Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
In contradistinction to a split-thickness or full-thickness skin graft, a flap receives its blood supply independent of the underlying bed on which it is laid. An axial-pattern flap derives its blood supply through an inflow artery and an outflow vein, which is termed the vascular pedicle. When this pedicle of an axial-pattern flap is divided, the flap becomes a free flap. Alternative terms include microsurgical free flap, microvascular flap, and free tissue transfer. A free flap is any segment of skin, fascia, muscle, or bone, or composite combination of some of these tissues, that is completely detached from its donor site and transferred to a distant site. Its blood supply is then immediately restored by microsurgical anastomoses of its inflow artery and outflow vein to recipient blood vessels in the vicinity of the defect. Almost 50 years ago, the first successful free tissue transfers were related to a new understanding of the axial pattern of certain vascular territories. Published review articles have since attested to the rapid and continuing growth in the number and variety of such free tissue transfers, throughout the body and specifically to the upper extremity.
The general indications for flap coverage (see Chapter 44 ) are (1) a defect with exposed joint cartilage, bone devoid of periosteum, or tendon devoid of paratenon that cannot support the neovascularization of a skin graft and (2) a defect at any specific location in which a skin graft might be subject to exposure and repeated ulceration or might impair later reconstruction.
But why use a free flap, which is the most complex technique in the reconstructive hierarchy? Local or regional flaps to the hand have a distinct advantage in that they provide skin of similar color, texture, and thickness and do not require a donor defect elsewhere in the body. However, they are limited in their size and availability. The larger the primary defect created by injury or excision, the less local skin remains to provide flap coverage. Even when the area is large enough, the thickness required to fill a defect is often lacking. Furthermore, raising a local flap inflicts additional injury on the traumatized limb. The resultant compound wound may impair hand function more than the initial defect.
If not local flaps, are there disadvantages to distant pedicle flaps? Distant pedicle flaps (e.g., the groin flap) can provide sufficient tissue for all but the most massive defects. However, circumferential defects are difficult to cover; the hand must be dependent during attachment; positioning may be uncomfortable; lack of exercise may result in more joint stiffness; the flap may be avulsed by a young or an uncooperative patient; and the thickness of the tube presents a later problem of inset and contour. Most importantly, a pedicled flap is no longer supported by its vascular pedicle after division, and the flap becomes something of a parasite, obtaining perfusion from the underlying scarred bed. In contrast, a free flap has its own independent vascular pedicle. Although the contention that a flap brings a new blood supply to relatively avascular tissue is fallacious, certainly at the time the flap is applied, and probably in the long term there is merit in the fact that it will not place additional vascular demands on the area of the defect, further compromising the limb. There is a strong similarity between tissues adjacent to a major wound and those that have been irradiated. The advantage of a “permanent pedicle, blood-carrying flap” in coverage of radiation ulcers was shown many years ago.
Free flaps offer an attractive alternative to the limitations of conventional techniques of skin coverage. They are now available in virtually any size and thickness and may be cut to fit the defect with incomparable precision. They permit elevation of the limb and early mobilization and are relatively unaffected by random movement in the infant or disoriented patient.
Free tissue transfer in its early years was a lengthy, unpredictable undertaking. In one series of over 500 cases performed in Ljubljana between 1976 and 1983, only 12 of the first 100 free flap procedures took less than 6 hours to complete. By contrast, 92 of the last 100 were finished in less than 6 hours, 24 of them in less than 4 hours. The failure rate in the first 100 free flaps was 26% and in the last 100, 4%. Similar success rates have been reported from other centers. Free tissue transfer is now a relatively rapid and highly reliable procedure when performed by an experienced team.
Free flaps are equally successful in both pediatric and elderly patients. The youngest patient on whom the authors have performed a free tissue transfer was 9 months of age; the oldest, 93 years. Both have done well, with no complications attributable to their ages. There were only two failures in a recent series of 433 free tissue transfers in children, a success rate of 99.8%.
Free flaps will not help heal unsuitable wounds. Adequate debridement and preparation are mandatory (see later). In emergency situations, especially in crush injuries, the wound may be so ill defined that radical debridement might needlessly destroy vital structures that otherwise may potentially survive and continue to function. In these circumstances, debridement of unquestionably dead and contaminated tissue should be followed by the application of wet dressings or, perhaps better, copious application of a petrolatum-based ointment such as Neosporin, or application of a vacuum-assisted closure (VAC) dressing. At a second look within 72 hours, it may be possible to create a clean wound by further debridement, and an “early” flap performed. If doubt remains much beyond 72 hours and further serial debridements are necessary, the chance of infection beneath the flap rises 12-fold. Split-thickness skin grafts should therefore be employed wherever possible after the final debridement, and secondary flap reconstruction is left until the wounds have completely healed.
