The Unfavorable Result in Plastic Surgery_ Avoidance and Treatment (2018, Thieme)

SECTION VII RECONSTRUCTIVE SURGERY FOR THE HEAD AND NECK, BODY, AND LOWER EXTREMITY
1116
the initial lymph node dissection. Thus we have found that
the best way to identify functioning lymphatics is to perform
indocyanine green fluorescence lymphography and mapping
of vessels just before surgery while in the operating room. In­
docyanine green is a water-soluble compound that has been
widely used to assess cardiac output, hepatic function, and
ophthalmic angiography.22 When it binds to protein in the
tissue, it emits energy in the near-infrared region between
750 and 810 nm. After indocyanine green has been absorbed
by lymphatic vessels, fluorescence lymphography can detect
the near-infrared light emitted by the dye, thereby demon­
strating the path of functioning lymphatic vessels. Lymphatic
vessels that are damaged do not always uptake indocyanine
green, and therefore this method has proven beneficial to
identifying functioning lymphatics for bypass targets.
We add 10 mL of sterile water in a 25 mg vial of in­
docyanine green and then inject intradermally into each
finger or toe web of the lymphedematous limb. The extrem­
ity is then elevated to enhance uptake of the dye into the
lymphatic vessels. Soon after the injections, a Hamamatsu
Photodynamic Eye (Hamamatsu Photonics K.K.) is used to
visualize fluorescent images of lymphatic vessels, and fluo­
rescent stains are identified proximal to the injection sites.
At this point, an objective classification of the severity of
the patient’s lymphedema can be made based on indocya­
nine green lymphangiographic findings22 (Fig. 59.3). As time
passes, even after just 10 to 15 minutes, indocyanine green
dye refluxes into the superficial and subdermal lymphatics
of the limb, creating various patterns of dermal backflow,
as described by Yamamoto and Koshima.23 Once this occurs,
the linear patterns of functioning lymphatic vessels are often
no longer visible, as they are overshadowed by dermal back­
flow. Therefore we advocate for real-time lymphography at
the time of surgery, rather than the day before. With the im­
ages as a guide, we use a pen to mark the visible lymphatic
pathways and future incision sites for LVBs. Before making
each incision, we inject local anesthetic with epinephrine at
the incision site to achieve optimal hemostasis. To aid with
visually identifying lymphatic vessels during the operation,
postoperatively by a lymphedema therapist for short-stretch
bandaging. Short-stretch bandages are worn for 1 month
postoperatively and then patients are cleared to wear their
own compression garments. Within the first month, LVBs
are quite delicate, and the struggle to put on a tight gar­
ment may disturb the bypasses; therefore we advocate for
atraumatic short-stretch wrapping during this time period.
Patients who received only VLNT are cleared to wear their
compression garment from postoperative day 1 if it fits ap­
propriately. Patients who undergo LVB or who have exten­
sive scar release may see improvement almost immediately.
Patients with lymph node transfer are counseled that it may
take up to 1 year to achieve measurable results from surgery.
Intraoperative Considerations
Our preferred lymph node donor site for the surgical treat­
ment of lymphedema is the supraclavicular region.21 The
supraclavicular free flap with lymph nodes are harvested
with a reliable vascular supply based off of the transverse
cervical artery and vein (Fig. 59.1).
For women who are seeking simultaneous breast recon­
struction with lymphedema treatment, we harvest lower
abdominal flaps (deep inferior epigastric perforator [DIEP]
or superficial inferior epigastric artery [SIEA]) with vascu­
larized inguinal lymph nodes, based on the superficial cir­
cumflex iliac vessels (Fig. 59.2). For upper extremity lymph­
edema, the thoracodorsal vessels are usually selected as
recipient vessels, thus allowing the positioning of the lym­
phatic tissue into the axillary space. To treat lower extremity
lymphedema, flaps are typically placed in the groin with the
femoral vessels or their branches serving as recipient vessels.
In certain cases we have transposed flaps to the popliteal or
ankle region if lymphedema is limited to these areas.
In the past, a major barrier to LVB was the precise iden­
tification of functioning lymphatic vessels in the affected
limb. Relying on knowledge of lymphatic anatomy alone is
not sufficient for identifying optimal lymphatic vessels for
bypass, because these vessels may have been injured during
Fig. 59.1 The supraclavicular free flap with lymph nodes are har­
vested with a vascular supply based off the transverse cervical ar­
tery and vein.
Fig. 59.2 Lower abdominal flaps (deep inferior epigastric perforator
[DIEP] or superficial inferior epigastric artery [SIEA]) with vascular­
ized inguinal lymph nodes based on the superficial circumflex iliac
vessels are used for simultaneous breast reconstruction with lymph
node transfer.

1117
Chapter 59 Microsurgical Treatment of Lymphedema
the indocyanine green (ICGN) pass from the lymphatic vessel
through the anastomosis and into the venule under fluores­
cent views of the microscope (Fig. 59.4).
The use of indocyanine green allows surgeons to avoid
unnecessary, “blind” incisions; decreases operative time;
and improves outcomes. One of the main challenges of per­
forming LVB in patients before the availability of indocya­
nine green fluorescence lymphangiography was accurately
identifying functioning lymphatic vessels for bypasses. In
these patients, we made incisions arbitrarily and conducted
random exploration under the microscope to identify the
vessels. We conducted an analysis of our own outcomes in
2010 and found that the mean number of bypasses per­
formed in patients who underwent LVB without indocy­
anine green fluorescence lymphangiography was 3.5; in
contrast, the mean number of bypasses performed in pa­
tients who underwent LVB with indocyanine green lym­
phangiography was 5.6.24 From this data we affirmed that
we use a 30-gauge needle to intradermally inject isosulfan
blue dye (Lymphazurin, United States Surgical Corp.) 1 to 2
cm distal to each incision.
The entire operation is performed under a surgical mi­
croscope. Using the microscope, 2- to 3-cm incisions are
made at the predetermined sites based on lymphography
performed earlier. The subdermal region is dissected to
identify lymphatic vessels. Lymphatic vessels appear either
blue with isosulfan blue dye or clear if no dye was taken up.
Once viable lymphatic vessels are identified, they are anas­
tomosed to a similarly sized adjacent recipient venule to cre­
ate the bypass. End-to-end or end-to-side anastomoses are
performed depending on the vessel size match. Some pre­
liminary data suggest end-to-side bypass may result in bet­
ter outcomes. We use superfine microsurgical instruments
for dissection and for creating end-to-end or end-to-side
anastomoses with 11–0 or 12–0 nylon sutures with 50-μm
needles. Patency of the bypasses is confirmed by observing
Fig. 59.3 Lymphedema classification based on indocyanine green lymphangiographic findings. (a) Stage 1: many patent lymphatic ves­
sels, with minimal, patchy dermal backflow. (b) Stage 2: moderate number of patent lymphatic vessels, with segmental dermal backflow.
(c) Stage 3: few patent lymphatic vessels, with extensive dermal backflow involving the entire arm. (d) Stage 4: no patent lymphatic ves­
sels seen, with severe dermal backflow involving the entire arm and extending to the dorsum of the hand. (Reproduced from Chang DW,
Suami H, Skoracki R. A prospective analysis of 100 consecutive lymphovenous bypass cases for treatment of extremity lymphedema. Plast
Reconstr Surg 2013;132:1305–1314.)
a
b
c
d
a
b
Fig. 59.4 (a) Confirmation of patency of end-to-side lymphatic-to-venous anastomosis with isosulfan blue dye. (b) Confirmation of patency
of end-to-side lymphatic-to-venous anastomosis with indocyanine green dye.

SECTION VII RECONSTRUCTIVE SURGERY FOR THE HEAD AND NECK, BODY, AND LOWER EXTREMITY
1118
abdominal flap (Video 59.1). This mapping technique is rou­
tinely used by breast surgeons to identify sentinel nodes
during axillary node dissections. On the day before or on the
morning of surgery, the patient’s donor limbs are injected
with Tm99 to facilitate identification of sentinel nodes using
gamma probe detection.
One of the most common problems we have encoun­
tered with the supraclavicular lymph node free flap is the
development of donor site seroma or lymph leak. To avoid
this, a careful surgical dissection with loupe magnification
is needed with meticulous clipping of lymphatic vessels. The
thoracic duct is often visible within the surgical site and
must be preserved. If it is damaged, it needs to be repaired
or bypassed to a vein under the microscope (Fig. 59.5). We
leave one drain in the donor site, which is pulled when the
output is less than 30 mL for 2 consecutive days. If a donor
site seroma does develop, we perform serial aspirations in
the clinic. Sometimes, in cases of large seromas, the neck
incision is opened a few centimeters for drainage followed
by a dressing change.
Like any surgical procedure, safe harvesting of this flap
requires full knowledge of the anatomy, complete under­
standing of potential donor site morbidity, and meticulous
attention to surgical technique. First, there is a significant
anatomic variation of arterial and venous supply in this re­
gion, both in terms of the location and the size. The surgeon
must be able to adjust the approach accordingly to ensure
blood supplies to the lymph nodes are preserved during the
dissection. Furthermore, there are many important struc­
tures in the region, including the carotid artery, internal
jugular vein, phrenic nerve, and thoracic duct (Fig. 59.6).
A meticulous surgical approach is required to avoid dam­
aging any key structures and to prevent devastating com­
plications.
The complication of primary concern when harvesting
vascularized lymph nodes is causing secondary donor site
lymphedema. The supraclavicular lymph nodes mainly drain
the breasts, lungs, esophagus, and thyroid. There has been
one anecdotal personal communication that upper limb
lymphatic drainage to the supraclavicular area has been
seen on an injection study. There has been one case report
of donor site arm lymphedema after a supraclavicular flap
of a same case by two different groups of authors. It is un­
certain what type of approach was used in this particular
case, but when the photo of the case is carefully examined,
the ­donor site scar appears very different than those we
create in terms of the location as well as the extent of the
deformity.
The supraclavicular flap is a difficult flap to harvest, be­
cause there are many crucial structures in that area and the
vascular anatomy can vary. The surgeon must be very famil­
iar with the anatomy in this region. Furthermore, utmost
care must be taken with precise surgical technique in har­
vesting nodes in this area.
Another common complication of surgery is transient
numbness at the donor site in the neck or anterior thigh
from recipient vessel dissection in the groin. During the
dissection, every attempt is made to preserve superficial
indocyanine green fluorescence lymphangiography provides
a clear benefit in lymphatic mapping.
Most Common Complications
Although the number of reported cases of secondary
lymphedema from harvesting the VLNT is low, it is still
the main concern of VLNT. In a study by Viitanen et al,25 a
decrease in the lymphatic transport index at the donor leg
compared with the contralateral leg after harvesting VLNT
from the groin was noted in 6 out of 10 patients. The lym­
phatic transport index of the donor site limbs ranged from
0.2 to 12.8, whereas healthy extremities show an index of
less than 10. Only two patients had lymphatic transport in­
dices of greater than 10. None of these patients, however,
demonstrated any clinical signs of lymphedema such as
physical complaints or differences in limb measurement.25
These results were considered unexpected, because it was
previously assumed that lymph nodes included in the groin
flap were not critical for lower limb lymphatic drainage.
In a retrospective review of 26 patients who underwent
VLNT, Vignes et al26 found that 6 patients (4 upper limb, 2
lower limb) had persistent lymphedema in the donor limb
at 1 year after surgery. These patients underwent VLNT with
donor sites from both the axilla (n = 4) and groin (n = 2).
No measurement of clinical symptoms was made, however.
Pons et al27 reviewed complications after VLNT at their cen­
ter, and out of 42 patients, 1 developed swelling in the do­
nor thigh. At 2 years follow-up, the patient showed a 2-cm
enlargement in thigh circumference in the donor leg com­
pared with the contralateral one. The percentage increase
of the affected thigh was 8.5% (6.2% when considering the
entire limb).27
In patients with a history of breast cancer, the most pop­
ular VLNT donor site is the inguinal region, because abdomi­
nal flaps can be combined with inguinal lymph nodes to
simultaneously accomplish delayed autologous breast recon­
struction and VLNT. The lymph nodes in the groin have been
described as being spread over five regions: central (saphe­
nofemoral junction), superomedial, superolateral, inferome­
dial, and inferolateral.8 Because nodes that drain the lower
limb are located medially and centrally, it is more advisable
to harvest the laterally based nodes that drain the suprailiac
region.28 The superior row of nodes is supplied by the super­
ficial circumflex iliac artery, whereas the medial column is
supplied by branches off the femoral artery. Becker et al17
stressed that the superficial inguinal lymph nodes located
along the superficial circumflex iliac vein drain lymph fluid
mainly from the abdominal wall and that their procurement
did not impair lymph drainage of the lower limb.
To avoid the potential risk of precipitating donor site
lymphedema during breast reconstruction, we use reverse
mapping to identify the sentinel lymph nodes critical for
lower extremity drainage. Preoperatively, a radioactive iso­
tope is injected into the limb, which then travels to the sen­
tinel draining nodes in the groin. An intraoperative gamma
probe/Geiger meter counter is used to detect the sentinel
inguinal lymph node(s) to avoid removing them with the

1119
Chapter 59 Microsurgical Treatment of Lymphedema
time, we found that the perforator to the skin on this flap
was not particularly reliable, and we experienced some in­
stances of skin necrosis despite patent anastomoses upon
operative exploration. We no longer routinely take a skin
paddle with the supraclavicular free flap for lymph node
transfer. Instead, flaps are monitored postoperatively with an
implantable Doppler placed on the artery. Patients remain in
the hospital for 3 days for flap monitoring. If at any point, the
sensory nerves in the area, although this is not always pos­
sible. We have observed that the numbness and neuropathic
pain usually resolves within 3 to 6 months after surgery. Pa­
tients are managed expectantly, and if pain is bothersome,
we have in some instances prescribed gabapentin for symp­
tom relief.
Early in our practice we harvested supraclavicular free
flaps with a small skin paddle (Fig. 59.7). However, over
Fig. 59.5 Lymphatic duct replanted to a branch of the
internal jugular vein.
a
b
Fig. 59.6 (a) Donor site after harvesting of lymph node transfer. The internal jugular vein (IJV) is seen medially. (b) The IJV is retracted to
show phrenic nerve coursing on top of scalene muscle.
Fig. 59.7 (a) Severe right leg lymphedema with volume excess of 190% compared with the left leg. The leg is heavy, firm, fibrotic, and causes
the patient significant pain, making it difficult for him to walk. (b) A supraclavicular lymph node transfer with a skin paddle. (c) Indocyanine
green injected to the skin paddle demonstrates lymphatic vessel draining into one of the lymph nodes. (d) A flap from the supraclavicular
region, including the lymph nodes, has been transplanted to the ankle. (e) The postoperative result demonstrates a significantly softer and
smaller leg (31% reduction in excess volume).
a
e
b
c
d

SECTION VII RECONSTRUCTIVE SURGERY FOR THE HEAD AND NECK, BODY, AND LOWER EXTREMITY
1120
follow-up visit, the patient complained of a hard, 3 cm by
3 cm mass in the right neck, although her incision had com­
pletely healed. There was concern for recurrent seroma, and
another bedside incision and drainage was planned. Upon
opening the neck cavity, there was no appreciable fluid;
therefore the incision was sutured closed. Paradoxically,
the swelling appeared improved. The decision was made to
follow up with the patient in 6 months to assess the area.
Chyle Leak and Lymphocele
A 67-year-old woman with a history of uterine cancer had
a hysterectomy and lymph node dissection in 1998. Eight
years later, she developed left leg lymphedema. She under­
went VLNT to the left groin via a right supraclavicular free
flap in addition to three LVBs in the lower extremity. Her
postoperative course was complicated by a chyle leak at the
donor site. Milky fluid was observed in the neck drain at
the 1-week postoperative visit. At 2 weeks postoperatively,
she was noted to have fluctuance and mild erythema at the
supraclavicular incision. The drain was pulled, and 30 mL of
fluid was drained from the neck at the bedside. The patient
was instructed to pack the neck. By 1 month, the neck was
completely healed. Nine months later, however, the patient
returned to the clinic with concerns of a right neck mass.
On examination, a 4 cm by 4 cm fluctuant mass was pal­
pable and visible superior to the right clavicle. The decision
was made to explore the area electively in the operating
room.
The previous skin incision from the supraclavicular flap
was opened and dissection was carried down to the mass,
identifying the capsule. After en bloc removal of the mass,
the capsule was entered and apparent chylous drainage was
Doppler does not function, portending acute anastomotic
thrombosis, the patient is taken back to the operating room
for flap salvage on an emergent basis. If a patient has severe
scarring and requires a large area of skin and subcutane­
ous tissue to be excised during the recipient site prepara­
tion, we use the latissimus dorsi myocutaneous flap with
thoracic lymph nodes or lower abdominal flap with groin
lymph nodes.
Managing Unfavorable Results
and Complications in the
Treatment of Lymphedema
Seroma
A 46-year-old woman with left lower extremity primary
lymphedema who previously had LVBs at an outside insti­
tution and did not see improvement was seeking VLNT. The
surgical plan was to harvest a supraclavicular lymph node
free flap from the right neck and transplant it to the left
groin.
Postoperatively, the patient continued to have high
drain output from the right neck, which was serosanguinous
(more than 350 mL in 24-hour period). When she returned
for a 10-day follow-up appointment, it was noted that the
right neck drain was no longer functioning, and there was
a small area of induration with underlying fluid. The drain
was removed and the incision was opened at bedside and
seroma fluid was drained. The patient was then sent home
with instructions for daily packing. The wound was allowed
to heal by secondary intention. Interestingly, at the 1-month
Fig. 59.8 (a) A 58-year-old woman with lymphedema in her left arm for 1.5 years. Her left arm is 18% larger than her right arm. (b) Preopera­
tive Indocyanine green lymphangiography (stage 3). (c) Seven bypasses (anastomosis range, 0.3–0.8 mm) were performed in this patient.
At 3 months, the patient said the arm was softer and lighter than it was before lymphovenous bypass, but there has been no measurable
reduction in volume. (Reproduced from Chang DW, Suami H, Skoracki R. A prospective analysis of 100 consecutive lymphovenous bypass
cases for treatment of extremity lymphedema. Plast Reconstr Surg 2013;132:1305–1314.)
c
b
a

