27. Sakhaii M, Groenewold U, Klonz A, Reilmann H. Ergebnisse nach palmarer Plattenosteosynthese mit der winkelstabilen T-Platte bei 100 distalen Radiusfrakturen. Eine prospektive Studie. Unfallchirurg 2003;106:272280.
28. Schtz M, Kolbeck S, Spranger A, Arndt-Kolbeck M, Haas
NP. Die winkelstabile palmare Plattenosteosynthese bei der
dorsal dislozierten distalen Radiusfraktur-Anwendung und
erste klinische Erfahrungen. Zentralbl Chir 2003;128:997
1002.
The Risk of Adverse Outcomes in ExtraArticular Distal Radius Fractures Is

Increased With Malalignment in Patients
of All Ages but Mitigated in Older Patients
Ruby Grewal, MD, Joy C. MacDermid, PhD
From the Division of Orthopedic Surgery, University of Western Ontario, Hand and Upper Limb Center, St.
Josephs Health Care, London, Ontario, Canada.
Purpose: The purpose of this study was to determine if malalignment after extra-articular
distal radius fractures influenced patient-reported pain and disability at 1 year and to
investigate how this relationship changes with age.
Methods: Two hundred sixteen subjects with extra-articular distal radius fractures were
followed. The influence of specific radiographic parameters and the overall acceptability of
alignment on Patient-Rated Wrist Evaluation (PRWE) and Disabilities of Arm, Shoulder and
Hand (DASH) scores were assessed. The relative risk (RR) of a poor outcome in the presence
of malalignment of the distal radius at various ages was calculated; the RR was then used to
calculate a number needed to harm.
Results: Malalignment of the distal radius was associated with higher reports of pain and
disability in patients 65 years of age. In patients aged 65 years, no isolated radiography
parameter was found to affect PRWE or DASH scores significanly; however, there was an
increased risk of a poor outcome in fractures with malalignment when compared with
fractures with acceptable alignment in all age groups. The RR of a poor outcome with
malalignment showed a decreasing trend with increasing age, with a significant reduction
after 65 years. In patients 65 years of age, 8 malaligned fractures would require correction
to prevent 1 poor outcome (based on DASH, or 9 based on PRWE); in younger patients, only
2 malaligned fractures would need correction to avoid 1 poor outcome (based on DASH, or
3 based on PRWE).
Conclusions: Patients 65 years of age showed no statistically significant relationship
between malalignment of the distal radius and PRWE or DASH scores when the radiography
parameters were examined in isolation and when clustered together. The relative risk data
demonstrates, however, that patients at all ages have a higher risk of a poor outcome with
malalignment of the distal radius when compared with those with acceptable alignment.
Therefore, we conclude that the relationship between outcome and alignment of the radius
should not be considered as an all-or-none phenomenon but rather considered as a decreasing gradient of risk, with the most significant change seen after patients reach 65 years of age.
(J Hand Surg 2007;32A:962970. Copyright 2007 by the American Society for Surgery of
the Hand.)
Type of study/level of evidence: Prognostic II.
Key words: Distal radius fractures, elderly, malalignment, patient-reported outcomes, risk.
here has been a great deal of literature published on the relationship of radiography variables and their influence on the final outcome
of distal radius fractures. Some authors report that
final functional outcome depends largely on the an-
T
962
atomic result,115 whereas others suggest that final

function is independent of residual deformity.16 25
These reports may differ for a variety of reasons.
Many of these trials used very small sample sizes (ie,
n 40),14 16,2224,26 27 and they also lacked a stan-
Grewal and MacDermid / Adverse Outcomes Risk in Extra-Articular Distal Radius Fractures
dardized method of reporting outcome. For example,

many of the functional outcomes reported have been
based on physician-reported measures of outcome (ie,
Gartland and Werley demerit system, or the Sarmiento
modification of this scale).1,13,16 17,2224,26 The reliability and validity of this scoring system has not been
reported, and some believe this system only assigns
poor outcomes to patients with severe problems.26
Also, traditional measures of impairment (ie, grip
strength, range of motion) do not necessarily reflect
patient-reported pain and disability.
Many trials also combined both intra- and extraarticular distal radius fractures.14,24,25,27 Although
both of these fractures involve the distal radius, they
can represent very different injuries. Most intra-articular fractures are higher energy injuries; they are
often associated with more soft tissue swelling and
comminution than the average fracture sustained by
falling from standing height. These factors are difficult to measure objectively and may also influence
final results. It is well documented that residual intraarticular step deformities lead to degenerative
change, and for this reason intra- and extra-articular
fractures should be studied as separate entities.26
Lastly, many published series in the literature report on results from a wide age range of patients (the
total age range covered by studies was at least 68
years [18 86 years] and was as large as 79 years
[1594 years] in some series).1,13,17,28 It would be
difficult to generalize results from these studies, as
most would agree the bone quality and physical
demands of an 18 year old are much different than
those of an 86 year old.
Some studies have focused on patients within a
narrower age range, usually evaluating elderly patients, who were defined as low demand or sedentary
subjects, with many medical comorbidities.22,24,29
These authors have shown that functional outcome is
not related to the final anatomic position of the
healed fracture in these patients.22,24,29 It would be
inaccurate to extrapolate the results of these lowdemand elderly cohorts22,24,29 to the average patient
over the age of 65 years, particularly as the elderly
population today is more active and has higher recreational and physical demands than in the past.
Despite the discrepancies in the literature, few
would debate that to ensure the best possible outcome in a young active patient, one should strive for
an anatomic reduction. Conversely, it is also accepted that elderly patients, with lower physical demands and multiple medical comorbidities, do not
require aggressive intervention to ensure an adequate
963
outcome. Clinical decision making becomes more

difficult when one is confronted with an older patient, who is still healthy and active, but presents with
some degree of residual deformity. The current literature offers little to guide decision making for both
the surgeon and the patient in this scenario.
We often assume that all older patients will tolerate residual pain, deformity, and functional impairments much better than their younger counterparts;
however, this has not been proved in a prospective
trial utilizing standardized patient-rated pain and disability scores. In addition, the relationship between
age and the risk of a poor outcome with a malaligned
distal radius fracture has not been previously reported in the literature, particularly for extra-articular
fractures.
The purpose of this study was to prospectively
follow a large cohort of patients with extra-articular
distal radius fractures and determine if the individual
radiographic parameters (dorsal angulation, radial inclination, and radial shortening) and the overall acceptability of alignment of the healed distal radius
fracture influence patient-reported pain and disability
at 1 year, as given by standardized patient-rated pain
and disability scores (Patient-Rated Wrist Evaluation
[PRWE] and Disabilities of Arm, Shoulder and Hand
[DASH]), and to determine if this relationship is the
same in different age groups. A secondary purpose
was to calculate the risk of a poor outcome for
patients with a malaligned distal radius fracture and
to examine the effect of age on this relationship.

This was a prospective observational study involving
a cohort of 216 patients with extra-articular distal
radius fractures. All patients were recruited from the
practices of 9 fellowship-trained hand surgeons at a
single tertiary care referral center from June 1997 to
June 2004. We excluded skeletally immature patients
and all intra-articular fractures. Subjects provided
informed consent for the use of their results in this
study. Testing was incorporated into scheduled clinic
visits. Data were collected at the initial visit (first
week after fracture), 3, 6, and 12 months. At each
visit, radiographs were obtained and patients were
requested to complete standardized patient-reported
outcome questionnaires (DASH and PRWE).
All patients received posterior-anterior and lateral
wrist x-rays before reduction, after reduction, and at
each follow-up appointment. All radiographs were
evaluated by 2 orthopedic surgeons. The following
parameters were assessed on each radiograph: dorsal
964
angulation, radial inclination, ulnar variance, and

presence of a fracture involving the ulnar styloid, the
ulnar head, or distal radioulnar joint (DRUJ).
An additional binary radiographic variable was
created to define the overall acceptability of alignment using guidelines set forth by the American
Society for Surgery of the Hand (ASSH) (eRadius
International Distal Radius Fracture Study Group:
ASSH Specialty Day at AAOS; available at http://
www.eradius.com). This variable was created because we believed that analysis of a single radiologic
variable in isolation was not as clinically useful as a
cluster of x-ray findings when dealing with a threedimensional structure such as the distal radius.
The overall alignment of the distal radius fracture
was designated as unacceptable if the dorsal angulation was 10, if the radial inclination was 15,
or if there was 3 mm of ulnar positive variance.
These guidelines were also used to evaluate the acceptability of each parameter in isolation.
We also gave consideration to using a different
cut-off point to define unacceptable alignment in
elderly patients as suggested by Kelly et al (5 mm
ulnar positive variance and 30 dorsal angulation).23 There were only 7 patients, however, who
were classified as having unacceptable alignment
based on these values, and if this criteria was used,
the study would have been underpowered.
The primary outcome, patient-rated pain and disability, was quantified with the DASH questionnaire
and the PRWE. Both the PRWE and the DASH were
chosen as outcome measures because they have been
shown to be valid, reliable, and highly responsive in
the distal radius fracture population.25,27,3133
Descriptive statistics were calculated for all
variables. All radiographic parameters (dorsal angulation, radial inclination, and radial shortening)
were analyzed as categorical variables (within acceptable parameters or not), and bivariate relationships were then determined between radiographic
parameters and outcomes (PRWE and DASH). The
overall acceptability of alignment was also analyzed in a similar fashion. For this portion of the
analysis, the sample was stratified into 2 groups by
age; those under 65 years and those 65 years and
older. We used 65 years as our cut-off point to
define elderly as this is usually the age of retirement, and we believe that previous definitions
(ranging between 50 and 60 years) have been too
young an age.22,24,25,27,30 To better understand the
practical significance of the relationship between
malalignment of the distal radius and the risk of a
poor outcome, we calculated the relative risk (RR)

the number needed to harm (NNH). The NNH
represents the number of people with malalignment needed before 1 person is harmed (ie, experiences a poor outcome based on DASH and
PRWE 20).
To more closely analyze the effect of age and
acceptability of alignment on final outcome, the sample was then stratified even further (age 40, 41 45,
46 50, 5155, 56 60, 61 65, 66 70, 7175, 76
years). We determined the relative risk of a poor
outcome (defined by PRWE and DASH 20) for
patients with a malaligned distal radius fracture at
each age group and also calculated the NNH.
The sample size required to answer the primary
question of this study, whether or not there was a
significant difference in 1-year PRWE and DASH
scores based on the acceptability of alignment, was
calculated based on a minimal clinically important
difference of 15 points (from the 100-point PRWE
score) and a standard deviation of 20. A sample of 56
subjects was necessary to ensure that 80% power was
achieved.
Results
A total of 297 patients met the inclusion criteria for
this study. Seventy-five patients were lost to follow-up evaluation and were excluded from analysis.
This left a total of 222 patients with extra-articular
distal radius fractures in our cohort. Of these 222
patients, 6 had x-rays that could not be located by our
radiology department. Thus, a total of 216 radiographs were available for review.
The mean age was 55.2 17.6 years. The range
was 18 89 years (71 years), and the median age was
58. The cohort consisted primarily of women; there
were 48 (22.2%) men and 168 (77.8%) women.
There were 19 (8.8%) subjects that had a third-party
claim surrounding their injury; either an existing or
pending insurance, legal or workers compensation
claim. There were 95 (42.8%) patients with associated ulnar styloid fractures, 6 (2.7%) with ulnar head
fractures, and 4 (1.8%) with involvement of the
DRUJ.
There were 73 patients aged 65 years in this
cohort. In this subgroup, the mean age was 74.4
5.8 years and the median age was 74.0 years. There
were 9 (12.3%) men and 64 (87.7%) women in this
subgroup. Forty (54.8%) patients had their dominant
hand affected. There were 2 (2.7%) subjects that
were involved in a third-party claim surrounding
their injury. Of the patients in this subgroup, 50.7%
965
Table 1. Radiography Parameters in Isolation (<65 Years)

PRWE
DASH
Acceptable*
Dorsal angulation
Radial inclination
Radial shortening
Acceptable*
Yes
No
Yes
No
16.6
15.9
16.1
23.2
46.2
29.0
.17
.04
.02
11.6
11.9
11.5
20.4
32.8
24.6
.06
.065
.002
*Acceptable dorsal angulation 10; unacceptable dorsal angulation 10. Acceptable radial inclination 15; unacceptable radial
inclination 15. Acceptable radial shortening 3 mm ulnar positive variance; unacceptable radial shortening 3 mm ulnar positive
variance.
(37/73) were healthy or had only 1 minor medical

problem (ie, isolated hypertension, hypercholesterolemia, or thyroid abnormalities). The remaining 36
(49.3%) had 1 or more of the following problems: 13
(17.8%) patients had heart disease, 23 (31.5%) had
arthritis, and 2 (2.7%) had diabetes. Eight (11%) of
these patients had 3 medical comorbidities.
There were 149 patients under the age of 65 years
in this cohort. In this subgroup, the mean age was
45.8 13.1 years and the median age was 48.0
years. There were 40 (26.8%) men and 109 (73.2%)
women. There were 18 (12.1%) patients involved in
a third-party claim and 70 (47.0%) with their dominant hand affected.
All patients were treated according to the attending
surgeons discretion. The majority of subjects 65
years were treated conservatively (56/73; 76.7%).
Fifteen patients had fractures that were well aligned
at initial presentation and were treated with casting
only (no reduction), 41 received a closed reduction
and casting, 4 were treated with intrafocal pinning
and external fixation, and 1 received an open reduction and external fixation. The mean initial displacement was 9.4 16.3 of dorsal angulation, 19.9
5.6 of radial inclination, and 1.6 mm 2.5 of ulnar
positive variance.
In the younger group, 84 (56.4%) subjects were
treated conservatively and the remainder was
treated operatively; 17 (11.4%) were treated with
an open reduction and internal fixation and the
remainder (48) was treated with a closed reduction
and percutaneous pinning, and 8 (5.4%) of these
patients required supplemental external fixation.
The degree of initial displacement was similar to
that in the older cohort. The mean initial dorsal
angulation was 11.6 18.2, the mean radial
inclination was 18.4 7.7, and the mean ulnar
positive variance was 0.83 mm 2.1.
Radiographic Parameters in Isolation

In the subgroup of younger patients, both the DASH
(p 0.002) and PRWE (p 0.02) scores were
significantly higher for patients with 3 mm of ulnar
positive variance than for those with 3 mm of ulnar
positive variance. Those with 15 of radial inclination had significantly worse PRWE scores than
those with 15 of radial inclination at final follow-up evaluation (p 0.04). The DASH score was
also markedly worse in patients with 15 of radial
inclination (11.9 vs 32.8); however, this difference
did not achieve statistical significance but did suggest
a trend with a p value of 0.065. Dorsal angulation did
not significantly influence either the PRWE (p
0.17) but showed a trend with the DASH score (p
0.06) (Table 1). No measured radiographic parameter
was found to influence patient-reported outcomes
(PRWE and DASH) in the subgroup aged 65 years
and older.
Overall Malalignment of Distal Radius Fracture
Based on the previously mentioned criteria, the majority (113/149, 75.8%) of patients aged 65 years
had acceptable positions of union, whereas (42/73,
57.5%) of patients aged 65 and older had evidence of
malalignment (Tables 2 and 3). In patients aged 65
years and older, the presence of malalignment of the
distal radius did not influence patient reports of pain
and disability; neither the PRWE (p 0.224) nor the
DASH (p 0.386) were significantly different between groups (Table 2). For patients under 65 years
of age, the presence of malalignment strongly influenced patient reports of pain and disability; both
PRWE (p 0.001) and the DASH (p 0.001) were
significantly higher (ie, more pain and disability) in
the group with malalignment at 1 year (Table 2).
Using the less stringent criteria for acceptability of
alignment reported by Kelly23 (5 mm ulnar positive variance and 30 of dorsal angulation), we see
966
Table 2. Overall Acceptability of Alignment and Effect on Patient-Reported Outcomes (PRWE and DASH)
PRWE
DASH
<65 Years
>65 Years
Acceptable Alignment*
Yes
No
Yes
No
13.1
9.2
29.3
23.2
.001
.001
11.9
15.4
18.3
19.7
.22
.39
*Acceptable alignment: dorsal angulation 10, radial inclination 15, radial shortening 3 mm ulnar positive variance.
that older patients (PRWE p 0.78; DASH p

0.73) were still able to tolerate malalignment while
their younger counterparts were not (PRWE and
DASH, p 0.001) (Table 3).
number needed to harm) supporting its earlier use as

an age cut-off point for acceptable and unacceptable
alignment. Figure 2A and B clearly displays how the
NNH increases with age. In patients over the age of
65 years, 8 patients would need a malaligned distal
radius before 1 poor outcome is experienced (based
on the DASH score). A similar trend is seen with the
PRWE, where the NNH at the age of 65 years is 9
(Fig. 2B). This contrasts with younger patients where
1 in 2 (based on DASH) or 1 in 3 (based on PRWE)
will experience a poor outcome if left with a malaligned distal radius. Although the trend with increasing age is still evident, the power of observations in
the older group decreases after age 70 years as the
sample size becomes smaller.
Risk of Poor Outcome With Malalignment of the

Distal Radius
Based on relative risk (RR), malalignment of the
distal radius increases the risk of having a poor
outcome (DASH 20 or PRWE 20) in all age
groups (RR 1); however, the RR is only considered
to be statistically significant when it is greater than 2.
Using this criteria, at each age interval, patients in the
younger group have a significantly higher risk (RR
2) of a poor outcome with malalignment, whereas
those in the older group do not. This even occurs at
the age intervals of 70 and 75 years (ie, RR 3.7 for
patients under 75 and RR 1.7 for patients older
than 75). When looking at the older group at each age
interval, we find that after the age of 60 years, the RR
of a poor outcome with malalignment of the distal
radius was not found to be statistically significant
(RR 2) (Fig. 1A). A similar trend was seen with the
PRWE (Fig. 1B). Figure 1A and B clearly demonstrates that the relationship between malalignment
and outcome is not an all-or-none phenomenon occurring at a specific age but rather presents as a
decreasing gradient of risk with advancing age.
This gradient appears to change most significantly
at the age of 65 (for both the relative risk and the
Discussion
Based on our literature search, this study represents
the largest published cohort study of isolated extraarticular distal radius fractures in the literature in
which standardized patient-reported outcomes are
used to prospectively quantify pain and disability.
We have focused on the risk of a poor outcome with
malalignment of the distal radius fracture because, in
our view, surgeons typically make decisions on the
overall fracture severity as indicated by x-ray rather
than on isolated radiographic measures. Previous
studies that have evaluated isolated measures may
have underappreciated the overall impact of mal-
Table 3. Overall Acceptability of Alignment and Effect on Patient-Reported Outcomes (PRWE and DASH)
Using Criteria of 5 mm Ulnar Variance and 30 of Dorsal Angulation
PRWE
DASH
<65 Years
>65 Years
Yes
No
Yes
No
16.7
12.5
73.0
50.0
.001
.001
15.8
18.1
13.4
15.2
.78
.73
*Acceptable Alignment: dorsal angulation 30 or 5 mm ulnar positive variance.
967
Figure 1. (A) Relative risk of poor outcome with malalignment of the distal radius: age-related trends (based on DASH
score). (B) Relative risk of poor outcome with malalignment
of the distal radius: age-related trends (based on PRWE
score).
on in the past; however most studies report only on

cohorts of low-demand elderly patients.22,24,29 These
studies have all shown that elderly, low-demand patients tolerate residual deformity well. Young and
Rayan24 and Chang et al22 found that unsatisfactory
radiographs did not correlate with poor function in
patients over 60 years of age with low physical
demands. In the study by Young and Rayan,24 patients also had multiple systemic conditions and were
considered to have high surgical risks.
Other studies have shown that attempts at reduction are not beneficial in older patients because even
if satisfactory reduction is obtained, it is difficult to
maintain.23,35 Beumer and McQueen studied older
patients with dementia and multiple medical comorbidities and found no correlation between fracture
classification, initial displacement, and final outcome
in patients who received a closed reduction and those
who did not.35 Kelly et al also reported no detectable
difference in radiologic or functional outcomes at 3
months between elderly patients who received a
closed reduction and those who did not. They concluded that up to 30 of dorsal angulation and 5 mm
of radial shortening can be accepted without requiring further manipulation.23
alignment and few have focused on risk; particularly

in terms of how many patients with malalignment
would need an adequate reduction to avoid one adverse outcome.
Our study demonstrates that the relationship between patient-reported pain and disability and distal
radial malalignment is different in patients younger
and older than 65 years of age. As expected, patients
under the age of 65 years demonstrate a very strong
link between poor outcomes (based on PRWE and
DASH) and the presence of a malalignment of the
distal radius. In particular, fractures with a loss of
radial length correlated with significantly higher reports of pain and disability, and there were trends
toward higher pain and disability with unacceptable
degrees of radial inclination and dorsal tilt. These
results are similar to other published reports in the
literature although different studies may stress the
importance of different radiographic parameters.2,15,16,28,34 Patients 65 years and older showed
no significant relationship between malalignment of
the distal radius and patient reports of pain and
disability. This was true when the parameters were
examined in isolation and when clustered together.
The relationship between distal radial malalignment and function in older patients has been reported
Figure 2. (A) Number needed to harm based on DASH score.

(B) Number needed to harm based on PRWE score.
968
Similar to this study, there have been other reports

on outcomes in elderly patients, without isolating
patients with low demands. Roumen et al reviewed
the results of 43 patients, over the age of 55 years,
whose distal radius fractures displaced after the initial reduction. These patients were randomized to
remanipulation and external fixation or acceptance of
deformity. They found there was no correlation between final anatomic and functional outcome; however, this was a retrospective review and included
primarily intra-articular fractures (95%).30
Anzarut et al investigated the effect of excessive
dorsal angulation on outcomes in conservatively
treated distal radius fractures. They studied 74 elderly patients (over the age of 50), all of whom were
living independently prior to their fracture. They
found that acceptable dorsal angulation was not associated with better generic physical or mental health
status, lesser degrees of upper-extremity disability, or
greater satisfaction than was unacceptable for dorsal
angulation at 6 months.25 In contrast with our study,
a significant selection bias was present as patients
requiring surgery or hospital admission were excluded from their sample. In addition, acceptability
of reduction was based on dorsal angulation only;
radial shortening was not considered.
Despite slight differences in methodology, the cohort of primarily healthy patients over the age of 65
years in this study demonstrated a similar relationship between malalignment and outcome as reported
by others.22,24,25,30 We found that patients over the
age of 65 years showed no statistically significant
relationship between radial length, radial inclination,
or dorsal tilt to pain and disability experienced at 1
year. When these variables were clustered together to
determine the overall acceptability of alignment,
similar results were seen. Although the reported
PRWE and DASH scores were higher (ie, more pain
and disability) in the group with malalignment (Tables 2 and 3), this was not found to be statistically
significant.
The clinical question, which is still unanswered,
is how one should use this data to influence clinical
decision-making in healthy, active elderly patients.
Beumer and McQueen postulated that as malunion
has been found to result in problems for younger
patients, it would be reasonable to assume that a
similar relationship exists between radiographic
and functional outcome in elderly patients, if similar demands were placed on their wrists.35 The
relative risk data reported in this study supports
this hypothesis.
The relative risk and the NNH can be used to

translate the relationship between malalignment
and patient-rated pain and disability into practical
guidelines. The data help demonstrate that the
likelihood of a poor outcome with malalignment in
elderly patients should not be regarded as an allor-none phenomenon but rather as a gradient of
risk. The risk of a poor outcome with malalignment is higher than with acceptable alignment at
all age groups, but this likelihood decreases with
advancing age. This is clearly depicted by the
NNH. Using the DASH score, patients over the age
of 65 years will experience 1 bad outcome for
every 8 patients who present with unacceptable
alignment of their distal radius, and patients over
70 years will experience 1 bad outcome for every
10 who present with unacceptable alignment.
These numbers can be presented to patients to help
guide their decision to pursue surgery or accept the
consequences of malunion. It also gives strength to
the use of the age 65 years as a cut-off point when
comparing older and younger patients.
In summary, our data demonstrate that isolated
measures of malalignment of the distal radius do
not have a statistically significant effect on selfreported pain and disability in patients 65 years.
The relative risk data demonstrate, however, that
all patients have a higher risk of a poor outcome
with malalignment of the distal radius, regardless
of age. This risk decreases with advancing age but
is still present. To illustrate, in patients 65 years
of age, 8 malaligned radii would have to be corrected to avoid 1 poor outcome, whereas this number is reduced to only 2 in patients under age 65
years (based on DASH).
One of the limitations of this study is that we only
considered radiographic variables and their influence
on outcome in isolation. There may be other variables responsible for influencing outcomes in this
population, and these may or may not be similar to
predictive factors in younger patients. In addition,
although we did not isolate low-demand elderly patients, we also did not isolate elderly patients who
place high demands on their wrist but rather included a cross section of patients 65 years of age.
The factors influencing outcomes in elderly patients are not completely understood at this point.
Many factors may be influencing outcomes in this
age group and should be investigated in the future:
the presence of home supports, underlying bone mineral density,36 and an objective assessment of baseline physical demands and activity level. A trial
investigating the importance of these and other factors in healthy, active elderly patients is necessary to
further evaluate the relationship between outcome
and malalignment in Colles fractures in this growing
population demographic.
Received for publication February 7, 2007; accepted in revised form May
11, 2007.
article.
Corresponding author: Ruby Grewal, MD, Assistant Professor, Division of Orthopedic Surgery, University of Western Ontario, Hand and
Upper Limb Center, St Josephs Health Care, 268 Grosvenor Street,
London, ON N6A 4L6, Canada; e-mail: rgrewa@uwo.ca.
0363-5023/07/32A07-0003$32.00/0
doi:10.1016/j.jhsa.2007.05.009
References
1. Altissimi M, Antenucci R, Fiacca C, Mancini GB. Longterm
results of conservative treatment of fractures of the distal
radius. Clin Orthop 1986;206:202210.
2. Aro HT, Koivunen T. Minor axial shortening of the radius
affects outcome of Colles fracture treatment. J Hand Surg
1991;16A:392398.
3. Bacorn RW, Kurtzke JF. Colles fracture: a study of two
thousand cases from the New York State Workmens Compensation Board. J Bone Joint Surg 1953;35A:643 658.
4. Bickerstaff DR, Bell MJ. Carpal malalignment in Colles
fractures. J Hand Surg 1989;14B:155160.
5. Field J, Warwick D, Bannister GC, Gibson AGF. Longterm
prognosis of displaced Colles fracture: a 10-year prospective review. Injury 1992;23:529 532.
6. Frykman G. Fracture of the distal radius including sequelaeshoulder-hand fingers syndrome, disturbance in the distal
radio-ulnar joint and impairment of nerve function. A clinical and experimental study. Acta Orthop Scand Suppl 1967;
108:1155.
7. Gartland JJ, Werley CW. Evaluation of healed Colles fractures. J Bone Joint Surg 1951;33A:895907.
8. Green JT, Gay FH. Colles fracture-residual disability. Am J
Surg 1956;91:636 641.
9. Jenkins NH, Mintowt-Czyz WJ. Malunion and dysfunction
in Colles fracture. J Hand Surg 1988;13B:291293.
10. Lidstrom A. Fractures of the distal end of the radius. A
clinical and statistical study of end results. Acta Orthop
Scand Suppl 1959;41:1118.
11. Mandell BB. Assessment of results of treatment of 100 cases
of Colles fracture. S Afr Med J 1965;39:171174.
12. Mason ML. Colles fracture: a survey of end-results. Br J
Surg 1953;40:340 346.
13. Porter M, Stockley I. Fractures of the distal radius intermediate and end results in relation to radiologic parameters.
Clin Orthop Relat Res 1987;220:241252.
14. McQueen M, Caspers J. Colles fracture: does the anatomical
result affect the final function? J Bone Joint Surg 1988;70B:
649 51.
15. Karnezis IA, Panagiotopoulos E, Tyllianakis M, Megas P,
Lambiris E. Correlation between radiological parameters
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
969
and patient-rated wrist dysfunction following fractures of the

distal radius. Injury 2005;36(12):14351439.
Rubinovich RM, Rennie WR. Colles fracture: end results in
relation to radiologic parameters. Can J Surg 1983;26:361
363.
Stewart HD, Innes AR, Burke FD. Functional cast-bracing
for Colles fractures. J Bone Joint Surg 1984;66B:749 753.
Cassebaum WH. Colles fracture, a study of end results.
JAMA 1950;143:963965.
Pool C. Colles fracture: a prospective study of treatment.
J Bone Joint Surg 1973;55B:540 544.
Smaill GB. Long-term follow-up of Colles fracture. J Bone
Joint Surg 1965;47B:80 85.
Tsukazaki T, Takagi K, Iwasaki K. Poor correlation between
functional results and radiographic findings in Colles fracture. J Hand Surg 1993;18B:588 591.
Chang HC, Tay SC, Chan BK, Low CO. Conservative
treatment of redisplaced Colles fractures in elderly patients
older than 60 years old - anatomical and functional outcome.
Hand Surg 2001;6(2):137144.
Kelly AJ, Warwick D, Crichlow TPK, Bannister GC. Is
manipulation of a moderately displaced Colles fracture
worthwhile? A prospective randomized trial. Injury 1997;
28(4):283287.
Young BT, Rayan MD. Outcome following nonoperative
treatment of displaced distal radius fractures in low-demand
patients older than 60 years. J Hand Surg 2000;25A:19 28.
Anzarut A, Johnson JA, Rowe BH, Lambert RG, Blitz S,
Majumdar SR. Radiologic and patient-reported functional
outcomes in an elderly cohort with conservatively treated
distal radius fractures. J Hand Surg 2004;29A:11211127.
Bradway J, Amadio P, Cooney W. Open reduction and
internal fixation of displaced, comminuted intra-articular
fractures of the distal end of the radius. J Bone Joint Surg
1989;71:839 847.
Beharrie AW, Beredjiklian PK, Bozentka DJ. Functional
outcomes after open reduction and internal fixation for treatment of displaced distal radius fractures in patients over 60
years of age. J Orthop Trauma 2004;18:680 686.
Trumble TE, Wagner W, Hanel DP, Vedder NB, Gilbert M.
Intrafocal (Kapandji) pinning of distal radius fractures with
and without external fixation. J Hand Surg 1998;23A:381
394.
Beumer A, McQueen MM. Fractures of the distal radius in
low-demand elderly patients Closed reduction of no value in
53 of 60 wrists. Acta Orthop Scand 2003;74(1):98 100.
Roumen RM, Hesp WL, Bruggink ED. Unstable Colles
fractures in elderly patients. A randomised trial of external
fixation for redisplacement. J Bone Joint Surg 1991;73B:
307311.
MacDermid J. Development of a scale for patient rating of
wrist pain and disability. J Hand Ther 1996;9:178 183.
MacDermid J, Turgeon T, Richards R, Beadle M, Roth J.
Patient rating of wrist pain and disability: a reliable and valid
measurement tool. J Orthop Trauma 1998;12:577586.
MacDermid J, Richards R, Donner A, Bellamy N, Roth J.
Responsiveness of the short form-36, disability of the arm,
shoulder, and hand questionnaire, patient-rated wrist evaluation, and physical impairment measurements in evaluating
recovery after a distal radius fracture. J Hand Surg 2000;
25A:330 340.
970
34. Batra S, Gupta A. The effect of fracture-related factors on

the functional outcome at 1 year in distal radius fractures.
Injury 2002;33(6):499 502.
35. Beumer A, McQueen M. Fractures of the distal radius in
low-demand elderly patients. Closed reduction of no
value in 53 of 60 wrists. Acta Orthop Scand

2003;74(1):98 100.
36. Hollevoet N, Verdonk R. Outcome of distal radius fractures
in relation to bone mineral density. Acta Orthop Belg 2003;
69(6):510 514.
Biomechanical Evaluation
of Locking Plate Radial Shaft
Fixation: Unicortical Locking
Fixation Versus Mixed Bicortical
and Unicortical Fixation in a Sawbone Model
Jason W. Roberts, MD, Steven I. Grindel, MD, Brandon Rebholz, MD,
Mei Wang, PhD
From the Department of Orthopaedic Surgery, Medical College of Wisconsin, Milwaukee, WI.
Purpose: Compression plating is a commonly accepted technique for treating diaphyseal

forearm fractures. The purpose of this study was to evaluate the stabilizing effects of two
hybrid fixations that replace the end screws of a locked unicortical fixation with bicortical
(locked or unlocked) screws and to compare these hybrid fixations to an unlocked bicortical
fixation.
Methods: Sixteen composite radius sawbones were equally divided into 4 groups. We
performed a midshaft osteotomy and plate fixation on the volar surface with 1 of 4 different
constructs: 3 unlocked bicortical screws on each side (unlocked bicortical), 3 locked unicortical screws on each side (locked unicortical), or with 2 unicortical locked screws near the
fracture and 1 bicortical unlocked (unlocked hybrid) or locked (locked hybrid) screw distant
from the fracture on each end (LCP system, Synthes USA, Paoli, PA). Specimens were tested
in nondestructive 4-point bending and torsion on a servo-hydraulic material testing system.
The construct stiffness was obtained from the linear portion of the load-displacement curves
after 3 cycles of preconditioning. The results from all groups were compared using analysis
of variance and post hoc Bonferroni tests.
Results: Under torsional loads, replacing the end screws of a locked unicortical configuration
with bicortical screws significantly improved the construct stiffness: 57.6% increase for the
locked screws and 51.6% increase for the unlocked. In anteroposterior (AP) bending, the
highest improvement over the locked unicortical configuration came from the locked hybrid
constructs (42.9% increase). When compared with the unlocked bicortical configuration,
both hybrid constructs provide equivalent stability in torsion but superior stability in AP
bending.
Conclusions: Replacing a single set of unicortical locked screws with locked or unlocked
bicortical screws distant from the fracture site improves torsional stability of the construct by
more than 50%, giving stability equal to standard unlocked plating. The hybrid fixation,
however, with locked bicortical end screws has the best stability in AP bending. (J Hand Surg
2007;32A:971975. Copyright 2007 by the American Society for Surgery of the Hand.)
Key words: Bicortical, construct stiffness, locked plating, radius, unicortical.
nternal fixation using compression plating techniques is a commonly accepted treatment for
diaphyseal forearm fractures. With the introduction of locking plates, there has been interest in using
these devices for diaphyseal fractures. Whereas the
benefits of fixed-angle locking constructs in periar-
ticular fractures are more established, there are

also scenarios where they offer advantages when
treating diaphyseal fractures. When compared with
traditional compression plating techniques, locking
plates can offer improved fixation in osteoporotic
bone or when treating highly comminuted fracThe Journal of Hand Surgery
971
972
tures.1,2 Locked plates also have potential benefits

with the current emphasis of biologically friendly
fracture treatment using indirect reduction and bridge
plating techniques with minimal periosteal stripping.
Because they are not compressed to the bone, the
periosteal blood supply may be better preserved.3,4
They also allow the effective use of unicortical
screws, which can be advantageous in certain clinical
scenarios.
We compared a locking construct fixed with all
unicortical screws to a compression plate construct in
a cadaver radius model. Whereas the locked devices
resisted bending forces equally, they were inferior
against torsional forces. Other authors have also
found that constructs fixed exclusively with unicortical screws were weak in torsion.5,6 Previous studies
of traditional compression plates have shown that
unlocked devices tend to fail in torsion at the point of
fixation farthest from the fracture and that bicortical
screws at this point greatly improve rotational stability.7,8 We hypothesized that this would prove true in
locking constructs as well.
Composite bone models that simulate the physical
properties of real bone have proved useful and valid
for mechanical studies of fracture fixation.5 Compared with cadaver bone models, they offer the advantage of interspecimen consistency that is especially beneficial for comparative studies.9,10 With
this low variability, much smaller sample sizes can
detect significant differences between constructs. Recently, a composite radius model has become available. Preliminary tests have shown that fixation of
these models behaves similar to fixation in cadaveric
specimens by being weaker in torsion when all unicortical screws are used. These preliminary tests also
indicate mechanical stiffness within the known range
for cadaver specimens, most closely simulating
healthy adult cortical bone. The objective of this
study was to test the hypothesis that use of a single
bicortical screw at both ends of a locking plate construct results in torsional stability similar to traditional compression plating techniques.
tions were used in each specimen with 3 screws

placed on either side of the osteotomy. The central 2
screw holes were not used in any of these configurations, thus holes 1, 2, 3, 6, 7, and 8 were filled in
each specimen. Group 1 was fixed with 6 bicortical
unlocked screws (unlocked bicortical), group 2 was
fixed with 6 unicortical locked screws (locked unicortical), and groups 3 and 4 were hybrid fixations
with 4 locked unicortical screws (holes 2, 3, 6, and 7)
combined with end bicortical screws on either end of
the construct (holes 1 and 8). These bicortical screws
were unlocked in group 3 (unlocked hybrid) and
were locked in group 4 (locked hybrid) (Fig. 1).
The plates were placed on the volar (anterior)
surface and were slightly contoured to fit the bow of
the radius models. All screws were placed according
to manufacturer recommendations; 3.5-mm self-tapping screws were used in the unlocking portion of the
combiholes for the unlocked screws and 3.8-mm
self-tapping locked screws were used for all of the
locked screws. For the 4 specimens in group 3, the
unlocked screws were applied prior to the locking
screws as is the recommended technique.
The specimens were then potted in customized jigs
using dental cement and were subjected to a series of
nondestructive mechanical tests with a servo-hydraulic, bimodal materials testing system (Model 809;
MTS Systems, Eden Prairie, MN). Nondestructive
4-point bend testing was carried out in 2 planes with
a maximum moment of 8 Nm. These planes were
designated as the anteroposterior (AP) and lateral
planes. The tests were run under displacement-control at a rate of 2.5 mm/s. Nondestructive testing was
then performed in torsion with a rate of 10 degrees/s
to a maximum torque of 5 Nm. Three cycles of
preconditioning were conducted in both bending and
torsional tests. Load-displacement behavior of each
construct was recorded, and the slope of the linear

A total of 16 composite radius sawbone specimens
were obtained for the study. We performed a transverse osteotomy of the midshaft using a manual saw
and then divided the specimens into 4 study groups
(n 4). All groups were fixed with 8-hole 3.5-mm
LCP plates with combination holes (combiholes) that
allow for locked or unlocked fixation (Synthes USA,
Paoli, PA). A total of 6 screws in various configura-
Figure 1. Screw configuration of the 4 study groups.
Roberts et al / Biomechanics of the Radius Locking Plate Fixation
973
Table 1. Bending Stiffness of the 4 Fixation Groups in AP and Lateral Directions
AP
Lateral
Unlocked Bicortical
(N/mm)
Locked Unicortical
(N/mm)
Unlocked Hybrid
(N/mm)
Locked Hybrid
(N/mm)
161.1 10.9
426.2 95.0
135.8 13.6
425.7 58.1
151.3 10.3
437.8 60.3
194.1 20.0
430.8 81.4
region of the curve was defined as the stiffness of the

construct. Statistical comparison of the construct
stiffness from the 4 study groups was carried out
using single-factor analysis of variance (ANOVA)
and post hoc Bonferroni/Dunn comparisons. The statistical significance was set at .05 for ANOVA and
.008 for Bonferroni/Dunn comparison.
Results
Four-Point Bending
Regardless of the type of fixation, the average values
of construct bending stiffness in the lateral direction
were 23 times the values of the bending stiffness in
the AP direction (Table 1). In the lateral direction,
the differences in bending stiffness among the 4
constructs were small and statistically non-significant
(ANOVA test, p 0.99). In the AP direction, the
stiffness of the locked hybrid configuration was significantly higher than the other 3 configurations (Fig.
2). The increase was 28.3% over unlocked hybrid (p
.001), 20.4% over unlocked bicortical (p .007),
and 42.9% over locked unicortical (p .0001). None
of the differences between the locked unicortical,
unlocked bicortical, and unlocked hybrid was statistically significant (Bonferroni test, p 0.03). A
retrospective analysis showed that the statistical
power for the ANOVA test was 99.6%.
Torsion
As expected, locked unicortical constructs were the
weakest construct under torsional loads. Replacing
the unicortical screws at the end holes with either
locked or unlocked bicortical screws significantly
improved the torsional stiffness of the plating construct (Fig. 3). Compared with the locked unicortical
constructs, the locked hybrid constructs showed
57.6% increase (p .0004), whereas the unlocked
hybrid demonstrated a comparable 51.6% increase
(p .0009). Torsional stiffness of both hybrid constructs was similar to that of the unlocked bicortical
constructs, with no statistically significant differences detected among the three (Bonferroni test, p
0.62). A retrospective analysis showed that the statistical power for the ANOVA test was 99.2%.
Discussion
Many variables contribute to a plate constructs stability. Among these are bone quality, plate length,
distance from the plate to the bone, working length,
and the material properties of the plate. We attempted to keep these variables constant and studied
the effects of varied screw configurations only. The
fixation construct of an 8-hole plate with a total of 6
screws was selected because it is commonly used for
a forearm fracture. In this model, a transverse osteotomy was made after the plates were applied leaving
a small (approximately 1 mm) fracture gap. An osteotomy with a wider gap may simulate a comminuted fracture more reliably when larger bending
loads are tested. In our study, the resultant deflection
angle at the osteotomy site was small enough (less
than 3) and the gap remained open, and any effect
that friction between the interfaces could have was
eliminated.
Conventional compression plates and locked
plates provide stability by different mechanical principles. Compression plates rely on friction between
the plate, screws, and bone, whereas locked plates
rely on the threaded plate-screw interface to lock the
bone fragments in place. In osteoporotic bone, the
screws cannot consistently provide the necessary
friction for stable unlocked fixation, thus a locked
bicortical screw will likely be more reliable in poorquality bone.1 Because of the different mechanism,
some have recommended not to combine the 2 types
of fixation for the same fracture fragment.1113
Sometimes it is preferable, however, to establish
friction fit between plate and bone with unlocked
Figure 2. Mean (SD) construct stiffness under 4-point bending in the AP direction.
974
Figure 3. Mean (SD) construct stiffness in torsion.
screws and then secure the fixation with additional

locking screws.14 In this study, we chose to add
bicortical screws to the end holes. The rationale for
this selection is 2-fold: bicortical purchase improves
the screws working length in torsion,6,15 and the
end-hole locations on the plate represent the largest
working length in bending. Clinically, in the cases of
spanning comminuted or pathologic bone, the end
screw positions allow the most reliable bicortical
screw purchase. When placing a combination of
locked and unlocked screws in a construct, it is
recommended that the unlocked screws be applied
first.14 We followed this recommendation when applying the plates.
Our study found increased torsional stability in
constructs with both locked and unlocked end bicortical screws as well as with all bicortical screws when
compared with constructs fixed exclusively with
locked unicortical screws. This indicates that the
dominant factor for improved torsional stability is
likely to be the increased working length of the
bicortical screws. Locked or unlocked fixation has
little influence. The factor contributing to the increased bending strength (AP plane) in the constructs
fixed with locked bicortical end-screws, however,
seems to be different from the other 3 types of
constructs. The working distance for the 6 screws
was identical among all constructs as the same
screws locations were employed. One probable explanation might be that improved local rigidity,
through the locked and bicortical fixation, from the
screw with the largest working distance (end screws)
strengthens the bending stability of the whole construct.
There are inherent limitations to any study using
synthetic bone models. The biological factors that
contribute to fracture healing cannot be studied.
These models have proved valuable in mechanical
studies of fracture fixation due to the reduced variability between specimens and their relatively low
cost. One disadvantage of the synthetic bone models
we used is that they simulate adult cortical bone more
closely than osteoporotic bone. Although the relative

weakness of the constructs with all unicortical screws
was apparent in this model, the advantages of the
locking bicortical screws may be more dramatic in
osteoporotic bone than was demonstrated in this
study.
In summary, findings from this study show that
when a locked device is chosen, use of exclusively
unicortical fixation results in considerable weakness
relative to traditional compression plating and may
not adequately stabilize fractures subjected to high
torsional forces. A single bicortical screw at both
ends of a locked plating construct significantly improves the torsional stability of these constructs. If
this bicortical screw is locked, it will improve further
the AP bending stability of the construct.
The authors would like to thank Synthes USA (Paoli, PA) for providing
the implants used in this study and Ms. Linda McGrady for her assistance
in mechanical testing and data analysis.
Received for publication April 17, 2007; accepted in revised form May
21, 2007.
article.
Corresponding author: Mei Wang, PhD, Department of Orthopaedic
Surgery, Medical College of Wisconsin, 9200 W. Wisconsin Avenue,
P.O. Box 26099, Milwaukee, WI 53226-0099; e-mail: meiwang@mcw.
edu.
0363-5023/07/32A07-0004$32.00/0
doi:10.1016/j.jhsa.2007.05.019
References
1. Egol KA, Kubiak EN, Fulkerson E, Kummer FJ, Koval KJ.
Biomechanics of locked plates and screws. J Orthop Trauma
2004;18:488 493.
2. Seebeck J, Goldhahn J, Stadele H, Messmer P, Morlock
MM, Schneider E. Effect of cortical thickness and cancellous bone density on the holding strength of internal fixator
screws. J Orthop Res 2004;22:12371242.
3. Perren SM. Evolution of the internal fixation of long bone
fractures. The scientific basis of biological internal fixation:
choosing a new balance between stability and biology.
J Bone Joint Surg 2002;84B:10931110.
4. Frigg R. Locking Compression Plate (LCP). An osteosynthesis plate based on the Dynamic Compression Plate and
the Point Contact Fixator (PC-Fix). Injury 2001;32(Suppl):
63 66.
5. Fulkerson E, Egol KA, Kubiak EN, Liporace F, Kummer FJ,
Koval KJ. Fixation of diaphyseal fractures with a segmental
defect: a biomechanical comparison of locked and conventional plating techniques. J Trauma Injury Infect Crit Care
2006;60:830 835.
6. Stoffel K, Dieter U, Stachowiak G, Gachter A, Kuster MS.
Biomechanical testing of the LCP how can stability in
locked internal fixators be controlled? Injury 2003;34(2):
B1119.
Roberts et al / Biomechanics of the Radius Locking Plate Fixation

7. ElMaraghy AW, ElMaraghy MW, Nousiainen M, Richards
RR, Schemitsch EH. Influence of the number of cortices on
the stiffness of plate fixation of diaphyseal fractures. J Orthop Trauma 2001;15:186 191.
8. Miclau T, Remiger A, Tepic S, Lindsey R, McIff T. A
mechanical comparison of the dynamic compression plate,
limited contact-dynamic compression plate, and point contact fixator. J Orthop Trauma 1995;9:1722.
9. Cristofolini L, Viceconti M. Mechanical validation of
whole bone composite tibia models. J Biomech 2000;33:
279 288.
10. Cristofolini L, Viceconti M, Cappello A, Toni A. Mechanical validation of whole bone composite femur models.
J Biomech 1996;29:525535.
975
11. Gardner MJ, Helfet DL, Lorich DG. Has locked plating
completely replaced conventional plating? Am J Orthop
2004;33:439 446.
12. Wagner M. General principles for the clinical use of the
LCP. Injury 2003;34(2):B31 42.
13. Krettek C, Haas N, Tscherne H. The role of supplemental
lag-screw fixation for open fractures of the tibial shaft treated
with external fixation. J Bone Joint Surg 1991;73A:893 897.
14. Gardner MJ, Griffith MH, Demetrakopoulos D, Brophy RH,
Grose A, Helfet DL, et al. Hybrid locked plating of osteoporotic fractures of the humerus. J Bone Joint Surg 2006;
88A:19621967.
15. Kubiak EN, Fulkerson E, Strauss E, Egol KA. The evolution
of locked plates. J Bone Joint Surg 2006;4A:189 200.
Distal Radius Osteotomy in the Elderly

Patient Using Angular Stable Implants and
Norian Bone Cement
Santiago Lozano-Caldern, MD, Michael Moore, MD,
Matthew Liebman, MD, Jesse B. Jupiter, MD
From the Hand and Upper Extremity Service, Department of Orthopaedic Surgery, Massachusetts General
Hospital-Harvard Medical School, Boston, MA; the Hand and Upper Extremity Center, Arkansas Specialty,
Little Rock, AR; and the Orthopaedic Hand Surgery Service, Massachusetts General Hospital, Harvard
Medical School, Boston, MA.
Purpose: To report our results after testing the combination of two technologiesangularstable locking screw implants and Norian SRS cementin corrective osteotomies of the distal
radius in the elderly. This technique eliminates donor site bone-graft morbidity and expands
the indications of corrective osteotomies to older patients with osteoporotic bone.
Methods: Our retrospective series include 6 patients (5 women and 1 man) with an average
age of 60 years. Three patients had corrections through a dorsal approach, 1 through a volar
approach, and 2 through a combined approach. Two corrections included an intraarticular
osteotomy. We used 2.4-mm volar T plates in patients approached volarly and 2.4-mm L and
T plates for those approached dorsally; the osseous defect was filled with bone cement
(Norian SRS). Range of motion and grip strength were measured at 16 months average
follow-up. Standard wrist radiographs were taken to evaluate alignment and determine
improvement. At final follow-up, patients completed the Modified Mayo Wrist score, the
Modified Gartland and Werley score, and the Disabilities of the Arm, Shoulder, and Hand
(DASH) questionnaire.
Results: There were no perioperative complications. All corrective osteotomies healed. One
patient required a Darrach procedure at 6 months. The average wrist and forearm motion was
77% of the opposite side and grip strength 88% of the opposite side. The average total
correction in the sagittal plane was 22 with all patients returning to neutral or better
alignment. The average ulnar variance improvement was 2 mm. Average postoperative
DASH was 28 points; average Modified Mayo Wrist score was 68; and the Modified Gartland
and Werley score averaged 9 points.
Conclusions: We believe that corrective osteotomy of the distal radius in the elderly using
angular stable implants and Norian calcium phosphate cement is a safe and predictable surgical
technique, even in patients with underlying osteoporosis. It eliminates donor site morbidity, and
patient-rated outcome measures demonstrated acceptable daily living function return. (J Hand
Surg 2007;32A:976 983. Copyright 2007 by the American Society for Surgery of the Hand.)
Type of study/level of evidence: Therapeutic IV.
Key words: Angle stable implants, corrective osteotomy, distal radius, malunion, Norian
cement.
nion with deformity is the most common complication after a distal radial fracture.1 This
deformity can be intra-articular, affecting either
the radiocarpal or radioulnar joints; extra-articular,
characterized by metaphyseal angulation and loss of
length; or it may be a combination of both.1
U
976
Corrective osteotomies have been proved to be an

effective treatment for symptomatic malunion.1,2 A
variety of techniques have been used; however, there
has remained concern regarding the indications for
surgical intervention in the presence of underlying
osteoporosis as well as the recognized morbidity
Lozano-Caldern et al / Distal Radius Osteotomy in the Elderly Patient
associated with autogenous iliac crest bone grafting.39

Technical advances including use of precontoured
internal fixation devices with angular stable fixation
locking screws, as well as use of osteointegration
biomaterials, have offered some advantages.
The contoured implants facilitate osteosynthesis
by providing higher stability even in osteopenic
bone.10 12 These implants afford osseous fixation
that allows early motion and rehabilitation,13,14 as
their precontoured shape maintains desirable patterns
of alignment, congruency, and inclination of the distal radius after corrective osteotomy.1318 These
properties reduce the probabilities of screw loosening
and consequent loss of reduction.10,16 18
After osteotomy and achievement of proper angulation and alignment, there will exist a three-dimensional defect that must be filled to adequately support
the bone fragments.19 23 Autogenous bone grafts
structural and nonstructural have been widely used
for this purpose and remain the gold standard.23
However, they have a recognized potential for donor
site morbidity, in particular those involving corticocancellous variants.39 Alternatives to bone graft can
be classified in different groups: demineralized bone
matrix (DBM), bone morphogenetic proteins (BMP),
osteoconductive bone graft substitutes, and materials
with osteogenic properties.23 The third group, osteoconductive bone graft substitutes, comprehends allografts and calcium phosphate synthetic substitutes. The
later can be divided into cements, ceramics, and composite grafts.23 Norian Skeletal Repair System (Norian
SRS, Norian Corp., West Chester, PA) belongs to the
cement category and differentiates from other cements by its injectable nature.19 24 Norian cement
properties and advantages include its biocompatibility with higher compressive strength than cancellous
bone but less than cortical; fast setting in vivo at
physiologic pH and temperature; its bioconductive
nature with physical and chemical properties comparable with those of bone minerals; and its easy visualization in radiographic studies.20 22,25
Previous studies have reported successful results
with corrective osteotomies of the distal radius and
posterior percutaneous fixation with K-wires and
Norian cement. Luchetti in 2004 reported in his case
series of 6 patients satisfactory functional outcomes
after using this technique.26 Additionally, the combination of volar plate fixation and Norian cement
has also been reported before. Yasuda et al in a case
report presented adequate results in a 37-year-old
patient who received a corrective osteotomy of the
977
distal radius and posterior fixation with volar plate

fixation and Norian cement.27 The purpose of this
study is to report our results after testing the simultaneous use of angular-stable locked implants and
Norian SRS cement in six elderly patients with subjacent osteoporosis who required a corrective osteotomy of the distal radius because of symptomatic
malunion.

Between 2002 and 2004, 6 patients (5 women and 1
man) with an average age of 60 years (range, 53 to 74
years) were treated at our orthopedic hand surgery
service by a single orthopedic hand surgeon because
of a symptomatic distal radius malunion (restricted
motion, pain, and disability). All of them agreed to
participate in the institutional review boardapproved retrospective study to evaluate the functional
outcomes of this technique for distal radius osteotomy. Their initial fractures according to the AO
Foundation classification system28 were classified as
type A.3.2 in 2 patients, A.3.3 in 1 patient, C.3.1 in
1 patient, and C.3.2 in 2 patients. According to the
Jupiter and Fernandez classification,29 3 patients
were type 1 (bending mechanism) and 3 patients
were type 3 (compression mechanism). All of them
were right handed. All patients defined their occupation as desk-based work. Four dominant hands were
involved. The mechanism of trauma was a fall from
standing height in 2 patients, higher-height fall in 2
patients, and lastly, 1 fracture occurred after an assault and 1 after a motor-vehicle collision. All patients had a diagnosis of osteoporosis before the
injury. The diagnosis of osteoporosis was previously
made by the primary care physician using bone densitometry in 3 patients. The treating orthopedic surgeon diagnosed osteoporosis in the remaining patients
by clinical assumption, combining radiographic evaluation, patient age, and history of pathologic fractures
(more than 2) (Table 1).
The patients presented at average 8 months after
trauma (range, 5 to 14 months) with symptomatic
distal radius malunions (pain, motion restriction,
functional limitation). Preoperative clinical and radiologic evaluations were done to assess range of
motion, grip and pinch strength, and to characterize
the malunion radiologically.
Three patients were classified as dorsal malunions. This pattern was typified as severe dorsal
tilt of the distal radius in the lateral plane (Fig. 1).
The physical exam of these patients consistently
showed excessive wrist extension and lack or im-
978
Table 1. Demographic Series Information
Case
Gender
Age
60
55
74
60
56
53
Hand
Dominance
Involved
Hand
Osteoporosis
Diagnosis
Mechanism of
Injury
Interval Between
Malunion and
Surgical Correction
Real estate
developer
Legal assistant
Self-height fall
14
High-height
fall
Self-height fall
Department store
supervisor
Retired arts
teacher
Software
engineering
Bone
densitometry
Clinical
assumption
Clinical
assumption
Clinical
assumption
Bone
densitometry
Manager
Occupation
Bone
densitometry
Assaulted
Motor-vehicle
collision
polytrauma
High-height
fall
5
10
6
Patients listed in boldface type are the patients that received intra-articular osteotomy.
pairment of wrist flexion. Three patients were classified as volar malunions. These were characterized as deformities with marked volar tilt of the
distal radius on the preoperative lateral view radiograph (Fig. 2). In contrast, these patients had a
lack of wrist extension and greater wrist flexion
Figure 1. Preoperative dorsal deformity of a patient with

distal radius malunion (A) (posteroanterior [PA] view). (B)
Lateral view. Postoperative status showing the correction of
the dorsal deformity with stable angle implants and Norian
cement (C) PA view. (D) Lateral view.
Figure 2. Preoperative volar deformity of a patient with distal

radius malunion (A) PA view. (B) Lateral view. Postoperative
status showing the correction of the volar deformity with
stable angle implants and Norian cement (C) PA view. (D)
Lateral view.
when compared with their control (same patient

opposite upper limb).
The osteotomy was performed through dorsal approach in 3 patients and via volar approach in 1
patient; 2 patients had a combined approach due to
removal of previously placed internal fixation (2
cases) as well as the need for median nerve decompression (2 cases). Two combined corrections included an intra-articular osteotomy. Dorsal deformities were treated in general with a dorsal osteotomy
and the volar ones volarly.
All cases were performed under regional block
anesthesia. For internal fixation, 3 patients were
treated with volar locking 2.4-mm T plates and 3
with dorsal 2.4-mm. T, L, and/or radial column locking plates (patients treated with intra-articular osteotomies or approached dorsally). After the osteotomy
and internal fixation, the created defect was filled
with Norian SRS cement (3 mL to 5 mL per case).
X-rays were taken to evaluate quality of reduction
and fixation after the osteotomy. At 10 to 14 days of
postoperative immobilization, all patients started active motion exercises.
Surgical Technique
Volar approach. A standard volar radial side approach described by Henry was used.30 A specially
designed, 2.4-mm AO locking volar plate was then
specially contoured, and 1 screw was placed in the
distal fragment. This was confirmed on fluoroscopy.
Using an osteotome through the original fracture site,
an osteotomy was created. It was slowly lengthened
with a lamina spreader, and this was confirmed under
fluoroscopy. Additional locking 2.4-mm screws were
placed proximally and distally once an acceptable
position in the frontal and sagittal planes was obtained.
Figure 3. Fixation of fragments after distal radius osteotomy.

