Vạt Da Và Phẫu Thuật Tái Tạo (Flaps and Reconstructive Surgery)

CHAPTER 25 — Vessels
317
could be dilated with a vessel dilator. When placing the
beveled end into the clamp, position the end in such a way
that the cut surfaces are parallel for suturing. If the beveling
is made with increased obliquity, positioning the end may
pose a problem. When the size discrepancy is >3 : 1, it is
better to resort to end-to-side anastomosis. In extreme situ­
ations, when end-to-side anastomosis is not feasible, we have
used the technique depicted in Figure 25.12 successfully.
The larger size vessel can be tapered at its end (Fig. 25.13)
to minimize turbulence. Generally, if the inflow from the
recipient artery is small and the flap artery is large, or if the
outflow of the flap is small and the recipient vein is large,
fewer problems are encountered.
SUTURE TECHNIQUES
CORRECT POSITION OF THE HAND
Correct positioning of the hand is important to avoid tremor
and reduce fatigue. The instruments are held in the “tripod­
pinch” between the thumb, index, and the middle finger
and stacked upon the ring and the little fingers. The hand
and the forearm in turn are rested on towels so that the limb
is comfortable. Resting the wrist and forearm greatly reduces
the coarse tremor, and supporting the fingers holding the
Usually, two single clamps are used during the procedure.
When doing flaps in the head and neck, if jugular veins are
chosen as recipient vessels, the size mismatch to some flap
veins could be great and end-to-side is the preferred option.
In such instances, a side biting Satinsky clamp can be used.
It is enough to make a slit in such big veins and not necessary
to excise a bit of the wall, as is done in an arteriotomy.
TECHNIQUES FOR USE IN VESSEL
SIZE DISCREPANCY
Various techniques are used for discrepancies in vessel
diameter. Some surgeons manage up to a 3 : 1 vessel mis­
match by careful placement of sutures; however, we do not
recommend attempting a mismatch of more than 2 : 1 in this
manner. Larger mismatches in diameter of veins (up to 4 : 1)
can be managed more safely and effectively than large mis­
matches in arterial diameter. The technique involves placing
interrupted sutures farther apart on the larger vessel and
this must be performed from the first stitches of the anas­
tomosis. When the discrepancy is between 2 : 1 and 3 : 1, we
prefer beveling or spatulation of the vessel.
When beveling an edge, the oblique cut should not be
more than 30° (Fig. 25.11). After beveling, the vessel end
Figure 25.7  Steps of continuous suture techniques. (A) A suture is tied in the near end. (B) A suture is made at the far end and the vessel
stretched. (C) Running sutures are taken at regular intervals. (D) A knot is made when it reaches the near end. (E,F) The vessels are turned
then running sutures are continued to completion.
A
B
C
D
E
F

318
PART 3 — RECONSTRUCTION
Figure 25.8  (A) Spiral interrupted suturing. 1. A decrescendo spiral on the anterior surface of the anastomosis has been formed. 2. The spiral
was cut tangentially to form a series of sutures of decreasing length. 3. The sutures were tied individually. (B) 1. The needle-bearing end of
the stay suture was used to approximate the anterior walls in a continuous fashion. 2. A vessel dilator was used to spread the loops to produce
a decrescendo spiral. 3. The anastomosis was completed.
1
2
3
A
1
2
3
B
Figure 25.9  (A,B) Technique of arteriotomy with a stay stitch at the site.
A
B
Figure 25.10  (A) Sutures must be taken radial to the center of the arteriotomy. (B) If sutures were to be taken transversely, it would lead to
leakage at the corners.
A
B

CHAPTER 25 — Vessels
319
instrument avoids fine tremor. Obtaining the right hand
position is easy when operating on limbs, but with some
practice and innovation, it is possible to achieve it even
while operating on the head and neck, or the trunk. Acland
considers the introduction of this hand position (Fig. 25.14)
as one of the greatest milestones in the development of
microsurgery since it led to control of tremor and subse­
quent development of microinstruments.
DEALING WITH DISCREPANCY IN VESSEL
WALL THICKNESS
This situation frequently occurs when a flap from the upper
limb is transferred to leg defects. The inner luminal diam­
eter may be the same but the thickness of the wall of one
vessel may make the other appear very big. In such situa­
tions, A-type clamps are useful to prevent slipping of the
vessel. The sutures through the thicker vessel are taken in
a radial fashion, so that the sutures when tied are matching
to the thinner vessel. The suture distances on the external
surface of the thicker vessel when compared with the exter­
nal surface of the thinner vessel will be greater (Fig. 25.15,
between arrows) but will be corresponding when seen on
the luminal side. In addition, grabbing the same amount of
intima on the thin and thick wall vessels and less of the
media and adventitia on the thick wall vessel is helpful in
dealing with the wall thickness discrepancy.
SUTURING A VERTICALLY ORIENTED ANASTOMOSIS
The most difficult lie of the vessels for anastomosis is when
the vessel lumens are vertically oriented. If that is the case,
first determine if there is any possibility of changing the
position of the body part or adjust the table to make the
anastomosis more horizontally oriented. For example, if
dorsal arteries of the foot are to be anastomosed, in the
normal position the anastomosis will be vertically oriented.
By bending the knee and getting the foot flat on the table,
the anastomosis becomes very easy. In situations where this
cannot be done, the following technique can be used. Place
Figure 25.12  Suturing of vessels with severe size discrepancy.
(A) The back wall is sutured first, followed by (B) suturing of the front
wall to the near corner. (C) The technique of inserting the corner
suture. (D) Direct suturing of the remaining larger vessel.
A
B
C
D
Figure 25.13  Proper technique for beveling the edge of the vessel
when performing a microvascular anastomosis.
Figure 25.11  Technique of managing a vessel size mismatch when the discrepancy is between 2 : 1 and 3 : 1. (A) An oblique cut is made, not
to exceed an angle of 30°. (B) The cut end is advanced within the clamp in order to align the ends for anastomosis.
Not more than 30°
30°
A
B

320
PART 3 — RECONSTRUCTION
the vessels under tension. Vessels sutured under tension
result in gross narrowing of the lumen, causing turbulence
in blood flow and spasm. As Acland said, “The time wasted
in doing vein grafts is usually the time that you take to
decide that you need one.” If certain principles are adhered
to, it is quick, safe, and sure.
Success in Vein Grafts Starts with Harvesting
the Vein Graft
This is preferably done under tourniquet and small side
tributaries of veins must be ligated or coagulated as the vein
is harvested. When used for arteries, in a high-flow system,
small open vessels, which may not cause problems in a low-
pressure venous system, may cause troublesome bleeding.
The gap in the arteries must be measured and the appropri­
ate length of vein graft must be marked before it is cut. Vein
grafts shrink in length after division. If a longer length is
used, in the presence of high-pressure flow there will be
kinking of the graft. If that happens, there is only one sure
way to solve the problem: resect and redo one anastomosis.
Veins have valves and even a short segment of veins may
have valves. They must always be reversed when used for
arteries. The proximal and distal ends must always be
marked and this point must never be trusted to memory. It
is better to have a routine in every unit. In our unit, the
proximal end is divided between ligatures and the distal end
is divided after one ligature. When harvested, one end will
be open and the other end will have the ligature (Fig.
25.16).We have the dictum that the blood flows through the
open end. The open end will be connected to the proximal
artery. Sometimes a long vein graft may be harvested and
used for multiple segmental defects. If this protocol is used,
the first two sutures at 180° with stay sutures. The far side
or the back wall must be sutured first. Each bite must be
taken with visualization, since there is a high risk of taking
a through stich. Backhand suturing technique may be of use
when placing some sutures. After completion of the far side,
the near side is sutured.
PERFORMING A FLOW-THROUGH ANASTOMOSIS
When performing flow-through flaps, the flap must be posi­
tioned in such a way that both anastomotic sites have good
access. The flap is retained with some stay sutures and
usually the proximal anastomosis is done first. We do both
the arterial and venous anastomosis and then let the distal
clamp open to see the perfusion. The distal anastomosis is
then done in the usual fashion. If the distal anastomosis lies
in a deep cavity, the distal arterial anastomosis is done first,
since the flap can be positioned and manipulated to help
in the anastomosis.
PERFORMING VEIN GRAFTS
Three points are important when doing vein grafts:
• Choosing the correct length of the vein graft
• There should not be any twisting of the vein graft.
• It must be reversed when it is being used for arteries and
not be reversed if used for bridging the defect in the
veins.
Vein grafts are used when there is a gap. Two anastomoses
need to be made, compared with one anastomosis in direct
repair. Still, it is better and safer to do vein grafts than suture
Figure 25.14  (A,B) Correct positioning of the hands.
A
B
Figure 25.15  Dealing with discrepancy in vessel wall thickness (see
arrows).
Figure 25.16  Technique of ligating the vein when harvesting a vein
graft.

CHAPTER 25 — Vessels
321
SUTURE TYING TECHNIQUES
Several techniques have been described for tying a micro­
vascular anastomosis including the “airborne” technique.
This technique avoids the suture ends from getting stuck to
surrounding tissues, therefore speeding up the anastomosis
(Video 25.1).15
ARTERIOVENOUS VASCULAR LOOPS
When the location of the defect is far from the site of the
availability of healthy vessels (over 20 cm away), long vein
grafts or arteriovenous loops are used. This is particularly
the case in the reconstruction of traumatic lower limb
defects. A long segment of saphenous vein, which offers low
resistance, is used. If the ipsilateral saphenous vein is used,
the distal end of the saphenous vein is sutured end-to-side
to a high flowing artery, such as the superficial femoral or
the profunda femoris. In this way, there is only one anasto­
mosis. If a vein graft from the contralateral side is chosen,
the vein must be anastomosed to the saphenous vein and
the high flowing artery, hence two anastomoses are required.
The tip of the loop can be divided and immediately fol­
lowed with free flap transfer, or the loop can be allowed to
mature for some days prior to the free flap transfer. Lin
et al. reported 65 cases of free tissue transfer using long
arteriovenous fistulas or vein grafts. They compared the
outcome between patients who underwent arteriovenous
fistula followed by free tissue transfer in one stage, with
patients who underwent free flaps following arteriovenous
fistula in two stages. In the two stage group the rate of occlu­
sion of the graft, re-exploration rate, and flap failure rate
were higher.16,17
Though they can have significant complications, arterio­
venous loops offer a good alternative when native vessels are
not available near the defect site, as may occur following
trauma, in patients with peripheral vascular disease like dia­
betes and in post-irradiation defects. Giving a bolus dose of
heparin at the time of creating the loop and the use of
dextran 40 in the postoperative period for 5 days to prevent
thrombosis has been advocated. Creating the arteriovenous
loop between large vessels with good flow, especially in the
leg, will maximize blood flow.17–19
SUPERMICROSURGERY
Supermicrosurgery is defined as the dissection and anasto­
mosis of vessels ranging from 0.3 mm to 0.8 mm in diame­
ter. The technique was first introduced by Koshima and now
has a range of applications including lymphedema treat­
ment, replantation of amputated fingertips, and free tissue
transfer.20,21 Working with such small caliber vessels requires
ultradelicate microsurgical instruments and a high-power
microscope (up to ×50 magnification). Typically, 11-0 or
12-0 nylon sutures are required.22 By using such instru­
ments, the microsurgeon can perform procedures such as
lymphaticovenous anastomosis and perforator to perforator
anastomosis reliably. This has led to the advent of free per­
forator to perforator flaps, which vastly increases the options
when it comes to selecting a recipient vessel and obviates
the need for prolonged pedicle dissection.23
When choosing a recipient perforator, it is important that
visible pulsation is observed when the vessel is examined
under the microscope. This is vital to ensure good flow. This
one end will still have the ligature marking it. If vein grafts
are used for veins they must not be reversed.
Using the system of keeping the “flowing end” open helps
avoid mistakes. If inadvertently a mistake occurs, the only
way to correct it is to redo it.
Vein grafts when collapsed appear thin and it is possible
that they might be twisted in their course. If the proximal
end is anastomosed first, blood allowed to flow and then the
distal anastomosis is done, this mistake can be avoided. This
sequence will also help to prevent using an inappropriately
long length of the vein graft. When choosing vein grafts, it
is good not only to choose the correct length but also to
choose the same thickness. For digital vessels, the best
match is found on the dorsum of fingers or on the volar
aspect of the wrist; upper limb defects are best managed
with upper limb veins. Saphenous veins from the lower leg
have thick muscular walls and it is better to pre-dilate them
if used for upper limb arteries.
Vein grafts when used appropriately have increased
success rates in microsurgery but one must be very conver­
sant with using them. If a mistake is made in any of the three
important points initially mentioned (length of graft, twist­
ing of graft, non-reversal of graft when used for arteries),
there is one definite and sure way to correct the problem.
That is to simply redo the anastomosis.
SUTURING ATHEROSCLEROTIC VESSELS
The location of the anastomosis is chosen after inspecting
and palpating the recipient vessel. If a significant plaque is
found, another location is chosen on that vessel or another
vessel is selected. After selecting the place with the least
amount of plaque, the artery is opened in a way that does
not allow for the vessel wall to collapse (a small area is
excised). It is important to make sure that the inside of the
vessel is clean and that there are no intimal flaps in that
location. Irrigation with heparinized saline is performed. It
is important to select the appropriate clamps that do not
put too much tension on the vessel. Excessive tension can
create fractures in the plaque or can induce damage to the
vessel wall. Vessel loops can be used on the proximal and
distal aspect of larger vessels if necessary.
A round body needle can be useful in avoiding damage
to the intima. It is best to choose the smallest microsuture
because the size of the needle and suture can be a factor in
the success of the anastomosis. It is preferable to perform
interrupted sutures, as this allows for perfect placement of
the needle with every bite. Additionally, it allows the surgeon
to see the intima and make sure that it is included with every
bite. The sutures should be passed from the inside of the
lumen of the atherosclerotic vessel to the outside in order
to avoid separating the intima from the rest of the vessel
wall. Sutures with double needles can help achieve this
purpose if both ends of the vessels have plaques. Gentle
handling of the vessels is extremely important and every
move should be purposeful and accurate without ever
placing the instruments into the vessels blindly. Vessel dilata­
tion is dangerous and if necessary, should be performed
with extreme care. Eversion of the vessel walls avoids leaving
raw surfaces exposed to blood flow. During the anastomosis,
the last 3–4 sutures can be tied at the end in order to be
able to visualize the lumen with each bite. The sutures
should be tied without too much tension, as tension can
cause erosion into the plaque and tearing of the vessel wall.

322
PART 3 — RECONSTRUCTION
Access video and further reading online at

REFERENCES
1. Jacobson JH 2nd, Suarez EL. Microvascular surgery. Dis Chest
1962;41:220–4.
2. Masia J, Olivares L, Koshima I, et al. Barcelona consensus on super­
microsurgery. J Reconstr Microsurg 2014;30(1):53–8.
3. Acland RD, Sabapathy SR. Practice manual for microvascular
surgery. 3rd ed. Coimbatore: Indian Society for Surgery of the
Hand; 2008.
4. Hallock GG. Liability of recipient vessels distal to the zone of injury
when used for extremity free flaps. J Reconstr Microsurg 1996;12:
89–92.
5. Kolker AR, Kasabian AK, Karp NS, et al. Fate of free flap micro­
anastomosis distal to the zone of injury in lower extremity trauma.
Plast Reconstr Surg 1997;99:1068–73.
6. Fernandez EJ, Nadal RD, Gonzalez SM, et al. The effect of stasis
on a microvascular anastomosis. Microsurgery 1983;4:176–7.
7. Ardehali B, Morritt AN, Jain A. Systematic review: anastomotic
microvascular device. J Plast Reconstr Aesthet Surg 2014;67(6):
752–5.
8. Watson JT. Couplers in arterial and venous repairs. J Hand Surg
[Am] 2013;38(7):1423–6.
9. Grewal AS, Erovic B, Strumas N, et al. The utility of the microvas­
cular anastomotic coupler in free tissue transfer. Can J Plast Surg
2012;20(2):98–102.
10. Jandali S, Wu LC, Vega SJ, et al. 1000 consecutive venous anasto­
moses using the microvascular anastomotic coupler in breast
reconstruction. Plast Reconstr Surg 2010;125(3):792–8.
11. Wong B, Patel N, Durrani A. The use of venous coupler device in
end to side anastomoses in head and neck free tissue transfer
(FTT). Int J Surg 2013;11(8):681.
12. Maraka J, Hausien O, Abreo N, et al. The microvascular coupler
device for end-to-side venous anastomosis in free tissue transfer for
head and neck reconstruction. J Plast Reconstr Aesthet Surg 2014;
67(12):1757–8.
13. Draper LB, Levine JP, Ahn CY. Routine use of microvascular cou­
pling device for arterial anastomosis in breast reconstruction. Ann
Plast Surg 2006;56(4):365–8.
14. Godina M. Preferential use of end-to-side arterial anastomosis in
free flap transfers. Plast Reconstr Surg 1979;64:673–82.
15. Chen HC, Mardini S, Tsai FC. “Airborne” suture tying technique
for
the
microvascular
anastomosis.
Plast
Reconstr
Surg
2004;113:1225–8.
16. Ethunandan M, Cole R, Flood TR. Corlett loop for microvascular
reconstruction in a neck depleted of vessels. Br J Oral Maxillofac
Surg 2007;45(6):493–5.
17. Lin CH, Mardini S, Lin YT, et al. Sixty-five clinical cases of free
tissue transfer using long arteriovenous fistulas or vein grafts.
J Trauma 2004;56(5):1107–17.
18. Angel MF, Chang B, Clark N, et al. Further clinical use of the
interposition arteriovenous loop graft in free tissue transfers.
Microsurgery 1993;14(8):479–81.
19. Lind B, McCarthy W, Derman G, et al. Arteriovenous loop grafts
for free tissue transfer. Vasc Endovascular Surg 2012;46(1):30–3.
20. Nagase T, Gonda K, Inoue K, et al. Treatment of lymphedema
with lymphaticovenular anastomoses. Int J Clin Oncol 2005;10(5):
304–10.
21. Koshima I, Inagawa K, Yamamoto M, et al. New microsurgical
breast reconstruction using free paraumbilical perforator adiposal
flaps. Plast Reconstr Surg 2000;106(1):61–5.
22. Mihara M, Hayashi Y, Iida T, et al. Instruments for supermicrosur­
gery in Japan. Plast Reconstr Surg 2012;129(2):404e–406e.
23. Koshima I, Yamamoto T, Narushima M, et al. Perforator flaps and
supermicrosurgery. Clin Plast Surg 2010;37(4):683–9.
must be confirmed by transection of the recipient perfora­
tor prior to final elevation of the flap. Inadequate flow may
necessitate a further dissection of the perforator beyond the
deep fascia, or using a different perforator. Careful use of
a vessel dilator to enlarge to the size of the lumen is also
necessary to allow adequate lumen size for suturing.
Supermicrosurgery is a technique that requires a great
deal of skill and a steep learning curve for the microsurgeon
to master. However, the possibilities it affords means it will
no doubt become more mainstream in the years to come.
CONCLUSION
Successful anastomosis of a vessel of 1 mm diameter com­
prises the execution of a series of well-defined steps. Could
any step be designated as more important than another?
Barring a “through stitch” where both walls are caught by a
suture, which definitely is fatal to the anastomosis, none of
the other steps can be said to be more important than
another. Taken individually, each component will allow sig­
nificant latitude of freedom. Nonadherence to the rules of
a single step is usually not the cause of a clinical failure. It
is the summation of many minor compromises that leads to
a major disaster. Hence, one needs to cultivate a sense of
discipline in performing every step correctly and never pro­
ceeding to the next step unless the previous step has been
done to one’s satisfaction. In the years to come, even if
mechanical devices take over the suturing of small vessels,
basic microsurgical skills will become even more important,
since only they will rescue a microsurgeon from a demand­
ing situation.
Expert Commentary
Joon Pio (JP) Hong, MD, PhD, MMM
Since the first introduction of vessel repair and microsurgery,
we have evolved into the era of supermicrosurgery suturing
ever smaller vessels. This evolution was possible in part by
state-of-the-art microscopes and sutures. A magnification
power of 5 times and more has paved the way to easier and
meticulous dissection. New microscopes are being devel-
oped to allow physiological evaluation of the flap immediately
after anastomosis of the vessels, leading to better predict-
ability of the results. Development of 11-0 and 12-0 sutures
overcame the modern barrier of microsurgery, allowing us to
suture vessels as small as 0.4 mm. Understanding and utiliz-
ing the recent advances in technology involved in microsur-
gery will allow you to challenge more difficult and complex
reconstructions.
Despite these technological breakthroughs, the core prin-
ciples of vessel selection, preparation, and technique have
changed very little. Different methods of practice have been
introduced, including artificial and virtual models, and these
pre-clinical practices have played a role in easing the learning
curve and obtaining predictable clinical results for learners.
Today, other forms of anastomosis are readily available, such
as coupling devices. These devices increase the efficiency
for microsurgery and reduce the stress involved in anasto-
mosis for vessels. However, one must be comfortable using
the suturing technique and must learn the basics before
moving on to using simpler anastomotic devices, as this will
remain the basics for anastomosis until we overcome the
limits of these devices in the future.

CHAPTER 25 — Vessels 322.e1
FURTHER READING
Bodor R, Yoleri L, Zhang F, et al. Blood-flow velocity as a factor in
postoperative microvascular patency. J Reconstr Microsurg 1997;
13A:463–70.
Chen HC, Chuang CC, Chen S, et al. Selection of recipient vessels for
free flaps to the distal leg and foot following trauma. Microsurgery
1994;15:358–63.
Chen HC, Coskunfirat OK, Ozkan O, et al. Guidelines for the optimiza­
tion of microsurgery in atherosclerotic patients. Microsurgery
2006;26(5):356–62.
Chen L, Chiu DT. Spiral interrupted suturing technique for microvas­
cular anastomosis: a comparative study. Microsurgery 1986;7(2):
72–8.
Chow SP, Zhu JK, So YC. Effect of bipolar coagulation and occlusion
clamping on the patency rate in microvascular anastomosis. J Recon­
str Microsurg 1986;2:111–15.
Delacure MD, Kuriakose MA, Spies AL. Clinical experience in end-to
side venous anastomoses with a microvascular anastomotic coupling
device in head and neck reconstruction. Arch Otolaryngol Head
Neck Surg 1999;125:869–72.
Fernandez EJ, Nadal RD, Gonzalez SM, et al. The effect of stasis on a
microvascular anastomosis. Microsurgery 1983;4:176–7.
Hallock GG. Liability of recipient vessels distal to the zone of injury
when used for extremity free flaps. J Reconstr Microsurg 1996;
12:89–92.
Kolker AR, Kasabian AK, Karp NS, et al. Fate of free flap microanasto­
mosis distal to the zone of injury in lower extremity trauma. Plast
Reconstr Surg 1997;99:1068–73.
Lorenzetti F, Suomien S, Tukiainen E, et al. Evaluation of blood flow
in free microvascular flaps. J Reconstr Microsurg 2001;17:163–7.
Nam DA, Roberts TL II, Acland RD. An experimental study of end
to side microvascular anastomosis. Surg Gynecol Obstet 1978;147:
339–42.
Neligan PC, She-Yue H, Gullane PJ. Reverse flow as an option in micro­
vascular recipient anastomoses. Plast Reconstr Surg 1997;100(7):
1780–5.
Park S, Han SH, Lee TJ. Algorithm for recipient vessel selection in free
tissue transfer to the lower extremity. Plast Reconstr Surg 1999;
103:1937–48.
Samaha FJ, Oliva A, Buncke GM, et al. A clinical study of end-to-end
versus end-to-side techniques for microvascular anastomosis. Plast
Reconstr Surg 1997;99:1109–11.
Shaffer JW, Kling TF, Wilber RG, et al. Patency of anastomosed arteries
and veins subjected to reapplication of microvascular clamps. J Hand
Surg [Am] 1984;9A:73–5.
Sullivan SK, Dellacroce F, Allen R. Management of significant venous
discrepancy with microvascular venous coupler. J Reconstr Microsurg
2003;19:377–80.
Tetik C, Unal MB, Kocaoglu B, et al. Use of continuous horizontal
mattress suture techniques in microsurgery: an experimental study
in rats. J Hand Surg 2005;30A:587–95.
Yajima K, Yamamoto Y, Nohira K, et al. A new technique of microvas­
cular suturing: the chopstick technique. Br J Plast Surg 2004;57:
567–71.
Zeebregets CJ, Kirsch WM, van den Dungen JJ, et al. Five years’ world
experience with non penetrating clips for vascular anastomosis. Am
J Surg 2004;187:751–60.
Zhang F, Pang Y, Buntic R, et al. Effects of sequence, timing of vascular
anastomosis and clamp removal on survival of microsurgical flaps.
J Reconstr Microsurg 2002;18(8):697–702.
Video 25.1 Airborne suture tying technique.