1121
Chapter 59 Microsurgical Treatment of Lymphedema
Conclusion
Various surgical procedures have been performed to treat
lymphedema. LVB and VLNT offer promising solutions for
the future. LVB is technically challenging and best serves
patients who have mild lymphedema with minimal perma­
nent tissue damage. Selecting functional lymphatic vessels
for bypass is critical to the success of the operation. VLNT
offers an exciting new horizon for the physiologic treat­
ment of advanced-stage lymphedema. In many cases, com­
bining both VLNT and LVB may optimize the chances for
improvement of lymphedema, because the two approaches
work via different mechanisms. More work is needed to ac­
curately evaluate lymphedema, properly define indications
for surgery, and refine the surgical procedures. Obtaining a
consensus in grading systems, reporting of outcomes, and
standardization of protocols will help to facilitate the next
phase of evolution in lymphedema treatment.
found. Pathology showed a 5 cm by 3 cm mass consistent
with a pseudocyst. The patient was left with a drain that
was eventually pulled. She had no recurrence as of 6 months
after the lymphocele excision.
Poor Outcome After Lymphovenous
Bypass
A 58-year-old woman with secondary lymphedema in
her left arm for 1.5 years presented for LVB22 (Fig. 59.8a).
Her left arm is 18% larger than her right arm. Preoperative
ICG lymphangiography demonstrates ICGN classification
stage 3 with significant dermal back flow and few visible
functioning lymphatic vessels (Fig. 59.8b) Seven bypasses
(anastomosis range, 0.3–0.8 mm) were performed in this
patient. At 3 months, the patient said the arm was softer
and lighter (Fig. 59.8c) than it was before LVB, but there has
been no measurable reduction in volume.
References
1. Rudkin G, Miller T. Lymphedema. In: Thorne C, ed. Grabb
and Smiths Plastic Surgery. 6th ed. Philadelphia, PA: Lip­
pincott, Williams & Wilkins; 2007
2. Gray JH. Studies of the regeneration of lymphatic vessels. J
Anat 1940;74(Pt 3):309–335
3. Koshima I, Kawada S, Moriguchi T, Kajiwara Y. Ultra­
structural observations of lymphatic vessels in lymph­
edema in human extremities. Plast Reconstr Surg
1996;97(2):397–405, discussion 406–407
4. Sakorafas GH, Peros G, Cataliotti L, Vlastos G. Lymphede­
ma following axillary lymph node dissection for breast
cancer. Surg Oncol 2006;15(3):153–165
5. Szuba A, Achalu R, Rockson SG. Decongestive lymphatic
therapy for patients with breast carcinoma-associated
lymphedema. A randomized, prospective study of a role
for adjunctive intermittent pneumatic compression. Can­
cer 2002;95(11):2260–2267
6. Charles R. Elephantiasis scroti. In: Latham A, English TC,
eds. A System of Treatment. J&A Churchill; 1912
7. Macey HB. A surgical procedure for lymphoedema of the
extremities; a follow-up report. J Bone Joint Surg Am 1948;
30A(2):339–346
8. Jacobson JH II, Suarez EL. Microvascular surgery. Dis Chest
1962;41:220–224
9. Sedlácek J. Lymphovenous shunt as supplementary treat­
ment of elephantiasis of lower limbs. Acta Chir Plast 1969;
11(2):157–162
10. Yamada Y. Studies on lymphatic venous anastomosis in
lymphedema. Nagoya J Med Sci 1969;32:1–21
11. O’Brien BM, Mellow CG, Khazanchi RK, Dvir E, Kumar V,
Pederson WC. Long-term results after microlymphati­
covenous anastomoses for the treatment of obstructive
lymphedema. Plast Reconstr Surg 1990;85(4):562–572
12. Shesol BF, Nakashima R, Alavi A, Hamilton RW. Success­
ful lymph node transplantation in rats, with restoration
of lymphatic function. Plast Reconstr Surg 1979;63(6):
817–823
13. Clodius L, Smith PJ, Bruna J, Serafin D. The lymphatics of
the groin flap. Ann Plast Surg 1982;9(6):447–458
14. Baker A, Kim H, Semple JL, et al. Experimental assessment
of pro-lymphangiogenic growth factors in the treatment
of post-surgical lymphedema following lymphadenecto­
my. Breast Cancer Res 2010;12(5):R70
15. Tammela T, Saaristo A, Holopainen T, et al. Therapeutic
differentiation and maturation of lymphatic vessels af­
ter lymph node dissection and transplantation. Nat Med
2007;13(12):1458–1466
16. Cheng MH, Huang JJ, Wu CW, et al. The mechanism of
vascularized lymph node transfer for lymphedema: natu­
ral lymphaticovenous drainage. Plast Reconstr Surg 2014;
133(2):192e–198e
17. Becker C, Vasile JV, Levine JL, et al. Microlymphatic sur­
gery for the treatment of iatrogenic lymphedema. Clin
Plast Surg 2012;39(4):385–398
18. Becker C, Arrive L, Saaristo A, et al. Surgical treatment
of congenital lymphedema. Clin Plast Surg 2012;39(4):
377–384
19. Yan A, Avraham T, Zampell JC, Haviv YS, Weitman E, Meh­
rara BJ. Adipose-derived stem cells promote lymphan­
giogenesis in response to VEGF-C stimulation or TGF-b1
inhibition. Future Oncol 2011;7(12):1457–1473
20. Tobbia D, Semple J, Baker A, Dumont D, Johnston M. Ex­
perimental assessment of autologous lymph node trans­
plantation as treatment of postsurgical lymphedema.
Plast Reconstr Surg 2009;124(3):777–786
21. Althubaiti GA, Crosby MA, Chang DW. Vascularized su­
praclavicular lymph node transfer for lower extremity
lymphedema treatment. Plast Reconstr Surg 2013;131(1):
133e–135e
22. Chang DW, Suami H, Skoracki R. A prospective analysis
of 100 consecutive lymphovenous bypass cases for treat­
ment of extremity lymphedema. Plast Reconstr Surg 2013;
132(5):1305–1314

SECTION VII RECONSTRUCTIVE SURGERY FOR THE HEAD AND NECK, BODY, AND LOWER EXTREMITY
1122
26. Vignes S, Blanchard M, Yannoutsos A, Arrault M. Com­
plications of autologous lymph-node transplantation for
limb lymphoedema. Eur J Vasc Endovasc Surg 2013;45(5):
516–520
27. Pons G, Masia J, Loschi P, Nardulli ML, Duch J. A case of
donor-site lymphoedema after lymph node-superficial
circumflex iliac artery perforator flap transfer. J Plast Re­
constr Aesthet Surg 2014;67(1):119–123
28. Teng E, Chang DW. Overview of surgical techniques. In:
Cheng MH, Chang DW, Patel KM, eds. Principles and Prac­
tice of Lymphedema Surgery. Elsevier; 2014
23. Yamamoto T, Yamamoto N, Doi K, et al. Indocyanine green-
enhanced lymphography for upper extremity lymphede­
ma: a novel severity staging system using dermal back­
flow patterns. Plast Reconstr Surg 2011;128(4):941–947
24. Chang DW. Lymphaticovenular bypass for lymphedema
management in breast cancer patients: a prospective study.
Plast Reconstr Surg 2010;126(3):752–758
25. Viitanen TP, Mäki MT, Seppänen MP, Suominen EA, Saaris­
to AM. Donor-site lymphatic function after microvascular
lymph node transfer. Plast Reconstr Surg 2012;130(6):
1246–1253
Discussion by Wei F. Chen
Recent advances in reconstructive microsurgery and
supermicrosurgery bring promise to treating lymph­
edema, a condition long considered untreatable. Al­
though these sophisticated procedures may be novel,
their underlying surgical principles are not. As in­
dicated in the chapter, favorable results come from
proper planning, patient and procedure selection, tech­
nical execution, and outcome assessment.
Surgical Planning
Lymphatic reconstruction commonly involves manipula­
tion of blood and lymphatic vessels that are significantly
smaller than those in routine microsurgical procedures
such as the deep inferior epigastric perforator (DIEP)
and anterolateral thigh (ALT) flaps. Supermicrosurgical
lymphaticovenular anastomosis (LVA) commonly re­
cruits lymphatic vessels ranging from 0.3 to 0.6 mm,1–3
whereas vascularized lymph node transfer (VLNT), al­
though still a microsurgical procedure, involves blood
vessels approaching 0.8 mm, the threshold of supermi­
crosurgery.3 Proper training facilitates comfortable and
efficacious performance of these procedures. Compared
to the simulation models using rat and silicone tubes,4,5
I prefer the chicken thigh model, because it provides the
simulation fidelity of a biologic model and the ease of
use and the economic advantage of a synthetic model.6
The simulation training allows the surgeon to complete
the learning curve on inanimate objects rather than on
patients. In our training program, our residents complete
20 consecutive 0.3-mm anastomoses with confirmed pa­
tency before performing the LVA in the operating room.
I agree with others that surgery is the rescue treat­
ment after failed complex decongestive therapy (CDT).
What constitutes failure of CDT, however, deserves dis­
cussion. Is it determined based on relief of symptoms,
functional improvement, or strictly volume reduction?
Is it a failure if the improvement, while notable, falls
short of the patient expectation? The decision of CDT’s
end point is not straightforward and is best made collab­
oratively with the treating therapist. In our lymphedema
center, the decision to triage a patient from therapy to
surgery is jointly made by our therapists and myself,
after having thoroughly reviewed past therapy and pre­
dicting an unfavorable prognosis of further therapy.
Patient and Procedure Selection
The chapter recommends LVA for mild disease, VLNT
in combination with LVA for moderate disease, and the
Charles procedure for severe disease. Although I agree
with the sequence of progression from the less-inva­
sive procedure to the more invasive procedures, I algo­
rithmize treatment decision based on the nature of the
disease instead of the generalized severity, based on
whether the condition is fluid- or solid-predominant
(Fig. 59D.1). Fluid-predominant disease is treated with
the drainage procedures (VLNT or LVA). Conversely,
solid-predominant disease is treated with the debulk­
ing procedures (liposuction or Charles procedure). Co-
predominant disease is treated based on patient expec­
tation—which disease component is most troublesome
to the patient. The drainage procedures “drain” and
would exert minimal impact on the pathologic adipose
deposit and fibrosis seen in the solid disease.7,8 As re­
ported previously,8 we were able to achieve more effec­
tive relief of symptoms and higher patient satisfaction
when specific disease components were targeted with
appropriate procedures. We use physical examination
in combination with bioelectrical impedance spec­
troscopy (BIS) to unequivocally differentiate fluid- and
solid-predominant states.9–11
“Primum non nocere” dictates that we consider not
just how well a procedure works but also how well it
fails. For this reason, between LVA and VLNT, I prefer the
former, because in addition to being minimally invasive,
in contrast to VLNT, LVA is free from the risks of donor
site morbidities,12–14 including donor site lymphedema.
I have not observed a notable difference in outcomes
between the two procedures when performed for fluid-
predominant disease. I especially favor LVA over VLNT
in patients with congenital lymphedema. The congenital

1123
Chapter 59 Microsurgical Treatment of Lymphedema
Fig. 59D.1 The UIHC Lymphedema Center Treatment Algorithm. Treatment decision hinges on whether the disease is fluid- or solid-
predominant, and patients are offered drainage or a debulking procedure, respectively. Lymphaticovenular anastomosis (LVA) is the
preferred drainage procedure, and liposuction is the preferred debulking procedure, whenever feasible. VLNT, Vascularized lymph
node transfer. (Algorithm courtesy Dr. Wei Chen, University of Iowa Hospitals and Clinics, Department of Surgery, Iowa City, IA.)
Patient evaluation with
1. History and physical
2. Quality of life survey
3. Bioimpedance spectroscopy
4. Indocyanine green lymphography
Fluid-predominant
disease
Solid-predominant
disease
Co-dominant
disease
Patient expectation
Maximum bulk
reduction
Maximum fluid
reduction
Healthy lymphatic
vessels?
Skin preserved?
LVA
VLNT
Suction
lipectomy +
skin excision
Suction
lipectomy
only
Charles
procedure
Offer VLNT
Patient accepting donor
site morbidities?
Suction lipectomy
Severe skin
excess?
Yes
No
No
Yes
Yes
No
Yes
No
disease is suggested by history and confirmed with in­
docyanine green (ICG) lymphography.15 In our experi­
ence, these patients may have concomitant asymptom­
atic lymphatic insufficiency, demonstrable with ICG
lymphography, in other limbs and are already predis­
posed to developing lymphedema in those limbs at a
later time.16 Additional lymphatic injury from lymph
node harvest may be the last straw needed for patho­
genesis.
My preference for LVA over VLNT ceases when the
ICG lymphography predicts a dismal chance of finding
healthy lymphatic vessels. The presence of a “diffuse”
lymphographic pattern in combination with the ab­
sence of a “linear” lymphographic pattern indicates se­
vere lymphatic injury and predicts a low probability of
locating functioning lymphatic vessels.17–20 Patients with
these lymphographic findings are offered VLNT. Those
that find the risk of donor site lymphedema unaccept­
able are still offered LVA, though with an understand­
ing that the surgery will be more challenging and less
efficacious.
The preferred lymph node donor site offered in the
chapter is the supraclavicular region because of the re­
ported morbidities associated with the groin flap.13,14,21
Although initially reported as a flap without the risk of
donor site lymphedema,22 it is now clear that the donor
site lymphedema can still occur after the supraclavicu­
lar VLNT.23 It should be noted that currently there is no
lymph node donor site that is free of donor site morbidi­
ties,24 and it is still unclear what the ideal lymph node
donor site is. I offer both the supraclavicular and groin
VLNTs. The choice between the two depends on whether
the patient has upper or lower extremity disease and
which donor site the patient finds more acceptable. I
agree that reverse lymphatic mapping is an invaluable
tool to decrease the risk of donor site lymphedema and
should be used when feasible.25 Relatively speaking, the
supraclavicular lymph node flap is more anatomically
variable and has an unreliable skin paddle. I agree that
this flap is best harvested without the skin paddle and
that the surgeon should be comfortable with head and
neck anatomy.
For patients with solid-predominant disease, I treat
with either liposuction or the Charles procedure.26,27
The Charles procedure is significantly more invasive
and morbid than liposuction. I therefore attempt to treat

SECTION VII RECONSTRUCTIVE SURGERY FOR THE HEAD AND NECK, BODY, AND LOWER EXTREMITY
1124
with liposuction whenever possible. If the skin quality is
at all preserved, I treat with liposuction. It is important
to note that elephantiasis, or fibrotic skin changes, oc­
curs in a distal-to-proximal sequence. This means that
even when the foot is elephantiatic, the leg and thigh are
still treatable with liposuction.
Technical Execution
The use of ICG lymphography to map the superficial
lymphatic vessels during the LVA procedure for stra­
tegic incision placement was discussed in the chapter.
I use the same technology to delineate the superficial
lymphatic anatomy. However, to perform an LVA, the
“V” is also important. I use an infrared vein visualiza­
tion device (VueTek Scientific) for superficial vein map­
ping (Fig. 59D.2). Knowing the precise locations of both
the lymphatic vessels and veins allows precise incision
placement and optimizes operative efficiency. This
guided approach results in creation of a higher number
of LVAs than both the “blind” anatomic approach and
the partially guided approach.1,28–30
I recommend 50-μm needles on 11–0 or 12–0 nylon
for LVA. This is important because needles with larger
diameters create excessively large holes that traumatize
the vessel edges. It is equally important that the needle
is as short as the surgeon is comfortable with. The lon­
ger the needle, the more distance it needs to traverse
through the fragile blood and lymphatic vessels, and
the more likely an accidental tear may occur. I prefer
2.5-mm-long needles. At this length the needle is long
enough to “take it in one” and still has an acceptably
short traverse length.
The standard microsurgery forceps have 0.3-mm
tips; most superfine microsurgery forceps have 0.1-mm
tips.31 I have found both insufficiently refined for smooth
handling of the submillimeter vessels. Instead, I prefer
the titanium supermicrosurgery forceps with 0.05-mm
tips (EMI Factory). These instruments are also lighter
and help to decrease hand fatigue. Another useful in­
strument is the locking needle holder. Although I prefer
a nonlocking needle holder for the regular microsurgery,
in supermicrosurgery, the locking mechanism relieves
the surgeon from having to constantly maintain grip­
ping pressure, allowing him or her to focus on precision
suture placement with a relaxed hand.
Many anastomotic configurations have been de­
scribed.32–36 To date, there is no good evidence showing
the superiority of one configuration over the others. My
preference is to perform end-to-end anastomosis when
the lymphatic and vein calibers are matching. In a vein–
lymphatic size mismatch in which the vein is larger,
end-to-side anastomosis (lymphatic end to the side of
the vein) works well. It is uncommon for the lymphatic
to be larger than the vein, but it can be resolved with
side-to-end or side-to-side anastomosis. For a quantity
mismatch in which there are more lymphatic vessels
than veins, the “octopus” technique allows pooling of
the drainage from multiple lymphatic vessels into a sin­
gle vein36 (Fig. 59.3). This technique has the advantages
of allowing for the creation of LVA when the standard
technique is not feasible and of creating a higher num­
ber of drainage pathways in a single LVA (Video 59D.1).
A common question is regarding how many stitches
to place. I determine the end point of suture place­
ment by when the anastomotic leak stops. In general,
a
b
c
Fig. 59D.2 Infrared vein localization. (a) The device is head-mounted, freeing up the surgeon’s hands. (b) The lymphatic mapping
had already been done, and the lymphatic vessels were marked with solid black lines. (c) The veins were indicated with dotted lines.