The arrow points to the defect that will be filled with Norian
cement.
979
The defect was then filled with Norian paste (3 mL

to 5 mL); complete filling of the defect was confirmed on x-rays. Following this, the tourniquet was
released. Hemostasis was obtained and the wound
was closed in layers after placing a drain. A sterile
dressing and splint were applied in all patients. Immobilization was used from 10 to 14 days before
starting rehabilitation movement protocol.
Dorsal approach. The dorsal approach consisted of
an incision through the skin over the third compartment. Dissection of subcutaneous tissue until the
extensor pollicis longus (EPL) visualization took
place. The EPL tendon was released from the third
dorsal compartment and transposed radially and dorsally into the subcutaneous tissues where it was left
at the end of the procedure. The second and fourth
compartments were elevated subperiosteally. In the
majority of cases, two Schantz screws were placed in
the radius, one in the proximal diaphyseal area and
another distal in the metaphyseal fragment. The osteotomy was performed on a parallel plane to the
articular surface with an osteotome at the previous
fractured site. Distraction forces were applied with a
small skeletal distractor or a lamina spreader to facilitate and transiently stabilize the realignment. Realignment was monitored and evaluated under fluoroscopy. In two cases of intra-articular involvement,
a dorsal capsulotomy was performed after which the
intra-articular osteotomy was created.
Fixation was accomplished with 2.4-mm T, L,
and/or radial column locking plates LCP (Synthes,
Inc., West Chester, PA) (Fig. 3). After fixation, injection of osteoconductive biomaterial (Norian SRS)
was done after distractor removal to fill the defect
re-created by the osteotomy. Filling with Norian was
verified under fluoroscopy (Figs. 4 and 5). After
hemostasis, we proceeded with wound closure. All
patients were immobilized with a splint from 10 to 14
days before starting rehabilitation protocol.
Radiographic follow-up was done at 2, 6, 12, and
24 weeks and at the final follow-up (average 16
months; range, 6 to 32 months) using standard anteroposterior and lateral projections of the wrist.
Healing and osteointegration was determined radiologically by the absence of radiolucency areas between the bone and Norian cement interface and
clinically by the absence of pain upon palpation at
the osteotomy site.
Preoperative and postoperative range of motion
and grip strength were obtained respectively from
clinical charts and clinical measurement by an
independent observer. Postoperative range of mo-
980
Results
Figure 4. Defect filling through Norian injection in a patient

who was treated through a dorsal approach.
tion was measured and compared with the contralateral noninjured wrist and objectively quantified with a goniometer (Orthofix, Inc., McKinney,
TX). Excellent range of motion was defined as
100% of wrist and forearm motion of the contralateral limb; good results as between 75% and 99%;
fair between 50% and 74%; and bad when
achieved motion was less than 50% of the uninvolved limb. Grip strength was also tested postoperatively comparing the injured and noninjured
wrist, using a hydraulic hand dynamometer (Baseline FEI; Irvington, NY) at the third station (elbow
at 90 of flexion and the wrist and forearm in
neutral). Patients were also tested after surgery for
instability of the distal radioulnar joint using the
radioulnar ballottement test. All patients received
postoperatively the Modified Mayo Wrist score31
and the Modified Gartland and Werley32 score to
evaluate outcomes in terms of pain, ability to return to work, mobility, grip strength, residual deformities, and complications. The Disabilities of
the Arm, Shoulder, and Hand (DASH) questionnaire was also applied at the postoperative visit.
Ulnar inclination, volar tilt, radial length, and
ulnar variance were measured in preoperative and
postoperative radiographies according to the standard technique for radiographic measurement in
the radius (Sarmientos modification of Lidstroms
grading system).32 Percentage quantification of
improvement and averages were calculated per
each patient.
Follow-up evaluation of our case series did not demonstrate perioperative or postoperative complications. All osteotomies healed in an average 7 weeks
after treatment (range, 6 to 8 weeks).
At an average follow-up of 16 months (range, 6 to
22 months), an average wrist and forearm motion of
77% of the opposite side was achieved. Average
achieved wrist and forearm motion was 50 of flexion, 47 of extension, 76 of supination, 85 of pronation, 15 of radial deviation, and 30 of ulnar
deviation. There was substantial improvement in
terms of range of motion from the preoperative period to the operative. However, differences were not
statistically significant (p 0.34).
In terms of grip strength, Modified Mayo Wrist
score scale was used. All patients were rated as good
strength (strength between 75% and 99%). The grip
strength on average was 88% when compared with
the uninvolved hand. The average grip was 32 kg.
None of the patients had signs of distal radioulnar
joint instability upon physical examination. Differences in improvement were not significant (p 0.57)
between the preoperatively and postoperatively obtained variables.
According to the Modified Mayo Wrist score, 1
patient rated as good result (75 89 points) and 5
patients as a fair outcome (50 74 points). There were
no patients scoring as a bad result (less than 50
points). The average Modified Mayo Wrist score was
68 points of 100 (range, 65 80).
When using the Modified Gartland and Werley
score, 3 patients scored as good results (3 8 points)
and 3 as fair (9 20 points). There were no poor
results (more than 21 points). The average Gartland
and Werley score was 10 points, ranging from 3 to 18
Figure 5. Final result after stable angle implants fixation and

defect filling with Norian cement.
981
Table 2. Functional Series Outcome After Corrective Osteotomy With Fixed-Angle Implants and Norian
Cement
Case
Follow-up
(Months)
1
2
14
16
3
4
5
6
Avg.
32
22
6
6
16
Complications
Mayo
None
Needed Darrach
procedure
None
None
None
None
65
65
65
65
70
80
68 (r, 6580;
SD, 6)
Mayo
Gartland
and Werley
Fair
Fair
Fair
Fair
Fair
Good
15
9
18
8
5
3
10 (r, 318;
SD, 6)
Gartland and
Werley Score
Fair
Fair
Fair
Good
Good
Good
DASH
29.1
41
25.92
38
17
15.3
28 (r, 15.341;
SD, 15)
Patients listed in boldface type are the patients that received intra-articular osteotomy.
points. The lower average score in the Modified

Mayo Wrist score reflects the stricter nature of this
evaluation instrument. The average DASH score was
28 points, ranging from 15 to 41 (Table 2).
One patient required a Darrach procedure at 6
months to increase motion and management of pain
at the distal radioulnar joint. Two patients required
plate removal due to pain and limitation in movement.
In terms of x-ray evaluation, the average preoperative volar tilt was 19 in extension in the dorsal
deformity group and 25 in flexion in the volar deformity group. Postoperatively, the average palmar
tilt was 13 in both groups. The average improvement after surgery was 32 in patients with dorsal
deformity and 13 in those with volar deformity.
Preoperative ulnar variance was on average 4 mm;
after surgery it corrected to 2 mm (50%). Ulnar
inclination averaged 15 preoperatively; after treatment it averaged 22 presenting an improvement of
7. Lastly, with the exception of 1 patient, restoration
after surgery achieved acceptable clinical outcomes
and radiologic parameters. Radiologic improvement
from the preoperative to the postoperative status was
substantial but statistically not significant (p 0.17).
Assessment for post-traumatic arthritis in the shortterm (less than 2 years) was negative in every case at
the time of follow-up.
Discussion
A corrective osteotomy in the older patient is more
difficult because of the associated osteopenia as well
as the limited autogenous bone graft to be obtained
from the iliac crest. Several technological advances
have made this procedure more predictable. The first
is the development of low-profile implants with angular stable screw fixation. This osteosynthesis sys-
tem device has shown good results in maxillofacial

and spine surgery, where stability is required without
bicortical screw purchase.11,33,34
The locking compression system offers a similar
mechanism of action with the mechanical advantage
of multiple points of screw fixation when compared
with fixed-angle devices. It is a point of crucial
importance in fractures with long working lengths,
short periarticular fragments, and the absence of osseous support on the contralateral side from where
the plate is placed.11 The fixed-angle constructs do
not affect the blood supply to the bone and do not
require good bone quality to render stability.3538 In
this system, threads on the screw heads lock into the
corresponding threads on the screw hole of the plate,
therefore eliminating toggling. Forces are transmitted
then from the bone to the threaded connection between the screws and the plate, converting compression unnecessary to get stability.3538 This lack of
compression preserves the blood supply to the bone
improving conditions for healing.3538 Disadvantages of this system include no tactile feedback to the
surgeon while tightening the screws. Previous reduction is needed before application of the device; once
the locked screw is placed below or above the fracture site, no further reduction is possible unless the
construct is totally removed.10,16,17 Clinical trials
have verified the efficacy of fixed-angle plates for the
treatment of distal radius fractures.10,16,17 Functional
outcomes are promising and the rate of complications
low, making this implant desirable also for the stabilization of osteotomies for the treatment of distal
radius malunion.
The second technical advance is use of cement and
biomaterials that can support and put together fragments of bone and that can fill defects after severe
982
comminution or osteotomies. The role of this material is particularly important in osteoporotic bone that
cannot adequately tolerate constructs and therefore
needs support while consolidation process takes
place. Norian SRS cement offers biocompatibility
and osteointegration; high compressive strength,
even higher than cancellous bone; fast setting that
cures in vivo at physiologic pH and temperature
avoiding local damage tissue characteristic of polymethyl methacrylate use; and injectable consistency
that allows percutaneous or open techniques usage.19 23 Additionally, advantages in imaging under
fluoroscopy and x-rays have been proved.25,39
Some studies have demonstrated better clinical
outcomes with Norian when compared with standard
treatment protocols; however, controversies exist in
terms of radiologic outcomes after both types of
treatment, conventional open reduction and internal
fixation versus percutaneous fixation and Norian cement use.19,22 Two prospective randomized studies
(Sanchez-Sotelo et al19 and Cassidy et al22) evaluating this technique showed good results. Clinical outcomes were significantly better than the standard care
(p 0.01 and p 0.05, respectively) however, none
of them defined what type of fractures get benefit
from this particular approach with cement and percutaneous fixation. Recent research has demonstrated
comparable results between percutaneous fixation
and open reduction and internal fixation for extraarticular and noncomplex intra-articular fractures;40
therefore, it is plausible that the role of Norian cement in distal radius fractures in previous studies is
related to bone quality, in other words, osteoporosis
and osteopenia.
According to the mechanism of action and previous description of these 2 surgical advances, we
consider them extremely useful for the treatment of
malunions in osteopenic patients that suffered a distal
radius fracture resulting in malunion. According to
the clinical and radiologic outcomes, we find this
technique useful and safe to treat malunions.
Norian cement has been used successfully in previous reports testing its utility in corrective osteotomies of
the distal radius. Luchetti in 200426 reported his experience with 6 patients treated with corrective osteotomies of the distal radius and K-wires fixation and calcium phosphate bone cement use instead of bone
grafting. Yasuda et al27 reported their experience in a
case report of a 37-year-old patient treated with an early
corrective osteotomy for a malunited Colles fracture
using a volar plate and calcium phosphate bone cement.
Both studies reported satisfactory results.
This series reports the results of treatment of distal

radius malunion with osteotomy plus internal fixation
with angular-stable locking compression plates and
Norian SRS in 6 patients. Our purpose is to present
this technique as an alternative in these complex
cases where we have to face the elderly patient with
osteoporotic bone. We are aware of the statistical
limitations of this study, but we consider it to be a
valuable surgical technique.
Received for publication February 7, 2007; accepted in revised form May
4, 2007.
article.
Corresponding author: Santiago Lozano-Caldern, MD, Hand and
Upper Extremity Service, Massachusetts General Hospital, YAW 2100,
55 Fruit Street, Boston, MA 02114; e-mail: slozanocalderon@
partners.org.
0363-5023/07/32A07-0005$32.00/0
doi:10.1016/j.jhsa.2007.05.005
References
1. Fernandez DL, Ring D, Jupiter JB. Surgical management of
delayed union and nonunion of distal radius fractures.
J Hand Surg 2001;26A:201209.
2. Jupiter JBF, Fernandez DL. Complications following distal
radial fractures. J Bone Joint Surg 2001;83A:1244 1265.
3. Kurz LT, Garfin SR, Booth RE Jr. Harvesting autogenous
iliac bone grafts. A review of complications and techniques.
Spine 1989;14(12):1324 1331.
4. Arrington ED, Smith WJ, Chambers HG, Bucknell AL,
Davino NA. Complications of iliac crest bone graft harvesting. Clin Orthop Relat Res 1996;329:300 309.
5. Goulet JA, Senunas LE, DeSilva GL, Greenfield ML. Autogenous iliac crest bone graft. Complications and functional
assessment. Clin Orthop Relat Res 1997;339:76 81.
6. Rajan GP, Fornaro J, Trentz O, Zellweger R. Cancellous
allograft versus autologous bone grafting for repair of comminuted distal radius fractures: a prospective, randomized
trial. J Trauma 2006;60(6):13221329.
7. Hill NM, Horne JG, Devane PA. Donor site morbidity in the
iliac crest bone graft. Aust N Z J Surg 1999;69(10):726
728.
8. Silber JS, Anderson DG, Daffner SD, Brislin BT, Leland
JM, Hilibrand AS, et al. Donor site morbidity after anterior
iliac crest bone harvest for single-level anterior cervical
discectomy and fusion. Spine 2003;28(2):134 139.
9. Laurie SW, Kaban LB, Mulliken JB, Murray JE. Donor-site
morbidity after harvesting rib and iliac bone. Plast Reconstr
Surg 1984;73(6):933938.
10. Orbay JL, Fernandez DL. Volar fixed-angle plate fixation for
unstable distal radius fractures in the elderly patient. J Hand
Surg 2004;29A:96 102.
11. Haidukewych GJ. Innovations in locking plate technology.
J Am Acad Orthop Surg 2004;12(4):205212.
12. Gesensway D, Putnam MD, Mente PL, Lewis JL. Design
and biomechanics of a plate for the distal radius. J Hand
Surg 1995;20A:10211027.
13. Jupiter JB. Complex articular fractures of the distal radius:
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
classification and management. J Am Acad Orthop Surg

1997;5(3):119 129.
Simic PM, Weiland AJ. Fractures of the distal aspect of the
radius: changes in treatment over the past two decades. Instr
Course Lect 2003;52:185195.
Trumble T, Culp RW, Hanel DP, Geissler WB, Berger RA.
Intraarticular fractures of the distal aspect of the radius. Inst
Course Lect 1999;48:465 480.
Orbay J. Volar plate fixation of distal radius fractures. Hand
Clin 2005;21(3):347354.
Orbay JL. The treatment of unstable distal radius fractures
Orbay JL, Touhami A, Orbay C. Fixed angle fixation of
distal radius fractures through a minimally invasive approach. Tech Hand Up Extrem Surg 2005;9(3):142148.
Sanchez-Sotelo J, Munuera L, Madero R. Treatment of
fractures of the distal radius with a remodellable bone
cement: a prospective, randomised study using Norian SRS.
J Bone Joint Surg Br 2000;82B:856 863.
Constantz BR, Ison IC, Fulmer MT, Poser RD, Smith ST,
VanWagoner M, et al. Skeletal repair by in situ formation of the
mineral phase of bone. Science 1995;267(5205):1796 1799.
Larsson S, Bauer TW. Use of injectable calcium phosphate
cement for fracture fixation: a review. Clin Orthop 2002;
(395):2332.
Cassidy C, Jupiter JB, Cohen M, Delli-Santi M, Fennell C,
Leinberry C, et al. Norian SRS cement compared with conventional fixation in distal radial fractures. A randomized
study. J Bone Joint Surg Am 2003;85A:21272137.
De Long WG Jr, Einhorn TA, Koval K, McKee M, Smith W,
Sanders R, et al. Bone grafts and bone graft substitutes in
orthopaedic trauma surgery. A critical analysis. J Bone Joint
Surg Am 2007;89A:649 658.
Cohen MS, Whitman K. Calcium phosphate bone cement
the Norian skeletal repair system in orthopedic surgery.
AORN J 1997;65(5):958 962.
Kopylov P, Jonsson K, Thorngren KG, Aspenberg P. Injectable calcium phosphate in the treatment of distal radial
fractures. J Hand Surg 1996;21B:768 771.
Luchetti R. Corrective osteotomy of malunited distal radius
fractures using carbonated hydroxyapatite as an alternative
to autogenous bone grafting. J Hand Surg 2004;2A9:825
834.
Yasuda M, Masada K, Iwakiri K, Takeuchi E. Early corrective osteotomy for a malunited Colles fracture using volar
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
983
approach and calcium phosphate bone cement: a case report.

J Hand Surg 2004;29A:1139 1142.
Kreder HJ, Hanel DP, McKee M, Jupiter J, McGillivary G,
Swiontkowski MF. Consistency of AO fracture classification
for the distal radius. J Bone Joint Surg Br 1996;78B:726
731.
Jupiter JB, Fernandez DL. Comparative classification for
fractures of the distal end of the radius. J Hand Surg 1997;
22A:563571.
Henry AK. Extensile exposure. Second ed. Edinburgh and
London: Churchill Livingstone, 1973.
Herzberg G, Comtet JJ, Linscheid RL, Amadio PC, Cooney
WP, Stalder J. Perilunate dislocations and fracturedislocations: a multicenter study. J Hand Surg 1993;18A:
768 779.
Sarmiento A, Pratt GW, Berry NC, Sinclair WF. Colles
fractures. Functional bracing in supination. J Bone Joint
Surg Am 1975;57A:311317.
Herford ASEI, Edward. Use of a locking reconstrucion bone
plate/screw system for mandibular surgery. J Oral Maxillofac Surg 1998;56:12611265.
Richter M, Wilke HJ, Kluger P, Claes L, Puhl W. Biomechanical evaluation of a newly developed monocortical expansion screw for use in anterior internal fixation of the
cervical spine. In vitro comparison with two established
internal fixation systems. Spine 1999;24(3):207212.
Stoffel K, Klaue K, Perren SM. Functional load of plates in
fracture fixation in vivo and its correlate in bone healing.
Injury 2000;31(Suppl 2):S-B3750.
Thielke KH, Wagner T, Bartsch S, Echtermeyer V. [Angularly stable radius plate: progress in treatment of problematic
distal radius fracture?] Chirurg 2003;74(11):10571063.
Wagner M. General principles for the clinical use of the
LCP. Injury 2003;34(Suppl 2):B31 42.
Zura RD, Browne JA. Current concepts in locked plating.
J Surg Orthop Adv 2006;15(3):173176.
Jupiter JB, Winters S, Sigman S, Lowe C, Pappas C, Ladd
AL, et al. Repair of five distal radius fractures with an
investigational cancellous bone cement: a preliminary report. J Orthop Trauma 1997;11(2):110 116.
Kreder HJ, Hanel DP, Agel J, McKee M, Schemitsch EH,
Trumble TE, et al. Indirect reduction and percutaneous fixation
versus open reduction and internal fixation for displaced intraarticular fractures of the distal radius: a randomised, controlled
trial. J Bone Joint Surg Br 2005;87B:829 836.
Rupture of the Flexor Pollicis Longus

Tendon After Volar Fixed-Angle Plating of
a Distal Radius Fracture: A Case Report
Raymond A. Klug, MD, Cyrus M. Press, MD, Mark H. Gonzalez, MD
From the Department of Orthopaedics, Mount Sinai Medical Center, New York, NY; Department of Orthopaedics, University of Illinois, Chicago, IL; and Department of Orthopaedics, John H. Stroger, Jr., Hospital
of Cook County, Chicago, IL.
We report a case of complete rupture of the flexor pollicis longus tendon 13 months after
volar fixed-angle plating of a distal radius fracture. Tendon disruption was associated with a
prominent distal volar lip of the plate. The plate was placed at the volar distal lip of the radius,
at the location recommended by the manufacturer. Most previous reports of flexor tendon
ruptures after volar plating of distal radius fractures have been in improperly placed plates,
custom-made plates that were later taken off the market, or in physiologically abnormal
tendons. This may be a unique case of flexor pollicis longus rupture with a currently
commercially available volar fixed-angle plate, placed at the site recommended by the
manufacturer, in a patient without other predisposition to tendon rupture. (J Hand Surg 2007;
32A:984 988. Copyright 2007 by the American Society for Surgery of the Hand.)
Key words: Complications, distal radius fracture, flexor pollicus longus (FPL), tendon, volar
plate.
endon injury is an uncommon complication of

distal radius fractures, ranging from irritation
to frank rupture.1 Injuries have been reported
to both flexor and extensor tendons in nonoperatively
treated1 6 as well as operatively treated79 patients.
Extensor tendon damage is thought to be more common than flexor tendon damage in patients treated
surgically.4,5,10 Injury to the flexor pollicis longus
(FPL) tendon after volar plating is rare. Our research
found only 12 such cases in the English language.
However, these are largely due to improperly
placed11,12 or poorly designed plates13 or in previously injured tendons.14,15 The fact that this patient
suffered a rupture of a previously healthy tendon and
was treated with a commercially available plate that
was placed per the manufacturers specifications is
what we believe makes this case unique.
Case Report
A 59-year-old, right-handed female had a fall onto an
outstretched arm and sustained a comminuted intraarticular fracture of the left distal radius (Fig. 1). She
had successful volar fixed-angle plating of the fracture at an outside institution using a Synthes Titanium Volar Distal Radius Plate (item no. 442483;
984
Synthes, Paoli, PA) and did well in the immediate

postoperative period (Fig. 2). The plate was placed
on the volar distal lip of the radius, per the manufacturers recommendation. The patient sought further
care at our institution for a nonmetastatic squamous
cell carcinoma of the tongue and was subsequently
followed up in our orthopedic hand surgery clinic.
She was first seen in our clinic approximately 11
months after the initial surgery on her wrist, at which
time she reported pain and weakness in the left wrist
and thenar area and pain with active flexion of
the thumb interphalangeal joint. Approximately 2
months later she developed complete inability to
actively flex the interphalangeal joint of the left
thumb. Her medical history included chronic obstructive pulmonary disease and hypertension, and
her medications included alendronate, trandolapril/
verapamil, salmeterol, and tiotropium. She had a 45
pack per year smoking history but quit using tobacco
and ethanol approximately 15 years prior to presentation. Of note, the patient did not then or previously
use inhaled, oral, or intravenous corticosteroids. The
carcinoma of the tongue was treated with local resection, without systemic chemotherapy.
Klug, Press, and Gonzalez / FPL Rupture After Volar Plating of Distal Radius Fracture
985
against the radius without appreciable gapping. All

hardware was removed. A tenolysis was performed,
and the FPL pseudotendon was excised. The tendon
was repaired using an ipsilateral palmaris longus
graft. Fraying of the flexor digitorum profundus to
the index finger was also noted. However, this was
minimally involved, and because the patient was
asymptomatic regarding this tendon, it was not
repaired.
Postoperatively, the patient was placed in a thumb
spica splint and started on our standard FPL repair
protocol with our occupational therapist. She progressed well with supervised hand therapy, and the
remainder of her postoperative course was unremarkable. At her last evaluation approximately 1 year
after tendon grafting, the patient had regained MRC
Grade 4 pinch strength, had functional active range
of motion of the thumb interphalangeal and metacarpophalangeal joints, and was completely asymptomatic regarding the remainder of her ipsilateral finger
flexors.
Figure 1. Lateral x-ray of left wrist showing a comminuted,
intra-articular, dorsally displaced fracture of the distal radius.
Her neurovascular exam was unremarkable in the

left upper extremity, and she had full passive but no
active flexion of the thumb interphalangeal joint. Her
strength was normal in adduction, opposition, and
abduction of the left thumb as well as flexion and
extension of the remainder of the ipsilateral fingers.
Although rupture of the FPL tendon was suspected,
an electromyogram with nerve conduction velocity
studies was ordered to rule out anterior interosseous
nerve palsy, and this test was normal. Radiographic
evaluation showed a healed distal radius fracture in
good position. The hardware appeared in place without signs of loosening, and the position of the volar
plate was on the prominent distal volar lip of the
distal radius. The patient was returned to the operating room to address the suspected FPL rupture.
The previous flexor carpi radialis approach was
used and extended to include the carpal tunnel. The
wrist was explored and complete rupture of the FPL
tendon was confirmed. The proximal and distal
stumps of the tendon were identified. The proximal
stump was retracted approximately 4 cm while the
position of the distal stump was found to correspond
with the level of the prominent distal volar lip of the
plate. There were no loose or prominent screws
noted, and the plate was found to be well seated
Discussion
Tendon complications of distal radius fractures are
well-known and described in the literature.13 The
majority of cases involve the extensor tendons of the
digits, most commonly the extensor pollicis longus.4,5 These complications may occur with displaced
or nondisplaced fractures treated nonoperatively, and
those nondisplaced fractures are presumed to be due to
increased pressure within an intact fibrous sheath that
inhibits blood flow or synovial fluid production.6 Tendon rupture after dorsal plating of distal radius fractures has also been well described. Jakob et al reported on 73 consecutive patients with 74 fractures.
Eight required plate removal for extensor tendon
complications3 due to tendon irritation and 5 with
complete rupture.7 Prior to this, Carter et al8 reported
use of an anatomically preshaped, low-profile dorsal
plate that was specifically designed to decrease extensor tendon complications in 71 patients (73 fractures). Although there were no tendon ruptures, 8
(11%) patients required plate removal due to extensor tendon irritation. Rarely, flexor tendon complications have occurred after dorsal plating with excessively long screws.9
Flexor tendon complications after distal radius
fractures are much less common2,10 16 and may occur
early with fracture displacement17 later after malunion
with bony prominence causing mechanical irritation4,18,19 or in physiologically abnormal tendons.3,14
Unlike the extensor tendons, the flexor tendons are
986
Figure 2. (A) Anteroposterior (AP) and (B) lateral views of the wrist with volar fixed-angle plate in place. Percutaneous K-wires
were removed at 6 weeks postoperatively.
not intimately associated with bone at the metadiaphyseal level of the distal radius and are less likely
to be damaged after fracture at this level.10 Two
factors contribute to this. First, the pronator quadratus muscle overlies the volar surface of the distal
radius and as such provides protection of the flexor
tendons from deeper structures such as prominent
bony fragments5 or hardware.14 Second, the normal
prominence of the volar lip in combination with the
bony anatomy of the pronator fossa creates a slight
bowstringing over this area, thus causing the resting
position of the flexor tendons to lie away from the
volar cortex.20 Both of these conditions create a
natural concavity to the volar surface of the distal
radius immediately proximal to the distal volar lip.
This concavity houses the pronator quadratus muscle
and allows for low-profile placement of volar plates
for distal radius fractures and plate coverage by the
pronator muscle. More distally, the pronator quadratus ceases to cover the distal radius, and the volar
flare of the epiphysis at the watershed area brings the
bony architecture in close apposition to the flexor
tendons. Volar plates may be placed at this level as
well. However, tendon irritation may be more common with more distal plate placement, as in this
location the plate may be more intimately associated
with the flexor tendons, especially in extension of the

wrist. Appropriate placement of a volar plate for
fixation of a distal radius fracture is dictated by the
fracture pattern, the plate design, and the manufacturers recommendation on its placement. In our patient, the plate was placed, per the manufacturers
recommendation, on the prominent distal volar lip of
the radius. Several locations for plate placement exist
and appropriate placement of any given plate is dictated by the manufacturers recommendations, which
can aid in the appropriate choice of implants. Use of
a lower profile plate designed for more proximal
placement into the volar concavity of the distal radius
may decrease the possibility of this complication in
the future.15,20
To our knowledge, only 12 cases of FPL tendon
rupture after volar plating of distal radius fractures
have been reported in the English literature.1115
Previous reports have described incorrect positioning of
plates,11,12 prominent screw heads with sharp edges,13
and/or pathophysiologic predisposition to rupture in
previously weakened tendons.14 In 1973, Fuller described use of an Ellis nonlocking volar butt plate for
displaced Smiths fractures.12 Of 31 patients, 1 patient in whom the plate was placed obliquely to
support a large radial fragment had delayed rupture
Klug, Press, and Gonzalez / FPL Rupture After Volar Plating of Distal Radius Fracture
of the FPL tendon 5 years after surgery. At the time

of tendon repair, it was noted that the tendon had
undergone attrition at the free edge of the plate. More
recently, Nunley and Rowan11 described FPL rupture
after inappropriate placement of a dorsal plate on the
volar surface of the distal radius. In 1998, Bell et al14
described 3 patients with complete rupture of the
FPL tendon after volar buttress plating of distal radius fractures. Two of these patients were on chronic
corticosteroid therapy and the other had previously
been treated with corticosteroids for pituitary insufficiency. All patients presented in delayed fashion
between 4 and 10 months after plate osteosynthesis,
and all cases were associated with prominent distal
lips of the respective plates. The authors concluded
that FPL rupture after volar plating of distal radius
fractures may occur in patients with tendons physiologically predisposed to rupture and recommended
early plate removal if the distal edge of the plate is,
or later becomes, prominent as a result of fracture
collapse. Although our patient had a considerable
smoking history, we do not believe that this contributed to tendon rupture as she quit using tobacco
products approximately 15 years prior to presentation.
With the advent of volar fixed-angle plates, dorsally unstable fractures could be addressed from a
volar approach, and extensor tendon complications
were reduced.21 With modern plating techniques,
only sporadic cases of flexor tendon complications
have appeared. In 2005, Douthit15 reported on 51
displaced fractures of the distal radius treated with
volar fixed-angle plates in 50 patients. Of these, 1
patient had an FPL rupture after 18 months. Interestingly, the authors reported this patients surgical result as excellent with no malunion, and his functional
result was graded as good. The patient refused surgery to address the tendon dysfunction. As such,
investigation into the quality of the tendon, its immediate environment, or the exact location of dysfunction was not possible. The authors did, however,
note flexor tendon irritation without rupture in an
additional patient and stated that their choice of plate
had changed as a result of these complications. After
noting this complication, the authors opted for larger
plates with smaller screw heads in an attempt to
decrease flexor tendon irritation. No follow-up was
given regarding this second group of patients. In a
later report, Orbay and Touhami20 stressed the importance of proper plate position and stated that loss
of reduction with reconstitution of dorsal deformity
may cause lift-off of the plate into the flexor tendons.
987
In this case, immediate reoperation was recommended to avoid flexor tendon rupture.
An additional report is that of Drobetz and
Kutscha-Lissberg,13 who reported their experience
with a custom-made fixed-angle volar plate of their
own design. They found FPL rupture in 6 of 50
(12%) fractures in their early series. The authors
believed that this was due to prominent screw heads
with sharp edges or plates positioned too far distally
and initially recommended plate removal on a routine
basis. They as well changed plate designs after noticing this complication, and the previously used
plate was removed from the commercial market.
Since changing plate designs, the authors have no
longer experienced FPL ruptures and no longer remove this hardware routinely.
Few reports have described rupture of the FPL
tendon after operative treatment of distal radius fractures. Many of these reports have involved oldergeneration plates, inappropriately placed plates,
physiologically abnormal tendons, or custom plates
that are not commercially available. To our knowledge, there are no documented cases of rupture of a
physiologically normal FPL tendon after placement
of a currently commercially available volar fixedangle locking plate in the location recommended by
the manufacturer. In operatively treated fractures,
tendon rupture may be predisposed in physiologically abnormal tendons, with improperly placed
plates, with prominence of the plate and/or screws or
with sharp edges to the plate and/or screws. All of
these factors should be considered in plate design and
implant choice prior to addressing this common fracture pattern. Loss of reduction may also cause prominence of an otherwise benign volar plate. In this
case, reoperation should be considered to avoid
flexor tendon complications.
Received for publication June 30, 2006; accepted in revised form May 4,
2007.
article.
Corresponding author: Mark H. Gonzalez, MD, 835 S. Wolcott Avenue, M/C 844, Chicago, IL 60612; e-mail: hand15@aol.com.
0363-5023/07/32A07-0006$32.00/0
doi:10.1016/j.jhsa.2007.05.006
References
1. Nana AD, Joshi A, Lichtman DM. Plating of the distal
radius. J Am Acad Orthop Surg 2005;13:159 171.
2. Ashall G. Flexor pollicis longus rupture after fracture of the
distal radius. Injury 1991;22(2):153155.
3. Roberts JO, Regan PJ, Roberts AH. Rupture of flexor pol-
988
4.
5.
6.
7.
8.
9.
10.
11.

licis longus as a complication of Colles fracture: a case
report. J Hand Surg 1990;15B:370 372.
Kato N, Nemoto K, Arino H, Ichikawa T, Fujikawa K.
Ruptures of flexor tendons at the wrist as a complication of
fracture of the distal radius. Scand J Plast Reconstr Surg
Hand Surg 2002;36(4):245248.
McMaster PE. Late problems of extensor and flexor pollicis
longus tendons following Colles fracture. J Bone Joint Surg
1932;14:93101.
Engkvist O, Lundburg G. Rupture of the extensor pollicis
longus tendon after fracture of the lower end of the radiusa
clinical and microangiographic study. Hand 1979;11(1):76 86.
Jakob M, Rikli DA, Regazzoni P. Fractures of the distal
radius treated by internal fixation and early function. J Bone
Joint Surg [Br] 2000;82B:340 344.
Carter PR, Fredrick HA, Laseter GF. Open reduction and
internal fixation of unstable distal radius fractures with a low
profile plate: a multicenter study of 73 fractures. J Hand Surg
1998;23A:300 307.
Krukhaug Y, Hove LM. Experience with the AO Plate for
displaced intra-articular fractures of the distal radius. Scand
J Plast Reconstr Surg Hand Surg 2004;38:293296.
Smith DW, Henry MH. Volar fixed-angle plating of the
distal radius. J Am Acad Orthop Surg 2005;13(1):28 36.
Nunley JA, Rowan PR. Delayed rupture of the flexor pollicis
longus tendon after inappropriate placement of the pi plate
on the volar surface of the distal radius. J Hand Surg 1999;
24A:1279 1280.
12. Fuller DJ. The Ellis plate operation for Smiths fracture.
J Bone Joint Surg 1973;55B:173178.
13. Drobetz H, Kutscha-Lissberg E. Osteosynthesis of distal
radius fractures with a volar locking screw plate system. Int
Orthop 2003;27(1):1 6.
14. Bell JS, Wollstein R, Citron ND. Rupture of flexor pollicis
longus tendon: a complication of volar plating of the distal
radius. J Bone Joint Surg 1998;80B:225226.
15. Douthit JD. Volar plating of dorsally comminuted fractures
of the distal radius: a 6-year study. Am J Orthop 2005;34(3):
140 147.
16. Akita S, Kawai H. Entrapment of the flexor digitorum superficialis in the radius fracture site. J Hand Surg 2005;30A:
308 311.
17. Cooney WP, Dobyns JH, Linscheid RL. Complications of
Colles fractures. J Bone Joint Surg 62A;4:613 619.
18. Lamas C, Proubasta I, Itarte J, Piero A, Majo J. Rupture of
all the flexor tendons in the hand due to malunion of a distal
radius fracture. Chir Main 2004;23(1):45 48.
19. Murase T, Hiroshima K. Rupture of the flexor tendon after
malunited Colles fracture. Scand J Plast Reconstr Surg
Hand Surg 2003;37(3):188 191.
20. Orbay JL, Touhami A. Current concepts in volar fixed-angle
fixation of unstable distal radius fractures. Clin Orthop Relat
Res 2006;445:58 67.
21. Orbay JL. The treatment of unstable distal radius fractures
Transfer of the Accessory Nerve to the

Suprascapular Nerve in Brachial Plexus
Reconstruction
Jayme Augusto Bertelli, MD, PhD, Marcos Flvio Ghizoni, MD
From the Department of Orthopedic Surgery, Governador Celso Ramos Hospital, Florianpolis, SC, Brazil;
and the Center of Biological and Health Sciences, University of Southern Santa Catarina (Unisul), Tubaro,
SC, Brazil.
Purpose: Transfer of the accessory nerve to the suprascapular nerve is a common procedure,
performed to reestablish shoulder motion in patients with brachial plexus palsy. We propose
dissecting both nerves via a distal oblique supraclavicular incision, which can be prolonged
up to the scapular notch. The results of the transfer to the suprascapular nerve are compared
with those of the combined repair of the suprascapular and axillary nerves.
Methods: Thirty men between the ages of 18 and 37 years with brachial plexus trauma had
reparative surgery within 3 to 10 months of their injuries. In partial injuries with a normal
triceps, a triceps motor branch transfer to the axillary nerve was performed. The suprascapular and accessory nerves were dissected via an oblique incision, extending from the point
at which the plexus crosses the clavicle to the anterior border of the trapezius muscle. In 10
patients with fractures or dislocations of the clavicle, the trapezius muscle was partially
elevated to expose the suprascapular nerve at the suprascapular notch.
Results: In all cases, transfer of the accessory to the suprascapular nerve was performed
without the need for nerve grafts. A double lesion of the suprascapular nerve was identified
in 1 patient with clavicular dislocation. In those with total palsy, the average improvement in
range of abduction was 45, but none of the patients with total palsy recovered any active
external rotation. Patients with upper-type injury recovered an average of 105 of abduction
and external rotation. If only patients with C5-C6 injuries were considered, the range of
abduction and external rotation increased to 122 and 118, respectively.
Conclusions: Use of the accessory nerve for transfer to the suprascapular nerve ensured adequate
return of shoulder function, especially when combined with a triceps motor branch transfer to the
axillary nerve. The supraclavicular exposure proposed here for the suprascapular and accessory
nerves is advantageous and can be extended easily to explore the suprascapular nerve at the
scapular notch. (J Hand Surg 2007;32A:989 998. Copyright 2007 by the American Society for
Surgery of the Hand.)
Key words: Accessory nerve, brachial plexus, nerve grafting, nerve transfer, suprascapular
nerve, supraspinatus muscle.
urgical reconstruction of the suprascapular nerve

is a high priority in brachial plexus injuries. This
is not only because functional control of the
shoulder is of paramount importance but also because
of the overall reduced success of reinnervation of muscles distal to the elbow. Therefore, in total palsy,
healthy roots and nerve transfers are preferentially used
to reconstruct shoulder girdle muscles because of their
greater potential for reinnervation when compared with

muscles distal to the elbow.
In the past, better results have been obtained
when the suprascapular nerve has been reconstructed via the transfer of the accessory nerve
versus grafts from the ruptured C5 root.1 Accessory nerve transfer to the suprascapular nerve now
is a standard procedure in brachial plexus repair.2 In
989
990
upper-type partial injuries of the brachial plexus,

hand function is spared, and shoulder reconstruction
should be maximized not only by suprascapular
nerve but also by axillary nerve reconstruction.35
Our preference is to use a triceps motor branch to
transfer to the axillary nerve.5,6
The suprascapular nerve routinely has been explored through an L type of incision in the supraclavicular region.7 Using this same access, the accessory nerve has been dissected proximally and tracked
distally.8 Recently, Bertelli and Ghizoni 9 introduced
a method of accessory nerve dissection close to the
clavicle, under the deep cervical fascia, which facilitates surgery and avoids use of nerve grafts. We now
propose also dissecting the suprascapular nerve distally by means of a separate oblique incision close to
the site of the accessory nerve dissection.
Some have proposed using the omohyoid muscle
as a landmark and digital palpation as a tool to locate
the suprascapular nerve.4 This is useful in the majority of cases. In brachial plexus traction injuries, however, distal migration of the origin of the suprascapular nerve can occur. If a clavicular fracture has
healed accompanied by exorbitant callus formation,
suprascapular nerve dissection can be difficult and
dangerous. Moreover, the suprascapular nerve can be
double injured in a scapular neck fracture, so that the
scapular notch should be explored. Indeed, suprascapular nerve lesions after clavicular fractures have
been reported.10 For these complicated cases, we
propose that surgeons extend the oblique incision, so
that they can detach the trapezius muscle from the
clavicle and explore the suprascapular nerve at the
scapular notch.
In the current paper, we describe our method of
suprascapular and accessory nerve dissection and
present the results of accessory to suprascapular
nerve transfer in 30 consecutive cases of brachial
plexus injury. We also compare the extent of recovery obtained in those with complete versus partial
palsies.

Patients
Thirty men with brachial plexus trauma who required
brachial plexus reconstructions were included in the
current study. Criteria for inclusion were age between 18 and 37 years, and surgery within 3 and 10
months of the initial injury. Patients with complete
avulsion injuries (ie, who had no graftable roots)
were excluded because, in these cases, we preferred
to transfer the accessory nerve to the musculocuta-
neous nerve. Also, patients with concomitant accessory nerve injury were excluded.
Fourteen patients had partial injuries with resultant
paralysis of shoulder abduction and external rotation
and elbow flexion. Among these 14 patients, 8 had
C5 and C6 injuries alone, whereas 6 also had involvement of the C7 root, producing triceps weakness. Twelve of the 14 had preserved hand function.
Two patients also had wrist and digital extension
palsy. The remaining 16 of 30 patients had total
palsy. Trapezius strength, evaluated by resisting
shrugging the shoulder, was normal in all patients.
Among those with complete palsy, mean patient
age and the time between the injury and surgery were
24 years (range, 18 37 years) and 5 months (range,
310 months), respectively. Among those with partial injuries, mean age and time between trauma and
surgery were 26 years (range, 18 36 years) and 6
months (range, 310 months), respectively.
All patients had surgical repair in the supine position under general anesthesia. In 20 patients, transfer
to the suprascapular nerve was performed through an
oblique supraclavicular incision. We used the extended approach in 8 patients with complete palsy
and in 2 with partial injury. Among these 10 patients,
we identified 4 clavicular fractures associated with
exuberant callous formation, 3 scapular neck fractures, 1 extended lesion of the suprascapular nerve, 1
patient with supraclavicular scarring, and 1 patient
with acromio-clavicular dislocation. In 7 of the 10
patients, the previous oblique supraclavicular incision was extended; and in the final 3 patients, dissection began at the suprascapular notch and continued proximally.
Brachial Plexus Exploration and Additional
Transfers
The roots of the brachial plexus were explored via an
oblique incision that was made slightly lateral to
Chassaignacs tubercle. Viable roots were grafted
either to the anterior or posterior division of the
upper trunk or to the musculocutaneous nerve in
cases of complete palsy.
In upper-type partial injuries, elbow flexion was
reconstructed by transferring ulnar nerve fascicles to
the biceps motor branch; whereas a motor branch of
the triceps muscle was transferred to the teres minor
motor branch and to the anterior and middle deltoid
branches of the axillary nerve, as reported elsewhere.5 Triceps motor branch transfers were performed via posterior arm5 or axillary access.11 In 3
patients with upper-type injuries associated with a
Bertelli and Ghizoni / Transfer of the Accessory Nerve to the Suprascapular Nerve
991
weak triceps (ie, M4 or M3), a branch of the long

thoracic nerve to the upper serratus muscle12 was
transferred to the suprascapular nerve, together with
the accessory nerve.
Surgical Technique
Exposure of the suprascapular nerve by a supraclavicular oblique incision. With the patient in
the supine position and the patients head turned
toward the contralateral side, the lateral limit of the
brachial plexus was palpated in the supraclavicular
region. A point was marked at which the brachial
plexus crosses the clavicle, and then a line over this
point and over a point 3 cm proximally to the clavicle
on the anterior margin of the trapezius muscle was
drawn (Fig. 1A, B). The skin was incised, the subcutaneous tissue was divided, and the omohyoid
muscle was identified and sectioned. The lateral limit
Figure 2. Intraoperative view of the suprascapular nerve (SN)

and the accessory nerve (AN) dissected with an oblique
supraclavicular incision. Between these 2 nerves, note the
deep cervical fascia grasped by the surgical clamp. The roots
of the brachial plexus were explored by means of a separate
incision. The head is to the right and the shoulder is to the
left.
of the brachial plexus was visualized, and the suprascapular nerve was isolated as the sole branch emerging from the brachial plexus at this location (Fig. 2).
The suprascapular nerve was dissected proximally
and then sectioned as shown in Video 1 (this video
may be viewed at the Journals Web site, www.
jhandsurg.org).
Figure 1. Design of the oblique incision to expose the suprascapular nerve. (A) The brachial plexus is palpated and its
junction with the clavicle determined. (B) An oblique line is
passed over the point determined in (A) and the anterior
border of the trapezius muscle. This line crosses the trapezius
muscle 3 cm cephalad to the clavicle.
Extended exposure of the suprascapular

nerve. The patient was placed in the supine position and the head of the operating table was elevated
to 45. Supports were placed under the shoulder. A
zizzag incision over the anterior border of the trapezius muscle and acromion was designed either as a
prolongation of or as separate from the previous
oblique incision (Fig. 3). The skin and subcutaneous
tissue were incised, the deep cervical fascia over the
trapezius muscles anterior border was divided, and the
trapezius muscle was disinserted from the clavicle and
acromion. Care was taken not to divide the fascia
over the supraspinatus muscle. The suprascapular
vessels were located; then the scapular notch was
identified. For this, the surgeons index finger was
introduced to the anterior side of the scapula and
moved from medial to lateral. The finger stopped at
the base of the coracoid process, which is the site of
the scapular notch, located almost directly under the
clavicle. With the help of a surgical clamp, the scapular notch was located and the suprascapular ligament divided. Alternatively, the suprascapular nerve
first was identified and the ligament divided after-
992
Figure 3. Line drawing of the surgical incision for extended

exploration of the suprascapular nerve, which includes partial elevation of the trapezius muscle. The line designed was
a zigzag extension over the anterior border of the trapezius
muscle.
ward. Dissection of the suprascapular nerve was carried out posteriorly to the suprascapular vessels,
which were preserved. After dividing the suprascapular ligament, the branches of the suprascapular
nerve were dissected and inspected for continuity.
The branch for the acromioclavicular joint, which
originates at the scapular notch, was divided. The
suprascapular nerve was tracked proximally and then
sectioned (Video 2; this video may be viewed at the
Journals Web site, www.jhandsurg.org).
Accessory nerve dissection. At the anterior margin of the trapezius muscle, the deep cervical fascia was separated from the trapezius muscle, 3 to 4
cm proximal to the clavicle. The deep cervical
fascia and the trapezius muscles were retracted,
like book pages. The accessory nerve was isolated
within the deep cervical fascia. Three to 4 cm
proximal to the clavicle, the accessory nerve is
more superficial and, therefore, easily identified.
The motor branches entering the trapezius muscle
confirmed the nerves identity. Before severing the
accessory nerve, it was electrically stimulated to
assess its functional integrity. The accessory nerve
was divided, as distal as needed, to ensure direct
coaptation to the suprascapular nerve. When the
trapezius muscle is detached, the accessory nerve
can be dissected up to the upper medial angle of
the scapula. The nerve coaptation was performed
under a microscope, using 9-0 nylon sutures. The
trapezius muscle was reinserted onto the clavicle
and acromion.
Postoperative Care and Evaluation

In the postoperative period, a sling was used, for 3
weeks. Patients with complete palsy had their final
evaluation an average of 27 months (range, 20 48
months) after surgery. Patients with partial injuries
were evaluated for up to a mean of 30 months (range,
18 48 months) after surgery. The extent of recoveryin terms of the return of range of abduction and
external rotationwas measured with a goniometer.
External rotation was measured from full internal
rotation. In partial lesions of the brachial plexus, the
shoulder resting position is not at neutral but is in
complete internal rotation. Any active rotation from
full internal rotation needs the reinnervation of the
external rotator muscles. Therefore, external rotation
was measured with the patient standing up with the
shoulder fully internally rotated and the forearm
placed transversally over the abdomen not to miss
any degree of recovery. Shoulder abduction and external rotation strength were graded based on a minimally modified Narakas scale.5 Shoulder abduction
was graded as follows: M0, no evidence of contractility;
M1, muscle contractions, but no active motion; M2,
abduction less than 60; M3, more than 60 of abduction, and the patient is able to keep the limb abducted at
60 for 10 seconds; M4, abduction to 60 against resistance applied to the elbow; and M5, abduction to 60
against resistance applied to the forearm.
External rotation strength was assessed with the
patient sitting, the arm abducted passively to 90, and
external rotation actively performed. Strength was
graded as follows: M0, no muscle contraction; M1,
muscle contraction, but no active motion; M2, in full
internal rotation, any active external rotation motion;
M3, the patient is capable of maintaining the forearm in
the horizontal position for 10 seconds; M4, active external rotation motion against resistance applied to the
proximal one-third of the forearm; and M5, active external rotation against resistance applied to the wrist.
Statistical Evaluation
The mean range of abduction recovery between the
complete palsy group and those with incomplete
palsy was compared using unpaired t-tests. The same
was done to assess the extent of recovery of abduction and external rotation among patients with partial
injuries with normal or less than normal triceps
strength. Among patients with less than normal triceps strength, the results again were compared between those in whom the suprascapular nerve received transfer from the accessory nerve and those in
whom the suprascapular nerve received transfer from
993
both the accessory nerve and a branch of the long

thoracic nerve.
Statistical significance was considered when p .05.
Results
Suprascapular and Accessory Nerve Dissection
The suprascapular nerve was dissected reliably and
without complications, both with the limited and the
extended approach. A double lesion of the suprascapular nerve was identified in 1 patient, who had
presented with acromio-clavicular dislocation. In this
instance, the C5 and C6 roots were injured and the
suprascapular nerve also was damaged at the suprascapular notch. The accessory nerve was dissected up to
the level of the medial angle of the scapula and trans-
Figure 5. Postoperative view of the typical recovery of abduction after accessory to suprascapular nerve transfer in a
patient with complete palsy, 30 months after surgery.
Figure 4. (A) Intraoperative view of the right suprascapular

nerve (SN), which was injured at the suprascapular notch.
This patient presented with an acromio-clavicular dislocation
that had been fixed previously with K-wires. The nerve was
adhered to the scapular notch and surrounding tissue. The
suprascapular ligament (Sl) was divided. (B) The accessory
nerve (AN) was sutured to the suprascapular nerve (SN),
distal to the suprascapular ligament (Sl). For this, the accessory nerve was dissected proximally to the upper medial
angle of the scapula. Trapezius muscle elevation facilitated
distal dissection of the accessory nerve.
ferred to the suprascapular nerve distal to the scapular

notch, without interposition of a nerve graft (Fig. 4A,
B). In 1 patient, we identified an extended lesion, from
the roots of the brachial plexus to the distal third of the
suprascapular nerve; but the suprascapular nerve was
not affected at the suprascapular notch.
In those patients with scapular fractures, the nerve
was not found to be entrapped at the site of the
fracture. Dissection of the accessory nerve within the
deep cervical fascia was straightforward; and, when
needed, detachment of the trapezius muscle from the
clavicle and acromion ensured a very long nerve
stump with direct coaptation to the suprascapular
nerve in all cases. Upper trapezius strength was preserved in all patients.
Motion Recovery
Patients with complete palsy recovered 30% of the
normal range of abduction. They experienced no
return of external rotation (Fig. 5). Patients with
994
Table 1. Data Summary for Patients With Complete Brachial Plexus Palsy
Range of Recovery
()
Patient
Sex
Age (Years)
Interval (Months)
Postoperative (Months)
Abduction
ER
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
22
19
30
22
18
23
18
18
34
37
19
19
33
27
21
27
3
5
6
7
10
6
8
7
5
3
3
4
5
4
4
4
42
48
24
30
36
36
30
24
30
30
30
20
20
22
20
24
45
60
40
45
80
50
20
60
30
45
50
45
45
45
60
90
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
The column Interval refers to the period between accident and surgery. The column Postoperative refers to the period between surgery
and last evaluation. ER, external rotation; X, no recovery.
motor branch transfers to the axillary nerve recovered better than those with less than normal (ie, M4
or M3) triceps strength. Only the difference in
abduction recovery, however, was statistically significant (p .001). In patients with partial palsy and
weak triceps, improvement among patients who had
combined transfer of the accessory nerve and the
long thoracic nerve branch to the suprascapular nerve
incomplete palsy recovered an average of 61% of the

normal range of abduction and 75% of normal range
of external rotation. Results are summarized in Tables 1, 2, and 3. In partial injury, results for abduction
and external rotation recovery were significantly better (p .001) than those observed in complete palsy.
In the partial palsy group (Figs. 6, 7), patients with
normal (ie, M5) triceps strength who also had triceps
Table 2. Data Summary for Patients With Upper-Type Partial Injury of the Brachial Plexus
Range of Recovery
()
Patient
Age (Years)
Interval (Months)
Postoperative (Months)
Abduction
ER
Triceps/Comments
1
2
3
4
5
6
7
8
9
10
11
12
13
14
35
24
36
18
31
20
21
18
25
24
30
35
22
24
4
4
4
9
4
8
5
6
6
6
5
7
6
5
36
42
30
36
36
24
18
30
20
18
38
24
48
24
170
90
90
70
120
100
90
70
170
90
170
80
90
70
95
140
60
30
140
140
140
140
140
90
90
100
140
30
NT
WT, XI-LT
WT-SW
WT
NT, mild SW
WT, SW, XI-LT
NT
WT, XI-LT
NT
WT, XI-LT, SW
NT
NT
NT
WT
The column Interval refers to the period between accident and surgery. The column Postoperative refers to the period between surgery
and last evaluation. ER, external rotation; NT, normal triceps (ie, M5); WT, weak triceps (ie, M4 or M3); SW, scapular winging; XI-LT, both
the accessory and a branch of the long thoracic nerve were transferred to the suprascapular nerve.
995
Table 3. Overall Results of Transfer to the Suprascapular Nerve in 30 Cases of Brachial Plexus Injury
Abduction
Total palsy
Partial palsy
External Rotation
Range of Motion ()
Strength
Range of Motion ()
Strength
45 (range, 2090)
105 (range, 70170)
3-M3, 13-M2
5-M5, 8-M4, 1-M3
No recovery
105 (range, 30140)
No recovery
7-M5, 4-M4, 3-M3
In the Strength columns, 3-M3 means that in 3 patients, external rotation was graded M3; and so forth. The results in the partial palsy
group were significantly better (p .05).
was better than those with the single transfer to the

suprascapular nerve. This difference was not statistically significant (p .05), however, likely because
of the limited number of patients (Fig. 7). Results for
Figure 6. Postoperative view of a patient who preoperatively had normal triceps strength. The accessory nerve
was transferred to the suprascapular nerve, whereas the
axillary nerve was supplied with a triceps motor branch.
Elbow flexion was reconstructed by transferring ulnar
nerve fascicles to the biceps motor branch. The C5 root
was healthy and was grafted in its entirety to the anterior
division of the upper trunk, with the goal of reinnervating
the brachialis muscle, which effectively occurred. Results
20 months after surgery.
the partial palsy groups with normal and weak triceps

are summarized in Table 4.
Scapular Winging
One patient with normal triceps strength had a mild
scapular winging that did not require treatment. This
patient had shoulder discomfort that largely resolved
over time. Three patients with partial palsy and triceps weakness presented with scapular winging. Two
had limitations in abduction and a decrease in elbow
flexion strength, which improved when the scapula
was manually stabilized by the examiner. These 2
patients had reoperation, during which the pectoralis
major was transferred to the scapula. One patient
experienced a good result, whereas in the other, in
Figure 7. Postoperative view of a patient who preoperatively

had weak triceps function. The suprascapular nerve was
supplied with both the accessory nerve and a branch to the
upper division of the serratus muscle. The C5 root was
grafted to the posterior division of the upper trunk. Elbow
flexion was reconstructed by transferring ulnar nerve fascicles to the biceps motor branch. Two years after the initial
surgery, the patient also had tendon transfers to correct wrist
and finger extension palsies. Results 42 months after the
initial brachial plexus surgery.
996
Table 4. Results of Abduction and External Rotation Reconstruction in the Partial Palsy Groups
Partial Palsy Groups
Abduction ()
External Rotation ()
Normal triceps (transfer to the suprascapular and axillary nerves)

Weak triceps (transfer to just the suprascapular nerve)
Weak triceps (double transfer to the suprascapular nerve)
122 (range, 80170)

76 (range, 7090)
88 (range, 70100)
118 (range, 90140)

50 (range, 3060)
128 (range, 90140)
The results of the normal triceps group were better than those from the weak triceps group. Only the results of the abduction recovery,
however, were statistically significant (p .05). In the weak triceps group, the results for the group with double transfer to the
suprascapular nerve were better than those after a single transfer. This was not statistically significant, possibly because of the reduced
number of patients.
whom the pectoralis major was weak, no improvement was observed.
Discussion
Surgical Approach of the Suprascapular and
Accessory Nerve
The approaches herein proposed were advantageous
because more direct dissection of the suprascapular
nerve was obtained. The cervical transverse artery
was not dissected or ligated. In fact, dissection was
carried out between the transverse cervical and suprascapular arteries. Preservation of the cervical
transverse vessels might be important in patients with
subclavian artery thrombosis or in the eventual need
of a secondary free muscle transfer as recipient vessels for microsurgical anastomosis. There also is a
decreased likelihood of skin breakdown, versus the
classic L incision, because the skin is not undermined.
It has been recommended that exploration of the
suprascapular notch for decompression be carried out
by means of a posterior approach, with an incision
over the spine of the scapula, and the patient in the
prone13 or supine14 position. Guan et al15 proposed
not only exposure but also transfer to the suprascapular nerve from the accessory nerve through the
posterior scapular approach. Their patient was placed
in a ventral decubitus position, which is a major
drawback. We previously have used the posterior
approach to the suprascapular nerve for motor rootlets transfer. We found that dissection of the suprascapular nerve is difficult in a deep plane,16 which
also was the impression of Kim et al.14 Of importance is that the scapular notch is located almost
directly under the clavicle. Anterior exposure of the
scapular notch by detachment of the trapezius muscle
is easier. The patient is in the dorsal decubitus position, and the suprascapular vessels can be controlled
safely. Moreover, the limited oblique approach proposed herein can be converted easily into the extended approach if necessary. We believe that the
scapular notch should be explored routinely in pa-
tients with scapular fractures or clavicular dislocation, because as observed here the suprascapular
nerve might be doubly injured.
The usefulness of accessory nerve dissection under
the deep cervical fascia was confirmed. Detachment
of the trapezius muscle allows for dissection of the
accessory nerve distally up to the medial angle of the
scapula, which ensures direct connection to the suprascapular nerve distal to the scapular notch and
preserves proximal branches to the upper trapezius
muscle. In fact, upper trapezius muscle function was
preserved in our patients.
Results of Nerve Transfers
In complete palsy, 45 of abduction but no external
rotation was obtained. This is in agreement with
results reported by Malessy et al17 and Chuang et al.3
Comparisons with other series are difficult, however,
because these other authors either did not specify the
degree of recovery or they failed to separate the
results for complete and partial injuries. There are
two explanations for the absence of external rotation
recovery observed in our patients with complete
palsy: (1) the supraspinatus muscle, being the first to
be reinnervated, attracts axons to the infraspinatus;
and (2) the antagonist musclefor instance, the subscapularis musclewhich is needed for humeral
head stabilization, remains denervated. When the suprascapular nerve was repaired using a contralateral
C7 motor rootlet, the mean recovery of abduction
was 86, and 2 of 12 patients also recovered some
external rotation.11 This suggests that increasing the
number of regenerating axons improves the regeneration rate. Indeed, even though not statistically significant, possibly because of the limited number of
patients, it seems that double transfer to the suprascapular nerve using the accessory nerve and a branch to the
upper portion of the long thoracic nerve improves results. This requires confirmation in a larger series.
For upper-type palsies, we performed transfers to
both the suprascapular and the axillary nerve. An
average of 105 of abduction/external rotation was
obtained. These results are significantly better than

those we observed with complete palsies, likely because we added a nerve transfer to the axillary nerve.
Preserved innervation of the antagonist muscles, better reconstruction of elbow flexion, and preserved
triceps function, however, also likely contributed to
improved shoulder motion.
The outcomes of the current study are slightly
better than those we have published previously5 (a
mean 92 recovery in abduction and external rotation) probably because in the current series, there
was longer follow-up evaluation. These combined
transfers resulted in 115 of abduction among the
first 7 patients reported by Leechavengvongs
et al.18 In his first report, the results for external
rotation were not mentioned. More recently,
Leechavengvongs et al19 later reported results on 8
additional patients who recovered an average of
115 of abduction and 97 of external rotation.
Different from the current study, however, only
C5-C6 injuries with normal triceps strength were
assessed. If results for patients with normal triceps
are considered exclusively, our recovery rate rises
to 122 of abduction and 118 of external rotation.
Of interest, the external rotation we observed was
20% greater than in Leechavengvongs series.19
This probably resulted from reinnervation of the
teres minor motor branch, as advocated by Bertelli
and Ghizoni5 but not done by Leechavengvongs
et al.18,19 These latter authors believed that the
teres minor functions as a shoulder adductor, so
that its reinnervation should decrease recovery in
abduction; we did not observe this.
Scapular Winging
Four of our patients with partial injuries presented
with postoperative winging of the scapula. This was
evident only after recovery of shoulder motion. In
patients with complete palsy, scapular winging also
would be noticed if better motion recovery could be
obtained. Of importance is that scapular winging
could be aggravated by our transfer of the accessory
nerve, because the lower portion of the trapezius
muscle and the rhomboid muscles, which may be
denervated, contribute to stabilization of the medial
border of the scapula. This aspect raises questions
about using the accessory nerve in certain patients.
Aggravation of scapular winging should be anticipated and probably prevented via reconstruction of
the long thoracic nerve, particularly the innervation
of the lower portion of the serratus anterior muscle,
which is the main stabilizer of the scapula.12 Not all
997
of the patients with scapular winging complained of

pain, and a few improved during the immediate postoperative period. In those with complaints, pectoralis
major transfers might be useful. In the case of failure
or weakness of the pectoralis major muscle, scapulothoracic arthrodesis might be proposed, but we lack
experience with such a procedure.
16, 2007.
article.
Corresponding author: Jayme A. Bertelli, MD, PhD, Praa Getulio
Vargas, 322, Florianpolis, SC, 88020030, Brazil; e-mail: bertelli@matrix.
com.br.
0363-5023/07/32A07-0007$32.00/0
doi:10.1016/j.jhsa.2007.05.016
References
1. Midha R. Nerve transfers for severe brachial plexus injuries:
a review. Neurosurg Focus 2004;16:110.
2. Alnot JY, Rostoucher P, Oberlin C, Touam C. Les paralysies
traumatiques C5-C6 et C5-C6-C7 du plexus brachial de
ladulte par lsions supraclaviculaires. Rev Chir Orthop
1998;84:113123.
3. Chuang DCC, Lee GW, Hashem F, Wei FC. Restoration of
shoulder abduction by nerve transfer in avulsed brachial
plexus injury. Evaluation of 99 patients with various nerve
transfers. Plast Reconstr Surg 1995;96:122128.
4. Terzis J, Kostas I. Suprascapular nerve reconstruction in 118
cases of adult posttraumatic brachial plexus. Plast Reconstr
Surg 2006;117:613 629.
5. Bertelli JA, Ghizoni MF. Reconstruction of C5-C6 brachial
plexus avulsion injury by multiple nerve transfers: XI to
suprascapular, ulnar fascicles to biceps branch, and triceps
long or lateral head branch to axillary nerve. J Hand Surg
2004;29A:131139.
6. Bertelli JA, Santos MA, Kechele PR, Duarte H, Ghizoni MF.
Triceps motor nerve branches as a donor or receiver in nerve
transfers. Neurosurgery (in press).
7. Tender GC, Kline DG. Anterior supraclavicular approach to
the brachial plexus. Neurosurgery 2006;58:360 364.
8. Chuang DCC. Neurotization procedures for brachial plexus
injuries. Hand Clin 1995;11:633 645.
9. Bertelli JA, Ghizoni MF. Improved technique for harvesting
the accessory nerve for transfer in brachial plexus injuries.
Neurosurgery 2006;58:366 370.
10. Huang KC, Tu YK, Huang TJ, Hsu RW. Suprascapular
neuropathy complicating a Neer type I distal clavicular
fracture:a case report. J Orthop Trauma 2005;19:343345.
11. Bertelli JA, Kechele PR, Santos MA, Duarte H, Ghizoni MF.
Axillary nerve repair by triceps motor branches transfer
through an axilla access: anatomical basis and clinical results. J Neurosurg (in press).
12. Bertelli JA, Ghizoni MF. Long thoracic nerve: anatomy
and functional assessment. J Bone Joint Surg 2005;87A:
993998.
998
13. Antoniadis G, Richter HP, Rath S, Braun V, Moese G.

Suprascapular nerve entrapment: experience with 28 cases.
J Neurosurg 1996;85:1020 1025.
14. Kim DH, Murovic JA, Tiel RL, Kline DG. Management and
outcomes of 42 surgical suprascapular nerve injuries and
entrapments. Neurosurgery 2005;57:120 127.
15. Guan SB, Hou CL, Chen DS, Gu YD. Restoration of shoulder abduction by transfer of the spinal accessory nerve to
suprascapular nerve through dorsal approach: a clinical
study. Chin Med J 2006;119:707712.
16. Bertelli JA, Ghizoni MF. Contralateral motor rootlets and
ipsilateral nerve transfers in brachial plexus reconstruction.
J Neurosurg 2004;101:770 778.
17. Malessy MJ, de Ruiter GC, de Boer KS, Thomeer RT.

Evaluation of suprascapular nerve neurotization after nerve
graft or transfer in the treatment of brachial plexus traction
lesions. J Neurosurg 2004;101:377389.
18. Leechavengvongs S, Witoonchart K, Uerpairojkit C,
Thuvasethakul P. Nerve transfer to deltoid muscle
using the nerve to the long head of the triceps, part
2 (a report of 7 cases). J Hand Surg 2003;28A:633
638.
19. Leechavengvongs S, Witoonchart K, Uerpairojkit C, Thuvasethakul P, Malungpaishrope K. Combined nerve transfers
for C5 and C6 brachial plexus avulsion injury. J Hand Surg
2006;31A:183189.
Anthropometry of the Human Scaphoid

Andrew D. Heinzelmann, MD, Graeme Archer, BSc,
Randy R. Bindra, MD
From the Orthopaedic Surgery Department, University of Arkansas for Medical Sciences, Little Rock, AR.
Purpose: Internal fixation has become a well-established alternative to casting for acute
scaphoid fractures. Screw design has evolved, and several different types of screws of varying
sizes are now available. The purpose of this study was to establish morphometric data for the
human scaphoid, document variation in scaphoid dimensions between genders, and to
evaluate symmetry in scaphoid measurements between the two sides.
Methods: We measured length, width, and morphology of the scaphoid in 30 paired
cadaveric specimens with reference to the long axis of the scaphoid from the proximal pole
to the distal articular surface. The width of the bone was compared with diameters of
commercially available screws.
Results: When measured along an axis from proximal pole to the distal articular surface, male
scaphoids (31.3 mm 2.1) were significantly longer than female specimens (27.3 mm
1.7). The male scaphoid was also significantly wider than the female specimen when
measured perpendicular to the long axis 2 mm from the proximal pole (4.5 mm 1.4 vs 3.7
mm 0.5) and at the waist (13.6 mm 2.6 vs 11.1 mm 1.2). There was no significant
difference in the distal pole diameter measured 2 mm from the tip between genders (7.2 mm
1.0 vs 7.2 mm 1.2). The diameters of most commercially available standard screws were
larger than the proximal pole of the female scaphoid.
Conclusions: Allowing for countersinking of the screw 2 mm beneath either pole, our data
suggest the usual screw length will be 27 mm and 23 mm for male and female scaphoids,
respectively. The small width of the proximal pole of the female scaphoid will not accommodate standard-sized screws from most manufacturers, and consideration must be given to
distal to proximal screw placement or use of mini screws if the implant is to be inserted in
a proximal to distal direction. (J Hand Surg 2007;32A:10051008. Copyright 2007 by the
American Society for Surgery of the Hand.)
Key words: Scaphoid, anatomy, fracture, internal fixation.
ith the availability of cannulated implants

and improved instrumentation, internal
fixation of scaphoid fractures is increasingly being suggested as the primary mode of treatment for acute injuries. The concept of scaphoid
fracture stabilization with a single intramedullary
screw has been promoted for several decades, but the
past decade has seen the introduction of several different screw designs from various manufacturers.
Besides differences in design, thread pattern, and
mode of compression, the various screws also have
different sizes (Table 1). Information on scaphoid
dimensions has been reported previously,1,2 but this
anatomic information has lesser surgical relevance
because the scaphoid is oriented obliquely in the
wrist and internal fixation is performed along the
central axis. In the senior authors experience, a

standard-size scaphoid screw may be relatively large
for a womans scaphoid especially when inserted
antegrade from a dorsal approach. Although the
scaphoid screw length to be used in a case is measured from the inserted guide wire at surgery, it
would be helpful to know the average length of
scaphoid in men versus women as a reference. The
purpose of this study was to establish the normal
range of morphometric data for the human scaphoid
with reference to screw fixation. Additionally, we
wanted to examine the difference in measurements
between genders and between the two sides in the
same individual. To our knowledge, there are no
previous reports that have documented scaphoid dimensions with reference to screw fixation.
1005
1006
Table 1. Screw Dimensions From Various Manufacturers

Leading Thread
Diameter (mm)
Trailing Thread
Diameter (mm)
Length Range
(Increments) (mm)
Acutrak II Std (Acumed)

Acutrak II mini (Acumed)
Acutrak II micro (Acumed)
4.0
3.5
2.5
4.1
3.6
2.8
Acutrak mini (Acumed)

Acutrak Std (Acumed)
Kompressor Mini (KMI)
Kompressor Std (KMI)
Twin-Fix (Stryker)
2.8
3.3
2.8
4.0
3.2
3.23.5*
3.84.6*
3.6
5.0
4.1
Herbert mini (Zimmer)

Herbert (Zimmer)
Headless Bone Screw Std (Orthosurgical Implants)
Headless Bone Screw Mini (Orthosurgical Implants)
Autofix (Small Bone Innovations)
3.0 HCS Short thread (Synthes)
2.5
3.0
3.0
2.5
2.0
2.5
3.0
3.2
3.9
3.9
3.2
3.0
3.3
3.5
3.0 mm cannulated Short thread (Synthes)
3.0
5.5 (washer)
1630 (2)
1630 (2)
814 (1)
1418 (2)
826 (2)
12.530 (2.5)
1026 (2)
1434 (2)
1420 (2)
2130 (1)
32, 34
1024 (2)
1230 (2)
1030 (1)
1030 (1)
1030 (1)
1030 (1)
1030 (1)
3240 (2)
830 (1)
3240 (2)
Screw (Manufacturer)
Source: Information obtained from manufacturers.

*Trailing thread diameter increases with increasing screw length.