e62
PURPOSES OF SKIN FLAP SURGERY
The purposes of flap transfer are classified into four catego­
ries: surface coverage; volume restoration; framework cre­
ation; and functional reconstruction. Category one is surface
coverage of the defect without the need for significant thick­
ness. In clinical situations, since every flap has some degree
of thickness, surface coverage inevitably is associated with
provision of volume (category two). The third category is recon­
struction of a framework. Transfer of vascularized bone, such
as the iliac bone or fibula, is classified into this category. The
fourth category is functional reconstruction, aiming to restore
defective function. Muscle flap transfer for reanimation of
the face and vascularized nerve flap transfer are examples
of this category.
In this chapter, the first, second, and third categories of
flap transfer, especially in relation to free flap transfer, are
described, with a focus on adaptation of the shape to the
needs of the recipient defect.
GEOMETRY OF THE DONOR SITE AND
THE FLAP
From a geometric viewpoint, the flap can be regarded as a
3-dimensional model specified by the XYZ coordinates (Fig.
26.1). Each axis corresponds to the specific properties of
the flaps for clinical use. The origin O indicates the surface
projection point of the vascular pedicle entry point to the
flap mass, and the XY plane coincides with the surface of
the skin flap. In the XY plane, the Y-axis represents the
direction of approximate axiality of the flap circulation. The
Z-axis characterizes the thickness of the skin flap. Point P
corresponds to the branching point of the pedicle vessel of
the flap from the main vessel and is usually found deep to
the XY plane. Point P′ is the projection of point P to the XY
plane. The distance between points P′ and O denotes the
approximate length of the vascular pedicle of the flap.
Clinically, the vascular pedicle of the flap will contract
after elevation and separation of the flap from the donor
site. Since anastomosis between the recipient and donor
vessel should be performed without tension, it is preferable
to underestimate the potential length of the pedicle. There­
fore, the distance P′O, which is actually shorter than the
length of the pedicle that is harvested, represents the
approximate clinical value.
In summary, the morphologic features of the flap can be
described by 3-dimensional variables (X, Y, Z) as well as the
distance and direction of P′O.
GENERAL PRINCIPLES IN SHAPING AND
MOLDING A FLAP
FREE FLAP RECONSTRUCTION
The first step in flap surgery is to evaluate which categories
of reconstruction, surface, volume, framework, or function
are required for the specific defect. For example, breast
reconstruction requires considerable volume as well as
surface coverage, whereas reconstruction for a burn scar
contracture of the neck needs surface coverage with minimal
volume, and a craniomaxillofacial defect frequently requires
some skeletal framework with volume reconstruction.
The second step is to choose the recipient vessels to be
anastomosed easily and safely, and estimate the required
length of the flap’s pedicle vessel. Since the maximum
length of the pedicle vessel for each specific flap to be har­
vested is nearly constant, flap selection should take this
important criterion into consideration.
The third step is to assess the match in color and texture
of the flap and the recipient site, and to balance the out­
comes that can be anticipated at the recipient site with the
morbidity that will be created at the donor site.
The donor site for the flap is selected after consideration
of these detailed requirements for reconstruction. Once the
donor site for flap transfer has been decided, the position
of point O, which is detected using the Doppler flowmeter
or estimated from the anatomic surface landmarks, is
marked on the donor skin. In addition, the X and Y axes
are marked across point O with reference to the approxi­
mate axiality of the blood flow in the skin, which is usually
parallel to the major axis on the extremities and to the
direction of the ribs on the back and the chest wall, radiat­
ing from the umbilicus on the abdomen, and radiating from
the femoral triangle in the inguinal area.
The location of point P is predicted from the palpable
point of the major arteries from the surface and point P′ is
also marked on the XY plane of the donor skin. For example,
the axillary artery gives off the subscapular artery that pro­
vides blood supply to the scapular flap and the thoracodor­
sal artery that provides blood supply to the latissimus dorsi
myocutaneous flap; the femoral artery gives off branches
Thinning and Tailoring
Naohiro Kimura
26 
PART 3  —  RECONSTRUCTION
SECTION B  —  FLAP HANDLING AND TECHNICAL TIPS

CHAPTER 26 — Thinning a nd Tailoring
e63
clinical situation. This is because twisting of the vascular
pedicle results from rotation of the flap and should be
limited to prevent interference with blood flow. Generally,
rotating the flap 90° or less does not obstruct the vascular
lumen.
If the angle of rotation exceeds 90°, the pedicle of the
flap should be positioned to form an arc in order to avoid
a harmful twist. However, this move shortens the arc of rota­
tion (Fig. 26.4). Therefore, the flap should be made longer
to compensate for loss of pedicle length created by forming
an arc.
THINNING OF THE FLAP
Another kind of flap tailoring is performed along the Z-axis
of the flap. It is frequently found that part of the flap needs
volume reduction or the entire flap should be made as thin
as possible
There are two approaches for removing excess adipose
tissue from a skin flap while preserving adequate circula­
tion. A thinning method that is regarded as conventional is
Figure 26.1  Three-dimensional description of the flap. The X and
Y axes are on the skin surface of the donor site. The Y represents
the axiality of the circulation of the specific donor site. The point O
represents the vascular insertion point to the flap projecting to the
surface of the skin. The point P is the branching point of the main
vessel to the flap pedicle, and the point P′ is the projection of the
point P to the XY plane.
Z
O
P
P
X
Y
Figure 26.2  The XY plane of each specific donor site of the flap.
(A) Lateral upper arm flap. (B) Thoracodorsal artery perforator flap.
(C) Deep inferior epigastric artery perforator flap. (D) Groin flap. (E)
Tensor fascia lata perforator flap. (F) Anterolateral thigh flap. (G)
Saphenous flap. (H) Peroneal flap.
Y
Y
Y
Y
Y
Y
Y
Y
X
X
X
X
X
X
X
X
O
O
O
O
O
O
O
O
P
P
P
P
P
P
P
A
B
C
D
E
F
G
H
such as the superficial circumflex iliac artery that supplies
the groin flap, and the deep femoral artery and the lateral
circumflex femoral artery that supply the anterolateral
thigh flap. Furthermore, the origin of the deep inferior
epigastric artery that supplies the rectus abdominis muscle
is from the femoral artery; and the popliteal artery gives off
the peroneal artery nourishing the fibula (Fig. 26.2).
Tailoring of the flap begins by deciding on the outline of
the flap to be elevated in the X, Y, Z coordinates of the
donor site by projecting the 3-dimensional relationship
between the recipient defect and the relative position of the
recipient vessels. If the recipient defect requires a flap with
a folded structure, such as reconstruction for an oropalato­
laryngeal defect, development of the structure is similarly
projected to the donor site. For a successful transfer of the
flap, the flap’s pedicle vessels must be kept in a natural posi­
tion after transfer to the recipient bed. Tension and twisting
of the anastomosed vessel should be strictly avoided.
Therefore, the 3-dimensional relationship is projected so
as to adjust the expected position for anastomosis to the
recipient vessel at point P. At the same time, vector OP′
should be positioned in nearly the same direction as the
venous flow in the recipient vessel. As a result, the coordi­
nates of the outline of the flap are determined. Arrangement
of the recipient and donor vessels is an essential component
of microsurgical tissue transfer (Fig. 26.3). In particular,
incorrectly arranged veins tend to kink and disrupt the
outflow of blood following revascularization of the flap.
Furthermore, it is preferable to adjust the major length
of the defect to the Y-axis on the donor site, especially in
the case of a large flap. If the recipient requirement obvi­
ously exceeds the surviving area based on one axial vessel,
combining a neighboring region of tissue based on another
axial vessel is advisable.
PEDICLED FLAP RECONSTRUCTION
For reconstructions using a pedicled flap, distance P′O + OY
represent the ideal maximum distance of the flap transposi­
tion. However, this distance becomes shorter in the actual

e64
PART 3 — RECONSTRUCTION
performed by secondary removal (in a subsequent proce­
dure) of adipose tissue after elevating a full-thickness skin
flap above the deep fascia.1–5 In this procedure, the location
of the vascular pedicle is predicted, and removal of adipose
tissue is limited to the area with the least expected vessel
distribution. Therefore, this is something of a blind proce­
dure in the strict sense and can sometimes jeopardize flap
circulation. However, most flaps are thinned using this
method, as the procedure is easy to perform. In contrast,
adipose tissue surrounding areas with a predicted high vas­
cular density cannot be removed by this method. Conse­
quently, a small amount of excess adipose tissue inevitably
remains in this type of flap thinning (Fig. 26.5).
Another method of thinning is performed by the micro­
dissection procedure.6–8 Microdissection represents the dis­
section of the blood vessels within the adipose layer under
operative microscopic magnification, and, through this
technique, vessel distribution in the adipose layer of the flap
can be directly identified. In addition to identification of
Figure 26.3  Example of flap planning (maxillofacial defect reconstructed with a rectus abdominis myocutaneous flap). Two-dimensional
outline of the necessary size of the flap is projected to the XY plane of the donor site considering the necessary length and adequate direction
of the pedicle vessel.
Tissue defect
Recipient vessel
Necessary length
of the pedicle
X
Y
P
P
O
Necessary size of the flap
Figure 26.4  Theoretical distance of flap transposition on condition
that the junction of the pedicle is fixed.
Figure 26.5  Conventional thin flap. A small amount of the adipose
tissue around the pedicle vessels cannot be removed.
the blood vessels within the adipose layer, the adipose tissue
surrounding the vessel is removed simultaneously by using
the microdissection technique. The flap is elevated in a one-
step procedure without damaging the blood vessels that
traverse the adipose layer. It should be noted that the micro­
dissected thin flap is not fabricated by thinning after elevat­
ing the full-thickness flap, but is elevated directly from the
donor site after the microdissection (Fig. 26.6).
Before the development of microdissection, there were
some reports of indirect investigation of the vascular distri­
bution in the adipose layer.9–12 These studies revealed that
the main vessel of the skin flap gives off some branches in
the middle of the adipose layer and these branches are
distributed into the subdermal network of the skin. However,
microdissection enables the vascular distribution in the
adipose layer to be observed directly. Based on the consider­
able number of observations from microdissection, the main

CHAPTER 26 — Thinning a nd Tailoring
e65
Through this small incision, suprafascial dissection is
carried out toward point O. Once the vascular bundle is
encountered at the hiatus of the deep fascia, an adequate
operative field for microdissection is prepared using a Gelpi
retractor. This retractor is a very convenient tool for spread­
ing between the fascia and the adipose tissue at the begin­
ning of the microdissection. Moreover, as the dissection
proceeds, the retractor can also spread adipose tissue to
expose the dissected vessels continuously (Fig. 26.9).
In the microdissection procedure, the vessel dissection
technique is somewhat different from that of the usual mac­
roscopic dissection. The vessels in the adipose layer are
isolated sharply and bluntly from the surrounding adipofas­
cial tissue in the usual macroscopic dissection, whereas the
surrounding adipofascial tissues are removed from the
bundle of the vessels and their branches in the microdissec­
tion procedure. As a result, a small hole surrounding the
thin vessels and their branches is made in the adipose layer
of the flap (Fig. 26.10).
Figure 26.7  The perforator of the groin flap emerges from the deep
fascia covering the sartorius muscle near the origin of the muscle.
Figure 26.8  Elevation of the microdissected thin tensor fascia
lata (TFL) perforator flap. An initial incision is made on the medial
margin of the temporal design of the flap. After microdissection (MD),
actual insertion points of the branched perforator vessel to the thin
layer of the flap are marked on the skin surface, and the outline of
the flap is corrected, based on these points.
Figure 26.6  Microdissected thin flap. Minimal adipose tissue is
observed around the bifurcation of the perforator vessel, and a con­
siderable amount of the adipose tissue remains on the donor site.
vessel nourishing the flap gives off some branches in the
adipose layer, and these branched vessels also have sufficient
thickness to tolerate the microsurgical manipulations.
Moreover, if only the microdissected vessels in the adipose
layer are preserved, a wide flap can be safely transferred
without excess adipose tissue.7 In the following section, the
specific technique of microdissection is described.
MICRODISSECTION
The blood circulation of most skin flaps is supplied from a
small bundle of axial vessels that penetrate the deep fascia
at the hiatus from the inside, which is represented as origin
O on Figure 26.1. Accordingly, dissection above the deep
fascia and detection of these vascular bundles constitute the
first procedure needed to prepare the microdissected thin
flap. This procedure is common to both the “perforator”
and “non-perforator” flaps. The only difference between
these two kinds of flaps is the tissue through which the
vessels traverse before penetrating the deep fascia.
This small vascular bundle consists of an artery and one
or two veins surrounded by connective tissue and a small
amount of adipose tissue. The bundle thickness usually
is approximately 2–3 mm at the hiatus of the deep fascia
(Fig. 26.7). This thin vascular bundle gives off several
branches and supplies blood to the subdermal vascular
network of the thin flap. Usually, this vascular bundle is
identified preoperatively using a Doppler probe; however,
the correct insertion points of branches into the thin layer
of the flap are sometimes far from the point of initial detec­
tion and are not identified precisely until completion of the
microdissection.
In the actual surgical procedure of elevating the microdis­
sected thin flap, the outline of the flap is temporarily
designed preoperatively around the origin O on the XY
plane of the donor site as described above. The initial inci­
sion for the suprafascial dissection is along the medial side
of the flap design and is about 10 cm in length. If a small
flap is required for the reconstruction, the incision is made
about 5 cm medially from this line, since the actual design
of the flap is modified according to the actual position of
the insertion point of the branches identified after the
microdissection (Fig. 26.8).

e66
PART 3 — RECONSTRUCTION
elevation of the flap is performed between the layer of small
and large fat lobules. Hence, a thin flap is elevated in a one-
step procedure after microdissection (Fig. 26.12), and a
considerable amount of the adipose tissue remains on the
donor site after elevating the flap.
ANATOMY OF THE INTRA-ADIPOSE VESSEL
Generally, as described above, the vascular bundle gives off
several branches in the adipose layer, and there are consid­
erable variations in these distributions. From a taxonomic
point of view, these variations can be classified into three
patterns according to the following branching patterns:
1. Spreading of the branches is minimal before they meet
the subdermal vascular plexus.
2. The branches spread out widely in the adipose tissue.
3. The vascular bundle itself runs obliquely after penetrat­
ing the deep fascia and the branches from the bundle
also spread obliquely in the adipose tissue.
Additionally, these patterns of vascular distribution are
closely related to the type of tissue through which the vessels
run before penetrating the deep fascia. Pattern 1 corre­
sponds to the vessel penetrating the muscle. Pattern 2 cor­
responds to the vessel running through the intermuscular
Generally, the adipose layer of the flap is composed of
the fat lobules and a network of fibrous tissues, which play
an important role in maintaining the structure of the integ­
ument. Since these adipofascial structures have a layered
arrangement, the microdissection proceeds by sharply dis­
secting the fascia tissue and bluntly removing the fat lobules
repetitively. As the procedure progresses, the size of the fat
lobules is reduced considerably near the superficial layer of
the adipose tissue. Microdissection into the adipose layer is
finished at this layer to preserve the subdermal vascular
network (Fig. 26.11). At this point, the insertion points and
the direction of each branched vessel into the subdermal
vascular network can be confirmed; therefore, the actual
design of the flap outline can be adjusted according to these
findings after completion of the microdissection.
Dissection of the pedicle vessels deep to the deep fascia
is also performed under microscopic magnification. After
dissecting the full length of the pedicle of the flap, the
entire flap is elevated from the circumference toward the
hole prepared by microdissection in the adipose layer, and
Figure 26.9  The perforator of the flap in Figure 26.8, which is
approximately 2.5 mm in thickness. The operative field for the micro­
dissection is prepared, spreading the suprafascial plane by two Gelpi
retractors.
Figure 26.10  After the microdissection, a small hole is made around
the perforator. Elevation of the flap is easily performed, dissecting
from the surrounding tissue toward the central hole of the flap into
the thin layer of the adipose tissue.
Perforator
Deep fascia
A hole after microdissection
Elevation of the thin flap
Figure 26.11  Completion of the microdissection of the perforator in
Figure 26.8.
Figure 26.12  The elevated microdissected thin tensor fascia lata
perforator flap.

CHAPTER 26 — Thinning a nd Tailoring
e67
The safe thickness of both types of thin flaps is also con­
sidered to be the same. The recommended thickness of a
thin flap is just beneath the superficial adipose layer, which
is characterized by smaller fat lobules than those of the deep
adipose layer.
SPECIFIC PRINCIPLES IN SHAPING AND
MOLDING A FLAP CONSIDERING INTRA-
ADIPOSAL VESSEL ANATOMY
FOLDING AND SPLITTING A FLAP
For flaps that are not laid on a flat defect, such as in the
case of craniomaxillofacial reconstruction, folding and
tubing and other maneuvers are sometimes necessary. If the
flap is to be folded 180°, the likelihood of disturbing blood
flow is high. Therefore, in an ideal situation, both segments
of a folded flap would be nourished by two separate perfora­
tors coming off the same pedicle. This would allow the
surgeon to maintain adequate blood flow to the entire flap
after folding because of the independent blood flow. Addi­
tionally, when two separate perforators nourish the folded
flap, the flap can be split into two segments that will allow
for further freedom in molding the flap.
Regarding the myocutaneous flap, the skin over the
muscle can be divided freely as long as it is attached to the
muscle belly. In contrast, in the skin flap, flap separation
should be based on the independent vessel to each part of
the flap. To consider this issue, perforator flaps in the thigh
area will be investigated as a specific example.
If two flaps are nourished independently by two separate
perforators, such as the ALTP and TFLP flaps, transfer is
possible by either combining the flaps or using two inde­
pendent flaps. Furthermore, in the ALTP flap, another
distal perforator of the lateral circumflex femoral artery can
nourish the more distal part of the flap and splitting of the
flap is also possible based on this distal perforator. It is rarely
observed that two adjacent perforators that have a common
trunk penetrate the deep fascia and nourish the same area
of the flap.
Generally, the so-called “random pattern flap” elevated
in the trunk and the extremities should follow the rule of
a 1 : 1 ratio of width and length to avoid flap necrosis. If a
flap is to be split vertically at the distal end of a flap in order
to accommodate two separate areas, the same rules that
apply to random pattern flaps should be followed. An
septum. Pattern 3 corresponds to the vessel running in
adipose or other tissue layers (Fig. 26.13).
For example, the intra-adiposal vessel in the groin flap is
classified as pattern 3, and the tensor fascia lata perforator
flap is classified as pattern 1 of an intra-adiposal vessel. In
the anterolateral thigh flap, two kinds of vessel anatomy
are observed among the vessels before they penetrate the
deep fascia: one type perforates the muscle belly and the
other traverses the intermuscular septum. These are closely
related to patterns 1 and 2, respectively.13
THINNING OF THE FLAP WITHOUT USING THE
MICRODISSECTION TECHNIQUE
The clinical significance of the above findings is that the
intra-adiposal distribution of the vessel can be predicted
beforehand. Therefore, if the conventional thinning tech­
nique is applied to the anterolateral thigh flap, removal of
the adipose tissue should be controlled according to the
subfascial anatomy of the vessels.
More specifically, if the perforator is found to be running
deep inside the muscle belly before it penetrates the deep
fascia, adipose tissue with a diameter of approximately 3 cm
around the perforator is sufficient to protect the intra-
adiposal branches of the perforator. If it runs through the
intermuscular septum, branches of the perforator tend to
spread out widely in the deep adipose layer and the adipose
tissue around the branches inevitably is wider than that of
the intramuscular type. In the anterolateral thigh perfora­
tor (ALTP) flap, two types of perforators can be observed:
the intermuscular septum perforator and intramuscular
perforator. Therefore, the size of the adipose tissue taken
around these perforators should be based on the type of
perforator found. In addition to these patterns, the supra­
fascial branch of the perforator is sometimes observed in
the ALTP flap. In this flap pattern, it is advisable to leave
additional adipose tissue around the suprafascial branch for
safe elevation of the thin flap.
In the tensor fascia lata perforator (TFLP) flap and deep
inferior epigastric artery perforator (DIEAP) flap, which
have an intramuscular perforator, adipose tissue with a
diameter of 3 cm is enough to protect the branches of the
perforator.
The safe dimension of the thin flap elevated either by
a conventional method or by microdissection seems to
be nearly identical, and is estimated to be approximately
17–18 cm from various clinical cases of thin flaps.5–7
Figure 26.13  Three anatomic patterns of the intra-adiposal perforators.
Muscle perforator
Fat perforator
Intermuscular septum perforator

e68
PART 3 — RECONSTRUCTION
Consequently, the size of the defect increases gradually
during the operation, and the initially estimated size of the
defect is less than at the time of transferring the flap.
However, at the donor site, swelling of the flap is not obvious
during elevation but increases in severity after transfer to
the recipient site, especially in the postoperative period.
Therefore, it is important to design the outline of the flap
advanced technique of longitudinal splitting based on the
additional neurocutaneous circulation of the flap has been
reported.14
In contrast, with partial horizontal splitting, significant
compromise to the circulation will result as part of the cir­
culation comes from the subdermal plexus. Therefore, hori­
zontal splitting should be avoided unless it involves a very
small area of the flap.
The techniques described above are thought to be rough
and mostly dependent on empiric knowledge. In contrast,
microdissected tailoring of the flap as described in the next
section, represents a maneuver based on direct observation
of the intra-adipose vascular anatomy.
MICRODISSECTED TAILORING OF THE FLAP
Preparation of a thin flap using the microdissection tech­
nique is described above; however, the essence of this tech­
nique lies not only in the preparation of a thin flap but also
clinical utilization of the detailed vascular anatomy observed
by microdissection of the flap. Specifically, microdissection
reveals the distribution and directions of the primary or
secondary branches of the perforator in the deep adipose
layer of the flap. Each branch runs in various directions:
running on the deep fascia, penetrating the adipose layer
and running parallel to the skin, and in most cases, directly
approaching the subdermal plexus.
By observing the direction of the exposed branches, the
nourishing patterns can be predicted beyond what is actually
seen and exposed using the microdissection technique.
These zones can be described as asymmetrical cones that
conceptualize the reliable blood flow in the suprafascial
tissue. Therefore, a flap can be divided into these cones
including each branch in the center, and the spaces among
these cones are considered watersheds areas. These cones
are named “territory cones,” and the notion that the flap is
divided into several cones according to the anatomy of the
intra-adipose vessels is termed the “territory cone concept”
(Fig. 26.14).
Using this concept, the flap can be safely divided, bended,
and folded according to the partition. Additionally, it is also
possible to divide the flap into several 3-dimensional zones,
i.e., the flap can be cut into vascularized fillets such as the
fascia flap, adipose flap, and thin flap. Microdissected tailor­
ing alters the flap under a specific blood flow depending on
the “territory cone concept”.13
In the operative field, the points and directions of each
microdissected branch vessel are correctly projected to the
skin side and marked by the vector arrow. These provide
the most important information for flap circulation consid­
ering the territory cones and every procedure of tailoring
– splitting, folding, cutting, and thinning – should be per­
formed on the basis of these marks (Fig. 26.15).
PRECAUTIONS FOR FREE FLAP SURGERY
All tissues that undergo a surgical procedure become
swollen. At the recipient site, both surface debridement of
the defect and dissection of the recipient vessel will induce
swelling. This effect is exacerbated by anesthesia, under
which the hydrokinetic balance is not maintained naturally.
Figure 26.14  A conceptual diagram of the territory cones.
Colored cones represent territories supplied by each branch arbo­
rized in the adipose tissue.
Figure 26.15  Insertion points and directions of each intra-adiposal
perforator into the subdermal plexus (arrows a, b, c, d) are marked
on the skin (a′, b′, c′, d′). The design and elevation of the flap should
take these points and arrows into consideration.

CHAPTER 26 — Thinning a nd Tailoring
e69
to be approximately 5–10% larger than the size of the defect
at the initial stage of the operation.
Flap tailoring should be performed before separation of
the vascular pedicle, preserving an adequate blood flow to
the flap. Subsequently, the color of the capillary bleeding
from the edge of the flap, not the venous bleeding, is
observed to estimate the distal blood flow of the flap. Light
red bleeding indicates an adequate circulation to the distal
part of the flap, whereas a dark color indicates an area of
flap that is compromised.
During elevation of the flap, meticulous hemostasis is
essential to control postoperative bleeding and the potential
for hematoma formation. Most flaps can be elevated safely
by careful dissection and use of an electric cautery for hemo­
stasis, without tourniquet control. However, if the flap is
elevated under tourniquet control, the author injects one
dose of approximately 0.5 unit/kg heparin to prevent intra­
vascular coagulation of the blood.
After elevating and severing the pedicle of the flap, the
flap should be soaked with heparinized saline on a plate.
The pedicle vessels are not clamped and are laid in a position
that allows for drainage from the flap. This procedure rinses
the inner lumen of the vessels and prevents clotting.
We believe that cooling of the flap below 10°C to prevent
reperfusion injury is not always necessary. The temperature
of the flap soaked in heparinized saline usually declines to
approximately 20°C in the standard environment of the
operation room, which is sufficient to reduce the metabo­
lism of the flap during the usual operation time.
Post-revascularization, warming of the flap with saline at
approximately 40°C is useful for recovering the circulation
of the flap, especially in the toe or partial toe transfer. If
warming of the flap does not effectively lead to recovery of
flap circulation, the patency of the anastomosis, as well as
the entire vascular pedicle should be examined.
It is important to carefully position the artery and the vein
throughout their course to prevent kinking and obstruc­
tion. In particular, if vessels are running across hard tissues
such as bone, tendon, and nerve, and if the artery and vein
intersect each other, care should be taken not to apply pres­
sure on these structures from the flap itself or by the sec­
ondarily formed hematoma. To prevent such an accident, it
is useful to build levees along the vascular pedicle, exploit­
ing the surrounding tissues.
After completing the vascular anastomosis and revascu­
larizing the flap, the course of the flap vessels should
be re-checked to ensure the absence of any kinking or
obstruction.
Postoperative anticoagulation is not routinely applied,
but when necessary, it should be limited in dose in order to
minimize the risk of hematoma formation and eventual
compression of the flap pedicle.
A vascularized iliac bone and microdissected tailored anterolat-
eral thigh flap was transferred simultaneously to reconstruct the
mandible and the oral floor. The soft tissue was required for
coverage of the neck defect and the transferred iliac bone was
used to create a new alveolar ridge (Fig. 26.16). As a conse-
quence of microdissection of the right anterolateral thigh area,
four branches of the perforator were identified and the flap was
designed projecting 3-dimensional relationships of the recipient
defect and vessel to the donor site (Fig. 26.15). The entire flap
was elevated above the deep layer of fat, leaving the greater part
of the adipose tissues at the donor site (Fig. 26.17). After anas-
tomosis of the deep circumflex iliac artery and vein to the
branch of the lateral circumflex femoral system as a flow through
flap, the main artery of the flap was anastomosed to the superior
thyroid artery with a venous anastomosis to the internal jugular
vein. The reconstruction was successful and acceptable shape
and function were restored as seen at 6 months postoperatively
(Fig. 26.18).
CASE 26.1  Reconstruction Using a Free Flap for Defects of the Left Mandibular and
Oral Floor
Figure 26.16  A patient with a neck skin defect and a left man­
dibular body defect after a cancer resection procedure.

e70
PART 3 — RECONSTRUCTION
CASE 26.1  Reconstruction Using a Free Flap for Defects of the Left Mandibular and
Oral Floor—cont’d
Figure 26.17  Planning of the flap and elevated microdissected tailored anterolateral thigh flap.
Figure 26.18  At 6 months postoperatively.