1125
Chapter 59 Microsurgical Treatment of Lymphedema
Fig. 59D.3 Dissection of the lymphatics and vein. (a) Multiple lymphatic vessels and a single vein are selected for creation of the
“octopus” lymphaticovenular anastomosis. (b) Division of the lymphatics and vein. The lymphatic vessels and the vein have been
divided. All have been skeletonized and aligned in preparation for anastomosis. (c) Creation of the anastomosis. A 12–0 nylon su­
ture has been placed transluminally on the vein, and through the adventitia only on the lymphatic vessels. Placing the suture back
transluminally through the vein will complete the suture loop and intussuscept the lymphatic vessels into the vein. (d) Completion
of the “octopus.” The lymphatic vessels have been intussuscepted into the vein. Additional reinforcing sutures have been placed.
Washout is clearly identified in the vein in this image, indicative of the lymphatic pressure overcoming the venous pressure. (Re­
produced from Chen W, Yamamoto T, Fisher M, et al. The octopus lymphaticovenular anastomosis: evolving beyond the standard
supermicrosurgical technique. J Reconstr Microsurg 2015;31:450–457.)
Lymphatic
vessels
Vein
a
b
c
d

SECTION VII RECONSTRUCTIVE SURGERY FOR THE HEAD AND NECK, BODY, AND LOWER EXTREMITY
1126
lymphatic vessels ranging 0.3 to 0.5 mm require only
three to four sutures to achieve a leak-proof anastomo­
sis. It is important to achieve a leak-proof anastomosis,
because it allows the lymphatic pressure to build up in­
stead of decompressing through a leak, “washing out”
the blood in the vein. Both a washout sign and a “back­
flow” sign (venous blood flowing into the lymphatic ves­
sel in a retrograde fashion) confirm anastomotic patency
(Fig. 59D.4). Obviously, the washout is more desirable.
The washout sign can occur in a delayed fashion a few
minutes after the anastomosis. This is especially likely
when compromised lymphatic vessels are used. It takes
time for a weak lymphatic vessel to build up sufficient
pressure to wash out the venous blood. Occasionally, a
backflow can be converted to a washout by compressing
the area distal to the incision. This simulates the effect
of postoperative compression.
I agree that recruiting healthy lymphatic vessels
for LVA construction is necessary. It is, however, impor­
tant to note that the lymphatic vessels are not either
“good” or “bad.” Their pathologic deterioration exists
in several gradations. As described by Mihara et al,19
the lymphatic vessels can be classified into normal, ec­
tatic, contracted, and sclerotic, in order of worsening
pathology (Fig. 59D.5). In my experience, functioning
LVA, as evidenced by the washout sign after anastomo­
sis, can be built with lymphatic vessels of all types ex­
cept for the sclerotic type. The sclerotic vessels can be
easily distinguished by their opaque appearance. When
cut, they demonstrate a pinhole lumen and absence of
Fig. 59D.4 The washout sign. (a) The blood in the vein (right side of the picture) has been washed out by the influx of the dye-
colored lymph from the lymphatic vessel (left side of the picture). (b) The backflow sign. This is a classic size mismatch situation.
Despite having performed an end-to-side anastomosis (end of the lymphatic vessel to side of the vein) in anticipation of a relatively
high venous pressure, the venous pressure still overcame the lymphatic pressure and the backflow occurred.
a
b
Fig. 59D.5 Pathologic deterioration of the lymphatic vessels. (a–d) The sequence is characterized by an initial lymphatic vessel
dilation from an increase in intraluminal pressure and subsequent progressive proliferation of smooth muscle cell and collagen
deposition in the lymphatic vessel wall. The normal and the ectasia types (a,b) are ideal for lymphaticovenular anastomosis con­
struction. (Reproduced from Mihara M, Hara H, Hayashi Y, et al. Pathological steps of cancer-related lymphedema: histological
changes in the collecting lymphatic vessels after lymphadenectomy. PLoS One 2012;7(7):e41126.)
a
b
c
d
Normal Type (Step 0)
Ectasia Type (Step 1)
Contraction Type (Step 2)
Sclerosis Type (Step 3)

1127
Chapter 59 Microsurgical Treatment of Lymphedema
lymph egress. Although the quality of the lymphatic ves­
sel is important, the quantity of LVA also contributes to
the overall efficacy of the procedure. When the qual­
ity is lacking, I attempt to compensate with quantity.
At this time, the minimum and maximum number of
LVAs needed per case are undetermined. In practice, I
routinely make five to eight incisions per limb, and all of
the lymphatic vessels found except for the sclerotic type
are used for LVA construction using various anastomotic
techniques to maximize the total drainage pathways.
Outcome Assessment
The success of lymphedema reconstruction is commonly
assessed by the degree of volume reduction.1,24,29,37–40
This is problematic because volume reduction is not
the only form of improvement. The relief of pain, the
increase in exercise tolerance, and the increase in hand
functions, for example, are qualitative improvements
not reflected by volume changes.41 I believe this ex­
plains the observation in the chapter of 96% qualitative
improvement and only 74% quantitative improvement
after LVA. Also noted was having observed significantly
more volume reduction in early disease. This is likely
because the early disease is fluid-predominant and
more responsive to drainage procedures. Relying on
volume reduction as the sole indicator of operative suc­
cess will likely lead to inaccurate conclusion of opera­
tive outcome. The outcome assessment should ideally
be multimodal, because no single tracking modality
tells the whole story.41 In our lymphedema center, we
evaluate all patients with the following:
1. A standardized history and physical examination,
including circumference measurement.
2. A validated quality-of-life (QOL) assessment that
establishes baseline functional level and identifies
domains of QOL that are adversely affected.41,42
3. Bioelectrical impedance spectroscopy (BIS) mea­
surement that quantifies the amount of limb fluid
by measuring the limb’s resistance to the flow of
electricity.9–11 High limb fluid content suggests fluid
predominance. Conversely, low limb fluid content
in a bulky limb suggests solid predominance.
4. ICG lymphography (Fig. 59D.6). In addition to being
a sensitive diagnostic modality,43 ICG lymphogra­
phy provides lymphographic staging that reflects
underlying lymphatic injury.17,18
Each of the mentioned assessment modalities evalu­
ates aspects of postoperative changes not evaluable by
the other modalities. Performing them preoperatively
Fig. 59D.6 (a) This patient had acquired Campisi IV, solid-predominant lymphedema of the left lower extremity. Preoperative in­
docyanine green (ICG) lymphography (left) demonstrated a pathologic “diffuse” pattern in the leg and “stardust” pattern in the
leg and thigh. The same patient at 1 year after lymphaticovenular anastomosis (LVA) demonstrated complete resolution of the
“diffuse” pattern and partial resolution of the “stardust” pattern, as well as emergence of a “linear” pattern in the foot (right).
(b) This patient had acquired Campisi III, fluid-predominant lymphedema of the right lower extremity. Preoperative ICG lymphogra­
phy (left) demonstrated a pathologic “stardust” pattern in the leg. Partial resolution of the pathologic pattern can be seen in the leg
at 1 year after LVA (right). (Reproduced from Chen W, Zhao H, Yamamoto T, et al. Indocyanine green lymphographic evidence of sur­
gical efficacy following microsurgical and supermicrosurgical lymphedema reconstruction. J Reconstr Microsurgery, in press, 2016.)
a
b

SECTION VII RECONSTRUCTIVE SURGERY FOR THE HEAD AND NECK, BODY, AND LOWER EXTREMITY
1128
and at regular intervals after the surgery allows sur­
geons to track progress.
Conclusion
Despite lymphedema reconstruction being a young and
evolving field, surgeons have come a long way from a
time when lymphedema is considered untreatable.
Surgeons will always have their own ideas and ways
to perform surgery; however, the underlying principle
is always the same: the favorable outcome is obtained
and the unfavorable result avoided by proper planning,
patient and procedure selection, technical execution,
and outcome assessment.
References
1. Koshima I, Inagawa K, Urushibara K, Moriguchi T. Su­
permicrosurgical lymphaticovenular anastomosis for
the treatment of lymphedema in the upper extremi­
ties. J Reconstr Microsurg 2000;16(6):437–442
2. Cheng MH, Chen SC, Henry SL, Tan BK, Lin MC, Huang
JJ. Vascularized groin lymph node flap transfer for
postmastectomy upper limb lymphedema: flap anat­
omy, recipient sites, and outcomes. Plast Reconstr
Surg 2013;131(6):1286–1298
3. Masia J, Olivares L, Koshima I, et al. Barcelona con­
sensus on supermicrosurgery. J Reconstr Microsurg
2014;30(1):53–58
4. Yamashita S, Sugiyama N, Hasegawa K, Namba Y,
Kimata Y. A novel model for supermicrosurgery train­
ing: the superficial inferior epigastric artery flap in
rats. J Reconstr Microsurg 2008;24(8):537–543
5. Matsumura N, Horie Y, Shibata T, Kubo M, Hayashi N,
Endo S. Basic training model for supermicrosurgery: a
novel practice card model. J Reconstr Microsurg 2011;
27(6):377–382
6. Chen WF, Eid A, Yamamoto T, Keith J, Nimmons GL,
Lawrence WT. A novel supermicrosurgery training
model: the chicken thigh. J Plast Reconstr Aesthet Surg
2014;67(7):973–978
7. Damstra RJ, Voesten HGJ, van Schelven WD, van der
Lei B. Lymphatic venous anastomosis (LVA) for treat­
ment of secondary arm lymphedema. A prospec­
tive study of 11 LVA procedures in 10 patients with
breast cancer related lymphedema and a critical re­
view of the literature. Breast Cancer Res Treat 2009;
113(2):199–206
8. Granzow JW, Soderberg JM, Kaji AH, Dauphine C. An
effective system of surgical treatment of lymphede­
ma. Ann Surg Oncol 2014;21(4):1189–1194
9. Ridner SH, Bonner CM, Doersam JK, Rhoten BA,
Schultze B, Dietrich MS. Bioelectrical impedance
self-measurement protocol development and daily
variation between healthy volunteers and breast can­
cer survivors with lymphedema. Lymphat Res Biol
2014;12(1):2–9
10. Erdogan Iyigun Z, Selamoglu D, Alco G, et al; Preliminary
Results of the Florence Nightingale Breast Study Group.
Bioelectrical impedance for detecting and monitoring
lymphedema in patients with breast cancer. Preliminary
results of the florence nightingale breast study group.
Lymphat Res Biol 2015;13(1):40–45
11. Bundred NJ, Stockton C, Keeley V, et al; Investigators
of BEA/PLACE studies. Comparison of multi-frequen­
cy bioimpedance with perometry for the early detec­
tion and intervention of lymphoedema after axillary
node clearance for breast cancer. Breast Cancer Res
Treat 2015;151(1):121–129
12. Viitanen TP, Mäki MT, Seppänen MP, Suominen EA,
Saaristo AM. Donor-site lymphatic function after mi­
crovascular lymph node transfer. Plast Reconstr Surg
2012;130(6):1246–1253
13. Pons G, Masia J. Pietro L, Nardulli ML, Duch J. A
case of donor-site lymphoedema after lymph node-­
superficial circumflex iliac artery perforator flap
transfer. Br J Plast Surg 2014;67:119–123
14. Vignes S, Blanchard M, Yannoutsos A, Arrault M. Com­
plications of autologous lymph-node transplantation
for limb lymphoedema. Eur J Vasc Endovasc Surg 2013;
45(5):516–520
15. Yamamoto T, Yoshimatsu H, Narushima M, Yamamo­
to N, Hayashi A, Koshima I. Indocyanine green lym­
phography findings in primary leg lymphedema. Eur
J Vasc Endovasc Surg 2015;49(1):95–102
16. Chen WF, Bowen M. Asymptomatic lymphatic insuffi­
ciency—an unrecognized entity. Unpublished raw data.
17. Yamamoto T, Yamamoto N, Doi K, et al. Indocyanine
green-enhanced lymphography for upper extremity
lymphedema: a novel severity staging system using
dermal backflow patterns. Plast Reconstr Surg 2011;
128(4):941–947
18. Yamamoto T, Matsuda N, Doi K, et al. The earliest finding
of indocyanine green lymphography in asymp­tomatic
limbs of lower extremity lymphedema patients second­
ary to cancer treatment: the modified dermal backflow
stage and concept of subclinical lymphedema. Plast Re­
constr Surg 2011;128(4):314e–321e
19. Mihara M, Hara H, Hayashi Y, et al. Pathological steps
of cancer-related lymphedema: histological changes
in the collecting lymphatic vessels after lymphad­
enectomy. PLoS One 2012;7(7):e41126
20. Hara H, Mihara M, Seki Y, Todokoro T, Iida T, Koshima
I. Comparison of indocyanine green lymphographic
findings with the conditions of collecting lymphatic
vessels of limbs in patients with lymphedema. Plast
Reconstr Surg 2013;132(6):1612–1618
21. Althubaiti GA, Crosby MA, Chang DW. Vascularized
supraclavicular lymph node transfer for lower ex­
tremity lymphedema treatment. Plast Reconstr Surg
2013;131(1):133e–135e
22. Sapountzis S, Singhal D, Rashid A, Ciudad P, Meo D,
Chen HC. Lymph node flap based on the right trans­
verse cervical artery as a donor site for lymph node
transfer. Ann Plast Surg 2014;73(4):398–401

1129
Chapter 59 Microsurgical Treatment of Lymphedema
23. Lee M, McClure E, Reinertsen E, Granzow JW. Lymph­
edema of the upper extremity following supraclavic­
ular lymph node harvest. Plast Reconstr Surg 2015;
135(6):1079e–1082e
24. Scaglioni MF, Arvanitakis M, Chen YC, Giovanoli P,
Chia-Shen Yang J, Chang EI. Comprehensive review of
vascularized lymph node transfers for lymphedema:
outcomes and complications. Microsurgery 2016:1–8
25. Dayan JH, Dayan E, Smith ML. Reverse lymphatic map­
ping: a new technique for maximizing safety in vascu­
larized lymph node transfer. Plast Reconstr Surg 2015;
135(1):277–285
26. Brorson H. Liposuction in lymphedema treatment. J
Reconstr Microsurg 2016;32(1):56–65
27. Karri V, Yang MC, Lee IJ, et al. Optimizing outcome
of charles procedure for chronic lower extremity
lymphoedema. Ann Plast Surg 2011;66(4):393–402
28. Koshima I, Nanba Y, Tsutsui T, Takahashi Y, Itoh S, Fu­
jitsu M. Minimal invasive lymphaticovenular anasto­
mosis under local anesthesia for leg lymphedema: is
it effective for stage III and IV? Ann Plast Surg 2004;
53(3):261–266
29. Nagase T, Gonda K, Inoue K, et al. Treatment of lymph­
edema with lymphaticovenular anastomoses. Int J
Clin Oncol 2005;10(5):304–310
30. Chang DW, Suami H, Skoracki R. A prospective analy­
sis of 100 consecutive lymphovenous bypass cases for
treatment of extremity lymphedema. Plast Reconstr
Surg 2013;132(5):1305–1314
31. Mihara M, Hayashi Y, Iida T, Narushima M, Koshima
I. Instruments for supermicrosurgery in Japan. Plast
Reconstr Surg 2012;129(2):404e–406e
32. Yamamoto T, Narushima M, Kikuchi K, et al. Lambda-
shaped anastomosis with intravascular stenting meth­
od for safe and effective lymphaticovenular anastomo­
sis. Plast Reconstr Surg 2011;127(5):1987–1992
33. Yamamoto T, Yoshimatsu H, Narushima M, et al. A
modified side-to-end lymphaticovenular anastomo­
sis. Microsurgery 2013;33(2):130–133
34. Yamamoto T, Yoshimatsu H, Narushima M, et al. Se­
quential anastomosis for lymphatic supermicrosur­
gery: multiple lymphaticovenular anastomoses on 1
venule. Ann Plast Surg 2014;73(1):46–49
35. Ayestaray B, Bekara F. π-shaped lymphaticovenular
anastomosis: the venous flow sparing technique for
the treatment of peripheral lymphedema. J Reconstr
Microsurg 2014;30(8):551–560
36. Yamamoto T, Chen W, Yamamoto N, Yoshimatsu H,
Tashiro K, Koshima I. Technical simplification of the
supermicrosurgical side-to-end lymphaticovenular
anastomosis using the parachute technique. Micro­
surgery 2015;35(2):129–134
37. Chen WF, Yamamoto T, Fisher M, Liao J, Carr J. The
“octopus” lymphaticovenular anastomosis: evolving
beyond the standard supermicrosurgical technique. J
Reconstr Microsurg 2015;31(6):450–457
38. Becker C, Vasile JV, Levine JL, et al. Microlymphatic
surgery for the treatment of iatrogenic lymphedema.
Clin Plast Surg 2012;39(4):385–398
39. Brorson H, Höijer P. Standardised measurements used
to order compression garments can be used to calcu­
late arm volumes to evaluate lymphoedema treat­
ment. J Plast Surg Hand Surg 2012;46(6):410–415
40. Armer JM. The problem of post-breast cancer lymph­
edema: impact and measurement issues. Cancer In­
vest 2005;23(1):76–83
41. Patel KM, Lin CY, Cheng MH. A prospective evaluation
of lymphedema-specific quality-of-life outcomes fol­
lowing vascularized lymph node transfer. Ann Surg
Oncol 2015;22(7):2424–2430
42. Keeley V, Crooks S, Locke J, Veigas D. A quality of life
measure for limb lymphoedema (LYMQOL). J Lymph­
oedema 2010;5(1):26–37
43. Mihara M, Hara H, Araki J, et al. Indocyanine green
(ICG) lymphography is superior to lymphoscintigra­
phy for diagnostic imaging of early lymphedema of
the upper limbs. PLoS One 2012;7(6):e38182

(Trang không có chữ)

The Hand and Upper
Extremity
S E C T I O N V I I I

Bronze Hands. NAM 15105, 15108, are considered to belong to a bronze philosopher statue,
dated around 230 B.C. or a little later, found in a 1st century B.C. shipwreck at the Island of Antikythera.
(National Archaeological Museum, Athens, Greece. ©Hellenic Ministry of Culture
and Sports/Archeological Receipts Fund. Photo: Kostas Xenikakis.)