We studied 30 pairs of embalmed cadaveric scaphoids (18 male and 12 female). The scaphoids were
excised by sharp dissection taking care to divide all
attached soft tissues while maintaining structure and
articular surface intact. Measurements were obtained
from each isolated specimen. To maintain consistency and clinical relevance, measurements were obtained in relation to an axis from the proximal pole to
the distal articular surface of the scaphoid where a
surgeon would normally insert a screw.3 Scaphoid
length and width were measured using a dial Manostat caliper with 0.05-mm scale (Manostat Corporation, New York, NY). Length was measured by placing one limb of the caliper at the most prominent
point of the proximal pole and the other tangential to
the distal articular surface of the scaphoid (Fig. 1).
Using this line as a reference, we then measured the
width of the scaphoid at 3 levels: proximal pole,
waist, and distal pole. The waist diameter was measured as the narrowest part of the scaphoid across its
capitate articular surface in a direction perpendicular
to the longitudinal axis (Fig. 2). As the scaphoid screw
is generally buried 2 mm beneath the articular surface of either pole, we measured the diameter of the
bone 2 mm from either pole. The proximal pole
diameter was thus measured 2 mm distal to the tip of
the pole and the distal pole was measured 2 mm

proximal to the distal articular surface.
We also classified the specimens by subjective
visual examination into the 3 morphologic types described by Compson et al1 as follows: 1, equally
sized poles; 2, small proximal pole; and 3, small
distal pole. A pole was considered small if it appeared visually smaller than the opposite pole. Poles
were considered equal if there was no visual size
difference.
We obtained implant dimensions for screws currently available for clinical use from the manufacturers to compare this data with scaphoid measurements
(Table 1).
Statistical analysis of the data was performed with
SigmaStat for Windows version 2.03 (Systat Inc, San
Jose, CA). Measurements were compared using the
unpaired t-test.
Results
The mean length of the scaphoid was 31.3 mm 2.1
for male and 27.3 mm 1.7 for female specimens,
which was a statistically significant difference (p
.001) (Table 2). We found the scaphoid to be narrowest at the proximal pole for both male and female
specimens, 4.5 mm 1.4 and 3.7 mm 0.5, respectively, which was a statistically significant difference
Heinzelmann et al / Scaphoid Morphometrics Related to Screw Fixation
1007
Table 2. Scaphoid Measurement Results for Male

and Female Specimens With Standard
Deviations and Corresponding p Values
Length
(mm)
Proximal
Width
(mm)
Distal
Width
(mm)
Waist
Width
(mm)
Male
31.3 2.1 4.5 1.4 7.2 1.0 13.6 2.6
Female 27.3 1.7 3.7 0.5 7.2 1.2 11.1 1.2
p value
.001
.010
.879
.001
ments were 13.6 mm 2.6 for male versus 11.1 mm

1.2 for female specimens. This difference was statistically significant (p .001).
We found that it was possible to classify the scaphoid specimens according to the 3 morphologic types.
Thirty-six specimens had equally sized poles (type
1), 8 scaphoids were type 2 with relatively smaller
proximal poles, and 16 specimens were type 3 and
demonstrated small distal poles.
There was no considerable difference between
proximal pole, distal pole, waist or length measurements between the right and left sides in the same
donor.
Discussion
Figure 1. The length of the scaphoid was measured from the
proximal pole to the distal articular surface of the scaphoid.
(p .010). The distal pole measured almost the same

regardless of gender, 7.2 mm 1.0 in male and 7.2
mm 1.2 in female specimens. The waist measure-
Figure 2. Waist width is determined at the narrowest part of

the scaphoid across the capitate articular surface in a direction perpendicular to the longitudinal axis.
The goal of internal fixation of the scaphoid is to

reduce the fracture and position a single compression
screw as close to the central axis as possible. Central
screw placement within the scaphoid has been shown
to be biomechanically advantageous,4 decreases healing time in acute fractures,5 and achieves union in
percutaneous treatment of ununited fractures.6 The
screw needs to be in the central one third of the
proximal pole to achieve central axis position.3 As
the scaphoid is narrowest at the proximal pole, it can
be likened to a cone. Placing a screw in a proximal to
distal direction from the apex of the cone allows the
surgeon to more precisely enter the middle of the
bone as opposed to attempting to direct the screw at
the narrow proximal pole from a distal to proximal
direction. Most currently available screws for scaphoid fixation are sized appropriate to our morphometric measurements with trailing thread diameters not
exceeding 4.1 mm. Some screws, however, such as
the standard-sized Kompressor (KMI Kinetikos
Medical Inc, Carlsbad, CA) with a 5-mm-diameter
trailing thread and the 3.0-mm cannulated screw with
5.5-mm threaded washer (Synthes, Paoli, PA) are
clearly too large for proximal to distal insertion
through the proximal pole. Furthermore, the small
proximal pole in female scaphoid specimens (3.7
mm) will not accommodate the larger diameter of
1008
standard-sized screws from most manufacturers. The

smaller proximal pole in type 2 scaphoids may also
be at risk of comminution with proximal to distal
screw insertion. In these situations, consideration
must be given to use of mini screws if the implant
is to be inserted in this direction. An alternative in
female patients would be to use distal to proximal
screw insertion stopping just short of the narrow
portion of the proximal scaphoid.
The average length of the scaphoid along the long
axis is 31 mm and 27 mm for men and women,
respectively. With a goal for screw placement 2 mm
beneath the cortical surfaces, the average length of
implant would be 27 mm for men and 23 mm for
women.
We did not find any difference in scaphoid measurement between the right and left sides in the same
individual. Radiographs of the uninjured wrist thus
offer an accurate template when planning reconstruction for scaphoid nonunion with collapse.
There are some limitations to this study. First, this
is a cadaveric study where it is easy to determine the
longitudinal axis. Intraoperative guide-wire placement may vary from the long axis resulting in different screw length if directed obliquely away from
the axis. Our sample size is relatively small with
unequal distribution among male and female specimens. We did not have information on racial demographics of the specimens and have not been able to
evaluate possible differences in scaphoid size among
different races.
Our data demonstrate significant difference in
scaphoid morphometry between the 2 genders. The
smaller size of the female scaphoid must be taken
into consideration when planning internal fixation.
Knowledge of the average lengths of screws may be

helpful in planning operating room inventory. When
designing new implants for scaphoid fixation, manufacturers should be cognizant of the dimensions of
the scaphoid.
Received for publication February 19, 2007; accepted in revised form
May 30, 2007.
article.
Corresponding author: Randy R. Bindra, MD, Orthopaedic Surgery Department, University of Arkansas for Medical Sciences, 4301 W. Markham
Street, No. 531, Little Rock, AR 72205; e-mail: BindraRandyR@uams.edu.
0363-5023/07/32A07-0009$32.00/0
doi:10.1016/j.jhsa.2007.05.030
References
1. Compson JP, Waterman JK, Heatley FW. The radiological
anatomy of the scaphoid. Part 1: osteology. J Hand Surg
1994;19B:183187.
2. Ceri N, Korman E, Gunal I, Tetik S. Morphological and
morphometric features of the scaphoid. J Hand Surg 2004;
29B:393398.
3. Trumble TE, Clarke T, Kreder HJ. Non-union of the scaphoid.
Treatment with cannulated screws compared with treatment
with herbert screws. J. Bone Joint Surg 1996;78A:1829
1837.
4. McCallister WV, Knight J, Kaliappan R, Trumble TE. Central
placement of the screw in simulated fractures of the scaphoid
waist: a biomechanical study. J Bone Joint Surg 2003;85A:
7277.
5. Trumble TE, Gilbert M, Murray LW, Smith J, Rafljah G,
McCallister WV. Displaced scaphoid fractures treated with
open reduction and internal fixation with a cannulated screw.
J Bone Joint Surg 2000;82A:633 641.
6. Slade JF III, Geissler WB, Gutow AP, Merrell GA. Percutaneous internal fixation of selected scaphoid nonunions with an
arthroscopically assisted dorsal approach. J Bone Joint Surg
2003;85A:20 32.
Association Between Lunate

Morphology and Carpal Collapse
Patterns in Scaphoid Nonunions
Steven C. Haase, MD, Richard A. Berger, MD, PhD,
Alexander Y. Shin, MD
From the Department of Surgery, University of Michigan Health System, Ann Arbor, MI; and the Department
of Orthopedic Surgery, Mayo Clinic, Rochester, MN.
Purpose: Type I lunates have a single distal facet for articulation with the capitate; type II
lunates have an additional (medial) hamate facet on the distal articular surface. We retrospectively reviewed a series of patients with scaphoid nonunions to determine if there was an
association between lunate morphology and the degree of carpal instability observed.
Association between lunate morphology and the location of the scaphoid fracture (proximal
or waist) was also investigated.
Methods: Radiographs were evaluated for 45 patients with established scaphoid nonunions.
Lunate morphology, scaphoid fracture location, and radiolunate angle were determined.
Results: Type I lunates were present in 21 patients. Of these, 15 were found to have a dorsal
intercalated segment instability pattern (radiolunate angle greater than 15). By contrast, only
4 of the patients with type II lunates exhibited this pattern of instability. No significant
association was found between lunate morphology and the scaphoid fracture location.
Conclusions: Type II lunate morphology is associated with significantly decreased incidence
of dorsal intercalated segment instability (DISI) deformity in cases of established scaphoid
nonunion (p .0002). Lunate morphology, however, was not significantly associated with
the location of the scaphoid fracture in these cases (p .19). (J Hand Surg 2007;32A:
1009 1012. Copyright 2007 by the American Society for Surgery of the Hand.)
Type of study/level of evidence: Prognostic IV.
Key words: Carpal instability, DISI, lunate, morphology, scaphoid nonunion.
he lunate has often been described as the cornerstone or keystone of the wrist.1,2 This title
is well deserved, as this carpal bone is
uniquely positioned at the middle of both the transverse and coronal arches of the carpus. Critical ligamentous attachments between the lunate, the scaphoid, and the triquetrum allow for stability within the
proximal carpal row. Additionally, the lunate is the
intercalated segment between the radius and distal
carpal row.
Variations in lunate morphology have been described in several different ways. In 1966, Antuna
Zapico divided lunates into three types (I, II, and III)
based on whether the proximal surface was curved or
angulated.3 Watsons three lunate types (D, V, and
N) are based on the lateral appearance on radiography.4 In 1990, Viegas classified lunates by their
distal articular morphology5,6: type I lunates with a

single capitate facet and type II lunates with an
additional hamate facet on the medial portion of the
distal articular surface, as shown in Figure 1. In the
same year, Burgess also described this finding, along
with matching variations of hamate morphology, referring to these as type I and type II midcarpal
joints.7
The presence of a medial (hamate) facet has been
linked to increased incidence of proximal hamate
arthrosis.8 10 Compared with type I lunates, type II
lunates have considerably different kinematics during radial-ulnar deviation of the wrist.11
Anecdotally, we have noticed a tendency for
wrists with type II lunate morphology to exhibit less
dorsal intercalated segment instability (DISI) deformity, in cases where this deformity is commonplace.
1009
1010
Figure 1. Type I lunates (L) have a single distal facet that

articulates with the capitate (C). Type II lunates have an
additional facet for articulation with the hamate (H). (T
triquetrum, S scaphoid) [Copyright Mayo Foundation].
DISI deformity12 describes the abnormally extended

posture of the lunate bone relative to the longitudinal
axis of the radius, as seen on lateral radiographs. This
radiographic manifestation of carpal instability is often seen in cases of scapholunate dissociation or
scaphoid nonunion. Both of these cases are forms of
the carpal instability dissociative (CID) pattern, and
both lead to abnormal extension of the lunate, as it is
dissociated from the flexion moment of the distal
scaphoid.
The purpose of this study was to evaluate the
association, if any, between lunate morphology and
the presence of DISI deformity. We also wanted to
determine if there was any association between lunate morphology and the site of the scaphoid fracture
(proximal pole or waist). We believe these associations, if present, may lead to new understandings
regarding instabilities of the wrist.
Our null hypotheses were as follows: (1) Lunate
morphology is NOT associated with the presence of
DISI deformity. (2) Lunate morphology is NOT associated with the location (proximal or waist) of
scaphoid fractures.
To test these hypotheses, we set out to perform a
retrospective review of a group of patients predisposed to DISI deformity (patients with scaphoid nonunions).

Institutional review board approval was granted for
this study. Surgical records for the years 1994
through 2003 were searched, and 52 consecutive
patients who had undergone vascularized bone grafts
for established scaphoid nonunion were identified.
The records for these patients were reviewed, with
specific regard to preoperative radiographic studies.
Seven patients were excluded due to lack of adequate

preoperative radiographs. The remaining 45 patients
were included in this analysis.
Preoperative wrist x-ray films were reviewed for
each patient, determining radiolunate angle, scaphoid
fracture location, and lunate morphology (Fig. 2).
The radiolunate angle was determined from lateral
wrist radiographs using the tangential method.13 The
patient was determined to have DISI deformity if
the radiolunate angle was greater than 15.14 Lunate
morphology and scaphoid fracture location were determined by examination of standard posterioanterior
wrist radiographs. If a medial lunate facet could be
identified, the lunate was classified as type II.5
Statistical analysis was performed using the chisquare test to determine if there was a statistically
significant relationship between the variables measured. Statistical significance was set at p .05.
Results
The study group was composed of 37 males and 8
females. The average age was 23 (range 13 66).
Type I lunates were identified in 21 (47%) cases.
Table 1 shows the distribution of lunate morphology and DISI deformity. Chi-square analysis shows
that this relationship is statistically significant (p
.0002).
Nine patients with type I lunates sustained a proximal pole fracture. Of the patients with type II lunates, 15 had proximal pole fractures. On chi-square
analysis, this distribution was not statistically significant.
Further analysis of the data was conducted to see if
there was an association between fracture location
and the development of DISI deformity. Of the proximal pole fracture patients, 5 developed DISI deformity, whereas 14 patients with scaphoid waist fractures developed DISI deformity. This was found to
be a significant association (p .002).
Discussion
Scaphoid nonunion is a condition well known to
predispose to carpal instability, specifically DISI deformity. Upon close examination of a cohort of
scaphoid nonunion patients, we found a significantly
decreased incidence of DISI deformity among those
Table 1. Lunate Morphology and DISI Deformity

Lunate Morphology
DISI Present
DISI Absent
Type I
Type II
15
4
6
20
Haase, Berger, and Shin / Lunate Morphology and Carpal Instability
1011
Figure 2. Posterioanterior (A) and lateral radiographs (B) of a typical patients wrist with a type I lunate, showing evidence of DISI
deformity. Posterioanterior (C) and lateral (D) radiographs of a type II lunate wrist, without evidence of DISI deformity.
patients with type II lunates. One possible interpretation of this finding is that type II lunate morphology
is protective against DISI deformity in this clinical
setting.
We theorize that the lunates added articulation
with the hamate lends some additional stability that
resists abnormal extension. It could be that the triquetrums usual extension moment, caused by
force transmitted through its screw-like articulation
with the hamate, is halted by the lunatohamate articulation present in type II midcarpal joints.
The reported incidence of type II lunate in the
literature ranges from 46% to 73%.5,7,8,15,16 Our
finding of 53% compares favorably with these reports, despite our relatively small sample size.
Sagerman et al specifically examined whether or
not lunate morphology could be accurately predicted
with routine radiographs.9 They found that only 74%

of type I lunates and 66% of type II lunates could be
correctly identified on plain x-rays. Relying on plain
radiographs for our determinations may have caused
us to overlook some type II lunates with very narrow
medial (hamate) facets. It is possible that these very
narrow medial facets are less clinically important,
but additional studies would be needed to confirm
this.
X-ray findings of an anatomic relationship do not
always have clinical implications. There is, however,
ample evidence that the lunatohamate articulation,
when present, is clinically relevant. Many authors
have noted that wrists with type II lunates are predisposed to proximal hamate arthrosis at this location.6,9 In a review of 186 wrist MRIs, Malik et al
found that 77% of lunates with type II morphology
1012
were physically apposed (articulated) with the

hamate on static magnetic resonance images. Furthermore, hamate edema, which was found in 5% of
studies, only occurred in wrists with type II lunates.15
Finally, a biomechanical study by Nakamura et al has
shown that there are real differences in the kinematics of wrists with type I versus type II lunates.11
All of this evidence points to the conclusion that
this lunatohamate articulation is clinically and biomechanically important. We believe it adds substantially to the stability of the proximal row in cases
where dissociative instability is likely. Further biomechanical studies are indicated to detect the exact
effect the type II lunate has on carpal instability.
Regarding the other findings of this study, it is
possible the rejection of our second hypothesis is
invalid due to a beta error. Sample sizes, although
satisfactory to make chi-square a valid test, were
indeed small in this study. Analysis of the data also
revealed an association between fracture location and
DISI deformity. The increased incidence of DISI
deformity in more distal fracture patterns correlates
nicely with the findings of previous studies,17 although this was not one of our studys a priori
hypotheses.
In conclusion, our primary null hypothesis is rejected. There is a statistically significant association
between lunate morphology and the development of
DISI deformity. There was, however, no significant
association detected between lunate morphology and
the location of the scaphoid fracture.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
Received for publication March 25, 2007; accepted in revised form June
11, 2007.
article.
Corresponding author: Steven C. Haase, MD, University of Michigan
Health System, 2130 Taubman Center, 1500 E. Medical Center Dr., Ann
Arbor, MI 48109-0340; e-mail: shaase@med.umich.edu.
0363-5023/07/32A07-0010$32.00/0
doi:10.1016/j.jhsa.2007.06.005
References
1. Sennwald G, Segmuller G. Base anatomique dun nouveau
concept de stabilit du carpe. [Anatomic basis of a new
15.
16.
17.
concept of stability of the carpus]. Int Orthop 1986;10:

2530.
Amadio PC. Carpal kinematics and instability: a clinical and
anatomic primer. Clin Anat 1991;4:112.
Taleisnik J. The wrist. New York: Churchill Livingstone,
1985:171172.
Watson HK, Yasuda M, Guidera PM. Lateral lunate
morphology: an x-ray study. J Hand Surg 1996;21A:759
763.
Viegas SF, Wagner K, Patterson R, Peterson P. Medial
(hamate) facet of the lunate. J Hand Surg 1990;15A:564
571.
Viegas SF. The lunatohamate articulation of the midcarpal
joint. Arthroscopy 1990;6:510.
Burgess RC. Anatomic variations of the midcarpal joint.
J Hand Surg 1990;15A:129 131.
Viegas SF, Patterson RM, Hokanson JA, Davis J. Wrist
anatomy: incidence, distribution, and correlation of anatomic
variations, tears, and arthrosis. J Hand Surg 1993;18A:463
475.
Sagerman SD, Hauck RM, Palmer AK. Lunate morphology:
can it be predicted with routine x-ray films? J Hand Surg
1995;20A:38 41.
Nakamura K, Patterson RM, Moritomo H, Viegas SF. Type
I versus type II lunates: ligament anatomy and presence of
arthrosis. J Hand Surg 2001;26A:428 436.
Nakamura K, Beppu M, Patterson RM, Hanson CA, Hume
PJ, Viegas SF. Motion analysis in two dimensions of radialulnar deviation of type I versus type II lunates. J Hand Surg
2000;25A:877 888.
Linscheid RL, Dobyns JH, Beabout JW, Bryan RS. Traumatic instability of the wrist. Diagnosis, classification,
and pathomechanics. J Bone Joint Surg 1972;54A:1612
1632.
Gilula LA, Weeks PM. Post-traumatic ligamentous instabilities of the wrist. Radiology 1978;129:641 651.
Nakamura R, Hori M, Imamura T, Horii E, Miura T. Method
for measurement and evaluation of carpal bone angles.
J Hand Surg 1989;14A:412 416.
Malik AM, Schweitzer ME, Culp RW, Osterman LA,
Manton G. MR imaging of the type II lunate bone: frequency, extent, and associated findings. AJR Am J Roentgenol 1999;173:335338.
Aufauvre B, Herzberg G, Garret J, Berthonneaud E, Dimnet
J. A new radiographic method for evaluation of the position
of the carpus in the coronal plane: results in normal subjects.
Surg Radiol Anat 1999;21:383385.
Moritomo H, Viegas SF, Elder KW, Nakamura K, Dasilva
MF, Boyd NL, et al. Scaphoid nonunions: a 3dimensional analysis of patterns of deformity. J Hand Surg
2000;25A:520 528.
Carpal Kinematics During Simulated Active

and Passive Motion of the Wrist
Rita M. Patterson, PhD, Laura Williams, BS, Clark R. Andersen, BS,
Shukuki Koh, MD, Steven F. Viegas, MD
From the Division of Biomechanics and Bone Physiology Research, Department of Orthopaedic Surgery and
Rehabilitation, The University of Texas Medical Branch, Galveston, TX; the Department of Hand Surgery,
Nagoya University School of Medicine, Nagoya, Japan; and the Division of Hand Surgery, Department of
Orthopaedic Surgery and Rehabilitation, The University of Texas Medical Branch, Galveston, TX.
Purpose: The purpose of this study was to investigate the effect of experimental control
mechanisms, simulated active (tendon-driven) and passive (externally assisted), on carpal
motion.
Methods: Kinematics of the carpal bones in five fresh-frozen cadaver upper extremities were
studied using an optical motion analysis system. The wrist extensors and flexors were
dissected and loaded. For passive motion, the tendons were loaded to simulate muscle tone
while the investigator passively moved the wrist using a pin placed in the third metacarpal.
To simulate active, patient-driven motion, the tendons were attached directly to guide bars
while the investigator used a puppeteer mechanism to move the wrist.
Results: There were no significant differences in carpal motion (flexion-extension motion or
radial-ulnar deviation) when the wrist was moved in simulated active motion through the
extensor and flexor tendons or in passive motion, with a constant force applied to the
tendons. Kinematics for simulated active motion, in general, was more difficult to control and
was less smooth than the kinematics for passive motion.
Conclusions: Carpal bone kinematics (excluding the pisiform) in a healthy normal joint are
similar in both simulated active (tendon-driven) and passive (externally assisted) wrist motion
because the carpal bones are passively moved during wrist motion (there are no direct
tendon-to-muscle attachments to the proximal carpal bones and minimal attachments to the
distal carpal bones). (J Hand Surg 2007;32A:10131019. Copyright 2007 by the American
Society for Surgery of the Hand.)
Key words: Carpal kinematics, passive motion, simulated active motion.
here have been a variety of cadaver and patient studies investigating the biomechanics
and kinematics of the wrist. Because there is
no good animal model for the human wrist, studies
have been limited to investigating kinematics in cadaver wrists or have used noninvasive imaging in
living subjects. The ultimate goal is to describe the
dynamic motion of the carpal bones in vivo. Cadaver
studies yield dynamic carpal kinematic data, but the
neuromuscular control of the extremity is questionable. Studies in living subjects using computed tomography (CT) or magnetic resonance imaging
(MRI) are limited to the estimation of kinematics
from static positioning of the hand because the technology is not fast enough to evaluate dynamic mo-
tion. The neuromuscular control used to position the

hand may be more physiologic, but the carpal kinematics may be different for static positioning. Therefore, there is a need to create realistic cadaver models
that simulate in vivo forces so that dynamic carpal
kinematic measurements can be made. The effect of
moving the wrist actively (tendon-driven motion)
versus passively (externally assisted motion) on carpal kinematics has not been determined.
Viegas et al1 investigated load contact areas
through the radiocarpal joint under different static
loading conditions in a cadaver model. They found
no difference in static load transmission if the load
was applied through the second and third metacarpals, through all five metacarpals, or through the
1013
1014
extensor and flexor tendons of the wrist. Dynamic

carpal kinematics has been studied in cadavers with a
variety of loading configurations. Some have produced wrist motion through a computer-controlled
servo hydraulic simulator2 and others by producing
motion passively with small forces applied to the
wrist extensor and flexor tendons to simulate normal
muscle tonus.35
Studies performed on living subjects have investigated carpal kinematics for sequential static positioning of the hand using CT6,7 or MRI.8 These
studies have provided a wealth of carpal position data
for different wrist positions in living volunteers under normal static muscle loading conditions. However, results from in vivo studies are affected by
motion artifact and mostly limited to the investigation of normal healthy volunteers. Kauffman4,5 compared the carpal kinematics in a quasi-dynamic cadaver model, using forces applied to the wrist flexor
and extensor tendons to position the hand, with those
reported in in vivo studies and found similar results.
Research7 comparing kinematics in cadaver models
with in vivo models has produced similar results, but
with large variability. Differences can be attributed
to nonrealistic loading and control of the wrist in
cadaver models or to differences because of the static
positioning required to image living subjects. The
effect of load configuration on carpal kinematics
remains unclear.
Dynamic cadaver testing with realistic tendon
loading will allow for more precise measurement of
carpal kinematics. In addition, this testing will allow
for study of soft tissue injury and possible surgical
repair techniques. There have been no studies that
have evaluated if passive and simulated active positioning of the cadaver wrist results in the same carpal
alignment. The purpose of this study was to investigate carpal kinematics of seven carpal bones (excluding the pisiform) during simulated active (tendondriven) and passive (externally assisted) motion of
the wrist to establish that the carpal bones are passively positioned by the tendons that move the hand
and carpal motion can be measured in a cadaver
model.

Specimen Preparation
Five fresh-frozen cadaver upper extremities free
from visible or radiographically identifiable deformities and/or degenerative changes were studied. Three
male and 2 female specimens were used (3 right
wrists and 2 left wrists). The average age at the time
of death was 42 years (range, 28 59 years). Titanium

screws (2.5-mm diameter) were placed into the dorsal surface of the radius, capitate, scaphoid, lunate,
triquetrum, hamate, trapezium, trapezoid, and third
metacarpal under fluoroscopic guidance (FluoroScan
Imaging Systems, Inc, Northbrook, IL). A 2-mmdiameter graphite rod approximately 5 to 8 cm long
was glued to the head of each screw. At the end of
each rod is a triad pin (TP) that has three 5-mmdiameter spheres placed in a cruciform arrangement
on top. The spheres are coated with 3M Scotchlite
reflective liquid (3M, St. Paul, MN) and sub 0.5-mmdiameter reflective beads (Rolco Labs, Carlstadt, NJ)
to enhance their reflectance to the video camera.
TPs were oriented to ensure that they would not
come into contact with each other during a full
passive flexion/extension range of motion (ROM)
of the wrist. The TPs are designed and placed so
they are rigidly attached to the screws in the bones
and do not flex or vibrate independently. They are
positioned to avoid tendons and avoid or minimize
tethering of the capsule and/or ligaments. The distal
flexor carpi ulnaris, flexor carpi radialis, extensor
carpi ulnaris, and extensor carpi radialis brevis and
longus tendons of the wrist were dissected and 0
polyethylene braided sutures were attached to each
tendon. The extensor carpi radialis brevis and longus
were sutured together as one unit. The remaining
length of the suture attached to each tendon was
passed under the skin along the muscle belly and
back out the skin at the elbow. The remaining lengths
of these four sutures that extended beyond the skin
were used to exert simulated muscle force in both the
simulated active and passive conditions. A special jig
(Fig. 1) was designed and constructed to maintain the
upper extremity in a neutral forearm orientation and
to allow wrist motion. The forearm was placed in
neutral pronation/supination parallel to the floor and
elbow flexed 90.
Passive Motion
To simulate passive, externally assisted motion, the
sutures that were attached to the tendons (the flexor
carpi radialis, the connected extensor carpi radialis
brevis and longus, the flexor carpi ulnaris, and extensor carpi ulnaris) and extended out beyond the
skin were used to simulate resting muscle force and
balance between the wrist extensors and flexors. The
radial wrist extensors (extensor carpi radialis brevis
and longus) were connected to the radial wrist flexor
(flexor carpi radialis) using the remaining length of
the sutures that extended beyond the skin by con-
Patterson et al / Carpal Kinematics
1015
pulley arrangement, one to the extensor and flexor of

the radial side and the second to the extensor and
flexor of the ulnar side. This was done to simulate a
resting in vivo tonus and maintain joint congruity.
Loading was verified via a digital hand-held scale
(Berkley, Spirit Lake, IA). Because of the pulley
arrangement (Fig. 1), the force from the elastic material was equally split between the extensor and
flexor tendons regardless of wrist position. The
weighted pulley arrangement allowed the flexor and
extensor tendons to move synergistically (together)
and to approximate normal muscle tone. The wrist
was moved by one investigator using a pin inserted
into the medullary canal of the third metacarpal.
Simulated Active Motion
To model simulated active, patient-driven motion, a
puppeteer mechanism was contrived in which the
radial and ulnar extensor tendon pair was attached to
a guide bar and the radial and ulnar flexor tendon pair
was attached to a second guide bar. These guide bars
allowed free movement of the tendon pairs during
simulated active motion. The bar arrangement allowed the flexor and extensor tendons to move synergistically (together) and to simulate normal muscle
control for trial motion (Fig. 2).
Figure 1. Photograph of the experimental setup showing the

cadaver arm and the Thera-Band tubing used to provide an in
vivo simulated muscle tonus. The insert shows a line drawing
of the pulley arrangement.
necting them to each other and passing the resulting

loop of suture around a free-floating pulley. The
ulnar wrist extensor (extensor carpi ulnaris) was connected to the ulnar wrist flexor (flexor carpi ulnaris)
in a similar looped fashion and passed around a
second free-floating pulley. These floating pulleys
allowed free movement of the tendon pairs during
wrist motion. Thera-Band tubing (Smith & Nephew
Rolyan Inc., Germantown, WI) connected each pulley to a jig post. Each post was adjusted to stretch the
Thera-Band and apply 22.2 N (5 lb) of weight to each
Kinematic Data Acquisition

The five wrists were tested twice each in flexion/
extension and radial/ulnar deviation. First, one investigator moved the wrist through a passive (externally
guided with the extensor and flexor tendons loaded)
arc. Second, the same investigator used lines attached
to the flexor and extensor pairs to move the wrist
though a simulated active (puppeteer controlled) arc.
Velocity of the movement was not specifically controlled for, but two cycles of flexion/extension or
radial/ulnar motion was achieved during a 6- or
4-second data acquisition, respectively.
Six black-and-white video cameras arranged
around and above the wrist tracked the TP reflective
surfaces throughout the range of motion, creating
motion path files. The original data were obtained at
60 Hz (60 frames/s). Data from each of 6 cameras
was processed by the Eva tracking program (Motion
Analysis Corp., Santa Rosa, CA), which creates
three-dimensional coordinate files for each target
identified. The points of each triad pin (the spheres)
in the radius, capitate, lunate, scaphoid, trapezium,
trapezoid, triquetrum, and hamate were tracked during the motion of the wrist for flexion/extension and
radial ulnar deviation for each condition (simulated
active or passive). The resultant raw three-dimen-
1016
Absolute orientation between corresponding pairs

of triads was determined using the method of Horn,9
capturing the complete relative rotation of the triads
(therefore, the motion of the bones to which they are
attached). The resulting angles were smoothed via a
running barycentric average10 over a 5-frame interval about each frame. This resulted in a threedimensional angular rate profile not artificially confined to the standard body planes (sagittal, coronal, or
transverse). The resultant angles generated from this
analysis are basically the rotation angle about the
discrete axis of rotation between the two objects.
Based on the ability of the system to track markers
and the resultant calculation of orientation between
pairs of triads (ie, motion between bones), the threedimensional angles between bones could be calculated to within 5.
Statistical analyses were performed on the data using
PC-SAS (SAS Institute, Cary, NC). A 2-way ANOVA
using the Proc GLM (General Linear Model) in PCSAS was used to test for differences in carpal angle due
to simulated active or passive movement of the wrist
(condition) at each global wrist position. A difference of
5 was considered clinically relevant. Unless explicitly
stated otherwise, a p value .01 was considered to be
statistically significant in this work.
Results
Figure 2. Photograph of the paddles used to create simulated
active motion by moving the flexors and extensors of the
wrist.
sional data were then checked for labeling and dropout errors using the tracking editor program. The
final path data contain the x, y, and z coordinates for
each of the three targets on the pin for each frame of
data acquisition. The path data error is dependent on
the data capture volume. The data acquisition volume
used for the study was a 15-cm square. The system
calculated a goodness of fit for each camera to
determine how well the objects could be tracked. The
goodness of fit averaged 0.5 mm (ie, the system
could define the position of each triad pin to within
0.5 mm). Estimated total error for the system has
been determined for both static and dynamic tests.
Position data calculated from a static test revealed
that the angles between static targets basically remained constant to within 0.01. Data taken from a
dynamic test when the targets were rotating through
a 360 motion revealed that the system was able to
record dynamic angle changes to within 3.1.
Positional data for each wrist motion (flexion/extension and radial/ulnar deviation) for each condition
(simulated active and passive) were recorded. To
confirm that the wrist was moved similarly for each
condition (simulated active and passive); out-ofplane motion was calculated. Our results found that
for flexion/extension of the wrist, the coronal (radial/
ulnar) out-of-plane component of wrist motion averaged 4.0 and 4.7, respectively, for the simulated
active and passive motions. The transverse (pronation/supination) out-of-plane component of the wrist
motion averaged 3.7 and 3.2, respectively, for the
simulated active and passive motions. Paired t-test
analysis of differences in average out-of-plane motion for each wrist revealed that there were no statistically different differences in out-of-plane motions between the 2 groups (p .2 and .5).
Three-dimensional angles for each carpal bone
were graphed with respect to the global wrist position
as measured by the capitate angle with respect to the
radius. Figure 3A and B shows lunate and scaphoid
motion during flexion/extension of the wrist for passive and simulated active motoring of the wrist. Figure 4A and B shows lunate and scaphoid motion
1017
Figure 3. (A) Lunate and (B) scaphoid motion during flexion/extension of the wrist for passive and simulated active motoring of
the wrist.
during radial/ulnar deviation of the wrist during passive and simulated active motoring of the wrist. Differences between simulated active and passive angles
for each carpal bone were graphed with respect to
global wrist position (calculated as the angle of the
capitate with respect to the radius). Each graph was
fit to a 10th order polynomial model in Sigmaplot
(Sigmaplot, Point Richmond, CA). This polynomial
was used to generate models for the data from all
specimens so that so that averages between wrists
could be calculated.
The percent total flexion extension motion (FEM)
for the lunate in the current study was 62.1% and
61.7% for simulated active and passive motion, respectively. The percent total FEM for the scaphoid in
the current study was 85.5% and 83.2% for simulated
active and passive motion, respectively. The percent
total radial ulnar deviation (RUD) for the lunate in
the current study was 72.3% and 79% for simulated
active and passive motion, respectively. The percent
total RUD for the scaphoid in the current study was
69.1% and 83.2% for simulated active and passive
motion, respectively.
Average differences between simulated active and
passive motion during flexion and extension of the

wrist and radial/ulnar deviation are reported in Table 1.
The mean difference between simulated active and
passive flexion/extension motion was less than 1.78.
The difference between simulated active and passive
radial/ulnar deviation was less than 2.46.
Discussion
There has been a variety of studies investigating the
kinematics of the carpal bones with disparate results.
The explanation for the varied results seems to revolve around differences between cadaver and livesubject measurements and static versus dynamic data
acquisition. Moojen et al7 provided a comprehensive
review of several different methods of measuring
carpal kinematics using either cadavers or live subjects. Specifically, they compared scaphoid and lunate percent contribution to wrist flexion and extension and radial and ulnar deviation. They found
consistent results reported for FEM of the wrist but
more varied results for RUD and were unable to
determine a single functional model of carpal kinematics as there was a great range of results reported
in the literature. This may be an artifact of wide
Figure 4. (A) Lunate and (B) scaphoid motion during radial/ulnar deviation of the wrist during passive and simulated active
motoring of the wrist.
1018
Table 1. Difference Between Simulated Active

and Passive Motion
Bone
Lunate
Scaphoid
Triquetrum
Trapezium
Hamate
Trapezoid
Flexion/Extension Motion
Differences (Mean SD)
1.78
0.63
0.76
0.42
0.76
1.58
5.31
3.75
3.07
3.86
3.76
5.41
Radial/Ulnar
Motion
Differences
(Mean SD)
1.64
2.46
1.16
1.02
0.57
1.69
3.54
3.42
2.51
3.15
2.02
3.79
normal variation and small specimen population. The

results of the current study show results for flexionextension to be within the range of previous studies;
however, the radial-ulnar deviation percent measurements are larger than those reported in the literature.
This is potentially due to the differences in how the
angles were reported. Typically, angles between
bones are reported as Euler angles. The current study
calculated a three-dimensional angle not artificially
confined to the standard Euler planes (sagittal, coronal, or transverse). Another, perhaps more likely
explanation is that the current study reports FEM
motion at 30 of flexion/extension, whereas previous
studies report motion at 60 of flexion/extension.7
Kauffman et al4 measured carpal kinematics in a
cadaver using CT-based noninvasive position registration methods similar to that measured in live volunteers.
They studied wrist motion using a quasi-dynamic cadaver model in which the wrist was positioned and held
through tension applied to the wrist flexor and extensor
tendons and reported similar results to those of in vivo
studies, which they felt validated their model.
Kinematic data in the current study found similar
results for lunate and scaphoid motion during simulated active and passive wrist FEM and RUD. Both
were similar to other published reports.3,11,12 In the
current study, average differences over the entire
wrist range of motion were less than 5 (Table 1).
The difference between passive and simulated active
angles, however, was not constant and varied
throughout the range of motion of the wrist (Figs. 3
and 4). Generally, the differences were greater toward the end of the range of wrist motion tested but
were not greater than 5 for any individual wrist.
Furthermore, data reported in the current study
found no significant differences in carpal motion
(FEM or RUD) when the wrist was moved actively
through the extensor and flexor tendons or passively
with a constant force applied to the tendons (p
.01). Kinematics for simulated active motion, in general, was more difficult to control and was less
smooth than passive motion. In addition, simulated
active range of motion was typically less than that
achieved through passive externally assisted motion.
The attachment sites of the main motors of the
wrist are distal to the carpal bones, which would
indicate that the wrist is passively moved throughout
FEM and RUD. They include the flexor carpi radialis, flexor carpi ulnaris, extensor carpi radialis brevis
and longus, and the extensor carpi ulnaris. The flexor
carpi radialis runs through a special groove deep to
the flexor retinaculum and inserts into the base of the
second metacarpal with small insertions into the third
metacarpal and trapezium. The flexor carpi ulnaris
passes anterior to the styloid process of the ulna and
inserts mainly into the proximal surface of the pisiform with smaller insertions in to the hook of the
hamate and the bases of the fourth and fifth metacarpals. The extensor carpi ulnaris passes anterior to the
ulnar styloid process and inserts into the posterior
aspect of the base of the fifth metacarpal. The extensor carpi radialis brevis inserts into the base of the
third metacarpal. The extensor carpi radialis longus
runs posterior to the anatomical snuffbox and inserts into the base of the second metacarpal.
Carpal bone kinematics in a healthy normal joint is
similar regardless of how the joint is moved because
the carpal bones are passively moved during wrist
motion (there are no direct tendon-to-muscle attachments to the proximal carpal bones and minimal
attachments to the distal carpal bones). The flexor
carpi radialis does insert into the trapezium. However, this attachment has been previously reported as
being only 1.1% of the total bone surface area.13 The
flexor carpi ulnaris main attachment is to the pisiform with minimal attachments to the hook of the
hamate, which minimally affects the kinematics of
the carpal bones. This information may be helpful in
decisions regarding early motion and rehabilitation in
the posttraumatic wrist.
Received for publication October 5, 2006; accepted in revised form May
4, 2007.
Supported by grant no. 5R01AR049354 from the National Institute of
Arthritis and Musculoskeletal and Skin Diseases at the National Institutes
of Health.
article.
Corresponding author: Rita M. Patterson, PhD, Department of Orthopaedic Surgery and Rehabilitation, 2.804 Rebecca Sealy, University of
Texas Medical Branch, 301 University Blvd., Galveston, TX 775550174. e-mail: rita.patterson@utmb.edu.

0363-5023/07/32A07-0011$32.00/0
doi:10.1016/j.jhsa.2007.05.004
7.
References
1. Viegas SF, Patterson RM, Tencer A, Peterson P, Roefs J,
Choi S. The effects of various load paths and different loads
on the load transfer characteristics of the wrist. J Hand Surg
[Am] 1989;14A:458 465.
2. Werner FW, Palmer AK, Somerset JH, Tong JJ, Gillison
DB, Fortino MD, et al. Wrist joint motion simulator. J Orthop Res 1996;14(4):639 646.
3. Patterson RM, Nicodemus CL, Viegas SF, Elder KW,
Rosenblatt J. High-speed three-dimensional kinematic analysis of the normal wrist. J Hand Surg [Am] 1998;23A:446
453.
4. Kaufmann R, Pfaeffle J, Blankenhorm B, Stabile K, Robertson D, Goitz R. Kinematics of the midcarpal and radiocarpal
joints in radioulnar deviation: an in vitro study. J Hand Surg
[Am] 2005;30A:937942.
5. Kaufmann R, Pfaeffle J, Blankenhorm B, Stabile K, Robertson D, Goitz R. Kinematics of the midcarpal and radiocarpal
joint in flexion and extension: an in vitro study. J Hand Surg
2006;31A:11421148.
6. Crisco JJ, McGovern R, Wolfe S. Noninvasive technique for
8.
9.
10.
11.
12.
13.
1019
measuring in vivo three-dimensional carpal bone kinematics.

J Orthop Res 1999;17:96 100.
Moojen TM, Snel JG, Ritt JPF, Venema HW, Kauer JM,
Bos KE. In vivo analysis of carpal kinematics and comparative review of the literature. J Hand Surg [Am] 2003;
28A:81 87.
Goto A, Moritomo H, Murase T, Oka K, Sugamoto K,
Arimura T, et al. In vivo three-dimensional wrist motion
analysis using magnetic resonance imaging and volumebased registration. J Orthop Res 2005;23:750 756.
Horn BKP. Closed-form solution of absolute orientation
using unit quaternions. J Optical Soc Am 1987;4(4):629
642.
Gramkow C. On averaging rotations. Int J Computer Vision
2001;42(1/2):716.
Werner FW, Green JK, Short WH, Masaoka S. Scaphoid and
lunate motion during a wrist dart throw motion. J Hand Surg
[Am] 2004;29A:418 422.
Berdia S, Short WH, Werner FW, Green JK, Panjabi M. The
hysteresis effect in carpal kinematics. J Hand Surg [Am]
2006;31A:594 600.
Nanno M, Buford WL, Patterson RM, Andersen CA, Viegas
SF. Three-dimensional analysis of the ligamentous attachments of the first carpometacarpal joint. J Hand Surg [Am]
2006;31A:1160 1170.
Long-Term Follow-Up of an Undiagnosed

Trans-Scaphoid Perilunate Dislocation
Demonstrating Articular Remodeling and
Functional Adaptation
Elizabeth A. Bathala, MD, Peter M. Murray, MD
From the Departments of Diagnostic Radiology and Orthopedic Surgery, Mayo Clinic, Jacksonville, FL.
The most common perilunate dislocation is the trans-scaphoid dorsal perilunate variant. It is
estimated that up to 25% of perilunate injuries are diagnosed late. We report 66-year
follow-up of an unreduced, previously undiagnosed trans-scaphoid perilunate dislocation of
the wrist. At follow-up, unique post-traumatic articular remodeling was seen on wrist
radiographs, and the patient had only a mild functional deficit. (J Hand Surg 2007;32A:
1020 1023. Copyright 2007 by the American Society for Surgery of the Hand.)
Key words: Articular remodeling, trans-scaphoid dorsal perilunate dislocation.
erilunate and lunate dislocations are diagnosed

late in up to 25% of cases1,2 presenting treatment challenges. Post-traumatic arthritis may
result after these injuries irrespective of treatment or
the timing of treatment. Little is known about the
actual incidence of perilunate fracture dislocations of
the wrist in general and trans-scaphoid perilunate
dislocations in particular. It has been estimated that
perilunate dislocations, fracture dislocations, and lunate dislocations as a whole represent approximately
10% of all wrist injuries.3 More than half of perilunate dislocations of the wrist are of the transscaphoid variety.2 Repair or reconstruction of the
scapholunate interosseous ligament in lunate and
perilunate dislocations is generally recommended.3
The results of late treatment of perilunate dislocations in general, however, are regarded as fair or
poor.4 8 We report a 66-year follow-up of an unreduced, previously undiagnosed trans-scaphoid perilunate dislocation, demonstrating unique post-traumatic articular remodeling and only mild long-term
functional deficits. This case demonstrates the potential for successful nonoperative treatment of the very
late presenting trans-scaphoid perilunate dislocation.
Case Report
An 83-year-old, right-handed, retiree presented
with painless left wrist swelling of 6 months duration. He remembered a remote history of a severe
1020
left wrist hyperextension injury during a football

game, 66 years prior, while serving in World War
II. His wrist was acutely painful and swollen.
Shortly after injury, he sought medical attention,
but no radiographs were obtained and no diagnosis
was made. He was immobilized in a long-arm cast
for 1 month. He remembered being asked to do
pushups after the cast was removed. The pushups
caused him pain, which lasted several months, but
after stopping the pushups his pain resolved and he
gradually returned to ship duty in the U.S. Navy.
After the injury, he noted that his wrist was deformed and was notably stiffer than before. After
his tour of duty in the Navy, he worked as an
accountant and played golf and tennis. During the
next 66 years, he remembers only occasional episodes of pain, and he adapted to the lack of left
wrist motion.
At original presentation to our clinic, physical
exam of the left wrist demonstrated deformity of
the left wrist (Fig. 1). He was medically stable. His
left wrist range of motion showed 20 of extension
and 15 of flexion compared with 50 of extension
and 60 of flexion on the right. Strength testing
revealed 27 kgf of grip strength right wrist and 14
kgf on the affected left wrist. Pinch strength was 4
kfg right thumb and 3 kgf left thumb, chuck pinch
5 kgf right thumb and 4 kgf left thumb. His upper
extremity neurologic examination was normal, in-
Bathala and Murray / Chronic Trans-scaphoid Dorsal Perilunate Dislocation
1021
Figure 1. Clinical photograph of an 83-year-old male presenting with unreduced left trans-scaphoid dorsal perilunate
dislocation.
cluding median nerve examination. Radiographs

demonstrated post-traumatic osteoarthritis of the
left wrist and an unreduced trans-scaphoid dorsal
perilunate wrist dislocation with a scaphoid nonunion (Figs. 2, 3, and 4). Also observed radiographically was remodeling of the dorsal lip of the
distal radius articular surface accommodating the
dorsally displaced carpus with articulation of the
capitate with the radius.
For his new onset of swelling, the patient was
treated with night-time splinting and oral antiinflammatory medications. His painless swelling
improved over the course of 6 weeks and he re-
Figure 2. Posteroanterior radiograph of unreduced transscaphoid dorsal perilunate dislocation.
Figure 3. Lateral radiograph of unreduced trans-scaphoid

dorsal perilunate dislocation.
turned to his normal level of function. Four years

after his initial presentation to our clinic, he has
had no return of his swelling and has no pain. He
gardens and works as a patient transport volunteer
at our institution, pushing patients in wheelchairs.
His left wrist range of motion has improved to 45
Figure 4. Sagittal computed tomography image of unreduced trans-scaphoid dorsal perilunate dislocation demonstrating distal radius articular remodeling to accommodate
the head of the capital.
1022
of extension and 15 of flexion. His neurologic

examination is unremarkable aside from left-hand
median nerve distribution Semmes-Weinstein
monofilament testing of 3.61.
Discussion
Late-presenting carpal dislocations typically have
a poor outcome.1,7 Patients with chronic carpal
dislocations may present with pain, deformity, limited motion, and median nerve symptoms.9 This
case is unique due to the very long delay in diagnosis, the resultant unique remodeling of the distal
radius articular surface, and the patients good
level of hand function after nonoperative care.
Poor radiographic appearance and malalignment
of the carpus does not necessarily portend a bad
outcome.1,2 The most common complication of
untreated trans-scaphoid perilunate fracture-dislocations is degenerative osteoarthritis.2,9 Additionally, untreated carpal injuries have been reported
to lead to median nerve dysfunction, pain, carpal
collapse, decreased range of motion, late carpal
instability, and osteonecrosis of the scaphoid or the
lunate.2,8 10
Open reduction and internal fixation after surgical immobilization is the treatment of choice in the
acute injury period. This treatment can be successfully employed up to 3 months after the injury.1,5,9,11 When attempting to reduce a chronically
dislocated carpus, the surgeon must take into account that shortening of the median nerve and the
radial and ulnar arteries has occurred.9,12 Chronic,
unreduced carpal dislocations can be treated with
wrist arthrodesis, excision of the lunate, or proximal row carpectomy.1,8,9 Inoue and Shionoya6 recommended proximal row carpectomy for patients
with chronic perilunate dislocations diagnosed
later than 2 months from injury. In their 16 patients
treated with proximal row carpectomy, however,
10 had a fair result and 6 a poor result. Siegert et
al7 reported similar results in patients with chronic
perilunate dislocations treated by proximal row
carpectomy or wrist arthrodesis and go on to mention that complications may occur after these procedures. Little is known about the long-term follow-up of the chronic nonoperatively treated transscaphoid dorsal perilunate dislocations. Tomaino
references diagnosing a patient with a previously
treated chronic perilunate dislocation.8 The patient
presented due to median nerve paresthesias, 60
years after injury. No follow-up after treatment
was provided. The long-term follow-up in our pa-
tient demonstrates potential adaptability of the

wrist joint after injury, including remodeling of the
distal articular surface. Our patient has had no
median nerve dysfunction. A shortcoming of our
report is that injury radiographs were never obtained. Our patient, however, graphically recalls
the injury as well as the postinjury sequelae and
has had no other wrist trauma.
In late presentations of the trans-scaphoid perilunate dislocations, the potential advantages of
chronic repair or salvage operations must be
weighed against the possible complications from
treatment. This case demonstrates that radiographic and functional adaptation to chronically
unreduced trans-scaphoid perilunate dislocations
of the wrist can occur. We conclude that when this
injury is diagnosed late, an enduring functional
result may be achieved by nonoperative treatment
of the chronically unreduced trans-scaphoid dorsal
perilunate dislocation of the wrist. This information may be of interest to surgeons and patients
faced with making treatment decisions when previously untreated trans-scaphoid dorsal perilunate
dislocations are diagnosed late.
May 3, 2007.
article.
Corresponding author: Peter M. Murray, MD, Department of Orthopedic Surgery, Mayo Clinic, 4500 San Pablo Road, Jacksonville, FL
32224; e-mail: murray.peter@mayo.edu.
0363-5023/07/32A07-0012$32.00/0
doi:10.1016/j.jhsa.2007.05.003
References
1. Givissis P, Christodoulou A, Chalidis B, Pournaras J. Neglected trans-scaphoid trans-styloid volar dislocation of the
lunate. Late result following open reduction and K-wire
fixation. J Bone Joint Surg 2006;88A:676 680.
2. Herzberg G, Comtet JJ, Linscheid RL, Amadio PC, Cooney
WP, Stalder J. Perilunate dislocations and fracturedislocations: a multicenter study. J Hand Surg 1993;88B:
768 779.
3. Minami A, Kaneda K. Repair and/or reconstruction of
scapholunate interosseous ligament in lunate and perilunate
dislocations. J Hand Surg 1993;18A:1099 1106.
4. Gellman H, Schwartz SD, Botte MJ, Feiwell L. Late treatment of a dorsal transscaphoid, transtriquetral perilunate
wrist dislocation with avascular changes of the lunate. Clin
Orthop Relat Res 1988;237:196 203.
5. Howard FM, Dell PC. The unreduced carpal dislocation. A
method of treatment. Clin Orthop Relat Res 1986;202:112
116.
6. Inoue G, Shionoya K. Late treatment of unreduced perilunate dislocations. J Hand Surg 1999;24B:221225.
Bathala and Murray / Chronic Trans-scaphoid Dorsal Perilunate Dislocation

7. Siegert JJ, Frassica FJ, Amadio PC. Treatment of chronic
perilunate dislocations. J Hand Surg 1988;13A:206 212.
8. Tomaino MM. Late management of perilunate fracture-dislocations. In: Trumble T, ed. Carpal fracture-dislocations.
Chicago: American Academy of Orthopedic Surgery, 2002:
718.
9. Murray PM. Dislocations of the wrist: carpal instability
complex. J Am Soc Surg Hand 2003;3:88 99.
1023
10. Johnson RP. The acutely injured wrist and its residuals. Clin
11. Pandit R. Proximal and palmar dislocation of the lunate and proximal scaphoid as a unit in a case of scaphocapitate syndrome. A
32-month follow-up. J Hand Surg 1998;23B:266268.
12. Mayfield JK, Johnson RP, Kilcoyne RK. Carpal dislocations: pathomechanics and progressive perilunar instability. J Hand Surg [Am] 1980;5:226 241.
Clinical Manifestations
of Type IV Ulna Longitudinal Dysplasia
Bassem El Hassan, MD, Sam Biafora, MD, Terry Light, MD
From the University of Illinois at Chicago, Chicago, IL; and the Department of Orthopaedic Surgery and
Rehabilitation, Loyola University, Stritch School of Medicine, Maywood, IL.
Purpose: Ulna longitudinal dysplasia is an uncommon congenital anomaly that demonstrates

a wide variety of clinical manifestations. The clinical manifestations and function of patients
with Bayne type IV ulna longitudinal dysplasia have not been well characterized. The
purpose of this study was to report the clinical features of type IV ulna longitudinal dysplasia
and the extent to which this affects a patients ability to perform activities of daily living.
Methods: The medical records of children diagnosed with ulna longitudinal dysplasia in our
institution between 1960 and 2004 were reviewed. The children found to have ulna longitudinal dysplasia with radiohumeral synostosis (Bayne type IV ulna dysplasia) were studied.
The laterality of the deformity, associated musculoskeletal and nonmusculoskeletal anomalies, and treatments were recorded. Patients were interviewed regarding their ability to
perform activities of daily living.
Results: One hundred twenty-five patients with 146 affected limbs were identified with ulna
dysplasia. Seventeen limbs in 14 patients (12% of affected limbs) demonstrated radiohumeral
synostosis (RHS). Three of 14 patients with RHS had bilateral involvement. The elbows were
fixed in 20 to 90 of flexion. No elbows were positioned in full extension. Eleven of the 17
involved limbs with RHS had digital anomalies. Nine of the 17 limbs had surgical reconstruction. The majority of these procedures were performed on the hand.
Conclusions: The elbow, forearm, wrist, and hand clinical findings associated with type IV ulna
longitudinal dysplasia are variable. Surgical treatment usually focuses on correction of hand
abnormalities. Many patients function satisfactorily and are able to perform daily activities
without surgical intervention. (J Hand Surg 2007;32A:1024 1030. Copyright 2007 by the
Type of study/level of evidence: Diagnostic IV.
Key words: Dysplasia, hemimelia, radiohumeral synostosis, type IV, ulna.
lna longitudinal dysplasia (ulnar club hand,

ulnar hemimelia, or ulnar longitudinal deficiency), an uncommon upper extremity anomaly, was described by Goller in 1698.1 The deformity
was classified by Swanson as a longitudinal failure of
formation of parts.2 The incidence of ulnar ray deficiency has been reported by Froster and Baird3 to be
about 1 per 5,000 live births. Though this condition is
usually regarded as a sporadic event, Roberts4 has
reported a case of familial occurrence.
The clinical findings associated with ulna longitudinal dysplasia vary widely with deficiencies ranging
from absence or hypoplasia of the ulna, to carpal
hypoplasia or coalition, to hypoplasia or absence of
digits, humerus, or shoulder girdle.518 Although the
1024
ulnar aspect of the distal upper extremity is more

substantially affected, thumb deficiencies are frequent.511 By contrast, in radial deficiency the
most profound abnormalities are along the radial
border, whereas the ulnar border is not profoundly
involved.
Ulnar ray deficiency classification schemes are
based on the extent or form of deformity in the
hand11,19 or forearm and/or elbow.5,6,8,20 22 The
scheme proposed by Bayne21 is primarily based on
forearm and elbow anomalies. He also describes associated wrist and hand involvement (Table 1). According to Baynes classification, all patients with
radiohumeral synostosis (RHS) are categorized as
ulna longitudinal dysplasia type IV.
El Hassan, Biafora, and Light / Clinical Manifestations of Type IV Ulna Longitudinal Dysplasia
1025
Table 1. Baynes Classification of Ulnar Ray Deficiency

Type
Ulna
Radius
Elbow
Hypoplasia
Distal and proximal
epiphyses present
Mild bowing
Stable
II
Partial aplasia
Proximal ulna
present
Anlage tethers radius
Total absence
No anlage
Bowing
Stability variable
Radial head may
dislocate posterolaterally
Straight
Unstable
Posterolateral dislocation
radius
Severe elbow flexion
deformity may coexist
Radiohumeral synostosis
Stable/extended
Humeral internal
rotation
Forearm pronation
III
IV
Usually completely
absent
Anlage present
Severe radial bowing
Synostosis of the radiohumeral joint suggests an

embryologic failure of cavitation, the process by
which joints develop embryologically.13 In type IV
ulna longitudinal dysplasia limbs, the ulna may be
completely absent, hypoplastic, or represented by a
cartilaginous ulnar anlage. The degree of associated
radial bowing and ulnar deviation of the wrist and
hand is variable.
Prior series have detailed the clinical manifestations of patients with all forms of ulna longitudinal
dysplasia.518 Literature focused on patient satisfaction and function after the treatment of type IV ulna
longitudinal dysplasia is limited. The purpose of this
study is to report the clinical manifestations of type
IV ulna longitudinal dysplasia and the impact of
these manifestations on activities of daily living.