CHAPTER 26 — Thinning a nd Tailoring
e71
REFERENCES
1. Thomas CV. Thin flaps. Plast Reconstr Surg 1980;65(6):747–52.
2. Akizuki T, Harii K, Yamada A. Extremely thinned inferior rectus
abdominis free flap. Plast Reconstr Surg 1993;91(5):936–41.
3. Koshima I, Fukuda H, Yamamoto H, et al. Free anterolateral thigh
flaps for the reconstruction of head and neck defects. Plast Recon­
str Surg 1993;92(3):421–8.
4. Kimura N, Satoh K. Consideration of a thin flap as an entity and
clinical applications of the thin anterolateral thigh flap. Plast
Reconstr Surg 1996;97(5):985–92.
5. Kimura N, Satoh K, Hasumi T, et al. Clinical applications of the
free thin anterolateral thigh flap in 31 consecutive patients. Plast
Reconstr Surg 2001;108(5):1197–208.
6. Kimura N. A microdissected thin tensor fasciae latae perforator
flap. Plast Reconstr Surg 2002;109(1):69–77.
7. Kimura N, Satoh K, Hosaka Y. Microdissected thin perforator flaps:
46 cases. Plast Reconstr Surg 2003;112(7):1875–85.
8. Kimura N, Saitoh M. Free microdissected thin groin flap design
with an extended vascular pedicle. Plast Reconstr Surg 2006;
117(3):986–92.
9. Taylor GI, Palmer JH. The vascular territories (angiosomes) of the
body: experimental study and clinical applications. Br J Plast Surg
1987;40(2):113–41.
10. Taylor GI, Minabe T. The angiosomes of the mammals and other
vertebrates. Plast Reconstr Surg 1992;89(2):181–215.
11. Nakajima H, Imanishi N, Minabe T, et al. Three-dimensional
reconstruction of cutaneous arteries of the back by computer
graphic imaging. Plast Reconstr Surg 1997;100(2):381–9.
12. Nakajima H, Minabe T, Imanishi N. Three-dimensional analysis
and classification of arteries in the skin and subcutaneous adipo­
fascial tissue by computer graphics imaging. Plast Reconstr Surg
1998;102(3):748–60.
13. Kimura N, Saitoh M, Okamura T, et al. Concept and anatomical
basis of microdissected tailoring method for free flap transfer. Plast
Reconstr Surg 2009;123(1):152–62.
14. Karacalar A, Ozcan M. Longitudinally-split radial forearm flap.
Scand J Plast Reconstr Surg Hand Surg 2005;39(2):109–12.
Expert Commentary
Hung-Chi Chen, MD, PhD, FACS
As stated by the author, there are four categories of flap trans-
fers. In this chapter, only the first, second, and third categories
have been addressed. I would like to comment on the fourth
category.
The fourth category is functional reconstruction, including
muscle flap for reanimation, vascularized nerve flap, and enteric
flap for esophagus and voice reconstruction.
The vascularized nerve flap usually has small volume and
does not require thinning. However, when it is used for simul-
taneous restoration of nerve continuity and revascularization of
a major artery of the affected limb, the flap should be designed
properly to prevent kinking of vessels. For example, following
trauma, the defects of the ulnar artery and median nerve can
be reconstructed at the same time with a flap incorporating
sural nerve and lesser saphenous vein as a vascularized nerve
flap. The defects of nerve and artery may be located at different
levels. Under such a circumference, the straight flow of artery
should be maintained to avoid turbulence and thrombosis.
Then the vascularized cable nerve grafts for median nerve
should be designed and tailored under this principle.
Microvascular transfer of free intestinal flap for reconstruction
of esophagus is becoming more and more popular. When a
free jejunal flap is used for reconstruction of cervical esophagus,
redundancy may cause dysphagia and should be avoided. This
is a common mistake for young microsurgeons. The intestine
becomes elongated after reestablishment of blood flow. There-
fore, if inset of free jejunal flap is done before vascular anasto-
moses, it may become too long so that the wound cannot be
closed primarily and the swallowing function is compromised.
Even a case of volvulus due to redundancy of a jejunal flap has
been reported in the literature. Therefore, the method sug-
gested is to do the inset of the upper end first, followed by
vascular anastomoses, and finally trimming of the lower part of
jejunal flap to the proper length before inset of the lower end.
The same principle can also be applied to the free ileocolon
flap, which is used for simultaneous reconstruction of esopha-
gus and voice. A common mistake is redundancy of the colon
segment, which is used for reconstruction of pharynx and
cervical esophagus, and the patient may suffer from dysphagia.
The other pitfall is the narrowing at the junction between the
colon segment and the upper end of thoracic esophagus. To
prevent redundancy, the colon segment should be shortened
to a proper length with removal of the shoulder roll; and a
longitudinal incision should be done at the upper end of tho-
racic esophagus, so that the suture line between these two
structures is longer to prevent narrowing in the future. The
lower part of the colon segment should be tailored with taper-
ing to match the width of thoracic esophagus. For voice recon-
struction the ileum segment should also be designed well to
avoid redundancy. To avoid leakage at the junction between
the colon segment and upper end of thoracic esophagus, an
excess of ileum segment of 3 cm in length can be tailored into
a seromuscular flap to be used as a second layer suture for
augmentation of the anastomotic site.
The functioning muscle flap for reanimation of the face
usually does not require an accompanying skin flap. However,
to avoid bulkiness of the cheek and to achieve adequate func-
tion, the muscle still needs thinning with the nerve distribution
as a guide to tailoring. On the other hand, the functioning
muscle flap for limb reconstruction often carries a skin island
with subcutaneous fat to facilitate gliding of muscle and
tendons. Therefore, the thinning and tailoring technique should
be done according to the description in this chapter.

323
Avoiding Complications
Jeremy L. Wilson  |  Andrew N. Morritt  |  Kenneth L. Fan  | 
Karen Kim Evans  |  Wayne A. Morrison 27 
INTRODUCTION
It is something of a dent to personal pride to be singled out
among so many internationally distinguished contributors
to write a chapter on complications in microsurgery. How
did the editors know we were so experienced in complica­
tions? We would like to be able to make the same boast as
an esteemed surgical mentor who, when asked to speak on
complications, warmed to the subject by stating that, “I am
an authority on complications – I saw one once.” Sadly, good
judgment comes from experience and experience comes
from bad judgment. We at least qualify from the experience
side of the equation.
Free flap failure is a traumatic experience for patient and
surgeon alike and sharply brings into focus a self-appraisal
of the wisdom of the operation in the first instance. Not only
has the reconstruction failed but now a donor site scar is
present – two holes instead of one. Perhaps also the telltale
limb scars of vein grafts, salvage trails of the desperate
surgeon, and hematomas from rashly considered anticoagu­
lation. Incredulity from inquiring relatives, the shattered
confidence and bitter disappointment of the patient. Self-
doubt, recriminations: “What more could I have done to
make this work? Was it bad tissues, the assistant, the anes­
thetist, the nursing, monitoring, or just a good surgeon out
of luck?” “Why am I doing microsurgery? Do I need this?
How did I get into this mess? Who can I share it with? Where
is the caustic soda? Which way to the river?”
The answers to these questions can be analyzed under
three headings:
• Failure of planning
• Failure of execution
• Failure of postoperative care
FAILURE OF PLANNING
For a brief summary of failure of planning, see Box 27.1.
In hindsight, flaps that do well are those that proceed
with comfortable familiarity; no change of plan; no unan­
ticipated need for vein grafts; access for microvascular
repair is easy; recipient vessels are large; and the artery has
good forward flow. The anastomotic site can be closed
without tension. The patient is warm, fluid filled, and
relaxed postoperatively. The flap can be well visualized,
kept warm, and monitored by experienced staff. Con­
versely, where there is indecision, too many cooks giving
advice, and general lack of purpose, results will be propor­
tionately unpredictable.
THE PATIENT
Many complications can be avoided by ensuring that each
patient undergoes a thorough clinical evaluation prior to
formulation of the management plan. Today, successful out­
comes in free tissue transfer and replantation are achieved
consistently in the presence of patient factors once consid­
ered to present a high risk for treatment failure, provided
that these factors are identified at the planning phase and
appropriate measures taken to reduce their effect. In a
historic, multicenter prospective survey of free flaps per­
formed over a 6-month period involving 493 flaps from 23
units, including our own, several factors emerged as impor­
tant to outcome.1 Overall, there was a failure rate of 4.1%
and the only factors adversely influencing survival were irra­
diated field and the use of muscle flaps with skin grafts.
Postoperative thrombosis rate requiring reexploration was
10% and more likely in chronic wound areas and if vein
grafts were used. The paper by Khouri et al., from 1998,
remains as one of the highest level evidence papers that has
been published on microsurgical outcomes.1
Box 27.1  Failure of Planning:
Preoperative
Fitness for Anesthesia
• Age, coagulopathy, general health, prolonged positioning
Poor Choice of Donor Flap
• Inappropriate characteristics: size, color, texture, thickness
• Access: prevents simultaneous resection and recipient site
preparation and donor site dissection
• Pedicle length too short
• Predisposes to postoperative complications, eg, tight chest
closure (LD flap) → respiratory problems, tight abdomen and
pelvic dissection (DIEP) → DVT
Recipient Vessels
• Access: too deep, too far away, needing vein grafts
• Damage: scar, edema, trauma, infection, radiation
Timing of Surgery
• Too early: inadequate planning and/or patient preparation
• Too late: excessive waiting prior to wound coverage
PART 3  —  RECONSTRUCTION
SECTION B  —  FLAP HANDLING AND TECHNICAL TIPS

324
PART 3 — RECONSTRUCTION
be expected to lose their replants.18 Our experience does
not support this low survival rate.
The effect of smoking on free tissue transfer outcomes is
less clear. Although numerous experimental studies have
linked cigarette smoking with delayed microvascular anas­
tomotic healing and free flap failure, we demonstrated sig­
nificant reductions in blood flow without reductions in
microvascular anastomotic patency in animal models follow­
ing nicotine administration.19 Furthermore, a number of
large clinical studies have demonstrated no significant dif­
ference in vessel patency, flap survival, or reoperation rates
in smokers compared with non-smokers. Of interest, perfo­
rator free flaps may differ from traditional free tissue trans­
fers in that reoperation rates, and anastomotic complications
appear to be higher in smokers than non-smokers.20 It is our
practice to recommend that smokers undergoing elective
reconstruction cease smoking at least 2 weeks before the
procedure, and refrain from smoking for at least 2 months
postoperatively. We have several anecdotal cases of both
digits and flaps undergoing irreversible arterial spasm as a
direct consequence of smoking.
DEFECT
SITE AND CAUSE
The site of the defect for reconstruction has not been shown
to significantly predict survival or microvascular complica­
tions in free flap surgery. In particular, free tissue transfer
to defects of the lower extremity does not seem to have a
higher incidence of complications than to other sites.1 Simi­
larly, outcomes do not seem to be predicted by the condi­
tion requiring reconstruction, whether it be trauma, chronic
wound, osteomyelitis, congenital anomaly, burns, cancer, or
other condition.1 With respect to trauma, this is inconsistent
with the reports of Godina21 and Byrd et al.,22 who convinc­
ingly showed that complications were higher and flap sur­
vival less when free flaps were transferred to compound
lower-leg wounds after the first week of injury. Empirically,
if there is poor vascularity of the defect bed, irradiation
damage, infection, or residual tumor, even though the
microvascular component of the procedure may be her­
alded as a success, the likelihood of non-microvascular com­
plications and hence failure to achieve the purpose of the
procedure, is higher.
DIMENSIONS
The surgeon must anticipate likely alterations in defect
dimensions: for example, after wound contractures have
been released, adjacent joints positioned in extremes of
functional range of motion, residual pathology, or antici­
pated die-back of surrounding skin margins are excised.
When resurfacing defects on convex surfaces, such as the
scalp, heel, or limb, or when myocutaneous flaps are
employed and the underlying muscle has to be accommo­
dated, a larger area of skin needs to be included to account
for the increased circumference. Skin size can become criti­
cally deficient once postoperative swelling occurs. Having to
release tight sutures postoperatively from a free flap can
lead to non-graftable bed exposure and an embarrassing,
surviving, islanded monument to incompetent design and
the need for further flap coverage. Especially in lower
extremity cases, planning must ensure coverage of the
FITNESS FOR ANESTHESIA
As with all major surgery, cardiac and respiratory fitness
should be appropriate before electing a microsurgical case.
Prolonged awkward postures need to be tolerated, which
may temper the choice of donor flap and extent of opera­
tion. A personal or family history of thrombosis should be
especially sought in the preoperative screening and consid­
eration given to preoperative anticoagulation in routine
cases.
AGE
Age per se is not a contraindication to microvascular surgery.
A number of studies have shown no difference in free flap
survival when extremes of age are evaluated.1–8 Pediatric
patients have excellent free flap survival rates, with very few
nonsurgical complications and no apparent growth-related
complications noted at donor or recipient sites, although
research indicates exposure to anesthesia before 3 years of
age may result in cognitive deficits.2,9 Prospective studies on
this are still ongoing.10 Although excellent flap survival rates
are observed in patients of advanced age, 75% of patients
older than 70 years have more than one medical issue con­
tributing to complications such as myocardial infarction,
pulmonary embolism, and prolonged intubation.11 In one
series,6 mortality was as high as 5.4% in the age group 60–69
years with significant preoperative medical. In other reports,
although medical complications were very high, mortality
was much lower.1–4
SYSTEMIC DISEASE
Diabetic patients do not seem to experience a higher inci­
dence of free flap thrombosis and/or failure.1 Most of the
robust research on diabetic free flaps has been reported
in the lower extremity with success rates around 92%.12
Numerous reports show the importance of tight glucose
control pre- and postoperatively to minimize surgical mor­
bidities in diabetic patients.13 Data from our institution
show poor long-term glycemic control in patients with
an HbA1c >6.5% leads to increase in wound dehiscence.14
Hypercoagulable patients should be managed in conjunc­
tion with hematology, as these patients have a much
higher rate of thrombosis and a lower rate of free flap
salvage.15 It is important to take a thorough history of
previous deep venous thrombosis (DVT), multiple miscar­
riages, or a family history of blood clotting abnormali­
ties.16 Obesity alone does not appear to be associated with
increased free flap failure rates, although the incidence of
hematoma or hemorrhage appears to be higher.1 A recent
meta-analysis of free autologous breast reconstruction
showed that patients with a BMI >40 had a much higher
prevalence of complications compared with non-obese
patients.17
TOBACCO SMOKING
The influence of smoking on the outcome of microvascular
surgery is a contentious issue. The detrimental effect of
nicotine on cutaneous blood flow and wound healing is
accepted.7 Chang and Buncke reported an extraordinarily
detrimental effect of smoking on replantation surgery out­
comes, stating that 80% of patients who smoke in the period
between 2 months before and 2 months after surgery can

CHAPTER 27 — Avoiding Complications
325
large part to the development of free tissue transfers with
pedicles of greater diameter and length than those of the
first free flaps described, such as the free groin flap. It could
be argued that microvascular free flap surgery would not
have had a long-term future if it were not for the discovery
of the latissimus dorsi and radial artery flaps, which took
anastomotic success from the preserve of the professional
gambler to the common man. However, with the advent of
supermicrosurgery and perforator-to-perforator techniques,
in some hands-free tissue transfer is not dependent on size.28
Pedicle length is important to facilitate proper access and
orientation of clamps so that microvascular anastomosis can
be performed accurately. Extra pedicle length will permit
elevation of the clamped vessels out of cavities and permit
orientation of the clamps to best suit the operator.
DONOR MORBIDITY
All flap surgery “robs Peter to pay Paul” and the art is to
strike the appropriate balance. The risk of donor site com­
plications varies with patient factors and site and composi­
tion of the flap to be raised. Complications are heightened
when flaps are dissected with the patient in an unconven­
tional position: for example, harvest of a latissimus dorsi
flap with the patient supine or an anterolateral thigh flap
with the patient on their side. Here the landmarks for per­
forators and connective tissue septae can markedly change,
and greater caution must be exercised in planning flaps and
anticipating pedicles. Unfamiliarity easily leads to errors.
Individual free flaps are associated with their own specific
complications and are not discussed here.
RECIPIENT VESSELS
Successful free flap surgery is surgery of the recipient vessels.
The majority of factors known to predict outcome in micro­
vascular surgery relate to the recipient vessels.
Access
Recipient vessels may be normal and of good caliber, but
unless they are accessible for microvascular anastomosis, the
stage is set for a cascade of technical errors that account for
most free flap failures. The choice of recipient vessels must
factor in access: not too deeply placed and in proximity to
the defect to avoid vein grafts.
Damaged Vessels
Irradiated, avulsed, edematous, scarred vessels or those in
an inflamed field have higher incidences of anastomotic
thrombosis and free flap failure. Microanastomosis to vessels
within irradiated fields has been shown experimentally to
be associated with poor patency rates and a high incidence
of flap complications.29 Khouri et al. showed clinically that
reconstruction of an irradiated recipient site is a significant
predictor of free flap failure.1 Inflamed, scarred, and avulsed
vessels similarly have poorer patency.21,30 Mitchell et al. dem­
onstrated histologically, in a rabbit avulsion model, that
significant arterial pathology extended up to 4 cm from the
rupture point, despite no detectable abnormality under the
operating microscope.30 By contrast, the venous injury did
not extend significantly beyond the site of rupture. In
avulsed human and monkey arteries, significant circumfer­
ential intimal and medial skip lesions were also observed to
extend many centimeters from the rupture site.
pedicle, which may be farther from the defect. Tension sur­
rounding the pedicle must be minimized to ensure that
there is no external pressure. It may be preferable to inten­
tionally accept a second-stage reduction of redundant tissue
once swelling has resolved. Additionally, the surgeon should
ensure that the chosen flap is of adequate volume to obliter­
ate dead space and not merely be of sufficient area to permit
wound closure. In breast reconstruction, the flap size may
be deliberately smaller than the opposite side, in anticipa­
tion of a secondary reduction on that side.
DONOR FLAP
SITE
Choice of flap donor site is determined primarily by the
defect with respect to size, contour, thickness, color, func­
tion, etc., but is tempered by the significance of the second­
ary defect and positioning of the patient to allow
simultaneous recipient and donor surgery. The donor
choice may influence outcome, in terms of flap failure rates
and anastomotic thrombosis.23 Free groin flaps seem to
have higher failure rates than other cutaneous free flaps
raised on larger-caliber pedicles. Similarly, rectus and
TRAM flaps have been observed to be associated with
reduced incidences of take-back and failure when com­
pared with flaps from other donor sites, although this asso­
ciation does not appear to be significant. The anterolateral
thigh flap has surpassed all others in popularity no doubt
because of its multiple attributes, size, thickness, long, large
diameter vessels, ease of access away from the site of primary
surgery, and usually direct closure.24
COMPOSITION
The composition of free tissue transfers influences outcome
in certain situations. Khouri et al. identified the use of a
skin-grafted muscle flap as a significant predictor of free flap
failure.1 Higher flap failure for these transfers appears to be
secondary to an impaired ability to detect postoperative
thrombosis, rather than a higher vascular thrombosis rate
per se. Muscle and musculocutaneous flaps have traditionally
been favored over fasciocutaneous flaps for coverage of
chronically infected wounds. Traditionally, musculocutane­
ous flaps were shown to be associated with improved oxy­
genation into the contaminated wound space and subsequent
greater reductions in bacterial counts over time, when com­
pared with fasciocutaneous flaps.26 However, recent osteo­
myelitis models demonstrate there are no convincing data
that muscle is superior to fascia following adequate
debridement.25
Machens et al. examined the vascular connections to the
recipient bed 10 years after muscle flap transfer to the lower
limb using Doppler and hydrogen gas washout. They noted
that only in those cases where there had been a salvageable
vascular complication at the time of transfer, in other words
an added angiogenic stimulus, did significant ingrowth
occur into the recipient bed to allow vascular independence
from the pedicle.27
PEDICLE LENGTH AND CALIBER
Free flap failures are higher when flaps with smaller-diameter
pedicles are utilized. Indeed, the increase in free flap success
rates observed over the past 25 years can be attributed in

326
PART 3 — RECONSTRUCTION
spasm sets in; a vicious downhill spiral to ruin. The surgeon
must take control throughout, allocate tasks, oversee simul­
taneous dissection of recipient and donor sites, be con­
stantly aware of time as the enemy, steady but careful and
purposeful. Persistence and determination are necessary
virtues for the microsurgeon. So too is the ability to quit in
the best interests of the patient. Not all flaps will work and
futile added hours spent primarily to salvage the surgeon’s
ego must be resisted. In a recent review analyzing multiple
patient risk factors, including diabetes, smoking, ASA status,
which utilized the American College of Surgeons’ National
Surgical Quality Improvement Program (NSQIP) database
of 639 patients who underwent free flap reconstruction,
prolonged surgical time was the only independent risk
factor for free flap failure.31
OPERATOR EXPERIENCE
A major determinant of success or failure in free tissue
transfer is experience of the operator. Accrued technical
expertise of the surgeon was identified as the most critical
factor related to improved free flap success rates observed
for nine microsurgeons when their early outcomes were
compared with their more recent outcomes.32
Where damaged or irradiated vessels must be used, it is
important to minimize mobilization and dissection of recipi­
ent vessels, employ careful debridement of vessel ends, and
microsuture from inside to outside, to prevent intimal sepa­
ration. This is especially important for calcified vessels. Cal­
cified vessels must be handled with care, as intimal separation
and intimal flaps can occur very easily if the calcified wall is
damaged. We routinely administer topical heparin (10,000
units in 100 mL of normal saline) and a one-time systemic
dose of heparin (5000 units IV) during the clamping and
anastomotic phase to prevent thrombosis in such circum­
stances. In some cases, vein grafts will be needed to anasto­
mose the flap vessels to recipients outside the zone of injury
but vein grafting brings its own risks and inherent
complications.
Number of Venous Anastomoses
Ideally, two venous anastomoses should be considered for
tissue transfer and at least two for digital replants whenever
possible, in an attempt to reduce flap loss, but for flaps this
is usually impractical. This is supported by Khouri, who
observed a significantly higher failure rate when only a
single vein was repaired (4.3%) versus two veins (0%) when
flaps requiring vein grafts were excluded from the study.1
Rarely, however, is it practical to have two recipient veins
close together; similarly, two flap veins may not be
available.
TIMING
Flap complications such as thrombosis and failure, as well
as osteomyelitis and hospitalization times are significantly
increased when flap coverage is delayed in cases of lower
extremity trauma. Godina showed improved outcomes
when early, stable soft tissue coverage of open tibial frac­
tures was achieved with free tissue transfers.21 Similarly, Byrd
et al. showed that complications worsened when open tibial
fracture cover was delayed until the subacute phase of
wound healing and contamination.22 These adverse out­
comes are likely to be a result of pathologic recipient vessels
during this unfavorable time period.
FAILURE OF EXECUTION
For a brief summary of failure of execution, see Box 27.2.
THE SURGEON
Replantation and flap surgery is a military exercise; a team
game with a designated strong leader. A time and motion
study that, if unsupervised, can lapse into Brownian motion,
the erratic random movements of unsupervised assistants,
and dabblers, all competing for a share of the microsurgical
extravaganza. A theater of the absurd may be: “let me do
this bit,” “my turn to do a vessel anastomosis,” “give me a go
at this,” “I’m sick of this operation,” “you don’t expect me
to stand here for 8 h holding a retractor, do you?” Hours
pass, shifts of bored anesthetists and nurses come and go,
and patient ownership and responsibilities become diluted.
The patient loses fluids and freezes from exposure, vascular
Box 27.2  Failure of Execution:
Intraoperative
Surgeon
• Disciplined, methodical
• Experience
Anesthetic
• Warm: theater temperature, warming blankets
• Wet: fluid-filled; CVP; hematocrit
• Happy: anxiolytics
Vessels
• Recipient
• Traumatic dissection
• Inadequate access
• Poor forward flow
• Size mismatch
• Donor
• Too short
• Tension
• Twists and kinks
Anastomotic Technique
• Vascular access too deep
• Orientation of clamps preventing visualization of lumens
• Damaged vessel resection
• Suture damage and catching back wall
Vein Graft
• To be avoided where possible because of increased failure
rate
Thrombosis
• Arterial
• Venous

