You are on page 1of 6


Superficial Temporal Artery and Vein as

Recipient Vessels for Facial and Scalp
Microsurgical Reconstruction
Scott L. Hansen, M.D.
Robert D. Foster, M.D.
Amarjit S. Dosanjh, M.D.
Stephen J. Mathes, M.D.
William Y. Hoffman, M.D.
Pablo Leon, M.D.
San Francisco, Calif.
Background: Although free flap transfer is commonly performed to reconstruct
defects of the upper two-thirds of the face and scalp, the superficial temporal
artery and vein have historically not been considered adequate for microsurgical
reconstruction and have rarely been described as recipient vessels. The purpose
of this study was to determine the indications for and effectiveness of using the
superficial temporal vessels for scalp and face reconstruction.
Methods: Retrospective chart review on all patients undergoing microsurgical
reconstruction for defects of the upper two-thirds of the face between 1996 and
2003 revealed45 free tissue transfers inwhichthe superficial temporal artery and
vein were considered for use as recipient vessels. Flap success rates and post-
operative course were evaluated.
Results: Forty-three patients underwent 45 free flap transfers. The superficial
temporal artery was used as the recipient artery in every case. In three cases, the
superficial temporal vein was not suitable as the recipient vein and required use
of a vein in the neck. The median length of follow-up was 4 years. Flap survival
was 96 percent. Five patients required reoperation for vascular compromise.
One of these patients ultimately had flap failure. In that patient, a subsequent
attempt at microvascular flap reconstruction was successful using the same
superficial temporal artery and vein as recipient vessels.
Conclusions: Use of the superficial temporal artery and vein for scalp and face
reconstruction is reliable and safe. The superficial temporal artery and vein
should be considered as primary recipient vessels in microsurgical reconstruc-
tion of the upper two-thirds of the face and/or scalp. (Plast. Reconstr. Surg. 120:
1879, 2007.)
icrosurgical free tissue transfer is now
considered the standard for reconstruc-
tion of defects resulting from tumor ab-
lation, congenital abnormalities, or traumatic in-
jury. The focus over the past two decades has
shifted to optimizing this process. Experience has
taught us that only a select number of donor sites
are needed for the majority of reconstructions.
For the upper two-thirds of the face and scalp, the
rectus abdominis and latissimus dorsi muscles,
andless commonly the scapular flap, provide great
flap reliability and flexibility in restoring the nor-
mal soft-tissue contour to the scalp and midface
while obliterating the maxillary and/or orbital
cavities when desired (Fig. 1). When thin resur-
facing of the scalp is indicated, the radial forearm
flap has been an ideal choice.
In contrast to donor-site selection, the optimal
choice of recipient vessels for scalp and midfacial
reconstruction is less well defined. When Schus-
terman et al. reviewed their experience with 308
microsurgical reconstructions, they documented
that greater than 90 percent of their recipient
vessels were large-caliber vessels in the neck.
felt that using large-caliber vessels enhanced their
success rate. Subsequent studies by members of
the reconstructive teamfromthe Memorial Sloan-
Kettering Cancer Center have endorsed the need
to use recipient vessels in the neck.
The rec-
From the Division of Plastic and Reconstructive Surgery,
University of California, San Francisco.
Received for publication February 6, 2006; accepted June 15,
Presented at the American Society for Reconstructive Micro-
surgery meeting, in Palm Springs, California, January 15
through 18, 2005.
Copyright 2007 by the American Society of Plastic Surgeons
DOI: 10.1097/ 1879
ommended recipient vessels for flaps commonly
included the facial, lingual, external carotid, su-
perior thyroid, and superficial cervical arteries
and their corresponding venous systems. How-
ever, Cordeiro et al. also acknowledged that the
long distance from the midface to the neck was a
challenging aspect of free flap reconstruction.
Despite techniques to increase pedicle length,
vein grafting may be necessary to span the 10 to 12
cm to the ipsilateral neck.
With all that has been written about head re-
construction for defects of the upper two-thirds of
the face, little has been written concerning the use
of the superficial temporal artery and vein as re-
cipient vessels. Hussussian and Reece used the
superficial temporal vessels in less than one-third
of their scalp reconstructions because they felt
they were of insufficient caliber.
More recently,
Lipa and Butler described the use of the superfi-
cial temporal vessels in five of their six scalp re-
constructions and advocated their use.
a recent study reported using the temporal vessels
in approximately 20 percent of their head and
neck reconstructions.