Review of the medical records of children seen at
Shriners Hospital for Children, Chicago, between
1960 and 2004 identified patients with the diagnosis
of ulna longitudinal dysplasia, isolated RHS, and
RHS in association with ulna longitudinal dysplasia.
The study was approved by the hospitals institutional review board.
One hundred twenty-five patients (146 limbs) with
ulna longitudinal dysplasia were identified. Seventeen limbs in 14 patients (12% of affected limbs)
demonstrated RHS, type IV ulna longitudinal dysplasia (Fig. 1). Five (36%) patients demonstrated anomalies in other extremities (Table 2): bilateral proximal
femoral focal deficiency (PFFD) with fibular ray
Wrist/Hand
Mild ulnar deviation at
wrist
Absence/hypoplasia of
digits common
Mild ulnar deviation at
wrist
Ulnar deviation at
wrist not great
Carpal/digital
deficiencies severe
Ulnar deviation at wrist
deficiency, hypoplastic or absent thumb, phocomelia,

radial head dislocation, and scoliosis. None of the
patients had visceral or hemopoietic anomalies. No
birth-related fractures were recorded. The average
age of the 14 included patients at the time of this
review was 22 years (range, 8 45 years).
The medical records as well as the radiographs of
the patients with type IV ulna longitudinal dysplasia
were reviewed to confirm the diagnosis and report
the associated findings. There were 9 men and 5
women, a ratio of 1.8:1. None of the patients had
a family history of congenital anomalies. Digital
anomalies, elbow, forearm, and wrist position, musculoskeletal and nonmusculoskeletal anomalies, and
surgeries performed were noted.
Determination of elbow, forearm, and wrist positions were based on the chart and radiography reviews. In most instances, it was very difficult to
measure these angles accurately because the elbow
was synostosed and there was frequently some degree of rotation in the radius. Essentially, the examiner (senior orthopedic resident) determined the armforearm and forearm-wrist angles. The position of
the elbow was determined radiographically using the
humeral and radial shafts as reference points. For the
forearm position, the degree of supination/pronation
was determined by the position of the thumb (if it
existed) or palm of the hand with respect to the arm.
Medical record documentation or follow-up visits
were deemed insufficient in 8 of the 14 patients.
These 8 patients were contacted by phone. A short
questionnaire was verbally administered to the pa-
1026
and 5 patients had right-side involvement. The affected extremity was shorter compared with the contralateral limb. Three of the 14 patients with RHS
had bilateral involvement. These limbs were proportionately short and symmetric in length.
The average position of the radiohumeral fusion
was 63o of flexion ranging from 10o to 90o. Five of
17 forearms were fixed in neutral rotation, 1 in 20o
pronation, and the other 5 were in 0o to 80o of
supination. Three limbs had complete absence of the
ulna, and the other 14 demonstrated varying degrees
of ulna hypoplasia. The wrist position was in neutral
in 71% of cases, and the rest were in 5 to 40o of
ulnar deviation. The wrist active range of motion was
within normal range in the group of patients with
wrist in neutral deviation. Patients with ulnar deviation, however, had limited active and passive radial
deviation (range, 30o ulnar deviation to 10o radial
deviation) and mild decrease in active wrist flexion
(average 42o) and extension (50o).
Digital anomalies were present in 11 of 17 limbs
(Table 3). Four limbs had absent thumbs. Eleven
limbs had ectrodactyly, simple or complex syndactyly often involving the thumb and index finger,
angular deformities, or intercarpal synostosis.
Six of 11 unilateral patients had surgical treatment.
Five of these surgeries involved the hand. The hand
procedures included first metacarpal rotational osteotomy, deepening of the first web space with fullthickness skin graft, index pollicization, and release
of syndactyly between the long and ring fingers. One
patient with an elbow synostosis positioned at 90o of
flexion and the forearm in 20o of supination had a
forearm osteotomy to rotate the forearm into neutral
rotation.
One patient had contralateral phocomelia, although recent literature suggests this may represent a
severe type V form of ulnar longitudinal dysplasia.23
In all other unilateral cases, the type IV ulnar dysplastic limb was shorter than the nonaffected limb.
Three patients had shoulder hypoplasia. One of these
children had such severe hypoplasia of the involved
Figure 1. Unilateral type IV ulnar deficiency demonstrating

ossified proximal hypoplastic ulna.
tients. The questionnaire defined limitations in their

activities of daily living including feeding, dressing,
showering, and hygiene. In addition, participation in
sports activities, work experience, any related pain
issues, and prosthetic use were also recorded. The
details in the medical records of the other 6 patients
provided answers to these questions.
Results
Eleven of the 14 patients with RHS had unilateral
involvement. Six patients had left-side involvement
Table 2. Skeletal Anomalies in Unaffected Extremities
Skeletal Anomalies
Unilateral
Bilateral
Total
Total Patients
None
PFFD
RHD
TH
FRD
PH
11
3
14
6
2
8
1
1
2
PFFD, proximal femoral focal deficiency; RHD, radial head dislocation; TH, thumb hypoplasia; FRD, fibular ray deficiency; PH, phocomelia.
1027
Table 3. Clinical Manifestations in the Affected Limbs
No. of limbs
Elbow Contracture
Wrist
No. of Digits
Total
Limbs
Flexion
Extension
Neutral
Ulnar Deviation
17
17
12
extremity that he was fit with an orthotic to provide

a longer limb post used against the contralateral
normal hand to hold objects and assist in performing
activities of daily living. This was the only patient
who used an orthotic to improve function. Two additional patients chose to use a hand prosthesis for
esthetic reasons.
Nine of the 11 unilateral patients reported no limitations in their activities of daily living. Five of 11
participated in sports and considered themselves capable of doing most activities compared with their
peers. Two patients, one with contralateral phocomelia and another with severe hypoplasia of the affected
extremity, were limited in some sports activities and
in performing fine motor activities such as tying
shoes and typing. Nonetheless, these individuals consider themselves highly functional.
Two of the 3 bilateral patients had surgical reconstruction. One had release of syndactyly between the
thumb and index fingers, and the other had bilateral
anlage excision. The presence of anlage in this patient was confirmed by magnetic resonance imaging
(Fig. 2). This patients wrists were in average 40o of
ulnar deviation and could be passively corrected to
20o of ulnar deviation. Her active wrists range of
motion averaged 44o of flexion and 50o of extension.
She had excision of the ulnar anlage in each forearm.
Several months after anlage excision, she could radially deviate her wrists to a neutral position, and her
wrists flexion improved to 55o and extension to 60o,
Figure 2. (A) Radiograph of an 8-year-old patient with type IV ulna longitudinal dysplasia with ulnar deviation of the wrist.
(B) Magnetic resonance image of the same patient performed 20 months after the radiograph showing the cartilaginous anlage
causing the tethering effect of the wrist and leading to ulnar deviation.
1028
thereby improving both the function and appearance

of her hands.
All 3 patients with bilateral involvement had symmetrically short upper extremities, with variable hypoplasia of the shoulder girdle and proximal humerus. The patient with PFFD used lower limb
prostheses. None of these patients use upper limb
prostheses.
All 3 patients with bilateral involvement were
satisfied with their upper extremity functional status
when questioned during the phone interview. One
patient, 44 years old at the time of the phone interview, reported playing football when he was in high
school. He has also been able to swim, ski, and
snowboard. Competitive participation in sports activities requiring use of both upper extremities was
difficult. He reported having strong upper limbs and
was working in construction. Another patient, 35
years old, reported that functionally she has almost
no limitations in using her upper extremities in
daily living activities. She did not participate in
sports except for swimming and horseback riding.
She believed that her limitations were mostly related to the associated bilateral PFFD. She is employed as a computer programmer. The third patient, currently 8 years old, was very pleased with
the appearance of her hands and was able to use
them in most activities of daily living. She was
involved in limited sports activities including
swimming and running.
Discussion
Isolated RHS is rare. Pfeiffer and Braun-Quentin24
distinguished 3 entities in which there is RHS: (1)
autosomal dominant ankylosis of the elbow as an
element of a systemic disorder with multiple joint
synostosis; (2) autosomal recessive RHS with dysgenesis of the ulna and possibly the fibula and femur
but without digital anomalies; and (3) nongerminal
RHS as part of ulna longitudinal dysplasia with digital anomalies.
The 1 patient in the first category was excluded
from this study. The patient with bilateral involvement and PFFD would be classified in the second
category, and the 13 other patients would be classified in the third category outlined by Pfeiffer.
Radiohumeral synostosis may also be associated
with Antley-Bixler syndrome, which is a rare disorder with musculoskeletal manifestations that include
RHS, ulnar bowing, camptodactyly, arachnodactyly,
joint contractures, and craniosynostosis.25 Other associated findings include brachycephaly, dysplastic
ears, midface hypoplasia, and choanal atresia. Most

cases of RHS found in this syndrome are bilateral
with the elbow fused in near 90o of flexion. We do
not suspect that any patient presented in this study
had Antley-Bixler syndrome.
Goldfarb et al23 proposed that an additional form
(type V) be added to Baynes existing classification
of ulna longitudinal dysplasia. This type would include cases of severe RHS with humeral bifurcation
or a large medial epicondyle. Although it is difficult
to distinguish between a greatly hypoplastic ulna
observed in type IV and a large medial epicondyle or
humeral bifurcation in type V, we do not believe that
any of the limbs presented in this study fall into this
category.
The incidence of type IV ulna longitudinal dysplasia in our study was 12%. We found 97 cases of
type IV ulna longitudinal dysplasia in the literature.5 8,10,13,16 18 This number accounts for 43% of
the ulna dysplasia cases reported in these series. The
incidence of type IV ulna longitudinal dysplasia
ranged from 13% to 53%. These differences may
reflect the varying nature of the reporting institutions,
childrens hospitals, prosthetic centers, or hand clinics. There are multiple series reporting on the associated manifestations of ulna longitudinal dysplasia
with RHS.5,6,8,10,16 18 These reports do not discuss
the ability of these patients to perform activities of
daily living at follow-up evaluation.
In our series, type IV ulna longitudinal dysplasia
was more common in men than in women (1.8:1),
and most of the cases were unilateral (79%). These
results are similar to those reported by previous studies. There are, however, differences in the rate of
associated clinical findings.
In our review of the literature, the largest reported
series of patients with ulna longitudinal dysplasia and
associated RHS was presented by Swanson et al.6
Swanson reviewed the records of 65 patients (88
limbs) with ulna longitudinal dysplasia and identified
47 (53%) limbs with RHS. He reported 76% of
patients with elbow synostosis were fixed in extension and 15% of the wrists were positioned in ulnar
deviation. In our series, none of the patients were
fixed in full extension; all of them were stiff in 10 to
90 of flexion. Similarly, Miller et al5 demonstrated
flexed elbow position in their patients with RHS
ranging from 15 to 100. The hand on flank deformity was present in 9 of their 17 limbs. Miller
describes this deformity as consisting of a hyperpronated forearm, bowing of the radius, and flexion and
rotation of the elbow so that the hand faces posteri-
orly and the hand tends to lie on the flank or buttock

at rest. These patients externally rotate at the shoulder to place the hand in a functional position in front
of the body. None of our patients exhibited this
deformity. Swanson et al6 reported 90% of his patients had absent ulna associated with RHS, 87% of
patients had 2 4 digits, 65% of patients had associated lower extremity involvement, and 35% had associated upper extremity involvement. In our series,
however, 27% of the patients with unilateral involvement and none of the patients with bilateral involvement had total absence of the ulna, 36% had associated anomalies either in the contralateral upper
extremity or the lower extremities, and 82% of the
patients had 2 4 digits. Carroll and Bowers10 and
Miller et al5 reported 2 4 digit hands in 67% and
88% of their patients, respectively.
Eighty-three percent of the procedures in our patients were performed on the hand to release syndactyly, deepen web spaces, and reorient digits by index
pollicization or first metacarpal osteotomy. In the
series of Miller et al,5 57% of the procedures were
performed on the hand for similar abnormalities. In 2
of our 17 affected limbs, the ulna anlage was excised.
Carroll and Bowers10 performed anlage excision in 3
of their 9 patients and Straub16 excised the anlage in
his 1 affected patient.
Miller et al5 also reported performing osteotomies
of the humerus or radius to address rotational and/or
angular deformities in 5 of 17 (29%) patients with
RHS. Carroll and Bowers10 performed osteotomies
in 3 of their 9 patients. Straub16 performed an osteotomy at the proximal radius in which the elbow
was placed in flexion with the forearm in slight
pronation. One of our patients had a rotational forearm osteotomy to place the flexed elbow in a more
neutral forearm rotation. Swanson6 expressed his belief that surgical intervention would be fruitless in
most of his patients and that a prosthesis would be
the treatment of choice.
All patients in this series expressed satisfaction
with their upper extremity function. Patients with
unilateral involvement were able to perform activities of daily living and were involved in many sports
activities without restrictions. Only one patient chose
to use an upper limb orthosis for functional assistance. Most of these patients used the contralateral
extremity as a dominant extremity and were able to
ski, play tennis, play basketball, and some of them
were also able to play volleyball. They did not report
limitations of activities of daily living. Patients with
bilateral involvement, however, were more restricted
1029
in terms of performing more complex tasks and various sports. Their participation in most types of
sports was more limited. Prior to excision, the patient
with bilateral ulnar anlage experienced functional
limitations in daily living activities. Both her hands
were in ulnar deviation, which in the presence of
fixed elbows limited her upper limb functions. According to Bayne,21 most patients with type IV ulna
longitudinal dysplasia demonstrate this anlage that
tethers the distal radius and influences the extent of
ulnar bowing of the radius. Only this patient in our
series, however, had both wrists contracted in 40o of
ulnar deviation. She did very well after her wrists
were corrected to neutral.
In all patients in our series with unilateral Bayne
type IV ulna longitudinal dysplasia, the involved
extremity was shorter than the contralateral limb.
The shorter limb in these patients may be advantageous. In patients with bilateral involvement, the
short limbs help patients place the hands closer to the
face and other parts of the upper body regardless of
the elbow position. These patients use shoulder,
wrist, and hand motion to address most of their
needed activities.
Ulnar longitudinal deficiency is a rare deficiency
that differs from other longitudinal deficiencies by its
widely varied clinical manifestations. Although there
have been numerous reports of type IV ulna longitudinal dysplasia, these reports fail to discuss the
ability of these patients to perform daily activities
at follow-up evaluations. Patients tend to function
well without surgical intervention. Most surgical
procedures are intended to address hand abnormalities.
Received for publication October 4, 2006; accepted in revised form
May 21, 2007.
article.
Corresponding author: Terry Light, MD, Loyola University, Stritch
School of Medicine, 2160 South First Avenue, Maywood, IL 60153; e-mail:
tlight@lumc.edu.
0363-5023/07/32A07-0013$32.00/0
doi:10.1016/j.jhsa.2007.05.020
References
1. Goller DC. Abortus humani monstros. His Anta Misc Acad
Nat Curios Norimb Decuria 1698;2:311.
2. Swanson AB. A classification for congenital limb malformations. J Hand Surg 1976;1:8 22.
3. Froster UG, Baird PA. Upper limb deficiencies and associated malformations: a population-based study. Am J Med
Genet 1992;44:767781.
1030
4. Roberts AS. A case of deformity of the forearm and hands

with an unusual history of hereditary congenital deficiency.
Ann Surg 1886;3:135139.
5. Miller JK, Wenner SM, Kruger LM. Ulnar deficiency.
J Hand Surg 1986;11A:822 829.
6. Swanson AB, Tada K, Yonenobu K. Ulnar ray deficiency: its
various manifestations. J Hand Surg 1984;9A:658 664.
7. Broudy AS, Smith RJ. Deformities of the hand and wrist
with ulnar deficiency. J Hand Surg 1979;4:304 315.
8. Ogden JA, Watson HK, Bohne W. Ulnar dysmelia. J Bone
Joint Surg 1976;58A:467 475.
9. Johnson J, Omer GE Jr. Congenital ulnar deficiency: natural
history and therapeutic implications. Hand Clinic 1985;1:499
510.
10. Carroll RE, Bowers WH. Congenital deficiency of the ulna.
J Hand Surg 1977;2:169 174.
11. Cole RJ, Manske PR. Classification of ulnar deficiency according to the thumb and first web. J Hand Surg 1997;22A:
479 488.
12. Spinner M, Freudlich BD, Abeles ED. Management of moderate longitudinal arrest of development of the ulna. Clin
13. Frantz CH, ORahilly R. Ulnar hemimelia. Artif Limbs 1971;
15:2535.
14. Kanavel AB. Congenital malformations of the hands. Arch
Surg 1932;25:153.
15. Pardini AG. Congenital absence of the ulna. J Iowa Med Soc
1967;57:1106 1112.
16. Straub LR. Congenital absence of the ulna. Am J Surg 1965;

109:300 305.
17. Laurin CA, Farmer AW. Congenital absence of the ulna.
Can J Surg 1959;2:204 207.
18. Blair WF, Shurr DG, Buckwalter JA. Functional status in
ulnar deficiency. J Pediatr Orthop 1983;3:37 40.
19. Ogino T, Kato H. Clinical and experimental studies on ulnar
ray deficiency. Handchir Mikrochir Plast Chir 1988;20:330
337.
20. Riordan DC. The upper limb. In: Lovell WW, Winter RB,
eds. Pediatric orthopaedics. Vol. 2. Philadelphia: JB Lippincott, 1978:685719.
21. Bayne LG. Ulnar club hand (ulnar deficiencies). In: Green
DP (ed): Operative hand surgery. 3rd ed. New York:
Churchill Livingstone, 1993:288 303.
22. Kummel W. Die missbildungen der extremitaten durch defect, verwachsung und ueberzahl. Bibliotheca Medica (Cassel) 1895;Heft 3:1 83.
23. Goldfarb CA, Manske PR, Busa R, Mills J, Carter P, Ezaki
M. Upper-extremity phocomelia reexamined: a longitudinal
dysplasia. J Bone Joint Surg 2005;87A:2639 2648.
24. Pfeiffer RA, Braun-Quentin C. Genetic nosology and counseling of humeroradial synostosis. Genetic Counseling 1994;
5(3):269 274.
25. Rumball KM, Pang E, Letts RM. Musculoskeletal manifestations of the Antley-Bixler syndrome. J Pediatr Orthop B
1999;8(2):139 143.
Objective Features and Aesthetic Outcome

of Pollicized Digits Compared With
Normal Thumbs
Charles A. Goldfarb, MD, Valerie Deardorff, MD, Ben Chia, BA,
Amy Meander, BA, Paul R. Manske, MD
From St. Louis Shriners Hospital for Children, St. Louis, MO; the Department of Orthopaedic Surgery,
Washington University School of Medicine, St. Louis, MO; Barnes Jewish Hospitals, St. Louis, MO.
Purpose: To evaluate the objective features and subjective aesthetic outcome of pollicized
digits compared with normal thumbs.
Methods: Thirty-one pollicized digits in 26 patients were evaluated at an average 41 months
after surgery. The length, girth, and nail width were measured and compared with previously
reported data for normal thumbs. A surgeon, therapist, and caregiver completed Visual
Analog Scales (VAS) to subjectively assess the aesthetic outcome; they also provided the
principal reasons for their assessment of the altered appearance compared with normal
thumbs. All data were statistically analyzed.
Results: The average length of the pollicized digit relative to the long finger proximal phalanx
was 90% (26%), compared with an age-matched normal average of 71%. The girth of the
pollicized digit relative to the long finger was 92% (8%), compared with an age-matched
normal thumb average of 132%. The nail width of the pollicized digit relative to the nail
width of the long finger was 96% (9%), compared with an age-matched normal thumb
average of 104%. The VAS scores averaged 7.3 for the caregiver, 6 for the therapist, and 6.4
for the surgeon. The most frequently cited (altered) features were narrow girth, angulation,
and excess length of the pollicized digit.
Conclusions: Pollicized digits are longer and have reduced girth and nail width compared
with age-matched normal thumbs. The most significantly abnormal features are decreased
girth, excess length, and angulation. (J Hand Surg 2007;32A:10311036. Copyright 2007
by the American Society for Surgery of the Hand.)
Key words: Aesthetics, appearance, outcome, pollicization, thumb size.
ollicization has been used to treat the absent or

hypoplastic thumb for more than 50 years. The
primary goal of pollicization is functional improvement. Outcome reports have provided details
on strength, range of motion, and usage patterns.1,2
The appearance of the pollicized digit has not been
the focus of previous reports, and assessments of the
physical characteristics of appearance compared with
normal thumbs have been limited. Aesthetic outcomes have been described as improved, less
objectionable, and satisfactory.3,4,5 There have
been only a few attempts to quantify the appearance
of the pollicized digit and there are no currently
widely accepted means of doing so.2,6
The physical characteristics (i.e., relative length,

girth, nail width) of normal thumbs of children 118
years have been recently reported.7 The first purpose
of this investigation was to objectively compare
the physical characteristics of pollicized digits
with those of normal thumbs. Additionally, we
evaluated the subjective aesthetic outcome using a
visual analog scale (VAS) completed by surgeon,
therapist and caregiver and compared the features of
the pollicized digit that contributed to its altered
appearance with a normal thumb. Finally, we correlated the objective data with the VAS scores to better
understand which objective features most affect subjective outcome.
1031
1032

Our review of the surgical registry at our pediatric
orthopaedic hospital identified a total of 53 index
finger pollicizations for thumb hypoplasia performed
between 1993 and 2005. Twenty-six patients with 31
pollicized index fingers agreed to participate and
returned for a clinical evaluation in this retrospective
investigation. Institutional Review Board approval
was provided for the investigation and informed consent was obtained from each patient.
There were 15 girls and 11 boys included, and the
average age at the time of assessment was 7 years
(standard deviation, 4.1 years; range, 218 years). All
patients had radial longitudinal deficiency (RLD); according to the Bayne and Klug8 classification as modified by James et al,9 there were 11 extremities with
a type 0 RLD, 3 extremities with a type 1 RLD, 4
extremities with a type 2 RLD, 3 extremities with a
type 3 RLD, and 10 extremities with type 4 RLD.
A single surgeon (C.A.G.) and therapist (A.M.)
evaluated each patient at an average 41 months (minimum 6 months) after pollicization.
Objective Measurements
Four recently reported, age-dependent measurements
of normal thumb size relative to the size of the index
finger (expressed as a percentage) were utilized in
this investigation.7 In this study, the pollicized index
finger represented the thumb and was compared with
the most radial adjacent digit (i.e., the long finger).
The contralateral hand was not used for comparison
due to the frequent incidences of bilateral involvement in RLD and an abnormal thumb on the opposite
side.10 The previously published measurements7 included relative length, girth, and width. Although the
previously reported length measurements related
thumb length to both the length of the proximal
phalanx and the length of the entire adjacent digit, in
the current study, we utilized only length relative to
the proximal phalanx because our surgical technique
adjusts the length of the pollicized digit equal to the
proximal interphalangeal (PIP) joint of the adjacent
finger.11
The hand therapist performed three, objective
measurements in a standardized fashion7: (1) the
relative length of the adducted pollicized digit to the
long finger proximal phalanx (expressed as a percentage), (2) the relative girth of the pollicized digit
compared with the long finger (expressed as a percentage), and (3) the relative nail width of the pollicized digit compared with the long fingernail width
(expressed as a percentage).
Figure 1. An example of deviation of the PIP joint in the

pollicized digit.
Additionally, angulation of the pollicized digit PIP

and distal interphalangeal (DIP) joints were measured (note that the PIP and DIP joints represent the
MP and IP joints of the pollicized thumb).
Finally, the therapist evaluated the shape of the
eponychial fold of the pollicized digit as normal
(type 1), asymmetrical (type 2), and asymmetrical
with deficient pulp on one side (type 3) (Fig. 1).
Subjective Assessment
A Visual Analog Scale (VAS) was used to evaluate
the overall appearance of the pollicized digit. It was
completed by the caregiver (typically mother or father), hand therapist (AM), and surgeon (CAG). Each
evaluator was asked to score the pollicized digits
appearance on a line with markings from 0 to 10. A
score of 10 represented a normal appearing thumb; a
score of 5 represented an abnormal appearing digit
that still resembled a thumb; and a score of 0 represented a severely malformed thumb.
Additionally, the caregiver, hand therapist, and
surgeon identified the features that negatively affected the appearance (and VAS score) of the pollicized digit compared with a normal thumb. The
choices included: too long, too short, too wide, too
narrow, abnormal nail shape, and angulation. The
objective measurement results were not available to
any rater at the time of subjective assessment, and
Goldfarb et al / Aesthetic Outcome of Pollicization
1033
Table 1. Relative Length, Girth, and Nail Width of Pollicized Digits Compared With Normal Thumbs
Relative length to proximal phalanx

Relative girth
Relative nail width
Age-Matched
Normal Thumbs*
Pollicized
Digits
% Difference Between
Pollicized Digit and Normal Thumb
71%
132%
104%
90% 26%
92% 8%
96% 9%
() 24%
() 30%
() 8%
p .01
p .0001
p .0001
*Goldfarb, et al. JHS 2005;30:1004 1008.
each evaluation was performed independently of the

other two. The reasons provided by each rater were
tabulated and compared with the objective measurements and VAS scores.
A statistical analysis was performed to assess the
data. All comparisons were performed using agematched data for normal thumb size. The chi-square
and unpaired t-tests were used to determine significance. Additionally, an analysis of variance was performed to compare the VAS scores between raters.
Finally, the subjectively identified abnormal features
and the objective measurements were assessed in
relation to the VAS scores (Pearson correlation).
Significance was set at .05.
Results
Objective Data
The relative length, girth, and nail width of the pollicized digits compared with normal thumbs are provided in Table 1. As noted, the pollicized digits were
longer than normal thumbs, but the girth and nail
width were smaller. All size differences were statistically significant.
Relative length. The average length of the pollicized digits relative to the long finger proximal phalanx was greater than normal thumbs, 90% compared
with 71% (p .0003). Thus, the pollicized digits
were 24% longer than normal thumbs relative to the
adjacent proximal phalanx. In 26 of the 31 pollicized
digits were longer than the age-matched normal
thumbs.
Relative girth. All of the pollicized digits had
smaller girths than the age-matched normal thumb
average. The average girth of the pollicized digits
relative to the long finger was 30% smaller than the
relative girth of normal thumbs, 92% compared with
132% (p .0001).
Nail width. The average nail width of the pollicized digits relative to the long finger was only
slightly (8%) narrower than the normal thumb-toindex finger ratio, 96% compared with 104%; nev-
ertheless, the difference was statistically significant

at p .0001. In 26 of 31 pollicized digits, the nail
width ratio was narrower than the ratio in the agematched normal thumbs.
Angulation. The results are noted in Table 2. Nineteen pollicized digits were angled at the PIP joint, the
DIP joint, or both; 12 had no angulation. Nine PIP
joints were angled an average of 9 (range 520),
15 DIP joints were angled an average of 8 (range
530); 5 pollicized digits were angled at both
joints. Seven were angled more than 5; 6 at either
the PIP or DIP joint and 3 at both joints.
Eponychial fold. The eponychial fold was type I
(normal) in 29 pollicized digits (94%) and type II
(asymmetrical) in two pollicized digits (6%). None
of the pollicized digits was type III (marked asymmetry with pulp defect).
Subjective Data
The average VAS score for all pollicized digits was
6.6, ranging from 4.4 to 9.7. The average VAS score
was 7.3 (1.8) for the caregiver, 6.0 (1.1) for the
therapist, and 6.4 (1.9) for the surgeon. The VAS
scores of the caregivers were significantly higher
than those of the therapist (p .04); there were no
other significant differences between the groups.
An assessment of the association between VAS
scores and objective measurements demonstrated that
as the size of the pollicized digit (i.e., length, girth, nail
width) increased, both the surgeon and the therapist
rated the pollicized digit appearance significantly higher
(i.e., higher VAS score) (Table 3). None of the caregiver measurements correlated with the VAS scores.
This statistical correlation between VAS scores and
Table 2. Joint Angulation
PIP Joint
DIP Joint
Both Joints
# Joints
Angulated
Average
Deviation ()
# Joints With
>5 Deviation
9
15
5
9
8
NA
4
2
7
1034
Table 3. Associations of VAS Scores and Clinical Measures

VAS Caregiver
r
p
Relative Length Pollicized Digit/
Long Finger Proximal Phalanx
Girth Thumb/Index
Nail Width Thumb/Index
0.25
0.03
0.22
0.17
0.88
0.23
VAS Therapist
p
0.38
0.49
0.5
0.03
0.005
0.004
VAS Surgeon
p
0.48
0.57
0.31
0.007
0.0009
0.09
r, correlation coefficient, () value implies negative correlation.
objective measurements suggests that the VAS score

evaluations of the surgeon and therapist have increased
validity over those of the caregivers.
Subjective reasons for altered appearance compared with normal. The primary and secondary
reasons cited by the caregiver, the therapist, and the
surgeon for lower VAS scores, thus detracting from
a normal appearance of the pollicized digit, were
length, girth, and angle.
Too long. Four pollicized digits were considered to
be too long by the caregiver, 11 by the therapist, and
8 by the surgeon. These responses represented a total
of 15 different pollicized digits. The average relative
length of these 15 was 96% compared with 81% for
the remaining 16 digits (p .05). However, there
was no significant difference in the average VAS
score (6.7 vs. 6.5).
Too short. Five pollicized digits were considered
to be too short by the caregiver, 4 by the therapist,
and 2 by the surgeon. These responses represented a
total of only 7 pollicized digits. Given the small
number of pollicized digits labeled as too short, no
further analysis was conducted.
Too wide. This response was not given as a primary or secondary response by any rater for the VAS
score.
Too narrow. Fourteen pollicized digits were considered to be too narrow by the caregiver, 18 by the
therapist, and 23 by the surgeon. These responses
represented a total of 26 different pollicized digits.
The average relative girth of these 26 was 89%
compared with 90% for the remaining 5, and the
comparative VAS scores were 6.5 vs. 6.7. Neither
difference was significant.
Angulation. Twelve pollicized digits were considered to be angulated by the caregiver, 6 by the
therapist, and 3 by the surgeon; these responses rep-
resented a total of 15 different pollicized digits.

Twelve of the 15 had measurable angulation at 1 or
both joints, including all 7 with greater than 5 angulation. Interestingly, 3 of the 15 had no measurable
angulation; in all 3 cases, the perceived angulation
was noted only by the caregiver. Of the remaining 16
pollicized digits for which angulation was not a negative feature, 5, in fact, had measurable angulation of
less than 5.
The angulation at the PIP and DIP joints of the 15
pollicized digits for which angulation was a negative
feature averaged 5. The angulation of the PIP and
DIP joints of the remaining 16 was 1 (p .003).
The average VAS score was not significantly different at 6.4 and 6.7, respectively.
Discussion
The purpose of this study was to compare the physical features and aesthetic appearance of pollicized
digits with normal thumbs.
Using objective measurements, the pollicized digits investigated in this study were significantly longer
than normal thumbs, the distal tip of the pollicized
digits was 90% of the length of the proximal phalanx
of the adjacent finger compared with 71% in normal
thumbs. The pollicized digits were also significantly
narrower, having a girth that was 8% less than the
adjacent finger, rather than 32% greater, as seen in
normal thumbs. Finally, the nail width was slightly
smaller than normal thumbs, being 4% narrower than
the adjacent fingernail compared with 4% wider than
in normal thumbs.
Subjectively, none of the 31 thumbs in this study
was considered to have the appearance of a normal
thumb, according to averaged VAS score assessments of the surgeon, therapist, and patient caregivers. The VAS scores of the caregivers were consistently higher than those of the surgeon and therapist,
suggesting that the surgeons and therapists assessments of the results of the procedure may underestimate the familys degree of satisfaction. On the other
Goldfarb et al / Aesthetic Outcome of Pollicization
hand, the VAS score of the surgeon and therapist

statistically correlated with objective measurements
for length, girth, and nail width, whereas those of the
caregivers showed no correlation. This finding challenges the validity of using parental or caregiver assessments in evaluating pollicization outcomes.6 The subjective assessment data also indicated that the
features of the pollicized digit that detracted from its
appearance as a normal thumb included its being too
long, too narrow, and or angled at either the PIP or
DIP joint, particularly angulation greater than 5 (a
surprising number, 12 pollicized digits, were angulated at one joint). Of these three, detracting features,
only digit length can be controlled by the surgeon at
the time of surgery; girth and joint angulation are
inherent components of the index finger.
Those pollicized digits considered too long extended nearly to the PIP joint of the adjacent finger
(ie, 96% of the proximal phalanx), which is the
length recommended by Buck-Gramcko11 whereas
the remaining pollicized digits that were not considered too long had lengths approaching those of
normal thumbs (81% vs 71%). Although it may be
appropriate for surgeons to consider reducing the
length of the pollicized digit from an aesthetic point
of view, such a consideration is probably inappropriate from a functional aspect. The added relative
length of the pollicized digit likely has functional
advantages, particularly in light of the recognized
marked reduction of pinch strength of pollicized digits.1 A shorter pollicized digit would potentially have
less function. Although we do not recommend that
the pollicized digit extend beyond the PIP joint, we
continue to adjust the length of the pollicized digit to
the level of or slightly proximal to the PIP joint as an
appropriate compromise between function and aesthetics.
It is arguable that a subjective evaluation of the
pollicized digit should include other assessment parameters, such as the abduction angle of the pollicized digit relative to the adjacent long finger, position at which the web skin attaches to the pollicized
digit,12 degree of rotation of the pollicized digit, etc.
We did not include these assessments because there
are no reliable methodologies to measure them. We
have chosen, therefore, to use objective measurements of size and VAS scoring as the basis for our
assessment. Likely, additional subjective assessment
parameters will be included in future studies.
Comparison of the results of this study with previous studies is difficult because most have provided
minimal objective description of the physical fea-
1035
tures and limited subjective assessment of the aesthetic outcome.

Staines et al reported on functional outcomes in 10
patients who underwent pollicization at a mean of 2.8
years (minimum 1 year after surgery) and suggested
that . . . parent questionnaires generally revealed
satisfaction with appearance and lack of ridicule
from others with good social interaction (p. 1319).3
Manske and McCarroll reported on 40 pollicizations
and noted that appearance was improved as the
thumb and 3-fingered hand was . . . less noticeable
and objectionable than a four-finger hand without a
thumb (p. 610).4 Egloff and Verdan reported on 30
pollicizations and reported that all of the thumbs
looked better, but there was no objective data to
support this claim.5
Two studies investigated the long-term functional
outcome after pollicization. Manske et al reported on
28 pollicizations, but appearance was not addressed.1
Kozin et al reported on 10 patients with an average
follow-up of 9 years and noted the length of the
pollicized digit averaged 6 mm (range, 218 mm)
distal to the adjacent PIP joint. The assessment methodology and the age/size relationship were not addressed.2 The authors reported that the parent, therapist, and surgeon agreed that the additional length of
the pollicized digit detracted from the appearance,
but no other data were provided.
Percival described a 22-point scale for the assessment of pollicization outcome based on function
(15 points), sensation (3 points), and appearance (4
points).6 The appearance score included one point
for length (if within 0.5 cm of the adjacent PIP joint),
2 points for position (based on palmar abduction of
45 to 80 and rotation of 90 to 160), and 1 point
for parental satisfaction.6 Although this system
attempts to provide an objective assessment of appearance, it is quite difficult to apply for several
reasons. First, using an absolute distance (0.5 cm) for
the length of the pollicized digit compared with the
long finger is inappropriate because of the marked
differences in digit size related to hand size, patient
age, and growth. Second, although position is an
important concept, objective measurements of palmar abduction and rotation is quite challenging and,
we believe, potentially inaccurate. Percival described
no methodology for performing these measurements.
Finally, although parental satisfaction is quite important, satisfaction was not defined or quantified in
this study; furthermore, we have not encountered
many parents who are unhappy or unsatisfied
after their child had undergone this procedure. As
1036
noted in our current study, the validity of caregiver

assessment is challenged by the absence of statistical
correlation between VAS scores and objective measurements. Therefore, we do not believe that including parental satisfaction as a point in the scoring
system helps to differentiate outcomes.
The difficulties we have highlighted in Percivals
system reflect the complexity in objectively assessing
subjective outcome in the childs hand. We have
been unable, thus far, to find a means to objectively
assess positioning, web space appearance, or rotation
of the pollicized digit.
A weakness of this study is the small sample
group; however, this is an inherent weakness in any
study of pollicization outcome. Additionally, we
used normal data (i.e., age-matched normal data of
thumb-to-index finger measurements) instead of using the contralateral extremity. We felt that given the
high rate of bilateral involvement, it would be inappropriate to use the contralateral extremity for comparison purposes. Furthermore, we compared the size
of the pollicized index finger to the radial-most digit,
the long finger (not the thumb and index finger as in
normal patients). We feel that this comparison is
appropriate because any assessment of size and shape
is contextual, and the appearance of the thumb is
always in relation to the associated hand and digits.
Finally, we did not have preoperative VAS data.
These data would likely have confirmed an improved
postoperative appearance of the hand (as has been
previously suggested); however, an assessment of
aesthetic improvement was not the purpose of this
investigation. Rather, we sought to evaluate aesthetic
outcome and the features affecting it.
May 25, 2007.
article.
Corresponding author: Charles A. Goldfarb, MD, Washington University Orthopedics, 660 South Euclid, Campus Box 8233, St Louis, MO
63110; e-mail: goldfarbc@wudosis.wustl.edu.
0363-5023/07/32A07-0014$32.00/0
doi:10.1016/j.jhsa.2007.05.028
References
1. Manske PR, Rotman MB, Dailey LA. Long-term functional
results after pollicization for the congenitally deficient
thumb. J Hand Surg 1992:17A:1064 1072.
2. Kozin SH, Weiss AA, Webber JB, Betz RR, Clancy M,
Steel HH. Index finger pollicization for congenital aplasia
or hypoplasia of the thumb. J Hand Surg 1992;17A:880
884.
3. Staines KG, Majzoub R, Thonby J, Netscher DT. Functional
outcome for children with thumb aplasia undergoing pollicization. Plast Reconst Surg 2005;116:1314 1323.
4. Manske PR, McCarroll HR Jr. Index finger pollicization for
a congenitally absent or nonfunctioning thumb. J Hand Surg
1985;10A:606 613.
5. Egloff DV, Verdan CI. Pollicization of the index finger for
reconstruction of the congenitally hypoplastic or absent
thumb. J Hand Surg 1983;8:839 848.
6. Percival NJ, Sykes PJ, Chandraprakasam T. A method of
assessment of pollicization. J Hand Surg 1991;16B:141
143.
7. Goldfarb CA, Gee AO, Heinze LK, Manske PR. Normative
values for thumb length, girth, and width in the pediatric
population. J Hand Surg 2005;3A:1004 1008.
8. Bayne LG, Klug MS. Long-term review of the surgical
treatment of radial deficiencies. J Hand Surg 1987;12A:169
179.
9. James MA, McCarroll HR Jr, Manske PR. The spectrum of
radial longitudinal deficiency: a modified classification.
J Hand Surg 1999;24A:11451155.
10. Goldfar CA, Manske PR, Busa R, Mills J, Carter P, Ezaki M.
Upper-extremity phocomelia reexamined: a longitudinal
dysplasia. J Bone Joint Surg 2005;87:2639 2648.
11. Buck-Gramcko D. Congenital malformations of the hand
and forearm. London, Philadelphia. Churchill Livingstone,
1998:379 402.
12. Manske PR, McCarroll HR Jr. Reconstruction of the congenitally deficient thumb. Hand Clin 1992;8:177196.
Isolated Wedge Osteotomy of the Ulna for

Mild Madelungs Deformity
Yann Glard, MD, Andr Gay, MD, Franck Launay, MD,
Didier Guinard, MD, Rgis Legr, MD
From the Department of Plastic and Reconstructive Surgery, Hpital de la Conception, Marseille, France;
and the Department of Pediatric Orthopaedic Surgery, Hpital dEnfants de la Timone, Marseille, France.
Purpose: Madelungs deformity is a characteristic pattern of anterior-ulnar bowing of the radius

and a dorsally prominent ulnar head. Even if this deformity is associated with a certain degree of
functional impairment, patients are satisfied with their function and mainly complain about the
appearance of their wrists. The purpose of this study was to report a new surgical procedure
(shortening combined with a slight anterior angulation osteotomy of the ulna) aiming to improve
the appearance of the wrist and to relieve pain if present without compromising the function of
the wrist. This technique is suitable for mild cases of Madelungs deformity.
Methods: This is a retrospective study of 4 wrists in 3 patients. All patients had a mild form of
Madelungs deformity (without any dislocation of the lunate). Even if it was not their primary
motivation to have surgery, all of the patients preoperatively experienced some wrist pain. An
anterior angulation and shortening osteotomy of the ulna shaft was performed through a dorsal
medial approach and fixed with a dynamic compression plate.
Results: At 24 months follow-up, all of the patients were satisfied with the appearance of their
wrists and forearms. The distal radioulnar joint was congruent radiologically in all cases, and the
range of active pain-free forearm rotation improved.
Conclusions: This technique seems to be safe and reliable in mild cases of Madelungs deformity.
(J Hand Surg 2007;32A:10371042. Copyright 2007 by the American Society for Surgery of the
Hand.)
Key words: Madelungs deformity, ulnar osteotomy.
adelungs deformity has various degrees of

severity and functional impairment.1 Madelungs deformity is a rare condition without
any available assessment of its incidence or prevalence
in the literature.2 It is defined as a characteristic pattern
of anterior-ulnar bowing of the radius and a dorsally
prominent ulnar head3 with a complex incongruency of
the distal radioulnar joint (Fig. 1A, B; Fig. 2). The term
anterior bowing is inappropriate, because in the pronated position, the direction of the bowing might be posterior. Nevertheless, it is commonly used in the literature. A partial closure of the distal radial growth plate
may cause this specific deformity.4,5 It may be isolated
or part of a syndrome such as dyschondrosteosis.6 8 It
may be inherited with autosomal dominance and variable penetrance9,10 or may be sporadic.10 Clinical features are a dorsal prominence of the ulnar head and a
certain degree of functional impairment.10,11 Range of
motion is decreased in flexion-extension, radial and

ulnar deviation, and forearm rotation.10,11 The age of
onset of the disease is reported to be late childhood or
teenage years.11 Conservative treatment is not effective,
failing to control or correct the deformity.1113 Our aim
is to describe a surgical procedure (isolated anterior
angulation and shortening osteotomy of the ulna) to
improve the appearance of the wrist and to relieve
pain if present without compromising the function of
the wrist in selected patients with mild forms of
Madelungs deformity who complain mainly about
the appearance of their wrists. We report a short
series of 4 wrists in 3 consecutive patients, with 24
months follow-up.

The study design was retrospective. To be included in
our series, patients had to meet the following criteria: to
1037
1038
extensor carpi ulnaris was dissected respecting a rectangular flap with a radial base in the dorsal carpal
ligament for further dorsal stabilization. The ulnar shaft
was shortened 6 cm proximal to the base of the styloid
process and fixed using a contoured dynamic compression plate after anterior angulation and shortening osteotomy of the ulna (Fig. 3B). The amount of ulnar
resection was planned on preoperative x-rays to bring
the ulnar variance close to 0. A slight anterior tilt was
performed to attempt a reduction of the distal radioulnar
joint. The extensor carpi ulnaris was stabilized using the
rectangular flap previously dissected. A protective cast
preventing the forearm rotation was applied for 3
weeks. The plate was removed after the ulnar union was
achieved. A postoperative view is shown in Figure
4A, B.
In each patient, the following items were recorded:
age at surgery, gender, involved side, preoperative and
24-months postoperative clinical and radiologic evaluation. The clinical evaluation was assessed preoperatively and postoperatively as follows: active pain-free
range of motion in pronation, supination, flexion, extension, radial deviation, and ulnar deviation. Pain was
assessed preoperatively and postoperatively using a personal pain score derived from the system described by
Jiranek et al,14 modified from Cooney et al.15 In this
comprehensive rating system, the pain is assessed using
a 30-point score (Table 1). All of the patients completed
the pain score 2 times: once prior to their surgery and
another time at 24 months follow-up evaluation. Even-
Figure 1. Preoperative radiograph in a patient with Madelungs deformity: (A) AP view, (B) lateral view.
be diagnosed with a mild form of Madelungs deformity (with no lunate dislocation, but with volar tilt of
the distal radius on the lateral radiograph of the wrist
and with increased radial inclination on the anteroposterior radiograph of the wrist), to be skeletally mature,
to have been operated on using the technique described
in the current document, and to have a follow-up of at
least 2 years. Four wrists in 3 consecutive patients were
included. There were 1 man and 2 women. The mean
age at surgery was 29 years, ranging from 27 to 32
years.
Surgical Procedure
All of the patients had the same surgical procedure. The
distal third of the ulna and the radioulnar joint were
reached through a dorsal approach, respecting the dorsal ulnar cutaneous nerve (Fig. 3A). The dislocated
Figure 2. Clinical preoperative appearance of the wrist in

Madelungs deformity (patient 2).
Glard et al / Ulnar Osteotomy in Madelungs Deformity
1039
Figure 3. Intraoperative view (patient 2). (A) Dorsal approach of the distal ulnar third and the radioulnar joint. (B) Fixation using
a contoured plate after anterior angulation and shortening osteotomy of the ulna.
tually, a tenderness at the distal ulna and at the distal

radioulnar joint was looked for preoperatively and postoperatively.
The radiologic evaluation was assessed preoperatively and postoperatively as follows: the distal radioulnar joint incongruency was assessed based on anteroposterior (AP) and lateral views of the involved wrist.
On the lateral view, the distal radioulnar joint was
defined as subluxated if less than 50% of the surface of
the ulnar head was superimposed with the distal radius.
A subluxated joint was noted , and a nonsubluxated
and nondislocated joint was noted . The ulnar variance was assessed preoperatively and postoperatively
according to the project a line technique described by
Steyers and Blair.16 The angle of the distal ulna at the
osteotomy site was assessed postoperatively on the lateral view.
Satisfaction of patients postoperatively was assessed
on the basis of their perception of the cosmetic result
according to a personal rating system derived from the
system described by Gangopadyay and Packer.17 Patients were asked to grade their wrists with a score
between 0 and 10, based on the following factors: (a)
length of the scar, (b) prominence of the radial head, (c)
interference with activities of daily living, (d) how
noticeable the deformity was to them, and (e) how
noticeable the deformity was to others. Each item is
rated as 0 (poor), 1 (medium), or 2 (good). A score of
between 9 and 10 was classified as excellent, between 7
and 8 as good, between 5 and 6 as fair, and below 5 as

poor.
The comparison was made based on preoperative
and 24 months follow-up data. Because of the very
small size of the sample, a simple description without
any statistical test was performed.
Results
Overall results are shown in Table 2. The data indicated
out that, at 24 months follow-up evaluation, the active
pain-free range of motion in pronation and supination
improved. Conversely, the active pain-free range of
motion in flexion, extension, radial deviation, and ulnar
deviation did not improve but did not worsen. The pain
score was higher postoperatively (meaning an improvement in pain). A tenderness at the distal ulna and at the
distal radioulnar joint was found in all cases preoperatively but in none postoperatively. The subluxation rate
of the distal radioulnar joint was 100% preoperatively
Table 1. Pain Score

Narcotic medication needed
Pain every day
Pain during gripping or impact loading
Aching after heavy work
Aching more than once a month
Aching once a month or less
No pain
0 poor; 30 excellent.
0
6
10
16
22
26
30
1040
Figure 4. Postoperative radiograph in patient 2 with Madelungs deformity: (A) AP view, (B) lateral view.
and 0% at 24 months follow-up. The ulnar variance

averaged 4 mm preoperatively and 1 mm postoperatively. The mean angle of the distal ulna at the
osteotomy site was 17. All of the patients were satisfied with the appearance of their wrists and their forearms at 24 months follow-up. At follow-up, all of the
patients rated their wrists as good according to our
personal satisfaction scale. The results of our personal
satisfaction scale are given in Table 3.
Discussion
The natural history of Madelungs deformity toward
osteoarthritis is not clear.11,18 Patients may suffer from
osteoarthritis late in their lives.11,19 Nevertheless, the
functional complaint is not common in patients with
Madelungs deformity. In these patients, the main motivation to have surgery is the appearance of the wrist.11
Conservative treatment is not effective, failing to control or correct the deformity.1113 Many surgical procedures were described in the literature. Dobyns et
al1 defined 3 groups of procedures in grown patients:
radial procedures,20,21 ulnar procedures,2226 and
combined radial and ulnar procedures.20,2228 Nevertheless, few procedures have been separately validated with a clinical series in the literature. To date,
we found only 6 series in which postoperative results
in patients with Madelungs deformity are clearly
given.10,11,20,23,24,29 The surgical procedure used is
homogenous in only 3 of these series. Bruno et al20
published a retrospective series of 9 adult patients.

All of them were suffering from ulnar-sided wrist
pain. They all had the same procedure: an ulnar
shortening osteotomy (without any attempt to reduce
the distal radioulnar joint subluxation). The authors
reported postoperative improvement in pain in all
cases without any notable changes in the range of
motion. The appearance of the wrist was not assessed
in this work.
dos Reis et al published a large series of 25 wrists in
18 patients.11 In 9 patients, the surgery was performed
due to pain and disability. For the other 9 patients, it
was done due to appearance. The procedure used was a
radial wedge osteotomy, combined with an ulnar shortening. The authors showed a notable postoperative improvement in pain, appearance, forearm rotation and
grip strength, and a slight improvement in the flexionextension range of motion. More recently, Harley et al
published a series of 26 wrists in 18 patients.29 The
surgical procedure used has been described in a previous work.30 It was a surgical release of the Vickers
ligament (abnormal thickened volar ligament that tethers the lunate and the triangular fibrocartilage complex
to the abnormal area of the volar-ulnar radial epiphysis
and metaphysis) associated with a dome osteotomy of
the distal radius. The authors showed improvement in
pain, appearance of the wrist, forearm supination and
wrist extension, with no loss of pronation or flexion. In
Glard et al / Ulnar Osteotomy in Madelungs Deformity
Table 2. Overall Results Showing Range of

Motion of the Wrist Pre- and Post-Operatively
Patient
Pronation
Preop
Postop
Supination
Preop
Postop
Flexion
Preop
Postop
Extension
Preop
Postop
Radial deviation
Preop
Postop
Ulnar deviation
Preop
Postop
Pain
Preop
Postop
Radioulnar dislocation
Preop
Postop
45
70
50
60
55
70
60
65
35
70
40
60
40
70
45
75
61
59
72
73
65
60
70
75
32
30
36
49
40
35
41
40
0
5
5
5
4
5
6
5
15
20
17
14
20
19
22
23
10
26
16
22
16
26
16
26
1
0
1
0
1
0
1
0
Preop, preoperatively; Postop, postoperatively.
the other 3 series available (Ranawat et al,23 Nielsen,24

and Murphy et al10), both radial and ulnar procedures
were variably associated causing some confusion about
the exact postoperative results.
Our series is made of selected patients with mild
forms of Madelungs deformity without completely dislocated lunate bone. Pain was not a criterion for inclusion. Nevertheless, we do not perform pure cosmetic
bone surgery. Therefore, all of the patients experienced
some kind of wrist pain preoperatively, as shown in
Table 1 (even if it was not their primary motivation to
have surgery). All of the patients preoperatively had an
ulnar-sided wrist pain and tenderness at the distal ulna
and at the distal radioulnar joint. In these patients who
complain mainly about the appearance of their wrists,
we chose to perform a reduction osteotomy of the ulna
diaphysis with anterior angulation aiming to reduce the
radioulnar incongruency and thereby improve the appearance of the wrist. The exaggerated palmar and ulnar
tilt of the distal radial joint surface is left undisturbed.
The aim was not to restore the whole anatomy of the
carpus but only to improve the distal radioulnar joint
anatomy, appearance, and function.
Our work assessed 4 points. Our results showed that
1041
there was a partial postoperative improvement in the

range of motion, with a forearm rotation improvement,
but without any effect on radial deviation, ulnar deviation, or flexion-extension. There was a postoperative
improvement of the pain, there was a radiologic postoperative improvement of the distal radioulnar joint
congruency, and overall there was an improvement of
the appearance of the wrist without any deterioration of
the appearance of the forearm at the osteotomy site.
This simple procedure seems to be safe and reliable in
selected patients with a mild form of Madelungs deformity. Union occurred within 4 months in all cases.
Although this was not the primary goal of this study, we
observed that no remodeling of the osteotomy site of
the ulna occurred during the follow-up period, and there
was no change in the radiographic relationship of the
radius and the ulna. Our series should not be compared
with those of dos Reis et al11 and Harley et al,29
because our operative procedure focuses only on the
distal ulna for Madelungs deformity, whereas dos Reis
et al11 used a combined radial and ulnar procedure and
Harley et al29 used a radial osteotomy associated with a
release of the Vickers ligament. Thus, the only series
available in the literature giving some postoperative
results after isolated ulnar procedure for comparison is
the one by Bruno et al.20 For the function of the wrist,
our procedure seems to have some similar outcomes
than the one described by Bruno et al20: there is no
dramatic postoperative change in the range of motion,
but the pain is relieved. Our work assessed the cosmetic
outcomes, which Bruno et al20 did not, and patients in
our series were satisfied with the appearance of their
wrists. The restricted range of motion in flexion-extension is not the main complaint in these selected patients.
We think that this procedure may be useful in selected
patients with a mild form of Madelungs deformity who
complain mostly about the appearance of their wrists
but are satisfied with its function. It is very important to
highlight that this simple procedure is suitable for mild
Table 3. Satisfaction Scale
Patient
Length of the scar

Prominence of the radial head
Interference with activities of
daily living
How noticeable the deformity
is to the patient
How noticeable the deformity
is to others
Total
0
2
1
2
1
1
0
2
2
8
2
8
2
7
2
8
1042
forms of Madelungs deformity, and we do not recommend performing it in case of severe Madelungs deformity with a completely dislocated lunate. We have
the feeling that these patients mainly complain about
the prominence of their ulnar head. They do not want to
know if the distal radioulnar joint is dislocated or subluxated. The only thing they see is that their wrist has a
hump on the ulnar side, and they want it off. We have
the feeling that the operation we are reporting addresses
this issue without compromising the function of the
wrist.
Further investigations with a larger sample of patients must be performed to assess the clinical effectiveness of this procedure, but this preliminary study
indicated that our procedure seems to be safe and has
not generated any postoperative joint stiffness or pain at
24 months follow-up in these patients with mild Madelungs deformity.
11.
12.
13.
14.
15.
16.
17.
18.
Received for publication January 19, 2007; accepted in revised form May
17, 2007.
article.
Corresponding author: Dr. Yann Glard, Department of Plastic and
Reconstructive Surgery, Hpital de la Conception, 147 Bd Baille, 13005
Marseille, France; e-mail: yann.glard@gmail.com.
0363-5023/07/32A07-0015$32.00/0
doi:10.1016/j.jhsa.2007.05.015
References
1. Dobyns JH, Doyle JR, Von Gillern TL, Cowen NJ. Congenital anomalies of the upper extremity. Hand Clin 1989;5:
321342; discussion 339 340.
2. Flatt AE. A test of a classification of congenital anomalies of
the upper extremity. Surg Clin North Am 1970;50:509 516.
3. Arora AS, Chung KC, Otto W. Madelung and the recognition of Madelungs deformity. J Hand Surg 2006;31A:177
182.
4. Vickers D, Nielsen G. Madelung deformity: surgical prophylaxis (physiolysis) during the late growth period by resection of the dyschondrosteosis lesion. J Hand Surg 1992;
17B:401 407.
5. Munns CF, Glass IA, LaBrom R, Hayes M, Flanagan S,
Berry M, et al. Histopathological analysis of Leri-Weill
dyschondrosteosis: disordered growth plate. Hand Surg
2001;6:1323.
6. Felman AH, Kirkpatrick JA Jr. Madelungs deformity: observations in 17 patients. Radiology 1969;93:10371042.
7. Golding JS, Blackburne JS. Madelungs disease of the wrist
and dyschondrosteosis. J Bone Joint Surg 1976;58B:350
352.
8. Gelberman RH, Bauman T. Madelungs deformity and dyschondrosteosis. J Hand Surg 1980;5A:338 340.
9. Mohan V, Gupta RP, Helmi K, Marklund T. Leri-Weill
syndrome (dyschondrosteosis): a family study. J Hand Surg
1988;13B:16 18.
10. Murphy MS, Linscheid RL, Dobyns JH, Peterson HA. Ra-
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
dial opening wedge osteotomy in Madelungs deformity.

J Hand Surg 1996;21A:10351044.
dos Reis FB, Katchburian MV, Faloppa F, Albertoni WM,
Laredo Filho J Jr. Osteotomy of the radius and ulna for the
Madelung deformity. J Bone Joint Surg 1998;80B:817 824.
Henry A, Thorburn MJ. Madelungs deformity. A clinical
and cytogenetic study. J Bone Joint Surg 1967;49B:
66 73.
Lamb D. Madelung deformity. J Hand Surg 1988;13B:3 4.
Jiranek WA, Ruby LK, Millender LB, Bankoff MS, Newberg AH. Long-term results after Russe bone-grafting: the
effect of malunion of the scaphoid. J Bone Joint Surg 1992;
74A:12171228.
Cooney WP, Bussey R, Dobyns JH, Linscheid RL. Difficult
wrist fractures. Perilunate fracture-dislocations of the wrist.
Clin Orthop Relat Res 1987:136 147.
Steyers CM, Blair WF. Measuring ulnar variance: a comparison of techniques. J Hand Surg 1989;14A:607 612.
Gangopadhyay S, Packer G. A comparative study between
longitudinal and T incisions for dorsal plating of the distal
radius. J Hand Surg 2003;28B:568 570.
Mansat M, Lebarbier P, Cahuzac JP, Gay R, Pasquie M.
[Madelungs disease. A study of nine wrists operated on
(authors transl)]. Ann Chir 1979;33:669 675.
Schmidt-Rohlfing B, Schwobel B, Pauschert R, Niethard
FU. Madelung deformity: clinical features, therapy and results. J Pediatr Orthop B 2001;10:344 348.
Bruno RJ, Blank JE, Ruby LK, Cassidy C, Cohen G, Bergfield TG. Treatment of Madelungs deformity in adults by
ulna reduction osteotomy. J Hand Surg 2003;28A:421 426.
Fernandez DL, Capo JT, Gonzalez E. Corrective osteotomy
for symptomatic increased ulnar tilt of the distal end of the
radius. J Hand Surg 2001;26A:722732.
Darrow JC Jr, Linscheid RL, Dobyns JH, Mann JM III,
Wood MB, Beckenbaugh RD. Distal ulnar recession for
disorders of the distal radioulnar joint. J Hand Surg 1985;
10A:482 491.
Ranawat CS, DeFiore J, Straub LR. Madelungs deformity.
An end-result study of surgical treatment. J Bone Joint Surg
1975;57A:772775.
Nielsen JB. Madelungs deformity. A follow-up study of 26
cases and a review of the literature. Acta Orthop Scand
1977;48:379 384.
Matev I, Karagancheva S. The Madelung deformity. Hand
1975;7:152158.
Watson HK, Pitts EC, Herber S. Madelungs deformity. A
surgical technique. J Hand Surg 1993;18B:601 605.
White GM, Weiland AJ. Madelungs deformity: treatment
by osteotomy of the radius and Lauenstein procedure.
J Hand Surg 1987;12A:202204.
Salon A, Serra M, Pouliquen JC. Long-term follow-up of
surgical correction of Madelungs deformity with conservation of the distal radioulnar joint in teenagers. J Hand Surg
2000;25B:2225.
Harley BJ, Brown C, Cummings K, Carter PR, Ezaki M.
Volar ligament release and distal radius dome osteotomy for
correction of Madelungs deformity. J Hand Surg 2006;31A:
1499 1506.
Harley BJ, Carter PR, Ezaki M. Volar surgical correction of
Madelungs deformity. Tech Hand Up Extrem Surg 2002;6:
30 35.
Surgical Treatment of the Pediatric

Trigger Finger
Donald S. Bae, MD, Samir Sodha, MD, Peter M. Waters, MD
From Crystal Run Healthcare, Rock Hill, NY; and the Department of Orthopaedic Surgery, Childrens
Hospital, Boston, MA.
Purpose: The purpose of this investigation is to assess the efficacy of a standardized surgical
technique in the treatment of symptomatic trigger fingers in pediatric patients.
Methods: A retrospective study was performed of 18 consecutive patients with 23 trigger
fingers treated at our institution between 1996 and 2006. Average age at the time of
presentation was 4.5 years (range, 112 years). Involved digits included 2 index, 12 long, 3
ring, and 6 small fingers. All patients had surgical treatment consisting of A1 pulley release
and resection of a single slip of the flexor digitorum superficialis (FDS) tendon. Average
clinical follow-up evaluation was 43 months (range, 3111 months).
Results: In almost half of the cases, triggering was noted to occur at the level of the FDS
tendon decussation. In 9 cases, specific tendon pathology was observed, including fusiform
thickening, nodular thickening, calcific tendonitis, and cyst formation. Overall, 21 of 23
(91%) fingers demonstrated successful resolution of triggering without recurrence after surgical treatment. One patient had recurrent triggering, which was successfully treated by a
second procedure to resect the remaining FDS slip. Another patient was successfully treated
with excision of an aberrant muscle belly from the FDS. Both of these patients remained
asymptomatic after their revision procedures. No other complications were observed. All
patients returned to full activities, and 17 of 18 (94%) patients were satisfied with the results
of surgery at most recent follow-up evaluation.
Conclusions: The pediatric trigger finger may be safely and predictably treated by surgical
release of the A1 pulley and resection of a single FDS tendon slip. (J Hand Surg 2007;32A:
10431047. Copyright 2007 by the American Society for Surgery of the Hand.)
Key words: Children, trigger finger.
he results of surgical treatment for pediatric trigger fingers have been less predictable than those
for trigger thumbs and for trigger digits in adults,
with recurrence rates reported as high as 44%.1 Historically, information regarding this uncommon condition
has come in the form of case reports and case series
combining both trigger thumbs and fingers.111 Recent
reports by Cardon et al and Tordai and Engkvist have
described a variety of surgical approaches in efforts to
achieve better outcomes, including widening of the
flexor digitorum superficialis (FDS) decussation, A1
and partial A2 pulley division, and partial or complete
resection of the FDS tendon.1,2 Abnormal anatomic
relationships between the FDS and flexor digitorum
profundus (FDP) tendons have been postulated to cause
triggering of digits in the pediatric population. Trigger
fingers have also been associated with central nervous

system disorders and mucopolysaccharidosis in children.3 As a result, more extensive surgical exposures
and interventions than simple A1 pulley releases have
been advocated.13 Indeed, Cardon et al have previously reported successful results with combined A1
pulley release with partial or complete resection of the
FDS in 6 patients.1
The purpose of this investigation is to assess the
efficacy of A1 pulley release and resection of one slip
of the FDS in the treatment of a larger consecutive
series of symptomatic trigger fingers in children.

A retrospective study was performed of 23 trigger
fingers in 18 consecutive children treated using a
1043
1044
standardized surgical technique. Patients were identified by querying the Department of Orthopaedic
Surgery computer database for all patients having
surgical treatment for trigger fingers. During this
study period, all patients having surgical release were
treated using the technique described later. Patients
with persistent triggering and/or fixed flexion contractures of the affected digit(s) associated with functional limitations despite observation, therapy, and
splinting were deemed candidates for surgery. All
patients were observed for a minimum of 6 months to
determine if spontaneous resolution might occur,
with the exception of those who presented with
locked, flexed digits associated with functional limitations. No children had prior surgery, and none had
concomitant medical conditions predisposing them to
trigger digits. Patients with trigger thumbs, predisposing medical conditions (eg, mucopolysaccharidoses), and children over 12 years of age were excluded.
From 1996 to 2006, 18 patients with 23 trigger
fingers had surgical release at our institution (Table
1). The average age at time of surgery was 4.5 years
(range, 112 years). There were 8 boys and 10 girls.
The involved digits included 2 index, 12 long, 3 ring,
and 6 small fingers. In 8 cases, the radial slip of the
FDS tendon was resected, whereas the ulnar slip was
removed in the other 15 cases. Average clinical follow-up evaluation was 43 months (range, 3111
months).
A standardized surgical technique was used in all
cases. After the administration of general anesthesia
and perioperative antibiotics, a Bruner-type incision
was made over the A1 pulley. After identification
and protection of the adjacent neurovascular structures, the A1 pulley was incised. After A1 pulley
incision, intraoperative traction on the flexor tendon
proximal to the A1 was then performed. If persistent
triggering or locking of the finger was observed,
attention was then directed to identify any anatomic
anomalies of the flexor tendon apparatus, particularly
of the decussation of the FDS. Resection of one slip
of the FDS tendon was then performed. If tendon
pathology, such as a nodule or thickening, was identified within one of the FDS slips, this slip was
resected; otherwise, the ulnar slip was chosen arbitrarily for resection. Prior to closure, full passive
motionas determined by passive digital extension
and flexion via traction on the flexor tendons proximal to the A1 pulleywithout triggering was confirmed. Postoperatively, a bulky soft dressing was
applied, and early motion was initiated after wound

healing.
The medical records of all 18 patients were carefully reviewed. Clinical presentation, associated
medical conditions, intraoperative findings, and postoperative course were noted. To maximize follow-up
evaluation, attempts were made to contact all patients. Twelve of the 18 patients were reached via
telephone interview or mailed a questionnaire. Specifically, patients/families were asked the following:
(1) Was there any recurrent triggering after the index
procedure? (2) Was there any pain or functional
limitations with activities of daily living involving
the affected digit? (3) Were any additional surgical
procedures performed on the affected digit (eg, revision release, scar revision, tenolysis, tendon repair,
etc.)? (4) Were the patients/parents satisfied with the
results of surgery? For those patients who were unavailable for telephone or mailed a questionnaire
follow-up evaluation, this information was obtained
from review of the medical records. This study was
approved by the Committee on Clinical Investigation
of our Institutional Review Board.
Results
As itemized in Table 1, tendon pathology was noted
intraoperatively in 9 digits. In 3 cases, the radial FDS
slip had fusiform thickening. Calcific tendonitis of
the ulnar slip of the FDS was noted in 1 case, as
confirmed by intraoperative inspection and subsequent histopathologic evaluation. In 2 cases, one of
the FDS slips was hypoplastic, resulting in relative
narrowing of the FDS decussation and thus a mechanical size mismatch between the decussation and
the triggering FDP tendon. There was a nodule noted
in one of the FDS slips in 2 digits. A cyst was
discovered at the decussation of the FDP tendon in
another case, which was subsequently excised.
Twenty-one of the 23 (91%) trigger digits went on
to successful healing with full return to function and
no recurrent triggering or pain at most recent follow-up evaluation. There were 2 cases of recurrent
triggering after surgical treatment. One patient (patient 12) had a second procedure 4 months after the
index operation to resect the remaining FDS tendon
slip. The other patient (patient 17) was discovered to
have an aberrant muscle belly arising from the FDS
tendon at the time of revision release. Both patients
had full return of motion without triggering after
their revision procedures. No other intra- or postoperative complications were noted. Interestingly, 1
patient (patient 5) presented with a new-onset index
Table 1. Patient Data
Patient
1
2
3
4
5
6
7
8
9
10
Affected
Digit
Structures Released
Intraoperative Findings
4
3
1
2
3
3
8
1
12
9
1
1
1
L long
L small
R long
L long
R long
L long
L index
R long
R long
L long
R ring
R ring
L long
A1 pulley, radial FDS slip

A1 pulley, ulnar FDS slip
4
5
10
4
4
9
R small
L small
L long
R small
L small
R ring

9
4
L index
R long

Nodule in radial FDS slip

NA
NA
Thickened radial FDS slip
NA
NA
NA
NA
NA
NA
NA
NA
Hypoplastic ulnar FDS
slip
NA
NA
Cyst of radial FDS slip
Nodule in ulnar FDS slip
Calcific tendonitis ulnar
FDS slip
NA
3
2
R small
L long

Hypoplastic ulnar FDS

slip
NA
11
12
13
14
15
16
17
18
NA, not applicable; Y, yes; N, no.
Follow-Up
Evaluation
(Months)
Recurrence?
Pain?
Functional
Limitations?
Patient/Family
Satisfaction?
Other
Procedures?
102
63
50
3
111
111
37
92
3
3
92
3
3
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
N
N
N
N
N
N
N
N
N
N
N
N
N
39
31
36
74
74
3
N
N
Y
N
N
N
N
N
Y
N
N
N
N
N
N
N
N
N
Y
Y
N
Y
Y
Y
N
N
Revision release
N
N
N
3
3
N
N
N
N
N
N
Y
Y
55
9
Y
N
N
N
N
N
Y
Y
N
N
Revision release
of aberrant
muscle belly
N
Bae, Sodha, and Waters / Trigger Fingers in Children
Age at
Surgery
(Years)
1045
1046
trigger finger 6 years after successful release of bilateral long trigger fingers; the index finger was not
triggering at the time of initial presentation and treatment of the long fingers.
At most recent follow-up evaluation, all patients
had returned to full scholastic and play activities
without functional limitations. Seventeen of 18
(94%) patients denied pain and were satisfied with
their procedure, including one of the patients requiring revision release. No additional operative procedures were performed on the affected digits in this
patient cohort.
Discussion
The exact etiology of the pediatric trigger finger
remains unknown. Several authors have described
flexor tendon abnormalities accounting for triggering
in their cases.9,10,1214 Whereas constriction at the
A1 pulley is more commonly the cause of triggering
in adults, it has been noted that nodular thickening or
fusiform swelling of the flexor tendon is more frequently found in pediatric trigger fingers.6,7,9,13,15
Calcifications or granulations within the tendon have
also been described less commonly.16,17 Furthermore, anatomic aberrations of the FDS terminal slips
and flexor tendon chiasm may also contribute to
mechanical triggering.1,2 All of these findings were
noted in the current investigation (Table 1). Given
the multiple etiologies of the pediatric trigger finger,
a diverse spectrum of surgical treatments have been
proposed.
Whereas trigger thumbs in children respond predictably to A1 pulley release, pediatric trigger fingers
frequently fail to resolve with this procedure
alone.1,2,12 As a result, surgical treatment has largely
been decided on a case-by-case basis, depending
upon the intraoperative findings encountered. There
are few reports of surgical strategies dealing with
trigger fingers in children.
Tordai and Engkvist advocated additional procedures, such as widening of the chiasm of the FDS
terminal slips and partial division of the A2 pulley as
needed, if A1 pulley release alone was insufficient.2
In their report, of the 5 digits that had A1 pulley
release alone, 2 developed recurrent triggering and 1
had a delayed response to surgery, with persistent
triggering seen for 6 months after surgery. The other
5 digits had widening of the FDS decussation and
partial A2 pulley release in addition to the standard
A1 pulley release. Two of these 4 cases had residual
triggering.
Cardon et al similarly recommended that the sur-
gery be tailored according to the intraoperative findings.1 In 6 of the 16 patients without concomitant
medical conditions, A1 pulley release was combined
with some type of FDS slip resection. Four had a
single slip resected, 1 had both slips removed, and
another had an A3 pulley release with FDS slip
resection. There were no recurrences or surgical failures. This technique of terminal FDS slip excision
has also been used with good results in children with
mucopolysaccharidoses.3 Although these studies
suggest that FDS terminal slip resection is effective,
there is little published information regarding the
success of this technique in a consecutive series of
patients.
In the current investigation, a standardized surgical technique involving A1 pulley release and resection of a single slip of FDS tendon was used in a
consecutive series of 23 digits, yielding 91% success
and 94% patient/family satisfaction. Indeed, use of
this standardized technique yielded predictable, successful outcomes regardless of the pathoanatomy encountered at the time of surgery. This unified surgical
strategy and high predictability encompass the main
inherent advantages of the proposed technique.
Of note, intraoperative confirmation of adequate
release of the offending anatomic structures was
based on inspection of the digit with manual passive
extension and flexion via traction on flexor tendon
proximal to the A1 pulley. No additional incisions in
the palm or forearm were used, and thus any pathology proximal to the tendon sheath may be missed
using this technique. Indeed, 1 recurrence was due to
an anomalous muscle belly arising from the FDS
tendon proximal to the A1 pulley. Although more
proximal pathoanatomy must be suspected in cases
of recurrence, further study is required to determine
the utility of routine palm/forearm exploration in
these rare cases.
There were a number of limitations to the current
investigation, including its retrospective design and
relatively small sample size. The sample size, however, may be attributed to the relative rarity of this
condition and does not differ from previously published reports. Furthermore, follow-up evaluation
clinical data were limited to 3 months in a number of
patients. Although this follow-up evaluation may be
considered relatively short, all of these patients had
regained motion and strength without recurrent triggering or other surgical complications and thus were
discharged from care. Therefore, 3-month follow-up
evaluation was sufficient to assess pain, restricted
motion, success (or failure) of surgical release, and
Bae, Sodha, and Waters / Trigger Fingers in Children
return to daily activities. In efforts to reduce the risk

of recurrent triggering, a slip of the FDS tendon was
removed, which raises the theoretical concern that
there may be loss of digital flexion and grip strength.
Although digital motion and grip strength were not
quantified, none of the patients in this series complained of grip weakness or functional limitations at
most recent follow-up evaluation. This is also consistent with other published reports. Finally, the results of our study only apply to those children presenting with trigger finger without predisposing
diagnoses or associated inflammatory flexor tenosynovitis.
In conclusion, surgical treatment with A1 pulley
release and resection of a single slip of the FDS
tendon resulted in successful resolution of trigger
digits in 91% of cases, regardless of the underlying
anatomic abnormalities encountered at the time of
release. This simple, safe, and straightforward technique allows children to return to their previous
functional status while minimizing the risk of recurrent trigger and need for reoperation.
Received for publication January 4, 2007; accepted in revised form May
31, 2007.
This study was conducted at the Department of Orthopaedic Surgery,
Childrens Hospital, Boston, MA.
article.
Corresponding author: Peter M. Waters, MD, Department of Orthopaedic Surgery, Childrens Hospital, 300 Longwood Avenue, Hunnewell
2, Boston, MA 02115; e-mail: peter.waters@childrens.harvard.edu.
0363-5023/07/32A07-0016$32.00/0
doi:10.1016/j.jhsa.2007.05.031
References
1. Cardon LJ, Ezaki M, Carter PR. Trigger finger in children.
J Hand Surg 1999;24A:1156 1161.
1047
2. Tordai P, Engkvist O. Trigger fingers in children. J Hand

Surg 1999;24A:11621165.
3. Van Heest AE, House J, Krivit W, Walker K. Surgical
treatment of carpal tunnel syndrome and trigger digits in
children with mucopolysaccharide storage disorders. J Hand
Surg 1998;23A:236 243.
4. Mulpruek P, Prichasuk S, Orapin S. Trigger finger in children. J Pediatr Orthop 1998;18A:239 241.
5. Nemoto K, Nemoto T, Terada N, Amako M, Kawaguchi M.
Splint therapy for trigger thumb and finger in children.
J Hand Surg 1996;21B:416 418.
6. Paaske BP, Soe-Nielsen NH, Noer HH. Release of trigger
finger in children: long term results. Scand J Plast Reconstr
Hand Surg 1995;29:65 67.
7. Rodgers WB, Waters PM. Incidence of trigger digits in
newborns. J Hand Surg 1994;19A:364 368.
8. Steenwerckx A, De Smet L, Fabry G. Congenital trigger
digit. J Hand Surg 1996;21A:909 911.
9. Tsuyuguchi Y, Tada K, Kawaii H. Splint therapy for trigger
finger in children. Arch Phys Med Rehabil 1983;64:7576.
10. Weilby A. Trigger finger: incidence in children and adults
and the possibility of a predisposition in certain age groups.
Acta Orthop Scand 1970;41:419 427.
11. Wood VE, Sicilia M. Congenital trigger digit. Clin Orthop
1992;285:205209.
12. Hirata H, Fujisawa K, Sasaki H, Morita A, Matsumoto M.
Congenital triggering of the index finger at the A2 pulley.
J Hand Surg 1996;21B:609 611.
13. Fahey JJ, Bollinger JA. Trigger-finger in adults and children.
14. Dobyns JH. Trigger digits. In: Green DP, ed. Operative hand
surgery. 3rd ed. New York: Churchill Livingstone, 1993:
374 378.
15. Oda Y, Uchida Y, Kojima T, Sugioka Y. Congenital, multiple, bilateral, trigger digits in a child. Int Orthop 1993;17:
20 22.
16. Seiler JG III, Kerwin GA. Adolescent trigger finger secondary to post-traumatic chronic calcific tendonitis. J Hand Surg
1995;20A:425 427.
17. Chia J, Pho RW, Sinniah R. Congenital trigger thumb
caused by intratendinous granulation tissue. J Hand Surg
1996;21B:612 613.
Replantation of Completely Amputated

Thumbs With Venous Arterialization
Lijie Tian, MD, Furong Tian, MD, Feng Tian, MD, Xiaochuan Li, PD,
Xianglu Ji, MD, Jiao Wei, BD
From the Department of Hand Surgery, Shengjing Hospital, China Medical University, Shenyang, China; and
the Department of Hand Surgery, Feng Tian Hospital, Shenyang Medical University, Shenyang, China.
Purpose: To report a new method of replanting completely amputated thumbs with venous
arterialization.
Methods: In 6 replantation surgeries of completely amputated thumbs performed during the
period 1999 2003, the proximal artery was anastomosed with a vein of the amputated part
to establish inflow and the proximal vein was anastomosed with several other veins in the
amputated part to establish outflow. This was because the proper palmar digital arteries were
seriously injured or anastomosis of proper palmar digital arteries failed many times.
Results: All the replanted thumbs survived, regained good sensory and motor functions, and
showed no difference from thumbs replanted conventionally.
Conclusions: Venous arterialization may salvage otherwise unreplantable thumbs. (J Hand
Surg 2007;32A:1048 1052. Copyright 2007 by the American Society for Surgery of the
Hand.)
Key words: Complete amputation, replantation, thumb, venous arterialization.
ome amputated digits cannot be replanted when

the arteries are severely injured. Generally, the
proximal arteries are anastomosed to the proper
digital arteries of the amputated part, and the proximal
veins are anastomosed to the veins of the amputated
part to restore circulation to the amputated part.
The purpose of this study is to report 6 replantation
surgeries of completely amputated thumbs performed
between 1999 and 2003 when the proper digital
arteries of the amputated digit were severely injured
and could not be repaired even by a long segmental
graft. In these 6 cases, the proximal artery was anastomosed with a vein in the amputated part to establish inflow, and a proximal vein was anastomosed to
several other veins in the amputated thumb. There
was 100% survival of all 6 thumbs and ultimately
useful function.