CHAPTER 27 — Avoiding Complications
327
tous plaques, and sclerosis. One should choose the most
optimal site for the anastomosis, managing the length of
pedicle with the healthiest part of the recipient vessel.
Tearing out of stitches is a sign of vessel damage. When
resecting vessels where the intima and media are continually
separating, it is helpful to “weld” the layers together by
cutting the wall progressively in a circumferential manner,
with one blade of the scissors inside the lumen and the
other outside. In this way, the layers of the wall are “crushed
together” as the vessel end is circumferentially trimmed
(Fig. 27.1). There is a tendency to underestimate the extent
of the vascular injury, particularly in avulsion injuries.
Unnecessary hydrodilation and instrumentation of the
vessel lumen should be avoided. For the end-to-side anasto­
mosis, a simple slit is the least invasive way to perform the
arteriotomy. The surgeon should ensure that vascular
clamps employed are of appropriate size, shape, and closing
pressure. Frequent irrigation with warm heparinized saline
prevents desiccation of exposed vessels and is a further
measure to prevent endothelial loss. It is also important to
release clamps to the artery and vein prior to performing
the anastomosis, in order to test inflow; is it retrograde or
antegrade and ensures that venous outflow is not under
pressure.
PEDICLE TENSION AND TORSION
A tension-free vascular anastomosis is imperative if the
repair is to be successful. Tension should be assessed after
decisions have been made regarding final wound debride­
ment, tumor ablation, skeletal shortening and fixation, joint
positioning, and flap inset. When functional muscle grafts
are inset it is important to anticipate that once one end of
ANESTHETIC COMPLICATIONS
Batchelor (pers. comm.) encapsulated the essence of a good
microsurgical anesthetic with the adage “the patient must be
warm, wet and happy.” Such patients are vasodilated, fluid
filled, and relaxed, all factors designed to avert spasm.
Peripheral vasoconstriction as a consequence of hypother­
mia, hypotension, anemia, or anxiety is highly undesirable
intraoperatively and every effort should be made to avoid it.
Poor forward flow from the recipient vessel leads to confu­
sion of interpretation. Is the vessel in spasm or is it damaged
and unsuitable for use? Following microvascular anastomo­
sis, is failure to perfuse the flap due to inadequate forward
flow from the recipient vessel, anastomotic thrombosis, or
an intrinsic flap perfusion problem? Is this because of
damage to the flap vasculature during dissection, intraflap
spasm, no reflow, hypotension, or just anemia and a pale
patient? Is it sympathetic overactivity from pain and anxiety,
or is it all of the above?
Anesthesia alters the normal thermoregulatory mecha­
nisms and, to prevent hypothermia, the temperature of the
operating theater should be raised, a warming blanket used,
and intravenous fluids warmed prior to administration.
Hypotension as a consequence of general anesthesia, and
intraoperative fluid losses must be prevented by adequate
fluid replacement. Anemia should be treated preemptively
with blood, and debate remains regarding the ideal hema­
tocrit for microvascular flow. It should be appreciated,
however, that it is large vessel spasm that is a concern and
not the microvasculature. Low hematocrit may benefit
microvascular flow, while a higher hematocrit may relieve
or prevent spasm. Inotropes should be condemned due to
unwanted vasoconstriction. It is interesting to note that the
data on these subjects are sparse and contradictory to our
common practices. If pressors are needed during free
flap surgery, experimental data demonstrate possible ben­
eficial effects on flap blood flow with dobutamine and
norepinephrine.33
Prolonged anesthesia may result in cardiorespiratory
problems. Prolonged surgery risks pressure problems and
special attention must be given to prevent neuropraxia. In
extreme cases, compartment syndrome has been reported,
especially in lithotomy position.34
Deep vein thrombosis (DVT) and pulmonary embolus
are the Damocles of prolonged surgery, and preventive anti­
coagulation has to be balanced against the risk of bleeding
and hematoma. We prescribe 5000 units of sodium heparin
subcutaneously with premedication and continue postop­
eratively at DVT prevention dosage.
VESSEL PREPARATION
RECIPIENT
As previously stated, poor forward flow should alert the
surgeon to the possibility of vascular damage or spasm.
Excessive mobilization and rough dissection are major
causes of this damage, especially injury at sites of tethered
side branches. Dissection must be obsessively gentle and
targeted to be just enough to allow adequate access for
clamping and elevation into the anastomotic field. Vessels
should be inspected for obvious tears, holes, bruising,
intimal dissection, intraluminal clot and debris, atheroma­
Figure 27.1  Cut circumferentially to prevent separation of intima
and media.
Crushing the vessel
wall from outside causes
repeat separation (lifting)
of the layers
Crush welding the lifted
medial layers to the vessel
wall by cutting from inside
progressively around
the circumference

328
PART 3 — RECONSTRUCTION
same time, and the operator is blinded to the entry point
of the suture. An experienced assistant can anticipate the
difficulties and compensate by showing the lumen, tempo­
rarily rotating the clamps into the correct axis. More time
needs to be invested in mobilizing the vessels further, rotat­
ing the microscope, changing the chair or limb position,
the operator switching to the opposite side of the micro­
scope, elevating one side of the clamps to lift them high
into a horizontal plane during passing of the suture, elevat­
ing the clamps out of the blood-filled sump with a gauze
swab, etc.
The vascular wall suffers damage during each needle and
suture penetration, and larger needles and oblique place­
ment of sutures result in larger endothelial lacerations.
Multiple needle punctures should be avoided. It is prefer­
able to accept a slightly misjudged siting of a suture than to
attempt adjustments that result in repeated merciless stab­
bings of the intimal wall. The number of sutures used in
the anastomosis should be just sufficient to avoid excessive
bleeding. Unequal intersuture distances result in endothe­
lial gaps, distortion, constriction, and exposed intimal flaps.
Loosely tied sutures may expose subendothelial elements
to the bloodstream and allow excessive anastomotic bleed­
ing and subsequent platelet plug formation. Too many
sutures or sutures tied too tightly can trigger endothelial
slough. Repeated irrigation of the lumens with heparinized
saline has been shown experimentally to enhance vascular
patency.
Although a number of authors have published results
showing no difference in outcomes using techniques other
than interrupted sutures, such as continuous sutures, micro­
vascular anastomotic couplers or staplers, or various cuff
techniques, we believe that simple interrupted suture tech­
nique remains the gold standard for microvascular anasto­
mosis. Godina observed a higher failure rate with end-to-end
anastomosis than end-to-side36 but this has not been our
experience, nor was it seen in Khouri’s survey.1
VASCULAR GRAFTING
Traditionally, it has been thought that vein grafts should be
avoided as much as possible by appropriate preoperative
planning, as they are associated with significantly increased
free flap failure rates.1 They are also thought to be associ­
ated with a significantly higher incidence of hematoma/
hemorrhage. It is debatable whether these complications
the muscle is anchored at its insertion, the other end needs
to be stretched under maximum tension for final fixation.
This results in a major shift in the pedicle (vessels and
nerve) and only after this is done can the anastomotic site
and pedicle length be accurately assessed. Prior to anasto­
mosis, each recipient and donor vessel needs to be meticu­
lously inspected for evidence of torsion, kinking, or
compression. These causes of vascular occlusion are often
dynamic and positional. The positions that the patient and
the part are likely to assume in the early postoperative
period need to be taken into account prior to final decisions
regarding length and positioning of proximal and distal
vessels for repair. The lengths of the flap vessels should
match those of the recipients. If a recipient artery has been
resected to a shorter length than the vein, then the flap’s
vessel lengths should be equally adjusted to avoid redun­
dancy and kinking of the longer vessel (Fig. 27.2).
The flap pedicle must be suitably checked to ensure no
twists or kinks have occurred. A verbal checklist during the
procedure should include “normal forward flow, no kinks
or twists, no tension; can I visualize both lumens during my
anastomosis?”
SIZE MISMATCH
When a size discrepancy exists between the donor and recip­
ient vessels, a number of options exist. A size discrepancy of
<3 : 1 may be handled by gentle dilation of the smaller vessel
and direct end-to-end repair. For size discrepancies of
greater magnitude than this, an end-to-side repair of the
larger vessel into the side of the smaller one gives the most
reliable anastomosis (Fig. 27.3).35
ANASTOMOTIC TECHNIQUE
Most technical failures occur because basic principles are
compromised in the interests of time. Basic technique
should not be learned in the operating theater but in the
laboratory on animals. Experience and wisdom are honed
on patients. The key to a successful anastomosis is to visual­
ize the needle entry and exit point for each suture. If an
anastomosis seems difficult, it means that something is
wrong and you should not persist. The axis of the vessels in
the clamps is not parallel to the operator or not in the hori­
zontal plane. The needle then cannot be parallel in the
axis of the vessels, the lumens cannot be visualized at the
Figure 27.2  Kinking can result from having differing pedicle lengths.
Recipient vessels
Donor pedicle
Kinking
Figure 27.3  End-to-side repair is advisable when there is a size
discrepancy of more than 3 : 1.

CHAPTER 27 — Avoiding Complications
329
MONITORING
Our flaps are monitored clinically by experienced nursing
staff on a half-hourly basis for the first 24 h; hourly for the
next 24 h; and then 4-hourly. The commonly observed
parameters include flap color, tissue turgor, and capillary
return. Commonly, handheld Doppler checks are also
employed. This is aided by a self-adhesive temperature tab
that is applied to the flap with a control on the adjacent
normal skin. A difference of 2°C between the two indicates
vascular compromise. Buried flaps are monitored by
Doppler, but venous obstruction can be present while the
arterial pulse is still present. Exposed skin island windows
of flap can be temporarily used as a monitor of buried skin
flaps but this is not possible for muscle flaps. Jejunal flaps
can be designed with separate vascularized island segments
sited externally through the neck skin. Open muscle flaps
can be difficult to monitor, especially if skin grafts have been
applied. In these cases, the nerve to the muscle can be
retained, and pinching this with forceps will cause the
muscle to contract provided it remains vascularized (P.
Callan, pers. comm.).
It is important to avoid circumferential bandaging of
limbs or tracheostomy tapes around the neck for head and
neck flaps and to have full exposure of the whole flap, so as
to detect hematoma or marginal discoloration. The flap
may be loosely covered between times with warm towels. In
some practices, increased salvage rates have been demon­
strated by placing a continuous venous Doppler probe for
flaps as compared with clinical monitoring alone without
increasing false-positive take-backs.41 This technique is also
useful in buried flaps. These probes are usually wrapped on
the venous segment, as arterial signals will persist for several
hours after venous thrombosis.
VASCULAR COMPROMISE
The great majority of vascular complications in free flap
surgery manifest themselves within the first 12–24 h.
Microvascular failure after this period is rare. Thromboses
most likely begin as an intraoperative event but are not
detected until later. Often, when circulation is suspiciously
are due to the grafts per se or, more likely, are a manifestation
of a deeper problem such as poor planning and the use of
inappropriate recipient vessels. Unplanned vein grafts have
a higher rate of thrombosis than planned vein grafts, which
in general have a similar free flap success rate to flaps
without vein grafts.37 It is important to recognize preopera­
tively when vein grafts are likely, and plan accordingly.
Zeeman et al. showed experimentally that reclamping
venous grafts that had been used to repair avulsed arteries
was safe for up to 30 min without increased risk of
thrombosis.38
INTRAOPERATIVE THROMBOSIS
Intraoperative thrombosis occurs in approximately 8% of
free flap cases, and historically has been reported to be
higher when skin-grafted muscle flaps and/or vein grafts are
utilized. The identification of intraoperative thrombosis is
not associated with a higher incidence of flap loss.1
ISCHEMIA REPERFUSION INJURY AND THE
NO-REFLOW PHENOMENON
Reperfusion injury occurs after the reestablishment of cir­
culation to a flap.39,40 Free radicals accumulate when the flap
is devascularized and cause endothelial cell damage, endo­
thelial swelling, and increased capillary permeability once
the blood flow is reestablished. Skin and subcutaneous
tissues are relatively tolerant to ischemia and will survive a
warm ischemia time of up to ≥6 h, and longer cold ischemia
times. Muscle is much more intolerant of ischemia and
develops irreversible changes after 3 h of warm ischemia.
Bowel is extremely intolerant.
The no-reflow phenomenon is characterized by failure of
tissue perfusion, despite apparent adequate arterial input,
but there is no venous drainage. Most cases of intraoperative
“no-reflow” are a reflection of poor planning: removing the
donor flap before the recipient vessels are fully prepared;
not checking the flap perfusion before removing it to
exclude the possibility of intraflap pedicle damage or spasm;
having to harvest vein grafts because the flap pedicle will
not reach the recipients; and permitting the patient to
become cold and fluid depleted. Once “no-reflow” has
occurred, our experience suggests that IV streptokinase or
urokinase has the best chance of salvaging the flap.
The underlying pathophysiology is believed to be swell­
ing of the vascular endothelium, platelet aggregation,
and leakage of intravascular fluid into the interstitial
space. It may be treated with fibrinolytic drugs or non-
steroidal anti-inflammatory drugs (NSAIDs). In addition, it
is important to consider whether the patient has a hyper­
coagulable condition that may be contributing to throm­
bosis, or very rarely heparin-induced thrombocytopenia
syndrome, which paradoxically causes thrombosis in the
presence of heparin.
FAILURE OF POSTOPERATIVE CARE
For a brief summary of failure of postoperative care, see
Box 27.3.
Box 27.3  Failure of Postoperative Care
• Monitoring
• Inexperienced staff
• Whole flap not visible
• Monitor probes malpositioned
• Dopplering wrong vessels
• Delayed recognition of thrombosis (as above)
• Exclude compression: tight skin closure; hematoma
• Re-exploration in theater
• Flap failure
• Late sequelae and/or complications
• Respiratory
• DVT
• Wound infection
• Cosmesis
• Pain

330
PART 3 — RECONSTRUCTION
We aim for a urine output of >30 mL/h in adults. There
are reports in the literature of severe complications associ­
ated with dextran,47,48 although these are rare in practice.
The patient should therefore be kept warm and pain-free
postoperatively to avoid vasoconstriction.
LATE COMPLICATIONS
Free tissue transfers are often long operations performed
on patients with malignancy, trauma, or congenital deformi­
ties with potential for a wide variety of general complica­
tions (Box 27.4). Painful, tight chest wound closure after
latissimus dorsi flaps will predispose to respiratory complica­
tions. In elderly sick patients, a flap from the anterior thigh
may be more appropriate. Although rare, extremely tight
closure of the anterior thigh may predispose to a compart­
ment syndrome or wound healing compromise.49 Medial
branch of the lateral cutaneous nerve paresthesia are
common. Studying each flap’s donor site morbidity is
important to discuss with each patient prior to surgery.
The prolonged surgery means that DVT and pulmonary
embolism are inevitable risks. Certain donor sites contribute
further to the risk. The free rectus abdominis muscle flap
based on the inferior epigastric artery and the deep circum­
flex iliac artery free flap require their vascular pedicles to
be dissected down to the external iliac vessels. The associa­
tion between pelvic surgery and increased risk of DVT is well
known. Furthermore, with the free TRAM flap, the risk of
DVT is increased when a very tight abdominoplasty is per­
formed as this can be painful and require hip flexion post­
operatively with consequent delayed mobilization. All adults
undergoing free tissue transfer should therefore receive
anti-DVT prophylaxis according to local guidelines. In
our practice, patients receive sequential graded pressure
stockings during the procedure and these are continued
postoperatively while in bed. Postoperative prophylactic
doubtful at the conclusion of the case, hope and denial
intervene. If you think a flap might be dead, it almost cer­
tainly is dead, and has been for a long time. Flap salvage
therefore has no room for procrastination or belief in mir­
acles. If you are looking at your flap too long and ques­
tioning and analyzing if there is problem, you should
highly consider taking it back to assess the pedicle.
The pale flap postoperatively must be interpreted as arte­
rial thrombosis until proven otherwise. The heart sinks.
Arterial insufficiency could be due to hypotension, vaso­
spasm, external compression, or thrombosis. A conclusion
other than thrombosis is probably rationalizing. The tem­
perature of the free flap can be measured with a surface
monitor and, if normal, there is unlikely to be an arterial
problem. Arterial blood supply can also be assessed at the
bedside by pricking the flap with a needle or rubbing a raw
wound edge with gauze. These maneuvers should elicit
bright red bleeding. Doppler monitoring over the course of
the pedicle may detect a pulse. It is advisable to map the
intraflap course of the vessel immediately postoperatively
and record at this site rather close to the pedicle anastomo­
sis to avoid misinterpreting a pulse in the recipient vessel
proximal to a thrombosed anastomosis. Other causes of a
pale flap postoperatively include generalized hypotension,
anemia, and nausea (as part of a vasovagal response). The
skin of some free flaps (eg, groin flap) is normally pale and
this may cause confusion when interpreting the pale flap. A
useful technique to detect subtle capillary return in pale
flaps is to press on the flap with the handle end of artery
forceps held flat against the skin. Visualizing the capillary
refill is easier in shaded light. There are situations when the
surgeon cannot be convinced the circulation is adequate.
In these instances the best approach is to return the patient
to theater.
The congested flap postoperatively implies venous
obstruction until proven otherwise. Causes include venous
thrombosis, a twisted or kinked pedicle or external com­
pression from adjacent tissues or hematoma. Excessive
bleeding from the flap edge and, especially, early hematoma
should suggest that venous drainage is compromised. Blood
is desperately seeking alternative avenues by which to
escape. The Doppler can be deceptive as the arterial pulse
may continue for a considerable period, despite venous
thrombosis. In addition, depending on the location, the
Doppler signal may be erroneously detecting a flowing vein
in the patient and not in the flap itself. Venous thrombosis
is lethal more quickly than arterial obstruction and needs
urgent rescue.42
Every institution and surgeon vary in the postoperative
protocol. Many do not give any rheologic agents except
routine DVT prophylaxis unless there has been a problem
intraoperatively with revascularization or where the recipi­
ent vessels are known to be damaged. It is controversial
whether heparin increases the risk of hematoma in free
flap surgery.43 In a trial of 88 patients undergoing free
flap surgery, we found an increased incidence of hema­
toma in those patients who had low-molecular-weight
heparin administered in the perioperative period.44
However, there are other reports of routinely and safely
administering ECASA and prophylactic dose of subcu­
taneous heparin for all patients undergoing free flap
surgery.45,46
Box 27.4  General Complications of
Prolonged Surgery
• Cardiovascular
• Hypotension
• Fluid overload and pulmonary edema
• Myocardial infarction
• Respiratory
• Atelectasis
• Bronchopneumonia
• Pulmonary edema
• Genitourinary
• Retention of urine
• Urinary tract infection
• Gastrointestinal
• Stress ulceration
• Vascular
• Anemia
• Coagulopathy
• DVT/PE
• Hypothermia
• Skin
• Pressure sore
• Neuropraxia

CHAPTER 27 — Avoiding Complications
331
7. Coskunfirat OK, Chen HC, Spanio S, et al. The safety of microvas­
cular free tissue transfer in the elderly population. Plast Reconstr
Surg 2005;115(3):771–5.
8. Miller CW. Free skin flap transfer by microvascular anastomosis.
Vet Clin North Am Small Anim Pract 1990;20(1):189–99.
9. Ing C, DiMaggio C, Whitehouse A, et al. Long-term differences in
language and cognitive function after childhood exposure to anes­
thesia. Pediatrics 2012;130(3):e476–85.
10. Sun L. Early childhood general anaesthesia exposure and neuro­
cognitive development. Br J Anaesth 2010;105(Suppl. 1):i61–8.
11. Blackwell KE, Azizzadeh B, Ayala C, et al. Octogenarian free flap
reconstruction: complications and cost of therapy. Otolaryngol
Head Neck Surg 2002;126(3):301–6.
12. Oh TS, Lee HS, Hong JP. Diabetic foot reconstruction using free
flaps increases 5-year-survival rate. J Plast Reconstr Aesthet Surg
2013;66(2):243–50.
13. Lazar HL, Chipkin SR, Fitzgerald CA, et al. Tight glycemic control
in diabetic coronary artery bypass graft patients improves periop­
erative outcomes and decreases recurrent ischemic events. Circula­
tion 2004;109(12):1497–502.
14. Endara M, Masden D, Goldstein J, et al. The role of chronic and
perioperative glucose management in high risk surgical closures; a
case for tighter glycemic control. Plast Reconstr Surg 2013;132(4):
996–1004.
15. Wang TY, Serletti JM, Cuker A, et al. Free tissue transfer in the
hypercoagulable patient: a review of 58 flaps. Plast Reconstr Surg
2012;129(2):443–53.
16. Davison SP, Kessler CM, Al-Attar A. Microvascular free flap failure
caused by unrecognized hypercoagulability. Plast Reconstr Surg
2009;124(2):490–5.
17. Schaverien MV, McCulley SJ. Effect of obesity on outcomes of free
autologous breast reconstruction: a meta-analysis. Microsurgery
2014;34(6):484–97.
18. Chang LD, Buncke G, Slezak S, et al. Cigarette smoking, plastic
surgery, and microsurgery. J Reconstr Microsurg 1996;12(7):
467–74.
19. Rao VK, Morrison WA, O’Brien BM. Effect of nicotine on blood
flow and patency of experimental microvascular anastomosis. Ann
Plast Surg 1983;11(3):206–9.
20. Hamdi M, Weiler-Mithoff EM, Webster MH. Deep inferior epigas­
tric perforator flap in breast reconstruction: experience with the
first 50 flaps. Plast Reconstr Surg 1999;103(1):86–95.
21. Godina M. Early microsurgical reconstruction of complex trauma
of the extremities. Plast Reconstr Surg 1986;78(3):285–92.
22. Byrd HS, Spicer TE, Cierney G 3rd. Management of open tibial
fractures. Plast Reconstr Surg 1985;76(5):719–30.
23. Kroll SS, Schusterman MA, Reece GP, et al. Choice of flap and
incidence of free flap success. Plast Reconstr Surg 1996;98(3):
459–63.
24. Wei FC, Jain V, Celik N, et al. Have we found an ideal soft-tissue
flap? An experience with 672 anterolateral thigh flaps. Plast Recon­
str Surg 2002;109(7):2219–30.
25. Salgado CJ, Mardini S, Jamali AA, et al. Muscle versus nonmuscle
flaps in the reconstruction of chronic osteomyelitis defects. Plast
Reconstr Surg 2006;118(6):1401–11.
26. Gosain A, Chang N, Mathes S, et al. A study of the relationship
between blood flow and bacterial inoculation in musculocutaneous
and fasciocutaneous flaps. Plast Reconstr Surg 1990;86(6):
1152–63.
27. Machens HG, Mailander P, Pasel J, et al. Flap perfusion after free
musculocutaneous tissue transfer: the impact of postoperative com­
plications. Plast Reconstr Surg 2000;105(7):2395–9.
28. Hong JP. The use of supermicrosurgery in lower extremity
reconstruction: the next step in evolution. Plast Reconstr Surg
2009;123(1):230–5.
29. Tan E, O’Brien BM, Brennen M. Free flap transfer in rabbits using
irradiated recipient vessels. Br J Plast Surg 1978;31(2):121–3.
30. Mitchell GM, Frykman GK, Morrison WA, et al. The nature and
extent of histopathologic injury in human avulsed arteries and
veins and in experimentally avulsed monkey arteries. Plast Reconstr
Surg 1986;78(6):801–10.
31. Wong AK, Joanna Nguyen T, Peric M, et al. Analysis of risk factors
associated with microvascular free flap failure using a multi-
institutional database. Microsurgery 2015;35(1):6–12.
32. Khouri RK. Free flap surgery. The second decade. Clin Plast Surg
1992;19(4):757–61.
low-molecular-weight heparin is given daily until discharge.
Debate exists if this should be continued for some time
beyond discharge.
Expert Commentary
Edward J. Caterson, MD, PhD,
Stephanie A. Caterson, MD, and
Julian J. Pribaz, MD
The microvascular anastomosis should be the simplest part
(of the free tissue transfer).
Julian Pribaz, MD
The key to avoiding complications with microsurgery is to
recognize and embrace opportunities to do things better.
This takes a measure of introspection and humility to gain
experience from both successes and failures. In short, hubris
will eventually lead to flap loss and inhibit a surgeon’s pro-
gression toward mastery. As eloquently articulated by Dr.
Wayne Morrison, Julian’s mentor, the avoidance of compli-
cations has many facets including flap planning (donor site),
recipient site, and patient factors. These factors can conspire
independently or in concert to cause complications. Detailed
planning and execution are critical; however, it is essential to
trust your instincts. If something doesn’t feel right then do
not leave the operating room. Taking the time to investigate
during the initial operation will only save you later by prevent-
ing immediate or eventual flap catastrophe.
The operation starts in the mind. It is the attention to
detailed planning and command of the current literature that
allows for the correct flap choice. A flap with the appropriate
pedicel length should be planned to match the appropriate
recipient vessels. Often the flap should be sized with insight
into the potential need for additional stages or complications.
It is this up-front mental exercise that decreases the need for
salvage maneuvers such as vein grafts or flap leeching. It
also decreases the stress on your own coronaries. In flap
planning it is imperative to think broadly; having not only a
Plan B but also a Plan C. The expansive mental “lifeboat”
preparation increases one’s repertoire, leads to innovation,
and often allows some fundamentals of the alternate plans
to be incorporated into the original operation. Finally, and
most importantly, free flap success requires a fundamental
regard for each individual patient. When at a crossroads, it
is best to ask “what is the safest plan?” and use this measure
to guide the patient toward recovery. It is hard to argue
against this prudent approach.
REFERENCES
1. Khouri RK, Cooley BC, Kunselman AR, et al. A prospective study
of microvascular free-flap surgery and outcome. Plast Reconstr
Surg 1998;102(3):711–21.
2. Canales F, Lineaweaver WC, Furnas H, et al. Microvascular tissue
transfer in paediatric patients: analysis of 106 cases. Br J Plast Surg
1991;44(6):423–7.
3. Shestak KC, Jones NF. Microsurgical free-tissue transfer in the
elderly patient. Plast Reconstr Surg 1991;88(2):259–63.
4. MacLeod AM, Cleland H. Free flaps in the aged and infirm. Med
J Aust 1994;160(11):679–81.
5. Serletti JM, Higgins JP, Moran S, et al. Factors affecting outcome
in free-tissue transfer in the elderly. Plast Reconstr Surg
2000;106(1):66–70.
6. Ozkan O, Ozgentas HE, Islamoglu K, et al. Experiences with micro­
surgical tissue transfers in elderly patients. Microsurgery 2005;25(5):
390–5.