To date, no center has evaluated a large group
of patients undergoing microvascular free tissue
transfer to the superficial temporal vessels. The
purpose of this study was to determine the indi-
cations for and the effectiveness of using the su-
perficial temporal vessels for microvascular face
and scalp reconstruction. In addition, we document
techniques to optimize the use of these vessels.
A retrospective chart review was performed on
all patients with defects of the scalp and upper
two-thirds of the face who underwent microsurgi-
cal reconstruction with free flap transfer between
January of 1996 and December of 2003 at the
University of California, San Francisco and affili-
ated hospitals. From that group of patients, those
cases in which the superficial temporal artery and
vein were used as recipient vessels were further
analyzed and form the basis for this study. Appro-
priate University of California, San Francisco
Committee on Human Research approval was ob-
tained before this study. Patient gender and age,
cause and size of the defect, flap choice for re-
construction, anastomotic technique, method of
flap monitoring, postoperative course, and com-
plications were recorded.
Between January of 1996 and December of
2003, 257 consecutive patients underwent micro-
surgical head and neck reconstruction. Of these,
45 microsurgical tissue transfers to the superficial
temporal vessels were attempted in 43 patients
after they were evaluated preoperatively to have a
palpable superficial temporal artery. Seventy-one
of 257 patients that underwent microsurgical re-
construction had defects in the scalp/midface re-
gion. Of these 71 patients, 12 had defects in the
parotid or ear region and thus the superficial tem-
Fig. 1. (Left) Preoperativeviewof a manwitha posterior scalpdefect demonstrates theextent of thedefect
after a failed local ap. (Right) Postoperative view several months after successful latissimus dorsi ap and
skin grafting.
Plastic and Reconstructive Surgery December 2007
poral vessels were not suitable for microsurgical
reconstruction purposes. The superficial tempo-
ral artery and vein were not used in 14 of 59
patients who had defects in the orbit/midface and
their flap was better positioned if the vessels in the
neck were used. This resulted in 45 of 59 patients
(76 percent) having their superficial temporal ar-
tery and vein used for scalp/midface reconstruc-
tion. There were 24 male and 21 female patients.
The ages of the patients ranged from2 to 91 years,
with a median age of 63 years. The areas of re-
construction included scalp (n 19), orbit (n
4), skull base (n 4), midface (n 4), and orbital
and midface (n 14) defects. The tissue trans-
ferred included the rectus abdominis (n 23),
radial forearm(n 9), latissimus (n 8), scapular
flap (n 4), and serratus (n 1) and was based
on the size and contour of the defect (Table 1).
The volume of our defects ranged from 32 cm
500 cm
, with a mean defect size of 129 cm
. Twenty-
eight of the patients (65 percent) had undergone
previous radiation therapy. All anastomoses were
performed in an end-to-end fashion using inter-
rupted 9-0 nylon suture. Bony replacement was
not necessary for any of the reconstructions. Be-
ginning in 2000, the majority of our patients had
an implantable Doppler device placed around the
recipient veindistal to the venous anastomosis that
provided continuous flap monitoring for 5 to 7
days postoperatively. Neither dextran nor heparin
was routinely administered postoperatively. Only
patients who required take-back for anastomotic
complications were given heparin following the
vascular revision.
In all 43 patients, the main trunk of the su-
perficial temporal artery was used as the recipient
artery. There was minimal atherosclerosis notedin
the superficial temporal artery for all patients. In
40 of the patients, we were able to identify the
main trunk of the superficial temporal vein, and
it had a less than 2:1 discrepancy with the donor
vein from the flap. In these situations, the super-
ficial temporal vein was used as the recipient ves-
sel. Three of the 43 patients had a greater than 2:1
discrepancy between the donor vein and the su-
perficial temporal vein on initial exploration. In
these three patients, an ipsilateral external jugular
vein, ipsilateral retromandibular vein, or con-
tralateral facial vein was used; the latter two re-
quired interposition vein grafts. The overall flap
survival rate was 96 percent. The overall compli-
cation rate was 31 percent (Table 2). Five patients
(11 percent) required a return to the operating
room for vascular compromise. All flaps requiring
vascular revision were taken back to the operating
room within 24 hours. Two of these patients were
monitored with an implantable Doppler device
and were returned to the operating room within
2 hours of identifying a change in the Doppler
signal. Two of the five patients were noted to have
an arterial thrombosis, and their anastomosis was
revised. The other three patients experienced a
venous thrombosis at the anastomotic site. Of
these three patients, one required a revision of the
anastomosis. Asecond patient had the venous out-
flow revised with an interposition vein graft to the
facial vein. The third patient with venous obstruc-
tion, in addition to revision of their venous anas-
tomosis with the superficial temporal vein, had
venous outflow optimized with the use of a vena
comitans of the superficial temporal artery. There
were no anatomical problems noted in the recip-
ient veins in the three patients who experienced
a venous thrombosis. Moreover, we did not see any
correlation between venous thrombosis and pres-
ence of a suitable superficial temporal vein. The
three patients who had venous thrombosis were
Table 2. Complications (n 14)
Complications No.