General Data
Six patients (5 men and 1 woman, 556 years old;
mean, 31 years) were reviewed. The causes of injury
included 1 sharp instrument injury, 2 rotary avul1048
sions, 1 contusion, 1 crush with avulsion, and 1

degloving injury. The level of amputation was at the
interphalangeal joint with the entire hand skin
avulsed (1), at the metacarpophalangeal joint (3), at
the carpometacarpal joint (1), and at the radiocarpal
joint (1). Five thumbs were replanted in situ replantation and 1 was replanted ectopically. The method
of venous arterialization was anastomoss between the
first dorsal metacarpal artery and a dorsal metacarpal
vein of the amputated thumb (1), anastomoss between the proper digital arteries and the palmar digital veins of the amputated thumb (2), and anastomoss between radial artery and a dorsal metacarpal
vein of the amputated thumb (3). Two veins were
repaired in 3 cases and 3 veins in 3 cases. (Table 1).
Surgical Techniques
If the proper digital arteries and principal artery of
thumb were severely injured, long segments were
injured, or arterial anastomoses failed many times,
the proximal first dorsal metacarpal artery, radial
artery, or proper digital arteries were anastomosed
with a vein in the amputated part to establish arterial
inflow. Then, 23 dorsal digital veins or dorsal meta-
Tian et al / Replantation of Completely Amputated Thumb
1049
Table 1. Six Patients Having Fingertip Transplantation

Age
Patient Gender (Years)
Type of Injury
Year of
Operation Level of Amputation
37
Sharp instrument
injury
1999
56
Rotary avulsion
2000
35
Contusion
2002
Rotary avulsion
2002
31
Contusion with
avulsion
2002
26
Injured by crush
roll
2003
Mode of Venous
Arterialization
First dorsal metacarpal

Carpometacarpal
artery dorsal
joints (including
metacarpal veins of
part of thenar
the amputated thumb
muscles)
Metacarpophalangeal Ulnar digital artery
joints
dorsal veins of the
amputated thumb
Radiocarpal joint
Radial artery dorsal
metacarpal veins of
the amputated thumb
Metacarpophalangeal Ulnar digital artery
joint
dorsal veins of
amputated thumb
Metacarpophalangeal Carpal dorsal branch of
joint
radial artery dorsal
metacarpal veins of
the amputated thumb
Radial artery
Digital articulation
cephalic vein
joint (the whole
skin of the hand
was avulsed)
carpal veins were anastomosed to establish outflow.

The treatment methods repairing the other tissues
were the same as those for conventional artery-toartery anastamoses.
Number of
Anastomosing
Vessels
2 dorsal metacarpal
veins
3 dorsal digital
veins
2 dorsal metacarpal
veins
3 dorsal digital
veins
2 dorsal digital
veins
3 dorsal carpal
veins
many blisters formed dorsally on the thumb and in 3

cases, there were a few (Figs. 1, 2). The swelling
began subsiding 1 week after surgery and disap-
Postoperative Treatment
After replantation the patients were placed in a relatively quiet and comfortable ward with room temperature maintained at around 25C. The patients
generally rested in bed for about 10 days with hands
elevated slightly above the heart. A 60-W side lamp
was 40 cm placed above the patients hand to maintain local warmth and to facilitate observation of the
hemodynamic status. A sterile syringe was used to
extract the exudates in any dorsal blisters. Low molecular dextran injection, 500 mL, intravenous drip,
twice daily; papaverine injection, 30 mg, intramuscularly, 4 times daily; and oral aspirin tablet, 100 mg,
3 times daily were administered for 10 days. The
pediatric dose was adjusted according to body
weight. A broad-spectrum antibiotic was used for 1
week. Analgesics were administered to prevent the
occurrence of vascular spasms triggered by pain.
Results
All the replanted thumbs survived. There was no
arterial crisis in any of the 6 cases but there was
evidence of venus congestion in 4 cases. In 1 case,
Figure 1. Vesiculation 2 days after surgery.
1050
Figure 2. The blisters resolved 2 weeks after surgery.
regular (high-low); and the soft, frail venous valves

swing back and forth. Some arterial blood passes
through the gaps in the valves and passes distally. In
addition, the arterial pressure is higher than the venous pressure. As compared with the artery, the wall
of venous blood vessels is thin with a low elasticity.
In the veins with a low pressure, the venous valves
can completely close so as to allow the flow into the
heart and prevent backward flow. Over time, atrophy
occurs in the valve and the arterial blood perfuses
into the distal capillary network without obstruction.
The animal experiment of treating ischemic diseases of lower extremities with arteriovenous bypass
in stages by Zhang1 also indicated the valves had
insufficiency, even atrophy, and the arterial blood
perfused the vessels inversely due to the persistent
impact of arterial blood flow and the gradual dilation
of venous segments within 2 weeks after surgery.
The microcirculation was normal at week 4.
What is the pathway of arterial blood entering
microcirculation inversely? The microcirculation of
the replanted thumb with venous arterialization and
that of the arterialized venous flaps have both similar
and different characteristics. The arterialized venous
flap circulation experiment by Chen et al2 showed
that, after arterial blood passed through the anasto-
peared by 2 weeks. In 2 cases, there was mild swelling and no blisters. Follow-up visits were performed
on all of the patients. The 2-point discrimination was
6 mm in 4 cases, 8 mm in 1 case, and 10 mm in 1
case. The patients had cold intolerance in the thumb
during cold weather 1 year after surgery and these
symptoms had disappeared by a year later. All the
replanted thumbs regained good sensory and motor
functions and showed no difference from thumbs
replanted with appropriate artery-to-artery and veinto-vein anastomoses (Fig. 3; Table 2).
Discussion
The Microcirculation in Replanted
Thumbs With Venous Arterialization
Is Different from Physiologic Circulation
How does the arterial blood pass venous valves and
reach the thumb tip in a retrograde fashion? In our
opinion, the palmar and dorsal veins of thumbs are
venules (diameter 2 mm) without valves. Dorsal
metacarpal veins are medium-sized (diameter 2
mm). Although there are a few valves in them, they
are soft and frail. When the arterial blood is perfused
into the venous system of the replanted thumb via the
anastomosis, it first depends on the contraction and
dilation of the heart; the systolic and diastolic flow is
Figure 3. Circulation was good at 9 months after surgery.
Tian et al / Replantation of Completely Amputated Thumb
1051
Table 2. Results and Complications

Motion Arc ()
Cold
Intolerance
Patient
Metacarpophalangeal
Joints
Interphalangeal
Articulations
Two-Point
Discrimination of
Finger Pulp (mm)
Pinch
Strength (kg)
Sweating of
Finger Pulp
2 years after
the operation
1
2
3
4
5
6
40
15
30
30
15
40
70
40
50
70
50
40
6
6
10
6
6
8
5
3
1
1
4
3.5
Normal
Normal
Light
Normal
Normal
Light
Yes
Yes
Yes
Yes
Yes
Yes
mosis and perfused the veins, the blood passed

through the venous system, small veins, and venules.
In the small veins and venules with flow in the
normal direction, the flow was fast and streamlined,
whereas in the small veins and venules with backflow
function, the flow was slow. Most of the flow reached
the venules in backflow function via the communicating branches between venules. Part of the flow
directly entered capillaries through venules (preferential
channel) inversely. In addition, some flow entered arterioles via arterioles and venules; however, this pathway was not the principal one. Therefore, the opening
of a great number of communicating branches and the
slow blood flow in the communicating branches are the
characteristics of the early changes in microcirculation
in the arterialized flaps at the early stage.
The research on the structure and function of veins
after arteriovenous bypass by Jiang et al3 suggested
the arterial blood entering veins went into the physiologic microcirculation channel by flowing into the
arterial side via arteriovenous shunts before reacting
the capillaries.
How is the metabolism in the replanted thumbs
with venous arterialization realized? The arterial
blood flows into the venous system and enters arterioles via venules and thoroughfare channel or arteriovenous communicating branches, then enters the
capillary network for metabolism and returns via the
veins. The arterial blood also can directly flow into
the capillary network from venules, perform metabolism, and return via another venule. The research by
Yuan et al4 also showed the walls of venules under
papillary of skin were similar to those of capillary
vessels, which played an important role in metabolism, and could replace capillary vessels in venous
arterialization.
Compared with other digits, the thumb is short and
large in diameter, the radial thumb vein and ulnar
thumb vein go along both sides of the back of the
thumb and form a first venous arch at the metacar-
pophalangeal joint and form dorsal metacarpal veins

proximally. Several dorsal metacarpal veins form a
secondary venous arch and run into the cephalic vein
on the back of the wrist. The diameter of the dorsal
digital vein can reach 1.0 mm or more at the metacarpophalangeal joint, 1.52.0 mm at the metacarpal,
and 2.0 mm or more at the carpometacarpal joint.
Because the diameters of veins are large and the
anastomosed vessels are unobstructed, the arterial
blood can inversely perfuse the venous system. Because there are several veins, 23 veins can be anastomosed for backflow while 1 vein is used for arterialization. The unobstructed venous return can lower
the venous pressure, alleviate venous congestion, reduce plasma exudation, thereby improve the microcirculation in the amputated thumb and decrease the
possibility of thrombosis in the venules. The thumb
can tolerate prolonged ischemia and requires little
nutrition, although part of the arterial blood perfused
inversely returns to the heart via vein-vein communicating branches without passing through the capillaries and allowing for metabolic exchange. If the
back of the thumb was pricked with needles immediately after surgery, however, there was continuing
exudation of bright-red blood, indicating most of the
arterial blood could be perfused into the capillary
network so as to satisfy the nutritional requirements
for the survival of the replanted thumb at the early
stage. In addition, there is abundant venous network
in the dorsal and palmar parts of the thumb providing
an anatomic base for retrograde perfusion of arterial
blood.
The Method of Replanting
Completely Amputated Thumbs With
Venous Arterialization Enlarges the Indication
In the past, extensive arterial damage precluded
thumb replantation. Only hallux flap or second toe
transplantation could be conducted to reconstruct the
thumb with the high risks associated with long op-
1052
eration time, serious injury, long course of treatment,

and high expense. Venous arterialization allows performing replantation of completely amputated
thumbs that was previously considered to be impossible, and this surgery has a low risk with little injury,
a short operation time, and a good functional recovery.
In thumb replantation the ulnar digital artery is
often used for anastomosis because it is a little larger
than the radial digital artery. Because the position of
the thumb is different from that of other digits, however, the operation is awkward, and because the
blood vessels are small, the surgery may fail. If
anastomosing the proper digital arteries fails, artery
transposition from the index finger is not required
and venous arterialization can be used. This surgery
is especially suitable for the thumb amputated at the
carpometacarpal joint level because there are many
veins with 1.52.0 mm diameters in this area. The
first dorsal metacarpal artery is near the radial artery
and dorsal metacarpal veins, the operation is convenient, the patency rate of vascular anastomosis is
high, and the operation time is short.
Degloving injuries of the hand are a challenge for
surgeons. The arteries remain on the hand and there
is no artery for blood supply in the avulsed flap.
Although there are some treatment methods (such as
thumbnail flap, dorsalsis pedis flap, anterior lateral
femoral flap, and abdominal pedicle flap), the effects
of these are not satisfactory. Case 6 proved that if the
superficial vein of the back of the hand was kept in
the avulsed flap, venous arterialization could be used
to rescue the entire thumb and most of the flap.
Sufficient Blood Return Is the Effective
Solution for Disturbance in Venous Return
In 4 cases, 1 dorsal metacarpal artery or radial artery
was anastomosed with the dorsal metacarpal veins
for blood supply, the arterial pressure was high, and
2 other dorsal metacarpal veins were used for venous
drainage. The diameters of the blood vessels were
large and the anastomoses remained patent. In case 1,
there was notable swelling in the amputated thumb
and vesiculation on the back of the thumb, indicating
the small veins and venules in the amputated thumb
were dilated, the pressure increased, and there was a
tendency of obstruction of venous return at the early
stage in this nonphysiologic circulation. There was
little exudation, however, at the border of skin in the

amputation plane after surgery, and relative balance
between arteries and veins could not be maintained
by the capillary hemorrhage and tissue fluid exudation at the border of the flap as at the early stage in
arterialized venous flap. Collateral circulation was
established and the swelling gradually subsided 1
week later.5
In 2 cases, proper digital arteries and palmar digital veins were anastomosed, the arterial pressure was
low, and 3 dorsal digital veins were used for venous
drainage. There was no notable swelling or blisters
indicating that there was sufficient venous return and
the circulation of arteries and veins was balanced.
Therefore, more veins should be anastomosed to
ensure sufficient drainage.
Previously, venous arterialization was only suitable for replanting the distal segment or the tip of the
amputated thumb. We have no experience in its use for
other digits. The characteristics of the microcirculation
with venous arterialization, the mode of nutriment, the
histologic changes in the walls of arterialized veins, the
operative precautions, and the treatment for postoperative swelling should be further investigated.
Received for publication September 15, 2006; accepted in revised form May
11, 2007.
article.
Corresponding author: Lijie Tian, MD, Department of Hand Surgery,
Shengjing Hospital, China Medical University, Shenyang 110004, China;
e-mail: tianlijie_sy@sina.com.
0363-5023/07/32A07-0017$32.00/0
doi:10.1016/j.jhsa.2007.05.011
References
1. Zhang P. Clinical vascular surgery. Beijing: Science Press,
2003:327329.
2. Chen J, Liang J, Chen Z. Experimental study on forms of
microcirculation of arterialized venous flap in rabbits. Chin J
Reparat Reconstr Surg 2002;16:170 172.
3. Jiang M, Lu M, Huang X. Structure and functions of vein
after arteriovenous bypass in stages. Chin J Exp Surg 1994;
11:289 290.
4. Yuan L, Zhong S. Applied anatomy of construction and
communication of flap veins. Chin J Microsurg 1991;1:163
165.
5. Tian L, Liu W, Wang X. Successful replantation of complete
amputation of thumb with venous arterialization: a case report. Chin J Microsurg 2000;23:160.
Digital Ischemia Due to Essential

Thrombocythemia: A Case Report
Anastasios Papadonikolakis, MD, George D. Chloros, MD,
Beth P. Smith, PhD, L. Andrew Koman, MD
From the Department of Orthopaedic Surgery, Wake Forest University School of Medicine, Winston-Salem, NC.
This report describes the case of a 34-year-old patient with essential thrombocythemia who
presented with Raynauds syndrome, was refractory to medical treatment, and developed
progressive digital gangrene. (J Hand Surg 2007;32A:10531057. Copyright 2007 by the
Key words: Cyanosis, digital ischemia, essential thrombocythemia, hand, periarterial sympathectomy, platelet apheresis, Raynauds, thrombocytosis.
ssential thrombocythemia (ET) is a rare

chronic myeloproliferative blood disorder
of unknown etiology characterized by proliferation of megakaryocytes leading to an overproduction of platelets.1 The diagnosis of ET is
established in the presence of a persistent thrombocytosis of greater than 600 109/L in the platelet
count in the absence of a known cause.
As many as two thirds of patients are asymptomatic and only receive medical attention fortuitously,
as a result of thrombocytosis found from a routine
blood cell count. The clinical course of ET is marked
by episodes of hemorrhage or thrombosis or both.
The major risk factors for thrombosis in ET are age
greater than 60 years and a history of previous thrombotic episodes. In most patients, treatment with a
single dose of aspirin usually results in a dramatic
improvement.2
A rare case of a young patient with ET and
Raynauds syndrome that was refractory to medical
treatment and progressed to digital gangrene is presented. Treatment included urgent therapeutic platelet apheresis (TP), which lowered his platelet count,
followed by periarterial sympathectomy35 with amputation of the necrotic tip of the middle finger.
This report emphasizes the importance of performing an early aggressive TP, when other medical measures fail, to rapidly decrease the platelet count to
help prevent or eventually reverse the ischemic
changes. Periarterial sympathectomy is performed
after a normal platelet count is achieved as an adjunctive procedure to increase nutritional collateral
flow. To our knowledge, there are no similar reports

in the English literature.
Case Report
The patient was a 34-year-old male with no significant past medical history other than splenectomy for
a previous motor vehicle accident. He presented with
a 2-week history of pain and bilateral bluish discoloration of the fingertips and toes. He reported that he
began to experience throbbing pain to his fingers and
toes, which started after sleeping on a cold floor. The
pain was associated with numbness and tingling with
decreased sensation to his fingers and toes and was
exacerbated by cold. He was started on aspirin and
nifedipine but experienced no significant relief. Prior
to his transfer to our institution, he was seen at an
outside hospital and found to have a platelet count of
1,034 103/L. It was determined that he had
Raynauds phenomenon related to ET because all
other causes of thrombocytosis were excluded. Treatment on hydroxyurea was initiated to reduce the
platelet count and to prevent thrombosis. The patient
was transferred to our institution for further management.
On admission, the patient had cyanosis and exquisite tenderness of his distal left second and third
fingers and transient areas of cyanosis and pain in his
toes and right hand that resolved spontaneously. Initial treatment consisted of hydroxyurea, aspirin, and
nifedipine to treat the ET.
On postadmission day 5, the pain was increasing,
there was cyanosis and discoloration of the left distal
1053
1054
long fingers. Through a fish-mouth incision, the necrotic skin and fingertip were amputated with a saw.
The middle phalanx was transected with a saw
through the mid-diaphysis, leaving the flexor digitorum superficialis intact. The digital nerves were identified and trimmed proximally. The flexor digitorum
profundus was pulled distally and transected. The
extensor mechanism was cut at the level of amputation. A periarterial sympathectomy (Fig. 3) was performed through a zigzag incision over the index
finger; both proper digital arteries were identified.
The incision was extended across the palm, and the
proper digital artery and the common digital artery to
the long finger were also identified. Vascular loops
were placed, and periarterial sympathectomy was
performed using the operating microscope under
magnification. Three months postoperatively, the patient is symptom-free with no evidence of ischemic
changes (Fig. 4A, B).
Figure 1. (A and B) Photographs showing partial necrosis of
the tip of the left index finger. There are skin changes of the
tip of the third left finger due to ischemia.
second and third digits, and a platelet count of 800

103/L was documented. At this point, the patient
was evaluated by the senior author who believed that
the patients fingertips were in danger of imminent
gangrenous necrosis. Measures to reduce the platelet
count and increase the blood flow to the distal fingertips were instituted. The patient thus received 2
TP treatments, which resulted in his platelet count
reaching 348 103/L on postadmission day 8.
There were no complications as a result of the procedure, and the toe ischemia resolved after the TP.
His digital ischemia, however, remained unchanged
(Fig. 1).
An upper extremity arteriogram was performed,
which showed multilevel occlusions of the proper
digital palmar arteries of the second and third digits
(Fig. 2). There was decreased flow to the distal
phalanx. Nitroglycerin was administered, and a repeat angiogram demonstrated no significant change
in the flow to the second and third digits. At this
point, the patient was diagnosed with peripheral vascular disease secondary to ET, associated gangrene
of his left index finger, and ischemic changes of the
left middle finger.
The patient was taken to the operating room, and
an amputation of his left index finger through the
middle of the middle phalanx was performed, along
with a periarterial sympathectomy of the index and
Discussion
Essential thrombocytosis is a slowly progressive disorder characterized by long asymptomatic periods
punctuated by thrombotic or hemorrhagic events.1 It
is a diagnosis of exclusion established in patients
Figure 2. Upper-extremity arteriogram showing multilevel

occlusions of the proper digital palmar arteries of the second
and third digits.
Papadonikolakis et al / Digital Ischemia Due to Essential Thrombocythemia
Figure 3. Periarterial sympathectomy: through a zigzag volar

incision, the adventitia is cleaned of sympathetic fibers and
connections with the nerve are severed. (Reproduced by
permission from Koman LA, ed. Wake Forest University orthopaedic manual. Winston-Salem, NC: Orthopaedic Press,
2007.)
who are rarely seen by hand surgeons. Although the

exact mechanism of the disease is unknown, it is
believed that platelets derived from the abnormal
megakaryocytes do not function properly and contribute to the clinical features of bleeding and thrombosis. There are approximately 1.5 to 2.4 per 100,000
individuals diagnosed annually with ET.6,7 The disease usually affects middle-aged to elderly individuals, with an average age at diagnosis of 50 60 years.
ET, however, may also affect children and young
adults. The major risk factors for thrombosis are age
older than 60 years and previous thrombotic episode,2 whereas the advent of thrombosis appears to
be unrelated to either the platelet count or hemostasis
tests.8,9
Almost two thirds of patients are asymptomatic,
and the most common symptoms at presentation are
due to disturbances of the microcirculation, particularly fingers, toes, and central nervous system manifestations including headache, dizziness, and visual
and acoustic symptoms.10 The term erythromelalgia,
specific to the myeloproliferative disorders, refers to
1055
the occlusion of the microcirculation by platelets and

is characterized by redness, congestion, and painful
burning sensations of the extremities. Symptoms are
characteristically relieved by cold or elevation of the
extremity and are exacerbated by warmth, exercise,
or standing.10 Essential thrombocythemia can also
lead to painful acrocyanosis and even peripheral gangrene.11 A single dose of aspirin usually results,
however, in reversal of the ischemic phenomena with
resolution of erythromelalgia in the vast majority of
patients within 2 to 4 days.2 Acute digital ischemia
may be caused by several entities including collagen
vascular diseases,12 Raynauds disease,13 Buergers
disease,14 peripheral atherosclerosis,15 heparin-induced
thrombocytopenia with thrombosis syndrome,16 consumption coagulopathy,17 and many others. Although
rare, hematologic disorders, however, have to be
considered in the differential diagnosis of acute digital ischemia.
The patient described in this report was atypical,
because he was free of all the aforementioned thrombosis risk factors. His painful acrocyanosis, which
progressed to digital gangrene, was probably due to
his Raynauds disease rather than to classic erythromelalgia, because his extremities were neither
warm nor congested. His symptoms were exacerbated by cold and were not relieved by aspirin.
Treatment of asymptomatic patients with ET remains controversial and largely problematic. Many
studies have shown that the degree of platelet count
elevation in ET is not an important determinant of the
Figure 4. (A and B) At 3 months, the patient is symptom-free

with no evidence of ischemic changes. The operation involved the patients left side.
1056
probability or frequency of thrombotic episodes.2

Furthermore, an acute reduction of platelet count by
TP is ineffective in the long-term, because of the
rapid rate of production of platelets.18 The high-risk
patient population (age 60 years, previous thrombotic episode), however, deserves therapeutic intervention. Cytoreductive therapy (platelet count to
600 109/L) with hydroxyurea19 or anagrelide20
has been effective in preventing additional thrombotic episodes in this patient population.
The current patient was managed on admission
with aspirin (to reverse the platelet-mediated thrombotic effect), nifedipine (to reverse the vascular
spasm), and hydroxyurea (to progressively decrease
his platelet count). At this point, TP was not performed, because it was believed that this treatment
would not provide a long-term benefit on the course
of the disease. The patients pain increased, however,
and there was apparent danger of imminent gangrene. Because medical treatment takes several days
or even weeks to lower the platelet count, TP represents an urgent intervention in patients with ET when
a rapid reduction in platelet counts is required,21 such
as in a thrombotic episode, and may result in dramatic clinical improvement,22 even recovery from
the gangrene.23 In addition, although a chemical
sympathectomy with an axillary block may be temporarily effective in controlling extremity neuropathic pain, it was not attempted in this case, as this
would not address the basis of the problem (increased
platelet count), which was responsible for the patients urgent condition.
Based on this case, TP should be initiated immediately when there are cutaneous or ischemic changes
that are not responding to alternative treatment. Further delay in treatment may lead to permanent ischemic damage or gangrene.
Periarterial sympathectomy is considered as an
adjunct to TP to (1) maximize nutritional flow by
decreasing arteriovenous shunting, (2) protect the
amputation flaps, and (3) possibly decrease the
intensity of pain. In the setting of worsening ischemic phenomena and a high platelet level, surgery
is not an option. The patient had complete resolution of his Raynauds phenomenon after sympathectomy.
This case highlights the need for appropriate timing of TP when other measures fail to limit the
ischemic phenomena. Periarterial sympathectomy
may provide a valuable adjunctive treatment option
in the setting of a normal platelet count to increase
the nutritional collateral flow.

11, 2007.
article.
Corresponding author: L. Andrew Koman, MD, Professor and Vice
Chair, Department of Orthopaedic Surgery, Wake Forest University
School of Medicine, Medical Center Blvd., Winston-Salem, NC 27157;
e-mail: lakoman@wfubmc.edu.
0363-5023/07/32A07-0018$32.00/0
doi:10.1016/j.jhsa.2007.05.010
References
1. Sanchez S, Ewton A. Essential thrombocythemia: a review
of diagnostic and pathologic features. Arch Pathol Lab Med
2006;130(8):1144 1150.
2. Fruchtman SM, Hoffman R. Essential thrombocythemia. In:
Hoffman R, Benz EJ Jr, Shattil SJ, Furie B, Cohen HJ, eds.
Hematology: basic principles and practice. 4th ed. New
York: Churchill Livingstone, 2005:12771296.
3. Koman LA, Smith BP, Pollock FE Jr, Smith TL, Pollock D,
Russell GB. The microcirculatory effects of peripheral sympathectomy. J Hand Surg 1995;20A:709 717.
4. Pollock DC, Li Z, Rosencrance E, Krome J, Koman LA,
Smith TL. Acute effects of periarterial sympathectomy on
the cutaneous microcirculation. J Orthop Res 1997;15(3):
408 413.
5. Ruch DS, Holden M, Smith BP, Smith TL, Koman LA.
Periarterial sympathectomy in scleroderma patients: intermediate-term follow-up. J Hand Surg 2002;27A:258 264.
6. Kutti J, Ridell B. Epidemiology of the myeloproliferative
disorders: essential thrombocythaemia, polycythaemia vera
and idiopathic myelofibrosis. Pathol Biol (Paris) 2001;49(2):
164 166.
7. Mesa RA, Silverstein MN, Jacobsen SJ, Wollan PC, Tefferi
A. Population-based incidence and survival figures in essential thrombocythemia and agnogenic myeloid metaplasia: an
Olmsted County Study, 1976 1995. Am J Hematol 1999;
61(1):10 15.
8. Barbui T, Cortelazzo S, Viero P, Bassan R, Dini E, Semeraro
N. Thrombohaemorrhagic complications in 101 cases of myeloproliferative disorders: relationship to platelet number and
function. Eur J Cancer Clin Oncol 1983;19(11):15931599.
9. Regev A, Stark P, Blickstein D, Lahav M. Thrombotic
complications in essential thrombocythemia with relatively
low platelet counts. Am J Hematol 1997;56(3):168 172.
10. Levine SP. Thrombocytosis. In: Greer JP, Foerster J, Lukens
JN, Rodgers GM, Paraskevas F, Glader B, eds. Wintrobes
clinical hematology. 11th ed. Philadelphia: Lippincott Williams & Wilkins, 2003:15911601.
11. Hussain S, Schwartz JM, Friedman SA, Chua SN. Arterial
thrombosis in essential thrombocythemia. Am Heart J 1978;
96(1):3136.
12. Herrick AL, Oogarah PK, Freemont AJ, Marcuson R,
Haeney M, Jayson MI. Vasculitis in patients with systemic
sclerosis and severe digital ischaemia requiring amputation.
Ann Rheum Dis 1994;53(5):323326.
13. Cooke JP, Marshall JM. Mechanisms of Raynauds disease.
Vasc Med 2005;10(4):293307.
14. Shionoya S. Buergers disease: diagnosis and management.
Cardiovasc Surg 1993;1(3):207214.
Papadonikolakis et al / Digital Ischemia Due to Essential Thrombocythemia

15. Longo GM, Friedman AC, Hollins RR, Buresh CJ, Baxter
BT. Distal radial artery lesion as a source of digital emboli.
J Vasc Surg 1998;28(4):710 714.
16. Chang JC. Review: postoperative thrombocytopenia: with
etiologic, diagnostic, and therapeutic consideration. Am J
Med Sci 1996;311(2):96 105.
17. Davis MP, Byrd J, Lior T, Rooke TW. Symmetrical peripheral gangrene due to disseminated intravascular coagulation.
Arch Dermatol 2001;137(2):139 140.
18. Goldfinger D, Thompson R, Lowe C, Kurz L, Belkin G.
Long-term plateletpheresis in the management of primary
thrombocytosis. Transfusion 1979;19(3):336 338.
19. Cortelazzo S, Finazzi G, Ruggeri M, Vestri O, Galli M,
Rodeghiero F, et al. Hydroxyurea for patients with essential
20.
21.
22.
23.
1057
thrombocythemia and a high risk of thrombosis. N Engl

J Med 1995;332(17):11321136.
Fruchtman SM, Petitt RM, Gilbert HS, Fiddler G, Lyne A.
Anagrelide: analysis of long-term efficacy, safety and leukemogenic potential in myeloproliferative disorders. Leuk
Res 2005;29(5):481 491.
Taft EG, Babcock RB, Scharfman WB, Tartaglia AP.
Plateletpheresis in the management of thrombocytosis.
Blood 1977;50(5):927933.
Schafer AI. Bleeding and thrombosis in the myeloproliferative disorders. Blood 1984;64(1):112.
Win N, Mitchell DC. Platelet apheresis for digital gangrene
due to thrombocytosis in chronic myeloid leukaemia. Clin
Lab Haematol 2001;23(1):65 66.
Aeromonas hydrophila Infection Causing

Delayed Vascular Thrombosis in a
Nearly Amputated Finger
Fernando A. Herrera, MD, Karen Horton, MD, Ahmed Suliman, MD,
Gregory M. Buncke, MD
From The Buncke Clinic and Department of Microsurgery, California Pacific Medical Center, San Francisco,
CA.
We discuss an unusual case of Aeromonas hydrophila infection in a partially amputated digit

from contaminated animal products resulting in delayed vascular thrombosis and
amputation. (J Hand Surg 2007;32A:1058 1060. Copyright 2007 by the American Society
for Surgery of the Hand.)
Key words: Aeromonas, thrombosis, partial amputation.
eromonas hydrophila is an oxidase-positive,

glucose-fermenting, gram-negative rod. This
bacterium is a ubiquitous microorganism responsible for infections in immunocompromised humans.1 It has been found in fresh and brackish water,
soil, poultry, pork, lamb, and other foodstuffs. Infections have been seen in healthy individuals exposed
to contaminated water sources with open wounds.
Aeromonas hydrophila infections are well documented in the plastic surgery literature.2,3 This microorganism has been associated with the medicinal
leech, which is used commonly in venous congested
flaps and replanted parts.4 To our knowledge, there is
only one other reported case series of Aeromonas
hydrophila infections associated with injured digits
exposed to raw meat.5 We report a case of a nearly
amputated digit from a band-saw injury in a healthy
woman who developed a delayed infection secondary
to contaminated machinery.
Case Report
A 35-year-old, otherwise healthy woman presented
with a near-amputation of her right dominant index
finger in zone II (Fig. 1). The patient was working at
her brothers butcher shop and sustained an injury
to her index finger using a band saw to cut lamb.
She was taken to a local emergency room where
she was evaluated by the microsurgery team. On
inspection, the digit was noted to have normal
capillary refill. The patient was taken to the operating room where the wound was explored under
1058
general anesthesia. The digits displayed normal

vascularity and were noted to be surviving on an
intact ulnar neurovascular pedicle. Osteosynthesis
was performed using 2 K-wires placed in a parallel
fashion. The flexor tendon was repaired using a
double opposing locking loop 4-strand technique using looped FiberWire (Arthrex, Naples, FL). The
wound was copiously irrigated followed by repair of
complex volar lacerations using 3-0 nylon. The finger was well vascularized at the completion of the
case as well as after application of the hand dressing
(Fig. 2). The next morning, the dressing was removed, and the index finger was viable with good
capillary refill; no signs of infection were present.
The patient was discharged with follow-up evaluation for daily dressing changes and continued on 1
week of prophylactic antibiotics to cover standard
skin flora (cephalosporin). Forty-eight hours after
surgery, the patient presented to the hand clinic with
wet gangrene of the finger distal to the injury (Fig. 3).
The patient was taken to the operating room where
the previous incision was opened and frankly purulent, foul-smelling fluid was expressed from the
wound. Emergent amputation was performed with
debridement of all questionable tissue. Wound cultures were taken intraoperatively. The patient was
admitted and placed on broad-spectrum antibiotics,
and daily dressing changes were performed. The
wound showed no signs of progressing cellulitis or
infection. Intraoperative tissue cultures grew 3
Herrera et al / Aeromonas hydrophila Infection in a Partially Amputated Digit
Figure 1. Partially amputated index finger.
Aeromonas hydrophila. Intravenous antibiotics were

tailored according to culture sensitivities, and the
patient went on to uneventful healing and rehabilitation.
1059
of soft tissue infections resulting from this organism.

Aggressive irrigation and debridement of devitalized
tissue and antibiotics are essential in suppressing
progression of infection. Several studies have documented the antibiotic resistance profile of Aeromonas
hydrophila.15,16 Therefore, recommended antibiotics
to treat this organism include intravenous fluoroquinolones or a third- or fourth-generation cephalosporin.17 This organism is known to cause severe
wound infections and to compromise revascularized
limbs when these injuries are sustained in an aquatic
environment.8,9,18 Therefore, when these injuries are
recognized, the clinician should alter antibiotic therapy based on mechanism of injury. To our knowledge, however, there is only one other report of hand
injuries complicated by Aeromonas infection after
exposure to contaminated meat.5 Our experience
with a delayed infection resulting in loss of a previously viable digit confirms the invasiveness of this
microorganism. Prophylactic treatment with fluoroquinolones or third- or fourth-generation cephalosporins should be instituted along with appropriate debridement of any nonviable tissue in injuries where
this invasive pathogen may be present. The risk of
Aeromonas hydrophila infection after leech therapy
is well-known among the plastic surgery community;
Discussion
Aeromonas hydrophila is a gram-negative, oxidasepositive bacterium that is ubiquitous in freshwater
and soil. This microorganism has been known to
cause many infections in humans, ranging from gastroenteritis to septic arthritis, soft tissue infections,
sepsis, and even death.6,7 Several reports of soft
tissue infections have been documented, most developing in the setting of water-related trauma.8 10
Aeromonas hydrophila has also been associated with
the medicinal leech and its use in the salvage of
congested flaps has resulted in soft tissue infection.4,11,12 In the immune-competent individual, infection is less likely to develop. Aeromonas species
have also been recognized as potential food-borne
pathogens. They have been linked to infections from
exposure to contaminated fish, shellfish, meats, and
fresh vegetables.13 Aeromonas infections are known
to be rapidly progressive and should be observed
closely when index of suspicion is high. In one study,
isolates from lamb meat were shown to produce
exotoxins at 37C.14 This may aid in the rapid spread
Figure 2. Perfused index finger after surgical repair and Kwire of distal phalanx.
1060
References
Figure 3. Necrotic distal index finger with nonviable soft

tissue.
however, soft tissue infections secondary to traumatic injuries caused by equipment used to cut contaminated meat are relatively uncommon.5 Due to the
potential dreaded complication of digit loss, a high
index of suspicion should be maintained in the setting of hand injuries with exposure to animal products just as for those injuries sustained in aquatic
environments where this organism is ubiquitous. If a
high index of suspicion is maintained and appropriate
antibiotics used to treat a more virulent infection are
employed, empirically further loss of injured digits
may be prevented. Close observation with frequent
dressing changes should also be performed to look
for early signs of progressive infection.
30, 2007.
article.
Corresponding author: Fernando A. Herrera, MD, 45 Castro, Suite
140, California Pacific Medical Center, Department of Microsurgery, San
Francisco, CA 94114; e-mail: fherrera@ucsd.edu.
0363-5023/07/32A07-0019$32.00/0
doi:10.1016/j.jhsa.2007.05.029
1. Gold WL, Salit IE. Aeromonas hydrophila infections of skin

and soft tissue: report of 11 cases and review. Clin Infect Dis
1993;16:69 74.
2. Lineaweaver WC, Hill MK, Buncke GM, Follansbee S,
Buncke HJ, Wong RK, et al. Aeromonas hydrophila infections following use of medicinal leeches in replantation and
flap surgery. Ann Plast Surg 1992;29:238 244.
3. Lineaweaver WC. Aeromonas hydrophila infections following clinical use of medicinal leeches: a review of published
cases. Blood Coagul Fibrinolysis 1991;2:201203.
4. Dabb RW, Malone JM, Leverett LC. The use of medicinal
leeches in the salvage of flaps with venous congestion. Ann
Plast Surg 1992;29:250 256.
5. Lowen RM, Rodgers CM, Ketch LL, Phelps DB. Aeromonas
hydrophila infection complicating digital replantation and
revascularization. J Hand Surg 1989;14A:714 718.
6. Geller HS, Tofte RW, Cunningham BL. Aeromonas hydrophila wound infection of the hand initially presenting as
clostridial myonecrosis. J Hand Surg 1983;8A:333336.
7. Adamski J, Koivuranta M, Leppanen E. Fatal case of myonecrosis and septicaemia caused by Aeromonas hydrophila
in Finland. Scand J Infect Dis 2006;38:11171119.
8. Semel JD, Trenholme G. Aeromonas hydrophila water-associated traumatic wound infections: a review. J Trauma
1990;30:324 327.
9. Weber CA, Wertheimer SJ, Ognjan A. Aeromonas hydrophila
its implications in freshwater injuries. J Foot Ankle Surg 1995;
34:442 446.
10. Sanger JR, Yousif NJ, Matloub HS. Aeromonas hydrophila
upper extremity infection. J Hand Surg 1989;14A:719 721.
11. Ardehali B, Hand K, Nduka C, Holmes A, Wood S. Delayed
leech-borne infection with Aeromonas hydrophila in escharotic
flap wound. J Plast Reconstr Aesthet Surg 2006;59:94 95.
12. Snower DP, Ruef C, Kuritza AP, Edberg S. Aeromonas
hydrophila Infection Associated with the Use of Medicinal
Leeches. J Clin Microbiol 1989;27:14211422.
13. Nishikawa Y, Kishi T. Isolation and characterization of
motile Aeromonas from human, food and environmental
specimens. Epidemiol Infect 1988;101:213223.
14. Majeed K, Egan A, MacRae IC. Enterotoxigenic Aeromonads on retail lamb meat and offal. J Appl Bacteriol 1989;
67:165170.
15. Lineaweaver WC, Follansbee S, Hing DN. Cefotaxime-sensitive Aeromonas hydrophila infection in a revascularized
foot. Ann Plast Surg 1988;20:322325.
16. Horii T, Morita M, Muramatsu H, Monji A, Miyagishima D,
Kanno T, et al. Antibiotic resistance in Aeromonas hydrophila and Vibrio alginolyticus isolated from a wound
infection: a case report. J Trauma 2005;58:196 200.
17. Gilbert DN, Moellering RC Jr, Eliopoulos GM, Sande MA,
eds. The Sanford guide to antimicrobial therapy 2004. 34th
ed. Hyde Park, VT: Antimicrobial Therapy, 2004:3750.
18. Noonburg GE. Management of extremity trauma and related
infections occurring in the aquatic environment. J Am Acad
Orthop Surg 2005;13:243253.
REVIEW ARTICLE
Spontaneous Flexor Tendon Ruptures of
the Hand: Case Series and Review of the
Literature
Aaron J. Bois, MD, Geoffrey Johnston, MD, Dale Classen, MD
From the Divisions of Orthopaedic Surgery and Plastic Surgery, Department of Surgery, The University of
Saskatchewan, Saskatoon, Saskatchewan, Canada.
Spontaneous flexor tendon ruptures within the hand are incompletely understood. We report 5
cases of spontaneous tendon rupture involving the flexor digitorum profundus tendon. One case
involves an abnormal intertendinous connection between the ring and small finger profundus
tendons and another involves a lumbrical muscle variant. To our knowledge, the latter has not
been reported in association with spontaneous tendon rupture. In reviewing the literature for
spontaneous flexor tendon ruptures, a total of 50 spontaneous ruptures in 43 cases was found.
The majority involve the profundus tendon of the small finger in the palm. The ruptures most often
occur during periods of peak strain but can also occur without identifiable trauma. The pathogenesis of spontaneous tendon ruptures is still unclear and is likely multifactorial. Spontaneous
flexor tendon ruptures of the hand occur more often than one might recognize. (J Hand Surg
2007;32A:10611071. Copyright 2007 by the American Society for Surgery of the Hand.)
Key words: Flexor digitorum profundus tendon, spontaneous tendon rupture, tendon anomaly, zone III.
ntratendinous flexor tendon ruptures are defined as

complete tears occurring within the tendon substance. Such ruptures are a result of intratendinous
fiber failure and occur most frequently in the small
finger.1
Spontaneous ruptures are uncommon, and the current
understanding of the basis for such ruptures is incomplete. Boyes et al2 in 1960 defined the term spontaneous tendon rupture to mean those ruptures that occurred
within the tendon substance in the absence of intrinsic
or extrinsic pathologic processes. Only 3 of the 80
tendon ruptures described in their series were spontaneous.
The purpose of this study was to report 5 cases of
spontaneous flexor tendon rupture in the hand and to
contrast the findings with those in the literature.

Data Sources and Study Selection
A literature review was conducted using PubMed and
Medline databases (January 1966 to June 2006) to obtain data on spontaneous tendon ruptures within the
flexor tendons of the hand. The search strategy included
published articles in English with the Medical Subject
Headings terms tendon injuries and spontaneous

rupture and the key words flexor tendons and flexor
digitorum profundus/superficialis. Combined searches
resulted in 154 abstracts. All studies were reviewed and
evaluated. Fourteen of these articles described spontaneous tendon ruptures within the hand as defined by
Boyes et al.2 The bibliographies of these studies were
reviewed to identify other articles that may have been
overlooked in the Medline search; 5 additional studies
were identified for a total of 19. Cases were excluded if
the history or location of the rupture was not clear.
Case Reports
A retrospective review of patients with flexor tendon
injuries identified 5 patients who sustained a spontaneous tendon rupture of the flexor digitorum profundus
(FDP) tendon as defined by Boyes et al.2 All 5 patients
were referred to our university-based hand specialists
for a presumed flexor tendon injury. The 5 patients
were all men, ranging in age from 38 to 65 years (mean,
55.8 years). The rupture occurred in the dominant hand
in 4 of the patients. The small finger was involved in 4
cases and the long finger in 1; all tendon ruptures
occurred in the palm.
1061
Yes
Longitudinal
force
65
5
Dom, dominant hand; ND, nondominant hand.
Left (ND)
Pop
56
55
3
4
65
38
1
2
M
M
Asymptomatic
Asymptomatic
Left (Dom)
Right (Dom)
FDP-5 (III)
FDP-5 (III);
anomalous
FDP
tendon
FDP-3 (III);
anomalous
lumbrical
muscle
Distal
Proximal
Origin
Gross inspection
Yes
Yes;
roughened
hamate
Normal
Normal
Gross inspection
Gross inspection
Atraumatic
Longitudinal
force
Atraumatic
Atraumatic
Proximal
Origin
FDP-5 (III)
FDP-5 (III)
Right (Dom)
Left (Dom)
Cramp
Sharp pain
Clinical
Symptoms
Gender
Age
(Years)
Case
No.
M
M
Histologic
Results
Mechanism of
Injury
Tendon-Digit
(Zone)
Relationship
to
Lumbrical
Attachment
Table 1. Rupture Characteristics of Case Series (N 5)
Case 1. A 65-year-old, right-handed, retired farmer

was assessed in the hand clinic 7 weeks after he felt
a cramp in his right palm between his little and ring
fingers. He rubbed this area and shortly thereafter
noted he could not flex his small finger. He could not
recall any previous injury to his finger. He had full
passive range of motion of his finger but had no
activity of the FDP or flexor digitorum superficialis
(FDS) of the small finger.
At surgery, the end of the distal ruptured FDP tendon to
the small finger was identified in the distal palm with the
origin of the lumbrical still firmly attached. At the level of
the superficial palmar arch, we identified the proximal end
of the ruptured profundus tendon. The rupture had occurred just proximal to the origin of the small finger
lumbrical muscle (zone III) (Fig. 1A). Macroscopically,
both the tendon and carpal tunnel appeared normal, with
no evidence of tendon attrition, and thus the tendon was
not sent for histologic examination. In addition, the superficialis tendon was intact, although it was small and thread-
Carpal Tunnel
Explored
At the time of rupture, 2 patients were involved in

activities that required forceful loading of the flexor
tendons such as lifting, whereas the other 3 patients
were at rest. All patients presented with a sudden inability to actively flex the involved finger at the level of
the distal interphalangeal (DIP) joint (passive flexion
was preserved). Three patients experienced an acute
onset of symptoms prior to their inability to flex the
involved finger and 2 were asymptomatic at the time of
flexor dysfunction. Clinically, there was no tenderness,
swelling, or ecchymosis evident in any of the cases,
which may in part be explained by the presentation of
most patients to the surgeon at an average of 3 weeks
after injury. No predisposing pathologic conditions or
bone abnormalities were found in these patients. In all
but 1 case, preoperative x-rays were normal. In 1 case,
a tiny fleck of bone just proximal to the distal phalanx
resembled an avulsion fracture fragment. All patients
were nonsmokers. A summary of patient characteristics
is presented in Table 1.
In all cases, a preoperative diagnosis of an FDP
avulsion injury from the distal phalanx was made. Surgical exploration started in zone I and extended proximally until the site of rupture was found. All 5 cases of
tendon rupture occurred within the palm of the hand
(zone III). Intraoperatively, the superficialis tendon was
present and intact, although thread-like and dysfunctional in 1 case and relatively weak in another case. The
first 3 cases described in the following sections consist
of normal FDP tendon anatomy, and the last 2 cases
described had a tendon anomaly.
Yes
Yes
Involved
Hand/
Dominance
1062
Bois, Johnston, and Classen / Spontaneous Tendon Ruptures of the Hand
1063
Figure 1. (A) Surgical exposure of the right hand: 1, rupture of the FDP-5 tendon in the palm, proximal to the origin of the small
finger lumbrical muscle; 2, superficial palmar arch. (B) Postoperative result: extension.
like. Postoperative range of motion in extension is demonstrated in Fig. 1B.

Case 2. A 38-year-old, left-handed farmer was referred to our clinic after injuring his left small finger.
Five days prior, he described attempting to open a car
door. He loaded the tips of his fingers, lifted up on the
handle, and immediately developed a sharp pain in the
palmar region of his hand. Clinical examination demonstrated an intact FDS and minimal flexion of the DIP
joint with the small finger fully extended.
Intraoperatively, proximal extension of the digital
incision was made into the midpalmar region, revealing the tendon rupture in zone III in the region of the
origin of the small finger lumbrical. The FDS tendon
was intact. Again, both the tendon and carpal tunnel
appeared normal. The tendon was not sent for histologic examination.
Case 3. A 56-year-old, left-handed businessman presented to our emergency department 4 days after playing hockey. He noted when getting changed after the
game that he could not flex the little finger of his left
hand. He could not recall any trauma or injury, nor
could he recall any specific episode of pain or discomfort within the hand or fingers. Clinical examination
demonstrated an intact FDS tendon, although weaker
than the other hand.
Intraoperatively, the tendon sheath was opened
proximal and distal to the A4 pulley, and at this level,
both the superficialis and profundus tendons were
intact. Proximal exploration revealed that the profundus tendon had ruptured just distal to the small finger
lumbrical muscle (Fig. 2). There appeared to be
evidence of tenosynovitis around the distal tendon
stump. Histologic examination of the tendon stump
was normal. There was no evidence of osteoarthritis
at the floor of the carpal tunnel or pathology involving the hook of the hamate.
Case 4. A 55-year-old, right-handed construction
worker presented to our clinic 6 weeks after he noted he
Figure 2. Surgical exposure of the left hand: 1, proximal and

distal tendon stumps of the FDP-5 tendon in zone III; 2,
fourth lumbrical muscle; 3, terminal branch of ulnar artery
and superficial branch of ulnar nerve.
1064
2
4
Figure 3. Preoperative resting posture of the right hand.

5
could not flex his right little finger. He could not recall
any previous specific injury to his finger. He had full
passive flexion of all joints of the involved digit; however, he was unable to actively flex his finger at the DIP
joint (Fig. 3).
Surgical exploration revealed that the superficialis
tendon of the small finger was intact, but the profundus
tendon of the small finger had been ruptured in the
proximal palm. The site of rupture was located at an
abnormal tendinous connection between the ring and
small fingers, just proximal to both the lumbrical attachment and superficial palmar arch (in zone III) (Fig.
4A). The distal tendon stump was found at the level of
the distal palmar crease. Proximally, in the carpal tunnel, the FDP of the ring finger appeared to be slightly
frayed but intact. Exploration of the floor of the carpal
tunnel revealed there was a lack of fascial coverage
over the hamate bone, leaving a portion of this bone
exposed and roughened (Fig. 4B). The hook of the
hamate was unremarkable. There was no abnormal
movement of the carpal bones with wrist motion. Histologic examination of the tendon stump was normal.
The tendon was reattached at the point of rupture in
zone III, re-creating the normal flexor tendon tone in the
small finger.
Figure 4. (A) Surgical exposure of the right hand: 1, distal

tendon stump of FDP-5; 2, FDS-5 tendon; 3, terminal branch of
ulnar artery and superficial branch of ulnar nerve; 4, proximal
attachment of FDP-5 to FDP-4 (anomaly). (B) Proximal dissection of the carpal tunnel: 5, the hamate bone is exposed and
appears roughened. (C) Schematic representation of the photographs in (A) and (B).
1065
Table 2. Demographic Characteristics of

Literature Review
4
5
3
Age (N 43)
Mean (range)
Gender (N 43)
Men
Women
Hand dominance (N 40)*
Dominant
Nondominant
47years (1579)
39 (90.7%)
4 (9.3%)
26 (65%)
14 (35%)
*Number of cases reporting hand dominance.