332
PART 3 — RECONSTRUCTION
33. Eley KA, Young JD, Watt-Smith SR. Epinephrine, norepinephrine,
dobutamine, and dopexamine effects on free flap skin blood flow.
Plast Reconstr Surg 2012;130(3):564–70.
34. Neagle CE, Schaffer JL, Heppenstall RB. Compartment syndrome
complicating prolonged use of the lithotomy position. Surgery
1991;110(3):566–9.
35. Hamilton S. VI International Society of Reconstructive Surgery.
Size mismatch in vessel repair. Melbourne; 1981.
36. Godina M. Preferential use of end-to-side arterial anastomoses in
free flap transfers. Plast Reconstr Surg 1979;64(5):673–82.
37. Cheng HT, Lin FY, Chang SC. Evidence-based analysis of vein graft
interposition in head and neck free flap reconstruction. Plast
Reconstr Surg 2012;129(5):853e–854e.
38. Zeeman BJ, Morrison WA, O’Brien BM. Effect of stasis on the
patency of microvenous grafts in avulsed rabbit femoral vessels. Br
J Plast Surg 1985;38(4):556–60.
39. May JW Jr, Chait LA, O’Brien BM, et al. The no-reflow phenome­
non in experimental free flaps. Plast Reconstr Surg 1978;61(2):
256–67.
40. Kerrigan CL, Stotland MA. Ischemia reperfusion injury: a review.
Microsurgery 1993;14(3):165–75.
41. Kind GM, Buntic RF, Buncke GM, et al. The effect of an implant­
able Doppler probe on the salvage of microvascular tissue trans­
plants. Plast Reconstr Surg 1998;101(5):1268–75.
42. Angel MF, Mellow CG, Knight KR, et al. Secondary ischemia time
in rodents: contrasting complete pedicle interruption with venous
obstruction. Plast Reconstr Surg 1990;85(5):789–95.
43. Kroll SS, Miller MJ, Reece GP, et al. Anticoagulants and hematomas
in free flap surgery. Plast Reconstr Surg 1995;96(3):643–7.
44. Jones SJ, Scott DA, Watson R, et al. Milrinone does not improve
free flap survival in microvascular surgery. Anaesth Intensive Care
2007;35(5):720–5.
45. Lemaine V, McCarthy C, Kaplan K, et al. Venous thromboembo­
lism following microsurgical breast reconstruction: an objective
analysis in 225 consecutive patients using low-molecular-weight
heparin prophylaxis. Plast Reconstr Surg 2011;127(4):1399–406.
46. Keith JN, Chong TW, Davar D, et al. The timing of preopera­
tive prophylactic low-molecular-weight heparin administration in
breast reconstruction. Plast Reconstr Surg 2013;132(2):279–84.
47. Ljungstrom KG, Renck H, Strandberg K, et al. Adverse reactions
to dextran in Sweden 1970–1979. Acta Chir Scand 1983;149(3):
253–62.
48. Kato A, Yonemura K, Matsushima H, et al. Complication of oliguric
acute renal failure in patients treated with low-molecular weight
dextran. Ren Fail 2001;23(5):679–84.
49. Collins J, Ayeni O, Thoma A. A systematic review of anterolateral
thigh flap donor site morbidity. Can J Plast Surg 2012;20(1):
17–23.

333
INTRODUCTION
Achieving success in microvascular free tissue transfers
requires more than the technical ability to proficiently
perform microvascular anastomoses. Preoperative plan­
ning, choice of recipient vessels, intraoperative execution of
dissection, elevation and insetting, along with postoperative
monitoring are all equally important to a successful outcome.
Although having a free flap survive may be a prerequisite
for success, a truly successful outcome is being able to
achieve the preconceived reconstructive goals of closing a
wound or building a part. Most large series report a micro­
vascular success rate of at least 90% and a salvage rate of the
flaps that initially “went down” of about 50%. In most cases,
salvage was predicated on a timely diagnosis of vascular
compromise and prompt intervention.
The decision to surgically re-explore a flap suspected of
vascular compromise may be clouded by significant psycho­
logical stress of the surgeon, patient, and the patient’s
family. The surgeon’s initial response frequently is “denial.”
All too often, precious time is wasted as the surgeon ratio­
nalizes early and sometimes not so subtle signs of vascular
compromise. The personal fortitude in recognizing what
may be perceived as one’s own impending “failure” is para­
mount. Overcoming this mental hurdle is the first step
in proactive resolution of a problem. Recognizing that
“even the masters” have failure rates of up to 10%, with
re-exploration rates of up to 16%, may even venture one to
state that re-exploration of some cases is part and parcel
with the reconstructive effort. Recognizing this, it is impor­
tant not to forget to discuss the possibility of the need to
return to the OR in the postoperative period, with the
patient and their family, preoperatively. We agree with
Schwabegger’s approach of dealing with this, as he states:
“Flap re-exploration should be approached with the attitude
of ‘let’s take a look and see what is going on’ instead of ‘I
have a disaster on my hands that I have to deal with and
everybody around me, including the patient and his family,
is disappointed in me.’ ”1
RECOGNITION THAT THERE IS A PROBLEM
Early recognition of the presence of a problem and rapid
action to address it, are key to flap salvage. While Chapter
29 addresses the topic of postoperative flap monitoring,
we will briefly underscore the importance of the clinical
assessment. A multitude of devices present on the market
definitely have a role in flap monitoring and, in many prac­
tices, have enabled tapering of strict and costly monitoring
routines (i.e., hourly flap checks in the intensive care units)
for flaps with exposed cutaneous paddles. Nevertheless,
basic tenets of clinical assessment and monitoring should
be second nature to all members of the operative and peri­
operative staff. Surgeons should take the lead in ensuring
that the clinical diagnostic art is not lost on the technologi­
cal science of the devices. Suspect changes in flap color,
turgor, temperature, and capillary perfusion in flaps with an
exposed cutaneous paddle, as well as drain output should
alarm the surgical team to a potentially serious problem,
even if ancillary monitoring modalities show no aberrancies.
Similarly, one should recognize that all monitoring devices
have a certain incidence of false-positive results and deci­
sions on intervention should be correlated with clinical
signs.2
All patients who have undergone a microvasculature free
tissue transfer, replant or revascularization should be moni­
tored by nursing staff specifically trained in assessing tissue
viability and flap monitoring. This entails being able to
provide clinical surveillance at appropriate intervals, includ­
ing hourly flap checks for the aforementioned clinical
parameters, as well as hourly Doppler checks for at least the
first 48 h in all complex reconstructions. All patients that
undergo free tissue transfer are not comparable in their
complexity or risk for failure and may require individualized
strategies for the type and duration of monitoring. Although
all our free tissue transfer patients go to the intensive
care unit (ICU) postoperatively, straightforward autologous
breast reconstruction in otherwise healthy patients is rou­
tinely monitored with a near-infrared tissue oximetry moni­
toring device, which can be monitored remotely with
smartphones or tablets and typically leave the ICU 12–24 h
postoperatively, after which clinical monitoring occurs on
the surgical floor every 4 h (many centers continue to
monitor every 1–2 h for the second 24 h period and every
2 h on the third day depending on the type of flap and the
concern for problems). These healthy patients are generally
discharged on postoperative day 3 or 4. Due to a number
of factors, patients who undergo reconstruction for complex
head and neck defects or lower extremity wounds usually
have a longer ICU stay. In comparison with healthy breast
reconstruction patients, head and neck patients as a group,
tend to have more comorbidities; their nutritional status
may be compromised due to their baseline pathology and/
or history of radiation; the presence of recently dissected
large, radiated vessels in fresh surgical field places the
patient at risk for devastating bleeding in the case of vessel
Flap Re-exploration
and Salvage
Jonathan Bank  |  Lawrence J. Gottlieb
28 
PART 3  —  RECONSTRUCTION
SECTION B  —  FLAP HANDLING AND TECHNICAL TIPS

334
PART 3 — RECONSTRUCTION
9. Preparation of blood products that may be needed
10. Expedient transport and preparation of the patient
11. If the surgeon performing the re-exploration is differ­
ent from the surgeon who performed the flap, they
should communicate about the surgical anatomy,
location of anastomoses, suggested sequence and
approach to the anastomosis, and other important
nuances that only the original team would know. In
addition, the original microsurgeon may have intuitive
suspicion about what might have gone wrong during
the initial surgery, and may provide useful advice for the
re-exploration.
12. Operative intervention: exploration, diagnosis, and
solution. The most experienced surgeon should lead
the re-exploration process.
If executed in a timely manner, flap salvage can be
achieved with a high rate of success. To facilitate an expedi­
tious process, one should consider having a standard flap
re-exploration protocol electively agreed on by the surgical
team, anesthesiologists, and operating staff. This would
include the understanding that flap exploration should be
considered a surgical emergency with urgency on par with
life- or limb-threatening emergencies.
EQUIPMENT
All of the standard equipment for free flap surgery must be
available immediately from the start of revision, including
an operating microscope (or at the very least loupes with
sufficient magnification, as a temporary alternative). Ideally,
one should have readily available a sterile surgical pack
with essential instruments saved for these emergencies, as
well as equipment (including mini-Fogarty catheters) and
medications (heparin, lidocaine, thrombolytics) that may
be needed.
MEDICATIONS
There is no consensus among microsurgeons on the peri­
operative use of anticlotting agents. This is particularly true
for elective straightforward microvascular transfers. In con­
tradistinction, when an anastomosis fails and requires mul­
tiple revisions or a salvage procedure, most surgeons will use
some medication to minimize the chances of recurrent
thrombosis.
ASPIRIN
Although there is compelling evidence demonstrating the
beneficial effect of aspirin on decreasing microvascular clot­
ting in animal studies, there is no statistically significant
clinical data in humans. Despite this, many surgeons use
low-dose aspirin to decrease platelet aggregation beginning
pre- or postoperatively and continue it for 2 weeks postop­
eratively when the endothelium has resurfaced the anasto­
mosis. Our protocol includes insertion of a 325 mg aspirin
suppository per rectum at the beginning of the case and
continues 81 mg by mouth daily postoperatively.
HEPARIN
Heparin is a powerful antithrombotic that can be used intra­
venously as a bolus or continuous drip, topically, or via the
or anastomosis disruption; and potential airway issues all
merit more intense ICU monitoring and care. In addition,
although Kroll et al.3 observed that the majority (80%) of
thrombi occurred within the first 2 postoperative days
and that 90% of arterial thromboses occurred within the
first 24 h, Bui et al. found head and neck patients were at
risk for up to 5 days postoperatively.4 Patients undergoing
lower extremity reconstruction are kept at strict bed rest
for the initial 5–7 days after surgery, and thus may benefit
from closer nursing vigilance that may not be available on
a regular surgical ward. This is mainly to permit leg eleva­
tion to facilitate venous drainage, which can be a major
source of problems in the early postoperative phase. In addi­
tion, these patients are more likely to have compromise of
their flap if they had a chronic wound and or required vein
grafts. Monitoring of patients with free flaps to the scalp,
trunk, or genitalia is individualized depending on the com­
plexity of the case and comorbidities of the patient. Buried
flaps (without an exposed cutaneous paddle), muscle flaps
covered with a skin graft, and skin flaps in very dark-skinned
patients are all typically difficult or impossible to monitor
clinically, and are therefore generally monitored using an
implantable Doppler probe.
PREPARING FOR A RE-EXPLORATION
PREOPERATIVE CONSIDERATIONS
The following elements should be incorporated in the
re-exploration process for a compromised flap:
1. Exclusion of problems related to external effects (com­
pression by dressings, positioning, etc.
2. Decision to surgically intervene
3. If anastomotic clotting is suspected, then consideration
should be given to empirically heparinize the patient
systemically at this point to try and stop propagation of
any clot while preparing to go back to the OR.
4. Informing the patient and family of the problem and
the need for surgical intervention
5. Obtaining informed consent, including informing the
patient of the possibilities that the wound may need to
be temporarily left open, other flaps may be required,
and vein grafts, skin grafts, or local tissue harvest may
be needed. They need to understand that new inci­
sions and donor sites may be required and the possible
need for more than one procedure to complete the
reconstruction.
6. Informing the operating room and anesthesia staff that
urgent re-exploration is required. An operating room,
microscope, microsurgical instruments, and specific
pharmacological preparations need to be prepared.
The team should be informed of the possibility that if
salvage is not possible, the flap may need to be replaced
by a new one, or an alternative reconstruction at that
time, which will significantly increase operative time.
7. Optimizing the medical condition and hemodynamic
status of the patient within a reasonable timeframe
8. Obtaining the necessary preoperative work-up, includ­
ing discussion of other treatment options (second free
flap, regional flaps, or other reconstructive options)

CHAPTER 28 — Flap Re-exploration and Salvage
335
some authors routinely use it systemically in the postopera­
tive period.14
THROMBOLYTICS
As a class, all thrombolytics act by converting plasminogen to
plasmin, an enzyme that degrades fibrin. None of the three
main thrombolytics (streptokinase, urokinase, or tissue plas­
minogen activator) commonly available have been shown to
have superiority to their counterparts. At the time of writing
this chapter, according to the FDA registry, urokinase has
been discontinued from availability in the United States. Dif­
ferent authors report equivalent success rates with varying
dosages of the same drugs. It seems that the main determi­
nant for drug selection is surgeon’s preference and institu­
tional availability. Since the re-exploration “event” can be
stressful on the surgical and ancillary team due to the high
acuity, the shortened timeframes, and the need for equip­
ment not routinely used, it is imperative that the lead surgeon
be familiar with what is available and how it should be pre­
pared and used (see Vascular Thrombosis, for details on use).
OPERATIVE SET-UP
Once the decision to take the patient back for exploration
has been made, the entire team should be notified and
made aware that the revision could be anything from a brief
inspection, to a prolonged revisional reconstruction (revis­
ing the initial flap or performing an alternate reconstruc­
tion). In managing personnel, factors such as fatigue and
extra surgical tasks should be accounted for. If available,
additional staff should be called in to facilitate the revision
(to simultaneously dissect an alternate flap, or harvest vein
grafts) or relieve fatigued members of the team.
Considerations in patient positioning should include pre­
venting any further damage to the flap, permit stepwise
exploration that will reveal the compromise culprit, allow
optimal ergonomics for the operator, and concomitantly
permit secondary maneuvers (i.e., a second flap, and vein,
or skin grafts). Notwithstanding the urgency of the case,
standard surgical preparation and safe positioning steps
should not be overlooked.
Preparing the surgical field should be performed care­
fully, particularly in the presence of an open wound from
suture release, dehiscence, or necrosis. Aggressive deter­
gents and alcohol-based preps should be avoided in open
wounds, as they will chemically harm the pedicle or open
wound. If Betadine (povidone-iodine) solution is used, it
may be washed off after it dries, so as not to interfere with
the flap color assessment.
Although not commonly performed, catheter – directed
thrombolysis under angiography has been advocated by a
few authors, following techniques similar to primary coro­
nary angioplasty.15
APPROACH TO A COMPROMISED FLAP
Timing of impending flap failure can be divided into four
time periods:
1. Immediate (intraoperatively)
2. Early (0–12 h)
subcutaneous route. Perioperative subcutaneous heparin is
the only route that has been demonstrated to decrease
thrombosis rates.5 The use of intravenous heparin in micro­
surgery is controversial and has been associated with an
increased hematoma rate. Despite this, some authors rec­
ommend systemically heparinizing with 3000–7000 units
intravenously at the time of anastomosis.6 Postoperative
doses of 100–500 units/h have been advocated by some
authors with or without titrating the dose up to a aPTT 2–2.5
times the normal.7 Our protocol includes heparin prophy­
laxis (5000 units subcutaneously) before induction of anes­
thesia, as routine DVT prophylaxis. Systemic heparinization
is usually limited to situations in which patients are known
or highly suspected of being hypercoagulable or ones that
have demonstrated intraoperative or postoperative throm­
bosis of the vascular anastomosis. If multiple revisions of the
anastomosis were performed (due to the increased risk of
thrombosis secondary to intimal damage), we continue a
heparin drip at a rate to 500 units/h without titrating the
aPTT, for several days after re-exploration. Transient (one
dose) systemic heparinization should be considered when a
major extremity artery is cross-clamped, especially in patients
with peripheral vasculopathy.
For both the initial procedure and the take-back, local
irrigation with heparinized saline has been used with varying
doses, ranging between 10 and 500 units/mL.
Irrigation with a concentration of at least 50 units/mL
helps prevent thrombosis without systemic side effects,
whereas topical solutions of 250 units/mL have been shown
to alter the partial thromboplastin time.8
In patients with known hypercoagulable states, intraop­
erative issues that may predispose to clot formation (such
as suspected vessel injury), or if long vein grafts were used,
an intravenous heparin drip at a rate of 500 units/h is used.
Currently, there is no clinical evidence to suggest this prac­
tice has outcome benefits.9 While animal studies suggest
that routine administration is beneficial for anastomotic
patency, it entails an increase in hematoma rates.10 Despite
this increased risk of hematomas, most surgeons heparinize
patients that have required multiple anastomotic revisions,
unless they are suspicious of the presence of heparin-
induced thrombocytopenia (HIT) or have undergone a
concurrent neurosurgical procedure in which anticoagula­
tion is associated with increased mortality.
DEXTRAN
This polysaccharide has been used in microsurgery due to
the reported function related to inhibition of platelet aggre­
gation, indirect pro-fibrinolytic activity, as well as rheological
function. Some surgeons use the low-molecular-weight prep­
aration as part of the perioperative protocol (dextran 40 at
25–50 mL/h).11 However, most seemed to have swayed from
this practice due to reports of adverse effects in the form of
anaphylactoid reactions and acute respiratory distress syn­
drome. Pretreatment with dextran 1 may avoid this.12
VASODILATORS
The most commonly used and reported smooth-muscle
relaxing vasodilator preparations are lidocaine (2% or 4%),
papaverine 0.25%, and verapamil (see Arterial Spasm,
below). There is limited experimental evidence that topical
administration of chlorpromazine has vasodilatory effects;13

336
PART 3 — RECONSTRUCTION
operative exercises, external reasons for low perfusion need
to be excluded. Some may be as obvious as neglecting to
release microvascular clamps that may be obscured in the
operative field or as mentioned above, a kink or twist of the
vessels either proximal or distal to the anastomosis. Extrem­
ity tourniquets should be completely released or removed.
Patient hemodynamics should ensure a strong perfusing
pressure head. Low perfusion pressure may be due to hypo­
tension, inadequate volume, and hypothermia, and rarely
from the administration of sympathomimetic drugs by the
anesthesiologist. Communication with the anesthesiologist
throughout the case is of paramount importance, as low
perfusion pressure from any of the above causes will not
only jeopardize the flap but will make flap monitoring dif­
ficult. This communication should reveal whether intravas­
cular fluid status is optimized and if any vasoactive
medications were given, which ones, and at what point.
Recognizing that except for flaps with small pedicle diam­
eters (<1 mm), most vasopressors seem to have little effect
on pedicle flow and perfusion pressure tends to be more
important than any vasoconstriction caused by pressors.16
Hypothermia is a frequent cause of decreased perfusion to
the microcirculation, either through spasm of the vascular
pedicle or through decreased perfusion to the peripheral
vascular system. Therefore, the temperature of the patient
should be continuously monitored throughout the case. If
the patient becomes cold, arrangements should be made to
warm the patient (see Intrinsic Flap Compromise, below.)
We advocate judicious administration of warmed crystalloid
fluids during the case, titrated to urine output, heart rate
and blood pressure, with colloid supplementation to limit
swelling. Excess swelling can compromise flap perfusion
and insetting.
REASSESSMENT
After the above steps have been taken, the pedicle is
reinspected. If no cause for the compromise is identified,
evaluation of the anastomosis should be performed. Whereas
a milking test proximal or distal to anastomoses is safe and
will show if there is flow across the anastomosis, the presence
of a small thrombus at the anastomotic site might be suffi­
cient to compromise flap perfusion due to a “stenosing”
effect, even though a milking test still suggests patency. A
milking test across an anastomosis can be dangerous. Having
said that, occasionally, when done, flow may be reestablished.
If flow decreases again, the most likely diagnosis is a clot that
was temporarily dislodged by the milking test. In this situa­
tion, the anastomosis should be redone. A ring resection of
the anastomosis should be performed, excising the throm­
bogenic culprit (e.g., intimal flap, back-wall stitch, or a plate­
let clot in the lumen). Although it is prudent to resect the
minimal length necessary to avoid shortening of the vessels
thus hindering reanastomosis, the new cut end of the vessel
should be pristine. To avoid tension, one should not hesitate
to use arterial or venous grafts to bridge any potential gaps.
To avoid prolonged warm ischemia time, consideration
should be given to cooling the flap, while preparation and
harvesting of vessels grafts is being done.17–19
On completion of the anastomotic revision, the assess­
ment is repeated. If there still is no perfusion or drainage,
anatomic anomalies of the recipient vessels should be con­
sidered. Such anomalies may include unrecognized prior
3. Delayed (12–48 h)
4. Late (beyond 48 h)
Diagnosis of flap compromise in each phase is discussed
here, along with the primary causes of each phase, and the
highlights of the authors’ approach to treating each particu­
lar issue.
INTRAOPERATIVELY: IMMEDIATELY AFTER
COMPLETION OF THE ANASTOMOSES
Technically, this time period would not be categorized as a
“re-exploration.” A brief outline is given here of our intra­
operative maneuvering, that adheres to principles that
would later guide a re-exploration, and serve as a reminder
of how to avoid the need for a take-back.
INITIAL INSPECTION
Following release of the microvascular clamps, flow should
be noted across the anastomoses and the flap is inspected
for perfusion. If the anastomoses appear to be open, the
surgeon should avoid early manipulation of the vessels, par­
ticularly around the area of the anastomosis for 10–20 min.
This may interfere with the vessels accommodating to reper­
fusion, and cause undue early trauma and thrombosis. If
there is any question about the flow, a distal milking test can
be performed to establish if there is flow (see Stripping the
Pedicle, below).
If no vascular inflow to the flap is detected, or if venous
outflow is inadequate, the vascular pedicle and recipient
vessels should be inspected for kinking, compression, or
twisting along its course. If kinking, compression, or twisting
is encountered and cannot be alleviated by manipulation of
the flap or its pedicle, then the anastomosis should be
divided and redone without kinking, compression, or twist­
ing. Flaps with small-caliber vessels (<1 mm) may be prone
to spasm. In this situation, consideration should be given to
irrigating the pedicle with warm fluid and smooth muscle
relaxants (such as lidocaine or papaverine). If there is no
kinking, compression, twisting, or spasm, the anastomosis
should be redone.
Most flaps are inset or at least tacked in place before the
microvascular anastomoses are completed. Periodic check­
ing of flap perfusion during the final insetting is important,
as tissue swelling or inaccurate flap design may preclude
safe wound closure. In this situation, the inset is undone
and the pedicle inspected; if flow improves and no direct
pedicle issues are found, the insetting is revised. Flaps that
do not tolerate insetting without compromise need to be
left partially open. The exposed soft tissue may be closed
with a skin graft, biologic dressing, or negative-pressure
wound therapy device.
MEDICAL OPTIMIZATION
Even the best-done microanastomosis with delicately placed
microvascular clamps will cause some intimal damage
subject to platelet adherence and aggregation. Low flow
in an injured vessel is more prone to clotting. High per­
fusion pressures seem to mitigate these small, localized
hypercoagulable regions around microvascular anastomo­
ses. One of the best ways to minimize clotting is to maintain
high perfusion pressure. Therefore, concomitant to the

CHAPTER 28 — Flap Re-exploration and Salvage
337
seem to occur relatively late in the failure of muscle flaps.
Other forms of monitoring should be employed such as
implantable Doppler probes. Recognizing the problems of
false-positive results of implantable Dopplers,2 one should
also consider incorporating an external skin or tissue
paddles for monitoring purposes. Similar issues exist with
buried adipofascial, bowel, and bone flaps.23
EXTRINSIC FACTORS OF FLAP COMPROMISE
This includes factors associated with the anatomy of the
wound bed, implanted devices, insetting, dressings, and
hematomas. These elements might cause problems that
manifest during the early postoperative period, and, to a
lesser extent, at other time points.
Wound Bed Anatomy
Ideally, this should be a non-issue if properly planned for
before, and executed during surgery. Failure to provide an
adequate geometric environment for flap inset can lead to
problems that manifest in the early postoperative phase.
Bone and cartilage ledges, as well as prominent muscle can
become the foci of kinking, stretch, or torsion compression
of the pedicle. Bone prominences or edges should be
excised or smoothed out; troughs in muscle or bone can be
created to accommodate a more direct lie of the pedicle.
Providing appropriate vessel length should accommodate
for dynamic structures such as tendons and joints.
Implanted Materials
These may cause issues arising early on related to place­
ment, or at the later timeframe on device extraction.
Placement of devices such as drains and implantable flow
Doppler monitors, may seem adequate intraoperatively, but
postoperative tissue changes, such as tissue edema, applica­
tion of external suction, and patient movement may cause
unfavorable alterations in the geometry of the surgical field
and result in kinking or pressure on the pedicle (Fig. 28.1).
These changes should be accounted for, and the task of
surgical alterations (i.e., previous ligations or bypasses),
post-traumatic vascular strictures, prior thrombosis, or con­
genital variances. Whereas arterial adequacy of the recipient
vessel is readily evaluated prior to the anastomosis, venous
drainage insufficiency is more difficult to ascertain. If it is
concluded that the arterial inflow or venous outflow prob­
lems originate proximal to the anastomosis site, the anasto­
moses are reopened, and micro-Fogarty catheters are
introduced into each recipient vessel, inflated and retracted
to remove clots. Intubating the vein in particular should be
done with exquisite delicacy due to the venous wall fragility;
any forceful insertion may preclude future utility of the vein
due to iatrogenic intimal damage. If a recipient vessel
patency cannot be confirmed, alternate vessels should be
selected, and the use of vessel grafts considered. If the recip­
ient vessels are patent, and the vessels of the pedicle are
clotted distal to the anastomoses, the thrombus should be
extracted from the vessels. Grabbing the thrombus with
microforceps and gently pulling it out of the vessel while it
is being gently irrigated with heparinized saline can be suc­
cessful in fresh thrombi. If the entire thrombus cannot be
removed, then considerations of using a Fogarty catheter in
the flap vessels should be considered. If the possibility of
embolization into flap is entertained, thrombolytic agents
should be employed into the flap. Once flow is established,
heparinizing the patient should be considered. If blood flow
through the pedicle is reestablished but the flap tissue is not
perfused, this is called a “no-reflow phenomenon.”20 If flap
perfusion cannot be reestablished, the flap will not survive.
If no technical or mechanical cause of repeated thrombosis
can be established, then the diagnosis of thrombophilia
should be considered (see below).
AFTER LEAVING THE OPERATING ROOM:
WITHIN THE FIRST 12 HOURS
If a problem is to occur, it will most likely manifest within
this first few postoperative hours. Anticipated “normal”
swelling of the flap may obscure the diagnosis of early flap
failure, and color changes in the flap are a relatively late
sign of vascular compromise. The diagnostic challenge
seems to be greater in patients of color, in which the natural
skin pigment further masks subtle color changes that would
signal the presence of a problem in lighter-skin patients.
External monitoring devices such as near-infrared tissue
oximetry may be useful in early detection. Emphasizing
once more that clinical acumen is the basis of problem
recognition and solving, the Achilles’ heel of clinical flap
monitoring exists in the change of shifts. This is of particu­
lar importance when less experienced staff are charged with
the clinical surveillance during the initial postoperative
period, in which the majority of problems arise. Accurate
bedside communication and information transfer are
crucial. To decrease observer error and shorten the time to
address flap issues, digital photography has been utilized in
flap monitoring.21 Some surgeons have further ventured to
normalize the image interpretation with the use of standard­
ized color cards photographed in the frame capturing the
flap to provide a reference for assessment and compensate
for variations in lighting, exposure, white balancing, etc.22
The issue of diagnosis can be more complex in non-skin
flaps. Color, volume, and turgor are all difficult to assess and
Figure 28.1  Plastic sleeve of implantable Doppler slipped toward
the anastomosis causing a kink (white arrow). Black arrow points to
the telltale dark blue thrombus that can be seen through the vein wall.