Venous thrombosis 3
Arterial thrombosis 2
Hematoma* 5
Late flap loss 1
Cellulitis 1
Necrosis of rectus skin island 2
*Two of five hematomas occurred after the patients were anticoag-
ulated following reexploration because of vascular compromise.
This flap loss was attributable to the patient wearing glasses post-
Cellulitis occurred at the inferior portion of a flap.
These skin islands had been debulked with liposuction.
Table 1. Choice of Flap by Site of Reconstruction in
43 Patients
Site of Reconstruction No. of Flaps
Scalp (n 19)
Latissimus 8
Radial forearm 5
Rectus 3
Scapular 3
Orbit and midface (n 14)
Rectus 13
Scapular 1
Orbit only (n 4)
Rectus 4
Midface only (n 4)
Rectus 1
Serratus 1
Radial forearm 2
Skull base (n 4)
Rectus 2
Radial forearm 2
Volume 120, Number 7 Temporal Vessels for Reconstruction
different fromthe three patients who did not have
sufficiently large superficial temporal veins and
required the use of veins in the neck. The second
patient with flap loss initially had arterial throm-
bosis that required revisionof the anastomosis and
the flap ultimately failed after a hematoma devel-
oped under the flap after the patient was antico-
agulated following the first reoperation. The over-
all complication rate was 31 percent.
The length of follow-up ranged from3 months
to 8 years, with an average of 4 years. Two flaps (4
percent) in this series failed. One was a patient
that initially required take-back for arterial throm-
bosis resulting from arterial kinking. After this
take-back, the patient was started on heparin and
developed a hematoma under the flap. This flap
subsequently failed and the patient underwent re-
construction with another flap performed to the
same superficial temporal vessels without further
adverse sequelae. A second patient developed flap
failure during the perioperative period despite
being discharged to home with a viable flap. Fur-
ther evaluation determined that the patient used
her eyeglasses (constricting the vascular pedicle),
despite instructions to the contrary. This patient
successfully underwent a second microvascular re-
construction using the facial vessels.
The ability to provide immediate reconstruc-
tion for large defects of the scalp and face has
significantly improved the quality of life for many
patients. Refining donor-site selection has further
improved results.
For the upper two-thirds of
the face, where bone flaps are rarely necessary, the
radial forearm, rectus abdominis, and latissimus
dorsi flaps are our flaps of choice. For coverage of
large wounds, the rectus abdominis and latissimus
dorsi musculocutaneous flaps are preferred.
These flaps are effective in skull base coverage,
particularly after a dural resection and repair
where muscle bulk is important to adequately pro-
tect the repair. The thin, broad muscle of the
latissimus flap (in combination with skin grafting) is
especially well suitedafter largescalpresections.
small defects, the radial forearm fasciocutaneous
flapprovides thin, pliableskinwithlimitedsoft-tissue
bulk, characteristics that make it ideal for resurfac-
ing the cheek and scalp.
Historically, concerns about vessel diameter
and supplying enough blood to support flaps with
a large surface area have limited the use of the
superficial temporal vessels for head and neck re-
construction. In contrast to previous reports, our
series suggests that virtually all flap reconstruc-
tions to the scalp and midface can be successful
using the superficial temporal vessels no matter
which flap is used. Anatomical studies in the lit-
erature examining vessel diameters in the head
and neck suggest that the difference in vessel di-
ameters of the superficial temporal and facial ar-
tery and vein may not be as significant, on average,
as has been understood.
We have found the anatomical course of the
superficial temporal vessels to be very reliable and
consistent with the descriptions in the literature.
The superficial temporal artery is one of the ter-
minal branches of the external carotid artery. It
begins in the substance of the parotid gland, be-
hind the neck of the mandible, and crosses over
the root of the zygomatic process of the temporal
bone. It consistently divides into two major
branches: the frontal and parietal branches.