1
the proximal interphalangeal joint. The rest of the exam

was unremarkable.
At surgery, the profundus tendon to the long finger
was ruptured just distal to the superficial palmar arch.
The rupture took place at the midpoint of the third
lumbrical muscle origin. The lumbrical muscle to the
long finger was absent. An abnormally large lumbrical
muscle belly originated from the ulnar side of the long
finger FDP tendon and inserted into the deep transverse
metacarpal ligament between the long and ring fingers
(Fig. 5). The latter tendon insertion has not been previously described in the literature. The ruptured tendon
and carpal tunnel appeared normal on gross
examination.
5
3
Figure 5. (A) Surgical exposure of the left hand: 1, proximal

and distal tendon stumps of the FDP-3 tendon in zone III; 2,
ruptured third lumbrical muscle; 3, FDS-3 tendon; 4, A1
pulley of long finger flexor sheath; 5, anomalous insertion of
the third lumbrical tendon into the deep transverse metacarpal ligament. (B) Schematic representation of the photograph
in (A).
Case 5. A 65-year-old, right-handed farmer was assessed 2 weeks after he injured his left hand. While
lifting a grain bin, he felt a sudden pop in his left palmar
region just proximal to the distal palmar crease in line
with the long finger. Soon thereafter, he could not flex
the long finger at the level of the DIP joint. Clinical
examination revealed normal flexor tone and strength at
Table 3. Rupture Characteristics of Literature

Review
Clinical symptoms (N 38)*
Snap/pop
Sudden sharp pain
Cramp
Asymptomatic
Mechanism of injury (N 40)*
Longitudinal force
Repetitive blunt trauma
Atraumatic
Digit involved (N 50)
Small
Ring
Long
Index
Thumb
Tendon ruptured (N 50)
FDP
FDS
Combined FDP/FDS
FPL
Tendon site of rupture (N 50)
Zone III
Zone II
Zone IV
26
10
1
1
(68.4%)
(26.3%)
(2.6%)
(2.6%)
37 (92.5%)
2 (5%)
1 (2.5%)
31
7
5
3
3
(62%)
(14%)
(10%)
(6%)
(6%)
41
6
4/43cases
3
(82%)
(12%)
(9.3%)
(6%)
40 (80%)
7 (14%)
3 (6%)
*Number of cases reporting symptoms and the mechanism of

injury at time of spontaneous tendon rupture.
1066
Table 4. Summary of Spontaneous Tendon Ruptures in the Hand

Dominant/
Nondominant
Relationship to
Lumbrical
Attachment
FPL (II)1
FDP/FDS-3
(III)1
FDP/FDS-5
(III)1
FDP-5 (III)
ORG-2
Resisted flexion
(2) and
hyperextension
Gross
inspection
only
? N/A (1)
operative
findings failed
to explain
ruptures
ORG
Resisted flexion
Gross
inspection
only
Normal
blood at distal
palmar crease
? N/A (2)
Other ruptures:
ring (31)
thumb (3) not
discussed
Anomalous FDS
muscle
Author (Year)
No. of Cases
Boyes et al
(1960)2
3/25 (total 5
ruptures)
32, 37, &

42 y.o. M
DOM-3
Kumar et al
(1985)4
44 y.o. M
Snap was felt in

the palm
ND
Imbriglia et al
(1987)5
7/44 (7/10
confirmed)
3063 y
(48.7) M
Snap or pop
followed by
pain (7)
DOM-4
ND-3
FDP-5 (II)2
FDP-5 (III) 4
FDP-5 (IV)1
DIST-4 others
N/A
All resisted flexion
Lillmars et al
(1988)6
32 y.o. M
Snap then pain

and swelling
ND
FDP-5 (II)
N/A
Resisted flexion
Wray et al
(1989)7
54 y.o. M
ND
FDP-5 (III)
DIST
Hyperextension
Walker et al
(1990)8
53 y.o. M
ND
FDP-3 (III)
PROX
Resisted flexion
de Roos et al
(1991)9
20 y.o. M
Sudden pain
while lifting
object
Sudden pain,
then swelling
& pain in palm
? (left hand)
FDP-5 (III)
DIST
Berglund et al
(1993)10
1 (total 2
ruptures)
15 y.o. F
? (patient was
unconscious)
? (right hand)
FDP-2 (II)1
FDP-3 (II)1
Coombs et al
(1993)11
55 & 74 y.o.
M
DOM-1
ND-1
Takami et al
(1993)12
16 y.o. M
Sudden pain only

complaint (1);
asymptomatic
(1)
Dull pain initially
Nakamichi et
al (1994)13
McLain et al
(1994)14
52 y.o. M
3 (total 5
ruptures)
50, 65,
78 y.o. M
Naam (1995)1
13
Yang et al
(1998)15
23 - 63 y
(43.5); 12
M, 1 F
34 y.o. M
Snap and slight

pain in wrist
Sudden cramp
(2); pain (3)
Snap or pop (13);
pain &
swelling (50%)
Sudden sharp
pain &
swelling
Mechanism of
Injury
Carpal
Tunnel
Explored
Histologic
Results
Gross
inspection
only
Gross
inspection
only
Normal
Yes; tendon
retrieval
only
No
Flexion of ring,
hyperextension
of little finger
N/A
N/A
FDP-5 (III)2
ORG-2
Blunt trauma with

forced
hyperextension
Normal
Yes; tendon
retrieval
only (1)
ND
FDP-5 (III)
ORG
Repeated blunt
trauma to palm
? (right hand)
FDP-5 (IV)
PROX
Resisted flexion
Gross
inspection
only
Normal
DM-3
FPL (III)1
FDP/FDS-5
(III)2
FDP-3 (II)1
FDP-4 (III)7
FDP-5 (III)5
FDP-5 (III)
? or N/A
Flexion (1),
hyperextension
(1), & ? (1)
All resisted flexion
Gross
inspection
only
Normal
? N/A (1)
Hyperextension of
flexed finger
Normal
Yes; normal
DOM-9
ND-4
ND
ORG-6
PROX-1
DIST-5
ORG
N/A
General
Comments
normal
appearing
tenosynovium
rupture in palm,
thus canal not
explored
complete
tendinous
fusion of FDP-5
and 4
Rupture followed
grand mal
seizure
Small FDS of
small finger
Yes; normal
no proliferative
synovitis
evident
Yes; normal
3
no evidence of
tenosynovitis
Presenting Signs/
Symptoms
Tendon-Digit
(Zone)No.
Cases
Age (Mean)/
Gender
Table 4. Continued.
Presenting Signs/
Symptoms
Dominant/
Nondominant
Tendon-Digit
(Zone)No.
Cases
Relationship to
Lumbrical
Attachment
No. of Cases
Hang et al
(2002)16
17 y.o. F
Sudden snap,
then swelling
DOM
FDP-5 (III)
DIST
Repeated blunt
trauma to palm
Davis et al
(2003)17
76 & 79 y.o.
M
DOM-2
FDP-5 (III)2
Resisted flexion
Imai et al
(2004)18
1 (total 3
ruptures)
DOM
FPL (III)
FDS-2 (III)
FDP-2 (IV)
Atraumatic
Krishnamurthy
et al
(2005)19
Tan et al
(2005)20
Current series
(2007)
74 y.o. F;
FPL
rupture at
age 71 y
55 y.o. M
Felt crack (1);

sudden pain
(1)
Slight pain &
swelling
Sudden pop (no

pain)
DOM
FDP-4 (III)
Resisted flexion
Normal
22 y.o. M
ND
Normal
38, 55, 56,

65 (2) y.o.
M
FDP/FDS-5
(I/III)
FDP-5 (III) 4
FDP-3 (III)1
Sudden pain &

swelling
Sudden sharp
pain (1), pop
(1), cramp (1)
PROX-2
ORG-3
Resisted flexion
(2); atraumatic
(3)
Normal
DOM-4
ND-1
Mechanism of
Injury
Histologic
Results
Gross
inspection
only
Gross
inspection
only
Abundant
regenerative
fibrosis
Carpal
Tunnel
Explored
?
Yes; normal
2
Yes; normal
Yes; tendon
retrieval
only
?
Roughened
hamate
in 1 case
Dom, dominant hand; ND, nondominant hand; PROX, proximal to lumbrical origin; ORG, at the origin of the lumbrical; DIST, distal to the lumbrical origin; y.o., years old.
General
Comments
evidence of
hypertrophic
scar
Anomalous FDP
tendon 2
Perinatal brachial
plexus injury
(use of right
hand only)
Small FDS
Anomalous FDP
tendon (1) &
lumbrical
insertion (1)
Author (Year)
Age (Mean)/
Gender
1067
1068
Discussion
Most flexor tendon ruptures result from avulsion of the
FDP tendon at its insertion. True spontaneous ruptures
are infrequent. In the past 50 years, tendon ruptures of
all types, including those of the hand, have increased in
incidence in most developed countries.3 In this study,
we have reviewed the current literature (English language) over a 50-year period and found a total of 50
spontaneous ruptures of normal flexor tendon in 43
cases1,2,4 20 (see Table 4) for which the demographic
results are summarized in Table 2. A descriptive summary of the 5 cases of spontaneous rupture in this series
is presented in Table 1.
Unique historical and physical features exist to help
differentiate spontaneous tendon ruptures from the
more common avulsion ruptures. Pooling the data from
the literature review of spontaneous ruptures, in 26 of
38 (68.4%) cases, the clinical history describes the
sensation of a pop or snap, and 11 (28.9%) patients
reported a sudden sharp pain or cramp in the region of
tendon rupture, most often within the palm. Only 1 case
was reported to be asymptomatic (2.6%) (Table 3). In
this case series, 3 patients experienced an acute onset of
symptoms within the palmar region and 2 patients were
asymptomatic. In contrast, most avulsion ruptures cause
discomfort within the region of the synovial sheath of
the digit. In type 1 avulsion injuries of the FDP tendon,
the proximal tendon stump usually retracts proximal to
the digital tendon sheath, prompting a tender mass in
the palm.21 Flexor digitorum profundus tendon avulsions, however, are not typically known to be preceded
by a snap or pop in the palm.
Diagnostic imaging for differentiating spontaneous
from avulsion ruptures can also be helpful adjuncts to
the clinical picture. Both ultrasound11,22,23 and magnetic resonance imaging (MRI)24,25 of the hand provide
accurate assessment of the location, integrity, and gap
distance of the ruptured tendon ends in both early and
late clinical presentations of flexor tendon rupture. Ultrasound and MRI not only assist in making the diagnosis of tendon rupture but may also provide important
preoperative information for surgical decision-making
and planning; this information may decrease postoperative morbidity by minimizing surgical dissection.
In 1960, Boyes review of 80 flexor tendon ruptures
first described ruptures as occurring in the palmar region, but only in a small percentage (3 of 80 cases).2 In
the 1980s, more spontaneous flexor tendon ruptures
were being reported, of which a notably higher proportion of such ruptures were within zone III (80%) compared with the other locations (14% zone II and 6%
zone IV) (Table 3). In the largest series of spontaneous
flexor tendon ruptures within the hand, Naam1 found

that 92% of ruptures occurred within the palm (Table
4). In our series of spontaneous ruptures, all took place
in the palm.
Spontaneous tendon ruptures in the hand not only
occur within a predictable location but also involve a
predictable tendon. Within the literature, the FDP was
involved in 82% of cases, FDS in 12%, and flexor
pollicis longus (FPL) in 6% of cases; in 4 (9.3%) cases,
multiple tendon ruptures were found (Table 3). In addition, the majority of tendon ruptures found in the
literature occurred within the ulnar three digits: small
(62%), ring (14%), and long (10%) fingers. In this
series, all ruptures involved the profundus tendon: 4 to
the small finger and 1 to the long finger. This predilection may in part be explained by the mechanical stresses
each tendon generates with loading. On average, in all
fingers the FDP is 50% stronger than the FDS, and
within the small finger, the FDP is 3 times stronger than
the FDS.26 Baker et al27 found a deficient small finger
superficialis tendon in 34% of the 263 adult subjects
tested; others have reported an absence of the small
finger FDS tendon in approximately 5% of cases.28
Imbriglia and Goldstein5 confirmed these findings
when they tested the grip strength of 100 patients and
found that the FDS provides a variable contribution to
small finger power grip and that the FDP absorbs the
majority of the stress load. This may cause the profundus tendon to be susceptible to degenerative changes
over time. In all cases in this series, the FDS of the
small finger was present. In 2 cases in the literature11,20
and 1 in this series, however, the FDS tendon was found
to be small. In another case in the literature,6 an anomalous superficialis muscle was found in the palm on
surgical exploration.
The theory that normal musculotendinous units infrequently fail within their midsubstance has been accepted for many years. As early as the 1930s, McMaster29 believed that healthy musculotendinous units did
not rupture within their midsubstance without the presence of predisposing factors that decrease the threshold
for tendon rupture. Boyes et al2 found that tendon
ruptures occurred at the tendon insertion in 43 (63%) of
their cases. Of the remaining 25 (31%) cases that occurred within the tendon substance, only 3 (3.7%) were
classified as spontaneous ruptures. Folmar et al30 reviewed 12 flexor tendon ruptures in nonrheumatoid
hands and found that all were caused by pathologic
states, such as a nonunion fracture of the lunate, exostosis of the distal radius, and chronic synovitis. Other
causes of intratendinous flexor tendon rupture previously described in the literature include, and are not
limited to, osteoarthritis of the pisotriquetral joint,31

nonunion fracture of the hook of the hamate,32 lunate
dislocation,33 accessory carpal bone,34 gouty infiltration
of the flexor tendon,35 and tumor.36
In 1991, Kannus and Jozsa3 evaluated tendon specimens from 5 different sites after spontaneous ruptures
in 891 patients. Each specimen was examined by light
microscopy, electron microscopy, and histochemical
techniques and compared with previously healthy ageand gender-matched cadaveric tendons. Pathologic
changes were detected in all of the spontaneously ruptured tendons and in 35% of the control tendons. Most
of the pathologic changes were classified as degenerative, including hypoxic (necrotic) and mucoid degeneration, tendolipomatosis, and calcifying tendinopathy.
Common to each type of degenerative lesion was alteration in the normal network of collagen fibers thereby
reducing the tensile strength of the tendon. In the 5
cases of this study and in all 43 cases reviewed in the
literature, the ruptures occurred within macroscopically
normal tendons. Only 2 (40%) cases in this series and
27 (54%) from the literature were confirmed microscopically. Tendon ruptures that appeared normal on
gross examination only could have had undetected histologic abnormalities. In addition, specimens confirmed
by light microscopy alone may not reveal all pathologic
changes as detected by more advanced techniques, and
normal and pathologic criteria to describe human
tendons have not been previously defined.
Although the etiology of spontaneous ruptures in the
hand is not known, several theories have been presented. Several investigators studying the effect of overuse injuries in animal loading models37,38 have reported
disruption in collagen fiber organization, tendon fibrillation, and tendon thickening. These abnormalities may
alter the tissues ability to resist tensile load. Patients
with physically demanding occupations may be more
susceptible to intrinsic tendon failure over time. In this
series (Table 1) and in the pooled literature review
(Table 2), the dominant hand was injured more often
than the nondominant hand. In addition, all 5 cases in
this series and 90.7% of reported cases of spontaneous
tendon rupture in the literature occurred in men and the
rupture occurred most frequently with the fingers engaged in some type of longitudinal force (Table 3).
Berglund et al10 described a case of spontaneous rupture of the long and index finger profundus tendons
within the flexor sheath in a 15-year-old girl during a
grand mal seizure. In this case, there were no pathologic
changes observed, and this tendon rupture was thought
to be a direct result of forceful muscular contraction. In
contrast, 3 cases of tendon rupture in this series and 1
1069
case found in the literature occurred during the performance of the usual activities of daily living. Imai et al18
reported a case of a 74-year-old woman with 3 flexor
tendon ruptures involving the same hand over a 3-year
period. The histopathology of the involved tendons
revealed an abundant regenerative fibrosis, suggesting a
chronic wearing out of the tendon fibers. In other
cases, repetitive impact forces to the palm alone12,16 or
in conjunction with tendon loading11 have been implicated as a cause for flexor tendon rupture. It may be
possible that each type of tendon rupture, traumatic or
atraumatic, and acute or chronic, represent different
subtypes of spontaneous tendon rupture of unknown
pathogenesis.
As early as 1959, Anzel et al39 believed that microscopic damage to the blood supply of a tendon causes
attritional changes and eventual tendon rupture. Lundborg and colleagues40 discovered 2 areas of avascularity within the profundus tendon in zone II as a result of
watershed regions that exist between the intricate vascular supply in the tendon. Such regions may be more
pronounced as one ages rendering the tendon dependent
on local diffusion for providing nutrition.41 The average
age of the 6 patients who experienced spontaneous
ruptures of the profundus tendon in zone II in the
literature was 29.6 years (15 to 45 years)1,5,6,10; this
may question the strength of the relationship between
poor vascular perfusion of tendons and tendon rupture
in this region. The flexor tendons in the palm receive
their blood supply via a longitudinal system of vessels
in the paratenon.41 Zbrodowski et al42 studied the blood
supply of the lumbrical muscles in 100 upper extremities from human cadavers using vascular injection techniques and discovered that each lumbrical muscle receives its arterial supply from 4 sources. There were,
however, no anastomoses found between the networks
supplying the lumbrical muscles and the tendons of the
FDP muscle within the palm, suggesting a possible
watershed zone between the FDP tendon and lumbrical
muscle origin. Eighty percent of spontaneous ruptures
in the hand reported in the literature and all 5 cases in
this series occurred in the palm in the region of the
lumbrical muscle origin. This may have occurred due to
a critical zone of poor blood supply. Such vascular
alterations may lead to lack of oxygen and thus hypoxic
alterations of the tendon. More investigations are warranted assessing tendon vascularization as a cause for
tendon rupture in the hand.
Anatomic variations within the deep flexor tendons
and lumbrical muscles have been well documented.43,44
In 4 cases in the literature6,9,17 and in 2 cases in this
series, anomalies involving the musculotendinous units
1070
are likely responsible for spontaneous rupture of flexor

tendons in the hand. One of our patients was found to
have an abnormal profundus tendinous connection between the small and ring fingers of his dominant hand.
In the forearm, the single muscle belly of the FDP
separates into an ulnar and a radial bundle. At the
musculotendinous junction, the radial bundle forms the
profundus tendon of the index finger, and the ulnar
bundle forms the remaining 3 tendons. Distally, the
tendons of the ulnar bundle have great variability in
their interconnectedness at the level of the carpal tunnel
and beyond.44,45 This variability and possible lack of
interdependence between the third, fourth, and fifth
profundus tendons may predispose them to rupture.
There have been 3 cases in the literature of spontaneous
flexor tendon rupture at regions where there was an
abnormal tendinous fusion between the ring and small
finger profundus tendons.9,17 The authors concluded
that the tendon bifurcation becomes a point of weakness
during activities that cause tendon lengthening in opposing directions. In all 3 cases in the literature and in
our single case, the tendon rupture occurred at the site
of abnormal tendinous connection. In addition, the abnormal tendinous connection found in our one case
likely led to abnormally high compressive forces within
the carpal tunnel in the region where the FDP tendon of
the little and ring fingers change direction causing fascial wear over the hamate bone.
In the second case of tendon anomaly in this series,
the lumbrical muscle inserted into the deep transverse
metacarpal ligament between the long and ring fingers
instead of the radial extensor hood. The lumbrical muscles usually arise from the radial and volar surface of
the FDP tendon; the extent of origin ranging from 1 to
3.5 cm depending on lumbrical number.46 The lumbrical muscle fibers run with the flexor tendons toward the
radial side of the fingers. At the level of the A1 pulley,
the lumbricals diverge radially and rest on the deep
transverse metacarpal ligament with the digital nerves
and vessels in their respective compartments.46 At the
level of the base of the proximal phalanx, the lumbrical
forms a flat tendon that joins the radial side of the
extensor apparatus. Invariably, the lumbrical inserts
into the extensor tendon, volar plate, or even bone.47
The anomalous insertion site into the transverse metacarpal ligament, however, that was observed in this
series has not been previously described. Under normal
conditions, as digital flexion occurs, the lumbrical origin is pulled proximally. We speculate that when the
patient attempted to lift the grain bin, maximal tension
was placed on the profundus tendon at the lumbrical
origin as a result of this abnormal insertion. We believe
this insertion created a tethering effect on the profundus

tendon, predisposing it to rupture.
Our understanding of spontaneous flexor tendon ruptures of the hand is limited, and they appear to occur
more commonly than previously believed. Virtually all
closed ruptures are thought to be associated with an
underlying pathologic process or direct trauma such as
a severe crush injury. When there is no definite cause
and the injury takes place within the intratendinous
substance, these ruptures are termed spontaneous.
Spontaneous ruptures form a small subgroup in contrast
with avulsion ruptures. When spontaneous ruptures of
the hand occur, they typically involve the profundus
tendon of the small finger in the area of the lumbrical
origin and may be associated with the sensation of a
pop or snap. The compromised tendons most often
rupture after periods of peak strain but can also rupture
without any inciting force. Preoperative ultrasound or
MRI should be considered a valuable aid to the surgeon
in determining the location of tendon rupture and an aid
in preoperative planning. Although the etiology of
spontaneous ruptures is unclear, these injuries likely
depend on the interplay of several factors including
vascular alterations, repetitive microtrauma, local anatomic features, tendon anomalies, and genetic or other
endogenous influences.
The authors wish to thank Yuka Yamaguchi for providing the illustrated
figures for this article.
Received for publication January 15, 2007; accepted in revised form
June 13, 2007.
article.
Corresponding author: Aaron J. Bois, MD, Division of Orthopaedic
Surgery, The University of Saskatchewan, Royal University Hospital,
103 Hospital Drive, Saskatoon, SK, S7N OW8, Canada; e-mail:
ajb157@mail.usask.ca.
0363-5023/07/32A07-0020$32.00/0
doi:10.1016/j.jhsa.2007.06.012
References
1. Naam NH. Intratendinous rupture of the flexor digitorum
profundus tendon in zones II and III. J Hand Surg 1995;20A:
478 483.
2. Boyes JH, Wilson JN, Smith JW. Flexor-tendon ruptures in the
forearm and hand. J Bone Joint Surg 1960;42A:637 646.
3. Kannus P, Jozsa L. Histopathological changes preceding
spontaneous rupture of a tendon: a controlled study of 891
patients. J Bone Joint Surg 1991;73A:15071525.
4. Kumar S, James R. Closed rupture of flexor profundus
tendon in the palm. J Hand Surg 1985;10B:193194.
5. Imbriglia JE, Goldstein SA. Intratendinous ruptures of the
flexor digitorum profundus tendon of the small finger.
J Hand Surg 1987;12A:985991.
6. Lillmars SA, Bush DC. Flexor tendon rupture associated
with an anomalous muscle. J Hand Surg 1988;13A:115119.

7. Wray RC, Parlin LS. Spontaneous flexor tendon rupture in
the palm. Ann Plast Surg 1989;23:352353.
8. Walker LG, Lesavoy MA. Traumatic rupture of the profundus tendon proximal to the lumbrical origin. J Hand Surg
1990;15A:484 486.
9. de Roos WK, Zeeman RJ. A flexor tendon rupture in the
palm of the hand. J Hand Surg 1991;16A:663 665.
10. Berglund HT, Newport ML. Flexor digitorum profundus
ruptures secondary to grand mal seizure. J Hand Surg 1993;
18A:652 654.
11. Coombs CJ, Mutimer KL. Closed flexor tendon rupture in
the palm: an unusual but predictable clinical entity. Aust NZ
J Surg 1993;63:910 913.
12. Takami H, Takahashi S, Ando M. Rupture of the flexor
digitorum profundus tendon in the palm caused by repeated,
chronic direct trauma. J Hand Surg 1993;18A:65 67.
13. Nakamichi K, Tachibana S. Closed rupture of the flexor
digitorum profundus tendon within the carpal tunnel. J Hand
Surg 1994;19B:197198.
14. McLain RF, Steyers C, Blair W. Spontaneous rupture of
digital flexor tendons. Orthopedics 1994;17: 292296.
15. Yang SS, McCormack RR, Weiland AJ. Closed rupture of
the flexor digitorum profundus tendon in the palm of a
non-rheumatoid patient. Orthopedics 1998;21:205208.
16. Hang YS, Hang DW, Liu TK, Hou SM. Rupture of the flexor
digitorum profundus tendon of the small finger. Am J Sport
Med 2002;30:903905.
17. Davis C, Armstrong J. Spontaneous flexor tendon rupture in
the palm: the role of a variation of tendon anatomy. J Hand
Surg 2003;28A:149 152.
18. Imai S, Kubo M, Kikuchi K, Ueba H, Matsusue Y. Spontaneous rupture of the flexor digitorum profundus and superficialis of the index finger and the flexor pollicis longus
without labor-associated tendon loading. J Hand Surg 2004;
29A:587590.
19. Krishnamurthy A, Golla D, Lee WPA, Rubin JP. Traumatic
intratendinous flexor digitorum profundus rupture: a case
report. Can J Plast Surg 2005;13:151152.
20. Tan V, Mundanthanam G, Weiland AJ. Traumatic simultaneous rupture of both flexor tendons in a finger of an athlete.
Am J Orthop 2005;34:505507.
21. Leddy JP, Packer JW. Avulsion of the profundus tendon
insertion in athletes. J Hand Surg 1977;2A:66 69.
22. Wang PT, Bonavita JA, DeLone FX Jr, McClellan RM,
Witham RS. Ultrasonic assistance in the diagnosis of hand
flexor tendon injuries. Ann Plast Surg 1999;42:403 407.
23. Lee DH, Robbin ML, Galliott R, Graveman VA. Ultrasound
evaluation of flexor tendon lacerations. J Hand Surg 2000;
25A:236 241.
24. Drape JL, Tardif-Chastenet de Gery S, Silbermann-Hoffman
O, Chevrot A, Houvet P, Alnot JY, et al. Closed ruptures of
the flexor digitorum tendons: MRI evaluation. Skeletal Radiol 1998;27:617 624.
25. Kumar BA, Tolat AR, Threepuraneni G, Jones B. The role of
magnetic resonance imaging in late presentation of isolated
injuries of the flexor digitorum profundus tendon in the
finger. J Hand Surg 2000;25B:9597.
26. Brand PW. Clinical mechanics of the hand. St. Louis:
Mosby, 1985:257272.
27. Baker DS, Gaul JS, Williams VK, Graves M. The little
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
1071
finger superficialis: clinical investigation of its anatomic and

functional shortcomings. J Hand Surg 1981;6A:374 378.
Furnas DW. Muscle-tendon variations in the flexor compartment of the wrist. Plast Recon Surg 1965;36:320 324.
McMaster PE. Tendon and muscle ruptures: clinical and
experimental studies on the causes and location of subcutaneous ruptures. J Bone Joint Surg 1933;15:705722.
Folmar RC, Nelson CL, Phalen GS. Ruptures of the flexor
tendons in hands of non-rheumatoid patients. J Bone Joint
Surg 1972;54A:579 584.
Grant I, Berger AC, Ireland DCR. Rupture of the flexor
digitorum profundus tendon to the small finger within the
carpal tunnel. Hand Surg 2005;10:109 114.
Hartford JM, Murphy JM. Flexor digitorum profundus rupture of the small finger secondary to nonunion of the hook of
the hamate: a case report. J Hand Surg 1996;21A:621 623.
Johnston GHF, Bowen CVA. Attritional flexor tendon ruptures
by an old lunate dislocation. J Hand Surg 1988;13A:701703.
Koizumi M, Kanda T, Satoh S, Yoshizu T, Maki Y, Tsubokawa N. Attritional rupture of the flexor digitorum profundus tendon to the index finger caused by accessory carpal
bone in the carpal tunnel: a case report. J Hand Surg 2005;
30A:142146.
Wurapa RK, Zelouf DS. Flexor tendon rupture caused by
gout: a case report. J Hand Surg 2002;27A:591593.
Masada K, Kanazawa M, Fuji T. Flexor tendon ruptures
caused by an intraosseous ganglion of the hook of the
hamate. J Hand Surg 1997;22B:383385.
Soslowsky LJ, Thomopoulos S, Tun S, Flanagan CL, Keefer
CC, Mastaw J, et al. Overuse activity injures the supraspinatus tendon in an animal model: a histologic and biomechanical study. J Shoulder Elbow Surg 2000;9:79 84.
Nakama LH, King KB, Abrahamsson S, Rempel DM. Evidence of tendon microtears due to cyclical loading in an in
vivo tendinopathy model. J Orthop Res 2005;23:1199 1205.
Anzel SH, Convey KW, Weiner AD, Lipcomb PR. Disruptions of muscles and tendons: an analysis of 1104 cases.
Surgery 1959;45:406 414.
Lundborg G, Myrhage R, Rydevik B. The vascularization of
human flexor tendons within the digital synovial sheath
region: structural and functional aspects. J Hand Surg 1977;
2A:417 422.
Manske PR, Lesker PA. Flexor tendon nutrition. Hand Clin
1985;1:1324.
Zbrodowski A, Mariethoz E, Bednarkiewicz M, Gajisin S.
The blood supply of the lumbrical muscles. J Hand Surg
1998;23B:384 388.
Perkins RE, Hast MH. Common variations in muscles and
tendons of the human hand. Clin Anat 1993;6:226 231.
Leijnse JNAL. A generic morphological model of the anatomic variability in the m. flexor digitorum profundus, m.
flexor pollicis longus and mm. lumbricales complex. Acta
Anat 1997;160:6274.
Goodman HJ, Choueka J. Biomechanics of the flexor tendons. Hand Clin 2005;21:129 149.
Kaplan EB, Hunter JM. Extrinsic muscles of the fingers. In:
Spinner M, ed. Kaplans functional and surgical anatomy of the
hand. 3rd ed. Philadelphia: JB Lippincott, 1984:109 112.
Eladoumikdachi F, Valkov PL, Thomas J, Netscher DT.
Anatomy of the intrinsic hand muscles revisited: part II
lumbricals. Plast Reconstr Surg 2002;110:12251231.
REVIEW ARTICLE
Immunologic Approaches to Composite
Tissue Allograft
Aurle Taieb, MD, Julio A. Clavijo-Alvarez, MD, PhD,
Giselle G. Hamad, MD, W. P. Andrew Lee, MD
From the School of Medicine, University of Pittsburgh, Pittsburgh, PA.
This article discusses the immunologic principles and the most promising immunologic
approaches for composite tissue allograft tolerance. We have previously reviewed some of
the pharmacologic approaches for composite tissue allo-transplantation. In this review, we
will summarize the range of options that may address the challenge of transplantation in
reconstructive surgery. (J Hand Surg 2007;32A:10721085. Copyright 2007 by the American Society for Surgery of the Hand.)
Key words: Allograft, composite, immunology, review, tissue.
wo formidable obstacles to composite tissue

allograft (CTA) transplantation are chronic rejection and the toxicity of immunosuppression, including opportunistic infections and malignancies. Therefore, the induction of tolerance to
donor limb tissue allografts would increase the feasibility of CTA in the clinical setting. Tolerance is
defined as a state of unresponsiveness to donor antigens while maintaining immune reactivity to other
foreign antigens. The following 3 strategies show
great promise in achieving tolerance in CTA.
Tolerance Through Major

Histocompatibility Complex Matching
Introduction
The major histocompatibility complex (MHC) is a
group of genes encoding the MHC molecules (or
proteins). These genes can be divided into 3 major
classes: class I, class II, and class III. The MHC
classes I and II code for antigens expressed on cells
and tissues, whereas MHC class III codes for antigens present on proteins in serum and other body
fluids that are critical to immune function, including
tumor necrosis factor (TNF), factor B, complement
component (C2), and heat shock proteins. Antigens
of MHC class III have no role in graft rejection,
whereas the molecules produced by MHC classes I
and II represent the main cell surface antigens of
organisms. Hence, they are the principal targets of
rejection in transplantation and therefore play an
important role in the acceptance or rejection of
1072
grafts. Depending on the species, the MHC is given

different names: for instance, in humans, the complex is known as the human leukocyte antigens
(HLAs); in swine, it is known as the swine leukocyte
antigens (SLAs).1
Barth et al observed that inbred strains of mice
rejected skin and tumors from identical donors with
the same MHC and therefore proposed that other
polymorphic proteins may play a role in rejection.2
These nonMHC antigens were called minor histocompatibility complex (mHC) because the reaction
of rejection that they produced was much less dramatic or vigorous than in cases of MHC mismatch3
as observed in skin grafts between HLA-identical
siblings.4
Experimental MHC Matching and
Transplantation Tolerance
In contrast with humans, rodents do not express
MHC class II antigens on their vascular endothelium.
Therefore, the swine model is preferred when MHC
matching studies are performed.5
Bourget et al demonstrated that tolerance to solid
organ could be achieved by MHC matching and a
short-term, 12-day course of cyclosporine. In their
study, the investigators failed to demonstrate that
chimerism was present in the peripheral blood of
these pigs, suggesting that the presence of donor cells
in the host was not the mechanism of tolerance
induction in this model.6
Then, Lee et al7 emphasized the importance of MHC
Taieb et al / Immunologic Approaches to Composite Tissue Allograft
1073
Table 1. Skin and Skeletal Muscle Graft Survival With MHC Mismatch and Minor Antigens Mismatch8
Group
12 Days CyA
Skin Island Survival (Days)
Skeletal Muscle Graft Survival (Days)
I-MM
II-MA
III-MA
IV-MA
4
4
7
4
Yes
No
Yes
Yes
N/A
N/A
N/A
39, 41, 50, and 120
Rejected 42 days
Rejected 6384 days
All grafts viable at harvest 178330 days
All grafts viable at harvest 98159 days
MM, MHC mismatch; MA, minor antigens mismatch.
matching for CTA. In one experiment, they applied

musculoskeletal allografts between MHC-matched, minor antigen-mismatched miniature swine with immunosuppression via cyclosporine for 12 days and showed
no histologic evidence of rejection in light of rejection
by group 1 at death between 178 and 372 days after
transplantation (group III; Table 1). Group I demonstrated the role of the MHC matching by rejecting the
allografts in less than 6 weeks while receiving cyclosporine, and group II illustrated the importance of the
short course of cyclosporine after the transplant. Therefore, Lee et al concluded that as with renal allografts,
tolerance could be the consequence of an alteration in
T-cell activity during the rejection cascade through the
inhibition of interleukin-2 (IL-2) by cyclosporine. In
addition, the recipient animals accepted skin grafts from
the original donor animals but rejected third-party control skin grafts, indicating that specific tolerance to the
musculoskeletal tissues had been induced.
More recently, to address the most antigenic component in CTA, known as skin, they added a vascularized skin component to the musculoskeletal allografts (group IV). In this group, the allografts were
MHC-matched and minor antigen-mismatched; the
swine were treated with a short-term, 12-day course
of cyclosporine. The results of the musculoskeletal
allografts were similar to the previous study, and no
rejection was observed until the end of the study.
Nevertheless, only 1 of 4 swine did not express any
signs of rejection; the 3 others rejected the epidermis,
whereas the dermis remained intact at 39, 41, and 50
days. Nevertheless, the donor graft dermis was reepithelialized by the host skin, indicating that although
it was much more difficult to induce tolerance to the
skin component using genetic matching and shortterm cyclosporine immunosuppression, it was clinically feasible. A topical treatment, however, may be
necessary for clinical application of this technique.9
the need for long-term immunosuppression. This

concept is supported by the lack of rejection in Chinese patients having limb transplant. In both cases,
there were only 3 HLA mismatches compared with
the first French and Louisville patients, both of
whom had a maximum of 6 HLA mismatches between the donor and recipient.10
Hence, the next challenge is where to find a donor
with MHC matching with the recipient. Two possibilities are family donors or the National Bone Marrow Registry.
MHC-Matched Transplant and Clinical

Application
According to these experimental data, MHC matching may induce tolerance in a clinical setting without
Conclusion
The induction of tolerance between a donor and an
MHC-matched recipient without long-term immunosuppression is experimentally feasible but is not yet
The Family. MHC genes are inherited as a group,

or haplotype, 1 from each parent. Thus, a heterozygous human inherits 1 paternal and 1 maternal haplotype. Consequently, in any family, 2 siblings from
the same parents have a 25% probability of sharing
both paternal and maternal haplotypes, a 50% probability of sharing at least 1 haplotype, and a 25%
probability of sharing no haplotype. Hence, a transplant performed between siblings sharing the same
paternal and maternal haplotypes is equivalent to the
experiment with MHC-matched antigens and mismatched minor antigens.9 In addition, in a transplant
between identical twins, no mismatched minor antigens will be present.
The National Bone Marrow Registry. Since 1995,
the National Marrow Donor Program has been recruiting volunteers from major cities in the United
States and therefore has the biggest donor registry in
the world, with more than 1.35 million HLA-typed as
of 2001.
Therefore, with a large-enough HLA-typed pool
of normal individuals and a mechanism for identifying a potential donor, this registry could allow
for MHC-matching transplantation between dissimilar individuals.11
1074
Figure 1. Schematic representation of the different types of chimerism.
clinically applicable because of the difficulty of finding a matching donor. In the future, gene therapy
may play a substantial role in tolerance induction.
Tolerance and Chimerism

Generality
Chimerism is the oldest strategy for tolerance induction. The term chimerism has its origin in Greek
mythology; the chimera was a creature bearing a
lions head, a goats body, and a serpents tail. We
can define chimerism as the coexistence of 2 genetically distinct cell populations in the same organism.12
Types of Chimerism
Microchimerism. Microchimerism refers to a state
in which donor cells are present at low levels in the
recipient organism (1 cell per 1 104 cells or less).13
This state results from the migration of passenger
leukocytes (characterized as immature dendritic
cells14) after transplantation of solid organs and cellular allografts into the recipients lymphoid organs.1517 Therefore, microchimerism does not require recipient preconditioning as no space is
needed for the engraftment. The cells involved in
microchimerism include passenger leukocytes but
not donor hematopoietic pluripotent stem cells.
Hence, there is no production of different lineages of
donor stem cells, and the level of donor cells found in

the recipients blood is very low.16 For this reason,
microchimerism can only be detected using highly
sensitive techniques such as polymerase chain
reaction.13,18,19
Macrochimerism. Macrochimerism arises from
the transplantation of donor hematopoietic stem
cells, usually from bone marrow, into a preconditioned recipient. As a result, the original cells are
depleted to create space for the donor bone marrow
(BM). These transplanted hematopoietic stem cells
engraft into the recipient BM and induce the production of all their lineages, resulting in a high level of
donor-specific cells in the recipient, hence the term
macrochimerism.12,20 To assess the level of donorspecific cells in the recipient, flow cytometry is usually used, because this technique allows detection of
cells at levels from 2% to 100%.13
Theoretically, BM transplantation can induce three
types of macrochimerism (Fig. 1). The first one is a
syngeneic chimerism. In this case, the BM transplanted is genetically identical to the recipient. This
occurs when BM is transplanted between identical
twins. The second type, the allogeneic chimerism,
arises when BM is transplanted from a genetically
distinct donor. Finally, there is xenogeneic chimer-
ism, which occurs when BM is transplanted between

different species.12
Moreover, two subtypes of allogeneic chimerism
are described in the literature, and each of them
requires a different preconditioning. Full donor chimerism is achieved when the recipient is completely
myeloablated before BM transplantation to destroy
the recipient hematopoietic system (eg, by total body
irradiation). The entire recipient hematopoietic system is replaced by donor cells, thereby inducing a
complete or near-complete donor hematopoietic reconstitution.21 Another subtype, termed mixed chimerism, describes a state wherein hematopoietic populations of both the recipient and the donor coexist in
the recipient.22,23 This state can be achieved by
milder forms of preconditioning (also known as nonmyeloablative form), which do not ablate totally the
host hematopoietic system and therefore are less
toxic (Fig. 1).12
Tolerance Induction Through Chimerism
Microchimerism and Tolerance. The role of passenger leukocytes in transplantation is complex.
These cells, characterized as immature dendritic
cells, can contribute to either rejection or tolerance of
the graft, depending upon the inflammation and cytokines production in the microenvironment present
within both the graft and the draining lymphoid tissue of the host. In the case of rejection, these immature dendritic cells will initiate an acute graft rejection by migrating through the host lymphoid tissue
where they will directly present their donor alloantigens to T cells and induce a process called direct
allorecognition.14,24,25 This can explain the results of
several studies where the depletion of these passenger leukocytes prevent graft rejection and induce
graft survival.14,26 28
It is also thought that these passenger leukocytes
have immunomodulatory properties on the
graft.14,29 32 Indeed, studies with liver grafts
known to contain large numbers of passenger leukocytes have not resulted in rejection.33,34 When
the livers were depleted of passenger leukocytes,
all of the animals demonstrated graft rejection.35,36
Furthermore, their presence during the first 24
hours after transplantation seems to indicate their
importance in immune modulation.29,31 Hence,
Wood et al31 have suggested that increasing the
number of donor-derived leukocytes in the early
stages of the immune response may be valuable.
Various studies have suggested that these cells
may play different roles at different stages in the
1075
Figure 2. Role of microchimerism in tolerance induction.
response, depending upon their concentration and

the microenvironment.14,37
The role of passenger leukocytes in immune modulation still remains obscure. They may mediate
tolerogenic properties in the absence of costimulatory molecules, which is characteristic of immature
cells. Alternatively, they may act as surrogate targets
of rejection and therefore protect the graft. Another
possibility is that they serve as an indicator of effective immunosuppression.13
It is still debated whether microchimerism is
responsible for tolerance or is a consequence of
tolerance. Indeed, acute rejection of allografts in
presence of microchimerism and long-term allograft survival in absence of microchimerism has
been reported.38 42 Furthermore, no correlation
has been clearly established between the degree of
donor microchimerism in the blood and the rate of
allograft survival.41 44 In conclusion, microchimerism can be considered as an epiphenomenon
rather than the cause of allograft survival (Fig.
2).13,41,42,44 46
Macrochimerism and Tolerance. The concept
that macrochimerism can induce tolerance arose in
the middle of the 20th century. In 1945, Owen described how fraternal bovine twins had a mixture of
two distinct types of erythrocytes despite sharing
the same placental circulation.47 This mixture remained present long after birth, therefore representing the first naturally occurring state of mixed chimerism.47 A few years later, Medawar demonstrated
that these chimeric twins tolerate skin grafts between
each other.48,49 These observations provided the basis for using hematopoietic cells to induce transplantation tolerance.50
1076
Advantages of Mixed Chimerism Over Full

Chimerism. There are several advantages of using mixed chimerism over full chimerism in tolerance induction:
Mixed chimerism can be induced with nonmyeloablative regimens. These conditioning regimens are
less toxic for the host than those necessary to
achieve full chimeras.51
A mixed chimera is more immunocompetent compared with a full chimera where a total ablation is
used.52
In the mixed chimera, the presence of donor and
recipient hematopoietic cells in the thymus induces
a peripheral tolerance to both host and recipient
tissues.51
There is a higher resistance to graft-versus-host
disease for the mixed chimerism.53
Therefore, in purpose to induce transplant tolerance, mixed chimerism is preferred over full
chimerism.50
Mechanism for Tolerance Induction Through

Mixed Chimerism. The principle of tolerance
induction requires 2 stages. The primary event is the
elimination or inactivation of preexisting mature donor-reactive T cells to prevent rejection of the donor
BM. The secondary event involves a lifelong inactivation or elimination of newly developing donorreactive T cells to maintain tolerance.19 An additional requirement is the creation of sufficient
space in the recipient to allow the engraftment of
the donor BM cells.12
Some preconditioning protocols for the induction
of BM engraftment involve total body irradiation
with or without thymic irradiation. Alternatives include use of cytotoxic drugs such as tacrolimus or
cyclophosphamide12,19 and exhaustive T-cell depletion with cytotoxic antibodies such as anti-CD4, antiCD8, and anti-natural killer (NK).23,54 More recently, the utility of reagents blocking T-cell
costimulatory pathways have been described.5557
Once the recipient is conditioned, the donor BM
containing the hematopoietic stem cells is infused
intravenously. As long as these donor and recipient
stem cells coexist, they will generate cells from all
hematopoietic lineages. Some of these progenitor
cells will migrate to the thymus and become thymic
dendritic cells.56 Once the thymus is engrafted with
these cells, a process of self-selection is established.
During T-cell maturation, self-selection creates a
state of central tolerance by deletion of T cells that
react specifically with either the host or the donor
Figure 3. Induction of tolerance through mixed chimerism.
antigens. Hence, at the end of their maturation, the

remaining T cells are not reactive to the new chimeric organism antigens and therefore are tolerant to
both sets of antigens (Fig. 3).25,58 60
Mixed Chimerism and CTA Tolerance Induction. In 1998, Foster et al showed that mixed chimeric rodents induced by BM transplantation accepted limb allografts across a full MHC barrier
without the need of long-term immunosuppression.61
In the experimental groups, the animals with a level
of donor chimerism greater than 60% were rejectionfree for more than 100 days. In contrast, animals with
a chimerism level less than 20% developed clinical
and histologic signs of moderate rejection; however,
their survival was prolonged when compared with
control groups.61 Therefore, this study shows a correlation between the degree of chimerism and the
presence of tolerance.10,12
More recently, Mathes et al demonstrated that in
the long-term tolerant recipient of limb allografts,
there is no evidence of donor BM cells in the hematopoietic tissues of the graft.62 It appears that recipients BM cells repopulate the donor marrow space
of the graft. This repopulation suggests the existence
of a favorable microenvironment within the limb
allograft. It is important to mention that they did not
find evidence of donor cell engraftment in the recipients lymphohematopoietic tissues.62
Advantages and Limitations of BM Transplantation. Several characteristics make macrochimerism attractive for clinical use:
Macrochimerism can be easily measured and may

indicate successful tolerance induction clinically.63
Thymic clonal deletion is known to be a robust and

reliable mechanism for tolerance induction.63
Tolerance has been induced successfully through a
large number of animal models, including rodents,
large animals, and nonhuman primate.22,64 66
A number of cancer patients who had standard BM
transplantation in the course of their hematologic
malignancy became tolerant and were able to accept a kidney from the same BM donor without
long-term immunosuppression.6771
Despite these arguments in favor of macrochimerism, a fully ablative conditioning of the recipient remains an obstacle. The risks of myelosuppression may
not compensate for the benefit of CTA transplantation
for nonlife-threatening conditions such as hand deficits.71 Additional specific complications related to BM
transplantation include the toxicity of ablative conditioning, failure of engraftment, graft-versus-host disease (GVHD), and a state of relative immunoincompetence if full ablation is used.12 More recently, Kean et al
have shown a strategy for producing high-level hematopoietic chimerism after nonmyeloablative conditioning in the rhesus macaque.72 This strategy relies on
hematopoietic stem cell transplantation after induction
with a nonmyeloablative dose of busulfan and blockade
of the IL-2 receptor in the setting of mammalian target
of Rapamycin (mTOR) inhibition with sirolimus and
combined CD28/CD154 costimulation blockade. Their
mean peripheral blood peak donor chimerism was 81%
with a median chimerism duration of 145 days; however, the reappearance of vigorous T-mediated alloreactivity accompanied rejection of the transplants. This
technique may be a platform upon which to evaluate
emerging tolerance-induction strategies.72
Vascularized Bone Marrow Transplant. Some

CTAs, such as limb, include BM. Consequently,
once the CTA is transplanted, it may be considered
vascularized bone marrow transplant (VBMT). This
unique situation has several advantages and disadvantages as described in the following.
Vascularized Bone Marrow Transplant: Advantages. First, the CTA brings BM within its
microenvironment; hence there is no need to create
BM space by host conditioning with irradiation or
radiomimetic drugs to allow stable engraftment.62,73
Second, this VBMT graft could be a continuous
source of donor cells, including specialized cells,
such as BM-derived dendritic cells, which have been
shown to modulate the host immune response in
certain animal models.74
1077
Third, the stromal microenvironment appears to be

essential for the proliferation and the differentiation
of the hematopoietic progenitors. Therefore, by providing the stromal hematopoietic microenvironment
as well as pluripotent progenitor cell source, the
VBMT seems to be more efficient and essential for
the post-transplant lymphopoiesis when compared
with cellular BM transplant.20,7577
Finally, VBMT has been proved to prolong the
survival of skin graft from the same donor.78
Whether the donor BM in a human hand transplant
might induce mixed chimerism in the recipient is
unknown. Nevertheless, this has not been seen in any
of the 24 reported hand-transplant patients to date.
Vascularized Bone Marrow Transplant: Disadvantages. A possible disadvantage of transplanting a limb allograft with functional immune
effector cells is the potential for these mature allogeneic T cells to attack the recipients tissue, resulting in the serious clinical reaction GVHD.79
In 1990, Hewitt et al used rat hind limb transplantation to evaluate the involvement of VBMT in inducing GVHD. They observed that approximately
35% to 40% of the recipients developed GVHD
when parental Lewis rat hind limbs were transplanted
to Lewis Brown Norway F1 hybrids. After additional investigation, it was shown that the recipients
became mixed chimeras after hind limb transplantation and that most of the animals expressing a T-cell
chimerism with a level of 11% demonstrated tolerance, whereas others with T-cell chimerism levels of
52% showed evidence of GVHD.80 Hence, this
group has shown that the level of mixed lymphocyte
chimerism plays an important role in the induction of
GVHD. Indeed, a low-level of mixed lymphocyte
chimerism is associated with the induction of tolerance after limb transplantation, but unstable highlevel lymphocyte chimerism is associated with the
development of GVHD, indicating that a very fine
balance exists between tolerance and GVHD.79
Then in 2003, the same group developed an
isolated VBMT (iVBMT) model to study the contribution of the BM component in a CTA. The
investigators demonstrated that it was not the vascularized BM within a CTA that caused the
GVHD. Although the reason for the lack of GVHD
in this model is still undetermined, the authors
hypothesize that the stromal microenvironment
may be responsible because of its multiple sources
of signaling mechanisms.81
1078
Conclusion
Although GVHD is an obstacle in CTA, GVHD has
not been observed in either experimental or the recent clinical hand transplants.82 84 This observation
could be explained by the fact that a very small
amount of active BM is expected to be transferred
with this type of CTA.79,80,85
Furthermore, since the first successful transplantation of BM in humans in 1968,86 much progress has
been made in clinical use as well as in experimental
models to treat nonlife-threatening conditions
through hematopoietic reconstitution. The extension
of lower-risk protocols, such as the nontotal body
irradiation approach, to establish the state of mixed
chimerism across MHC barriers seems to be an important point to reduce the toxicity occurring from
the conditioning of the host.65
Monoclonal Antibodies and Induction of

Tolerance
The immunosuppressive drugs used to prevent graft
rejection target the immune system in a nonspecific
way. This approach increases the risk of infection as
well as malignancies. When we look closely at this
phenomenon, only a small percentage of lymphocytes are responsible for graft rejection. Therefore, it
would be preferable to control the responsible lymphocytes specifically rather than weakening the entire immune system. Novel therapies aim to target
specific key lymphocyte molecules by using a variety
of monoclonal antibodies (mAbs). These mAbs interfere with the interaction between the T cells and
the antigen-presenting cells (APCs).87 We will review two mAbs that may be of relevance:
Blockage of Costimulatory Molecules (Blockage
of the Second Signal)
The activation of lymphocytes requires at least two
signals. The first signal is elicited by the interaction
between a foreign antigen and the T-cell receptor
(TCR) on the surface of the T cell. The antigens are
presented by APCs via MHC molecule. The second
signal is produced by costimulatory molecules expressed on the surface of activated APCs (Fig. 4).87
B7-1 (CD80), B7-2 (CD86), and CD40 are some of
the costimulatory molecules on APCs that have been
described. They interact with their counterparts
(termed ligands) located on the T cells (CD28 for B7,
and CD154 or CD40L for CD40). Therefore, T cells
that encounter antigen without the appropriate second signal become nonreactive or tolerant to this
antigen.88 91 In this strategy, monoclonal antibodies
Figure 4. T-cell activation requires two signals.
are used to block the costimulatory molecules located

on the APCs.
The B7/CD28 pathway provides a second signal
that activates nave T cells, induces their proliferation, and increases their production of IL-2 (a proinflammatory molecule).92 On the other hand, the
CD40/CD154 pathway upregulates B7 molecules,
induces the production of IL-12, promotes the differentiation of CD8 T cells into cytotoxic T lymphocytes, and costimulates B cells, macrophages, and
endothelial cells.9397 Table 2 summarizes the costimulatory molecules and their actions.
These diverse effects on the immune system explain the role of this pathway in acute allograft rejection as demonstrated in a number of rodent experiments.98,99
Two reagents have been used in experimental transplantation for blocking the secondary signal. Cytotoxic
T lymphocyte-associated antigen-4 (CTLA4)-Ig interferes with the B7-1, -2/CD28 pathway, and anti-CD154
mAbs inhibit the CD40/CD154 pathway (Fig. 5).
Various groups have evaluated different protocols for
costimulatory blockage.100,101 It has been demonstrated
that by blocking either the CD40/CD154 pathway or
the B7/CD28 pathway, a prolongation of allograft acceptance is achieved; however, tolerance is not observed.102 Larsen et al showed that simultaneous but
not independent blockade of the B7/CD28 and CD40/
CD154 pathways effectively aborts T-cell clonal expansion in vitro and in vivo, promotes long-term survival of
fully allogeneic skin grafts, and inhibits the development of chronic vascular rejection of primarily vascularized cardiac allografts longer than 50 and 70 days,
respectively.103 Kirk also demonstrated an extended
kidney allograft survival in rhesus monkeys for longer
1079
Table 2. Costimulatory Molecules and Their Actions8

Molecule
CD40
Ligand
Action
Monoclonal Antibody
CD40 ligand or CD154
Upregulates B7-1 and B7-2

Induces IL-12 production
Enables dendritic cells to
stimulate CD8 cells to
differentiate into CTL
Acts on B cells to allow
immunoglobulin isotype
switching from IgM to IgG
Upregulates cell adhesion
molecules on endothelial cells
Increases the proinflammatory
activities of monocytes
Induces nave T-cell activation
Leads to increased IL-2
production and T-cell
proliferation
Induces nave T-cell activation
Allows for sustained TCR signaling
Leads to increased IL-2
production and T-cell
proliferation
Anti-CD40 ligand or anti-CD154 mAb
B7-1 (CD80)
CD28
B7-2 (CD86)
CD28
CTLA4-Ig
CTLA4-Ig
CTL, cytotoxic T-lymphocyte antigen
than 180 days without the need of additional conventional immunosuppresion.104

Although recent experimental studies have demonstrated CTLA4-Ig to be a promising immunosup-
pressive drug in vascularized solid organ allografts,

little work has been done in CTAs. Iwasaki et al
recently described the potential of this monoclonal
antibody therapy for limb allograft transplanta-
Figure 5. Costimulatory blockade in the APC.
1080
Figure 6. T-cell receptor structure.
tion.105 By injecting CTLA4-Ig (0.5 mg/animal),

they significantly prolonged limb survival when
compared with controls (P 0.01). Because the
treatment only lasted for a few days, they were only
able to suggest the potential benefits of this treatment.105 Moreover, it is important to consider the
prothrombotic properties of anti-CD154 mAb because they will limit its use in a field where graft
survival depends on the microvascular anastomosis.106,107 Clearly, further studies should be done to
clarify the benefits of mAbs in the tolerance of CTA.
Anti-TCR mAbs
As seen earlier, TCRs are receptors on the surface of
T lymphocytes that recognize peptides from foreign
antigens. These antigens are displayed by MHC molecules of APCs. This liaison induces the first signal
necessary for the activation of T lymphocytes. In this
approach, mAbs are used against the subregion of
the TCR to deplete the potentially alloreactive T
cells, leading to an unresponsive environment and
peripheral anergy to the graft (Fig. 6).75,108 Furthermore, it was also shown that these mAbs decreased
the expression of the TCR on the surface of the T
cell.109,110
Initial studies using anti-TCR mAbs on an experimental cardiac allograft model demonstrated that they
abrogated accelerated rejection (in a hamster-to-rat
model) and prolonged cardiac allograft survival in a
dose-dependent fashion.111 Recently, several studies
used anti-TCR mAbs in CTA.75 When the transplant
recipients were treated with 21-, 7-, and 5-day protocols
of anti-TCR mAbs combined with cyclosporin A (antiTCR/CyA), all the transplants survived more than 350
days. After cessation of immunosuppression, the clinical tolerance and immunocompetence were confirmed
by skin grafting in vivo and mixed lymphocyte reaction

(MLR) in vitro.
The mechanism for this unresponsiveness is not
well defined. The fact that the 5-day protocol was as
successful as the 35-day protocol can be explained by
a prolonged deletional effect. Indeed, treatment with
anti-TCR/CyA resulted in depletion of more than
95% of TCR-positive T cells. This occurs as early as
day 7 and for 35 days after cessation of treatment.75
These results may be explained by the fact that the
maturation of new T cells through the thymus takes
approximately 28 days.112 Therefore, the deletion of
potentially alloreactive TCR-positive cells creates an
immunologic silence period in which newly developed T cells (both from donors and recipients) are
educated in the thymus, thereby allowing durable
engraftment of the donor-derived stem cells inducing
chimerism. Furthermore, the results show that a short
course of anti-TCR/CyA therapy (5 days) leads to the
development of donor-specific chimerism only when
immunomodulating therapy is introduced shortly before antigenic exposure.75
More recently, Klimczak et al demonstrated the
benefits of bilateral VBMT-induced donor-specific
chimerism. Bilateral femoral bones were bilaterally
anastomosed to the abdominal aorta and inferior vena
cava of the recipient across the MHC barrier
[BN(RT1) donors and Lewis(RTI) recipients]. Using
the immunomodulatory protocol of -TCR mAb/
CyA for 7 days, they were able to confirm the viability of BM cells for up to 35 days.113
The Programmed Death-1/PD-L1/PD-L2
Pathway
Programmed death-1 (PD-1) is related to CD28 and
CTLA4 but lacks the membrane proximal cysteine
required for homodimerization. Programmed death-1
is induced on peripheral CD4 and CD8 cells, NK
cells, B cells, and monocytes.91
Programmed death-1 has two ligands, PD-L1 and
PD-L2. Programmed death-L1 is expressed on resting APCs and T, B, and endothelial cells and is
upregulated on activation of these cells. Programmed
death-1 ligation transmits a potent inhibitory signal
in the early stages of T-cell activation, resulting in a
decrease in cytokine production and cell-cycle arrest
in the G0/G1 phase (also in B cells).91
In cardiac allograft studies, PD-L1-Ig agonist fusion proteins plus CyA significantly enhanced allograft survival over CyA or PD-L1-Ig alone (P
0.01). Similarly, PD-L1-Ig markedly reduces cardiac
transplant arteriosclerosis.114 Its protective effects
are unclear. Some theories proposed that PD-L1-Ig

may trigger a negative signal through PD-1 or block
a positive signal for T-cell activation.91
Tumor Necrosis Factor/TNF-R Family
Tumor necrosis factor-receptor (TNF-R) family
members are expressed by T cells and their TNFfamily ligands are expressed by APCs. Tumor necrosis factor-R/TNF interactions are critical in the clonal
expansion/effector phases of immune responses and
involve deatritic cell (DC)/T cell and B/T cell relations.
Tumor necrosis factor-receptor may be either constitutively expressed by nave T cells (CD27) or
induced after antigen recognition (OX40, 4-1BB,
CD30, and CD27). Their corresponding ligands (L)
are induced simultaneously with their receptors on T
cells, one to several days after activation. This delayed expression may help to sustain the ongoing
response by a broad range of professional APCs.91
Curry et al used an OX40-Ig fusion protein to
block the OX-40/OX-40L pathway in mouse cardiac
allograft rejection across MHC and mHC barriers.
Heart survival for fully MHC-mismatched allografts
was unaffected by OX40 blockade alone, but
OX40-Ig treatment in the mHC-mismatched model
resulted in long-term graft survival.115
Demirci et al blocked the OX40/OX40L pathway
(anti-OX40 ligand) markedly prolonging skin graft
survival in CD28/CD154 knockout mice.116
Finally, Cho et al have shown that 4-1BB deficient mice displayed delayed heart allograft rejection
compared with control mice. Moreover, treating
wild-type mice with a blocking anti-4-1BBL mAb
(TKS-1) resulted in substantial prolongation of heart
allograft survival (median survival time 42 days vs 8
days for control), with 40% of the recipients displaying long-term (60 days) survival.117
Achievement of tolerance represents the primary
objective for the success of transplantation of CTA or
solid organs. Immunologic manipulations including
MHC matching, induction of chimerism, and mAbs
may be useful tools for CTA tolerance. Combined
with pharmacologic immunosuppression, immunomodulating therapy may address the challenge of
transplantation in reconstructive surgery.
June 13, 2007.
No benefits in any form have been received or will be received from a
commercial party related directly or indirectly to the subject of this article.
Corresponding author: W. P. Andrew Lee, MD, School of Medicine,
University of Pittsburgh, 690 Scaife Hall, 3550 Terrace Street, Pittsburgh, PA 15261; e-mail: leewpa@upmc.edu.
1081

0363-5023/07/32A07-0021$32.00/0
doi:10.1016/j.jhsa.2007.06.013
References
1. Goldsby RA, Kindt TJ, Osborne BA, Kuby J. Immunology.
5th ed. W. H. Freeman, 2002.
2. Barth R, Counce S, Smith P, Snell GD. Strong and weak
histocompatibility gene differences in mice and their role in
the rejection of homografts of tumors and skin. Ann Surg
1956;144:198 204.
3. Laylor R, Cannella L, Simpson E, Dazzi F. Minor histocompatibility antigens and stem cell transplantation. Vox
Sang 2004;87(Suppl 2):11 4.
4. Ceppellini R, Mattiuz PL, Scudeller G, Visetti M. Experimental allotransplantation in man. I. The role of the HL-A
system in different genetic combinations. Transplant Proc
1969;1:385389.
5. Pescovitz MD, Sachs DH, Lunney JK, Hsu SM. Localization of class II MHC antigens on porcine renal vascular
endothelium. Transplantation 1984;37:627 630.
6. Bourget JL, Mathes DW, Nielsen GP, Randolph MA,
Tanabe YN, Ferrara VR, et al. Tolerance to musculoskeletal allografts with transient lymphocyte chimerism in miniature swine. Transplantation 2001;71:851 856.
7. Lee WP, Rubin JP, Bourget JL, Cober SR, Randolph MA,
Nielsen GP, et al. Tolerance to limb tissue allografts between swine matched for major histocompatibility complex
antigens. Plast Reconstr Surg 2001;107:14821490; discussion 14911492.
8. Mathes DW, Randolph MA, Lee WP. Strategies for tolerance induction to composite tissue allografts. Microsurgery
2000;20:448 452.
9. Mathes DW, Randolph MA, Solari MG, Nazzal JA,
Nielsen GP, Arn JS, et al. Split tolerance to a composite
tissue allograft in a swine model. Transplantation 2003;75:
2531.
10. Jones NF. Concerns about human hand transplantation in
the 21st century. J Hand Surg 2002;27A:771787.
11. Mori M, Beatty PG, Graves M, Boucher KM, Milford EL.
HLA gene and haplotype frequencies in the North American population: the National Marrow Donor Program Donor Registry. Transplantation 1997;64:10171027.
12. Prabhune KA, Gorantla VS, Maldonado C, Perez-Abadia
G, Barker JH, Ildstad ST. Mixed allogeneic chimerism and
tolerance to composite tissue allografts. Microsurgery
2000;20:441 447.
13. Monaco AP. Chimerism in organ transplantation: conflicting experiments and clinical observations. Transplantation
2003;75(9 Suppl):13S16S.
14. Wood KJ. Passenger leukocytes and microchimerism: what
role in tolerance induction? Transplantation 2003;75(9
Suppl):17S20S.
15. Starzl TE, Demetris AJ, Trucco M, Ramos H, Zeevi A,
Rudert WA, et al. Systemic chimerism in human female
recipients of male livers. Lancet 1992;340:876 877.
16. Starzl TE, Demetris AJ, Murase N, Ildstad S, Ricordi C,
Trucco M. Cell migration, chimerism, and graft acceptance.
Lancet 1992;339:1579 1582.
17. Starzl TE, Demetris AJ, Trucco M, Murase N, Ricordi C,
Ildstad S, et al. Cell migration and chimerism after whole-
1082
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.

organ transplantation: the basis of graft acceptance.
Hepatology 1993;17:11271152.
Ferrand C, Perruche S, Robinet E, Martens A, Tiberghien
P, Saas P. How should chimerism be decoded? Transplantation 2003;75(9 Suppl):50S54S.
Wekerle T, Sykes M. Mixed chimerism and transplantation
tolerance. Annu Rev Med 2001;52:353370.
Siemionow M, Ozer K. Advances in composite tissue allograft transplantation as related to the hand and upper
extremity. J Hand Surg 2002;27A:565 80.
Main JM, Prehn RT. Successful skin homografts after the
administration of high dosage X radiation and homologous
bone marrow. J Natl Cancer Inst 1955;15:10231029.
Ildstad ST, Sachs DH. Reconstitution with syngeneic plus
allogeneic or xenogeneic bone marrow leads to specific
acceptance of allografts or xenografts. Nature 1984;307:
168 170.
Sharabi Y, Sachs DH. Mixed chimerism and permanent
specific transplantation tolerance induced by a nonlethal
preparative regimen. J Exp Med 1989;169:493502.
Game DS, Lechler RI. Pathways of allorecognition: implications for transplantation tolerance. Transpl Immunol
2002;10:101108.
Abbas A, Lichtman A. Basic immunology: Functions and
disorders of the immune system, 2nd ed. Sidcup. W.B.
Saunders, 2003:560.
Lechler RI, Batchelor JR. Immunogenicity of retransplanted rat kidney allografts. Effect of inducing chimerism
in the first recipient and quantitative studies on immunosuppression of the second recipient. J Exp Med 1982;156:
18351841.
Lechler RI, Batchelor JR. Restoration of immunogenicity
to passenger cell-depleted kidney allografts by the addition
of donor strain dendritic cells. J Exp Med 1982;155:31 41.
Lafferty KJ, Cooley MA, Woolnough J, Walker KZ. Thyroid allograft immunogenicity is reduced after a period in
organ culture. Science 1975;188:259 261.
Ko S, Deiwick A, Jager MD, Dinkel A, Rohde F, Fischer R,
et al. The functional relevance of passenger leukocytes and
microchimerism for heart allograft acceptance in the rat.
Nat Med 1999;5:12921297.
Fabre JW, Morris PJ. The effect of donor strain blood
pretreatment on renal allograft rejection in rats. Transplantation 1972;14:608 617.
Wood KJ, Evins J, Morris PJ. Suppression of renal allograft
rejection in the rat by class I antigens on purified erythrocytes. Transplantation 1985;39:56 62.
Josien R, Heslan M, Brouard S, Soulillou JP, Cuturi MC.
Critical requirement for graft passenger leukocytes in allograft tolerance induced by donor blood transfusion. Blood
1998;92:4539 4544.
Zimmermann FA, Davies HS, Knoll PP, Gokel JM,
Schmidt T. Orthotopic liver allografts in the rat. The influence of strain combination on the fate of the graft. Transplantation 1984;37:406 410.
Qian S, Demetris AJ, Murase N, Rao AS, Fung JJ, Starzl
TE. Murine liver allograft transplantation: tolerance and
donor cell chimerism. Hepatology 1994;19:916 924.
Sriwatanawongsa V, Davies HS, Calne RY. The essential
roles of parenchymal tissues and passenger leukocytes in
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
the tolerance induced by liver grafting in rats. Nat Med

1995;1:428 432.
Sriwatanawongsa V, Davies HF, Calne RY. Successful
retransplantation of rat liver grafts. Transplantation 1995;
59:908 910.
Rifle G, Mousson C. Donor-derived hematopoietic cells in
organ transplantation: a major step toward allograft tolerance? Transplantation 2003;75(9 Suppl):3S7S.
Ishida H, Kawai T, Tanabe K, Hayasaka Y, Yasuo M,
Toma H, et al. Status of microchimerism in recipients 15
years after living related kidney transplantation. Transplantation 1996;62:126 128.
Triulzi DJ, Nalesnik MA. Microchimerism, GVHD, and
tolerance in solid organ transplantation. Transfusion 2001;
41:419 426.
Hisanaga M, Hundrieser J, Boker K, Uthoff K, Raddatz G,
Wahlers T, et al. Development, stability, and clinical correlations of allogeneic microchimerism after solid organ
transplantation. Transplantation 1996;61:40 45.
Schlitt HJ, Hundrieser J, Ringe B, Pichlmayr R. Donor-type
microchimerism associated with graft rejection eight years
after liver transplantation. N Engl J Med 1994;330:646
647.
Schlitt HJ, Hundrieser J, Hisanaga M, Uthoff K, Karck M,
Wahlers T, et al. Patterns of donor-type microchimerism
after heart transplantation. Lancet 1994;343:1469 1471.
Hamano K, Rawsthorne MA, Bushell AR, Morris PJ,
Wood KJ. Evidence that the continued presence of the
organ graft and not peripheral donor microchimerism is
essential for maintenance of tolerance to alloantigen in vivo
in anti-CD4 treated recipients. Transplantation 1996;62:
856 860.
Elwood ET, Larsen CP, Maurer DH, Routenberg KL, Neylan JF, Whelchel JD, et al. Microchimerism and rejection in
clinical transplantation. Lancet 1997;349:1358 1360.
Wood K, Sachs DH. Chimerism and transplantation
tolerance: cause and effect. Immunol Today 1996;17:584
587; discussion 588.
Sivasai KS, Alevy YG, Duffy BF, Brennan DC, Singer GG,
Shenoy S, et al. Peripheral blood microchimerism in human
liver and renal transplant recipients: rejection despite donor-specific chimerism. Transplantation 1997;64:427 432.
Owen RD. Immunogenetic consequences of vascular anastomoses between bovine twins. Science 1945;102:400
401.
Anderson D, Billingham RE, Lampkin GH, Medawar PB.
The use of skin grafting to distinguish between monozygotic and dizygotic twins in cattle. Heredity 1951;5:379
397.
Billingham RE, Lampkin GH, Medawar PB, Williams HL.
Tolerance to homografts, twin diagnosis, and the freemartin
condition in cattle. Heredity 1952;6:201212.
Wekerle T, Blaha P, Koporc Z, Bigenzahn S, Pusch M,
Muehlbacher F. Mechanisms of tolerance induction
through the transplantation of donor hematopoietic stem
cells: central versus peripheral tolerance. Transplantation
2003;75(9 Suppl):21S25S.
Sykes M. Mixed chimerism and transplant tolerance. Immunity 2001;14:417 424.
Singer A, Hathcock KS, Hodes RJ. Self recognition in
allogeneic radiation bone marrow chimeras. A radiation-
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
65.
66.
resistant host element dictates the self specificity and

immune response gene phenotype of T-helper cells. J Exp
Med 1981;153:1286 1301.
Ildstad ST, Wren SM, Bluestone JA, Barbieri SA, Stephany
D, Sachs DH. Effect of selective T cell depletion of host
and/or donor bone marrow on lymphopoietic repopulation,
tolerance, and graft-vs-host disease in mixed allogeneic
chimeras (B10 B10.D2B10). J Immunol 1986;136:
28 33.
Pelot MR, Pearson DA, Swenson K, Zhao G, Sachs J, Yang
YG, et al. Lymphohematopoietic graft-vs.-host reactions
can be induced without graft-vs.-host disease in murine
mixed chimeras established with a cyclophosphamidebased nonmyeloablative conditioning regimen. Biol Blood
Marrow Transplant 1999;5:133143.
Markees TG, Phillips NE, Gordon EJ, Noelle RJ, Shultz
LD, Mordes JP, et al. Long-term survival of skin allografts
induced by donor splenocytes and anti-CD154 antibody in
thymectomized mice requires CD4() T cells, interferongamma, and CTLA4. J Clin Invest 1998;101:2446 2455.
Wekerle T, Sayegh MH, Hill J, Zhao Y, Chandraker A,
Swenson KG, et al. Extrathymic T cell deletion and allogeneic stem cell engraftment induced with costimulatory
blockade is followed by central T cell tolerance. J Exp Med
1998;187:20372044.
Wekerle T, Sayegh MH, Ito H, Hill J, Chandraker A,
Pearson DA, et al. Anti-CD154 or CTLA4Ig obviates the
need for thymic irradiation in a non-myeloablative conditioning regimen for the induction of mixed hematopoietic
chimerism and tolerance. Transplantation 1999;68:1348
1355.
Wekerle T, Sykes M. Mixed chimerism as an approach for
the induction of transplantation tolerance. Transplantation
1999;68:459 467.
Nikolic B, Sykes M. Clonal deletion as a mechanism of
transplantation tolerance. J Heart Lung Transplant 1996;
15:11711178.
Tomita Y, Khan A, Sykes M. Role of intrathymic clonal
deletion and peripheral anergy in transplantation tolerance
induced by bone marrow transplantation in mice conditioned with a nonmyeloablative regimen. J Immunol 1994;
153:10871098.
Foster RD, Fan L, Neipp M, Kaufman C, McCalmont T,
Ascher N, et al. Donor-specific tolerance induction in composite tissue allografts. Am J Surg 1998;176:418 421.
Mathes DW, Randolph MA, Bourget JL, Nielsen GP, Ferrera VR, Arn JS, et al. Recipient bone marrow engraftment
in donor tissue after long-term tolerance to a composite
tissue allograft. Transplantation 2002;73:1880 1885.
Wekerle T, Sykes M. Induction of tolerance. Surgery 2004;
135:359 364.
Huang CA, Fuchimoto Y, Scheier-Dolberg R, Murphy MC,
Neville DM Jr, Sachs DH. Stable mixed chimerism and
tolerance using a nonmyeloablative preparative regimen in
a large-animal model. J Clin Invest 2000;105:173181.
Fuchimoto Y, Huang CA, Yamada K, Shimizu A, Kitamura
H, Colvin RB, et al. Mixed chimerism and tolerance without whole body irradiation in a large animal model. J Clin
Invest 2000;105:1779 1789.
Kawai T, Cosimi AB, Colvin RB, Powelson J, Eason J,
Kozlowski T, et al. Mixed allogeneic chimerism and renal
67.
68.
69.
70.
71.
72.
73.
74.
75.
76.
77.
78.
79.
1083
allograft tolerance in cynomolgus monkeys. Transplantation