The alternative to this approach provides the rationale for an “emergency free flap.” , The initial injury or subsequent debridement may expose vital tissues that will not support a skin graft: bare bone, cartilage, ligament or tendon, and exposed vessels or nerves. A flap will therefore be required at some stage. Even with the wettest of wet dressings, exposed bone and tendon continue to desiccate and die and eventually have to be excised. It follows, therefore, that with exposure of such important but poorly vascularized structures, emergency or early flap coverage is mandatory to preserve function. Radical debridement with emergency free flap coverage has become accepted for both upper and lower extremity traumatic wounds. Similarly, early free flap coverage of high-voltage electrical burn injuries has become the accepted standard of care.
Although early flap coverage of traumatic wounds of the extremities has been advocated within 72 hours of injury, a recent systematic review of timing of upper extremity free flap reconstruction found no significant association between timing of reconstruction and rates of flap loss, infection, or bony nonunion. However, early reconstruction seemed to decrease the length of hospital stay and limit associated medical costs.
Free tissue transfer should not be used to salvage a limb with insufficient potential for eventual function. Such function can only be attained by an acceptably sensate hand with mobile joints controlled by sufficient musculotendinous units. In complex injuries, the potential salvage of a limb can only be assessed by an experienced hand surgeon, who must weigh the relative benefits of the three options of primary amputation, radical debridement and emergency free flap coverage, or serial debridements and secondary free flap coverage.
The recognition that theoretically any vascular territory or angiosome could be transferred on its axial pedicle led to a renaissance in studies on the vascular anatomy of the skin and underlying muscles and the development of a wide armamentarium of potential free flap donor sites. Only those flaps that have stood the test of time or are currently evolving as potential free flaps for use in the upper extremity are included in Table 45.1 . The classification of axial-pattern flaps into cutaneous, fasciocutaneous, and musculocutaneous flaps has been outlined in Chapter 44 and is reiterated here for convenience.
Classification | Flap | Artery | Composite Tissues | References |
---|---|---|---|---|
Cutaneous | ||||
Dorsalis pedis | Anterior tibial–dorsalis pedis artery | Bone: second metatarsal Tendon: extensor tendons Sensory: superficial peroneal nerve |
McCraw, 1977 ; Ohmori, 1976 ; Robinson, 1976 ; Zuker, 1986 | |
Groin | Superficial circumflex iliac artery Deep circumflex iliac artery |
Bone: iliac crest | Chuang, 1992 ; Harii, 1975 ; Hough, 2004 ; Smith, 1972 | |
Scapular | Transverse branch of circumflex scapular artery | Dabernig, 2007 ; Gilbert, 1982 ; Hamilton, 1982 | ||
Parascapular | Descending branch of circumflex scapular artery | Bone: lateral border of scapula | Burns, 1986 ; Nassif, 1982 ; Swartz, 1986 | |
Anterolateral thigh | Descending branch of lateral circumflex femoral artery | Sensory: lateral cutaneous nerve of thigh | Baek, 1983 ; Javaid, 2003 ; Koshima, 2003 ; Song, 1984 ; Wang, 2005 ; Wei, 2002 | |
First web space/toe pulp | Dorsalis pedis artery/first dorsal metatarsal artery | Sensory: fibular digital nerve to great toe; tibial digital nerve to second toe; deep peroneal nerve | Buncke, 1979 ; Gu, 2014 ; Koshima, 2000 ; Lin, 2007 ; May, 1977 ; Morrison, 1978 | |
Fascial/fasciocutaneous | Radial forearm | Radial artery | Bone: radius Tendon: palmaris longus |
Bardsley, 1990 ; Hentz, 1987 ; Jones, 2008 ; Muhlbauer, 1982 ; Partecke, 1984 ; Song, 1982 ; Timmons, 1986 |
Sensory: medial and lateral antebrachial cutaneous nerves | ||||
Ulnar forearm | Ulnar artery | Sensory: medial and lateral antebrachial cutaneous nerves | Christie, 1994 | |
Posterior interosseous | Posterior interosseous artery | Chen, 1996 ; Shibata, 1997 | ||
Lateral arm | Posterior radial collateral artery | Bone: humerus | Katsaros, 1984 ; Katsaros, 1991 ; Ulusal, 2007 ; Yousif, 1990 | |
Tendon: triceps | ||||