The recipient arterial anastomosis inour study was
always at the main trunk of the superficial tem-
poral artery, anterior to the tragus, before the
branch point (Fig. 2). The spectrum of size dis-
crepancies between recipient and donor vessels
ranged from 1:1 to 1:2. We were able to accom-
modate the size discrepancies between recipient
and donor arteries for all patients. The same su-
perficial temporal artery and vein were used in
one patient whose flap failed after initial difficul-
ties with the arterial anastomosis, in the subse-
quent free flap reconstruction.
The superficial temporal vein demonstrates
more variability with respect to its branching pat-
Fig. 2. Intraoperativedemonstrationof thesupercial temporal
artery and vein dissected just anterior to the tragus.
Plastic and Reconstructive Surgery December 2007
tern (it can divide into one, two, or three major
branches) and its relationship to the superficial
temporal artery.
Park et al. noted that the su-
perficial temporal artery and vein ran in parallel
to supply temporoparietal flaps in only 63.4 per-
cent of patients studied. The remainder had ve-
nous drainage through the posterior auricular,
occipital, or diploic veins.
In cadaver studies, it
was noted that the major branches of the super-
ficial temporal vein were different from those of
the superficial temporal artery. The thin parallel
veins to the superficial temporal artery are venae
However, we overcome this variabil-
ity by performing the venous anastomosis using
either the closest branch to the superficial tem-
poral artery or, more reliably, to the main trunk of
the superficial temporal vein at its initial branch-
ing point. During the initial operation, we were
not able to identify a suitable recipient vein in
three patients. In these cases, a suitable recipient
vein in the neck was used. Two of these patients
required a vein graft to bridge the distance. The
primary use of the superficial temporal vessels ob-
viates the need for a longer pedicle or vein graft.
Although vein grafting does not decrease overall
flap survival, it necessitates the use of additional
donor sites and has been associated with a higher
revision rate.
As our experience with the superficial tempo-
ral artery and veinincreased, we have made several
modifications to our technique. Preoperatively,
we evaluate for the presence of the superficial
temporal artery by direct palpation and Doppler
imaging if a pulse is not easily identified. Other
preoperative imaging studies are not performed.
Intraoperatively, we always explore the superficial
temporal artery and vein by means of a face-lift
incision before looking for an alternative in re-
construction involving the upper two-thirds of the
face or scalp. After gross dissection, we perform
the majority of our vessel preparation under the
operating microscope. To avoid spasm, we handle
the vessels minimally and inject papaverine into
the adjacent adventitia with a 30-gauge needle. The
adventitia of the artery and vein is minimally
stripped. All patients routinely receive an implant-
able venous Doppler device for postoperative vas-
cular monitoring. Althougharterial monitoring of
the superficial temporal vessels is possible without
the use of an implantable Doppler device, arterial
monitoring does not allow for rapid detection of
venous thrombosis, as an arterial pulse will persist
for as long as several hours after venous throm-
bosis. By the time the arterial signal is lost, salvage
of the flap is less likely. In addition to postoper-
ative vascular monitoring, the Doppler device has
been valuable during closure of the recipient site.
The strain being placed on the pedicle by the
closure can be monitored directly by the venous
Doppler device and the closure can be adjusted as
needed. Since adopting this practice, no flaps have
failed during the immediate postoperative period.
Our overall complication rate of 31 percent is
comparable to previous studies that have docu-
mented an overall complication rate ranging from
17.5 to 59 percent.
Other than administra-
tion of daily aspirin begun immediately postoper-
atively, blood thinners (heparin, dextran) are not
part of the routine postoperative care. The venous
Doppler device is used for 5 to 7 days.
The superficial temporal vessels are reliable
and safe recipient vessels for microsurgical recon-
struction of the upper two-thirds of the face and
scalp. Free tissue transfer was successfully carried
out in a patient as young as 2 and a patient as old
as 91 years. Flap perfusion was adequate for mus-
cles as large as the latissimus dorsi and defects as
large as 500 cm
. Incorporation of the principles
presented has allowed us to reliably reconstruct
the middle to upper face and scalp by a simpler
approach compared with previously described
Scott L. Hansen, M.D.
Division of Plastic Surgery
Surgical Research Division, Box 1302
University of California, San Francisco
San Francisco, Calif. 94131-1302
None of the authors has any financial interests re-
garding the products, devices, or drugs mentioned in this
1. Foster, R., Anthony, J., Singer, M., et al. Reconstruction of
complex midfacial defects. Plast. Reconstr. Surg. 99: 1555,
2. Disa, J., Pusic, A., Hidalgo, D., et al. Simplifying microvas-
cular head and neck reconstruction: A rational approach to
donor site selection. Ann. Plast. Surg. 47: 385, 2001.