1995;59:256 262.
Spitzer TR, Delmonico F, Tolkoff-Rubin N, McAfee S,
Sackstein R, Saidman S, et al. Combined histocompatibility
leukocyte antigen-matched donor bone marrow and renal
transplantation for multiple myeloma with end stage renal
disease: the induction of allograft tolerance through mixed
lymphohematopoietic chimerism. Transplantation 1999;68:
480 484.
Millan MT, Shizuru JA, Hoffmann P, Dejbakhsh-Jones S,
Scandling JD, Grumet FC, et al. Mixed chimerism and
immunosuppressive drug withdrawal after HLA-mismatched kidney and hematopoietic progenitor transplantation. Transplantation 2002;73:1386 1391.
Sayegh MH, Fine NA, Smith JL, Rennke HG, Milford EL,
Tilney NL. Immunologic tolerance to renal allografts after
bone marrow transplants from the same donors. Ann Intern
Med 1991;114:954 955.
Jacobsen N, Taaning E, Ladefoged J, Kristensen JK, Pedersen FK. Tolerance to an HLA-B,DR disparate kidney
allograft after bone-marrow transplantation from same donor. Lancet 1994;343:800.
Dey B, Sykes M, Spitzer TR. Outcomes of recipients of
both bone marrow and solid organ transplants. A review.
Medicine (Baltimore) 1998;77:355369.
Kean LS, Adams AB, Strobert E, Hendrix R, Gangappa S,
Jones TR, et al. Induction of chimerism in rhesus macaques
through stem cell transplant and costimulation blockadebased immunosuppression. Am J Transplant 2007;7:320
335.
Suzuki H, Patel N, Matthews M, DelRossi AJ, Doolin EJ,
Hewitt CW. Vascularized bone marrow transplantation: a
new surgical approach using isolated femoral bone/bone
marrow. J Surg Res 2000;89:176 183.
Talmor M, Steinman RM, Codner MA, Chen M, Harper
AD, Witmer-Pack MD, et al. Bone marrow-derived chimerism in non-irradiated, cyclosporin-treated rats receiving
microvascularized limb transplants: evidence for donorderived dendritic cells in recipient lymphoid tissues. Immunology 1995;86:448 455.
Siemionow MZ, Izycki DM, Zielinski M. Donor-specific
tolerance in fully major histocompatibility major histocompatibility complex-mismatched limb allograft transplants
under an anti-alphabeta T-cell receptor monoclonal antibody and cyclosporine A protocol. Transplantation 2003;
76:16621668.
Siemionow M, Oke R, Ozer K, Izycki D, Prajapati R.
Induction of donor-specific tolerance in rat hind-limb allografts under antilymphocyte serum and cyclosporine A
protocol. J Hand Surg 2002;27A:10951103.
Janczewska S, Ziolkowska A, Durlik M, Olszewski WL,
Lukomska B. Fast lymphoid reconstitution after vascularized bone marrow transplantation in lethally irradiated rats.
Transplantation 1999;68:201209.
Durlik M, Lukomska B, Ziolkowska H, Namyslowski A,
Janczewska S, Cybulska E, et al. Microchimerism following allogeneic vascularized bone marrow transplantation
its possible role in induction of posttransplantation tolerance. Ann Transplant 1998;3:24 26.
Ramsamooj R, Llull R, Black KS, Hewitt CW. Composite
tissue allografts in rats: IV. Graft-versus-host disease in
1084
80.
81.
82.
83.
84.
85.
86.
87.
88.
89.
90.
91.
92.
93.
94.
95.
96.
97.

recipients of vascularized bone marrow transplants. Plast
Reconstr Surg 1999;104:13651371.
Hewitt CW, Ramsamooj R, Patel MP, Yazdi B, Achauer
BM, Black KS. Development of stable mixed T cell chimerism and transplantation tolerance without immune
modulation in recipients of vascularized bone marrow allografts. Transplantation 1990;50:766 772.
Tai CY, Strande LF, Eydelman R, Sheng X, VanTran JL,
Matthews MS, et al. Absence of graft-versus-host disease
in the isolated vascularized bone marrow transplant. Transplantation 2004;77:316 319.
Dubernard JM, Owen E, Lefrancois N, Petruzzo P, Martin
X, Dawahra M, et al. First human hand transplantation.
Case report. Transpl Int 2000;13(Suppl 1):S521S524.
Dubernard JM, Owen E, Herzberg G, Martin X, Guigal V,
Dawahra M, et al. [The first transplantation of a hand in
humans. Early results]. Chirurgie 1999;124:358 365; discussion 365367.
Dubernard JM, Owen E, Herzberg G, Lanzetta M, Martin
X, Kapila H, et al. Human hand allograft: report on first 6
months. Lancet 1999;353:13151320.
Hewitt CW, Black KS, Dowdy SF, Gonzalez GA, Achauer
BM, Martin DC, et al. Composite tissue (limb) allografts in
rats. III. Development of donor-host lymphoid chimeras in
long-term survivors. Transplantation 1986;41:39 43.
Bach FH, Albertini RJ, Joo P, Anderson JL, Bortin MM.
Bone-marrow transplantation in a patient with the WiskottAldrich syndrome. Lancet 1968;2:1364 1366.
Waldmann H. The new immunosuppression. Curr Opin
Chem Biol 2003;7:476 480.
Wells AD, Walsh MC, Bluestone JA, Turka LA. Signaling
through CD28 and CTLA-4 controls two distinct forms of
T cell anergy. J Clin Invest 2001;108:895903.
Sho M, Sandner SE, Najafian N, Salama AD, Dong V,
Yamada A, et al. New insights into the interactions between
T-cell costimulatory blockade and conventional immunosuppressive drugs. Ann Surg 2002;236:667 675.
Li XC, Strom TB, Turka LA, Wells AD. T cell death and
transplantation tolerance. Immunity 2001;14:407 416.
Snanoudj R, de Preneuf H, Creput C, Arzouk N, Deroure B,
Beaudreuil S, et al. Costimulation blockade and its possible
future use in clinical transplantation. Transpl Int 2006;19:
693704.
Viola A, Schroeder S, Sakakibara Y, Lanzavecchia A. T
lymphocyte costimulation mediated by reorganization of
membrane microdomains. Science 1999;283:680 682.
Ridge JP, Di Rosa F, Matzinger P. A conditioned dendritic
cell can be a temporal bridge between a CD4 T-helper
and a T-killer cell. Nature 1998;393:474 478.
Bennett SR, Carbone FR, Karamalis F, Flavell RA, Miller
JF, Heath WR. Help for cytotoxic-T-cell responses is mediated by CD40 signalling. Nature 1998;393:478 480.
Schoenberger SP, Toes RE, van der Voort EI, Offringa R,
Melief CJ. T-cell help for cytotoxic T lymphocytes is
mediated by CD40-CD40L interactions. Nature 1998;393:
480 483.
Grewal IS, Flavell RA. CD40 and CD154 in cell-mediated
immunity. Annu Rev Immunol 1998;16:111135.
Grewal IS, Flavell RA. The CD40 ligand. At the center of
the immune universe? Immunol Res 1997;16:59 70.
98. Bluestone JA. New perspectives of CD28-B7-mediated T

cell costimulation. Immunity 1995;2:555559.
99. Banchereau J, Bazan F, Blanchard D, Briere F, Galizzi JP,
van Kooten C, et al. The CD40 antigen and its ligand. Annu
Rev Immunol 1994;12:881922.
100. Kirk AD, Burkly LC, Batty DS, Baumgartner RE, Berning
JD, Buchanan K, et al. Treatment with humanized monoclonal antibody against CD154 prevents acute renal allograft rejection in nonhuman primates. Nat Med 1999;5:
686 693.
101. Iwakoshi NN, Mordes JP, Markees TG, Phillips NE,
Rossini AA, Greiner DL. Treatment of allograft recipients
with donor-specific transfusion and anti-CD154 antibody
leads to deletion of alloreactive CD8 T cells and prolonged graft survival in a CTLA4-dependent manner. J Immunol 2000;164:512521.
102. Gudmundsdottir H, Turka LA. T cell costimulatory
blockade: new therapies for transplant rejection. J Am Soc
Nephrol 1999;10:1356 1365.
103. Larsen CP, Elwood ET, Alexander DZ, Ritchie SC, Hendrix R, Tucker-Burden C, et al. Long-term acceptance of
skin and cardiac allografts after blocking CD40 and CD28
pathways. Nature 1996;381:434 438.
104. Kirk AD, Harlan DM, Armstrong NN, Davis TA, Dong Y,
Gray GS, et al. CTLA4-Ig and anti-CD40 ligand prevent
renal allograft rejection in primates. Proc Natl Acad Sci U
S A 1997;94:8789 8794.
105. Iwasaki N, Gohda T, Yoshioka C, Murakami M, Inobe M,
Minami A, et al. Feasibility of immunosuppression in composite tissue allografts by systemic administration of
CTLA4Ig. Transplantation 2002;73:334 340.
106. Kawai T, Andrews D, Colvin RB, Sachs DH, Cosimi AB.
Thromboembolic complications after treatment with monoclonal antibody against CD40 ligand. Nat Med 2000;6:114.
107. Elster EA, Blair PJ, Kirk AD. Potential of costimulationbased therapies for composite tissue allotransplantation.
Microsurgery 2000;20:430 434.
108. Heidecke CD, Hancock WW, Jakobs F, Westerholt S,
Sewzcik T, Deusch K, et al. TCR alpha/beta targeted therapy in the rat: pretreatment with R73 monoclonal antibody
induces profound immunological anergy and long-term allograft survival. Transplant Proc 1993;25(1 Pt 1):540 542.
109. Alcover A, Alarcon B. Internalization and intracellular fate
of TCR-CD3 complexes. Crit Rev Immunol 2000;20:325
346.
110. Ozer K, Izycki D, Zielinski M, Siemionow M. Development of donor-specific chimerism and tolerance in composite tissue allografts under alphabeta-T-cell receptor monoclonal antibody and cyclosporine a treatment protocols.
Microsurgery 2004;24:248 254.
111. Heidecke CD, Hancock WW, Jakobs F, Zantl N, Kurrle R,
Westerholt S, et al. alpha/beta-T cell receptor-directed therapy in rat cardiac allograft recipients. Treatment prior to
alloantigen exposure prevents sensitization and abrogates
accelerated rejection. Transplantation 1995;59:78 84.
112. Kampinga J, Schuurman HJ, Pol GH, Bartels H, Broekhuizen R, Vaessen LM, et al. Vascular thymus transplantation
in rats. Technique, morphology, and function. Transplantation 1990;50:669 678.
113. Klimczak A, Agaoglu G, Carnevale KA, Siemionow M.
Applications of bilateral vascularized femoral bone marrow

transplantation for chimerism induction across the major
histocompatibility (MHC) barrier: part II. Ann Plast Surg
2006;57:422 430.
114. Ozkaynak E, Wang L, Goodearl A, McDonald K, Qin S,
OKeefe T, et al. Programmed death-1 targeting can
promote allograft survival. J Immunol 2002;169:6453
6546.
115. Curry AJ, Chikwe J, Smith XG, Cai M, Schwarz H, Bradley JA, et al. OX40 (CD134) blockade inhibits the co-
1085
stimulatory cascade and promotes heart allograft survival.

Transplantation 2004;78:807 814.
116. Demirci G, Amanullah F, Kewalaramani R, Yagita H,
Strom TB, Sayegh MH, et al. Critical role of OX40 in
CD28 and CD154-independent rejection. J Immunol 2004;
172:16911698.
117. Cho HR, Kwon B, Yagita H, La S, Lee EA, Kim JE, et al.
Blockade of 4-1BB (CD137)/4-1BB ligand interactions
increases allograft survival. Transpl Int 2004;17:351361.
REVIEW ARTICLE
Stability of the Distal Radioulna Joint:
Biomechanics, Pathophysiology, Physical
Diagnosis, and Restoration of Function
What We Have Learned in 25 Years
William B. Kleinman, MD
From The Indiana Hand Center and the Indiana University School of Medicine, Indianapolis, IN.
quarter century of research by some of the

brightest minds in the field of hand surgery
has illuminated what some have considered
in the past to be a black box, namely, rotational and
translational mechanics at the distal radioulna joint
under load. Lack of understanding of the details of
anatomy and biomechanics at the distal end of the
ulna resulted in 75 years of simple resection of the
distal ulna as treatment for most disabling pathology
in this part of the distal forearm. Over the past 25
years, brilliant contributions have been made to further our understanding of this complex arena of hand
surgery, especially focused on the bony anatomy of
the distal radioulna joint, the extrinsic dynamic stabilizers of forearm rotation, and the critical components of the triangular fibrocartilage. With a deeper
understanding of anatomy, clinical evaluation of disabling pain at the distal end of the ulna has become
more precise. Provocative maneuvers have been designed to increase the effectiveness of physical diagnosis following injuries to the triangular fibrocartilage. A greater appreciation of distal forearm
biomechanics has provided more reliable reconstructive procedures to the hand surgeon. This Clinical
Perspective emphasizes how far our understanding of
disorders at the distal end of the ulna has come in the
past 25 years.
One of the most exciting areas in hand surgery
during the past quarter century has been the study of
anatomy, biomechanics, and pathophysiology at the
distal end of the ulna.1 Painless rotational stability of
the forearm under load is now recognized by hand
surgeons as critical to the functional capacity of the
entire upper limb.219 We also now recognize that
stiffness of forearm rotation may be as functionally
incapacitating as painful instability in this area of the
distal forearm.20
During the past 25 years, a plethora of landmark
1086
basic science papers and published clinical investigations have provided insight into: (1) normal kinematics at the distal radioulna joint (DRUJ)21; and (2)
how the critical ligaments that support the DRUJ can
fail.22 We have been instructed in a broad spectrum
of surgical procedures recommended to restore lost
DRUJ function. As published details of DRUJ anatomy and biomechanics have continued to illuminate
some of the academic darkness surrounding the distal
end of the ulna, more precise and successful evidence-based reconstructive procedures have been designed.
Significant Advances in Our Understanding

of Anatomy and Biomechanics
The longitudinal axis of forearm pronosupination
passes through the center of the radial head proximally
and through the foveal sulcus at the lateral base of the
ulna styloid distally (Fig. 1). Af Ekenstam and Hagert21
defined the pole of the distal ulna as that portion adjacent to the triangular fibrocartilage (TFC), covered by
hyaline cartilage, and responsible for absorption of load
transferred to the forearm from the medial carpus
through the articular disc of the TFC onto the ulna. The
fovea is the recess lying between the hyaline cartilage
of the ulna pole and the ulna styloid. This fossa is richly
vascularized; it serves as a point of insertion of the
major DRUJ stabilizing ligamentous components of the
TFC. Twenty-five years of research have now enabled
us to understand how the anatomic position of the fovea
and the deep components of the TFC are critical to the
rotational and translational stability of the DRUJ. Anatomic works by Bednar et al in 1991,23 and ThiruPathi et al in 198624 demonstrate the rich vascularity
within the fovea. These authors also described the
source of vascular nutrition of all peripheral tissues of
the TFC, both dorsal and palmar. Their two landmark
papers also describe the avascular, central articular disc
William B. Kleinman / Stability of the Distal Radioulna Joint
1087
Figure 1. The longitudinal axis of rotation of the forearm (red bar) passes through the head of the radius proximally and through
the fovea of the ulna distally.
of the TFC, nourished by synovial fluid from both the

DRUJ and the ulnocarpal joint.
At the ginglymus ulnotrochlear joint of the elbow,
the ulna participates only in forearm flexion and
extension; it does not participate in rotation. Forearm
pronosupination involves rotation of the radiocarpal
unit (with the attached hand) around a rotationally
fixed and stable ulna. The complex architecture of
the ulnotrochlear articular surfaces and the collateral
ligament stabilizers of the elbow allow the ulna to
move as a hinge only, without a rotational component. The longitudinal axis around which rotation
takes place can be seen in Figure 1. The anatomy of
the sigmoid notch of the radius, the seat of the ulna,
which articulates with the notch, and the guiding,
check-rein potential of the components of the TFC,
allow 90 of forearm supination, at which point the
two forearm bones are essentially parallel, and the
interosseous space the widest, to 90 of pronation, at
which point the radius has rotated across the anterior
surface of the fixed ulna (Fig. 2). The principal axis
of load bearing at the DRUJ tracks obliquely across
the sigmoid notch as the radiocarpal unit pronosupinates around the fixed ulna (Fig. 3). (The principal
axis is an engineering term used to define an
imaginary point at the center of an infinite number of cluster points between two loaded surfaces,
in contact with each other.) The tracking line is
from distal/dorsal in pronation to proximal/palmar in
supination.
In the human bipedal condition, with elbows at the

side and flexed 90 (prepared for work), the radiocarpal unit rests on top of the ulna seat, with
gravity pulling the hand and its load towards the
ground (Fig. 4). This so-called zero-rotation position of the forearm has been nicely described by
Palmer and Werner.25
In this positionthe most common for human
forearm functiona significant joint reaction force
(JRF) can develop between the sigmoid notch of the
radius and the rotationally fixed ulna seat (Fig. 4).
Standard teaching and practice advises viewing xrays of the carpus with the fingers towards the ceiling. The reorientation of Figure 4 allows the reader to
appreciate the zero-rotation position of the working forearm, with the radius and ulna styloids at their
widest anatomic separation and the forearm in neutral rotation (0 pronosupination). In this position, the
principal axis of load bearing at the DRUJ is between
the center of the sigmoid notch and the center of the
seat of the ulna. In zero rotation, the various radioulna ligament components of the TFC are under the
least amount of tension.21,26
The structural presence and health of the articular
surface of the ulna seat is critical in providing a
painless mechanical fulcrum for all radioulna loadbearing activity. In a state of equilibrium (no forearm
motion), all moments around a fixed fulcrum (ie, the
seat of the ulna) must be equal (Fig. 5). The loaded
hand (F) times the length of the loaded hand from the
1088
Figure 2. As the radius rotates from full supination to full pronation around a fixed ulna, the radiocarpal unit shortens relative
to the ulna, resulting in ulna-plus variance in the pronated position.
fulcrum (L) must be equal to the moment (F L)

on the proximal side of the fulcrum, where F is the
stability provided at the radiocapitellar joint by the
annular ligament encircling the radial head, and L is
the entire length of the forearm. Because F L F
L, the relatively long forearm makes the requirement for proximal radius stability by the annular
ligament (F) relatively small, regardless of the load
in the hand.
The sum of all moments distal and proximal to
this fulcrum equals the total load on the fulcrum
(the ulna seat) itself, defined as the JRF at the
DRUJ (Fig. 6). Based on how much load is in the
hand (F), and the size of the moment distal to the
DRUJ, the reader begins to appreciate the potential
magnitude of the JRF at the DRUJ and the importance of surface hyaline cartilage health at both the
sigmoid notch and the ulna seat for painless DRUJ
function.
Using elegant cadaver dissections, in 1985 Af Ekenstam and Hagert et al.21,27 demonstrated that the
concave radius-of-curvature of the sigmoid notch is
greater than that of the ulna seat (Fig. 7). This incongruity of articular surfaces creates a geometrically
nonconstrained articulation at the DRUJ, subject to
translational dorsal and palmar instability. Not only
does the radiocarpal unit rotate around the fixed ulna

seat in pronosupination, but also the flatter surface of
the sigmoid notch has enough inherent instability
through incongruous cartilage surfaces to allow
translation of the notch palmarly or dorsally on the
fixed ulna as the forearm rotates into pronation or
supination, respectively. The DRUJ is not a ball-andsocket joint. Figure 3 reveals the exposed cadaver
sigmoid notch (the ulna has been dissected free and
hinged palmarly out of the notch to the readers left),
showing the tracking line of the principal axis of
forearm load bearing at the DRUJ (see above). Note
that the line is oblique. Because rotation of the radius
across the ulna from supination to pronation results
in relative ulna-minus variance25 at the end arc of
pronation, the tracking line of the principal axis of
load bearing across the sigmoid notch is oblique
(from slightly more proximal at the palmar edge of
the notch in full supination, to slightly more distal at
the dorsal edge of the notch in full pronation).
Stability of the Distal Radioulna Joint

With inherently unstable, nonconstrained articular
surfaces, anatomic stability of the DRUJ is achieved
through extrinsic extracapsular as well as intrinsic
intracapsular structures. Extrinsic stability is pro-
1089
Figure 4. The seat of the ulna is the fulcrum for all distal
radioulna joint mechanics. Because most upper limb activities in the bipedal human occur with the radiocarpal unit on
top of the ulna seat, the joint reaction force (JRF) at the distal
radioulna joint can be enormous. The joint reaction force is
proportional to the load in the hand, the force of all muscles
acting to pull the radius and ulna together for stability, and
the force of gravity acting on the hand-forearm unit.
Figure 3. The principal axis of load bearing tracks across the

sigmoid notch from proximal/palmar in supination to distal/
dorsal in pronation.
vided principally by (1) dynamic tensioning of the

extensor carpi ulnaris as its tendon crosses the distal
head of the ulna,17,28 (2) the semirigid sixth dorsal
compartment itself, constraining the extensor carpi
ulnaristendon,17,28 (3) dynamic support provided by
the superficial and deep heads of the pronator quadratus,29 and (4) the interosseous ligament of the midforearm30 (Fig. 8). These extrinsic DRUJ stabilizers
are of relatively minor consequence to rotational
forearm stability, compared with the more biomechanically effective intrinsic radioulna components
of the TFC.21,23,27 Dorsal and palmar radioulna TFC
fibers arise from the medial border of the distal radius
and insert on the ulna at two separate and distinct
sites: the fovea at the base of the ulna styloid and the
ulna styloid itself. Figure 9 emphasizes the wellvascularized nature of the dorsal and palmar radioulna ligaments. Figure 10 illustrates the critical and
clinically significant difference between the angle of
attack of the dorsal and palmar superficial radioulna
fibers inserting on the ulna styloid and the deep fibers
inserting onto the fovea. Well-vascularized, longitu-
dinally oriented connective tissue fibers of the TFC

anchor the radius to the ulna along both its dorsal and
palmar margins. The blood supply to both these areas
of the periphery of the TFC are through branches of
the posterior interosseous artery (Fig. 9).23,24 These
vessels course along the dorsal and palmar radioulna
ligaments penetrating and nourishing the dorsal 20%
and palmar 20% of the TFC.23,24 Between these two
Figure 5. In equilibrium, the moments on the distal and

proximal sides of the ulna seat fulcrum must be equal. The
load in the hand (F) times the distance of the load from the
fulcrum (L) must be equal to the length of the forearm from
the fulcrum (L) times the resistance to displacement provided
by the annular ligament at the radial head (F). F x L F x L.
1090
Figure 6. The sum of the moments on the distal and proximal

sides of the fulcrum equal the joint reaction force (JRF) at the
distal radioulna joint.
sets of DRUJ check-reins, the articular disc is nourished by synovial fluid. The disc is principally responsible for load transmission from the medial carpus to the forearm, especially with the wrist in ulnar
deviation. In neutral deviation of the hand-forearm
unit, the principal axis of load transmission passes
from the hand, through the head of the capitate,
through the scapholunate ligament, and onto the distal radius articular surface at the interfossal ridge,
which separates the elliptical (scaphoid-bearing) and
spherical (lunate-bearing) fossae of the distal radius
(Figs. 10 and 11). Af Ekenstam, Palmer, etal27 have
Figure 8. Extrinsic stabilizers of the inherently unstable distal

radioulna joint include: (1) the tendon of the extensor carpi
ulnaris, (2) the sixth dorsal compartment subsheath, (3) the
superficial and deep heads of the pronator quadratus, and (4)
the interosseous ligament of the forearm. Even considered all
together, the extrinsic DRUJ stabilizers are fairly ineffective in
physiologically maintaining DRUJ stability through the arc of
pronosupination under load.
demonstrated in the laboratory that when the wrist is

in neutral position, 84% of hand load is transferred to
the radius, and only 16% is transferred through the
central articular disc of the TFC. With ulnar deviation of the hand-forearm unit (Figs. 12 and 13), the
principal axis of load bearing shifts medially, placing
more load on the articular disc and the pole of the
distal ulna, rather than on the interfossal ridge of the
distal radius when the hand-forearm unit is in neutral
deviation.
Figure 7. Transverse section through the distal radioulna

joint. The radius of curvature of the sigmoid notch is greater
than the radius of curvature of the seat of the ulna, leading to
an inherent instability of the distal radioulna joint through an
arc of pronosupination. Rotation of the forearm around a
longitudinal axis of rotation (see Fig. 1) is manifest at the
distal radioulna joint by rotation and translation of the sigmoid notch against the ulna seat (from Af Ekenstam and
Hagert21).
How The TFC Components Guide the

Radiocarpal Unit Around a Fixed Ulna: The
Significance of Rotation and Translation
The dorsal and palmar radioulna ligaments consist of
superficial components inserting directly onto the
ulna styloid and deep components inserting more
lateral, into the fovea adjacent to the articular surface
of the pole of the distal ulna (Fig. 14, from Af
Ekenstam and Hagert21). These two components of
the TFC are distinct in both their anatomy and func-
1091
Figure 9. The prime intrinsic stabilizer of the distal radioulna joint is the triangular fibrocartilage (TFC). The TFC complex
consists of superficial (green) and deep (blue) radioulna fibers, the two disc-carpal ligaments (disc-lunate and disc-triquetral), and
the central articular disc (white). The articular disc is responsible for transferring load from the medial carpus to the pole of the
distal ulna. The vascularized, peripheral radioulna ligaments (green and blue) are nourished by dorsal and palmar branches of
the posterior interosseous artery and are responsible for guiding the radiocarpal unit around the seat of the ulna.
tion. Scrutiny of Figure 14 reveals that fibers of the

superficial component form an acute angle as they
converge on the ulna styloid from the medial radius.
This narrow angle of attack gives the superficial TFC
a poor mechanical advantage for guiding the radiocarpal unit through an arc of pronosupination. The
deep components of the TFC (arising along the medial border of the distal radius but inserting into the
fovea), however, form an obtuse angle of attack,
much more mechanically advantageous in stabilizing
rotation of the radius around the fixed ulna (Fig. 14).
The deep components of the TFC have been referred
to by wrist investigators as the Ligamentum subcruentum. In his landmark 1975 article on the Articular disc of the hand, Kauer31 gives credit to
Henle32 and Fick33 for describing a vascularized
fissure between the superficial and deep components
of the TFC, which they called the ligamentum sub-
cruentum, technically not a ligament at all. Over the

past 20 years, however, the term Ligamentum subcruentum has come to represent the deep fibers of the
TFC (inserting into the fovea) and is now used commonly by many investigators as interchangeable with
the term deep TFC radioulna ligaments.
Figures 11 and 13 schematically represent the superficial radioulna ligaments (green) and the deep
Ligamentum subcruentum (blue) in transverse section through the ulnocarpal joint. The distinct differences in fiber orientation and mechanical advantage
of the deep (blue) radioulna ligaments in controlling
forearm rotation are apparent. It might be easier,
though, to understand the relative difference in the
effectiveness of the superficial (green) and deep
(blue) radioulna ligaments of the TFC in controlling
physiologic forearm rotation and translation using an
analogy of a team of horses, a buckboard, and a
1092
driver (Fig. 15): The radius is represented by the

team of horses; the buckboard is the fixed ulna. The
buckboard driver holds the reins at the buckboard
seat (the foveal sulcus of the ulna). The angle of
attack of the blue reins to the two outside horses of
the team represents the angle of attack of the deep
dorsal and palmar fibers of the Ligamentum subcruentum from the radius (the team of horses) to the
ulna fovea (the seated buckboard driver). Their orientation makes the angle of attack on the outside
horses of the team much more effective in turning the
team of horses (the radius) around the buckboard (the
seat of the ulna). In Figure 16, the narrower, acute
angle of attack of the superficial (green) TFC fibers is
represented as green reins connecting the driver to
the central horses of the team. This angle of attack is
much less effective in controlling rotation of the team
relative to the driver. Like the diagrammatic obtuse
blue fibers of the Ligamentum subcruentum in Figures 11 and 13, the blue reins are much more effective in controlling the team of horses, by virtue of
their obtuse angle of attack on the two outside horses.
Af Ekenstam and Hagert first suggested in 1985,21
that the deep fibers of the TFC (blue) were the
principal intrinsic stabilizers of the DRUJ. Although
these investigators used cadaver dissections and experimental techniques, which might be considered
somewhat basic by todays standards, their conclusions were sound: in forearm supination the dorsal,
deep fibers of the Ligamentum subcruentum tighten
significantly, while the deep palmar fibers remain
lax. This suggests a pulling, tethering mechanism for
controlling stability during DRUJ rotation (Fig. 17).
Conversely, these authors found that the palmar,
deep fibers of the Ligamentum subcruentum were the
principal restraints against superphysiologic palmar
migration of the radiocarpal unit at the sigmoid notch
in pronation (Fig. 17).
In 1991, 6 years after general consensus throughout the hand surgery world about the principal intrinsic stabilizers of forearm rotation, a landmark paper
was published by Schuind et al26 from the Mayo
Clinic Biomechanics Laboratory, refuting Af Ekenstams and Hagerts21 1985 work and suggesting a
totally different mechanism for forearm stability. Using a sophisticated stereophotogrammatic technique
with phosphorescent markers and computer analysis,
the authors concluded that the dorsal fibers of the
TFC tighten in pronation, and the palmar fibers
tighten in supination, conclusions opposite of those
published by Af Ekenstam and Hagert, 6 years earlier.
Figure 10. In neutral deviation, 84% of load the load in the

hand passes to the forearm through the radius, with the
principal axis of load bearing passing through the scapholunate ligament and onto the interfossal ridge of the distal
radius. Only 16% of the entire load transferred from the hand
to the forearm is borne by the articular disc of the TFC in
neutral deviation.
Controversy raged in the academic hand surgery

world until 1994, when Hagert, carefully studying
the conflicting conclusions, recognized that both
research groups were correct but that each was
examining a different piece of the puzzle. In a
small, but brilliant work published in 1994, Hagert
clarified the biomechanical effect of each component of the TFC for the first time.34 In this work, he
reasoned that his earlier publication21 studied only
the deep components of the TFC, those inserting
into the fovea of the ulna. This was a consequence
of aggressive excision of the central articular disc
for the 1985 study, which affected the integrity of
the dorsal and palmar superficial TFC radioulna
ligaments. He also implied in his 1994 article that
the phosphorescent markers applied to the surface
of the TFC by Schuind et al26 at the Mayo Clinic
in their 1991 article measured developing tension
1093
Figure 11. The deep radio-ulna fibers of the TFC (dorsal and palmar) originate at the dorsal and palmar edges of the medial
border of the distal radius, and insert onto the fovea of the ulna (blue fibers, referred to as the Ligamentum subcruentum). Their
obtuse angle-of-attack makes them particularly effective in guiding the radius around the ulna through a functional arc of
pronosupination. In compressive mode, with the hand-forearm unit in neutral deviation, the principal axis of load-bearing passes
through the scapholunate ligament onto the articular surface of the distal radius.
or tightening of the superficial radioulna ligaments

only, as these fibers enveloped the articular disc.
The surface phosphorescent markers, however, did
not consider the biomechanics of the radiofoveal
deep fibers, with their much more mechanically
effective angle of attack. Hagert clearly stated in
1994 that in forearm pronation, the dorsal superficial fibers of the TFC must tighten for stability, as
do the deep palmar fibers of the Ligamentum subcruentum. Conversely, in supination, the palmar
superficial TFC radioulna fibers (to the ulna
styloid) tighten, as do the deep dorsal fibers of the
Ligamentum subcruentum, making both theories
correct (Fig. 18).
Two critical anatomic factors are responsible for
a more significant stabilizing effect of the Ligamentum subcruentum on forearm rotation than the
superficial radioulna ligaments of the TFC: first,
full forearm pronation allows the principal axis of
load bearing at the DRUJ to track distally and
dorsally along the sigmoid notch to a point where
less than 10% of the dorsal notch is still in contact
with the articular seat of the ulna.21 In this position, most of the hyaline cartilage-covered distal
end of the ulna has herniated against the dorsal
DRUJ capsule,20 out from under the confining
cover of the dorsal superficial radioulna fibers of
the TFC that attach to the bony ulna styloid. In

maximum pronation, DRUJ stability is based almost entirely on the restraining, pulling action of
the palmar, deep fibers of the Ligamentum subcruentum, preventing superphysiologic translation
from occurring. Second, the obtuse angle of attack
of these deep fibers (recall the analogy of the team
of horses and the buckboard) is perfectly oriented
to prevent DRUJ subluxation. More than any other
intrinsic or extrinsic DRUJ-stabilizing components
discussed above, these two anatomic considerations are responsible for DRUJ stability.
Conversely, in full supination (as the sigmoid
notch migrates dorsally on the ulna seat to less
than 10% articular surface contact21) the superficial palmar TFC fibers (green) become ineffective
as the ulna head rolls out beyond their effective
confines, stretching the palmar DRUJ capsule.20 At
the same time, the dorsal Ligamentum subcruentum tightens, and its obtuse orientation becomes
mechanically advantageous in restraining the radiocarpal unit from superphysiologic dorsal migration on the seat of the ulna. The profound stabilizing importance of the Ligamentum subcruentum
relative to the superficial fibers of the TFC becomes quite clear.
1094
mentum subcruentum using precise provocative maneuvers and physical diagnostic techniques.
If, on the basis of 25 years of accumulated knowledge, it is assumed that in full forearm supination the
dorsal radiofoveal fibers of the Ligamentum subcruentum are under maximum tension (the palmar
superficial fibers are also under tension but less mechanically advantageous), and in full forearm pronation the palmar radiofoveal fibers of the Ligamentum
subcruentum are under maximum tension, then the
dorsal and palmar components of the deep TFC can
be stress tested, employing these two provocative
maneuvers:
Figure 12. With hand-forearm unit in ulnar deviation, the

principal axis of load transfer is shifted medially onto the
articular disc of the TFC, away from the distal radius. The
principal axis now passes through the triquetrohamate joint
onto the pole of the distal ulna.
Provocative Maneuvers for Determining the

Health and Integrity of the Ligamentum
Subcruentum: The Perpetual Role of
Physical Diagnosis in Hand Surgery
Direct arthroscopy of the TFC through the ulnocarpal
joint can, unfortunately, only provide information
about the integrity and health of the superficial dorsal
and palmar fibers of the TFC and the articular disc
(Fig. 19). Ulnocarpal arthroscopy will not allow the
surgeon visualization of the deeper Ligamentum subcruentum, unless the superficial TFC has been ruptured and retracted from its moorings on the bony
ulna styloid or if a large, central degenerative hole
has been worn through the articular disc.37 The clinician can, however, examine the health and integrity
of the dorsal and palmar components of the Liga-
(1) The examiner sits opposite the patient, with the

patients elbow on the examining table and the
fingers towards the ceiling. The patients forearm
is rotated into full supination. In this position, the
dorsal fibers of the Ligamentum subcruentum
should be under maximum tension. The examiner then pushes the distal ulna towards the patient while pulling the radiocarpal unit towards
himself (Fig. 20). This maneuver introduces a
superphysiologic load into the DRUJ. It will be
painless only if the dorsal fibers of the Ligamentum subcruentum are healthy. If inflamed or suffering from relatively minor injury, the two forearm bones will be grossly stable on stress testing,
but the patient will experience considerable pain
on loading the DRUJ beyond its physiologic
limits. If the deep dorsal fibers have been severely sprained and detached from the fovea, this
maneuver will result not only in pain but also in
superphysiologic movement of the sigmoid
notch off the seat of the ulna, resulting in subtle
subluxation or even gross instability, depending
on the magnitude of injury to the dorsal fibers of
the Ligamentum subcruentum.
(2) Figure 21 demonstrates the clinician stress testing the deep palmar Ligamentum subcruentum
by applying a dorsally directed superphysiologic
load to the distal ulna (along the vector of the
arrow), while supporting the forearm in full pronation. The hand-forearm unit is gently pulled
towards the examiner, while the examiners
thumb pushes the ulna towards the patient. If the
deep palmar radioulna portion of the TFC is
either ruptured or attenuated, the sigmoid notch
will translate beyond its normal end-arc relationship to the seat of the ulna, resulting in painful
instability in full pronation.
Physical findings of pain, instability, or both with
1095
Figure 13. Deep fibers of the TFC (Ligamentum subcruentum) insert onto the ulna fovea at an obtuse angle-of-attack (blue).
Superficial radioulna fibers (green) insert onto the ulna styloid and have little function in controlling forearm rotation at the distal
radioulna joint. As the principal axis of load-bearing shifts onto the TFC in ulnar deviation, the articular disc is supported by the
superficial palmar and dorsal (green) radioulna ligaments, attaching directly to the bony ulna styloid.
these two provocative maneuvers must always be

compared with the opposite, uninjured extremity.
These physical diagnostic techniques are highly
reliable. They are precise in detecting painful insta-
Figure 14. The obtuse angle of attack of the Ligamentum

subcruentum (blue) from the medial radius to the fovea is
much more mechanically advantageous in guiding the radius
around the seat of the ulna through a full arc of pronosupination. The superficial radioulna ligaments (green) are much
less effective in controlling distal radioulna rotation and
translation (from Af Ekenstam and Hagert21).
bility of the DRUJ, whether secondary to inflammatory arthropathy or trauma. The importance of provocative maneuvers that can isolate the deep TFC
Ligamentum subcruentum from the superficial radioulna ligaments cannot be overstated, especially when
arthroscopic evaluation of the Ligamentum subcruentum through an ulnocarpal portal cannot be achieved
(Fig. 19). In full supination, essentially all load introduced by the examiners hand stresses the dorsal
Ligamentum subcruentum (Fig. 22, blue), as the seat
of the ulna is translated along the sigmoid notch
beyond the confines of the superficial palmar TFC
fibers (green). In full pronation, introduction of a
superphysiologic load by the examiners thumb
against the distal ulna stress tests the palmar Ligamentum subcruentum, eliciting pain and/or instability
in its pathologic state (Fig. 23, blue). Physical diagnosis of injury to the deep fibers of the TFC should
now be considered a gold standard in the clinical
evaluation of the TFC. Once the clinician understands the significance of TFC fiber orientation and
can appreciate the importance of the critical angle of
attack of the deep TFC fibers (blue) from the medial
radius to the ulna fovea, injuries to the TFC leading
to either subtle or gross instability can be readily
identified by careful physical examination.
Over the past 15 years, a new generation of wrist
coils has improved the diagnostic potential of mag-
1096
Figure 15. The TFC Buckboard analogy. Outer reins from the driver seated on the buckboard (the ulna) to the outside horses
of the team easily control the entire team because of their wide angle of attack from the seat of the buckboard to the horses. This
wide angle represents the same effectiveness of the angle of attack of the deep Ligamentum subcruentum of the TFC from the
radius to the fovea of the ulna.
Figure 16. A narrower angle of attack of green reins on the central horses of the team is much less effective in controlling the
team. These more acutely angled reins represent the acutely angled and less-effective green fibers of the TFC (superficial and
inserting onto the ulna styloid).
1097
Figure 17. (A, B) Cadaver photos illustrating palmar dislocation of the radius in pronation following division of the TFC palmar
ligaments. (C, D) Cadaver photos illustrating dorsal dislocation of the radius in supination following division of the TFC dorsal
ligaments (from Af Ekenstam and Hagert21).
netic resonance imaging (MRI). There are many published studies corroborating MRI findings by open
surgical exploration; diagnostic accuracy has become
much more reasonable in recent years. But, MRI with
or without gadolinium contrast should not be regarded as a substitute for a thorough physical examination of the DRUJ and the TFC, particularly
through use of direct tissue palpation and stress testing techniques. The MRI in Figure 24A clearly
shows integrity of the superficial radioulna ligaments
arising from the medial radius and inserting into the
ulna styloid, but contrast can be observed streaming
around the pole of the distal ulna, across the fovea to
the base of the styloid, suggesting complete avulsion
of the Ligamentum subcruentum deep TFC fibers
from the foveal sulcus. This massive injury to the
deep components of the TFC resulted in painful
dorsal and palmar translational instability at the
DRUJ in this patient. Provocative stress tests (described above) were clinically positive in both pronation and supination. Figure 24B demonstrates intra-operative findings of complete avulsion of the
Ligamentum subcruentum from the ulna fovea.
While possibly corroborating the clinicians physical
findings, MRI should be considered only helpful, but
not definitively diagnostic, in assessing the condition
of a patients TFC. The incidence of false-positive
and false-negative MRI findings still remains very
high today. Imaging always requires clinical correlation by the examining physician.
Displaced Basilar Ulna Styloid Fractures

With new-found understanding of the integral role
played by the Ligamentum subcruentum in DRUJ
mechanics, more critical attention should be given to
displaced basilar ulna styloid fractures, so often oc-
1098
Figure 18. Reproduction of Hagerts 1994 work illustrating

the effectiveness of the deep fibers of the TFC in controlling
distal radioulna rotation and the relative ineffectiveness of
the superficial fibers. The critical factor is the angle of attack
of the Ligamentum subcruentum from the radius to the fovea
of the ulna (from Hagert34).
curring in association with fractures of the distal

radius. Laboratory cadaver studies by Viegas et al in
199035,36 demonstrate certain conditions necessary
for traumatic ulna styloid failure and displacement.
Their results suggest that with certain losses of radius
length, palmar tilt, angle of inclination, or all three,
associated with distal radius fractures, enough tension could be placed on the ulna styloid through the
intact superficial radioulna ligaments of the TFC that
the bony styloid could be avulsed from its base as the
fractured distal radius fragments displace. With the
ulna styloid (and TFC) intact, the authors could cut
the radius but achieve only the following in their
laboratory: (1) radius shortening of 2 and 4 mm, (2)
an angle of radius inclination of 10, and (3) a radius
tilt of 0. If, however, they first cut the ulna styloid
through its base, the radius could be shortened beyond 4 mm, the inclination could be reduced to at
least 0, and a dorsal tilt of 15 or 30 could be
achieved. The clinical implication of their work is
clear: a distal radius fracture with metaphyseal collapse and shortening or dorsal tilt beyond the authors experimental methods, or both, could result in
either a concomitant displaced fracture of the ulna
styloid (an avulsion fracture via intact superficial
TFC components inserting onto the bony ulna
styloid) or an avulsion or tear of the TFC superficial
ligaments from the styloid itself, with the bones
remaining intact.35,36 In the clinical setting, one can
readily appreciate that with displaced distal radius
Figure 19. The Ligamentum subcruentum (deep dorsal and palmar fiber of the TFC) cannot be visualized by ulnocarpal
arthroscopy. Only injuries of the superficial radioulna ligaments and articular disc can be seen.
1099
Consequences of Failing to Appreciate TFC

Anatomy When Treating Displaced Distal
Radius Fractures
Figure 20. Stress testing the dorsal, deep fibers of the Ligamentum subcruentum for pain, mechanical instability, or
both (findings must be compared with the opposite, uninjured side).
fractures, if the ulna styloid has been avulsed at its

base by the superficial TFC fibers (Fig. 25B, green)
that insert directly into the bony ulna styloid, then, by
definition, the deep fibers (blue) of the Ligamentum
subcruentum must also be avulsed and displaced
from the fovea of the ulna, precisely the same distance as the ulna styloid displacement from its bony
base (Fig. 25A, B). The potential effect of a Ligamentum subcruentum injury of this nature on DRUJ
stability, with or without concomitant avulsion of the
ulna styloid through its base, should be obvious. The
hand surgeon must exercise judgment in determining
whether the degree of ulna styloid displacement is
sufficient to justify surgical open reduction and stabilization. By surgically anchoring the displaced
basilar ulna styloid fracture back to its bony base by
open reduction and internal fixation (at the same time
as any required anatomic open reduction and internal
fixation of the displaced distal radius fracture), the
surgeon essentially assures that the concomitantly
avulsed Ligamentum subcruentum is most closely
approximated to the fovea. Once healed, the DRUJstabilizing function of the deep radioulna component
of the TFC has been restored. Following anatomic
alignment of the bony ulna styloid, the deep TFC
fibers can be maintained in this position by long-arm
cast immobilization for 6 weeks, preventing forearm
rotation, with or without adjunctive percutaneous
0.062 K-wires through the ulna into the radius in
zero-rotation. A more secure and predictable reattachment of the Ligamentum subcruentum can be
performed using bone-anchoring techniques directly
to the foveal sulcus (see below).
Over the past 25 years, we have learned that the

details of TFC anatomy must be respected. An example of the consequences of failing to respect this
type of disruption of anatomy, and potential disturbance of normal DRUJ biomechanics, is seen in
Figures 26 and 27. A 19-year-old woman college
student fell off her bicycle onto her outstretched
dominant hand, incurring this grossly displaced, extra-articular distal radius fracture. The extent of
shortening and dorsal angulation is clear in Figure
26A and B. Also readily seen is the significant displacement of the ulna styloid, avulsed through its
base by the intact superficial radioulna components
of the TFC. The patient was treated elsewhere with
closed reduction of her distal radius fracture and cast
immobilization until radius bony union (Fig. 27). She
presented for a second opinion in my office 9 months
following her initial treatment (Fig. 28), complaining
of disabling pain at the distal end of the ulna, particularly with loaded pronation and supination. Physical
examination revealed subtle hypermobility at the sigmoid notch on stress testing the Ligamentum subcruentum using the provocative maneuvers described
above (relative to the opposite side); full pronation
and supination; and 65 wrist extension/75 wrist
flexion. There was no tenderness on direct palpation
of the ulna styloid fracture nonunion. The piano
key sign was negative.
X-rays of this young woman show a 6-mm, radially displaced malunion of the distal radius, with a
6-mm displacement of her ulna styloid, avulsed
Figure 21. Stress testing the palmar, deep fibers of the Ligamentum subcruentum for pain, mechanical instability, or
both (findings must be compared with the opposite, uninjured side).
1100
Figure 22. An illustration of tightening of the dorsal, deep fibers of the Ligamentum subcruentum as the radius rotates and
translates dorsally off the seat of the ulna in supination. The head of the ulna translates along the sigmoid notch, and herniates
out from under cover of tightening superficial palmar TFC fibers, rendering these (green) fibers ineffective in controlling DRUJ
mechanics.
Figure 23. An illustration of tightening of the palmar, deep fibers of the Ligamentum subcruentum as the radius rotates and
translates palmarly off the seat of the ulna in pronation. The head of the ulna translates along the sigmoid notch, and herniates
out from under cover of tightening superficial dorsal TFC fibers, rendering these (green) fibers ineffective in controlling DRUJ
mechanics.
1101
Figure 24. Improved wrist coils have made wrist MRI more useful in establishing a working diagnosis of a triangular
fibrocartilage injury. (A) In this example, it appears that the superficial TFC components attaching directly to the bony ulna styloid
are still intact (arrow); but contrast material streams across the ulna pole and through the fovea, suggesting that the Ligamentum
subcruentum has been completely avulsed from the fovea. (B) These MRI findings were confirmed at the time of surgery. The
arrow points to the stump of the avulsed Ligamentum subcruentum, deep in the ulna fovea.
through its bony base by the superficial dorsal and

palmar radioulna components of the TFC (Fig. 26).
The styloid is displaced from its base exactly the
same distance as the lateral displacement of the distal
radius fragment at the malunion site. The original
displacement of the metaphyseal radius fracture at
the time of injury pulled the ulna styloid from its base
by force transmission through the intact superficial
dorsal and palmar (green) components of the TFC.
As the bony styloid was avulsed from its base, the
deep foveal attachments of the Ligamentum subcruentum failed as well, destabilizing the DRUJ by
loss of this critical anchor point. Failure by the first
treating surgeon to anatomically reduce the distal
radius not only resulted in a radius malunion and ulna
styloid nonunion (Fig. 26) but also left the avulsed
fibers of the deep Ligamentum subcruentum adjacent
to hyaline cartilage covering the distal ulna pole.
There was no potential for the Ligamentum subcruentum to heal to the fovea because of the magnitude of the initial fracture displacement.
In this patient, failure of the Ligamentum subcruentum to heal properly to its anatomic insertion
onto the ulna fovea resulted in chronic, painful DRUJ
instability under load.
The patients 6-mm radius malunion not only shifted

the ulna styloid from its anatomic base but also left the
Ligamentum subcruentum displaced to a position from
which it could not be reanchored to the fovea without
corrective osteotomy of the radius. A dome osteotomy
of the radius was performed (Fig. 28), allowing a 6-mm
medial shift of the distal radius, along with the superficial radioulna ligaments and articular disc, all still
attached to the ulna styloid. The restored anatomy of the
radius allowed the ulna styloid fibrous union to be taken
down, reduced, and anchored anatomically to its base
using a tension-band wiring technique. With the ulna
styloid anatomically reduced, the deep fibers of the TFC
could be restored to their anatomic insertion into the
fovea. A bone anchor placed directly into the fovea
(with an additional 2-0 Fiberwire added to the 2-0
Ethibond suture already attached) assured an anatomic
insertion of the deep fibers of the TFC. Fixation techniques for the radius dome osteotomy (four 0.062
K-wires in this case) are of secondary importance to
understanding the profound importance of repairing the
deep components of the TFC to the fovea in this patient.
The critical concept was to shift the entire distal forearm unit medially, enabling anatomic restoration of the
Ligamentum subcruentum to the fovea and restoration
1102
Figure 25. Basilar ulna styloid fractures can be displaced in association with displaced distal radius fractures through tension on
the intact superficial dorsal and palmar radioulna ligaments (green). As the radius fracture fragments displace, the intact
superficial radioulna ligaments (green) can either rupture (see Palmer et al classification of traumatic TFC lesions37) or remain
intact, avulsing the entire styloid from its base (as in this example). If the superficial radioulna ligaments (green) remain intact as
the styloid displaces from its base, the deep fibers of the Ligamentum subcruentum (blue) will be avulsed from their insertion at
the ulna fovea, and displace the same distance as the initial displacement of the ulna styloid. (A) X-ray of the displaced radius
and ulna styloid fractures; (B) schematic representation of the tearing and separation of the critical Ligamentum subcruentum
(blue) as its foveal insertion fails.
of normal DRUJ mechanics. Figure 29A & D demonstrates the intra-operative technique of preparing the
fovea for reattachment of the Ligamentum subcruentum, installing the bone anchor (with additional 2-0
Fiberwire attached) and passing sutures through the
TFC prior to pulling the deep fibers securely into the
fovea. As seen in the postoperative x-ray (Fig. 28), two
additional percutaneous 0.062 K-wires were passed
through the distal ulna into the radius prior to tying the
bone anchor sutures as tightly as possible. The K-wires
maintain rigid stability of the DRUJ during the early
post-operative healing process. All eight percutaneous
K-wires seen in Figure 30 were removed 6 weeks after
surgery. Following rehabilitation, the patient regained
full, painless pronosupination, with normal load-bearing capacity at the DRUJ. Her final x-ray is seen in
Figure 29.
Conclusions
The intent of this Clinical Perspective has been to
give the reader fresh insight into the critical function
of each of the components of the triangular fibrocartilage and how these intrinsic DRUJ stabilizers function in guiding the distal radioulna joint through a
physiologic arc of pronosupination. We have come a
long way in 25 years. In the recent past, some of our
colleagues referred to disorders at the distal end of
the ulna as a black box. I believe that considerable
light has been shed on that box over the past 25 years,
mainly through key scientific contributions that have
covered anatomy, biomechanics, and reconstruction
at the DRUJ. The perspective we now have enables
usas hand surgeonsto better understand our patients complaints of painful instability at the distal
end of the forearm. We also now have tools to
1103
Figure 26. Anteroposterior and lateral x-rays of a 19-year-old woman who fell off a bicycle onto her outstretched, dominant
hand. The magnitude of radius shortening, dorsal angulation, and complete loss of inclination is readily seen.
diagnose subtle instabilities that so often result in

functional incapacitation for our patients. We can
now make precise diagnoses by applying our knowledge of the details of TFC anatomy and biomechanics and by using new physical examination techniques. We can envision these injuries so much more
clearly now; we can design precise surgical tech-
niques to repair damaged tissue in ways that restore

anatomic relationships and that allow our patients to
return to full functional capacity. For this new clinical perspective, we are grateful to so many who have
enhanced our understanding.
In this Clinical Perspective, I have particularly
tried to emphasize the importance of the deep fibers
of the TFC, the Ligamentum subcruentum, in guiding
the DRUJ through a full physiologic arc of pronosu-
Figure 27. The patient (see Fig. 26) was treated elsewhere
by closed reduction and cast immobilization for 6 weeks.
Anteroposterior x-rays taken after cast removal reveal lateral displacement of the distal radius fragment, with similar lateral displacement of the ulna styloid through its
base, avulsed by the intact superficial radioulna ligaments
of the TFC.
Figure 28. Nine months following the initial injury, the radius was healed, the displaced ulna styloid fracture nonunion
was ankylosed and not tender, but the distal radioulna joint
was painful and unstable through a full arc of pronosupination. Provocative maneuvers that stress the deep dorsal and
deep palmar fibers of the Ligamentum subcruentum were
positive for pain and instability.
1104
pination. This movement allows the radiocarpal unit

to rotate around the forearm axis of rotation (center
of radial head proximal; ulna fovea distal), with the
seat of the ulna acting as a perpetual fulcrum for load
transmission from the rotating and translating sigmoid notch of the radius to the fixed ulna seat.
The reader should now understand why the works of
af Ekenstam et al21 and Schuind et al26 deemed so
controversial 15 years ago, should now seem so corroborative. We should in particular appreciate the insight of Hagert in 1994,34 providing an explanation of
the controversy and insight into how the different components of the TFC dovetail to provide functional stability at the DRUJ. The superficial radioulna components of the TFC (green) maintain stability of the
hypovascular articular disc for load transmission from
the medial side of the carpus to the pole of the ulna. The
clinical manifestation of chronic, superphysiologic load
on the superficial components of the TFC is progressive
deterioration of the articular disc, the ulna pole, and,
eventually, the lunatotriquetral joint, referred to by
Milch in 19413739 as ulnocarpal abutment syndrome
and classified by Palmer as progressive degenerative
lesions of the TFC.40 Deep peripheral tears of the TFC
have an effect on DRUJ rotational stability that is quite
different from the effects of central degenerative TFC
tears classified by Palmer in 1989.
The deep components of the TFC (palmar and dorsal
Ligamentum subcruentum), with their obtuse and mechanically advantageous angle of attack from the palmar and dorsal medial edges of the distal radius to the
ulna fovea, principally guide the radius around a fixed
ulna through a full functional arc of pronosupination.
That the DRUJ can be destabilized by rupture of the
deep fibers of the TFC alone, with the superficial radioulna ligaments still intact, is now well recognized. In
the past 3 years, we have treated 10 patients at The
Indiana Hand Center who had intraoperative documentation of Ligamentum subcruentum avulsion from the
ulna fovea but normal ulnocarpal arthroscopic examinations, with intact superficial radioulna ligaments and
articular disc. Recent improvements in contrast MRI
have made our presumptive diagnoses more clear, but
the value of preoperative physical diagnostic maneuvers cannot be overstated. Bone anchoring of the
avulsed Ligamentum subcruentum to the ulna fovea is
now a regularly performed surgical procedure in our
practice, in those cases where provocative stress testing
of the Ligamentum subcruentum has been positive. Certainly, the chronicity and magnitude of the patients
symptoms as well as a failure of conservative manage-
Figure 29. Dome osteotomy of the radius malunion allowed

a 6-mm medial shift of the distal radius fragment in this
19-year-old woman. The ulna styloid could then be anatomically reduced at its base and held with tension-band wires.
With the bony anatomy restored, the critical, deep radioulna
ligaments of the Ligamentum subcruentum were now juxtaposed to the fovea (red circle) and could be securely reattached with a bone anchor. To eliminate any forearm rotation for 6 weeks, two 0.062 K-wires were introduced
percutaneously through the distal shaft of the ulna into the
radius.
ment are factors that enhance the indications for surgical intervention.
The reader can also now clearly recognize that
avulsion of the deep fibers of the Ligamentum subcruentum from the ulna fovea, with intact superficial
TFC components, cannot be treated by operative
arthroscopy. The repair of this critical, deep TFC
tissue must be performed using open techniques.
It is my sincere hope that the 25-year overview I
have provided in this Clinical Perspective will help
hand surgeon colleagues everywhere in their efforts
to treat patients with painful, debilitating DRUJ instability. Being able to appreciate the details of
DRUJ biomechanics and the anatomy of the TFC is
the first step in effectively managing these patients
and in surgically reconstructing function anatomy.
Our goal is restoration of full, painless forearm pronosupination under loaded conditions.38-39
Received for publication March 12, 2007; accepted in revised form June
13, 2007.
article.
Corresponding author: William B. Kleinman, MD, The Indiana Hand
Center, 8501 Harcourt Rd., Indianapolis, IN, 46280; e-mail: mgould@
indianahandcenter.com.
1105
Figure 30. A sequence of intraoperative steps used to prepare the ulna fovea for a bone anchor. Once placed deeply into the
fovea, the standard 2-0 suture material manufactured with the anchor can be reinforced by the surgeon by adding additional
suture material to the anchor (C), giving a potential four-strand attachment of the avulsed, deep radioulna fibers of the
Ligamentum subcruentum to bone (D).