Sensory: posterior cutaneous nerve of arm | ||||
Fibular osteocutaneous | Peroneal artery | Bone: fibula | Wei, 1986 | |
Medial plantar | Medial plantar artery | Morrison, 1983 ; Ninkovic, 1996 | ||
Temporoparietal | Superficial temporal artery | Hing, 1988 ; Hirase, 1994 ; Upton, 1986 | ||
Musculocutaneous | Gracilis | Medial circumflex femoral artery | Jones, 1990 | |
Latissimus dorsi | Thoracodorsal artery | Chen, 2004 ; Godina, 1987 ; Gordon, 1982 ; Lee, 2008 | ||
Rectus abdominis | Deep inferior epigastric artery | Pennington, 1980 ; Press, 1990 ; Rao, 1994 ; Taylor, 1984 | ||
Serratus anterior | Thoracodorsal artery | Bone: rib | Brody, 1990 ; Harii; 1982 | |
Omentum | Right gastroepiploic artery | Iglesias, 2014 ; Seitz, 2009 |
The vessel of an axial pattern flap may be purely cutaneous ( Fig. 45.1A ), proceeding directly to and supplying skin alone. Free flaps based on these cutaneous vessels are termed free cutaneous flaps or free skin flaps. Typical free cutaneous flaps used for reconstruction of the upper extremity include the dorsalis pedis flap, the groin flap, , , and the first web space neurosensory flap. ,
The axial vessel may first supply fascia before reaching its cutaneous territory (see Fig. 45.1B ). The vessels commonly reach the fascia by running in an intermuscular septum. Because the septum is attached to bone, a segment of bone may be harvested as a composite flap with the overlying skin and be vascularized by the same pedicle, for example, the fibular osteocutaneous flap. Axial flaps whose vessels first pass along the fascia are termed fasciocutaneous flaps. Just as with a musculocutaneous flap, the fascia itself may be transferred alone without the overlying skin , but skin cannot be transferred without the underlying fascia.
Typical fasciocutaneous flaps used for coverage of the upper extremity include the radial forearm flap, ulnar forearm flap, and lateral arm flap. Pure fascial flaps include the radial forearm fascial flap, lateral arm fascial flap, scapular fascial flap, dorsal thoracic fascial flap, serratus fascial flap, , and temporoparietal fascial flap. ,
The axial vessel may first supply muscle (see Fig. 45.1C ) with multiple secondary perforators supplying the overlying skin. The muscle must be harvested with the skin to ensure survival of the overlying skin, but the muscle may be transferred itself without the overlying skin and covered with a split-thickness skin graft. Flaps containing muscle and skin are termed musculocutaneous flaps or myocutaneous flaps, whereas transfer of the muscle alone is referred to as a free muscle flap. Useful muscle and musculocutaneous flaps for reconstruction of the upper extremity include the latissimus dorsi, serratus anterior, rectus abdominis, and gracilis muscles. In certain cases, free muscle flaps can be reinnervated by coaptation of the motor nerve to a recipient motor nerve in the forearm to restore otherwise irretrievable motor function.
More recently, perforator flaps have been developed in which a specific muscle perforator is dissected in continuity between the overlying skin and the main axial vessels so that the skin component of a musculocutaneous flap can be transferred without the bulk of the underlying muscle; the base is only the single muscle perforator. There are two options for dissection and elevation of perforator flaps and their anastomoses to recipient vessels. The skin flap can be harvested based on multiple perforators that remain in continuity with the axial vessel in order to provide a long pedicle that can be anastomosed to a recipient vessel of similar caliber. Alternatively, the same skin flap can be harvested based only on a single individual perforator without dissecting the main vessel pedicle more proximally; this obviously simplifies the dissection and shortens the operative time for harvesting the flap. However, anastomoses of these smaller diameter vessels in the shorter single perforator pedicle may require supermicrosurgical techniques. Perforator flaps include the thoracodorsal artery perforator (TDAP) flap, , , , , circumflex scapular artery perforator flap, superficial circumflex iliac artery perforator (SCIP) flap, , and the anterolateral thigh perforator flap.