338
PART 3 — RECONSTRUCTION
small hematomas, patient movement, and positioning may
lead to an imbalance in tissue pressures that may propagate
toward the vascular pedicle, primarily being hazardous
to the lower-pressured venous component. As mentioned
above, prevention is better than treatment. Preoperative
planning of the size and shape of the flap and location of
the recipient vessels and plans for vessel draping and flap
insetting should obviate problems of tight skin closure, com­
pressive tunnels, or exposed pedicles. The thickness of the
designed flap should be taken into account. If a template is
used, adding the third dimension (i.e., thickness of the flap)
may allow recognition of volume discrepancies between the
flap and the space it is to fill. Release of a tunnel to accom­
modate the pedicle may cause difficulty in closure. Designing
the flap with a triangular extension (i.e., teardrop shape for
a circular defect) can obviate insetting difficulties (Fig. 28.2).
Expected flap swelling should further be accounted for in
the 3-dimensional planning. Tissue edema typically peaks at
postoperative day 2.
placing these devices should not be relegated to a junior
member of the surgical team. Correspondingly, planning of
device placement should also allow for safe removal, so as
to not disrupt critical elements of the procedure. Poorly
placed sutures for flap inset or skin closure can unintention­
ally distort the pedicle anatomy.
These factors should be investigated during the explor­
ative surgery, which should be approached in a manner that
enables identification of these interferences, and that helps
to avoid repeating the initial mistake on completion of the
exploration.
Skin Closure and Tunneling
Pedicle compression or flap compromise due to these
reasons will usually be apparent intraoperatively, but it may
only become clinically significant later, with the onset of
postoperative swelling.
Despite seemingly tension-free inset, or noncompressive
tunneling intraoperatively, changes related to tissue edema,
Figure 28.2  (A) Patient with invasive squamous cell carcinoma of scalp. (B) Round surgical defect with opening to avoid tunneling (arrow).
(C) Flap designed with triangular extension (arrow) to facilitate inset without tension on the pedicle. (D) Flap inset with triangular extension
(arrow) used to minimize and tension or pressure on pedicle.
A
B
C
D

CHAPTER 28 — Flap Re-exploration and Salvage
339
transected and transferred vessels are essentially sympathec­
tomized and therefore less prone to systemic vasoregulatory
mechanisms. Experience shows that one cannot rely on that
theory. Preference is given to volume repletion over the
administration of sympathomimetic drugs by the anesthesi­
ologist in the efforts to restore normotension. However, as
mentioned above, high perfusion pressure seems to be
more important than any vasoconstriction caused by the
systemic use of most pressors.
Schramm et al. contend that normovolemic hemodilu­
tion improves oxygenation in the ischemic flap, with relative
anemia at hemoglobin level of 9 g/dL.24 Transfusion to
higher hemoglobin levels bears a hypercoagulable risk
related to hyperviscosity. In our practice, in light of poten­
tial infectious and immunologic risks,25,26 of blood transfu­
sion, we accept hemoglobin levels as low as 7 g/dL in the
absence of other transfusion indications such as hemo­
dynamic instability of the presence of cardiac risk factors.
Hyperviscosity related to excessive blood transfusion cou­
pled with inadequate fluid resuscitation or diuresis, may
provoke arterial or venous thrombosis.
Dressings, Bandages, and External Compression
Tightly conforming dressings may cause venous outflow
obstruction. Elastic wraps should be avoided or placed very
loosely at the end of surgery, with the anticipation that
tissues will swell with postoperative third spacing. Similarly,
care should be taken that splints are fashioned and secured
in a noncompressive manner. This should be specifically
evaluated on the postoperative exam several hours after
the case.
The initial intervention may be done at the bedside by
releasing sutures to allow the flap to accommodate. The
method of how and when to address the resultant open
wound, re-inset the flap, alter the flap, if and when to
explore the flap if it is still in peril, should be taken account
with the patient’s condition in mind. Temporary coverage
can be performed with autologous or allogeneic skin grafts,
or bioengineered skin substitutes. Providing physiologic
coverage of the wound will facilitate resolution of tissue
inflammation and edema (Fig. 28.3). In our experience,
both arteries and veins covered with skin grafts will usually
maintain their patency.
Hypovascularization Due to Systemic Factors
Hypothermia may mimic decreased flap perfusion. More­
over, it may instigate arterial spasm, truly decreasing flow to
the flap, and initiating a thrombotic cascade. If spasm due
to hypothermia is the only cause for decreased inflow,
rewarming of the patient and the flap should resolve the
problem, if performed in a timely fashion. Fluid warmers,
room heating, warming blankets, and radiant heat lamps
can be used. If warmed wet towels or irrigation are used,
they should be changed frequently, as they rapidly lose heat
to the environment and may result in cooling. One should
be cautious not to spend too much time on warming the
flap, as this will increase its metabolic demands and increase
the warm ischemia time should it not relieve the spasm and
anastomotic revision is required.
Systemic hypotension can cause decreased flap perfusion
due to total decreased intravascular volume and a decrease
in perfusion pressure. An argument may be made that
Figure 28.3  (A) Defect requiring pharyngoesophageal reconstruction along with anterior neck closure. (B) Reconstruction with double paddle
thigh-based flap. (C) Difficulty in obtaining arterial pulse led to removing some sutures at bedside. (D) Full release of all sutures performed in
OR. (E) Dressed with negative-pressure wound therapy device. (F) Open wound ultimately closed with split-thickness skin graft.
A
D
B
E
C
F

340
PART 3 — RECONSTRUCTION
congestion because the venous hypertension caused the
venous bleeding, which then decompressed the venous
system. If indeed the bleeding was decompressing an unsus­
pected venous anastomotic occlusion, the flap will invariably
demonstrate venous congestion a few hours after hemostasis
was secured if the venous anastomosis was not revised.
Hematomas warranting take-back may not always be asso­
ciated with flap compromise. Expanding hematomas in the
neck or mediastinum may cause airway obstruction or even
exsanguination if large vessels are involved. Similarly, large
volumes of blood loss in the lower extremity may go unrec­
ognized and ultimately require evacuation. However, impli­
cations on flap survival, aside from direct compression of
the pedicle, are related to hypotension, resuscitative efforts,
fluid and blood product transfusion, vasoactive medica­
tions, etc. In virtually all situations, flap survival should be
secondary to life- or limb-saving efforts.
INTRINSIC FACTORS OF FLAP COMPROMISE
Arterial Spasm
There is an old adage in microsurgery: How do you spell
spasm in microsurgery? Answer: C-L-O-T. Be that as it
may, spasm occasionally really happens, especially in small
vessels.
Once other causes of inflow decrease are excluded and
typical causes of vasospasm are identified and corrected,
arterial spasm may still persist. The pedicle should be
inspected for excessive tension that may cause vasospasm.
Prior to performing surgical manipulation of the vessels,
irrigating the pedicle with warm heparinized saline and
vasodilatory medications may release the spasm. The most
commonly used and reported smooth-muscle relaxing vaso­
dilator preparations are lidocaine (2% or 4%), papaverine
0.25%, and verapamil28 irrigation performed on the exter­
nal surface of the vessels, along their exposed course. Lido­
caine is a voltage-gated sodium channel antagonist. There
is evidence that at lower doses lidocaine (i.e., ≤2%) may
have a vasoconstrictive effect – some authors caution its use
for this indication.29 At the higher concentrations the vaso­
dilatory effects are more pronounced. Conrad and Adams
describe using a 20% solution of lidocaine applied with
cotton pledgets.8 Verapamil, a calcium channel blocker, and
papaverine, a phosphodiesterase inhibitor, have both been
shown to increase blood flow in animal models.29,30 Although
all three medications have an acidic pH, there is no evi­
dence revealing detrimental effect on vessels intima. None­
theless, most surgeons seem to resort to the cautious practice
of extraluminal application alone.
Another mechanical way to relieve spasm is by adventitial
stripping. The mechanism may have to do with a local sym­
pathectomy effect, though this may apply only to the donor
vessels. The effect on the pedicle vessels may just be due to
bulk reduction and removal of constricting bands that
become critical in the new position of the filled vessels on
inset.
Failure of the above to relieve spasm necessitates revision
of the anastomosis. If the vessels still seem spastic, they may
be cautiously dilated with dilating forceps or needle holders.
This is performed by introduction of the instrument into
the lumen, while ensuring the tips do not disrupt the intima.
The dilation is performed with the sides of the instrument’s
When recipient vessels are in the neck or preauricular
area (superficial temporal vessels), care should be taken in
securing tracheostomy or endotracheal tubes. These tubes
should be sewn in place rather than secured with circum­
ferential ties or devices that cross over underlying pedicles.
Postoperative patient positioning should account for areas
in which pressure may be detrimental. Markings should be
made on the patient, cautioning avoidance of pressure on
key locations (Fig. 28.4).
The first step in addressing a compromised flap would be
to simply alleviate any and all of these potentially compres­
sive factors.
Hematomas
Compressive hematomas may arise at any postoperative
point. Assuming adequate hemostasis was achieved during
surgery, resolution of intraoperative hypotension, or tran­
sient hypertension on emergence from anesthesia may
bring on early postoperative hematomas. Consequently, one
should request that emergence from anesthesia should
be smooth without “bucking” or transient hypertension.
Mechanical disruption of clotted or ligated vessels during
patient transfer may be another reason for hematomas.
Hematomas should be evacuated, preferably in the OR.
Separate from any compressive effect, hematomas are toxic
to tissue most likely due to free radical production.27 It is
not unusual to note a hematoma under a segmental or
partial flap loss. In addition, acute hematoma could repre­
sent vessel disruption or venous anastomotic occlusion.
Hematomas in the presence of a decreasing or changing
arterial signal should lead one to suspect a venous anasto­
motic occlusion. Patency of the venous anastomosis should
be directly evaluated at the time of hematoma evacuation.
The flap may not show any clinical signs of venous
Figure 28.4  Marking warning to avoid any compression over the
pedicle (arrow).

CHAPTER 28 — Flap Re-exploration and Salvage
341
re-performing the anastomosis. If a coupler was used for the
venous anastomosis that thrombosed, one should reassess if
an appropriately sized coupler was selected. Consideration
should be made to performing a hand-sewn anastomosis if
there is a significant size mismatch and or a thick vessel wall
relative to the coupler diameter.
Thrombi that form in the area of the anastomosis or in
vessels that are relatively larger in caliber, may be amenable
to mechanical extraction with Fogarty catheters, in combina­
tion with thrombolytic agents. If gross thrombus is removed
and flap perfusion is still lacking, distant microthrombi
within the flap that cannot be removed mechanically may be
present, and thrombolytic therapy should be considered
(see below). As mentioned above, a distinct phenomenon
that is not primarily thrombotic in etiology, is the “no reflow
phenomenon” caused by intimal edema, generally resulting
from prolonged warm ischemia.33 This, in turn, may lead to
microthrombosis that propagates proximally.
Three agents are available to dissolve thrombi: urokinase,
streptokinase, and recombinant tissue plasminogen activa­
tor (rTPA). After correcting potential causes of thrombosis,
the agents may be instituted to aid in resolving inflow or
outflow obstruction. These medications are expensive (rTPA
in particular) and have the inherent side effect of bleeding.
There are no consensus guidelines for use of these medica­
tions in flap salvage, or validated dosages. There are no
evidence-based studies defining superiority of one agent
over the other in the context of microsurgery. Numerous
case reports and experimental studies report the use and
efficacy of the different agents and dosages, and derivations
are made from the cardiac and vascular literature regarding
restoration of flow. Therefore, usage should rely on the
dosages recommended for units per kg body weight, as
individually described in every medical/drug information
sheet.
Streptokinase.  Derived from Streptococci, this is the least
expensive thrombolytic. Doses of 50,000–250,0007 to 750,000
units of streptokinase,8 diluted in normal saline at a concen­
tration of 5000 units/mL can be used to infuse the throm­
bosed vessel.
Urokinase.  Lending from the vascular literature on lower
extremity salvage, doses of 100,000,34,35 to 250,000 units of
urokinase36 in 50 mL of sterile water, used in repeated
administrations of smaller aliquots until clot dissolution.
Recombinant Tissue Plasminogen Activator.  This is typi­
cally the most costly preparation. Once more, different
authors use different concentrations: 2 mg37 in 50 mL; 5–20
(diluted as 1 mg/mL);7,35 to 2 mg in 2 mL.38
Small aliquots of appropriately concentrated solutions of
the thrombolytics should be prepared and infused into the
clotted lumen (or into vessel proximal to clot). The throm­
bolytics may be infused through the cut end of the resected
anastomosis, via a side branch, or through an arteriotomy
proximal to the thrombosis. To minimize intimal damage,
we usually use an epidural catheter attached to a three-way
stopcock to infuse the thrombolytic solution. Choosing the
portal entry depends on what other maneuvers need to be
performed such as use of Fogarty catheters, revision of flap
inset, etc. The thrombolytics should be allowed to take
tines, so as to not generate excessive point pressure on the
vessel’s wall. Alternatively, if a small 1–2 mm slit is made in
the vessel opening, it will open up, creating a flange. If there
is suspected spasm distant to the arteriotomy, mini-Fogarty
catheters can be inserted, gently inflated, and withdrawn to
dilate the vessel, although this carries the risk of intimal
damage. Decrease of vessel diameter as it enters the flap
warrants particular caution in attempting to dilate the
pedicle. Simultaneous irrigation should be performed with
introduction of the catheter.31
Inadequate analgesia and anxiolysis may also have an
effect on vascular tone through generalized sympathetic
mechanisms. While it is difficult to objectively assess, clini­
cally we find that addressing these issues reverses clinical
vessel spasm at the bedside. This seems to be more apparent
in flaps in the pediatric population, thought to have higher
sympathetic tonicity versus adults.32 Vasospasm in children
may also be more critical, given the size of the vessels,
increasing the propensity for clotting with decreased flow.
Kinking and Twisting
Factors intrinsic to the vascular pedicle that may cause
mechanical obstruction to inflow (and more commonly and
rapidly to outflow) are primarily attributed to incorrect posi­
tioning, while setting up for the anastomosis. Ensuring that
the pedicle is not twisted on itself is mandatory. Long ped­
icles and tunneled pedicles, vein grafts, and venous loops
are at particular risk for this to occur (Fig. 28.5). With
reestablishment of flow and filling of the vessel, an incom­
plete twist or kink can be exacerbated and lead to a com­
plete obstruction. Marking the correct orientation of the
vessels, noting the location of side branches and the arrange­
ment of the vessels within the pedicle prior to transfer, can
prevent such errors.
If one of these factors is found to have been affecting flow
on re-exploration, de-rotating the flap may solve the
problem. Frequently, it may be simpler to takedown and
redo the anastomosis (especially in cases of complex flap
inset). If the vessels appear to kink due to length redun­
dancy, they should be shortened or the flap inset adjusted,
in order to allow the pedicle to assume a gentle curve. One
should be cautious in attempting to fixate the pedicle by
suturing it to the surrounding tissue, as this may in and of
itself beget a kink. Without sutures, a natural repositioning
of the vessels will take place and the chances of kinking are
lessened. Pedicles in the neck or across joints should be
evaluated while the neck or joint is moved through its
normal range of motion before closure.
If the exploration was delayed and the vessels are throm­
bosed, subsequent steps in the salvage algorithm should be
followed.
Vascular Thrombosis
Prior to attempting a reanastomosis, any potential cause of
thrombosis should be sought and corrected. The vascular
lumen should be inspected for the presence of intimal flaps
that might cause an area of turbulent flow begetting a
clot. Suboptimal coaptation of the recipient and donor
vessels resulting in buckling of the vessels at the site of the
anastomosis can have a similar effect to an intimal flap. In
some cases, loose suture or other micro-debris can have a
similar effect. Any of the causes warrant resection and

342
PART 3 — RECONSTRUCTION
most of the enzyme is consumed by the clot.36 Unfortu­
nately, as mentioned above, urokinase is no longer available
in the United States. If the arterial anastomosis was left
intact, leaving the vein open to drain can affirm
reestablishment of flow thorough the flap, and permit resid­
ual clot and thrombolytics to be expressed and not return
to the systemic circuit. Most surgeons irrigate the vessel with
heparin solution after infusing the thrombolytic and then
systemically heparinize.
Leeches are of little if any benefit in frank venous throm­
bosis. Their main indications are for small fasciocutaneous
flaps or replanted parts with venous insufficiency in which
effect over several minutes (some authors recommend irri­
gation up to 30 min), with the vein clamped to allow the
medication to “bathe” the thrombus, followed by thorough
irrigation with saline prior to subsequent infusion attempts
with the same or an alternative preparation. To avoid sys­
temic introduction of the agents, the venous anastomosis
should be taken down, even if not thrombosed. The open
vein should be left to drain onto gauze to collect the efflu­
ent and prevent inadvertent absorption from the wound
bed. Serletti et al. believes that “venting” the vein is unnec­
essary practice if urokinase is the chosen thrombolytic,
given its short half-life (14 min), and the assumption that
Figure 28.5  (A) Venous loop ( black arrow) prior to being passed under a subcutaneous tunnel. (B) Acute arterial ischemia of flap noted 12 h
postoperatively. (C) After tunnel opened up, vessels noted to be twisted with acute kink in artery (white arrow). Note thrombus visible on either
side of kink (black arrows). No flow in artery or vein. (D) Flow reestablished after dividing artery, untwisting of vessels, thrombectomy, throm­
bolysis with rTPA, and revision of arterial and venous anastomoses.
A
C
B
D

CHAPTER 28 — Flap Re-exploration and Salvage
343
count; PT, PTT; prothrombin G20210; antithrombin III
activity; protein C activity; protein S activity; antiphospho­
lipid antibody testing; and plasma homocysteine levels. If
therapeutic anticoagulation has already been instituted,
some of the work-up may need to be deferred to a later time.
Patients with a personal or family history of thromboem­
bolic events should undergo a preoperative work-up. This
will not necessarily preclude microvascular reconstruction,
but it may help with perioperative management37 (even to
the extent of operating under anticoagulation in rare
instances, balancing the risks and benefits).
Given that the surveillance frequency is decreased by
this time, flap compromise may go undetected for longer
periods, reducing the likelihood of salvage. Skin, fat, fascial,
and bone flaps are more likely to survive a prolonged
ischemic insult, and re-exploration with intent to salvage
should be done recognizing that subsequent fat necrosis or
bone resorption (from ischemic loss of osteocytes) will likely
ensue. Muscle and intestinal flaps are less resilient and the
chances of failure are much higher: the surgeon should
plan for alternative reconstruction.
In practice, the likelihood of flap salvage with a delayed
presentation is low.3 The inflammatory phase is at its
peak, making exploratory attempts and tissue manipulation
unforgiving, and potentially hazardous.
SPECIFIC STEPS IN RE-EXPLORATION
If not already done at the bedside, sutures are released first
over the location of the pedicle. Using loupe magnification,
the edges of the flap are gradually separated from the
surrounding tissue, taking care not to damage underlying
structures such as anastomosed vessels or coapted nerves,
which might be concealed by adherent blood clots.
First, an attempt should be made to locate the artery and
vein and inspect the anastomosis. Attention should be paid
to the presence of clot along the course of the recipient and
donor vessels, as this may hint to the origin of the problem
(such as a bleeding side branch). The lay of the vessels
should be noted for kinks and twists, and these reversed as
previously delineated. Visual inspection of the color, diam­
eter, and vein may suggest or exclude a clot. Visible or pal­
pable pulsations and Doppler signals along the artery are
sought.
STRIPPING THE PEDICLE
To confirm flow, one might perform a strip test (or milking
test). The strip test can be done on the artery or the vein.
One smooth microforceps gently occludes the vessel just
beyond the anastomosis (on the flap side for artery, and on
the non-flap side for vein). A second microforceps gently
occludes the vessel immediately downstream to the first
forceps, and is gently shifted away from the anastomosis for
0.5–1 cm, while continuing to occlude the vessel, thus emp­
tying or “milking” out the luminal content from the inter­
vening segment. The first forceps (the one closest to the
anastomosis) is then briskly released while the second
forceps maintains pressure (Fig. 28.6). Attention is paid to
the diameter of the released segment that should increase
with the refilling of blood, if this is to occur. The velocity
of the refilling provides information regarding the inflow
pressure.
venous outflow may be reestablished of the subdermal
venous connection of the flap to wound edges or bed, and
for marginal venosomes in an otherwise well-perfused, well-
drained flap. Leech decongestion of the flap without revers­
ing the causal factor may lead to false security that the flap
is salvaged – exploration should not be delayed more than
necessary.
AT 12–48 HOURS AFTER COMPLETION OF THE
RECONSTRUCTIVE PROCEDURE
This timeframe is characterized by increasing mobility
of the patient, with increased mobilization of fluids and
peaking of postoperative edema. The aforementioned
factors, particularly the mechanical disruptors, probably
play the major role in flap compromise during this period.
The most important preventive intervention is cautious
clinical management by the nursing and ancillary staff.
If delayed diagnosis occurs with partial or complete flap
necrosis, surgical debridement should take place to avoid
inflammation and infection that may hinder subsequent
reconstructive efforts. Exposed critical vessels or structures
(brain, prosthetics, etc.), presence of infection, potential
need for hypercoagulable work-up, etc. will affect the deci­
sion whether or not to perform a second flap in the immedi­
ate setting, or to provide an alternative method of temporary
or definitive wound closure. We are of the opinion that if
there was an identified reason for flap failure, a subsequent
flap can be performed without a hypercoagulable investiga­
tion. If not, a work-up is warranted prior to attempting
another microvascular transfer.
BEYOND 48 HOURS
Isolated arterial perfusion problems seldom arise after 48 h.
Most issues would be related to venous outflow obstruction,
either from tight dressings, patient movement, or vessel
stretch or compression due to flap and patient positioning
(i.e., dependent positioning in lower extremities, compres­
sion of flaps to the head and neck, gravity effects on large
breast reconstruction, etc.). A large seroma or late hema­
toma may cause compression as well. Infection at the flap
site will compromise microperfusion and instigate a clot in
the pedicle. High-risk situations are in the acute trauma
reconstructions, in which debridement may not have been
complete, and salivary leak in head and neck reconstruc­
tion. Ultrasound evaluation may help diagnose the etiology,
but surgical exploration should not be deferred, unless
edema seems to clearly be the issue, in which case partial
release of sutures may resolve the problem by reducing
tension and allowing the flap to expand. Infected flaps are
best reconstructed in a delayed fashion once the wound bed
is deemed clean.
Thrombosis due to a hypercoagulable state may present
during this interval. In a series that included 58 flaps in
known hypercoagulable patients, the overall thrombosis
rate was 20.7% (12 flaps), of which four occurred intraop­
eratively, and eight postoperatively. Seven of the eight
delayed thromboses took place beyond 48 h.39 Once all
other thrombogenic factors are excluded, hypercoagulable
work-up should be performed in consultation with a hema­
tologist.40 A typical work-up should include complete blood