0363-5023/07/32A07-0022$32.00/0
doi:10.1016/j.jhsa.2007.06.014
References
1. Kleinman WB, Graham TJ. Distal ulnar injury and dysfunction. In: Peimer CA, ed. Surgery of the hand and upper
extremity. New York: McGraw-Hill, 1996:667709.
2. Kleinman WB, Greenberg JA. Salvage of the failed Darrach
procedure. J Hand Surg 1995;20A:951958.
3. Bowers WH. Surgical procedures for the distal radioulnar
joint. In: Lichtman DM, ed. The wrist and its disorders.
Philadelphia: WB Saunders, 1988:232243.
4. Bowers WH. Distal radioulnar joint arthroplasty: the hemiresection-interpostion technique. J Hand Surg 1985;10A:169178.
5. Dibenedetto MR, Lubbers LM, Coleman CR. Long-term
results of the minimal resection Darrach procedure. J Hand
Surg 1991;16A:445 450.
6. Watson HK, Ryu JY, Burgess RC. Matched distal ulna
resection. J Hand Surg 1986;11A:812 817.
7. Darrach W. Anterior dislocation of the head of the ulna. Ann
Surg 1912;56:802 803.
8. Darrach W. Partial excision of lower shaft of ulna for deformity
following Colles fracture. Ann Surg 1913;57:764 765.
9. Dell PC. Distal radioulnar joint dysfunction. Hand Clin
1987;3:563582.
1106
10. Hartz CR, Beckenbaugh RD. Long-term results of resection

of the distal ulna for post-traumatic conditions. J Trauma
1979;19:219 226.
11. Noble J, Arafa M. Stabilization of distal ulna after excessive
Darrachs procedure. Hand 1983;15:70 72.
12. Watson HK, Brown RE. Ulnar impingement syndrome after
Darrach procedure: treatment by advancement lengthening
osteotomy of the ulna. J Hand Surg 1989;14A:302306.
13. Bieber EJ, Linscheid RL, Dobyns JH, Beckenbaugh RD.
Failed distal ulna resections. J Hand Surg 1988;13A:
193200.
14. Dingman PVC. Resection of the distal end of the ulna
(Darrach operation); an end result study of twenty four
cases. J Bone Joint Surg 1952;34A:893900.
15. Field J, Majkowski RJ, Leslie IJ. Poor results of Darrachs
procedure after wrist injuries. J Bone Joint Surg 1993;75B:
5357.
16. Breen TF, Jupiter JB. Extensor carpi ulnaris and flexor carpi
ulnaris tenodesis of the unstable distal ulna. J Hand Surg
1989;14A:612 617.
17. Goldner JL, Hayes MG. Stabilization of the remaining ulna
using one-half of the extensor carpi ulnaris tendon after
resection of the distal ulna. Orthop Trans 1979;3:330 331.
18. Tsai TM, Stilwell JH. Repair of chronic subluxation of the
distal radioulnar joint (ulnar dorsal) using flexor carpi ulnaris
tendon. J Hand Surg 1984;9B:289 294.
19. Hui FC, Linscheid RL. Ulnotriquetral augmentation
tenodesis: a reconstructive procedure for dorsal subluxation
of the distal radioulnar joint. J Hand Surg 1982;7:230 236.
20. Kleinman WB, Graham TJ. The distal radioulnar joint
capsule: clinical anatomy and role in posttraumatic limitation of forearm rotation. J Hand Surg 1998;23A:588 599.
21. Af Ekenstam F, Hagert CG. Anatomical studies on the
geometry and stability of the distal radio ulnar joint. Scand
J Plast Reconstr Surg 1985;19:1725.
22. Adams BD, Berger RA. An anatomic reconstruction of the
distal radioulnar ligaments for posttraumatic distal radioulnar joint instability. J Hand Surg 2002;27A:243251.
23. Bednar MS, Arnoczky SP, Weiland AJ. The microvasculature of the triangular fibrocartilage complex: its clinical
significance. J Hand Surg 1991;16A:11011105.
24. Thiru-Pathi RG, Ferlic DC, Clayton ML, McClure DC.
Arterial anatomy of the triangular fibrocartilage of the wrist
and its clinical significance. J Hand Surg 1986;11A:258
263.
25. Palmer AK, Werner FW. Biomechanics of the distal radioulnar joint. Clin Orthop Relat Res 1984;187:26 35.
26. Schuind F, An KN, Berglund L, Rey R, Cooney WP, Linscheid
RL, Chao EYS. The distal radioulnar ligaments: a biomechanical study. J Hand Surg 1991;16A:1106 1114.
27. Af Ekenstam FW, Palmer AK, Glisson RR. The load on the
radius and ulna in different positions of the wrist and forearm:
a cadaver study. Acta Orthop Scand 1984;55:363365.
28. Spinner M, Kaplan EB. Extensor carpi ulnaris: its relationship to stability of the distal radio-ulnar joint. Clin Orthop
Relat Res 1970;68:124 128.
29. Ruby LK, Ferenz CC, Dell PC. The pronator quadratus
interposition transfer: an adjunct to resection arthroplasty of
the distal radioulnar joint. J Hand Surg 1996;21A:60 65.
30. Hotchkiss RN, AN KN, Sowa DT, Basta S, Weiland AJ.
An anatomic and mechanical study of the interosseous
membrane of the forearm: pathomechanics of the proximal migration of the radius. J Hand Surg 1989;14A:256
261.
31. Kauer JMG. The articular disc of the hand. Acta Anat
1975;93:590 605.
32. Henle J. Handbuch der bnderlehre des menschen.
Braunschweig: Friedrich Vieweg, 1856.
33. Fick RA. Handbuch der anatomie und mechanik der gelenke
unter bercksichtigung der bewegenden muskeln. Vol 1.
Anatomie der gelenke. Jena: Fischer, 1904.
34. Hagert CG. Distal radius fracture and the distal radioulnar
joint anatomical considerations. Handchir Mikrochir Plast
Chir 1994;26:2226.
35. Viegas SF, Pogue DJ, Patterson RM, Peterson PD. Effects of
radioulnar instability on the radiocarpal joint: a biomechanical study. J Hand Surg 1990;15A:728 732.
36. Pogue DJ, Viegas SF, Patterson RM, Peterson PD, Jenkins
DK, Sweo TD, Hokanson JA. Effects of distal radius fracture malunion on wrist joint mechanics. J Hand Surg 1990;
15A:721727
37. Milch H. Cuff resection of the ulna for malunited Colles
fracture. J Bone Joint Surg 1941;23:311313.
38. Deitch MA, Stern PJ. Ulnocarpal abutment. Hand Clin 1998;
3:251263
39. Friedman SL, Palmer AK. The ulnar impaction syndrome.
Hand Clin 1991; 7:295310
40. Palmer AK. Triangular fibrocartilage complex lesions: a
classification. J Hand Surg 1989;14A:594 606.
TECHNIQUE ARTICLE
Reconstructive Hand Surgery for
Scleroderma Joint Contractures
Ananthila Anandacoomarasamy, MBBS (Hons),
Helen Englert, MBBS, PhD, Nicholas Manolios, MD, PhD,
Stuart Kirkham, MBBS
From the Department of Rheumatology, Royal North Shore Hospital, St. Leonards, North South Wales,
Australia; the Departments of Rheumatology and Orthopaedics, Westmead Hospital, Westmead, New South
Wales, Australia.
Systemic scleroderma can cause significant hand deformity and functional impairment.
Surgery is often avoided due to the perceived risks of wound healing. The most common
surgical procedures have been digital sympathectomy, arthrodesis or arthroplasty of the
proximal interphalangeal (PIP) or both, and metacarpophalangeal (MCP) joints. We describe
herein successful soft tissue hand surgery in 2 patients for treatment of scleroderma claw
deformities without the use of arthrodesis or arthroplasty. At the MCP joint, the tight capsules
were excised, and the collateral ligaments and volar plates were released. At the PIP joints,
the volar plates were released and the tight palmar skin was released, resulting in marked
improvement of joint position. Intensive hand therapy was used to maximize function. In
these 2 patients with claw deformity, we found that tight volar skin was the main contributor
to flexion contracture at the PIP level. In contrast, joint capsule contracture was the main
contributor to hyperextension deformity at the MCP level. (J Hand Surg 2007;32A:
11071112. Copyright 2007 by the American Society for Surgery of the Hand.)
Key words: Hand surgery, scleroderma.
cleroderma, also known as systemic sclerosis

(SS), is a heterogeneous connective tissue disorder characterized by excessive collagen production within the skin and internal organs.1 The 2 clinical
subsetslimited cutaneous and diffuse cutaneousare
distinguished by the extent of skin involvement, autoantibody profile and pattern of organ involvement. In
limited disease, hand involvement is thought to relate to
micro- and macrovascular disease. In diffuse disease,
however, hand involvement also includes joint contractures. Claw deformity is common.
The pathophysiology of claw deformity is unknown. When scleroderma causes hand contractures,
the most typical pattern of clinical deformity is fixed
flexion deformities at the proximal interphalangeal
(PIP) joints, with associated fixed extension, and
even hyperextension deformities at the metacarpophalangeal (MCP) joints. The distal interphalangeal
(DIP) joints may show fixed flexion, while the thumb
may show some adduction contracture with narrowing
of the first web space.2 Contractures can also occur at
the wrist. Foot involvement is less common and tends

to occur later in the disease process.
Typically, the deformities that are observed have
excessive collagen affecting the skin (scleroderma),
fascia, ligaments, tendons, and joint capsules. Radiological joint destruction, however, is not common,
especially in early disease.
As a result of prolonged fixed flexion posturing,
trauma, and associated poor micro-circulation, these
patients also commonly suffer with dorsal PIP joint
ulceration. In the hyperextended MCP joints, fascial
tissue between extensor tendons and the joint capsule
becomes sclerotic, secondarily contributing to extension contracture.3
A study by Herrick et al looked at the functional
ability of patients with SS.4 In 140 patients, using an
11-item functional questionnaire, they found that functional limitations were largely related to contractures
and decreased range of motion, especially in the hand.
Although their functional questionnaire had not been
validated and was weighted heavily to upper extremity
1107
1108
function,5 functional disability seems related, at least in

part, to flexion contractures.
We report the cases of 2 women with diffuse SS
who underwent soft tissue reconstructive surgery to
correct finger contractures, which were significantly
impeding function. Good results were obtained by
releasing the contractures at the PIP and MCP joints.
This method carries several functional advantages
over arthrodesis. The factors contributing to MCP
and PIP joint contractures are discussed.
Case 1
A 24-year-old woman was diagnosed with diffuse
SS. Initial treatment consisted of prednisone, methotrexate and D-penicillamine. Ten months after the
onset of symptoms, the scleroderma had progressed
to involve the limbs, anterior chest wall, and trunk.
She also had early pulmonary disease. She was then
treated with monthly pulse cyclophosphamide, pulse
methylprednisolone, and cyclosporin.
She underwent autologous stem cell transplantation (SCT) 14 months after diagnosis, as she continued to have rapidly progressive diffuse disease with
inflammatory features. Prior to SCT, she had diffuse
scleroderma involving the limbs and most of her
trunk. Her modified Rodnan skin score (a measure of
skin thickening) was 27/51. Following the procedure,
the woman began to experience reversal of the skin
thickening and reduction in polyarthralgia after 2
months. Follow-up evaluation at 15 months revealed
marked clinical improvement in her musculoskeletal
disease status as measured by skin scores, health
assessment questionnaire, C-reactive protein, and visual analogue scales. Her skin softened dramatically,
and all joint contractures resolved, with the exception
of the MCPs and PIPs. In fact, these contractures
were progressive, and she had great difficulty in
performing activities of daily living. Pain was not a
prominent feature.
Preoperative evaluation revealed varying degrees
of finger deformity (Fig. 1a). The worst affected
joints (fifth PIPs, bilaterally) had fixed flexion deformity of 85 to 90. She had hyperextension of the
MCPs and narrowing of the first web spaces, bilaterally. Sensory examination was normal. Allens test
was slightly abnormal with delay in capillary refill
time (4 5 seconds). Doppler ultrasound of the digital
vessels (after SCT) showed good blood flow to both
hands. Plain radiographs of both hands showed no
evidence of erosive change or loss of joint space.
She proceeded to reconstructive surgery of the
right hand under general anesthesia.
Figure 1. (A) Prereconstructive hand surgery. (B) Postoperative image demonstrating flexed MCP joints and extended
PIP joints.
For exposure of the MCP joints, a transverse incision was made over all 4 MCPs on the dorsal aspect.
The sagittal band of the extensor hood was incised
for each digit. The extensor tendon was then temporarily sublaxated from the midline of the metacarpal
head towards the ulna. Then, the thickened dorsal
capsule was excised completely and sent for histological assessment. Following this, the Freer elevator
(Aesculap Inc., Center Valley, PA) was used to, first,
tease off the collateral ligaments from the metacarpal
head and the base of the proximal phalanx. Then, the
Anandacoomarasamy et al / Hand Surgery for Scleroderma
volar plate was released from the distal end of the

metacarpal head and neck. Intraoperatively, the following passive ranges of motion were obtained: index finger 0 to 100, middle finger 0 to 110, ring
finger 0 to 110 and little finger 0 to 110.
The PIPs were then each approached with 4 separate midlateral incisions. The interval between the
accessory collateral ligament and the volar plates was
incised. The volar plate was then removed as much
as possible, though in some digits (especially the ring
finger) it was only possible to incise the volar plate.
Palmar skin was released at the skin crease for digits
3, 4, and 5. Intraoperatively, the following passive
ranges of motion were obtained at the PIPs: index
finger 0 to 125, middle finger 0 to 125, ring finger
0 to 125, and little finger 0 to 125.
To keep the fingers splinted in the intrinsic-plus
position, intra-osseous K-wires were passed through
each of the metacarpal heads to keep the MCPs
flexed and the PIPs extended. Extending and straightening the previously pathologically fixed and flexed
PIPs resulted in a defect of 12 cm on each digit at
the PIP joint, requiring skin grafts. A full thickness
skin graft measuring 18 1.5 cm was taken from the
volar distal forearm. The skin was used to graft the
dorsal aspect of the metacarpal heads and the volar
aspect of the PIP joint skin creases (Fig. 1b). The
donor sites were not sclerodermatous.
The pathological findings at surgery were as follows:
At the MCP joint level, the dorsal skin was tight.
There were adhesions between the extensor apparatus and the capsule. To a lesser extent, there were
some tight subcutaneous fibrous bands superficial to
the veins in line with the ring and little fingers only.
The dorsal capsule was very thick. The capsules were
more responsible for lack of MCP joint flexion than
the skin. Once the skin, capsule, and adhesions were
addressed, the joint underwent a large increase in
passive flexion. This corrected posture left a skin
defect measuring approximately 18 mm in maximal
width. At the PIP joint level, volar relieving skin
incisions over each PIP joint allowed the joints to
gain considerable passive extension. The volar plate
and joint capsule, once released, also contributed to
overall joint stiffness to a lesser extent than the skin.
These joints did not become free until these structures were all released. The tight volar skin was the
main contributing factor to lack of extension.
Histopathology showed marked fibrosis of the volar plate of tissue (PIP joint) with hyalinized bundles
of collagen fibers (Fig. 2). The dorsal capsule at the
1109
Figure 2. Histology of the volar plate of the PIP joint obtained at surgery demonstrating marked fibrosis with hyalinized bundles of collagen.
MCP joint had dense avascular fibrous tissue with

prominent hyalinization of the collagen. In our volar
plate biopsy specimen, vessels were found that
showed hyperplasia of the tunica media. The smaller
vessels showed a predominant lymphocytic infiltrate
around and within the vessel wall.
Postoperatively, there were no complications, specifically no ischemic or wound healing problems.
Her skin grafts took completely in a normal time
frame. Three weeks after surgery, the 4 K-wires were
removed. She underwent regular hand therapy sessions with static and dynamic splints. Four months
after surgery, she had maintained her hand in the new
intrinsic-plus posture. Also, she achieved noticeable
improvements in hand function, with return of key
grip and ability to wash her hair. She had not, however, regained power grip. Overall the patient was
satisfied that she had achieved a modest gain.
Eighteen months following reconstructive hand
surgery, the patient is able to use her operated hand
for functional tasks, which she cannot perform with
the non-surgically treated claw hand. The operated
hand allows her to open jars and drive a car holding
the steering wheel, neither of which she can do with
her un-operated hand. She is unable, however, to
perform certain tasks with her operated hand, such as
reaching into her small handbag. Overall, postoperative hand function has improved. The patient would
recommend the procedure to her friends and would
agree to have it again under the same circumstances.
Her overall satisfaction was relatively high. The current range of motion is only modest. There remains
marked bilateral hand stiffness. On the operated side,
1110
her fingers are stiff in a position of much greater PIP

joint extension.
Case 2
A 52-year-old woman was diagnosed with scleroderma. She had diffuse disease with no major organ
involvement. Her medications were methotrexate, cyclosporine, and D-penicillamine. Her function was severely impaired by severe contractures of both hands.
Preoperatively, the MCP joints of both hands were
fully extended while all the PIP joints were fully
flexed. The DIP joints were flexed to a lesser extent.
The dorsal skin over the MCP joints was very tight.
Sensory examination was normal. Angiography of
the digital vessels showed good blood flow. Plain
radiographs of both hands were normal.
Reconstructive surgery of the left hand was performed under general anesthesia. The four MCP
joints were approached through a single, dorsal,
transverse skin incision. The MCP joint capsules
were thickened, and these were excised dorsally.
There were minor adhesions between the skin and the
tendons. There was definite shortening of the skin
requiring full-thickness skin grafts once the defect
was corrected. An elliptical full-thickness graft was
sutured into place over the dorsal aspect of the MCP
joint.
The 4 PIP joints were then addressed. A transverse, relieving skin incision was made at the flexion
skin crease at the PIP joint level. The incision was
then extended by making an anchor-shaped incision
with proximal and distal extensions centered in the
midlateral aspect of the PIP joint. This was done on
the radial border for the index, middle, and ring
fingers. The ulnar border was used at the little finger
due to difficult access. The neurovascular bundle was
identified and protected throughout the procedure. A
Beaver 6400 scalpel blade (Robbins Instruments Inc.,
Chatham, NJ) was used to incise the volar plate and
pathological ligaments. The PIP joints were then
manipulated from their abnormally flexed position
into a fully extended position. This was performed
while observing the tension on the neurovascular
bundles. Once all 4 fingers were corrected, K-wires
were placed across the metacarpal heads and into the
proximal and middle phalanges, creating an intrinsic
plus position for the entire hand.
With the fingers in the extended position, the tourniquet was released, and perfusion in the 4 digits was
observed. The index, middle and ring fingers perfused well, but the little finger did not. Therefore, the
K-wire was pulled back to allow the finger to flex.
This permitted immediate reperfusion. Following

this, the K-wire was repositioned into the little finger
but only across the MCP joint (not across the PIP
joint to avoid excess tension across the artery). Fullthickness skin grafts obtained from the volar proximal forearm were sutured into place on the volar
aspect of the four PIP joints. Each graft measured
approximately 23 cm in longitudinal length. The
total procedure took 4 hours. The intraoperative
pathological findings were similar to those seen in
the first case.
Postoperatively, there were no complications. The
skin grafts took completely. Sutures were removed 2
weeks after surgery and K-wires 4 weeks thereafter.
Subsequently, she underwent intensive hand therapy
to maintain hand position and improve strength.
Since removal of K-wires, the fingers have remained
in the corrected positions (MCPs flexed and PIPs
extended).
At the 12-month follow-up visit, her overall hand
function showed only small gains. She could hold
onto a rowing machine at the gym with the operated
hand. Marked bilateral hand stiffness remains, with
the operated side showing a much less clawed posture and also a mildly improved ability to perform
mono- and bimanual tasks.
Discussion
Joint contractures are a major cause of morbidity and
disability in scleroderma. Despite this, reconstructive
hand surgery in patients with scleroderma has been
reported infrequently, probably reflecting the fear
that wound healing is compromised in these patients
and that surgical intervention might cause further
deterioration in the vascularity of the hands and
digits.2 A few series have been reported. Most authors advocate and report the outcomes of joint arthrodesis or arthroplasty at the PIP joint and MCP
joint capsulotomy or arthroplasty (with or without
metacarpal head resection).2,6 10 Adduction contracture of the thumb requires an opening of the first web
space to release the adductor attachment. Resection
of the trapezium has been used successfully to restore
thumb metacarpal abduction.3 With disease progression, wound healing and the possibility of tissue loss
is a genuine concern. The skin over the dorsal aspect
of these joints is stretched tightly, and the dermal
capillary bed can be reduced by as much as 80%.6
Implant prostheses carry the additional risk of implant failure as well as a higher rate of wound healing
complications.3
The earliest small series was reported by Lips-
Anandacoomarasamy et al / Hand Surgery for Scleroderma
comb et al in 6 patients who underwent surgery from

1962. The patients reported an improvement in overall function. The range of motion obtained at the
MCP joints, however, was not greatly improved by
capsulotomy and arthroplasty.7 Swanson flexible implant arthroplasties have been used at the PIP joints,
but the ultimate gain in motion reported is small.7
Jones et al performed 53 PIP joint fusions in 12
patients.2 Arthrodesis was accomplished in a position
of 45 to 55 using a combination of K-wires, intraosseous wires, and bony shortening. This improved the position of the fingers for both thumb
index finger pinch and grasp. They also performed 19
MCP capsulotomies and 4 metacarpal head resections in conjunction with PIP joint arthrodesis in 4
patients. The resultant motion at the MCP joints,
however, was reported to be disappointing, averaging
less than 20.2 Gilbart et al reported 13 PIP joint
fusions and 5 MCP joint excisional arthroplasties.6
Nalebuff reports that in selected cases, capsulotomy
will increase the range of motion in the MCP joints.9
The largest series reported thus far by Melone et al
included 70 scleroderma patients requiring 272 hand
operations.10 They performed 211 interphalangeal
joint arthrodeses after complete capsulotomy and jdicious skeletal shortening. Radiographic union of arthrodesis was achieved within 8 weeks of surgery.
Twenty-eight implant arthroplasties were performed
after wide resection of the metacarpal heads, condyles, and collateral ligaments. The authors report a
50 average arc of motion for these arthroplasties
after follow-up evaluations ranging from 1 to 15
years.
In most reports to date, severe fixed finger-in-palm
deformities have been treated with a combination of
MCP joint excisional arthroplasty and PIP joint fusion, involving resection of bone from both the distal
metacarpal and proximal phalanges.11 While this
procedure improves finger position and function,
there is resultant axial shortening of the hand with a
loss of the normal MCP joint contour.11
In our patients, a satisfactory functional outcome
and aesthetic appearance was gained by positioning
the MCP joints in flexed position with the PIP joints
in extension. By carefully excising the capsule, and
separating the collateral ligaments and volar plates,
replacement arthroplasty at the MCP joints was obviated. Likewise, at the PIP joints, careful incision
and removal of the volar plates and release of the
palmar skin at the skin crease resulted in marked
improvement in passive range of motion. Bone resection and joint fusion was avoided by releasing the
1111
tissues responsible for joint contractures. We believe

that soft tissue releases offer obvious advantages
compared with bony procedures such as arthrodesis
and arthroplasty, especially in light of the fact that so
many scleroderma patients have normal joint radiographs.
There were 3 important findings from this small
case series of 2 patients with relatively early diffuse
scleroderma. First, we believe that reconstructive
surgery for one hand, leaving the other hand with
flexed PIP joints would appear to confer the greatest
functional advantage. Second, we were able to elucidate and rank in order of importance the factors
contributing to sclerodermatous claw hand at the
MCP and PIP joints. At the MCP joint level, capsular
tightening and adherence between the latter and overlying tendons was the prime limiting factor. At the
PIP level, however, the predominant joint deformity
occurred because of overlying skin tightness rather
than joint pathology per se. Hence therapeutic intervention to prevent or ameliorate such features in
other patients will need to focus on both aggressive
control of sclerodactyly (PIP), and joint and periarticular tissue (MCP). Finally we were able to successfully perform surgery in patients with early aggressive diffuse disease without complications
relating to vascular compromise, either in the hands
or at the area of donor skin.
This type of surgery, however, is not without considerable risks. These patients should be advised,
preoperatively, of the risks of arterial intimal tears
and digital ischemia (including the possibility of
losing one or more digits) and that the 3 4 months of
rehabilitation is only likely to achieve a small-tomoderate gain of their overall hand function. Cosmetically the improved posture needs to be weighed
against the appearance of the skin grafts.
Prolonged flexion may also cause attenuation of
the central slip of the extensor digitorum communis.
Unlike the normal digit, however, this attenuation
does not necessarily lead to any boutonniere deformity because of the very tight skin and joints that
prevent the passive imbalance of a boutonniere posture.
Our cases also suggest that scleroderma patients
should not be denied the benefits of reconstructive
hand surgery when they have significant functional
limitation from joint deformities. In patients with
advanced hand contractures, a nominal measured improvement in position and function may lead to a
substantial improvement in the patients adaptive
ability to perform certain activities of daily living.11
1112
Hand surgery can improve function but needs to be

considered together with the specific surgical risks
and a realistic discussion of how long the recovery
takes. Hand surgery is an adjunct to medical management and should be considered in the overall
management of scleroderma patients with disabling
contractures. Further studies may allow us to determine the optimal timing of both medical and surgical
treatment in the disease process.
Received for publication December 29, 2006; accepted in revised form
June 13, 2007.
article.
Corresponding author: Stuart Kirkham, MBBS, Department of Orthopaedics, Westmead Hospital, Westmead, NSW 2145, Australia; e-mail:
vanstu@yahoo.com.
0363-5023/07/32A07-0023$32.00/0
doi:10.1016/j.jhsa.2007.06.011
References
1. Lin ATH, Clements PJ, Furst DE. Update on disease-modifying antirheumatic drugs in the treatment of systemic sclerosis. Rheum Dis Clin N Am 2003;29:409 426.
2. Jones NF, Imbriglia JE, Steen VD, Medsger TA. Surgery for
scleroderma of the hand. J Hand Surg 1987;12A:391 400.
3. Jakubietz MG, Jakubietz RG, Gruenert JG. Scleroderma of
the hand. J Am Soc Surg Hand 2005;5:42 47.
4. Herrick A, Rooney B, Finn J, Silman A. Lack of relationship
between functional ability and skin score in patients with
systemic sclerosis. J Rheumatol 2001;28:292295.
5. Pope JE. Musculoskeletal involvement in scleroderma.
Rheum Dis Clin N Am 2003;29:391 408.
6. Gilbart MK, Jolles BM, Lee P, Bogoch ER. Surgery of the
hand in severe systemic sclerosis. J Hand Surg 2004;29B:
599 603.
7. Lipscomb PR, Simons GW, Winkelmann RK. Surgery for
sclerodactylia of the hand. J Bone Joint Surg 1969A;51:112
117.
8. Norris RW, Brown HG. The proximal interphalangeal joint
in systemic sclerosis and its surgical management. Br J Plast
Surg 1985;38:526 531.
9. Nalebuff EA. Surgery in patients with systemic sclerosis of
the hand. Clin Orthop 1999;366:9197.
10. Melone CP, McLoughlin JC, Beldner S. Surgical management of the hand in scleroderma. Curr Opin Rheumatol
1999;11:514 520.
11. Bogoch ER, Cross DK. Surgery of the hand in patients with
systemic sclerosis: outcomes and considerations. J Rheumatol 2005;32:642 648.
INSTRUCTIONAL COURSE LECTURE

Arthrodesis of the Interphalangeal Joints
With Headless Compression Screws
Stephen J. Leibovic, MD
From the Virginia Hand Center, Medical College of Virginia, Virginia Commonwealth University, Richmond, VA.
Interphalangeal arthrodesis is a reliable method of pain relief for arthritic proximal and distal
interphalangeal joints in the fingers. Indications include osteoarthritis, acute trauma, chronic
reconstruction for trauma, rheumatoid and other inflammatory arthritides, and at the distal
interphalangeal joint, chronic mallet finger deformity and unreconstructible flexor tendon defects. Solid arthrodesis imparts stability to the digital skeleton. Headless compression screws can
be reproducibly inserted and are a good method to provide fixation adequate to accomplish
interphalangeal arthrodesis. Surgical technique involves a dorsal incision and preparing the
skeleton for good bony apposition. Exact technique for screw insertion depends on the specific
screw used. Union rates range from 85% to 100% in published studies, with time to union of 7
to 10 weeks. (J Hand Surg 2007;32A:11131119. Copyright 2007 by the American Society for
Surgery of the Hand.)
Key words: Arthrodesis, screw, proximal interphalangeal joint, distal interphalangeal joint,
osteoarthritis.
rthrodesis of the interphalangeal joints of the

fingers is a successful and time-tested treatment
for a variety of painful or disabling conditions.
At the proximal interphalangeal joint, indications include arthritides of all kinds (osteo-, psoriatic, rheumatoid, post-traumatic). It is also vital in reconstruction
from trauma, both acute and chronic, and from unstable
joints where soft tissue stabilizing procedures are contraindicated or have failed. It is sometimes used in
severe contractures from Dupuytrens disease or other
causes where release is unlikely to restore useful motion. At the distal interphalangeal (DIP) joint, similar
indications exist, and in addition, arthrodesis is used for
treatment of chronic mallet finger deformity and flexor
digitorum profundus avulsion or laceration when soft
tissue repairs are impossible or unlikely to succeed.
In chronic painful cases, nonsurgical measures
should normally be exhausted before surgery is performed. These include anti-inflammatory medication,
intra-articular steroid injections, and splinting. In acute
traumatic situations, treatment need is more immediate.
Familiarity with a variety of methods of surgical arthrodesis equips the surgeon with the means to treat
acute as well as chronic cases with the techniques
and hardware best suited for the particular case.
Motion-preserving procedures for the interphalan-
geal joints exist, but they have limitations. Biologic

resurfacing techniques include volar plate arthroplasty,
chondral resurfacing, and vascularized free joint transfer. Each has met with limited success; the volar plate
arthroplasty has certainly been the most widely used.
Volar plate arthroplasty is suited for some situations
with destruction limited to the volar portion of the
middle phalangeal base, particularly after dorsal proximal interphalangeal (PIP) joint dislocations.1 Chondral
arthroplasty2 and free joint transfer,3 however, both
have significant donor morbidity for limited functional
gains.
Synthetic silicone interposition arthroplasty has enjoyed popularity at the metacarpophalangeal joints as
well as the PIP joints and, to a limited extent, the DIP
joints.4 More recently, total joint arthroplasty with anatomic replacement of both articulating surfaces has
seen increasing popularity with the advent of metal on
plastic or pyrocarbon implants. Enthusiasm for motionpreserving procedures at the interphalangeal joints is
tempered, however, due to problems with stability, especially in the index finger during pinching activities,
and implant failure. In such cases, or to avoid these
problems, arthrodesis remains a gold standard. Moberg5
eloquently expressed in 1960 that the prime requisite
of a good digital arthrodesis is a painless and stable
1113
1114
union in proper position occurring in a reasonable space

of time. Proximal and distal interphalangeal arthrodesis remains today a fundamental and frequently used
technique for relief of pain and disability in the fingers.
Methods for achieving interphalangeal arthrodesis
have evolved. It is generally accepted that techniques
that provide structural rigidity and compression are
likely to have the highest success rate. K-wires have
been used for years, although they suffer from lack of
structural rigidity. The tenon method6 and the bone peg
technique7 are time-consuming methods of bone carpentry that impart considerable rigidity to an arthrodesis
when used alone or with just K-wires. More recently,
tension band wiring and headless compression screw
fixation of interphalangeal joint fusions have demonstrated consistent success with relatively low complication rates. At the present time, they represent the product of 3 decades of investigations to increase the rigidity
of interphalangeal arthrodesis while simplifying surgical techniques.
Technique
Headless compression screws used for interphalangeal
arthrodesis are either screws with leading and trailing
threaded portions with different thread pitch separated
by a smooth shaft, such as the TwinFix (Stryker-Leibinger, Kalamazoo, MI), Herbert or Herbert/Whipple
(Zimmer, Warsaw, IN), HBS (Orthosurgical, Miami,
FL), or Millennium (Millennium Medical, Santa Fe,
NM), or with a fully threaded, continuously variable
tapered thread, such as the Acutrak (Acumed, Hillsboro, OR). Surgical techniques are similar for all.
Surgical Approach
The surgical approach to the PIP joint is through a
dorsal, curvilinear incision around the joint (Fig. 1). The
extensor tendon is encountered directly under the skin,
covered with thin filmy epitenon. There is no macroscopically defined separation between the extensor tendon and the joint capsule at the PIP level. The combined
structure of tendon and capsule is divided in the midline
to enter the joint. Care should be taken to ensure that the
central slip is not detached from the dorsal base of P2,
but that it is rather subperiosteally dissected so that it
will re-adhere to this region without proximal retraction
upon closure. The lateral bands are left undisturbed and
should continue to function to extend the DIP joint.
To fully expose the joint surfaces for appropriate
bone carpentry, both collateral ligaments are sectioned.
This can be simply done from inside the joint, either
peeling them off the condyles of P1 or by direct section.
In either case, any redundant tissue after sectioning
Figure 1. Curved incision on the dorsal side of the joint

avoids the midline. This minimizes the potential for adhesions that might impede tendon function (Leibovic16).
should be excised so that it does not interfere with bony

apposition. Care must be taken to protect the neurovascular bundles on both sides of the finger. Once the
collateral ligaments are excised, the joint can be hyperflexed for further exposure. The volar plate will remain
and will prevent complete separation of P1 from P2. As
both P1 and P2 are slightly shortened by subsequent
bone carpentry, the volar plate may act as a spacer
preventing apposition of P1 to P2. It should either be
removed, or a central portion of it excised, at this stage
with joint hyperflexion.
Approach to the DIP joint is through a Y- or Hshaped incision over the joint. Care must be taken to
avoid damage to the germinal matrix, close by the
surgical field. The extensor tendon is sharply divided
directly over the joint, and the collateral ligaments are
excised as in the PIP joint. The volar plate is excised.
Bone Preparation
Fixation techniques currently available impart considerable stiffness and strength to the arthrodesis immediately, unlike in years past when fixation consisted of all
bone carpentry with no metal, or K-wires only. Therefore, elaborate bone carpentry with tenons or bone pegs
is unnecessary. Typically, straight cuts are made with
Stephen J. Leibovic / Arthrodesis of the Interphalangeal Joints
Figure 2. Straight saw cuts at the condyles of P1 and the base

of P2 can be fashioned to produce the appropriate amount of
joint flexion after fusion. If the angles of each cut with respect
to the perpendicular axis of each bone is represented by
and , the final angle of fusion will equal the sum of the
angles and , as shown here. The technique is unforgiving,
however, as the final angle of flexion as well as angulation in
the coronal plane are completely determined by the saw cuts
and they cannot be modified without recutting the bone
(Leibovic16).
an oscillating saw on both phalanges, or rongeurs are

used to fashion cup and cone or convex-concave opposing surfaces for arthrodesis.
Preoperative planning must include a decision on the
appropriate angle for arthrodesis. Proximal interphalangeal joints of the index and middle fingers are typically
more functional when fused at 15 to 30 of flexion. As
they are used more in fine pinch and manipulation, a
more extended position improves dexterity. The ring
and small fingers are more functional when fused at 30
to 45, as they are more involved in grip. As one moves
from the radial to the ulnar side of the hand, optimal
fusion angle increases. Individual patient function,
however, must be considered; a guitar player, in the
fretting hand, may want the joints more flexed than a
data entry operator. Encouraging the patient to wear
custom fabricated orthoplast splints at different angles
prior to operative planning will help them be part of the
decision making process. At the DIP joint, 0 to 50 of
flexion is generally most useful.
When the desired fusion angle is determined, implementation must be planned depending upon the carpentry method chosen. If saw cuts are to be used, both cuts
should be angled. The sum of the angles on each bone
equals the total amount of joint flexion obtained (Fig.
2). This produces the largest surface area of bone ap-
1115
position for the arthrodesis. Coronal plane alignment,

however, is completely determined by the saw cuts; no
further adjustment can be made without recutting the
bone, which further shortens the digital skeleton.
An alternative is the use of rongeurs to fashion cup
and cone or convex-concave opposing surfaces on the
two bone ends (Fig. 3). Final positioning can be determined after machining of the bone surfaces. Small
changes in the angle of flexion as well as rotation can be
adjusted without having to recut or recontour the bone
ends. Bone mills are available that can produce accurate
cup and cone configurations on the opposing surfaces,
but their use is cumbersome in the fingers; I prefer
careful use of a rongeur.
The digit is necessarily shortened somewhat by removal of cartilage and subchondral bone from opposing
joint surfaces. Especially in osteoarthritis with preoperative angulation in the coronal plane, the shortening can
be considerable. If a sufficient amount of the volar plate
has not been removed, it may prevent apposition of the
opposing surfaces. A frequent complaint in osteoarthritis is enlargement of the proximal or distal interphalangeal joints by osteophytes, which may be deemed unsightly or prevent the patient from wearing rings. These
should be removed at this time.
Screw Insertion
Successful arthrodesis with a headless compression
screw requires attention to detail in planning and
execution. At the PIP joint, the screw is inserted
antegrade from the dorsal surface of the proximal
phalanx. It is critical that the dorsal entry site is at
least 6 or 7 mm proximal to the joint surface to
prevent fragmentation of the dorsal cortex upon drilling. A K-wire is used to start a hole obliquely
through the dorsal proximal phalangeal cortex. The
angle of the hole in the sagittal plane is equal to the
desired angle of arthrodesis. The K-wire is followed
by the drill sequence required for the particular screw
used. In the case of the Herbert and mini-Herbert
Figure 3. Rounded cup and cone configurations can be easily fashioned at the proximal interphalangeal joint with a
rongeur. Then, the final angle of fusion can be adjusted
during insertion of the fixation device, and any rotational
malalignment can be easily corrected (Leibovic16).
1116
screws, the long, skinny drill is used first to drill

through the proximal phalanx into the medullary
canal of the middle phalanx (Fig. 4). This is followed
by the short, fat drill for the trailing threads in the
proximal phalanx only. The only two-piece headless
compression screw, the TwinFix, which has the theoretical advantage of providing more compression
across the arthrodesis site, and the Acutrak screw
have a single cannulated drill that will follow the
K-wire through the proximal phalanx into the middle
phalanx. The drill must pass the cortical isthmus of
the middle phalanx.
A small 1.1 or 1.5 mm rongeur is used to enlarge the
dorsal hole in the proximal phalanx to prevent fracture
and fragmentation upon screw insertion. With a rongeur, small bites can be taken around the hole in the
dorsal cortex. Cortical purchase of the trailing thread of
the screw is not necessary for adequate fixation. If the
screw being used requires a tap, it is used at this point.
Screw length is then estimated by eye after experience
or using fluoroscopy and a K-wire. Measurement need
not be precise, as the leading end of the screw may rest
anywhere at or beyond the cortical isthmus of the middle phalanx. Usually, screws between 16 and 22 mm
are appropriate for PIP joint arthrodesis. The assistant
then aligns the joint and maintains firm compression

while the screw is advanced until it is flush with or
beneath the dorsal cortex. Final rotational alignment
can be adjusted after screw insertion (Fig. 5).
Insertion of a headless compression screw into the
DIP joint is done retrograde. A preliminary K-wire is
necessary only if the hardware being used has a cannulated drill. It is drilled first antegrade from the joint
surface distal. The drill will find its way down the canal
of the distal phalanx and exit through the tuft. As it tents
the skin at the pulp, a small incision is made large
enough for exit of the drill. The drill is then withdrawn
and reinserted retrograde through the pulp incision, into
the tuft hole and into the joint. Under direct vision, it is
then directed into the middle of the denuded and decorticated head of the middle phalanx. Up to 10 of
flexion at the arthrodesis site is usually possible if at this
time the drill is directed from slightly volar to the center
of the head of the middle phalanx dorsally toward the
dorsal cortex until it reaches just beyond the middle
phalangeal isthmus, seen on fluoroscopy (Fig. 6). If the
screw requires a larger diameter trailing thread hole, it
is made in the distal phalanx only. Then, while the
assistant reduces and compresses the joint, the screw is
inserted (Fig. 7).
Technical Pitfalls
Figure 4. The long, skinny drill from the Herbert screw set
is inserted first. The hole must be made sufficiently far back
from the joint to avoid fragmentation of the dorsal cortex.
Usually 6 7 mm is enough. The short, fat drill is then used
to enlarge the trailing hole in the dorsal cortex of P1. Finally,
a rongeur is used to enlarge the hole sufficiently that the
trailing head of the Herbert screw can easily pass deep to the
dorsal cortex (Leibovic16).
The major complication of PIP joint arthrodesis with

these screws is fragmentation of the dorsal cortex of the
proximal phalanx. If this occurs, all purchase is lost and
screw fixation must be abandoned or supplemented.
This is best avoided by enlarging the dorsal cortical
hole with a rongeur until it is almost as large as the
thread diameter of the trailing threads.
At the DIP joint, a significant complication can occur
if the thread diameter exceeds the anteroposterior diameter of the distal phalanx, which it may. The average
anteroposterior diameter of the distal phalanx is 3.55
mm,8 while the trailing thread diameter of the Herbert
screw is 3.9 mm and the TwinFix screw, 4.0 mm. In
their cadaver study, Wyrsch et al8 found penetration of
the dorsal distal phalangeal cortex by screw threads in
83% of their specimens and in all female specimens
during joint instrumentation with a Herbert screw. This
can lead to stretching or disruption of the nail bed with
deformity of the nail.
Occasionally, the diameter of the metaphysis of
the middle phalanx is so large that the screw will not
gain purchase. This occurs most often in the thumb.
If preoperative radiographs suggest this, alternate
means of fixation should be used.
1117
Figure 5. (A) Posteroanterior and (B) lateral radiographs of a PIP joint fusion using a Herbert screw. Note the enlarged hole in
the dorsal cortex of P1 to prevent cortical fragmentation.
Postoperative Care
A well-padded dressing on the digit or a small bulky
dressing on the hand incorporating a splint will protect
the digit from stress and allow for postoperative swelling. After 7 to 10 days, the dressing and sutures can be
removed, and a thermoplastic splint can be fitted to just
the involved joint. A scar pad may be included in the
splint, which is applied with Velcro straps. Early mobilization of the adjacent joints prevents stiffness. The
screw imparts substantial stability to the joint, allowing
minimal external protection. The splint can be discontinued after 3 weeks during light activity, although for
any stressful activities, it may be left in place. Six weeks
is usually sufficient for clinical union, judged by a
painless joint on stress, although radiographic union,
judged by presence of trabeculae crossing the arthrodesis site, may take 3 months or more. If the arthrodesed
joint is pain free, lack of radiographic union should not
impede functional use of the hand. Resistive exercises
should be begun, however, only after radiographic
union is evident.
Nonunion is rare, but should be declared if there
remains pain and instability 6 months after surgery.
Delayed unions may be splinted for longer than
usual, though splinting for more than 3 months is
generally ill advised.
Figure 6. For DIP joint arthrodesis, the drill is passed antegrade from the joint surface distally through the end of the
tuft and the pulp of the finger. It is then withdrawn and
reinserted retrograde, from where it enters the remains of the
head of P2. This can sometimes be done in approximately
10 of flexion at the DIP joint.
1118
Figure 7. Radiographs of a DIP joint fusion using an Acutrak tapered screw. An additional K-wire has been inserted for added
stability. The large diameter (3.5 mm) of the trailing threads may break through the cortex of the distal phalanx, which will result
in loss of purchase by the screw in the bone. Sometimes this can be salvaged with the addition of a K-wire.
Results
Interphalangeal joint arthrodesis using headless compression screws has enjoyed high success rates. At the
PIP joint, reported nonunion rates vary between 0% and
2% (Table 1). Leibovic and Strickland,9 in a study
comparing PIP fixation with K-wires, tension band,
Herbert screw, and plate found nonunion rates of 21%,
4.5%, 0%, and 50%, respectively. There was a clear
decrease in nonunion rate with the Herbert screw compared to other means of fixation. Stern and Fulton,10
however, in a study of complications of DIP joint
arthrodesis found nonunion rates unchanged between
different surgical techniques of K-wire, interfragmentary loop wire plus K-wire, and Herbert screw. In their
study, the nonunion rate was 11% to 12%, independent
of fixation used. Both studies agreed that patients with
psoriatic arthritis had the highest nonunion rates at the
two joints, although the rates of nonunion with other
diagnoses differed (Table 2).
Complications are not common. Brutus et al11 had 4
nonunions, 2 from infection in the DIP joint. There
were 4 postoperative infections, easily treated. There
was a 7% incidence of nail deformity, but no significant
complaints of fingertip tenderness. Stern and Fulton10

had no osteomyelitis in DIP arthrodeses with Herbert
screws, although there was a 2% incidence of osteomyelitis across all fixation methods. There was, however,
deep wound infection in 4% of fusions with Herbert
screws. There was an incidence of dorsal skin necrosis
in 15% of joints fused with a Herbert screw. Leibovic
Table 1. Nonunion Rate and Time to Union in

Interphalangeal Arthrodesis With Headless
Compression Screws
Joint
Nonunion
Rate (%)
Time to
Union
(weeks)
PIP
PIP and DIP
2
0
8.1
PIP
DIP
DIP
DIP
11
5
15
10
8
810
Author
13
Ayres et al
Katzman et al14
Leibovic and
Strickland9
Stern and
Fulton10
Gomez et al15
Brutus et al11
Table 2. Radiographic Nonunion Rate by

Diagnosis in Interphalangeal Arthrodesis
(Various Fixation Methods)
Diagnosis
Osteoarthritis
Post trauma
reconstruction
chronic
Acute trauma
Rheumatoid
arthritis
Psoriatic arthritis
Stern and Fulton10

(DIP Joint)
Nonunion Rate (%)
Leibovic and
Strickland9
(PIP Joint)
Nonunion (%)
13
7
22
9
12
13
22
15
29
and Strickland9 had 1 case of superficial infection after

screw fixation, or 3%, and no osteomyelitis.
Established nonunion, while rare, should be treated
with reoperation using cancellous bone graft. Good
apposition of vascular cancellous surfaces should be the
goal of treatment, and fixation should be as secure as
possible. Malunion can occur, but it is best avoided by
attention to detail. Angular alignment in all planes and
rotational alignment must be assessed intraoperatively
and corrected when needed.
Dorsal skin necrosis is more likely to occur over the
DIP joint than the PIP joint. Blood supply is less dense
and the skin is tighter. Tight skin closure and overly
compressive dressings must be avoided. As the digit is
shortened somewhat by the resection required for arthrodesis, skin flaps should not normally be needed at the
time of operation.
Interphalangeal joint arthrodesis has proven effective
in relieving pain and instability in cases of osteoarthritis, inflammatory arthritides, and trauma. The PIP joint
contributes 85% of intrinsic digital flexion and 30% of
the combined overall flexion of the finger,12 and painful
dysfunction at the joint can be debilitating. The morbidity associated with a fused joint is a worthwhile price
to pay for the increased function and comfort associated
with a painless joint. At the DIP joint, arthrodesis is also
tolerated very well. As it is accepted that rigid fixation
facilitates healing of an arthrodesis, fixation techniques
have evolved over the last 4 decades with a goal of
increasing the rigidity of the fixation, while simplifying
its application and insertion. Fixation with headless
compression screws accomplishes rigid fixation with a
straightforward surgical technique. It is consistent with
Mobergs requirements for a good digital arthrodesis
and is my preferred technique for digital arthrodesis
when there is adequate bone stock.
1119
Received for publication February 14, 2007; accepted June 13, 2007.
article.
Corresponding author: Stephen J. Leibovic, MD, Virginia Hand Center, 2819 N. Parham Rd., Richmond, VA 23294; e-mail: sleibo@mail1.
vcu.edu.
0363-5023/07/32A07-0024$32.00/0
doi:10.1016/j.jhsa.2007.06.010
References
1. Eaton RG, Malerich MM. Volar plate arthroplasty for the
proximal interphalangeal joint: A ten year review. J Hand
Surg 1980;5A:260.
2. Hasegawa T, Yamano Y. Arthroplasty of the proximal interphalangeal joint using costal cartilage grafts. J Hand Surg
1992;17B:583585.
3. Foucher G, Lenoble E, Smith D. Free and island vascularized
joint transfer for proximal interphalangeal reconstruction: a
series of 27 cases. J Hand Surg 1994;19A:8 16.
4. Zimmerman NB, Suhey PV, Clark GL, Wilgis EF. Silicone
interpositional arthroplasty of the distal interphalangeal joint.
J Hand Surg 1989;14A:882 887.
5. Moberg E. Arthrodesis of finger joints. Surg Clin N. Am
1960;40:465 470.
6. Lewis RC, Nordyke MD, Tenny JR. The tenon method of
small joint arthrodesis in the hand. J Hand Surg 1986;11A:
567569.
7. Potenza AD. A technique for arthrodesis of finger joints.
8. Wyrsch B, Dawson J, Aufranc S, Weikert D, Milek M.
Distal interphalangeal joint arthrodesis comparing tensionband wire and Herbert screw: A biomechanical and dimensional analysis. J Hand Surg 1996;21A:438 443.
9. Leibovic SJ, Strickland JW. Arthrodesis of the proximal
interphalangeal joint of the finger: Comparison of the use of
the Herbert screw with other fixation methods. J Hand Surg
1994;19A:181188.
10. Stern PJ, Fulton DB. Distal interphalangeal joint arthrodesis:
An analysis of complications. J Hand Surg 1992;17A:1139
1145.
11. Brutus JP, Palmer AK, Mosher JF, Harley BJ, Loftus JB.
Use of a headless compressive screw for distal interphalangeal joint arthrodesis in digits: Clinical outcome and review
of complications. J Hand Surg 2006;31A:85 89.
12. Little JW, Herndon JH, Thompson JS. Examination of the
hand. In: Reconstructive Plastic Surgery. Vol 6.
Philadelphia: WB Saunders Inc., 1977:2973.
13. Ayres JR, Goldstrohm GL, Miller GJ, Dell PC. Proximal
interphalangeal joint arthrodesis with the Herbert screw.
J Hand Surg 1998;13A:600 603.
14. Katzman SS, Gibeault D, Dickson K, Thompson JD. Use of
a Herbert screw for interphalangeal joint arthrodesis. Clin
Ortho 1993;296:127132.
15. Gomez CL, Proubasta I, Escriba I, Itarte J, Caceres E. Distal
interphalangeal joint arthrodesis: Treatment with Herbert
screw. J Southern Orth Assoc 2003;12:154 159.
16. Leibovic SJ. Arthrodesis of the proximal interphalangeal
joint of the finger using tension band wiring or Herbert
screws. Atlas of the Hand Clinics 1998;3:1730.
LETTERS TO THE EDITOR

Locked Intramedullary Nailing of Metacarpal
Fractures Secondary to Gunshot Wounds
the authors have a hypothesis as to why this is the

case? What antibiotic regime did they use?
Tim Stansfield, MBChB
Department of Orthopaedics
Hope Hospital
Salford, Manchester, UK
To the Editor:
Bach, Gonzalez, and Hall report good long-term

results from their 10 case series retrospective study
of locked intramedullary nailing in low-velocity gunshot metacarpal fractures. Where the wound circumstances permit, the arguments for such a fixation are
highly convincing. We would like to raise a few
points for clarification.
1. The study follow-up period ranged from 1 year to
9 years; were any patients lost to follow-up evaluation?
2. It has been reported elsewhere1 that limited
debridement is associated with low rates of complications for typical civilian handgun wound
fractures; how aggressive were the initial debridements? We acknowledge the authors recommendation for locked intramedullary (IM) fixation where the soft tissue sleeve is largely intact,
but in their series what soft tissue repairs were
necessary?
3. The nail was inserted through the fracture site. To
do this, the fracture should be distracted and
angulated to an extent that allows siting of the
nail. We would like to know, in those patients
with intact index and little metacarpals, how easy
was it to nail the middle and ring metacarpal? Did
any of the 9 patients with grafting have removal
of the nail and, if so, how easy was it to remove
the nail? The locking jig was screwed on to the
middle of the nail. Was there any problem with
failure of the targeting device? Was there any
locking screw breakage?
4. Bone graft incorporation depends on many factors of which graft stability and recipient bed are
crucial.2 Would they describe the technique they
used for graft stability? What proportion of grafts
were sutured as opposed to slotted? Would the
authors tell us about the size of the bone defect?
Bridging of bone defects combined with IM nailing has been reported in gunshot wounds to the
forearm.3
5. None of the series developed a superficial or deep
infection. The literature suggests that fractures
associated with civilian low-velocity gunshot
wounds have relatively low infection rates4,5; do
1120
doi:10.1016/j.jhsa.2007.05.013
References
1. Kiehn MW, Mitra A, Gutowski KA. Fracture management of
civilian gunshot wounds to the hand. Plast Reconstr Surg
2005;115(2):478 481.
2. Eyre-Brook AL, Baily RA, Price CH. Infantile pseudarthrosis
of the tibia. J Bone Joint Surg 1969;51B:604 613.
3. Spira E. Bridging of bone defects in the forearm with iliac
graft combined with intramedullary nailing. J Bone Joint Surg
1954;36B:642 646.
4. Knapp TP, Patzakis MJ, Lee J, Seipel PR, Abollahi K, Reisch
RB. Comparison of intravenous and oral antibiotic therapy in
the treatment of fractures caused by low-velocity gunshots.
J Bone Joint Surg 1996;78A:11671171.
5. Ryan JR, Hensel RT, Salciccioli GG, Pedersen HE. Fractures
of the femur secondary to low-velocity gunshot wounds.
J Trauma 1981;21(2):160 162.
In Reply:
I thank Dr. Stansfield for his interest in our study. I
will address the comments as numbered in his letter.
1. We were unable to get longer term follow-up

evaluation on several of the patients who had
only a year of follow-up evaluation.
2. Debridement was performed on clearly devascularized tissue. Generally damaged skin around
the entry and exit sites was also debrided. In all
of the cases, the skin could be closed without
undue tension. Any gunshot wounds that required
a soft tissue reconstruction would be treated initially with an external fixator and/or a cement
antibiotic spacer.
3. The placement of the rod was only performed in
fractures with notable comminution. The central
area of comminution made introduction of the
nail rather easy. Once the rod is in place, the
fracture can then be brought to the proper length
and the rod locked. The only hardware complication was failure of the locked rod that was
placed without grafting. None of the patients had
removal of the nail. There were no fractures of
the targeting device or the locking screws.
4. The grafts were generally slotted and sutured

together. In the case of the corticocancellous
grafts, the cancellous bone can be left protuberant
beyond the edge of the cortex so that the cancellous graft acts as a filler in the space between
the graft and host bone. The defects were approximately 1.53 cm. Because the ends are jagged or
oblique, this is only an approximation.
5. We believe that the incidence of infection was
low because (1) gunshot wounds to the hand
usually have little contamination from clothing,
dirt, and so forth; (2) the weapon favored by
urban criminals is a handgun with relatively low
kinetic energy and limited soft tissue destruction;
and (3) the patients were treated with early debridement and antibiotics. We have documented
the relatively low rate of infection in gunshot
wounds to the hand in a previous report.1
Mark H. Gonzalez, MD
Wolcott Avenue, M/C 844
Chicago, IL 60612
doi:10.1016/j.jhsa.2007.05.012
Reference
1. Gonzalez MH, McKay W, Hall RF. Low-velocity gunshot
wounds of the metacarpal: Treatment by early stable fixation
and bone grafting. J Hand Surg 1993;18A:267270.
Costochondral Autograft as a Salvage

Procedure After Failed Trapeziectomy in
Trapeziometacarpal Osteoarthritis
in Table 1 are incorrect or if there was another

reason for not being able to write.
The 2 patients that had a good result (no. 1 and no. 3)
could write postoperatively with their operated hands
(the left hands) in spite of the fact that they were
right-handed (M&M). What was the question the
authors actually posed to their right-handed patients to asses their ability to write with their left
hands? I also wonder how was the question about
the ability to use a contact key (to switch on a car
engine) posed. Most cars have the starting key on
the right side (apart from Porsche cars, which have
the key on the left side, perhaps UK cars as well).
I found surprising the reluctance of the authors to
recommend their operation in spite of the fact their
reported results were so good (see Discussion in
Ref. 1). Why the Cooney et al2 approach? If one
compares their reported results with those of
Cooney et al, theirs are better. For example, they
referred that their patients had 0 pain (over 100,
the maximum pain), and the average postoperative
pinch strength was 5.5 kg. Conversely, in the
Cooney et al study, the average postoperative pain
was 0.6 (over 5, the maximum pain) and the average pinch 3.4 kg.
The authors asses their patients at 2 years, but nothing is said about the follow-up evaluation afterward.
Did their results deteriorate as time went by?
I appreciate very much the authors help and insight and congratulate them on their innovative operation for this very difficult problem.
Francisco del Pial, MD
Caldern de la Barca 16-entlo.
39002 Santander, Spain
E-mail: drpinal@drpinal.com
doi:10.1016/j.jhsa.2007.05.014
To the Editor:
I read with interest the paper by Glard et al on

trapeziometacarpal chondral arthroplasty.1 I would
like to congratulate the authors on their innovative
work. There are, however, some inconsistencies
when assessing their results, and I wonder if they
could clarify some issues.
When assessing pain, the authors said that the pain

score was an average from the resting pain and the
pain during activity. Some of the patients could
hardly write with their hands yet they had 0 pain
(see Table 1 of Ref. 1). I wonder if the data given
1121
Note: Glard et al chose not to respond to this letter to the editor.
References
1. Glard Y, Gay A, Valenti D, Berwald C, Guinard D, Legre R.
Costochondral autograft as a salvage procedure after failed
trapeziectomy in trapeziometacarpal osteoarthritis. J Hand
Surg 2006A;31:14611467.
2. Cooney WP III, Leddy TP, Larson DR. Revision of thumb
trapeziometacarpal arthroplasty. J Hand Surg 2006;31A:219
227.
LETTERS TO THE EDITOR

Locked Intramedullary Nailing of Metacarpal
Fractures Secondary to Gunshot Wounds
the authors have a hypothesis as to why this is the

case? What antibiotic regime did they use?
Tim Stansfield, MBChB
Department of Orthopaedics
Hope Hospital
Salford, Manchester, UK
To the Editor:
Bach, Gonzalez, and Hall report good long-term

results from their 10 case series retrospective study
of locked intramedullary nailing in low-velocity gunshot metacarpal fractures. Where the wound circumstances permit, the arguments for such a fixation are
highly convincing. We would like to raise a few
points for clarification.
1. The study follow-up period ranged from 1 year to
9 years; were any patients lost to follow-up evaluation?
2. It has been reported elsewhere1 that limited
debridement is associated with low rates of complications for typical civilian handgun wound
fractures; how aggressive were the initial debridements? We acknowledge the authors recommendation for locked intramedullary (IM) fixation where the soft tissue sleeve is largely intact,
but in their series what soft tissue repairs were
necessary?
3. The nail was inserted through the fracture site. To
do this, the fracture should be distracted and
angulated to an extent that allows siting of the
nail. We would like to know, in those patients
with intact index and little metacarpals, how easy
was it to nail the middle and ring metacarpal? Did
any of the 9 patients with grafting have removal
of the nail and, if so, how easy was it to remove
the nail? The locking jig was screwed on to the
middle of the nail. Was there any problem with
failure of the targeting device? Was there any
locking screw breakage?
4. Bone graft incorporation depends on many factors of which graft stability and recipient bed are
crucial.2 Would they describe the technique they
used for graft stability? What proportion of grafts
were sutured as opposed to slotted? Would the
authors tell us about the size of the bone defect?
Bridging of bone defects combined with IM nailing has been reported in gunshot wounds to the
forearm.3
5. None of the series developed a superficial or deep
infection. The literature suggests that fractures
associated with civilian low-velocity gunshot
wounds have relatively low infection rates4,5; do
1120
doi:10.1016/j.jhsa.2007.05.013
References
1. Kiehn MW, Mitra A, Gutowski KA. Fracture management of
civilian gunshot wounds to the hand. Plast Reconstr Surg
2005;115(2):478 481.
2. Eyre-Brook AL, Baily RA, Price CH. Infantile pseudarthrosis
of the tibia. J Bone Joint Surg 1969;51B:604 613.
3. Spira E. Bridging of bone defects in the forearm with iliac
graft combined with intramedullary nailing. J Bone Joint Surg
1954;36B:642 646.
4. Knapp TP, Patzakis MJ, Lee J, Seipel PR, Abollahi K, Reisch
RB. Comparison of intravenous and oral antibiotic therapy in
the treatment of fractures caused by low-velocity gunshots.
J Bone Joint Surg 1996;78A:11671171.
5. Ryan JR, Hensel RT, Salciccioli GG, Pedersen HE. Fractures
of the femur secondary to low-velocity gunshot wounds.
J Trauma 1981;21(2):160 162.
In Reply:
I thank Dr. Stansfield for his interest in our study. I
will address the comments as numbered in his letter.
1. We were unable to get longer term follow-up

evaluation on several of the patients who had
only a year of follow-up evaluation.
2. Debridement was performed on clearly devascularized tissue. Generally damaged skin around
the entry and exit sites was also debrided. In all
of the cases, the skin could be closed without
undue tension. Any gunshot wounds that required
a soft tissue reconstruction would be treated initially with an external fixator and/or a cement
antibiotic spacer.
3. The placement of the rod was only performed in
fractures with notable comminution. The central
area of comminution made introduction of the
nail rather easy. Once the rod is in place, the
fracture can then be brought to the proper length
and the rod locked. The only hardware complication was failure of the locked rod that was
placed without grafting. None of the patients had
removal of the nail. There were no fractures of
the targeting device or the locking screws.
4. The grafts were generally slotted and sutured

together. In the case of the corticocancellous
grafts, the cancellous bone can be left protuberant
beyond the edge of the cortex so that the cancellous graft acts as a filler in the space between
the graft and host bone. The defects were approximately 1.53 cm. Because the ends are jagged or
oblique, this is only an approximation.
5. We believe that the incidence of infection was
low because (1) gunshot wounds to the hand
usually have little contamination from clothing,
dirt, and so forth; (2) the weapon favored by
urban criminals is a handgun with relatively low
kinetic energy and limited soft tissue destruction;
and (3) the patients were treated with early debridement and antibiotics. We have documented
the relatively low rate of infection in gunshot
wounds to the hand in a previous report.1
Mark H. Gonzalez, MD
Wolcott Avenue, M/C 844
Chicago, IL 60612
doi:10.1016/j.jhsa.2007.05.012
Reference
1. Gonzalez MH, McKay W, Hall RF. Low-velocity gunshot
wounds of the metacarpal: Treatment by early stable fixation
and bone grafting. J Hand Surg 1993;18A:267270.
Costochondral Autograft as a Salvage

Procedure After Failed Trapeziectomy in
Trapeziometacarpal Osteoarthritis
in Table 1 are incorrect or if there was another

reason for not being able to write.
The 2 patients that had a good result (no. 1 and no. 3)
could write postoperatively with their operated hands
(the left hands) in spite of the fact that they were
right-handed (M&M). What was the question the
authors actually posed to their right-handed patients to asses their ability to write with their left
hands? I also wonder how was the question about
the ability to use a contact key (to switch on a car
engine) posed. Most cars have the starting key on
the right side (apart from Porsche cars, which have
the key on the left side, perhaps UK cars as well).
I found surprising the reluctance of the authors to
recommend their operation in spite of the fact their
reported results were so good (see Discussion in
Ref. 1). Why the Cooney et al2 approach? If one
compares their reported results with those of
Cooney et al, theirs are better. For example, they
referred that their patients had 0 pain (over 100,
the maximum pain), and the average postoperative
pinch strength was 5.5 kg. Conversely, in the
Cooney et al study, the average postoperative pain
was 0.6 (over 5, the maximum pain) and the average pinch 3.4 kg.
The authors asses their patients at 2 years, but nothing is said about the follow-up evaluation afterward.
Did their results deteriorate as time went by?
I appreciate very much the authors help and insight and congratulate them on their innovative operation for this very difficult problem.
Francisco del Pial, MD
Caldern de la Barca 16-entlo.
39002 Santander, Spain
E-mail: drpinal@drpinal.com
doi:10.1016/j.jhsa.2007.05.014
To the Editor:
I read with interest the paper by Glard et al on

trapeziometacarpal chondral arthroplasty.1 I would
like to congratulate the authors on their innovative
work. There are, however, some inconsistencies
when assessing their results, and I wonder if they
could clarify some issues.
When assessing pain, the authors said that the pain

score was an average from the resting pain and the
pain during activity. Some of the patients could
hardly write with their hands yet they had 0 pain
(see Table 1 of Ref. 1). I wonder if the data given
1121
Note: Glard et al chose not to respond to this letter to the editor.
References
1. Glard Y, Gay A, Valenti D, Berwald C, Guinard D, Legre R.
Costochondral autograft as a salvage procedure after failed
trapeziectomy in trapeziometacarpal osteoarthritis. J Hand
Surg 2006A;31:14611467.
2. Cooney WP III, Leddy TP, Larson DR. Revision of thumb
trapeziometacarpal arthroplasty. J Hand Surg 2006;31A:219
227.
BOOK REVIEW
Fractures of the Hand and Wrist. Ring DC,
Cohen MS, eds. New York: Informa Healthcare
USA, Inc., 2007, 195 pages, $199.95.
Fractures of the Hand and Wrist is an attractive,

well-illustrated, hardcover review of bony injuries
from the fingertip to the distal radius. It is designed to
be an easy-to-read and practical reference of common fractures and dislocations and their management. Not meant to be a comprehensive discussion
on upper extremity trauma, this text is focused and
concise, providing a practical review of current concepts (p. v) to a target audience of orthopaedic and
hand surgeons who treat these injuries.
Written by a panel of internationally acclaimed
hand surgeons, and presented in 7 clearly written
chapters, the text flows seamlessly from distal to
proximal; beginning with a discussion of fingertip
and thumb injuries in Chapter 1, including a lucid
section on soft tissue coverage options, and moving
on to a thorough review of phalanx shaft fractures in
Chapter 2. Chapter 3 on dislocations and fracture/
dislocations of the metacarpophalangeal and proximal interphalangeal joints includes many radiograph
images that help elucidate a challenging and complex
topic. Chapter 4 discusses operative management of
metacarpal fractures based on the location of the
fracture, and Chapter 5 is a tour de force on carpal
dislocations and fracture/dislocations. As Fractures
of the Hand and Wrist concludes with elegant re-
1122
views of both scaphoid (Chapter 6) and distal radius

(Chapter 7) fractures, the uniformly excellent quality
of the images and text continues, and the reader is left
wanting more.
At $199.95, this text may indeed leave some readers wishing it contained a greater breadth of topics.
However, it is precisely the concise nature of this text
that may make it so attractive to many others searching for a user-friendly and practical reference when
approaching challenging clinical topics such as proximal interphalangeal joint fracture/dislocations or
complex distal radius fractures. It is important to note
before buying this text that it is in no way intended to
be a comprehensive treatise on all fractures of the
hand and wrist. In fact, Drs. Ring and Cohen chose
not to devote exclusive chapters to distal ulnar injuries or isolated carpal fractures other than the scaphoid. Rather than aiming to cover every topic imaginable, Ring and Cohen sought to create a readable
text for easy reference . . . of the practical and up-todate aspects of fracture care (p. iii). With remarkable clarity, an internationally flavored panel of authors enable Ring and Cohen to meet their goal.
Fractures of the Hand and Wrist is a concise and
practical reference that would make a welcome addition to any orthopaedic or hand surgeons library.
Craig M. Rodner
University of Connecticut
doi:10.1016/j.jhsa.2007.05.026

Jurnal On Hand Surgery

Hochgeladen von

Dokumentinformationen

Originaltitel

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

Jurnal On Hand Surgery

Hochgeladen von

Copyright:

Verfügbare Formate

THE JOURNAL OF

electronic or mechanical, including photocopy, recording, or any

Roy A. Meals Los Angeles, California

Mark E. Baratz Pittsburgh, Pennsylvania

Vincent R. Hentz Palo Alto, California

Richard A. Berger Rochester, Minnesota

Kevin C. Chung Ann Arbor, Michigan

Associate Editors: Scientific

Associate Editors: Review/Technique

Reid A. Abrams, San Diego, California

Kimberly Amrami, Rochester, Minnesota

British and European Volume

David Shewring, Cardiff

VOLUME 32A, NUMBER 7, SEPTEMBER 2007

History of Hand Surgery

Clifton G. Meals and Roy A. Meals

Susanne Rein, Hartmut Schikore, Wolfgang Schneiders, Michael Amlang,

The Risk of Adverse Outcomes in Extra-Articular Distal

Ruby Grewal and Joy C. MacDermid

Biomechanical Evaluation of Locking Plate Radial Shaft

Jason W. Roberts, Steven I. Grindel, Brandon Rebholz, and Mei Wang

Distal Radius Osteotomy in the Elderly Patient Using

Santiago Lozano-Calderon, Michael Moore, Matthew Liebman, and

Rupture of the Flexor Pollicis Longus Tendon After Volar

Raymond A. Klug, Cyrus M. Press, and Mark H. Gonzalez

Clinical-Electromyography Correlation in Infants With

Carlos O. Heise, Mario G. Siqueira, Roberto S. Martins, and

Association Between Lunate Morphology and Carpal 1009

Carpal Kinematics During Simulated Active and Passive 1013

Long-Term Follow-Up of an Undiagnosed Trans-Scaphoid 1020

Objective Features and Aesthetic Outcome of Pollicized 1031

Isolated Wedge Osteotomy of the Ulna for Mild Madelungs 1037

Surgical Treatment of the Pediatric Trigger Finger 1043

Digital Ischemia Due to Essential Thrombocythemia: 1053

Aeromonas hydrophila Infection Causing Delayed Vascular 1058

Immunologic Approaches to Composite Tissue Allograft 1072

Stability of the Distal Radioulna Joint: Biomechanics, 1086

Instructional Course Lecture

Letters to the Editor

section that animal care complied with the guidelines of the

The Journal of Hand Surgery

dress. If the name or address for reprint requests is different, so state.

Letters to the Editors of sufficient interest will be considered for

must appear at the end of the manuscript. Please refrain from

LEVELS OF EVIDENCE FOR PRIMARY RESEARCH QUESTION

Types of Clinical Studies

Economic and Decision

Testing of previously developed

Sensible costs and

Systematic review2 of Level-I

Systematic review2 of Level-I

Systematic review2 of Level-I

Sensible costs and

Untreated controls from a

Systematic review2 of Level-II

Systematic review2 of Level-II

Study of nonconsecutive patients Analyses based on limited

Systematic review2 of Level-III

Poor reference standard

SUBMIT MANUSCRIPTS ONLINE AT HTTP://EES.ELSEVIER.COM/JHS

The Journal of Hand Surgery

History at Our Fingertips

The Journal of Hand Surgery