Vascularized bone grafts and vascularized nerve grafts can also be transferred on their vascular pedicle (see Chapter 46 ). Apart from these “pure” single tissue flaps, composite transfers usually contain two tissue components: skin-bone (osteocutaneous), , skin-nerve (neurosensory), , , , and skin-tendon (tendocutaneous). , These composite flaps have very specific indications: osteocutaneous flaps such as the fibular osteocutaneous flap and the deep circumflex iliac artery osteocutaneous (DCIA) flap , may be indicated for reconstruction of combined bone and soft tissue defects of the upper extremity; the radial forearm flap with the palmaris longus tendon , and the dorsalis pedis flap with vascularized long extensor tendons may occasionally be indicated for reconstruction of a combined skin and extensor tendon defect due to a dorsal degloving injury of the hand; and neurosensory flaps , , may be specifically indicated for restoring sensation to an important tactile area of the hand (the nerve within the flap itself can be used as a vascularized nerve graft and possibly provide more rapid return of sensation in the distal extremity).
Venous flaps and arterialized venous flaps have provoked sporadic interest during the past decade. In a pure venous perfusion flap, a small flap of skin and subcutaneous tissue containing a superficial vein that projects a few millimeters both proximally and distally beyond the flap is harvested from the flexor surface of the distal forearm. The flap is then reversed and can be used for simultaneous coverage of a defect over the dorsal aspect of the fingers or hand as well as for restoration of venous outflow by anastomosis of both the proximal and distal ends of the vein to a vein proximal and distal to the defect.
There are two types of arterialized venous flaps. In the first type, the flap inflow vein is anastomosed to a digital artery and the outflow vein is anastomosed to a vein proximal to the defect, in essence creating an arteriovenous fistula. The second type of arterialized venous flap can be used for simultaneous coverage of a defect over the palmar surface of a digit as well as for restoration of a segmental defect in a digital artery, by anastomosis of both the proximal and distal ends of the vein to the digital artery to create a flow-through flap. Clinically, arterialized venous flaps have better survival rates than pure venous perfusion flaps. The largest successfully surviving venous perfusion flap measured 10 × 8 cm, so it is unlikely that these flaps survive purely as a composite graft. Until the hemodynamics are more fully understood, pure venous flaps should only be considered for coverage of small dorsal or palmar defects of the fingers which require simultaneous restoration of venous outflow or arterial inflow.
Complete examination of any limb on which surgery is being considered is mandatory. Where free tissue transfer is proposed, particular attention should be paid to the wound itself, the bony skeleton, the vascular anatomy, and any structures that may require simultaneous repair or reconstruction.
Preliminary examination of the wound should be directed toward those factors that will influence flap selection: size, depth, and special needs. In the fresh wound, the area may have been reduced somewhat by approximation, but the surgeon should remember two facts. First, one of the major benefits of importing additional tissue is that tension, with all its adverse effects of delayed wound healing, edema, and limitation of early motion, can be eliminated completely. Second, depending on the flap selected, additional area is required to accommodate the bulk of the flap itself. This bulk has its merits because it serves to fill dead space in the wound. In the old wound that is to be excised, the full extent of excision should ideally include all scar tissue, with the incision being made in adjacent normal tissue throughout. If this is impractical (e.g., in extensive healed burns), the excision should include not only ulcerated skin but also all the skin that is immobile due to fixation to deep tissues. It should be recognized that the size of flap required will be considerably larger than the area of skin excised because of scar contracture affecting the surrounding tissues. For this reason, as emphasized below, the final template for the flap should not be made until after wound excision. In those cases in which some scar tissue has to remain on the limb, the surgeon must realize that its lack of elasticity may prevent closure of contiguous incisions, such as that over the pedicle, and make allowances for that fact by planning for “pseudopods” on the flap to close such extensile incisions.
The bony skeleton should be fully evaluated by radiography and, if necessary, computed tomography (CT). Rigid fixation is an essential prerequisite, not only because a rigid skeleton supports the soft tissue reconstruction but also because motion at unstable fractures creates a dead space in which infection can develop. In chronic wounds, the presence and extent of any osteomyelitis can be assessed by four-phase technetium-99m bone scanning and indium-111–labeled leukocytes or magnetic resonance imaging (MRI) with gadolinium enhancement.