1133
Congenital hand anomalies represent a spectrum of
­conditions that many plastic surgeons treat. They range in
­complexity and scope from simple extra digits to far more
complex and challenging conditions. It is imperative that
an accurate diagnosis be made, because it has implications
on treatment and appropriate genetic counseling. Impor­
tant considerations include appropriateness of surgery,
timing of surgery, associated medical problems, and type of
procedure to be performed. If the first procedure is not per­
formed adequately it can be very difficult, if not impossible,
to obtain a satisfactory result even with revision surgery.
Avoiding Unfavorable Results
and Complications in Congenital
Hand Surgery
The ideal outcome after surgery would obviously be nor­
mal function and appearance. Although this is obtainable
in some cases, many times improvement in function alone
has to be the realistic goal. If not performed properly, these
surgeries can easily result in a deleterious outcome causing
decreased function in addition to worsening of the overall
cosmetic appearance. This can be of considerable clinical
importance, because additional surgeries (revisions) can
cause significant psychological distress for the patient (and
his or her family) and result in functional outcomes that
may be inferior to those that could have been obtained
initially. Each congenital hand difference case has its own
unique considerations; therefore thoughtful preoperative
assessment and planning is of utmost importance to obtain
the optimal result.
Appropriate Timing of Surgical
Intervention
Operations for congenital hand differences are rarely a
medical emergency. The pediatric upper extremity under­
goes rapid changes in size during the first few years of life,
and as the structures get larger they are easier to identify
and reconstruct.1 As such, allowing the patient time to
grow before performing the initial surgical correction al­
lows for the opportunity for greater precision in the execu­
tion of the surgical treatment plan. Notable exceptions to
this are cases of neonatal compartment syndrome and im­
pending limb loss (vascular compromise) associated with
amniotic disruption sequence.2,3
An additional aspect of surgical timing to be cognizant
of is the recognition that, because of the constant anatomic
changes resulting from growth and development, even if a

Congenital Hand Deformities
Andreas D. Weber, Lesley C. Butler, and Scott N. Oishi
CHAPTER
60
Summary Box
Unfavorable Results and Complications
in Congenital Hand Surgery
•• Ischemia
•• Joint instability
•• Axil deformities
•• Nail deformities
•• Unrealistic expectations
•• Hypertrophic scarring
•• Contractures
•• Web creep

SECTION VIII THE HAND AND UPPER EXTREMITY
1134
ever, by 9 months, as the child begins the exploration of
the environment in earnest, the thumb gains its indepen­
dence. Mobility from the palm begins commensurate with
new functional demands. At 12 months, the thumb has be­
come a critical portion of the hand and serves a multitude
of functions.5 Therefore a clear assessment about strength
and mobility cannot accurately be achieved before reaching
this level of cognitive and motor development. This can be
very important, because determination of prehensile grasp
patterns is mandatory in determining proper surgical treat­
ment for these patients.
Aspects Associated with Anesthesiology
Another aspect in determining the appropriate timing for
a surgical intervention is to weigh the risks and benefits
associated with the exposure to general anesthetics in the
infant. In the past, anesthesia was avoided in very young
patients (younger than 6 months), if possible, because of
perceived increased risks related to the metabolism of the
anesthetic. In addition, recent preclinical and animal trials
have yielded data that demonstrate a possible correlation
with decreases in neurodevelopment when multiple anes­
thetics are given at a young age; however, conclusive stud­
ies are still ongoing. The exact age that this risk decreases is
still to be determined, but presumably the older the patient
the less the risk.6,7
Additional Considerations
Significant external pressure can be placed on the surgeon
by the patient’s family to perform early surgical interven­
tion because of the aesthetic appearance, but this should
not influence timing of surgery. An extensive discussion
with the parents usually is sufficient to allay concerns and
delay surgery until a more appropriate time.
Also importantly, the hand surgeon must be aware of the
comorbidities that are often present in patients with con­
genital hand differences. Such comorbidities (e.g., Fanconi
anemia, thrombocytopenia absent radius) often have the po­
tential to significantly affect treatment in myriad ways and
are discussed in the following section.
Associated Anomalies
Completing a detailed medical history, comprehensive re­
view of systems, and thorough clinical examination are a
critical first step in the evaluation of a child with a complex
hand anomaly. Children with hand anomalies can also have
serious visceral and systemic anomalies that can take pre­
cedence over the treatment for anomalies of the hand. As
well, medical conditions associated with complex hand dif­
ferences are often not directly related to the affected limb.
As such, they can easily be missed or “glossed over” if not
properly evaluated, with the resultant consequences of this
error leading to severe morbidity or even mortality.
result looks aesthetically flawless immediately after the pro­
cedure, there is (unfortunately) no guarantee that this de­
gree of perfection will persist ad infinitum. An example of
this type of “postponed surgical complication” is clearly evi­
dent when, for example, damage is incurred to the growth
plates or neurovascular structures. The underlying damage,
though not immediately seen postoperatively, can result in
angular deformities or undergrowth of a bone or part of the
hand months or years later. Conversely, leaving osteogenic
tissue after removing extra bones can result in regrowing
of part of the removed bone with functional and cosmetic
impairment.
In summary, important considerations regarding surgi­
cal timing include the following:
1. Anatomy of the growing hand
2. Physiology of wound healing
3. Development of functional skills
4. Aspects associated with anesthesiology
Anatomy of the Growing Hand
A defined period of rapid hand growth occurs from birth
through the first few years of life.1 Therefore delaying sur­
gery, if possible, until the patient is 2 to 3 years old makes
the procedure technically easier and yields a result that
may be at decreased risk to deteriorate over time (i.e., at
skeletal maturity).
Physiology of Wound Healing
As discussed in Tachdjian’s Pediatric Orthopaedics, “another
unique characteristic of a child’s hand is the increased risk
for a disproportionately thick scar created in infants’ skin
at the site of hand incisions . . . Fetal skin has the remark­
able characteristic of healing without scarring . . . soon af­
ter birth the infant skin may make an unexpectedly robust,
hypertrophic scar at the site of surgical incisions. This is
in contrast to the more favorable nature of deep scarring
around the tendons and joints of older children. This super­
ficial hypertrophy of the scar in the skin incision is particu­
larly a problem for the hand surgeon because the hand’s
great mobility demands absence of scar in areas of the skin
that change length with hand motion.”4
Development of Functional Skills
The developmental stage of the child is also an important
factor in achieving a good result after surgery. This is par­
ticularly true when a certain level of cognitive develop­
ment is imperative to successful treatment of a condition.
A prime example of this “developmental requirement” is
in patients with a hypoplastic thumb. Throughout the first 3
months of life, the thumb is primarily in a palmar position.
At this stage, the thumb is merely an accessory appendage
and functions more as a pacifier than anything else. How­

1135
Chapter 60 Congenital Hand Deformities
Decision for Incision
The design of skin incisions and flaps is paramount and
more critical in a child’s hand than in an adult’s because of
the degree of scarring that can occur. It is well known that
tension on a scar leads to scar hypertrophy and can result
in restricted movement. Fig. 60.1 illustrates the fundamen­
tal principles for the correct design of incisions along fin­
gers.4 The incisions are based on the axis of rotation of the
interphalangeal (IP) joints. The connection of these lines
results in three digital diamonds. These diamond-shaped
fields contain skin that changes length with finger motion.
Any longitudinal scar located within these diamonds will
therefore be under significant tension, possibly causing
scar hypertrophy and contracture.4
In particular, scars placed in the first web space lead to
restricted movement of the thumb. The great mobility of
the human thumb is conferred by a specific geometric ar­
rangement of the first web space. The volume of the tissue
in the first web space is a tetrahedron covered with supple
palmar and dorsal triangles of skin. As such, no suture line
should cross the leading edge of the first web space. This
is particularly important in pollicization operations. In this
instance, we recommend using a modified racquet incision9
(Fig. 60.2), so that the sutures will align on the dorsal tetra­
hedron border of the first web space.
Table 60.1 Congenital upper limb
differences associated with other
malformations
Condition
Malformations
Radial
longitudinal
deficiency
Anomalies of the heart, kidneys,
spine, hematopoietic system,
gastrointestinal tract
Ulnar
longitudinal
deficiency
Musculoskeletal anomalies
such as proximal femoral focal
deficiency, fibular and tibial
ray deficiency, phocomelia,
scoliosis, clubfeet, absent patellae,
congenital dislocation of the hip,
coxa vara, or spina bifida8
Symbrachydactyly
Poland’s syndrome
Amniotic band
syndrome
Constriction band of the leg
Fig. 60.1 The illustrative principles underlying the correct design of finger incisions. (a) The incisions are based on the axis of rotation for
the interphalangeal joints. (b) The palmar view shows the three digital diamonds containing skin that changes length with finger motion.
Any longitudinal scar located within these diamonds will hypertrophy and contract. (c) In full digital flexion, points 1, 2, 3, and 4 touch, as
do the side limbs of the diamonds, to maintain length. (Reproduced from Herring JA, ed. Tachdjian’s Pediatric Orthopaedics. 4th ed. Phila­
delphia, PA: Elsevier Saunders; 2008:490.)
a
b
c
Fig. 60.2 Modified racquet incision as described by Marybeth Ezaki, MD, and Peter Carter, MD. (a) Volar. (b) Radial. (c) Dorsal.
a
b
c
Examples of congenital upper limb differences that are
associated with other malformations of the musculoskeletal
system are listed in Table 60.1.

SECTION VIII THE HAND AND UPPER EXTREMITY
1136
the proximal interphalangeal (PIP) joint, because this pro­
vides adequate length to prevent an incision line in the web
space. If undue tension exists at the incision line, this even­
tually will retract into the web space and possibly cause a
web space contracture.
Further considerations in syndactyly reconstruction in­
clude designing appropriate opposing zig zag flaps to avoid
linear scars that would otherwise result in contracture (Fig.
60.4). If not designed correctly, this can lead to increased
scar contracture and the need for more skin grafting than
usual, which can affect donor site location. An exception to
this may be in the complex syndactyly reconstruction in an
Apert’s patient with no significant IP motion when straight-
line separation may be appropriate (Fig. 60.5, Video 60.1).
Skin Grafts
Skin grafts are required when there is not enough local skin
to cover a defect. The use of skin grafts requires both func­
tional and cosmetic assessment of the recipient site. Ide­
ally, a skin graft should have a similar color scheme to the
recipient site, minimal contracture during healing, mini­
mal hair growth, and ability to grow with the patient. We
have found that full-thickness skin grafts obtained from the
volar wrist crease meet these requirements best; however,
if a substantial amount of graft is required (as is sometimes
the case in Apert’s or complex polysyndactyly reconstruc­
tion), an alternate site must be used. In particular, the vo­
lar-ulnar side of the wrist is ideal because it offers hairless
skin, even in the hairiest male, that many other donor sites
do not provide.
The area is very forgiving in young patients because it
can be closed with minimal risk of wrist contracture. In ad­
dition, this area can be repeatedly used in cases of staged
reconstruction or revisions.
An additional benefit is that the entire area is covered
by the same dressing or cast throughout the healing pe­
riod. As in any full-thickness skin graft, proper preparation
is required, as is minimal tension when inserting. Loss of
graft can result in web creep, contracture, or hypertrophic
­scarring.
Another example of the need for careful incision plan­
ning involves syndactyly reconstruction, which is one of the
most common congenital hand differences encountered. The
success of any syndactyly reconstruction is largely depen­
dent upon the design of the skin incision and flaps. In the
majority of cases a dorsal flap is used to reconstruct the web
space. There are many designs for the dorsal flap, but the
primary objectives are for a tension-free closure and appro­
priately placed scars. We prefer a standard dorsal flap design,
which is designed as shown in Fig. 60.3. It is important to
extend the tip of the flap at least two thirds of the way to
Fig. 60.3 Flap design for syndactyly. The dorsal flap is designed first. To define the starting point of the flap, the metacarpal heads are identi­
fied by flexing the metacarpophalangeal joints. (a) It is important to extend the tip of the flap at least two thirds of the way to the proximal
interphalangeal (PIP) joint to obtain a flap of adequate length to prevent an incision line in the web space. (b) Then, the dorsal incision is
designed. (c) Finally, the palmar incision is made.
a
b
c
Fig. 60.4 A linear scar resulted in a flexion contracture
Fig. 60.5 Straight-line separation may be appropriate for patients
with Apert’s syndrome because of the lack of significant interpha­
langeal motion.

1137
Chapter 60 Congenital Hand Deformities
Skin flap necrosis can easily occur in this patient pop­
ulation if care is not taken during the design and mobili­
zation of the flaps. It is important to keep the base of the
flaps as wide as possible with mobilization of the flap in
the proper plane to provide adequate perfusion all the way
to the tip of the flap. Failure to perform this correctly will
lead to flap death with secondary healing of the area that
can cause severe contracture. This is especially important in
cases of syndactyly reconstruction where flap loss can result
in web creep and/or finger contracture which can be severe
(Fig. 60.6). It is also important in cases of hypoplastic thumb
reconstruction when the first web space reconstruction is
performed using z-plasties. If these flaps do not survive, the
result can be a severely contracted first web space which can
be very difficult to treat by revision surgery.
Joint Instability
Joint instability can be a significant component of many con­
genital hand difference conditions. Depending on whether
this involves a finger or a thumb as well as which specific
joint is involved has significant implications on hand func­
tion. If untreated, this can lead to joint degenerative changes
along with pain. This complication is most often seen in the
setting of complex syndactylies, preaxial (radial) polydacty­
lies, and hypoplastic thumb reconstruction.
The bony synostosis that exists in patients with complex
syndactylies usually is between the tufts of the distal pha­
langes. Tip separation should be from proximal to distal after
identification and preservation of the collateral ligaments of
the DIP joints. If done in the opposite direction, injury to the
collateral ligaments can occur, resulting in gross instability
of the DIP joint that is not easily repaired (Fig. 60.7).
In preaxial (radial) polydactyly, specifically in the most
common variation (Wassel type IV), reconstruction of liga­
ments and tendons is especially critical, unlike a simple ex­
cision performed in postaxial (ulnar) type B polydactyly. It
is imperative that the radial collateral ligament of the meta­
carpophalangeal (MCP) joint is transferred from the thumb
to be excised to the thumb that is being reconstructed. Dis­
tally, the insertion is taken down with a strip of periosteum
to facilitate reattachment, while proximally care must be
taken during the chondroplasty procedure of the metacarpal
head to avoid detachment of the origin. Failure to do this can
Ischemia
After release of a tourniquet, an underperfused digit can
be encountered, especially when extensive surgery is per­
formed involving either stretching or mobilization of the
vascular pedicle. There are several potential causes of hy­
poperfusion that do not involve transection of the vessels:
•• Vasospasm
•• Tension on the vessel
•• Kinking of the vessel
The initial maneuvers to treat this should always be non­
operative and follow a logical sequence that includes the
following:
1. Warming of the digit
2. Removal of tight dressings, including Steri-Strips
3. Injection of local anesthetic in the nerve distribution to
provide sympathetic blockade
4. Removal of pins that may be causing severe stretching
of the vessels
5. Removal of sutures and exploration of the vessels to
check for kinking or injury
Especially in children, this scenario can be seen in cases
of camptodactyly when the flexion contracture of the PIP
joint is released and the joint is pinned in extension. If
warming of the digit does not completely restore perfusion,
usually removal of the pin is required with stabilization of
the joint in a slightly flexed position. Although syndactyly
reconstruction is thought of as a very innocuous procedure,
ischemia can occasionally occur and must be looked for.
This can be the result of tight sutures and/or a tight dress­
ing (i.e., Steri-Strips). To avoid the latter it is imperative to
avoid placing the strips circumferentially around the finger
at any point, because this may cause a constriction. Index
pollicization procedures can also result in hypoperfusion
of the transposed digit when the tourniquet is released.
If the conservative measures fail to restore perfusion, it is
mandatory to release the sutures and explore the pedicle,
because it may be kinked or twisted, resulting in the hypo­
perfusion. No matter what the cause, the surgeon must be
convinced it has been adequately addressed (i.e., complete
restoration of perfusion) before covering the digit.
A surgeon must always be aware that congenital
­malformations of the upper extremity can comprise ab­
normal vascular neural anatomy, specifically in radial longi­
tudinal deficiency. Careful attention to venous outflow, arte­
rial inflow, design, rotation, and tension of flaps is critical to
avoiding an ischemic finger. Abnormal vascular anatomy can
exist in patients with congenital hand differences; therefore
a careful approach must be taken when performing surgery.
This is particularly true in patients with radial longitudinal
deficiency and central deficiencies where vessels can be ei­
ther absent or in very aberrant positions.
Fig. 60.6 Severe web creep and contracture after loss of graft.

SECTION VIII THE HAND AND UPPER EXTREMITY
1138
procedure enables us to split the end of the tendon to re­
construct the UCL, RCL, or both as part of the same proce­
dure10 (Fig. 60.9).
A free tendon graft reconstruction is another possibility,
although care must be taken in patients with open physes
to avoid creating a bone tunnel that injures this important
area, resulting in growth arrest. At times global instability
of the MP joint exists that is not amenable to effective re­
construction. Although technically difficult, a chondrode­
lead to significant joint instability and degenerative changes.
Both are very difficult to treat and usually require a chon­
drodesis or fusion procedure4 (Fig. 60.8).
One of the specific problems associated with type II and
III hypoplastic thumbs in radial longitudinal deficiency is
ulnar or radial MCP joint instability. Stabilization of this lax
joint can be accomplished by several modalities. Because
these patients also require opponensplasties, our preferred
technique is to use the FDS ring opposition transfer. This
Fig. 60.7 (a) Complex syndactyly with bony synostosis at the distal phalanx seen on a
radiographic study. (b) Dorsal view of the complex syndactyly. (c) Flexion creases on the
volar view indicate the presence of flexor tendons.
a
b
c
Correct
Incorrect
c
a
b
Fig. 60.8 (a) Transfer of the collateral ligament of the metacarpophalangeal (MCP) joint in preaxial (radial) polydactyly. (b) The collateral
ligament has to be carefully detached as far distally as possible while preserving the origin of the ligament (off the broad metacarpal head);
a cuff of radial periosteum together with the collateral ligament is carefully sharply dissected. (c) The chondroplasty technique. (c Repro­
duced from Herring JA, ed. Tachdjian’s Pediatric Orthopaedics. 4th ed. Philadelphia, PA: Elsevier Saunders; 2008:490.)