344
PART 3 — RECONSTRUCTION
more hazardous in edematous flaps, due to the amount of
force required to remove the clot.
USE OF FOGARTY CATHETERS
The use of a Fogarty catheter for thrombus extraction is
safer for recipient vessels than flap vessels. As the donor
(flap) vessels narrow during their course into the flap, the
risk of tearing the vessels increases, especially if an inap­
propriate catheter size is selected. Size 2 and 3 French cath­
eters may be introduced through a side branch, a slit in the
vessel, or via an opened anastomosis. After introduction of
the catheter, it is passed through and beyond the clot, the
balloon is carefully and gently inflated, and the catheter
retracted, extracting the clot from the vessel lumen. The
catheter should not be pulled against any resistance. If
it does meet resistance, the balloon should be partially
deflated so it can be pulled without resistance. The proce­
dure should be repeated until no more clot is retrieved.
Instillation of thrombolytics before and after Fogarty cath­
eter thrombectomy may facilitate catheter insertion and clot
extraction. The vessels should then be flushed with heparin
solution to wash out any residual thrombolytic and mini­
mize any residual clot propagation prior to reanastomosing
the vein.
USE OF VASCULAR GRAFTS
Vein or artery grafts may bridge vessel gaps created by revi­
sions and partial resections due to vessel damage, interfer­
ing venous valves, old mural thrombi, or intractable spasm.
Other indications include insufficient vessel length and size
mismatches that an interposition graft helps overcome.
WHEN ONE SHOULD USE NEW RECIPIENT VESSELS
In the event of poor arterial inflow, or obstructed venous
outflow, the surgeon may need to select alternative recipient
vessels. Unrecognized prior trauma to the recipient vessels,
radiation effects, iatrogenic injury to the vessels, or shorten­
ing from revisions of the anastomoses can all be causes. If
alternate recipient vessels are needed, long vessel grafts
may be required and anticoagulation should be considered.
In some scenarios where an alternate venous outflow is
required, venous turndowns (i.e., external jugular, cephalic,
or saphenous vein) can be used as a bridge without the need
If there is a suspected problem at the anastomosis, it
should be opened, resected, and redone.
OPENING THE ANASTOMOSIS
If there is weak or no flow on the milking test, the anasto­
mosis must be opened and inspected to exclude a narrowing
from a technical error or the presence of any obstructing
material. If only a portion of the anastomosis was opened
(i.e., several sutures removed), cause of the obstruction may
be revealed. However, it is unlikely that simply removing a
thrombus at the anastomosis will correct the problem if the
underlying cause is not addressed (such as turbulent flow
from an inadequately performed anastomosis, an intimal
flap, or debris). Occasionally (typically only during the initial
procedure), simple replacement of a poorly placed suture
may be done at this juncture and will correct the problem.
If not, the anastomosis should be revised.
VESSEL DILATION
On rare occasions vessel spasm may cause frank thrombosis.
If application of vasodilatory medications and adventitial
stripping fail to relieve the vasospasm, the anastomosis
should be opened or resected; gentle dilation with a vessel
dilator should be performed. This should be performed
under optimal visualization in order not to damage the
vessels. This maneuver is particularly beneficial when treat­
ing arteries, but thick-walled veins (mainly lower extremity)
can be manipulated similarly. Thinner-walled veins are
more susceptible to injury from instrumentation. As men­
tioned previously in the chapter, the dilation is performed
with the sides, and not the tips of the instrument tines, in
order not to puncture or tear the intima. In addition, slit­
ting the open end of the vessel may alleviate the spasm and
provide a larger circumference for anastomosis.
COMPRESSING THE FLAP
Squeezing, compressing, or milking a flap in an attempt to
remove a thrombus is potentially dangerous and rarely
helpful (if ever), but the author has included this for com­
pleteness. If a venous clot is visible within the pedicle but
cannot be extracted with instruments, gentle pressure on
the flap may help to evacuate the thrombus, with or without
thrombolytics. This may be less effective and potentially
Figure 28.6  Milking test. (A) Gently grasping vessel distal to anastomosis with two microforceps. (B) Slide most distal forceps away from
anastomosis, milking the blood out of the vessel. (C) Releasing the forceps closest to anastomosis and observing if the emptied vessel fills
from blood crossing the anastomosis.
A
B
C

CHAPTER 28 — Flap Re-exploration and Salvage
345
a back-up plan and “lifeboats” to cover critical structures
or providing physiologic closure of the wound as needed
(Fig. 28.7).
CONCLUSION
In short: “When in doubt – there is no doubt: Explore!” One
of the most difficult steps in flap re-exploration is making
the decision to take the patient back to the operating room.
If there is ambiguity, consulting a colleague may facilitate
making a decision, bearing in mind that the goal is a com­
pleted reconstruction. Once the emotional task is overcome,
and while there will always be unknowns, the ensuing steps
are largely logistical and technical, and a fairly straightfor­
ward thought process and approach can be followed.
The steps and considerations outlined in this chapter are
to serve as general guidelines, and should be secondary to
following plastic surgery principles and surgical common
sense. Because success rates are generally high, there is a
lack of good data comparing the specific components of
flap re-exploration. While the absence of data and formulas
remains a deficiency in microsurgery, it leaves room for
research, and provides an interesting challenge to further
better our field.
to perform two anastomoses as in a classic vein graft. Vein
loops can be used to extend both the artery and vein with
the advantage of observing the flow through the loop for a
period of time prior to dividing it.
WHEN TO ABANDON THE SALVAGING
OF A FLAP
Hemodynamic instability, severe coagulopathy, hypother­
mia, or the need to perform extensive resuscitative maneu­
vers before or during efforts to salvage a flap should prompt
consideration of aborting the reconstruction at that point
in time. How “extensive” is extensive should be determined
on a per-case basis, weighing the risks and benefits, the
alternatives, etc. As mentioned previously, different flaps
have differing ischemic resilience: functional muscle flaps
cease to be optimally viable (or functional) beyond 2 h of
warm ischemia time, intestinal flaps have similar tolerance,
etc. In other words, the surgical goal again enters the
decision-making – reconstructing a functional part versus
closing a wound. Unsalvageable portions of chimeric or
composite tissue flaps may be discarded in a way that may
achieve the reconstructive goal, albeit not as initially
planned. This point reiterates the importance of having
Figure 28.7  (A) At 55 h after an immediate left breast reconstruction with a deep inferior epigastric artery perforator flap, with loss of arterial
signal. Note that although the mastectomy skin looks compromised, the flap skin paddle shows little sign of compromise. (B) On inspection
of the pedicle, the artery appeared to be open partway up the pedicle; the vein was clearly thrombosed (arrow). The flap itself was ecchymotic
with thrombosed veins. (C) Clot retrieved with a Fogarty catheter. Despite thrombolytics, flow could not be reestablished in flap. (D) Flap and
compromised native breast skin debrided. The wound was left open and treated with negative-pressure wound therapy until closure with a
pedicled latissimus myocutaneous flap, 1 week later.
A
B
C
D
A

346
PART 3 — RECONSTRUCTION
8. Conrad MH, Adams WP. Pharmacologic optimization of microsur­
gery in the new millennium. Plast Reconstr Surg 2001;108(7):
2088–97.
9. Froemel D, Barker JH, Frank J, et al. A review of thrombosis and
antithrombotic therapy in microvascular surgery. Eur Surg Res
2013;50(1):32–43.
10. Greenberg BM, Masem M, May JW. Therapeutic value of intrave­
nous heparin in microvascular surgery: an experimental vascular
thrombosis study. Plast Reconstr Surg 1988;82:463–72.
11. Riva FMG, Chen YC, Tan NC, et al. The outcome of prostaglandin-E1
and dextran-40 compared to no antithrombotic therapy in head
and neck free tissue transfer: analysis of 1,351 cases in a single
center. Microsurgery 2012;32(5):339–43.
12. Renck H, Ljungström KG, Hedin H, et al. Prevention of dextran-
induced anaphylactic reactions by hapten inhibition. III. A Scandi­
navian multicenter study on the effects of 20 ml dextran 1, 15%,
administered before dextran 70 or dextran 40. Acta Chir Scand
1983;149:355–60.
13. Notodihardjo HW, Ogawa Y, Kusumoto K. The blood flow pat­
terns of microsurgical procedures in rats with topical and systemic
vasodilators. Scand J Plast Reconstr Surg Hand Surg 1998;32:
249–54.
14. Karamursel S, Bagdatly D, Markal N, et al. Versatility of the lateral
arm free flap in various anatomic defect reconstructions. J Reconstr
Microsurg 2005;21:107–12.
15. Trussler AP, Watson JP, Crisera CA. Late free-flap salvage with
catheter-directed thrombolysis. Microsurgery 2008;28:217–22.
16. Ibrahim AMS, Kim PS, Rabie AN, et al. Vasopressors and recon­
structive flap perfusion: a review of the literature comparing the
effects of various pharmacologic agents. Ann Plast Surg 2013;00:
1–4.
17. Francel TJ, Vander Kolk CA, Yaremchuk MJ. Locally applied hypo­
thermia and microvascular muscle flap transfers. Ann Plast Surg
1992;28(3):246–51.
18. Shaw WW, Ko CY, Ahn CY, et al. Safe ischemia time in free-flap
surgery: a clinical study of contact-surface cooling. J Reconstr
Microsurg 1996;12:421–4.
19. Lee DT, Lee G. Cold ischemia in microvascular breast reconstruc­
tion. Microsurgery 2010;30:361–7.
20. May JW, Chait LA, O’Brien BM, et al. The no-reflow phenomenon
in experimental free flaps. Plast Reconstr Surg 1978;61:256–67.
21. Varkey P, Tan NC, Girotto R, et al. A picture speaks a thousand
words: the use of digital photography and the Internet as a cost-
effective tool in monitoring free flaps. Ann Plast Surg 2008;60(1):
45–8.
22. Engel H, Huang JJ, Tsao CK, et al. Remote real-time monitoring of
free flaps via smartphone photography and 3G wireless internet: a
prospective study evidencing diagnostic accuracy. Microsurgery
2011;31:589–95.
23. Cho BC, Shin DP, Byun JS, et al. Monitoring flap for buried free
tissue transfer: its importance and reliability. Plast Reconstr Surg
2002;110:1249–58.
24. Schramm S, Wettstein R, Wessendorf R, et al. Acute normovolemic
hemodilution improves oxygenation in ischemic flap tissue. Anes­
thesiology 2002;96:1478–84.
25. Rohde JM, Dimcheff DE, Blumberg N, et al. Health care-associated
infection after red blood cell transfusion: a systematic review and
meta-analysis. JAMA 2014;311(13):1317–26.
26. Cata JP, Wang H, Gottumukkala V, et al. Inflammatory response,
immunosuppression, and cancer recurrence after perioperative
blood transfusions. Br J Anaesth 2013;110(5):690–701.
27. Angel MF, Zhang F, Rogers B, et al. Role of free radicals in necrosis
of skin graft compromised with hematoma. Ann Plast Surg 2002;
48:665–9.
28. Yu JTS, Patel AJK, Malata CM. The use of topical vasodilators in
microvascular surgery. Br J Plast Surg 2011;64(2):226–8.
29. Evans GR, Gherardini G, Gürlek A, et al. Drug-induced vasodila­
tion in an in vitro and in vivo study: the effects of nicardipine,
papaverine, and lidocaine on the rabbit carotid artery. Plast Recon­
str Surg 1997;100:1475–81.
30. Kim DC, Chen D, Perrotta VJ, et al. Porcine gastroepiploic artery
as an in vitro experimental model to study vasodilators in micro­
surgery. Ann Plast Surg 1996;36(5):502–7.
31. Meyer VE. Smahel J, Valka J. Spherical dilators to resolve arterial
spasm. Acta Chir Plast 1999;41:117–19.
REFERENCES
1. Schwabegger AH. Flap re-exploration. In: Wei F-C, Mardini S,
editors. Flaps and reconstructive surgery. Philadelphia: Elsevier
Inc.; 2009. p. 125–36.
2. Rosenberg JJ, Fornage BD, Chevray PM. Monitoring buried free
flaps: limitations of the implantable Doppler and use of color
duplex sonography as a confirmatory test. Plast Reconstr Surg
2006;118(1):109–15.
3. Kroll SS, Schusterman MA, Reece GP, et al. Timing of pedicle
thrombosis and flap loss after free-tissue transfer. Plast Reconstr
Surg 1996;98(7):1230–3.
4. Bui DT, Cordeiro PG, Hu Q-Y, et al. Free flap reexploration: indica­
tions, treatment, and outcomes in 1193 free flaps. Plast Reconstr
Surg 2007;119(7):2092–100.
5. Khouri RK, Cooley BC, Kunselman AR, et al. A prospective study
of microvascular free-flap surgery and outcome. Plast Reconstr
Surg 1998;102:711–21.
6. Egozi D, Fodor L, Ullmann Y. Salvage of compromised free flaps
in trauma cases with combined modalities. Microsurgery 2011;
31:109–15.
7. Bui DT, Cordeiro PG, Hu QY, et al. Free flap reexploration: indica­
tions, treatment, and outcomes in 1193 free flaps. Plast Reconstr
Surg 2007;119(7):2092–100.
Expert Commentary 
Roman Skoracki, MD, FRCSC, FACS
Any center engaged in microvascular surgery should have
an algorithm in the event that a thrombosis is suspected.
This thoughtful and practical chapter serves as the perfect
resource to design a 1–2 page algorithmic template for such
a purpose. All team members, ICU or floor nursing staff,
residents and midlevel providers, and perioperative staff,
should practice the activation and execution of such an algo-
rithm on a regular basis. When such an algorithm is executed,
the preparedness for such an event makes the difference
between salvage and flap loss, even when the incidence of
thrombosis is low. This should be considered similar to a BLS
protocol for flap units. As such, it is the responsibility of the
microvascular surgeon to create an educational framework
to continuously educate and check the competency of the
staff for flap monitoring and salvage procedures.
In addition to this operational framework, there are a few
additional pearls in this chapter that deserve to be under-
scored. As prevention is better than cure, so it should be
stated that one should never leave the OR unhappy or even
with remaining doubt. After a long arduous day of complex
reconstructive microsurgery, the relief of completing the
case and exhaustion can easily lead us to ignore or rationalize
the early warning signs of a failing flap. However, there is no
time like the initial surgery to correct or avoid potential com-
plications. Sometimes this may be as simple as rearranging
the course of the pedicle, widely resecting a thrombosed
artery (beyond the area of what under the operating micro-
scope looks like “Velcro” of the intima after clot extraction),
or as complex as finding new recipient vessels, vein grafts,
and additional flap procedures. This may be an important
time to call for help if the length or complexity of the proce-
dure is clouding our judgment at its conclusion. A quick
examination of the flap vessels prior to complete flap inset
can ease the surgeon’s mind or serve as an early warning
sign of thrombosis. This examination includes confirming
visible pulsation in the artery, confirming the presence of a
soft vein with adequate flow, and checking for the absence
of engorgement of the second vena comitans or other
clipped venous branches.

CHAPTER 28 — Flap Re-exploration and Salvage
347
32. Pinder RM, Hart A, Winterton RIS, et al. Free tissue transfers in
the first 2 years of life – a successful cost effective and humane
option. J Plast Reconstr Aesthetic Surg 2010;63:616–22.
33. Calhoun KH, Tan L, Seikaly H. An integrated theory of the
no-reflow phenomenon and the beneficial effect of vascular
washout on no-reflow. Laryngoscope 1999;109:528–35.
34. D’Arpa S, Cordova A, Moschella F. Pharmacological thrombolysis:
one more weapon for free-flap salvage. Microsurgery 2005;25(6):
477–80.
35. Yii NW, Evans GRD, Miller MJ, et al. Thrombolytic therapy: what is
its role in free flap salvage? Ann Plast Surg 2001;46:601–4.
36. Serletti JM, Moran SL, Orlando GS, et al. Urokinase protocol for
free-flap salvage following prolonged venous thrombosis. Plast
Reconstr Surg 1998;102(6):1947–53.
37. Nelson JA, Kim EM, Eftekhari K, et al. Late venous thrombosis in
free flap breast reconstruction: strategies for salvage after this real
entity. Plast Reconstr Surg 2012;129(1):8e–15e.
38. Casey WJ, Craft RO, Rebecca AM, et al. Intra-arterial tissue plas­
minogen activator: an effective adjunct following microsurgical
venous thrombosis. Ann Plast Surg 2007;59(5):520–5.
39. Wang TY, Serletti JM, Cuker A, et al. Free tissue transfer in the
hypercoagulable patient: a review of 58 flaps. Plast Reconstr Surg
2012;129(2):443–53.
40. Herrera FA, Lee CK, Kryger G, et al. Microsurgery in the hyperco­
agulable patient: review of the literature. J Reconstr Microsurg
2012;28(5):305–12.

348
Postoperative Care
Jonathan Rimler  |  Audrey Nguyen  |  Gregory R.D. Evans
29 
INTRODUCTION
Microsurgical reconstruction with free tissue transfer has
become standard practice for plastic surgeons around the
world with varied uses and applications. However, periop­
erative management of these complex patients is diverse. As
the understanding of flap physiology and refinement of
microsurgical techniques has evolved, both pedicled flaps
and free tissue transfer have become reliable and effective
methods of producing sound functional and aesthetic
reconstructive outcomes.
The continued refinement of microvascular instruments,
magnification, and monitoring techniques has vastly im­
proved the overall reported free flap success rates. Unfortu­
nately, the rare complication or flap failure may prove
devastating and create significant patient morbidity or even
mortality. It is therefore essential to reduce the risks of
postoperative complications as well as identify and manage
them early when they occur. Factors associated with the
development of complications following tissue transfer may
include the patient’s age, comorbid medical conditions,
tobacco usage, and surgical technique.
This chapter presents guidelines for perioperative risk
reduction, flap monitoring, complication management, and
discharge care for patients undergoing flaps, with a focus
on free tissue transfer.
PERIOPERATIVE RISK REDUCTION
A patient’s comorbid conditions related to age, underlying
disease, and cardiopulmonary or vascular status have an
important effect on medical and surgical complications.
For example, medical complications and overall length of
hospital stay are higher in the elderly population. In addi­
tion, tobacco use has been associated with a significant
increase in the overall incidence of medical morbidity.
While complications associated with tissue transfer surgery
can never be completely avoided, the risk of their occur­
rence may be reduced by thoughtful preoperative eval­
uation, perioperative prophylactic strategies, meticulous
attention to surgical technique, and diligent patient moni­
toring (see also Chapter 27, Avoiding Complications).
PREVENTION OF POSTOPERATIVE
CARDIAC COMPLICATIONS
Myocardial ischemia occurring within the first postoperative
week after non-cardiac surgery is a major risk factor for
future morbidity and mortality in surgical patients.1 The
incidence of a serious cardiovascular event occurring within
2 years for this patient population is up to 20-fold higher.1
Postoperative myocardial ischemia is a potentially reversible
risk factor for future complications and mortality.
The role of perioperative beta-blockade in non-cardiac
surgery has matured in recent years. The development and
validation of the revised cardiac risk index (RCRI) (Table
29.1) has allowed for more accurate use of beta-blockers
with respect to stratified risk. Early meta-analysis demon­
strated decreased cardiac complications (50% reduced mor­
tality rates) using atenolol. However, this was not stratified
by cardiac risk.2,3 High powered studies have revealed that
beta-blocker use may be harmful to some patients, causing
hypotension, bradycardia, or even stroke.4,5 For patients
with low cardiac risk (RCRI = 0), potential complications of
beta-blocker use outweigh their benefits. High cardiac risk
patients with RCRI ≥3 show a clear benefit with periopera­
tive use of a beta-blocker. Intermediate risk patients with 1–2
risk factors have uncertain benefits, except vascular surgery
patients.6,7
Perioperative cardiac management is challenging. We
recommend a multidisciplinary care approach with anesthe­
sia and medicine to help choose the best cardiac prophy­
laxis for each patient.
DEEP VEIN THROMBOSIS/PULMONARY
EMBOLISM PROPHYLAXIS
Deep venous thrombosis (DVT) and pulmonary embolism
(PE) pose a clinical challenge for the reconstructive surgeon.
Table 29.1  Revised cardiac risk index criteria
and probability of cardiac events
Revised cardiac risk index (RCRI)
Predictors
History of ischemic heart disease
History of congestive heart failure
History of cerebrovascular disease (stroke
or transient ischemic attack)
History of diabetes requiring preoperative
insulin use
Chronic kidney disease (creatine > 2 mg/dL)
Undergoing suprainguinal, vascular,
intraperitoneal, or intrathoracic surgery
Risk for non-cardiac death, non-fatal myocardial infarction,
non-fatal cardiac arrest: 0 predictors = 0.4%, 1 predictor =
0.9%; 2 predictors = 6.6%; >3 predictors = 11%.
PART 3  —  RECONSTRUCTION
SECTION C  —  PRE AND POSTOPERATIVE MANAGEMENT

CHAPTER 29 — Postoperative Care
349
The risk for a DVT/PE and its sequelae must be weighed
against the risk of hematoma formation. Although venous
clots can form in any part of the venous system, they most
commonly develop in the calves.8–10 Approximately 50% of
patients with untreated proximal DVTs will develop a symp­
tomatic pulmonary embolism within 3 months.11 Long-term
morbidity from post-thrombotic syndrome may be substan­
tial. Most fatal events of venous thromboembolism (VTE)
occur within 3–7 days of surgery.11 As such, deep vein throm­
bosis and pulmonary embolism are among the best exam­
ples of preventable postoperative complications.
Plastic surgeries vary widely in terms of risk strata, and
overall VTE incidence, and is reported to be on average
1.69%.12 Few studies have examined the VTE incidence
in free tissue transfer, however, DVT rates of 3.4% (range
of 2–6%) have been reported for free flap breast
reconstruction.13–15 More complicated free flap procedures
seemingly may have a higher rate of VTE due to increased
operative time and more complicated patients.
Primary thromboprophylaxis is effective in reducing the
incidence of asymptomatic and symptomatic VTE in sur­
gical patients via mechanical measures or anticoagulant
drugs. According to the American College of Chest Physi­
cians’ Antithrombotic Therapy and Prevention of Thrombosis,
9th edition (AT9) practice guidelines, VTE falls into four
general risk categories: very low, low, moderate, and high.16
Several studies have also demonstrated that certain factors
place patients at higher risk, including age >60 years, malig­
nancy, prior history of VTE, bed rest >4 days, and anesthesia
>2 h.17,18 The Caprini score, which was developed in the
early 2000s, assigns various points to patient characteristics
that put them at increased risk of thrombosis and is incor­
porated into the AT9 guidelines for plastic surgery patients
(Table 29.2).12,16,19–22 The Caprini score was subsequently
validated in plastic surgery patients.12
Most hospitals use guidelines similar to those in Table
29.2, and in the United States VTE prophylaxis is required
within 24 h of surgery unless contraindicated. The risks of
VTE must be weighed against those of hematoma, since free
tissue transfer often entails wide undermining in natural
tissue planes. Most plastic surgery patients fall into the
moderate–high category with recommended chemoprophy­
laxis. Hematoma rates are 0.5–1.8% without prophy­
laxis,14,23,24 and combined use of mechanical prophylaxis
and low-molecular-weight heparin (LMWH) does not create
a significant difference in hematoma formation compared
with mechanical means alone.25
Suspected DVT should be diagnosed promptly. In patients
with low clinical suspicion, a D-dimer (fibrin degradation
product) test can be completed and if negative, no further
testing is needed.26 In patients with multiple risk factors
and high clinical suspicion, compression ultrasound is the
test of choice. Ultrasonography has a sensitivity of 97–100%
and a specificity of 98–99% for detection of proximal
DVT.27 Therefore, full compressibility of the femoral and
popliteal veins essentially excludes proximal deep vein
thrombosis. The rate of VTE in patients with a negative
ultrasound was 0.7%, indicating that few thromboses were
missed and that therapeutic anticoagulation may be safely
withheld when ultrasonography is normal.28 On the other
Table 29.2  AT9 and Caprini risk stratification for VTE and recommended regimen
Very low risk
Low risk
Moderate risk
High risk
Caprini score
0–2
3–4
5–6
7–8
Deep vein
thrombosis
N/A
0.60%
1.30%
2.80%
Prophylaxis
regimen
Ambulation and mechanical
compression
Mechanical
compression and/
or
chemoprophylaxis
Mechanical
compression +
chemoprophylaxis
Mechanical
compression +
chemoprophylaxis
Caprini
stratification
1 point each
2 points each
3 points each
5 points each
Risk factors
Age 41–60 years
Minor surgery
History of major surgery (<1 month)
Varicose veins
Inflammatory bowel disease
Swollen legs
Obesity
Acute MI
CHF exacerbation (<1 month)
Serious lung disease (<1 month)
Abnormal pulmonary function
Current bed rest
Oral contraceptive (women)
Hormone replacement (women)
Pregnancy/Postpartum (women)
History of antiphospholipid
antibody syndrome (women)
Age 60–74 years
Arthroscopic surgery
Malignancy (present
or previous)
Major surgery
(>45 min)
Laparoscopic
surgery (>45 min)
Confined to bed
(>72 h)
Immobilizing plaster
cast (<1 month)
Central venous
access
Age over 75 years
History of DVT/PE
Family history of
thrombosis
Factor V Leiden
Positive prothrombin
2021QA
Elevated serum
homocysteine
Positive lupus
anticoagulant
Elevated anticardiolipin
antibodies
Heparin-induced
thrombocytopenia
Acquired thrombophilia
Elective major lower
extremity
arthroplasty
Hip, pelvis, or leg
fracture (<1 month)
Multiple trauma
(<1 month)
Acute spinal cord
injury (paralysis)
(<1 month)