The vascular anatomy of the limb should be evaluated by the surgeon by palpation of the brachial, radial, and ulnar pulses and also by using a handheld 8-MHz Doppler pencil probe, making sure to determine the direction of flow and any possible occlusion in the radioulnar arch system by selective occlusion. Blood flow distal to the wound in the recently injured limb should be assessed both before surgery and after fixation of any fractures. If blood flow is deemed to be inadequate, revascularization should be considered, by means of either an interposition vein graft or a flow-through flap, , such as the radial artery of the radial forearm flap. , For old injuries, information regarding the vascular anatomy should be obtained by noninvasive vascular studies and transfemoral angiography. Whereas some concerns have been expressed in the past regarding the possible adverse effect of angiographic dye injection in the days immediately preceding free tissue transfer, the authors have routinely obtained such studies during the 48 hours before surgery without adverse effect. The newer noninvasive techniques of magnetic resonance angiography and computed tomographic angiography may eventually supersede the need for invasive angiography of the recipient limb. Angiography of the flap donor site is not necessary, except possibly for some partial and total toe transfers. Instead the surgeon should examine all potential flap donor sites to ensure that there are no previous scars that might compromise harvesting the flap.
As with all free tissue transfers, the patient should be brought to the operating room well perfused. To this end, the patient should be kept warm during the night preoperatively, and intravenous fluids should be commenced at midnight at a rate that ensures a urinary output of more than 100 mL/hour in the adult patient at the beginning of surgery. The operating room must be kept at a temperature above 70°F. Some surgeons use aspirin or heparin on a routine prophylactic basis, but this is not our custom.
The use of two surgical teams, one to harvest the flap and the other to prepare the recipient site, is common practice and provides the opportunity for the shared experience of two microsurgeons. However, the anecdotal experience of several microsurgeons suggests that little or no time is saved by using two teams and the incidence of complications may be somewhat higher than when one microsurgeon carries the responsibility for the entire procedure.
A wound that is to be covered by a free flap should lie over a stable skeleton. It should not contain any tissue that remains contaminated or has a compromised blood supply. The defect should present as flat a bed as possible for application of the flap. These prerequisites are achieved by rigid fixation and radical debridement ( Fig. 45.2 ). Such debridement may be performed with confidence only in the knowledge that good soft tissue cover is available. Similarities to tumor surgery exist in that radical resection of a malignant tumor with adequate clear margins is best achieved by the surgeon who knows he or she can close the resulting defect. As with cancer ablation, radical debridement demands incision through normal tissue. This should be done under tourniquet control; otherwise, the bleeding from muscle and viable soft tissue will conceal the lack of bleeding from tissues that are not viable. If a major vessel has to be repaired, tissues distal to the arterial injury will be better perfused after the vascular repair, and therefore debridement should be performed after reperfusion.
In severe limb trauma, adequate bone shortening can sometimes facilitate approximation of muscle and soft tissue, as in macroreplantation. In these cases, debridement must precede the arterial anastomosis. Those tissues that remain viable can be demonstrated by cannulating the distal end of the vessel to be repaired and perfusing the ischemic part with heparinized saline. Weeping of this fluid will indicate potential viability; absence of this weeping fluid indicates inevitable ischemia.
When immediate flap coverage is planned (see Fig. 45.24B later), , , contaminated bone can be retained; it is scrubbed vigorously, and burs and a rongeur are used to remove contamination. Even free fragments can be similarly treated and used as bone grafts, but only if they are rigidly fixed and are covered with well-vascularized soft tissue. In late cases, in which osteomyelitis has developed, two criteria are used for assessing bone viability. The paprika sign denotes the punctate bleeding that occurs when cortical bone has been burred back to the level at which there is adequate perfusion. The presence of adherent periosteum is usually indicative of viable bone, and therefore bone should be resected through adjacent adherent periosteum.
Use a tourniquet!
The wound should be debrided similar to “tumor resection” from outside-in.
All foreign material must be removed.
All nonviable tissue must be removed.
Neurovascular structures should be spared.
Deflate the tourniquet and assess the bleeding from skin edges, muscle, and bone.
Remaining muscle should contract with stimulation.
The extent of debridement should not be predicated on size.
A free flap can cover almost any defect in the hand or arm.
All marginally viable skin should be excised with the exception of highly specialized areas such as the fingertip or palm. If marginal tissue is retained, such skin may harbor infection and will certainly heal with profuse scarring. When debridement appears adequate, tissue samples are taken from several areas in the wound for rapid gram-stained smear and quantitative cultures. Such tissue samples are only necessary in open wounds and where there has been previous infection. Samples are taken from tissues that are considered clinically not to be infected at the completion of the debridement and therefore, if positive, are indicative of inadequate debridement. They also identify the probable organism in the event of later infection.