1139
Chapter 60 Congenital Hand Deformities
wedge osteotomy may be appropriate, as reported previ­
ously.12 Angular deformities of the thumb can also be seen,
especially in patients with Apert’s or Rubinstein-Taybi syn­
dromes. When identified, this can be treated as shown in
Fig. 60.12. This involves designing VY-YV skin flaps, releasing
abnormal insertion of thenars (if present), and performing a
dome osteotomy of the proximal phalanx. Careful preopera­
tive evaluation of patients with Rubinstein-Taybi syndrome
must be performed, because cardiac anomalies may be pres­
ent and prohibit safe surgery.
sis can be an option in young patients. The cartilaginous
surface is carefully removed from the base of the proximal
phalanx up to the level of the physis, as is the surface of the
metacarpal head. This can be a technically challenging pro­
cedure and may eventually require bony fusion when the
patient is older.
Children with higher grades of thumb hypoplasia (IIIB
or greater) are candidates for index finger pollicization. The
MCP joint of the index finger becomes the CMC joint of the
new thumb.
Despite finding radiographic nonunion at this juncture
in up to 20% of our patients following pollicization, only 3.5%
of our patients had clinical instability that required addi­
tional procedures.11
Axial Deformities
Axial deformities can be encountered frequently in con­
genital hand difference diagnoses. This can manifest itself
as angulation or malrotation. Radial polydactyly patients
frequently have angular deformities that can not only af­
fect function but also the aesthetic result. The cause is
not the same in every instance and must be analyzed and
addressed thoroughly during the initial reconstructive
procedure. In Wassel type IV radial polydactyly, a zig zag
­deformity may be present because of aberrant position­
ing of the flexor and extensor tendons or positioning of the
­bifid metacarpal head (Fig. 60.10). In an attempt to cor­
rect this, recentralization of flexor and extensor tendons,
in ­addition to osteotomies, is performed4 (Fig. 60.11). In
some cases the angular deformity is related to joint insta­
bility and may require fusion at a later date.
Patients with clinodactyly present with angulation that
may be the result of either an abnormally shaped phalanx or
a delta phalanx (bracket epiphysis). Before the age of 2 years
this may be very difficult to assess radiographically. If a true
bracketed epiphysis is noted, this can be effectively treated
with excision of the bracket and fat interposition. If abnor­
mal phalanx shape is noted without bracket, then opening
a
b
Fig. 60.9 FDS ring opposition transfer. (a) The ring finger FDS tendon
is isolated and cut at the ring finger base. (b) Then the FDS tendon is
brought around the FCU tendon and passed through a subcutaneous
tunnel to the radial side of the thumb.
Fig. 60.11 With marked angulation of the remaining articular facet, a
supracondylar osteotomy of the metacarpal is indicated. Importantly,
this is performed proximal to the origin of the collateral ligaments.
(Reproduced from Herring JA, ed. Tachdjian’s Pediatric Orthopaedics.
4th ed. Philadelphia, PA: Elsevier Saunders; 2008:490.)
a
b
Fig. 60.10 Wassel type IV preaxial (radial) polydactyly. (a) An aber­
rant positioning of the flexor and extensor tendons exacerbates the
zig zag deformity. (b) Radiograph of the deformity.

SECTION VIII THE HAND AND UPPER EXTREMITY
1140
The Bilhaut-Cloquet procedure, or modifications of it,
has been used effectively in correction of preaxial (radial)
polydactyly Wassel/Flatt types I and II. However, nail clo­
sure is critical. Removal of the nail plate and careful, accu­
rate closure of the sterile and germinal matrix are essential
to provide the best nail later. Even then, a groove or ridge is
essentially always present (Fig. 60.14).
Unrealistic Expectations
Patients and their families often evaluate the success (or
failure) of their surgery based on assumptions that have
very little to do with the technical quality of the care they
received.14 This is particularly true given that most parents,
understandably, lack a sophisticated understanding of the
underlying pathology and associated factors affecting the
ultimate outcome. Recent evidence has suggested that
Patients with complex polysyndactylies can also have
severe angular, rotatory deformities secondary to the under­
lying anatomy. The goal in these patients is to align the phy­
ses in a transverse position, as well as provide joint stability.
Nail Deformities
Nail deformities can be a significant complaint of patients
and parents after reconstruction of complete syndacty­
lies. At the initial reconstruction, the hyponychial and
paronychial folds need to be reconstructed with tissue that
resembles what would normally be present. The best op­
tion is glabrous tissue, which can be obtained from either
composite grafts from the hypothenar eminence or toe or
Buck-Gramcko tip flaps4,13 (Fig. 60.13). Full-thickness skin
grafts alone are not adequate and will lead to significant
nail deformity.
Fig. 60.12 (a) Angular deformities of the
thumb. (b) VY-YV flaps design. (c,d) Re­
sult after thenar release, dome osteoto­
my, and closure of the VY-YV skin flaps.
a
b
c
d
Fig. 60.13 (a) Illustration of Buck-Gramcko technique. (b,c) Preoperative and postoperative clinical example. (a Reproduced from Herring
JA, ed. Tachdjian’s Pediatric Orthopaedics. 4th ed. Philadelphia, PA: Elsevier Saunders; 2008:490.)
b
c
a

1141
Chapter 60 Congenital Hand Deformities
Evaluating Expectations and Effective
Communication
An article by Wilson et al16 demonstrated that evaluating
patient expectations before surgery rarely occurs in the
clinical setting. This is particularly troubling, because a pa­
tient’s or caregivers’ ability to accurately predict the out­
come for surgery is the cornerstone of informed decision
making.17 To that end, a family’s evaluation of the relative
success (or failure) of surgery is often based on how well
their preprocedure expectations correlate with their post­
procedure results.18
Perhaps the most important ingredient to enhance how
well physician–patient/family expectations correlate with
each other is to evaluate the quality and quantity of relevant
information presented and the individuals’ style of commu­
nication.19 Research has indicated that informational and
communication factors are inexplicably tied to quality of
care delivered. As such, both the Institute of Medicine (IOM)
and the Accreditation Council for Graduate Medical Educa­
tion (ACGME) have made physician–patient communication
a top priority in training new physicians.20
Effective communication (in both the adult and pedi­
atric populations) between patients, parents or caregivers,
and physicians can be enhanced (or diminished) by certain
behaviors21–27 (Box 60.1).
there are a multitude of individual differences and exter­
nal variables that underlie how patients (and their fami­
lies) evaluate their satisfaction with surgery.15 Although a
comprehensive review of all of these variables is outside
the scope of this section, some important clinical pearls
are addressed that can be easily incorporated by treating
physicians to maximize patient satisfaction and establish
realistic expectations for the aims (and possible outcomes)
for surgery.
a
b
Fig. 60.14 (a) Nail groove after a Bilhaut-Cloquet procedure for pre­
axial (radial) polydactyly repair. (b) The preoperative appearance of
the thumb.
Box 60.1
Attentiveness
With an almost endless array of distractions that characterize an individual’s everyday life, inattentiveness can
easily be misattributed to a lack of interest. To overcome this obstacle in patient care, physicians should offer a
greater level of engagement to uncover the source of the behaviors (e.g., inattentiveness can easily mask height­
ened anxiety, untreated pain, or even as a response to an uninviting environment).23
Use of Nontechnical Language
Many patients (as well as parents and caregivers) fail to comprehend even the most rudimentary aspects of tech­
nical language that is used by physicians.24,25 To overcome this barrier to effective communication, physicians
should endeavor to provide an adequate amount of information at a level of understanding that is commensurate
with the comfort of patients and their families. In addition, asking questions that check for patients’ and family
members’ level of understanding by using techniques such as reflective listening can be instrumental to facilitat­
ing this process.
Awareness of Patient Attitudes and Anxiety
One of the greatest ways to evaluate patient expectations with surgery is for physicians to actively demonstrate
an interest for patients to disclose what their concerns are about surgery and ways in which the entire health care
team can address them (e.g., aesthetic concerns such as scars, the possible need for additional surgeries, the ideal
hopes from surgery and how they compare to realistic expectations).26
Personalization and Empathy
Engaging patients as a partnership in their care is one of the best tools that will enhance patient outcomes, adher­
ence to treatment, and satisfaction.27
Physician Behaviors That Are Key for Successful Communication

SECTION VIII THE HAND AND UPPER EXTREMITY
1142
An important factor predisposing patients for this appears to
be abnormally large fingers (Fig. 60.17). If this is recognized
in a patient with syndactyly, the surgeon should be cogni­
zant of keloid scarring as a possible sequelae after surgery
and should counsel the parents accordingly. If in fact nor­
mal healing occurs, the parents (and the surgeon) are happy.
However, more importantly, if keloid scarring does occur,
the parents were adequately prepared for this and it is not
a surprise or the surgeons “fault.” Pediatric patients usually
tolerate methotrexate treatment well but must receive folic
acid supplementation and be monitored for liver function
abnormalities and blood count irregularities. The length of
treatment is dependent on the response of the patient and
is not always the same.28,29
However, if a nonsurgical approach fails or inadequate
response occurs, the patient may require excision and repeat
grafting (Fig. 60.18). In this case, we start the methotrexate
one week before surgery and continue it for a minimum of
6 weeks postoperatively. We have had good results with this
treatment strategy in this subset of patients. It should be
further reiterated that most of these patients can be identi­
fied preoperatively based on finger size and thus counsel­
ing regarding this potential complication will have already
taken place.
Managing Complications and
Correcting Unfavorable Results
in Congenital Hand Surgery
Hypertrophic Scars, Contractures,
and Web Creep
In some cases, significant scarring can occur, which often
presents as either hypertrophic or keloid scars (Figs. 60.15
and 60.16). Once identified, the hypertrophic or keloid
scars should be treated with usual measures, including
massage and scar pads. One particular complication of hy­
pertrophic scars (if persistent) is the possible development
of a significant contracture that has the potential to impede
the normal function of the affected digit. In these cases,
surgical release with local flaps or grafting is the mainstay
of treatment after allowing a satisfactory period of conser­
vation care (9–12 months).
Keloid scarring in pediatric patients with congenital
hand differences can be extremely severe and can even ex­
tend into nonoperated areas. In this particular subset of pa­
tients, nonsurgical treatment with short-term methotrexate
(as well as local scar care) has been shown to be beneficial.
Fig. 60.15 Clinical examples of hypertrophic scar after syndactyly release.
Fig. 60.16 Keloid formation after syndactyly release.

1143
Chapter 60 Congenital Hand Deformities
one or both of the fingers. If this is present it is tempting to
perform z-plasties for correction. Although this sometimes
can be adequate treatment, the surgeon should realize that
these are very small flaps that can be technically challeng­
ing, and it may be more appropriate to perform full-thick­
ness skin grafting instead. The volar wrist crease can be used
again as a donor site because it is very forgiving in children,
as discussed previously. However, temporary pin fixation
can also be used in some patients with flexion contractions
during the initial healing period.
Web creep (with or without scar contracture) is the
most commonly encountered complication after syndac­
tyly reconstruction. Treatment of web creep is not the same
in every case, and the deformity must be assessed carefully.
In some cases, when the webbing is up to half the distance
to the PIP joint, a butterfly-type z-plasty (with or without
small skin grafts) may solve the problem (Fig. 60.19). How­
ever, if the webbing extends laterally near the PIP joint, a
dorsal flap will be required with skin grafting. Again, the
mainstay of this approach is appropriate dorsal flap design.
At the same time there may be a linear scar contracture on
Fig. 60.17 Preoperative clinical photographs (a,b) and radiographs (c,d). The involved digits are enlarged.
a
b
c
d
Fig. 60.18 Clinical views of the patient shown in Fig. 60.16 after keloid excision, full-thickness skin grafting, and methotrexate therapy.
a
b

SECTION VIII THE HAND AND UPPER EXTREMITY
1144
wound and the hand. Placing a neural blockade using 2%
lidocaine to the median and superficial radial nerves and/
or application of lidocaine-soaked plugs directly to the ves­
sels may interrupt the vasospasm and restore proper circula­
tion of the finger. However, if all of these approaches fail to
show improvement, the wound must be explored, and the
surgeon must evaluate the location of the kinked, twisted,
or stretched vessel. Once identified, a microsurgical inter­
vention should be undertaken to save the finger, such as
microstripping of the adventitia, microvascular repair, or
interposition of a vein graft of an artery.
Ischemia
A necrotic finger is one of the worst complications that
can occur in pediatric hand surgery. After release of the
­tourniquet, the fingers should pink up immediately. If the
circulation is in doubt it must be restored. The first step is
the ­removal of the dressing. If a tight dressing is not the
cause of the impaired circulation, the vasculature of the
­finger may be vasospastic.
Treatment in this situation would include warming the
patient and applying warm saline sponges directly to the
Fig. 60.19 Flap design (a,b) and result (c,d) for a butterfly-type z-plasty in a patient with web creep after syndactyly reconstruction.
(e) Illustrative graphic of the procedure.
a
b
c
d
e

1145
Chapter 60 Congenital Hand Deformities
of the thumb. Therefore revision procedures that are aimed
at correcting malalignment can also include corrective oste­
otomies and tendon realignment and reinsertion.
The preferred method for opening–closing wedge oste­
otomy is through the metaphysis, or the site of the maximal
convexity of the bone. This is followed by placing a combina­
tion of longitudinal and crossed K-wires to provide adequate
fixation and compression across the osteotomies.
Conclusion
Unfavorable results after surgery for treatment of congeni­
tal hand differences may occur in four different situations:
(1) The wrong operation was performed. A thorough his­
tory and physical examination along with correct diagnosis
is cornerstone in decision-making regarding surgery. Fur­
thermore, each case must be evaluated individually, be­
cause there can be significant phenotype variation among
patients with the same diagnosis. (2) Technical failures
have occurred. The surgeon has to be aware of pitfalls of
the procedure and possible anatomic variations of a patient
with congenital hand differences of the upper extremity.
(3) The timing was wrong. Understanding and respecting
the anatomic, physiological, and psychological develop­
ment of a patient allows the surgeon to choose the optimal
timing for a surgical intervention. (4) The patient is not able
to adapt functionally to his or her operated hand. Thorough
preoperative evaluation of the patient and his or her family
will give clues regarding how the patient will adapt after
surgery. For example, pollicization of a stiff index finger
that the child does not use for prehensile activities rarely
results in a favorable outcome, because the child tends to
continue to ignore the pollicized digit after surgery.
A good rapport with parents is imperative and of utmost
importance. Proper counseling and discussions before sur­
gery regarding likely outcomes and realistic expectations
is obviously important. Failure to achieve this can lead to
significant unhappiness in parents and the patient despite
the surgeon evaluating the postoperative result as “good.”
If flap necrosis occurs—depending on the severity—it
can be treated with dressing changes. However, if it is se­
vere, débridement and revision surgery with skin grafts or
radial forearm or groin flaps may be necessary.
Nail Deformity
Nail deformity can occur after reconstruction of a complete
syndactyly if the initial surgery fails to address this area ad­
equately. Correction of this is very challenging and must be
based on the deformity present. There may be inadequate
tissue in the hyponychial or paronychial area that can be
treated with a composite graft from the hypothenar emi­
nence or toe4 (Fig. 60.20). However, other times the defor­
mity can involve the nail or nail fold region in addition to
the hyponychial and paronychial folds. This is a very diffi­
cult deformity to adequately treat and sometimes requires
eponychial–germinal matrix revision as well as composite
grafting. Again, no matter what technique is used, the re­
sultant appearance is unlikely to be as good as if this area
had been adequately addressed at initial surgery.
Angular Deformity and Joint Instability
Malalignment and instability (primarily of the interpha­
langeal joint) tend to be the primary factors that affect the
long-term outcome in patients with radial polydactyly.
This is especially true in patients affected with Flatt type IV
polydactyly. As such, for these patients the likelihood that a
revision surgery will be necessary increases over time.30,31
One of the reasons underlying the need for revision sur­
gery involves the role of extrinsic tendon malalignment. If
the malalignment is underestimated (and therefore did not
receive an adequate level of correction at the time of the
index surgery), this condition may induce the thumb into a
zig zag deformation.
In a pollex abductus anomaly, the flexor pollicis longus
(FPL) attaches to its insertion point but also to the extensor
tendon via a tendon slip that passes around the radial side
Fig. 60.20 (a) Composite grafts of skin and adipose tissue from the hypothenar eminence. (b) Illustrative depiction of procedure. (b Repro­
duced from Herring JA, ed. Tachdjian’s Pediatric Orthopaedics. 4th ed. Philadelphia, PA: Elsevier Saunders; 2008:490.)
a
b

SECTION VIII THE HAND AND UPPER EXTREMITY
1146
16. Wilson KA, Dowling AJ, Abdolell M, Tannock IF. Percep­
tion of quality of life by patients, partners and treating
physicians. Qual Life Res 2000;9(9):1041–1052
17. Waljee J, McGlinn EP, Sears ED, Chung KC. Patient expec­
tations and patient-reported outcomes in surgery: a sys­
tematic review. Surgery 2014;155(5):799–808
18. Bogart LM. Relationship of stereotypic beliefs about phy­
sicians to health care-relevant behaviors and cognitions
among African American women. J Behav Med 2001;24(6):
573–586
19. Ambady N, Laplante D, Nguyen T, Rosenthal R, Chaume­
ton N, Levinson W. Surgeons’ tone of voice: a clue to mal­
practice history. Surgery 2002;132(1):5–9
20. Talen MR, Grampp K, Tucker A, Schultz J. What phy­
sicians want from their patients: Identifying what
makes good patient communication. Fam Syst Health
2008;26(1):58–66
21. Taylor S. Health Psychology. 9th ed. New York: McGraw-
Hill Publication; 2015
22. Frederikson LG. Exploring information-exchange in con­
sultation: the patients’ view of performance and out­
comes. Patient Educ Couns 1995;25(3):237–246
23. Street RL Jr. Information-giving in medical consultations:
the influence of patients’ communicative styles and per­
sonal characteristics. Soc Sci Med 1991;32(5):541–548
24. Fields AM, Freiberg CS, Fickenscher A, Shelley KH. Patients
and jargon: are we speaking the same language? J Clin
Anesth 2008;20(5):343–346
25. LeBlanc TW, Hesson A, Williams A, et al. Patient under­
standing of medical jargon: a survey study of U.S. medical
students. Patient Educ Couns 2014;95(2):238–242
26. Street RL Jr, Marengo MF, Barbo A, et al. Affective tone in
medical encounters and its relationship with treatment
adherence in a multiethnic cohort of patients with rheu­
matoid arthritis. J Clin Rheumatol 2015;21(4):181–188
27. Kessels RPC. Patients’ memory for medical information. J
R Soc Med 2003;96(5):219–222
28. Muzaffar AR, Rafols F, Masson J, Ezaki M, Carter PR. Keloid
formation after syndactyly reconstruction: associated
conditions, prevalence, and preliminary report of a treat­
ment method. J Hand Surg Am 2004;29(2):201–208
29. Tolerton SK, Tonkin MA. Keloid formation after syndacty­
ly release in patients with associated macrodactyly: man­
agement with methotrexate therapy. J Hand Surg Eur Vol
2011;36(6):490–497
30. Stutz C, Mills J, Wheeler L, Ezaki M, Oishi S. Long-term
outcomes following radial polydactyly  reconstruction. J
Hand Surg Am 2014;39(8):1549–1552
31. Patel AU, Tonkin MA, Smith BJ, Alshehri AH, Lawson RD.
Factors affecting surgical results of Wassel type IV thumb
duplications. J Hand Surg Eur Vol 2014;39(9):934–943
References
1. Feingold M. Clinical evaluation of the child with skeletal
dysplasia. Orthop Clin North Am 1976;7(2):291–301
2. Ragland R III, Moukoko D, Ezaki M, Carter PR, Mills J. Fore­
arm compartment syndrome in the newborn: report of
24 cases. J Hand Surg Am 2005;30(5):997–1003
3. Ho CA, Richards BS, Ezaki M. Congenital band syndrome
with pseudarthrosis of the radius and ulna and impend­
ing vascular compromise: a case report. J Pediatr Orthop
2014;34(6):e14–e18
4. Carter PR, Oishi SN, Ezaki M. Disorders of the upper ex­
tremity. In Herring JA, ed. Tachdjian’s Pediatric Orthopae­
dics, 4th ed. Philadelphia, PA: Saunders Elsevier; 2008:
490
5. Edgerton MT, Snyder GB, Webb WL. Surgical treatment
of congenital thumb deformities (including psychologi­
cal impact of correction). J Bone Joint Surg Am 1965;
47(8):1453–1474
6. Loepke AW, Soriano SG. An assessment of the effects of
general anesthetics on developing brain structure and
neurocognitive function. Anesth Analg 2008;106(6):1681–
1707
7. Block RI, Thomas JJ, Bayman EO, Choi JY, Kimble KK, Todd
MM. Are anesthesia and surgery during infancy associat­
ed with altered academic performance during childhood?
Anesthesiology 2012;117(3):494–503
8. Tonkin M. Ulnar longitudinal deficiency. In Green DP,
Hotchkiss RN, Pederson WC, eds. Green’s Operative Hand
Surgery, Philadelphia, PA: Churchill Livingstone; 1999:358
9. Carter PR. Making a thumb—the story of pollicization.
Surgeon DVD. Texas Scottish Rite Hospital for Children,
2006
10. Vuillermin C, Butler L, Lake A, Ezaki M, Oishi S. Flexor
digitorum superficialis opposition transfer for augment­
ing function in types II and IIIA thumb hypoplasia. J Hand
Surg Am 2016;41(2):244–249, quiz 250
11. Lochner HV, Oishi S, Ezaki M, Malungpaishrope K, Moore
RB. The fate of the index metacarpophalangeal joint fol­
lowing pollicization. J Hand Surg Am 2012;37(8):1672–
1676
12. Piper SL, Goldfarb CA, Wall LB. Outcomes of opening
wedge osteotomy to correct angular deformity in little fin­
ger clinodactyly. J Hand Surg Am 2015;40(5):908–13.e1
13. Buck-Gramcko D. Congenital malformations. In: Nigst H,
Buck-Gramcko D, Millesi H, Lister GD, eds. Hand Surgery.
Vol 1. New York, NY: Thieme; 1988:22–23
14. Waljee JF, Ubel PA, Atisha DM, Hu ES, Alderman AK. The
choice for breast cancer surgery: can women accurately
predict postoperative quality of life and disease-related
stigma? Ann Surg Oncol 2011;18(9):2477–2482
15. Birkmeyer JD, Dimick JB, Birkmeyer NJ. Measuring the
quality of surgical care: structure, process, or outcomes? J
Am Coll Surg 2004;198(4):626–632