3. Cordeiro, P., and Disa, J. Challenges in midface reconstruc-
tion. Semin. Surg. Oncol. 19: 218, 2000.
4. Schusterman, M., Miller, M., Reece, G., et al. Asingle centers
experience with 308 free flaps for repair of head and neck
cancer defects. Plast. Reconstr. Surg. 93: 472, 1994.
5. Hidalgo, D., Disa, J., Cordeiro, P., et al. A review of 716
consecutive free flaps for oncologic surgical defects: Refine-
Volume 120, Number 7 Temporal Vessels for Reconstruction
ment in donor-site selection and technique. Plast. Reconstr.
Surg. 102: 722, 1998.
6. Hussussian, C., and Reece, G. Microsurgical scalp recon-
struction in the patient with cancer. Plast. Reconstr. Surg. 109:
1828, 2002.
7. Lipa, J., and Butler, C. Enhancing the outcome of free la-
tissimus dorsi muscle flap reconstruction of scalp defects.
Head Neck 26: 46, 2004.
8. Nahabedian, M., Singh, N., and Deune, E. G. Recipient vessel
analysis for microvascular reconstruction of the head and
neck. Ann. Plast. Surg. 52: 148, 2004.
9. Park, C., Lew, D., and Yoo, W. An analysis of 123 temporopa-
rietal fascial flaps: Anatomic and clinical considerations in
total auricular reconstruction. Plast. Reconstr. Surg. 104: 1295,
10. Zhao, Z., Li, S., Xu, J., et al. Color Doppler flow imaging of
the facial artery and vein. Plast. Reconstr. Surg. 106: 1249,
11. Imanishi, N., Nakajima, H., Minabe, T., et al. Venous drain-
age architecture of the temporal and parietal regions: Anat-
omy of the superficial temporal artery and vein. Plast. Re-
constr. Surg. 109: 2197, 2002.
Online CME Collections
This partial list of titles in the developing archive of CME article collections is available online at www. These articles are suitable to use as study guides for board certification, to help readers refamiliarize
themselves on a particular topic, or to serve as useful reference articles. Articles less than 3 years old can be taken for CME
Head and Neck
The Role of Open Rhinoplasty in the Management of Nasal Dermoid CystsRod J. Rohrich et al.
The Evolution of the Hughes Tarsoconjunctival Flap for Lower Eyelid ReconstructionRod J. Rohrich and
Ross I. S. Zbar
Subunit Principles in Midface Fractures: The Importance of Sagittal Buttresses, Soft-Tissue Reductions, and
Sequencing Treatment of Segmental FracturesPaul Manson et al.
Microsurgical Replantation of the Amputated NoseDennis C. Hammond et al.
Advances in Head and Neck ReconstructionGeoffrey C. Gurtner and Gregory R. D. Evans
Auricular Reconstruction: Indications for Autogenous and Prosthetic TechniquesCharles H. Thorne et al.
Reconstruction of the CheekFrederick J. Menick
Management of the Recurrent, Benign Tumor of the Parotid GlandDavid L. Larson
The Superiorly Based Nasolabial Flap for Simultaneous Alar and Cheek ReconstructionRod J. Rohrich and
Matthew H. Conrad
Pharmacologic Optimization of Microsurgery in the New MillenniumMatthew H. Conrad and William P.
Adams, Jr.
Understanding the Nasal Airway: Principles and PracticeBrian K. Howard and Rod J. Rohrich
Auricular Reconstruction for Microtia: Part I. Anatomy, Embryology, and Clinical EvaluationElisabeth K.
Beahm and Robert L. Walton
Auricular Reconstruction for Microtia: Part II. Surgical TechniquesRobert L. WaltonandElisabethK. Beahm
Rhinophyma: Review and UpdateRod J. Rohrich et al.
Velopharyngeal Incompetence: A Guide for Clinical EvaluationDonnell F. Johns et al.
Basal Cell Carcinoma: An Overview of Tumor Biology and TreatmentDavid T. Netscher and Melvin Spira
Nasal Reconstruction: Forehead FlapFrederick J. Menick
Reconstruction of Acquired Scalp Defects: An Algorithmic ApproachJason E. Leedy
Further Clarification of the Nomenclature for Compound FlapsGeoffrey G. Hallock
Current Options in Head and Neck ReconstructionKeith A. Hurvitz et al.
Plastic and Reconstructive Surgery December 2007