350
PART 3 — RECONSTRUCTION
Control of postoperative pain is a vital component for
improving pulmonary function by reducing inspiratory
splinting and allowing for early ambulation. There are
many strategies available for pain control (oral, IV,
patient-controlled anesthetic [PCA] and regional). These
approaches are highly subjective and must be tailored to
each patient. Paramount is frequent reassessment and
adjustment of medications to allow pain control without
oversedation. Intravenous or oral narcotics are the first
choice. However, these medications have significant
side effects, including somnolence, constipation, urinary
retention, respiratory depression, nausea, and altered
mental status, especially in the elderly. Recently, there has
been increasing use of regional anesthesia (pain pumps,
nerve blocks, epidurals) for tissue transfer patients. In fact,
epidural use in free flap surgery significantly reduces the
percentage of patients with postoperative atelectatic fevers.38
In addition, disposable pain pumps can infuse 0.25% bupi­
vacaine at a rate of 2.08 cc/h directly into the surgical site
for 48–72 continuous hours. Patients who use these devices
report a two-fold decrease in their overall pain level and
have a five-fold decrease in use of oral and intravenous pain
medications with fewer episodes of nausea, emesis, and
other side effects.39 Epidural anesthesia has also been
shown to improve pulmonary function by giving excellent
pain control, improved pulmonary compliance, and allow­
ing normal diaphragm function. However, epidural anes­
thesia has not been proven to decrease the incidence of
pneumonia.40,41
HYPERGLYCEMIC CONTROL
Hyperglycemia in response to physiologic stress occurs in
patients who undergo major surgery, trauma, or sepsis, as
well as exogenous causes of hyperglycemia, including steroid
administration and other therapeutic interventions.42–44
This “stress hyperglycemia” applies to non-diabetic patients;
however, patients with preexisting diabetes are also at risk
for hyperglycemic events in the perioperative period.42,45,46
Hyperglycemia is associated with poorer outcomes, in­
creased risk for wound infection, and increased morbidity
and mortality in both diabetic and non-diabetic patients.43,47–51
While intensive insulin therapy (blood glucose, BG 80–
110 mg/dL) was initially believed to reduce mortality,
blood­stream infections, renal failure, and prolonged
mechanical ventilation,51 it is now apparent that intensive
control results in a six-fold increase in severe glycemic
events (BG <40 mg/dL). This is an independent risk factor
for mortality, especially in critically ill patients compared
with conventional control.52,53 It is recommended that sur­
geons aim for blood glucose levels that are slightly higher
than 80–110 mg/dL and <180 mg/dL, though no threshold
value is currently defined in the literature.45,54
WOUND INFECTION PROPHYLAXIS
To achieve optimal prophylaxis against surgical site infec­
tions, adequate concentrations of an appropriate antibiotic
must be present in the serum, tissue, and wound for the
entire time that the incision is open and at risk for bacterial
contamination. Antibiotics should be active against the type
of bacteria that are likely to be encountered during a
hand, if the ultrasound is positive, treatment should begin
with LMWH, fondaparinux, or IV, or subcutaneous unfrac­
tionated heparin.26 If a pulmonary embolism is suspected,
CT angiography is the diagnostic test of choice with clinical
validity similar to that of conventional angiography.29 The
total rate of subsequent venous thromboembolic events
after a negative CT scan at 3 months is 1.4%.29
The AT9 recommends initial treatment with LMWH or
fondaparinux instead of unfractionated heparin.26 Unfrac­
tionated heparin is inconvenient for both patients and
medical staff, due to frequent injection and monitoring.
Low-molecular-weight heparin does not require laboratory
monitoring due to a more predictable dose–response rela­
tionship of the medication. Its half-life is longer than unfrac­
tionated heparin, which allows for once or twice daily dosing,
and it has a reduced risk for immune mediated thrombocy­
topenia or osteoporosis.30–32 In severely obese or renal insuf­
ficiency patients, monitoring anti-factor Xa activity 4 hours
after injection may be useful.33 Thus, LMWH is the drug of
choice. Daily doses of LMWH are preferred over twice daily.
If the patient has a contraindication to anticoagulation,
a retrievable inferior vena cava filter may be placed to
prevent fatal pulmonary embolism. This device may be
removed at a later date when the danger has passed.
PNEUMONIA AND PULMONARY PROPHYLAXIS
Postoperative patients, especially the elderly and those with
chronic obstructive pulmonary disease, are prone to devel­
oping acute pulmonary complications from either poor
inspiratory effort while lying supine or respiratory splinting
from incisional pain. Pulmonary complications after surgery,
much as with cardiac complications, can result in longer
hospitalization, increased cost, and increased morbidity/
mortality.
Atelectasis and pulmonary infiltrates can rapidly develop
and progress to pneumonia in the early postoperative
period. An incentive spirometer measures the inspiratory
capacity of the lungs and provides patients with visual cues
on their progress. This encourages daily goals and deeper
inspiratory effort. Meta-analysis demonstrates that incentive
spirometry is the only strategy of proven benefit. Selective
nasogastric tube use has probable benefit reducing pneu­
monia and atelectasis in patients with nausea, emesis, or
distention when there has been an abdominal component
to surgery.34–36
If pneumonia is suspected, obtaining a chest X-ray and
sputum culture followed by immediate initiation of antibi­
otic therapy is essential. Empiric antibiotic therapy should
treat bacteria that are common for hospital-acquired pneu­
monias such as Pseudomonas, Enterobacteriaceae, Streptococci,
and Haemophilus. Antibiotic-resistant pathogens such as
Pseudomonas, Acinetobacter, and methicillin-resistant strains of
Staphylococcus aureus are much more common after prior
antibiotic treatment, prolonged hospitalization, or mechan­
ical ventilation. After 48–72 h, antibiotics should be tailored
appropriately to the organisms growing on culture. Correct
empirical antibiotic treatment is vital, and local resistance
patterns should be discussed with the infectious disease
team as improper antibiotic use shows higher mortality.37
Postoperative nosocomial pneumonias should be treated
for 1–2 weeks with appropriate antibiotic therapy.

CHAPTER 29 — Postoperative Care
351
to assess, despite the techniques available today. The major­
ity of surgical complications after tissue transfer surgery are
related to vascular thrombosis, which usually occurs within
3 days of surgery. However, late thrombosis can occur and
is often associated with local infection or mechanical com­
pression of the vascular pedicle. Therefore, nursing staff
routinely monitor all free flaps hourly for the first 24 hours
postoperatively (the authors admit patients to the ICU for
the first night), then every 2 hours for the next 1–2 days,
and thereafter every 4–8 hours until discharge. There are
many techniques available for evaluation of the flap viability.
The most common techniques are clinical observation and
Doppler vascular pedicle monitoring.
CLINICAL ASSESSMENT
To construct a skin paddle, a portion of the free flap’s skin
is exteriorized and may be used to evaluate perfusion by
monitoring skin temperature, capillary refill, turgor, color,
and bleeding. This skin paddle technique is one of the most
accurate and reliable ways to evaluate flap perfusion.65 It
allows the surgeon to monitor for both arterial insufficiency
as well as venous obstruction. Two techniques are commonly
used for examining bleeding from the flap, either pinprick
with a 25-gauge needle or a partial dermal incision with a
15 blade. Arterial insufficiency is present when the skin is
pale and cool and fails to bleed after a needle stick or partial
dermal cut. Venous congestion usually results in edema and
darkening of the skin color. During early venous obstruc­
tion, a needle stick or partial dermal cut will cause rapid
bleeding of dark blood. A skin paddle should be large
enough to make correlation with the vascular pedicle more
reliable. Early signs of vascular compromise may be subtle
while monitoring a skin paddle and require an experienced
examiner to recognize problems.
Clinical observation is indispensable in monitoring flaps.
However, most surgeons rely on adjunctive methods to opti­
mize flap salvage because continuous observation by the
surgeon is currently not possible. Adjunctive monitoring
methods allow the nursing staff to observe the flap and alert
the surgeon at the first sign of vascular insufficiency.
EXTERNAL DOPPLER
Currently, there is no single adjunctive monitoring tech­
nique widely accepted as the method of choice, but the
handheld Doppler is the most common device in use. We
prefer the ES-100x Mini-Doppler with the 8 MHz probe by
Koven Technology Inc (St Louis, MO). Its most important
limitation is that a clinician may detect the recipient vessels’
Doppler signal instead of the flap’s vascular pedicle due to
proximity, which may mislead the observer to believe that
the flap’s pedicle is patent when in reality, a thrombosis has
occurred. Therefore, it is recommended to place a marking
suture in the exact location of the vascular pedicle’s location
relative to the skin. Handheld Dopplers are also useful for
evaluating buried flaps. However, this is not as easy or as
accurate, and sometimes not even possible.
The arterial and venous signals can both be monitored
using this Doppler device. External Doppler monitoring
of the vein will demonstrate flow that increases when the
flap is compressed (venous augment) Additionally, the
particular type of operation. The choice and duration of
antibiotic prophylaxis should have the least effect on the
patient’s normal bacterial flora and be based upon expected
bacterial contamination, as well as fit within the microbio­
logic ecology of the hospital.
A beta-lactam antibiotic (e.g., cefazolin) is usually the
first choice for perioperative prophylaxis and is recom­
mended to begin 60 min prior to the incision, However,
research has demonstrated that infusion at the time of anes­
thesia induction is safe and results in adequate serum and
tissue antibiotic levels prior to the incision. If the patient
has a beta-lactam allergy, vancomycin or clindamycin is
recommended.55
The majority of published research demonstrates that
antimicrobial prophylaxis after wound closure is not neces­
sary, and most studies, including those in the plastic surgery
literature, that have compared single- to multiple-dose pro­
phylaxis, have not revealed additional benefits.56,57 However,
prolonged use of prophylactic antibiotics is associated with
the emergence of resistant bacterial strains.58 Antibiotic pro­
phylaxis should end within 24 hours after the operation.55
Prophylaxis should be used for cases with prosthetics,
microsurgical cases, clean contaminated cases in the oral
cavity or genitourinary system, ventral hernias, or contami­
nated cases.59 While these data are promising for evidence-
based practice in plastic surgery, there is still no consensus.
Preventing intraoperative hypothermia, perioperative sup­
plemental oxygen, and aggressive fluid resuscitation, and
prepping the operative field with chlorhexidine may
decrease surgical site infection rates.
SMOKING CESSATION
Smoking in the perioperative time period is related to
higher complications, especially in relation to wound
healing. There is a high correlation between smoking and
rates of reoperation and tissue necrosis.60 Urine cotinine
tests clearly demonstrate a high preponderance of decep­
tion regarding this habit as well. Of patients who stated that
they quit smoking, 4% will have positive urine tests.
It is well documented that smoking increases the overall
complication rate for tissue transfer surgery.61 For example,
smoking patients undergoing reconstruction with pedicle
TRAM flaps have increased flap, abdominal skin, and umbil­
ical necrosis.61 Heavy smoking is generally considered to be
a contraindication for pedicled TRAM flap breast recon­
struction.61 However, free TRAM flap reconstruction in
smokers is not associated with increased pedicle thrombosis,
flap loss, or fat necrosis, but does have increased risk for
mastectomy skin flap necrosis, abdominal flap necrosis,
and hernia formation.62,63 These perioperative complica­
tions are significantly reduced when the reconstruction
is delayed at least 4–6 weeks after smoking cessation, and
some studies even suggest 8 weeks as a more appropriate
timeframe.63,64
FLAP MONITORING
Accurate assessment of free or pedicle flap perfusion has
always been a challenge for surgeons and nursing staff. The
intricacies of flap microcirculation are frequently difficult

352
PART 3 — RECONSTRUCTION
88%, and in two cases, the laser Doppler was the only clini­
cal measure of flap decline.73 This system can be helpful in
darkly pigmented skin or a pure muscle flap, where clinical
evaluation is not straightforward. However, the laser Doppler
flowmeter equipment is expensive, and some studies have
found it to be inaccurate as a result of probe dislocation
artifacts. The laser Doppler flowmeter, as other monitoring
systems, may provide the reconstructive surgeon with an
early indication of impending flap failure, thus providing
enhanced opportunity for flap salvage.
NEAR INFRARED ANGIOGRAPHY
Near infrared angiography is a semi-quantitative measure of
microvascular perfusion within a soft tissue flap.74 This tech­
nique is relatively new and is based on the detection of a
systemically administered dye (indocyanine green) by
exposing the tissue to near infrared light and detecting
emitted signals through an optic filter. Not only can this
technique determine the quality of perfusion throughout
the flap, but it also may detect the area of tissue nourished
by a particular vascular pedicle on which a flap is raised,
thus assisting in the flap design.74 This method has been
shown to be useful in evaluating zonal perfusion of free
TRAM flaps intraoperatively as a way to avoid partial flap
necrosis.75 Another advantage of near infrared angiography
is that it provides discrete visualization of perforator blood
vessels. This method, however, is a static evaluation of the
perfusion at one particular point in time and cannot be
used as a continuous form of monitoring. A flap with a skin
paddle or meshed split-thickness graft over muscle is neces­
sary for monitoring, which prevents evaluation of a buried
flap. Near infrared angiography may appear too sensitive,
with a 66% false-positive rate found in a small series attempt­
ing to detect postoperative venous congestion.74
PULSE OXIMETRY
This method relies on two diodes that emit light at a wave­
length of 660 nm (visible) and 940 nm (infrared) and a
photodiode receiver. It measures oxygen saturation by
detecting a difference in light absorption between oxyhe­
moglobin and reduced hemoglobin. It provides continuous
monitoring and is particularly useful in monitoring digit
replants and toe-to-hand transfers.
PHOTOGRAPHY
A useful tool used by some surgeons is photography. They
take a picture of the flap when the flap appears well per­
fused and without congestion and post the picture near the
bedside. Nurses are instructed to call the surgeon if there is
a change in appearance of the flap when compared with the
picture.
SURFACE TEMPERATURE
For intraoral flaps and thin flaps placed on the trunk,
surface temperature monitoring may be less accurate. A
surface temperature probe (a liquid crystal temperature
probe) may be placed on the flap and another on a
normal adjacent skin. Temperature differences of 3°C are
type of arterial signal can be distinguished by an experi­
enced listener. A triphasic signal that becomes monophasic
may indicate a change in the status of the flap and venous
occlusion.
INTERNAL DOPPLER
The implantable Doppler can measure blood flow across a
microvascular anastomosis and is an effective tool to monitor
flap perfusion and improve salvage rates, especially in
buried flaps.66 Initial experience with this device was not
positive with a 3% false-positive rate, leading to unnecessary
re-explorations, and a 5% false-negative rate when the probe
was placed on the artery.67 Later studies showed a false-
positive rate of 6.8% and a false-negative rate of 0%.68
Further, up to a 5-h delay was found between a venous
obstruction and the loss of the arterial signal in large muscle
flaps.66 This is also a disadvantage for the handheld Doppler.
However, some studies suggest that placing an implant­
able probe on the vein instead of the artery is a more rapid
and accurate method of detecting a pedicle thrombosis.
This technique detected any vascular compromise without
delay.68,69 These probes may be useful in patients with darkly
pigmented skin in whom clinical detection of venous con­
gestion is difficult.
With buried flaps, the implantable Doppler has been
shown to have similar flap success rates (98.2%), and has
allowed for salvage of buried flaps that are difficult to
monitor.70 Inexperienced nursing and house staff can
quickly detect a perfusion problem in the flap with the
internal Doppler probe. However, the probe is costly, with
reports of malfunction or it becoming displaced during the
early postoperative course.71 These problems require further
monitoring techniques, such as clinical observation or
handheld Doppler examination to be performed. For
buried flaps, clinical examination includes monitoring for
increase in edema. This can indicate congestion of the flap
due to venous thrombosis or inadequate drainage. It could
also indicate hematoma formation, which could be the
result of a venous obstruction leading to bleeding from the
buried flap.
LASER DOPPLER
A laser Doppler flowmeter provides an objective measure­
ment for flap perfusion, but the observer cannot rely on
absolute values. It requires experience to interpret, as rela­
tive perfusion values are different for every tissue and
patient, and may fluctuate because of physiologic microcir­
culation variation or artifacts. Therefore, the observer must
monitor the trend rather than the absolute values. This fact
is of particular importance during venous occlusion, when
a drop in value is not as abrupt and steep compared with
arterial obstruction.
Yuen and Feng had a series of nursing staff record values
every 15 min for 1 h, every 30 min for 2 h, and then every
4 h for 4–5 days. When a vascular occlusion occurred, the
perfusion values decreased by 50% or more for longer than
20 min.72 Using these criteria, they reported no false-positive
values with no unnecessary re-explorations and one false-
negative when the observation protocol was violated. In
another series, the flap salvage rate with laser Doppler was

CHAPTER 29 — Postoperative Care
353
that results in vessel thrombosis. This reinforces the need
for meticulous microsurgical technique and the rationale
for prophylactic antithrombotic therapy. The necessity
for antithrombotic therapy is frequently debated and cur­
rently not universally employed. Some antithrombotic
agents used are heparin, dextran, aspirin, and platelet gly­
coprotein IIb/IIIa inhibitors. Low-molecular-weight heparin
is usually used after re-exploration and thrombectomy for
vascular failure. It is rarely used for primary thrombosis
prevention because of the risk for hematoma formation.
However, low-dose unfractionated heparin has not been
demonstrated to increase the rate of hematomas. Low-
molecular-weight heparin increases vascular pedicle patency
rates, without increasing hematoma formation in animal
studies.81 There is no difference in vascular patency rates
when dextran is used regularly, but there is an increased risk
of anaphylactoid reactions, adult respiratory distress syn­
drome, cardiac overload, hemorrhage, and renal damage.82
Low-dose aspirin inhibits anastomotic venous thrombosis
and improves microcirculatory perfusion,83 while glycopro­
tein IIb/IIIa inhibitors prevent reperfusion injury after a
period of ischemia.84 Some antithrombotic agents have
been proven to help prevent thrombosis but no ideal
regimen has been established.
THROMBOLYTIC THERAPY
Thrombolytic drugs such as urokinase, streptokinase, and
recombinant tissue-plasminogen activator (T-pa) may be
used to treat vascular pedicle thrombosis or act as a compli­
ment to reestablish flow when surgical exploration fails to
restore circulation. Intraoperatively, 100 000–250 000 IU of
urokinase/streptokinase or 15 mg of T-pa are infused over
30 min through a side branch or resected end of the venous
or arterial anastomosis.85,86 To prevent the undesired sys­
temic side effects of thrombolytic therapy, the venous anas­
tomosis may be taken down and allowed to bleed freely
during infusion.85,86
To further reduce the risk of systemic side effects for
thrombolytic therapy, local intra-arterial and intravenous
infusions, as well as regional soft tissue injection techniques
have been developed.87 Arterial thrombi can be treated with
intra-arterial thrombolytic infusion with the catheter placed
proximal to the arterial anastomosis, while venous thrombi
are lysed through an engorged vein of the replanted limb
or flap.87 In replanted digits, where direct intravenous infu­
sion is not possible, injection directly in the soft tissues may
result in successful salvage.87 Currently, there is no large
series investigating this technique.
DISCHARGE CARE
In general, the same hospital precautions apply when a
patient is discharged home. All patients are sent home with
an incentive spirometer and instructions to stay active by
participating in light physical activity 3–4 times daily (e.g.,
walking). They are all given oral narcotics and laxatives for
pain control and bowel care. All patients are instructed on
wound and drain care if needed, and the signs/symptoms
for a wound infection. After discharge, patients are seen in
the clinic per individual physician preference.
associated with arterial insufficiency and 1–2°C may be
more associated with venous insufficiency.
CELLULAR AND WIRELESS DEVICES
The wide dissemination of cell phones and wireless technol­
ogy is also having implications on free tissue transfer moni­
toring. Recently, cell phone imaging was used to clinically
evaluate flap perfusion in the early perioperative time
period, allowing for senior staff surgeons to more rapidly
assess a flap with questionable vascular status. Flap survival
rose to 100%, flap salvage rose from 50–100%, and time
from flap compromise to reoperation was also drastically
decreased.76,77 When compared with traditional monitoring,
remote smart phone photography shows similar accuracy in
assessing flaps.77 These are promising findings, and further
integration with wireless devices will surely lead to stream­
lined care for tissue transfer patients. It is also likely that
advances in telemedicine will allow senior surgeons to
remotely view a flap in high definition to allow accurate and
experienced assessment in real-time. As these technologies
continue to evolve, data encryption and HIPAA compliance
will be essential for long-term viability of these unique and
powerful monitoring.
MANAGEMENT OF SURGICAL
COMPLICATIONS
SURGICAL EXPLORATION
No matter the technique for flap monitoring, prompt surgi­
cal exploration is mandatory if a flap’s vascular integrity is
in question. The majority of tissue transfer complications are
related to vascular pedicle thrombosis. Currently, free flap
thrombosis rates are approximately 3.2–4%, with 80% of
occlusions arising within the first 48 hours.78–80 While an
arterial or venous occlusion may occur, venous thrombosis
occurs more frequently due to its low flow, low pressure
nature.79 Arterial occlusion is a rapid event, while venous
thrombosis may evolve over several hours. Early detection of
vascular occlusions is essential for flap salvage. Surgical
re-exploration is mandatory and should be undertaken
without delay. Unfortunately, the reported flap survival rate
after an arterial thrombosis and thrombectomy is only 40%.80
In contrast, after a venous occlusion and thrombectomy, the
flap survival rate is 71%.80 A theory as to why venous throm­
bosis has a better salvage rate may be because conventional
monitoring detects venous congestion more easily.79
Hematoma, seroma, and partial flap loss are additional
complications that may require further surgical interven­
tions. Hematomas or seromas may form rapidly filling large
dead spaces. Their expansion may place tension on the vas­
cular pedicle, causing occlusion. Drain placement intra- or
postoperatively with ultrasound guidance can minimize fluid
collections and alert the team to continued surgical bleed­
ing. Partial flap loss may require multiple surgical debride­
ments to remove necrotic tissue and prevent infection.
ANTITHROMBOTIC THERAPY
Vascular endothelial damage from surgery results in platelet
aggregation and the initiation of a complex clotting cascade

354
PART 3 — RECONSTRUCTION
instructed to progressively increase their dangling time until
ambulating normally.
CONCLUSION
Free tissue transfer is technically complicated, and patients
undergo risk stratification and perioperative prophylaxis,
just as in any other major surgery. While complications in
free tissue transfer patients cannot be completely prevented,
they can be minimized. Careful perioperative screening and
medical management, initiation of sound prophylactic strat­
egies, meticulous surgical technique, and diligent use of
perioperative monitoring all help to reduce overall periop­
erative risk. When complications do occur, it is also essential
to act in a definitive and appropriate way to minimize mor­
bidity and maintain patient safety.
For lower extremity free flaps, patients are sent home on
a prolonged extremity elevation protocol. Recent literature
supports that lower extremity free tissue transfers develop
decreased tissue oxygenation with dangling, that wrapping
the leg can decrease this phenomenon, and that a more
aggressive early dangling protocol can be instituted without
adverse outcomes with the tissue transfer.88,89 While no con­
sensus exists for perioperative dangling protocols, plastic
surgeons average 7.8 days prior to initiation. A more aggres­
sive protocol is initiated by 42% of surgeons, starting dan­
gling between 2–5 days, while 37.8% of surgeons initiate
dangling at 6–13 days.90 A typical protocol would be 1 week
of continuous elevation above the heart in the hospital, at
which point patients are instructed to dangle their leg at the
bedside for 5 min every hour for 5 days with a compression
stocking in place. If flap edema is minimal, patients are
Expert Commentary
Lucie Lessard MD, CSPQ(O), CSPQ(P), FRCSC(O),
FRCSC(P), FACS
In plastic surgery/microsurgery, the evolution of flaps has
evolved greatly in the last 30 years. In microsurgery, from a
success rate of around 70–80% originally, we are now trying
for almost 95–100% success rate, many years later.
This chapter reviews very thoroughly the perioperative risk
reduction, antibiotic strategy, flap monitoring, early complica-
tion management, and discharge care. It also provides excel-
lent in-depth references.
In most centers,1 patients are sent for the first 24–48 h to the
intensive care unit (ICU) for an hourly monitoring. In several
countries, ICU beds are not always available because of finan-
cial limitations. At our teaching institution, we have overcome
that issue by having dedicated plastic surgery nurses trained
to follow flaps routinely with residents overseeing their hourly
checks on the ward most of the time or in the recovery room
overnight if faced with bed shortage.
Our protocol is similar to that mentioned in the chapter, with
the following important “top 10” items (in chronology):
1. Regional anesthesia and axillary/spinal/epidural block
preoperatively in upper and lower extremity to control
spasm and achieve vasodilatation to keep a good perfusion
is our routine.
2. Preoperative antibiotherapy loading dose to decrease
infection rate. (Third party payers in the United States are
also monitoring precise protocols in relation to complica-
tion rate and standard of care.)
3. DVT prophylaxis (LMWH) is also important and we use a
sequential pneumatic pump to the lower extremities on all
our microsurgery elective cases and replantations. ASA
81 mg/QD is our only Rx for 1 month in elective cases, with
the exception of hand trauma, where we use full anticoagu-
lation if trauma is severe and take-back.
4. Vein grafts are limited to strict essential to bridge trau-
matic zone in replant.
5. In vein anastomosis, the coupler technique has been
favored to keep a good patency with the device rigidity
itself, 95.9% in lower extremity2 and other sites, as well as
to save time.
6. In head and neck cases, to avoid misplacement and
kinking of vessels, head hyperextension has to be reversed
prior to skin closure to place the vessels/anastomosis stra-
tegically right without excessive length.
7. In head and neck/cervical, anastomoses have to be stra-
tegically placed to follow the vessels with a Doppler away
from the carotid vessels.
8. Warming device (Bair-hugger) always to keep patient’s
operated site well perfused and vasodilated; in fingers
replantation, I will sometimes add an electrical pad proxi-
mal at the shoulder level in a cold room/winter.
9. If all fails, the use of leeches has been interesting to the
author. Prior to using leeches we proceeded with an animal
study of an ischemic/reperfusion injury rat model. The
leeches were successful at salvaging a larger portion of the
flap significantly.3 Chemical leeching is another option but
may require blood transfusions.4
10. To insist on patient smoking cessation and family non-
smoking collaboration5 around the patient, is essential.
As an anecdotal situation, one of my patients thrombosed
his flap on his way home on day 13, while his wife was
smoking for 2 h in the car driving him back; he was not
smoking himself!
In summary, a good technique and early exploration is still
the best strategy to salvage flaps with complications. Diligent
hourly clinical follow-up and Doppler monitoring is also very
important to identify the complication efficiently for a better
outcome for the patient.
REFERENCES
1. Mangano DT, Browner WS, Hollenberg M, et al. Long-term cardiac
prognosis following noncardiac surgery. JAMA 1992;268:233–9.
2. Mangano DT, Layug EL, Wallace A, et al. Effect of atenolol on
mortality and cardiovascular morbidity after noncardiac surgery. N
Engl J Med 1996;335(23):1713–20.
3. Auerbach AD, Goldman L. Beta-blockers and reduction of cardiac
events in noncardiac surgery: scientific review. JAMA 2002;287(11):
1435–44.
4. Bangalore S, Wetterslev J, Pranesh S, et al. Perioperative beta block­
ers in patients having non-cardiac surgery: a meta-analysis. Lancet
2008;372(9654):1962–76.