1147
Chapter 60 Congenital Hand Deformities
Discussion by Joseph Upton, Brian I. Labow, and Amir Taghinia
Children with congenital upper limb differences
­present both routine and unique problems, which are
amenable to a plethora of surgical approaches. It is not
­surprising that many of the more imaginative members
of our specialty gravitate toward this corner of hand
surgery.
Trust
A solid relationship of trust between the surgeon,
­patient, and family is the greatest deterrent to untow­
ard, unwanted, and unfavorable results. The initial con­
sultation is pivotal. Plenty of time should be allotted
for concerned and often guilt-stricken parents. There
should be no phone calls, no beepers, and no inter­
ruptions, and the surgeon must give the family his or
her undivided attention. They must perceive that he or
she is there for them. All questions should be answered
clearly in a fashion the family is able to comprehend.
Patience and lack of visible frustration are important
when parents and grandparents are repeatedly ask­
ing the same questions. In extreme cases it is wise to
­record and transcript the entire consultation and have
the families listen instead of taking notes. An attentive
surgeon is not an oxymoron. The family must be reas­
sured that they will have access to the surgeon in the
future. These are not young patients coming in for a
simple carpal tunnel release or triangular fibrocartilage
complex (TFCC) repair. With the exception of those with
simple polydactylies, surgeons retain these patients for
life and will often treat their children and occasionally
grandchildren in the setting of familial deformities.
Adjunctive aids include arranging for the family to see
follow-up patients, clinical pictures before and after a par­
ticular procedure, and the use of hand molds (Fig. 60D.1).
Pictures can be a double-edged sword, because it is im­
perative to show the family both good and bad results as
well as potential complications. The effect of actual molds
is most dramatic as the family sees a three-dimensional
image before and after comparison. At this point many of
their questions are answered, because they can see the
projected result.1 Those who attend to these preoperative
details are likely to be technically proficient surgeons who
follow their patients carefully as they grow.
Fig. 60D.1 Hand molds give a three-dimensional picture and answer many parental and family questions. (a) A sequence of recon­
struction in a child with Apert type II hands from 1 to 15 years of age. (b) A sequence of a child who lost two digits and a thumb
in a neonatal vascular accident. These were replaced with microvascular toe-to-hand transfers. (c) The right hand of a child with
radial dysplasia before and 3 years after pollicization of the index ray.
a
b
c

SECTION VIII THE HAND AND UPPER EXTREMITY
1148
Timing
Debate over timing for intervention is always lively and
prolific. Of the factors raised by the authors some are
more important than others.2–6 In general, we try to
complete major reconstructions by the time children
enter first grade at age 6 years. Of course, there are ex­
ceptions. Knowledge of anomalous anatomy, associated
malformations, and comorbidities is important and is
constantly evolving in the literature.
Arguments centered around earlier acquisition of
functional skills and plasticity of the cerebral cortex
continue for this age group because children adapt so
rapidly. It is interesting that functional magnetic reso­
nance imaging (fMRI) scans in elderly stroke patients
show similar but less extensive reorientation—an age
group thought to have very little regenerative potential.7
The pediatric brain is capable of constant, rapid change,
and we don’t believe there is any absolute best time to
integrate new skills. Early in childhood is preferred.
The most important factor regarding timing is the
surgeon. He or she should decide when the patient is
old enough and well enough for the best outcome. Con­
fidence, knowledge of anomalous anatomy, and expe­
rience vary among surgeons and performing complex
operations too early on very small children can be del­
eterious. Waiting can be a virtue. If the trust exists, the
family will wait.
Ischemia
Fortunately, ischemia is not common in this patient
population. Compromised arterial inflow does exist in
five specific settings:
1. A white digit after tourniquet release is usually
indicative of spasm but can be the result of direct
injury, compression, twisting, or kinking. The
microvascular protocol of warm saline, topical anti­
spasmotic, patience, flow test, adventitial dissec­
tion, and pedicle repositioning is usually sufficient
to reverse the problem. If a thrombus is present,
resection and microvascular repair or vein graft
may be necessary.
2. An ischemic flap may have been excessively thinned,
closed under tension, or just anatomically unsound.
The caveats here are that “mini flaps yield mini
results,” especially within the first web space. If
there is any doubt about tension, a full-thickness
skin graft should be used. These grafts can always
be excised when the child is older and tissue has
become more supple.
3. A white or mottled distal extremity after a proce­
dure by interventional cardiology or radiology usu­
ally involve small infants or children with multiple
medical problems. The diagnosis is characteristi­
cally delayed and not suspected early; aggressive
surgery is needed.
4. A diagnosis is rarely made early in a newborn with
dorsal forearm blistering indicative of a neonatal
Volkmann injury, which, like postoperative forearm
and hand compression lesions, require immediate
fasciotomies and nerve decompression. Any sign
of ischemia in the arm, forearm, or hand of a child
must be taken seriously.
5. Venous congestion is much less common but just as
ominous and can occur after free tissue transfer or
digital transposition procedures (i.e., pollicization),
revascularization, or debulking of vascular malfor­
mations. The same approach is used for problems
on the arterial side. The threshold for repeating
a venous anastomosis or inserting a vein graft to
bridge a zone of injury should be low.
Syndactyly
The principles of syndactyly release (Box 60D.1) are
well established.5 More than 50 techniques for release
of the first web in addition to those for the second to
fourth webs have been described. Some are preferred
over others. It is important for the surgeon to become
expert in one or two and not “shop around” with ev­
ery newly described method.5,8 Technical proficiency is
important. We prefer to distribute the grafts equally,
use 6–0 chromic sutures, to evert all graft to flap edges,
to take necessary time to complete the surgery, to ap­
ply a careful dressing, and to adequately immobilize all
patients. Anything less than a 100% take of a skin graft
is indicative of a technical problem. The overwhelming
majority of problems following syndactyly correction
(and indeed all pediatric hand surgery) are secondary
to inadequate immobilization. With exceptions small
children are held in well-padded long arm casts with
the elbow flexed 90 degrees.5
Box 60D.1
•• Careful planning
•• Full-thickness flaps in commissure
•• Full-thickness skin grafts
•• Zig zag palmar incisions
•• Straight-line incisions dorsal
•• Excision of intradigital fibrous bands
•• Equal distribution of grafts
•• Meticulous technique
•• Adequate, conforming dressing
•• Immobilization
Surgical Principles of
Syndactyly Separation

1149
Chapter 60 Congenital Hand Deformities
Technique
Incision technique is a personal preference. After 40
years and more than 7,500 web releases, I (J.U.) pre­
fer either (1) dorsal rectangular or dorsal and palmar
triangular interdigitating flaps for commissure con­
struction, (2) sharp zig zag palmar incisions using the
transverse flexion creases, and (3) straight-line dorsal
incision (Fig. 60D.2). The latter is used for appearance
and avoids visible pigmented skin grafts on the dorsal
surface of a digit or thumb. Of course, straight-line pal­
mar incisions can be made on a patient with Apert’s
syndrome who has no interphalangeal joint motion
of the central three digits. The triangular commissure
flaps (see Fig. 60D.2) provide greater versatility with
managing the width of the commissure and in the cre­
ation of a dorsal-to-palmar 45-degree angle. Single
dorsal rectangular flaps are also commonly used for the
digital web spaces, especially incomplete, simple syn­
dactylies. In simple cases the bifurcation of the com­
mon digital artery marks the depth of the web space.
Overcorrection may be considered with brachydactyly
with which the surgeon desires to provide the appear­
ance of relative length.
Incisions
The origin of the “diamond” incisions deserves some
clarification. The long-term observations that palmar
incisions made perpendicular or obliquely across
transverse skin creases generated needless flexion
contractures were well appreciated in the early days
of the specialty of hand surgery when most opera­
tions consisted of incision and drainage of infections.
Many of these incisions were left to heal by secondary
intention. Consequently, hypertrophic scars with joint
contracture were common. Incisions made within the
diamond patterns were more susceptible to contrac­
ture, because with digital motion these areas moved
and were not tension free. In contrast, incisions made
around the borders of these configurations did not
move and remained stationary.5,9 Additional safe areas
that did not move were the transverse flexion creases
(MP, proximal interphalangeal [PIP], distal interpha­
langeal [DIP] joints) and along the side of the digit
in either the midaxial or midlateral lines. All hand
surgeons must understand Littler’s classic review9 of
these principles as published in a Plastic Surgery Sym­
posium publication instead of a hand journal article
Fig. 60D.2 Syndactyly. (a,b) A large dorsal rectangle has been used for the commissure lining. Comparison measurements have
been made on the adjacent web space. The dorsal dimples (blurred) mark the position of the underlying joint space, and the limbs
of the flap are directed toward these dimples. Palmar zig zag incisions use the normal flexion creases. (c–e) The dorsal incision is a
straight line. Commonly preferred commissure flap designs include a large dorsal rectangle, dorsal and palmar rectangular flaps,
(f) and dorsal-palmar interdigitating triangular flaps.
a
b
c
d
e
f

SECTION VIII THE HAND AND UPPER EXTREMITY
1150
Fig. 60D.3 Palmar incisions. (a) The midaxial line connects the center of rotation of the interphalangeal joints. The midlateral
line is below it in the midportion of the digit. Two approaches to the neurovascular bundle illustrate both dorsal (top) and palmar
(bottom) Cleland’s ligament. (b) Left: Safe palmar incisions are made along transverse (T), oblique (O), and longitudinal (L) direc­
tions. Center: The neurovascular bundles are particularly vulnerable near the lateral borders of these diamond patterns. Right: Many
combinations of exposure to the digit can be made using these lines. (Modified from Littler JW. Hand, wrist and forearm incisions.
In: Littler JW, Cramer LM, Smith JW, eds. Symposium on Reconstructive Hand Surgery: Educational Foundation of Plastic Recon­
structive Surgery. Vol. 9. St. Louis, MO: CV Mosby;1974:87–100. Original drawings by Dr. J.W. Littler.)
Radial aspect
Ulnar aspect
Cleland’s
ligament
Ob.l.
Ob.l.
Midaxial
Midlateral
Midaxial
Midlateral
a
b

1151
Chapter 60 Congenital Hand Deformities
(Figs. 60D.3 and 60D.4). Our preference for syndac­
tyly release uses these transverse creases. One caution
about incisions along the sides of small digits is that
in children there is a tendency to migrate palmar with
growth.
For this reason we prefer the high midaxial location
just above the interface of the glabrous and hirsute dor­
sal skin (Fig. 60D.5).
Grafts
The type of skin graft used varies, and some are bet­
ter than others. The ideal is to have glabrous skin po­
sitioned palmar and nonglabrous (often hirsute) skin
positioned dorsal. The hypothenar eminence is best
suited as a donor, but the size is limited. Spare parts
from amputated polydactyly are excellent but usu­
ally not available. The groin flexion crease is utilitar­
ian, and the lower abdominal flexion crease is useful
for repeated graft sessions in the hand of a person with
Apert’s syndrome or other types of mitten hand. These
grafts should not include central portions within the
future hair-bearing escutcheon. Teenagers with hair-
bearing skin grafts along the sides of their digits are not
satisfied with the appearance.5,6
In males preputial grafts are an option and have
been used many times.5 A single preputial foreskin can
be used to resurface two complete web spaces in the
a
b
Fig. 60D.4 Preputial grafts. (a) Following circumcision the preputial skin has been stretched to its maximal limits. Enough skin was
available to resurface two complete syndactyly releases. (b) This skin remains thin, often shiny and commonly hyperpigmented.
Fig. 60D.5 Fibrous bands. Dense fibrous bands are invariably seen with varying degrees of density within the syndactyly. They
contain oblique and transverse components and are most consistent along the proximal and distal phalangeal segments. They
may insert along the digit above and below the neurovascular bundles.

SECTION VIII THE HAND AND UPPER EXTREMITY
1152
Fig. 60D.6 (a) This child presented with a tight palmar contracture secondary to a burn. (b) During release cutback incisions were
made to the midaxial lines (red dots) and all soft tissue was released. (c) The proximal interphalangeal joint has been immobilized
with a pin and a precisely contoured full-thickness skin graft harvested from the groin region. (d,e) A subcuticular closure has been
performed with points x–y and a–b in direct opposition. With time and growth the incisions along the side of the digit will migrate
more palmar. (Modified from Littler JW. Hand, wrist and forearm incisions. In: Littler JW, Cramer LM, Smith JW, eds. Symposium on
Reconstructive Hand Surgery: Educational Foundation of Plastic Reconstructive Surgery. Vol. 9. St. Louis: CV Mosby;1974:87–100.)
a
b
c
d
e
hand of a patient with Apert’s syndrome. The outcome is
a thin, shiny, hyperpigmented surface. Many patients ask
for revision during adolescence, but most do not mind
the appearance5 (Fig. 60D.6).
Skin grafts harvested from the antecubital flexion
crease, wrist flexion crease, or volar forearm are usu­
ally made as a matter of convenience and can be aes­
thetically troubling later.5 When subcuticular suturing
techniques are not used, conspicuous cross-hatching re­
sembling a railroad track will be the result. These are
not “forgiving” donor sites, and alternative donor sites
are always available. It is cruel to subject a child to the
potential psychological nuances of a visible transverse
scar in the distal third of the forearm.
Our use of glabrous grafts harvested from the dis­
tal palmar flexion crease and thenar crease have been
restricted to their use as composite grafts at the distal
phalangeal level. However, the hypothenar eminence is
another valuable donor site from which glabrous skin,
dorsal (hirsute) skin, or combinations can be harvested.5
Web Creep
Web creep, defined as recurrent webbing or as a distal
migration of the commissure at the base of the digit,
occurs more frequently than reported. A conservative
estimate of 15 to 20% of released web spaces demon­
strate some degree of web creep by the time the child

1153
Chapter 60 Congenital Hand Deformities
reaches skeletal maturity.5 It is very difficult to reach an
accurate figure, because most (almost all) reported se­
ries do not extend beyond 5 years. We have seen many
recurrences in our own patients as well as those of
many other surgeons. There are many causes; the fol­
lowing are the most well known:
•• Flawed original design
•• Partial flap or graft loss within the commissure or at
the volar base of the digits, usually caused by loss of
immobilization
•• Infection (rare)
•• Discordant growth of the incisional scars and skin
grafts versus the other tissues within the digit
•• Intradigital fibrous bands
Intradigital fibrous bands have not been described in
detail but have been studied5 (see Fig. 60D.3). These
bands are distinct from Cleland’s and Grayson’s liga­
ments dorsal and palmar to the neurovascular bundles
along the sides of the digits. They extend across the
webbing in varying densities and have origins both
above and below the neurovascular bundles. There
are oblique and transverse components with very few
longitudinal portions. They are most dense along the
diaphyses of the phalanges and less developed at the
level of the DIP and PIP joint spaces. These bands will
become contiguous with the well-described Cleland’s
and Grayson’s structures, which insert along the sides
of the retinacular sheath. They remain a source of con­
tracture along the sides of each digit and at its base and
are best removed using a microvascular technique at
the time of the initial syndacytly release. The potential
for injury to the neurovascular bundles is significant.
They can be found in all web spaces.
The management of complex syndactyly, mitten
hands, syndactyly associated with central synpolydac­
tyly (SPD), and Apert’s syndrome is well beyond this
discussion, and the reader is referred to texts covering
these problems in depth. Complications are common,
and unfavorable outcomes abound.2,4,5,8 The problems
seen in long-term follow-up of these children has as
much to do with the original anatomic abnormalities
of the growth plates, intrinsic muscle origins and in­
sertions, previous scars and incisions, and abnormal
joint configuration as it does with surgical technique.
SPD ­patients are the most perplexing, and their residual
problems are predictable with growth.
Joint Instability
Children will not use an unstable finger or floppy thumb
or digit; they will transfer function to a stable ray of
the hand. The border rays (thumb and fifth digit) are
the most commonly affected. The FDS (ring) transfer
for hypoplastic (II, IIIA) thumbs is an excellent choice,
because it solves two problems: (1) joint instability
and (2) lack of palmar abduction. The procedure is
demanding, because the surgeon must avoid injury to
the growth plates at the base of the proximal phalanx,
position one limb of the tendon graft above and one
below the joint axis of rotation, secure bone-to-bone
fixation, and adjust tension. We pass the tendon grafts
(FDS tails) through subperiosteal tunnels prepared
with K-wires. Direct suturing of these tendon grafts
will not provide permanent correction. The degree of
difficulty for reconstruction of a small type IIIA thumb
is similar to that for an index pollicization. A good as­
sistant makes a difference.
For polydactyly correction, the periosteal sleeve is
an important component in any reconstruction involv­
ing polydactyly excision anywhere in the hand.10 We
prefer a much wider base than that shown in Fig. 60.8
and actually raise periosteum on both ends of the col­
lateral ligament to be saved as well as a portion of the
palmar plate. Often the proximal sleeve must be elevated
for complete excision of the extra metacarpal head.
Congenital hand surgeons do not like chondrodesis
procedures because of the high nonunion rate, especially
at the interphalangeal joint level. The articular surfaces
need to be shaved down to bleeding bone. Small pins
need to be left in longer than 1 month, and injury to a
phalangeal epiphysis may easily occur. Very small bones
simply will not fuse. One absolute indication is the MP
joint in type IV radial polydactyly, in which the meta­
carpal head is angulated in two directions and can devi­
ate with divergent–convergent phalangeal segments.10
In the past many surgeons (including ourselves)
have erred in trying to preserve thumb MP joint mo­
tion in polydactyly and hypoplastic thumbs. A thumb
with a stable MP joint that has only IP joint and CMC
joint motion is very functional.
The instability of the fifth (ulnar) border ray MP
joint is a particular problem with no easy solution. This
occurs most commonly in symbrachydactyly. Ligament
reconstruction using tendon grafts often pull out. Non­
vascularized toe phalangeal transplants are unpredict­
able. Ultimately, these unstable joints are fused when
patients are older and function very well. Deficient
length of these rays is corrected using distraction and
bone grafting techniques.
Axial Deformities
We were initially taught that articular deformities at
birth were capable of remolding similar to a fracture
malunion involving the disphysis of a tubular bone.
This does not occur in the congenital population. For
complex syndactyly we were initially taught to “. . . wait
until the baby fat is gone and separate by 2 years and

SECTION VIII THE HAND AND UPPER EXTREMITY
1154
correct skeletal deformities at 7 years.”8 Early correc­
tion of skeletal deformities can be safe, easy, and pre­
dictable. We prefer to keep these osteotomies as simple
as possible. Closing wedges works better than opening
wedges with bone grafts. A saw is used to make the cut
on the opening side, and an osteotome is used to fa­
cilitate a greenstick fracture of the deep cortex. A small
oblique wire is usually all that is needed to hold the os­
teotomy. Before discarding extra polydactylous parts,
the surgeon must make sure that this bone would not
be helpful elsewhere within the hand.
Long-term follow-up (more than 20 years) of the
deviation seen in our radial and central polydactyly pa­
tients has been consistent.10 What you see at the time
of initial repair is what you get. There is no significant
change in the axial deviation. Therefore, we try to cor­
rect as much abnormal deviation and rotation as pos­
sible initially.
The radial clinodactyly of the thumb seen in many
syndromic patients is of particular interest to us; as
of this printing, I (J.U.) have treated 147 children with
Apert’s syndrome over 40 years. The cause of the de­
viation is, of course, the hypoplastic proximal phalanx
containing a longitudinal epiphyseal bracket (LEB) on
its radial border, which acts as a deterrent to growth.
The clinodactyly progresses as the child grows. We still
lengthen with opening wedge osteotomies, bone grafts,
and a large z-plasty on the radial side. We prefer a V-Y
lengthening, which leaves the same problem, a soft tis­
sue depression at the site of lengthening. This problem
can be alleviated with the placement of a composite
graft (dermis and fat) beneath the closure. The soft tis­
sue available usually restricts the degree of lengthen­
ing. We have also placed a composite graft in the central
limb of a z-plasty after transposition of the flaps. By add­
ing tissue the soft tissue deficiency will not compromise
the potential length.
We have used the Vickers procedure, LEB excision,
and interpositional fat grafting many times with vari­
able results. Most (more than 60%) have been success­
ful, but it takes a long time to see the results. Some have
resulted in the formation of an osseous bar on the edge
of the epiphysis that will restrict growth. This operation
is much more predictable at the distal radial or humeral
levels. It is our procedure of choice for early Madelung
deformities when detected early in childhood. We prefer
corrective, complex osteotomy when seen late.
The best correction for severe thumb radial clino­
dactyly in Apert’s syndrome is distraction lengthening
in two stages. The procedure is performed during the
preteen years when the phalanges are larger. An oste­
otomy is placed just proximal to the location of what
would be the interzone of the IP joint. Pins are placed in
the distal phalanx and the distal portion of the thumb
metacarpal. Intercalated, preferably autogenous bone
grafts are always inserted within the gap; the process
of callotasis is not reliable at this level. However, there
may be an occasional case that does fill with bone by the
second stage. Distraction lengthening techniques used
for midface and thumb correction have significantly
transformed the outlook for these children, who now
can achieve an improved appearance of both faces and
thumbs (see Fig. 60D.1).
Clinodactyly encountered within the central por­
tions of the hand, as well as the common isolated ra­
dial deviation of the fifth ray, are treated with the same
principles. Attention to rotational deformities is often
neglected and is very easily checked by flexing the digit
at the time of osteotomy and pin fixation.
Nails
Although little can be done to correct dysmorphic nails,
attention to nail folds and configuration is important.
Over the years we have tried many modifications of the
nail sharing procedure (Bilhaut operation). The most
difficult portion of the operation is the creation of the
proper convexity of the underlying distal phalangeal
segments and avoidance of a longitudinal ridge. We use
this procedure primarily if both nail partners are very
small. Hopefully, distraction techniques will solve this
unfavorable outcome in the future.
Paronychial folds are amenable to improvement
in most syndactyly patients. First we eliminate the
sharply angulated portions of the conjoined nail ad­
jacent to the ridge with strip excisions of sterile and
germinal matrix. The sidewall pulp flaps (if present)
are advanced at least 2 mm to 3 mm above the level of
the nail plate. This overhanging portion will epithelial­
ize and contract and form the semblance of a nice fold.
The V deformity at the juncture of the paronychial and
eponychial folds can be avoided with careful technique.
Eponychial folds can be created by advancing dorsal
tissue as a bipedicle flap and placing a full-thickness
skin graft over the DIP joint level. This method is well
known to burn surgeons but may be less familiar for
congenital hand surgeons.
We are also strong advocates of the composite
great toe (lateral surface) pulp free grafts to the lateral
­borders of the distal phalanx and routine use of Dieter
­Buck-Gramcko triangular flaps for paronychial fold con­
struction. Fig. 60.13a depicts these flaps, which are very
narrow. This technique has several caveats:
•• Broad base of flap on palmar surface
•• Conservative undermining
•• Trimming of necessary bone or sterile matrix to get
the flap along the digit
•• Removing sutures and repositioning if still
blanching at the end of the procedure

1155
Chapter 60 Congenital Hand Deformities
With too many tight sutures placed along the border of
a thin flap, “perfection” becomes the enemy of “good.”
Communications and Expectations
Empathy, communication styles, attentiveness, percep­
tiveness, and susceptibility for distraction are some of
the behaviors that are usually covered in grand round
presentations. We teach our trainees to sit down with
the families, look them in the eye, and in common-
sense terms describe the diagnosis and treatment in
the same fashion they would if the patient was a mem­
ber of their family.
Expectations can be a problem, particularly with
overbearing parents. Unrealistic desires are usually ob­
vious after a comprehensive initial visit with the family,
which often includes grandparents, aunts, and uncles.
There is no substitute for a well-structured and well-
presented explanation about the condition and its natu­
ral history. This is where the molds and pictures of both
good and bad results are particularly helpful. What they
want to see most is confidence and reassurance that the
surgeon can help their child. A physician who is well
known and respected locally, nationally, or internation­
ally must work harder to convey realistic outcomes. The
ego must not get in the way. The hardest people to treat
are 5- and 6-year-old children who come to see their
hand surgeon thinking that they are going to magically
receive a new finger or a new hand.
Contractures
Treatment for contractures is similar to that for other
conditions seen by the hand surgeon (see Fig. 60D.4)
The most common theme has been “. . . when in doubt,
use a skin graft.”5 Those seen in syndromic patients are
particularly well established and involve all of the soft
tissues as well as the joint configuration. Repeated soft
tissue releases may be disappointing. We have success­
fully treated severe MP contractures with deviation
in Freeman-Sheldon, distal arthrogryposis, hemihy­
pertrophy, windblown hand, and other patients with
metacarpal osteotomy and shortening procedures. An
additional caveat for a patient with tight PIP flexion
contractures seen in cleft hand or foot, isolated fifth
digit camptodactyly, synpolydactyly, or radial dysplasia
is to look at the extensor side of the digit to rule out
a tight oblique retinacular ligament (ORL), attenuated
central slip, or tight lateral intrinsics. Rapid recurrence
of the contracture can be expected if these structures
are not adjusted or released.
Hypertrophic Scars and Keloids
Hypertrophic scars are clinically distinguished from
keloids because they do not extend outside the origi­
nal zone of injury. They are commonly seen in patients
with overgrowth conditions, CLOVES syndrome (con­
genital, lipomatous, overgrowth, vascular malforma­
tions, epidermal nevi, and spinal/skeletal anomalies
and/or scoliosis), Proteus syndrome, lymphatic malfor­
mations, and other syndromes.11,12 Our treatment has
been the same as that described in the chapter: inter­
lesional steroid or methotrexate. Steroids do work, but
many serial injections are necessary over a long period
of time. The side effects of hypopigmentation and soft
tissue atrophy are well known.
Keloids remain an unsolved problem. They may re­
spond to injections. We have used skin grafts as a last re­
sort in the hand only to find a clear picture frame of new
keloid on the border of the graft. Glabrous skin surfaces
with thick dermal layers generate consistently large le­
sions and are surfaces within which to avoid incisions.
Conclusion
Children are a very special group, and the treatment
of their upper limb anomalies is a unique subspe­
cialty, because what is often believed to be a satisfac­
tory outcome after early surgery will change during
growth. Adaptation, though remarkable at times, will
not provide a permanent correction. Some unfavorable
outcomes are avoidable, and some are not. We strongly
agree with the chapter regarding the primary culprits:
1. Lack of judgment with performing the wrong
operation
2. Poor timing
3. Technical failures
4. Unrealistic expectations for patient recovery and
adaptation.
Experience is cumulative in this field. Surgeons see a
limited number of specific case types because of the
increasing density of surgeons. Consequently, contin­
ued, compulsive follow-up and critical evaluation are
paramount if outcomes are to be improved. The most
important ingredient is knowledge of anatomy and the
variations seen in these patients. If a surgeon becomes
an astute clinical anatomist and pays attention to detail
in the treatment of these small hands, he or she will ex­
perience the joy of watching these children grow with
improved function.
Acknowledgment
It has been a pleasure to contribute to all of the editions
of The Unfavorable Result in Plastic Surgery, edited by our
friend, neighbor, mentor, and chief at the Beth Israel Dea­
coness Hospital for 30 years. The ever-curious Bob Gold­
wyn would be proud of the continued growth of this seed
he planted and nutured many years ago.

SECTION VIII THE HAND AND UPPER EXTREMITY
1156
References
1. Rayan GM, Upton J. Congenital Hand Malformations
and Associated Syndromes. Berlin and Heidelberg:
Springer Co; 2014
2. Upton J. Classification of Upper Limb Congenital dif­
ferences and General Principles of Management.
Chapter 202. In: Mathes S, Hentz VR, eds. Plastic Sur­
gery. 2nd ed. Philadelphia, PA: WB Saunders; 2006:
25–50
3. Buck-Gramcko D. Congenital malformations. In: Nigst
H, Buck-Gramcko D, Millesi H, Lister GD, eds. Hand
Surgery. Vol. 1. New York, NY: Thieme, 1988:22–23
4. Herring JA, Ed. Tachdjian’s Pediatric Orthopaedics. 4th
ed. Philadelphia, PA: Saunders Elsevier; 2008:1–490
5. Upton J. Management of disorders of separation-
syndactyly. In: Mathes S, Hentz VR, eds. Plastic Sur­
gery. 2nd ed. Philadelphia, PA: WB Saunders; 2006:
139–184
6. Upton J. Management of transverse and longitudi­
nal deficiencies (Failure of formation). In: Mathes S,
Hentz VR, eds. Plastic Surgery. 2nd ed. Philadelphia,
PA: WB Saunders; 2006:51–138
7. McDonnell MN, Koblar S, Ward NS, Rothwell JC,
­Hordacre B, Ridding MC. An investigation of cortical
neuroplasticity following stroke in adults: is there
evidence for a critical window for rehabilitation?
BMC Neurol 2015;15:109
8. Flatt AE. The Care of Congenital Hand Anomalies. St.
Louis, MO: CV Mosby; 1977:170–212
9. Littler JW. Hand, wrist and forearm incisions. In: Lit­
tler JW, Cramer LM, Smith JW, eds. Symposium on Re­
constructive Hand Surgery. Educational Foundation
of Plastic Reconstructive Surgery. Vol. 9. St. Louis,
MO: CV Mosby; 1974: 87–100
10. Upton J. Disorders of duplication. In: Mathes S, Hentz
VR, eds. Plastic Surgery. 2nd ed. Philadelphia, PA: WB
Saunders; 2006:215–264
11. Upton J. Hypoplastic and absent thumb, Chapter 208,
In: Mathes S, Hentz VR, eds. Plastic Surgery. 2nd ed.
Philadelphia, PA: WB Saunders; 2006:323–368
12. Upton J. Failure of differentiation and overgrowth.
In: Mathes S, Hentz VR, eds. Plastic Surgery. 2nd ed.
Philadelphia, PA: WB Saunders; 2006:265–322

1157
Flexor tendon surgery is a challenging area that has seen
modest progress over the years. Comparing the 1977 series
by Lister et al1 of two-strand repairs and simple circum­
ferential epitendinous sutures, combined with the Kleinert
rehabilitation regimens,1 with the 2011 data from Sandow
et al2 on four-strand repairs with early active mobilization
regimens reveals the rupture rates and functional results
to be almost identical. Our management has completely
changed, built on new scientific foundations, but our out­
comes have not changed dramatically. Around 25% of pa­
tients still have poor to fair outcomes with an approximate
5% rupture rate, even in the best centers. Surgeons seem
to have accepted this situation in hand surgery for trauma,
whereas it is unlikely to be acceptable in elective surgery.
With this range of outcomes, much more needs to be
known about why the results are disappointing. Over the
years, we have established that failures are related to both
the mechanics and the cell biology of tendon repair. The
challenge has been to achieve a strong repair in a biologi­
cally active structure that is able to withstand friction-free
tensile load at the earliest possible time point. It is also im­
portant to appreciate that failure occurs in the context of an
injured, inflamed, confined environment. Once this concept
is understood, it is possible to understand how to avoid un­
favorable results.
Restoration of functional outcome is a fine balance be­
tween strength of the repair to prevent rupture and mobiliza­
tion to prevent adhesions. Adhesions is a clinical description
of loss of range of motion that can be related to true adhesion
formation between the tendon and sheath or scarring and fi­
brosis of the surrounding structures. Early mobilization with
good functional glide is designed to prevent adhesions and
stiffness, but mobilization carries a risk of rupture. Paradoxi­
cally, gapping, elongation, and attenuation may also promote
adhesion because of a reduction of glide, secondary healing to
the surrounding tissues, and fibrosis between tendons.
Avoiding an Unfavorable Result
in Flexor Tendon Repairs
It is essential to understand the causes of failure to avoid
an unfavorable result. This requires a root cause analysis of
every case that deviates from a satisfactory outcome and
close cooperation between the surgeon, hand therapist,
and patient to ensure the best results can be achieved.

Flexor Tendon Injuries
Jason Wong, Wee Sim Khor, Martin Franz Langer,
and Duncan Angus McGrouther
CHAPTER
61

SECTION VIII THE HAND AND UPPER EXTREMITY
1158
leys repeatedly come into contact with the tendon. These
structures are constantly changing their cross-sectional
profile as they glide; hence it is easy to appreciate that any
disruption to the anatomy will result in functional impair­
ment (see Fig. 61.1). Tendon injuries by convention have
been subdivided into zones originally proposed by Verdan,3
although the junctions between them are not precise. Zone
I, for example, is defined as distal to the FDS insertion, and,
as it extends all along the middle phalanx, it incorporates
many different situations proximal, through, or distal to the
A4 pulley.
Tang4 realized that zone II could be better subdivided in
to zone IIa, IIb, IIc, and IId to highlight the complexities of
repair in this zone5,6 (Fig. 61.2). However, no classification
system characterizes the effect of whether injuries of the
tendon are in flexion or extension, or in tension or relax­
ation, and this ultimately dictates the degree of retraction
through the sheath.
Tendon acquires its tensile strength by having a hierar­
chal arrangement of collagen extending from its ultrastruc­
ture through to its macroscopic anatomy6–9 (Fig. 61.3). Colla­
gen has a triple helical arrangement, grouped as periodically
staggered fibrils to form fibers, which are then packaged as
bundles or fascicles, interspersed with proteoglycan mol­
ecules, which have a high water affinity.6 This arrangement
is necessary to allow the tendon to flex around corners, de­
form, and stretch, which can only happen if the individual
molecules and fibrils can slide over one another. Critical to
tendons’ longitudinal morphology is the epitenon covering,
which is a biologically active single layer of epithelial-like
Anatomy
Tendon is an extremely complex tissue that is rarely fully
appreciated in anatomical and surgical texts. Given the
ubiquitous nature of tendon injuries and pathologies, it is
surprising how little attention tendon has received in med­
ical training. Few surgeons realize that tendon has precisely
defined areas of anatomical specialization that include vas­
cularity, elasticity, and fibrocartilaginous differentiation, all
of which influence how the tendons heal at different points
along its length. There are multiple vincula that enter the
dorsal aspect of the flexor tendons in the zone II region,
which are surgically important for a variety of reasons:
•• Preventing the tendon from retraction when the tendon
is divided
•• Providing nutrition to the tendon as it heals
•• Bleeding into the sheath once the tourniquet is released,
acting as a source of clot and subsequent adhesion
formation
In “no-man’s-land” (zone II) of the flexor tendon sheath, the
flexor digitorum profundus (FDP) has surface areas of fi­
brocartilage that resist compression from the flexor digito­
rum superficialis (FDS) as this tendon splits to surround the
FDP between the bifurcation before rejoining at the Camper
chiasm (Fig. 61.1). This careful arrangement of the pulleys
allows the FDS to engage and lock the FDP in the hook grip.
Fibrocartilaginous differentiation is also found at the inser­
tion/entheses of the FDP and in the region where the pul­
Summary Box
External
factors
Patient factors
Clinical factors
Unfavorable results
Preoperative
Mechanism
of injury
Timing/delay
Concomitant
injury
Occupation
Personality
Comorbidities
Primary wound
management
Surgical delay
Surgical expertise
Clinical assessment
Infection
Wound breakdown
Adhesions
Stiffness
Poor scar
Scar contracture
Flexion contracture
Poor glide
Triggering
Quadriga
Intraoperative
Tolerance
Intraoperative assessment
Cleaning and débridement
Dissection
Surgical technique
Active and passive
assessment after repair
Postoperative
Patient compliance
with rehabilitation
Hand therapy
Causes of Unfavorable Results

1159
Chapter 61 Flexor Tendon Injuries
Biology of Healing
The current teaching of flexor tendon healing that refers to
extrinsic and intrinsic pathways gives the misconception
that these processes occur independently. The concepts
of intrinsic and extrinsic healing were borne from ex­
perimental studies that showed tendons needed external
vascularization through adhesions to heal (extrinsic) but
also could heal in an avascular nutrient-rich environment
(experimentally in a knee joint) without external factors
(intrinsic). The current paradigm acknowledges that these
complex cellular processes occur together.11 The tendon,
sheath, subcutaneous tissue, and skin are all injured in con­
tinuity through actual trauma or surgery. They therefore
heal in an overlapping fashion and if immobilized result
in a continuous mass of scar tissue. The process involves
the classic phases of wound healing, including coagula­
tion, inflammation, synthesis, and remodeling; however,
the different tissues heal at different rates. The peak of
skin and subcutaneous tissue collagen synthesis occurs
around 1 week, whereas tendon generates the most colla­
gen at around 3 weeks and therefore gradually increases in
strength. Protection from full active loading is required for
10 to 12 weeks. Normal tissue boundaries are breached and
fibrin clot between the tendon and sheath are infiltrated by
fibroblasts at around day 5, leading to new collagen depos­
ited as fibrous adhesions. To some extent this is unavoid­
able, and the role of therapy and motion is to disrupt or
lengthen any adhesions that form.
Like all wound-healing processes, there will be individ­
ual variations in the tendency to fibrosis. Prediction of those
at most risk remains elusive, but considering parallels with
scar patients, we can assume that individual patient phe­
notypes, mechanical forces, and inflammation are likely to
play a key role. In addition to this already complex biology,
sutures grasping the tendon ends create areas of high strain,
resulting in acellular tendon zones.
We have found that a suture grasping the tendon leads
to tenocyte necrosis and prolongs inflammation that can
lead to increased adhesion formation12 (Fig. 61.4).
The tendon healing process involves a significant re­
cruitment of cells throughout the various phases, which is
perceived as tissue swelling. In the early phases of repair
the tendon actually softens because of matrix ­enzymes
cells and meshed collagen, which contains the internal fibers
to give the tendon a tubular form.10 This epitenon in intrasy­
novial tendon is especially important to appreciate, because
it is a layer in which sutures place a hold in the tendon sub­
stance; this explains the different suture-holding capacities
of extrasynovial tendon and intrasynovial tendon.
2d
2c
2b
2a
1c
1b
1a
A1
A2
C1
A3
C2
C3
A4
A5
Fig. 61.2 Flexor tendon subzones of the digit. Subzones of injury in the hand are modified from the Tang subclassification of zone II and Moi­
eman and Elliot subclassification of zone I. Image adapted from Langer illustration. (Modified from Langer MF, Oeckenpöhler S, Hartensuer
R, et al. Ringbandrekonstruktion an der Hand. Der Orthopäde 2015;44(10):757–766.)
Fig. 61.1 Cross section of the flexor tendon in the digit and
the flexor digitorum profundus and flexor digitorum superficialis ar­
rangement.

SECTION VIII THE HAND AND UPPER EXTREMITY
1160
A Repair Strong Enough for Early Mobilization
Ex vivo biomechanical studies have shown that the strength
of repair is influenced by the number of core strands, su­
ture caliber, purchase length, and suture material.14 How­
ever, it is perplexing to find that meta-analyses of rupture
rates worldwide could not identify significant differences
between two-strand and multistrand repairs in clinical
studies,15 and complication rates over the last 30 years have
remained static.
Our surgical attempts to join severed ends with ­multiple
strands of suture are simplistic, as attempting to join
­analogous engineering structures, such as cables, end-to-
end is destined to fail. A suture-based repair experimentally
can resist load forces in the region of 40 to 80 N, ­although
gapping usually occurs at forces around 10 N, which falls
far short compared with normal tendon strength (usu­
ally greater than 150 N). The amount of force transmission
permissible during active mobilization must function well
within these values. However, the tensile strength achiev­
able at the outset will not necessarily reflect the pattern over
the next few days as the surrounding tissue swells, tendon
softens, and glide characteristics change. The configuration
of most suture repairs dramatically alters as soon as load is
placed through the repair. Urbaniak16 showed a considerable
loss of tensile strength 5 days after repair. This data is cur­
rently being overlooked in the search for an ever-stronger
suture repair configuration, because what happens to suture
tensile strength in clinical practice over time is unknown.
­liberated by the injury process and suturing. The increase
of strength above baseline in the tendon repair is only seen
after 2 weeks; up to this time it is reliant on the suture
strength alone.13 From this point on tendons gain their
structural organization and strength by tensile load; hence
active motion with some loading is a beneficial factor in the
repair process while collagen is being deposited. A care­
ful balance of rehabilitation with the biology is required
by allowing motion, with slowly increasing load over the
course of healing.
Fig. 61.4 Cell mapping of a single suture in a tendon (gold) with
resultant acellular zone formation (blue) and inflammatory cell in­
flux (red).
Fig. 61.3 Ultrastructure and microstructure of a partially damaged mouse flexor tendon. (a) Transmission electron microscopy cross ­section
of a mouse flexor digitorum profundus tendon. (i) Magnified view of undamaged epitenon, seen one cell layer thick. (ii) Magnified view of
undamaged endotenon and collagen fibers made up of fibrils. (iii) Damaged tendon epitenon showing proliferation of the surface cells.
(iv) Damaged endotenon, showing loss of organized bundle and fiber architecture. (This image courtesy of Yinhui Lu and Karl Kadler, whose
work is funded by the Wellcome Trust.)
100 μm
i
A
ii
iii
iv

1161
Chapter 61 Flexor Tendon Injuries
inexperienced emergency department staff, such as closed
FDP rupture related to avulsion from the distal phalanx or
division of one of the two tendons in the digit. Early ac­
curate diagnosis and treatment are the first line in avoiding
unfavorable results, although on rare occasions a primary
repair may still be successful even months after the injury if
the divided tendon has not retracted. In our experience the
longest time delay between tendon division and a primary
repair has been 10 years (Fig. 61.5).
Mechanism of Injury
Domestic injuries to the flexor tendons tend to be related
to mishandling of sharp instruments, slips, and episodes of
impaired judgment. Work-related injuries often arise from
heavy or violent machinery like circular saws. These are as­
sociated with high energy transfer and tend to have more
concomitant injuries, fractures, neurovascular injury, and
skin loss. High-energy saw-type injuries that macerate the
tendon ends result in poorer functional outcomes compared
with clean-cut blade injuries, which tend to have a bet­
ter outcome because of a more limited zone of trauma and
lesser degree of inflammation.17 For untidy injuries, early
intervention, adequate débridement, antibiotic therapy, and
additional hand therapy are needed to prevent a poor out­
come. Ensuring the injury is clean of debris, necrotic tissue,
and environmental contamination minimizes the chance of
infection, which is a serious early unfavorable outcome.
The Patient: Collateral Pathologies
Studies have shown that the typical demographic for
tendon injuries of the hand is young, active adults who
The key point is that the immediate suture strength as de­
termined from current cadaver or animal tendon studies is
very different to the in vivo strength at 5 days. More studies
to determine the dynamic accrual of tensile strength after
tendon repair are required to underpin a logical rehabilita­
tion program of motion and loading.
Rehabilitation programs need to institute motion to pre­
vent adhesion but avoid any resistance until the tendon has
regained functional tensile strength. Better techniques to
investigate tendon loading noninvasively would be beneficial
to tailoring rehabilitation regimens.
Preoperative Planning
Timing and Delays in Diagnosis and
Treatment
Functional outcomes tend to fare better with early repair
of flexor tendon injuries compared with delayed interven­
tions. When tendons have retracted and become adherent,
muscle fibers have been shown to shorten, and tendon
sheaths collapse. It is not clear at what precise point delay
will begin to have a negative effect on functional outcomes.
We have found that delays beyond 7 days are subsequently
related to a lesser total active range of motion.14 All tissues
appear nearest to normal on the day of injury. Within hours
there are profound biologic changes in both the wound and
tendon, with edema and inflammatory infiltrate. Clini­
cal services should ideally be organized to provide surgi­
cal repair within 24 hours to reduce the risk of infection
from contamination. However, in reality this can be diffi­
cult to provide because of delays in referral and availabil­
ity of acute services. Many tendon injuries are missed by
a
b
c
d
Fig. 61.5 Primary flexor tendon repair after 10-year delayed presentation.

SECTION VIII THE HAND AND UPPER EXTREMITY
1162
GA remains a common method of choice for analgesia in
upper limb surgery, because most anesthetists are familiar
with its use. RA is also contraindicated when there is clear
patient preference toward GA or if there is a local infection
at the site of RA injection.
RA has been shown to decrease postoperative nausea
and vomiting, reduce early postoperative analgesic require­
ments, and result in less time spent in the hospital. Patients
receiving RA ambulate sooner and commence normal oral
intake quicker. RA requires a longer induction time, but the
time in operating theater is no different from GA. If using a
separate room to institute the RA blockade, the total anes­
thetic time can be reduced even further, thereby minimizing
the delay between cases. The success rate of RA is dependent
on experience and increases with ultrasound control, which
also considerably reduces the complications of pneumotho­
rax, hematoma, and postoperative pain from needle damage
to the fascicles.
Long-standing concerns about adrenaline use in local an­
esthetic ring blocks of the finger and hand have led to wide­
spread avoidance out of fear of digital necrosis. This is still
evident in current textbooks, medical teaching, and surgical
practice. However, there is level 1 evidence demonstrating
that local anesthetic with adrenaline can be safely used with­
out the risk of digital infarction. The cases of digital necrosis
in the 1950s may be attributed to expired batches of pro­
caine before mandatory expiration dates were instituted. Pro­
caine degraded into a harmful acidic solution, which, when
injected, would lead to tissue necrosis by chemical damage.21
This has led to the development of wide-awake surgery
using adrenaline to control bleeding without tourniquet
control21 (Fig. 61.6). Wide-awake anesthesia has been de­
scribed to be safe and effective for flexor tendon surgery in
the hand with acceptably low rupture rates.20 Wide-awake
primary tendon repairs enable the surgeon and therapist to
assess range of motion and quality of the repair before skin
closure. Inadequate repair (poor glide, triggering, gapping, or
bunching) can be corrected immediately, and therapy regi­
mens can be adjusted to any intraoperative findings.
Although wide-awake surgery may have potential ben­
efits financially and surgically, uptake has proven to be slow.
The surgeon requires familiarity with local anesthesia in
conscious patients and familiarity with tendon surgery to
instill confidence in an anxious patient.
Incisions and Tendon Retrieval
Exposure of the tendon and other damaged structures is
an important part of the overall outcomes in flexor tendon
surgery (Box 61.1). Skin incisions should be carefully de­
signed to incorporate the existing laceration and develop
flaps that maintain good perfusion. Ensuring good primary
wound healing reduces the incidence of scar contracture,
another unfavorable result.
Our starting point is usually to incorporate the wound
with either a Bruner’s22 or midaxial incision, although varia­
tions of these incisions have been described. Generally there
will be a traumatic wound, which requires extension. Before
are relatively free of comorbidities; hence impairment of
functional and working capacity is significant.18 Patients’
behavior and compliance can have a major effect on out­
comes, so care must be taken to optimize the management
of anxieties, impatience, and motivational aspects of re­
covery. Surgeons should be wary of a patient whose injury
has been associated with drugs or alcohol. Curiously, failed
suicide attempts give little indication of motivation, and
yet many such patients work hard to reverse self-inflicted
damage, and the rehabilitation effort often distracts them
from their other life circumstances.
The Surgeon: Training and Experience
Depending on the hand unit organization, some centers man­
age flexor tendon injuries during day time by consultant
­supervised operating lists. Alternatively, trauma may be man­
aged outwith routine operating hours by less-experienced
surgeons. Some studies have shown that outcomes are poorer
with less-experienced surgeons, whereas others have failed to
demonstrate a difference.19 It is clear that there is huge varia­
tion in how surgeons approach the problem, and it is impor­
tant that within each center a surgical protocol matches the
rehabilitation regimens. Group-wide training and evaluation
of repairs in cadaveric or animal models is valuable for junior
trainees and should be mandatory. However, nothing is better
than supervised clinical experience to allow trainees to make
judgments about management of skin incisions, the pulleys,
and the associated damaged structures.
Patient Tendon Examination
The location of the tendon ends can be inferred from the
history and mechanism of injury. A digit engaged in grasp
at the time of injury is likely to have retracted, whereas
a tendon relaxed at injury does not retract far from the
wound. A meticulous examination commencing with ob­
servation of the normal cascade posture is also important
in preoperative planning. Several passive maneuvers on
the injured patient can provide an indication of which ten­
dons are divided, such as a tenodesis test (on supination
the wrist extends and digits flex if tendons are intact) and
forearm compression test. Isolated testing of distal inter­
phalangeal (DIP) joint and proximal interphalangeal (PIP)
joint movement is also of value but is usually hampered by
pain at the wound site.
Intraoperative Factors that Influence
Outcome
Anesthesia
Anesthetic options for flexor tendon surgery are general an­
aesthesia (GA); regional anaesthesia (RA) (brachial plexus
or regional nerve blockade, under ultrasound control); and
the wide-awake local anaesthetic, no tourniquet technique
(WALANT).20

1163
Chapter 61 Flexor Tendon Injuries
a zig-zag to prevent narrow-based flaps that lead to necrosis
(Fig. 61.7). The planning of incisions requires the surgeon to
weigh the surgical exposure for the repair against creating a
larger wound with the attendant problems of postoperative
swelling and later fibrosis. Large incisions for exposure will
lead to increased swelling during the immediate postopera­
tive period, which makes it harder to mobilize the digit and
can affect the rate of recovery.
Outcomes can be impaired by excessive dissection just
to find the ends. The surgeon should be able to judge where
the tendon ends lie and then plan the incisions, exposure,
and management of associated fractures, neurovascular in­
juries, or need for skin cover.
mapping out the incision line it is important to assess any
associated digital vascular injuries, because basing skin flaps
on divided vessels is likely to lead to skin necrosis. Flaps
should be broad-based and, as advised by Bruner,22 there
should be a linear extension from the wound before forming
2 mL
2 mL
10 mL
10 mL
Fig. 61.6 Recommended injection sites and volumes for wide-awake anesthesia with no tourniquet.
Box 61.1
Pitfalls for Incisions
•• Inadequate exposure on primary incision
requiring further extension and a large final scar
•• A midlateral incision that is too far volar and
therefore does not confer the safe sites for incision,
placing the digital neurovascular bundle at risk
•• Distal incisions that compromise flap viability
beyond the trifurcation of digital vessels
•• Skin flaps that are too thin, resulting in necrotic
flaps, suture pullout, or wound dehiscence
•• Crossing the skin crease with the incision as a
perpendicular line
Steps to Avoid These Pitfalls
•• Appropriate clinical assessment and history to
anticipate the position of the distal and proximal
stump in relation to the site of skin injury
•• Planning of the distal incision to allow adequate
purchase and pulley venting at the expected
distal stump site
•• Incisions distal to the digital vessel trifurcation
planned for the midline to preserve skin flap
vascularity
•• Skin flaps taken with fat in full thickness down
to the tendon sheath to preserve vascularity
Pitfalls for Incisions and Steps
to Avoid Them
Bruner’s
Midlateral
Limited incision
Bruner’s of thumb
Fig. 61.7 Flexor tendon approaches: Common incisions for exposure
of the digital flexor tendons.

SECTION VIII THE HAND AND UPPER EXTREMITY
1164
a
b
c
d
e
f
g
h
Fig. 61.8 Flexor tendon retrieval. (a) Exposure planned over the zone of injury extending distally and a separate incision at the distal pal­
mar crease. (b) The incisions made preserve skin over zone II. (c) A 6-Fr pediatric feeding tube is passed distal to proximal down the flexor
tendon sheath. (d) The tube is pulled through into the proximal wound, and a big grasping stitch with 3–0 Prolene is passed through ten­
don (small image in inset). (e) Needle cut of long suture length and both suture ends are passed into the eye of the feeding catheter until
it passes through the distal wound. (f) The suture ends are grasped and the tendon is snared onto the feeding catheter. (g) The tendon is
retracted back through the sheath and used as retraction while performing the repair with a new suture. (h) The snare and catheter are
removed after the repair has been performed. (Reproduced from Wong J, McGrouther DA. Minimizing trauma over ‘no man’s land’ with
flexor tendon retrieval. J Hand Surg Eur Vol 2014;39:1004–1006.)