After debridement, the wound may be uneven. Additional excision of vital but nonessential tissue is justified to create a flat wound surface. Close contact between flap and bed is thus assured, and dead spaces are eliminated that could later become the sites of infection or fibrosis. The only exceptions to the rule of radical debridement are longitudinal vital structures that carry the promise of function: intact tendons or nerves and major patent arteries. These should be preserved and cleaned with the help of magnification.
Once debridement is complete, rigid skeletal fixation should be applied. Compression plates are preferred for the radius and ulna, low-profile miniplates for the metacarpals, and type A intraosseous wiring or 90-90 intraosseous wiring for the phalanges. External fixation is employed for segmental bony defects. Small defects can be reconstructed primarily with cancellous bone grafts, whereas large defects can be obliterated with either the muscle component of the free flap or a temporary antibiotic-impregnated methylmethacrylate spacer, pending later definitive bony reconstruction.
After radical debridement of chronic osteomyelitis, secondary cancellous bone grafting should be delayed until 6 weeks after application of the flap, provided that soft tissue healing is uneventful.
The skin margins should be freshened and made even by excising any redundant promontories and peninsulas of no functional importance. A pattern of the defect is taken using any suitable material, such as a piece of an Esmarch bandage that has been moistened with alcohol (see Fig. 45.2C ) or the paper of a glove wrapper.
An additional incision will invariably be required to expose the recipient vessels. In the extremities, primary closure of this incision over the vascular pedicle may potentially compromise blood flow to the flap. If this possibility has not been anticipated and a tongue or “pseudopod” of flap has not been incorporated, it may become necessary to apply a split-thickness skin graft over the vascular pedicle. In situations in which no extension of the flap is planned, a zigzag incision is used to expose the recipient vessels, with a 60-degree angle between each limb. This not only improves exposure but also permits closure of the wound as a series of “V-Y” advancements, thus increasing the circumference of the limb and avoiding any possibility of vascular compromise of the flap.
With the use of loupe magnification, the recipient vessels are dissected in a preliminary manner, meaning that the vessels are exposed but are not prepared definitively for the microvascular anastomoses. This is for two reasons: The precise location of the anastomoses can only be determined with the flap sutured in place, and dissection is best achieved under relatively higher magnification under the operating microscope.
The key to dissection of the recipient vessels is that exploration should proceed from normal tissue toward the wound, not in the more obvious centrifugal fashion, working outward from the wound. This is especially true in reconstruction of chronic wounds because the inflammation and subsequent fibrosis that follows the original injury spreads farthest from the wound along the vascular planes. If the surgeon dissects outward from the wound, stopping when the vessels clearly improve, the point at which one stops will be closer to the wound than if one dissects toward the wound, stopping when the vessels clearly deteriorate. The distance between these two stopping points may be small, but it is highly significant. Because the vessels still contain scar that is scarcely evident to the examiner, it makes them friable and inelastic, which causes problems in performing the anastomoses, and furthermore the vessels may fail to dilate in response to increased flow, which is especially significant in the venous outflow of a free flap.
If prior examination has demonstrated only one artery to the limb is patent, that artery should be inspected with particular care for the presence of (1) large and expendable side branches into which an end-to-end anastomosis could be performed and (2) atherosclerosis. The latter may be seen as irregular, transverse striations on the vessel or felt as unusual hardness in the vessel wall detected with either the gloved finger or a jeweler’s forceps. If atherosclerosis is detected in a single artery extremity in which an end-to-side anastomosis would normally be the technique of choice, the surgeon may elect to perform two end-to-end anastomoses by resecting a segment of the vessel and incorporating a “T”-junction in the flap pedicle (see Fig. 45.23B and C later). Another option is to cut out a small portion of the diseased vessel and place a short segment of vein graft to reconnect the ends. The flap vessel can then be sewn into the nondiseased vein graft.
As a final step, the length of vascular pedicle required should be measured between the point selected for the anastomoses and the closest margin of the defect. This should be done before the precise location of the flap on the donor site has been drawn and dissection of its pedicle completed. Pedicle length can be increased by extending the dissection of the pedicle farther than is customary, often with increasing difficulty (see Fig. 45.6A later). In some flaps, the pedicle can be lengthened relatively, by moving the skin pattern of the flap further away from the vascular pedicle but keeping the majority of the skin paddle within the vascular territory.
If the pedicle of the selected flap is not long enough, three solutions exist. First, choose a different flap. Second, harvest a vein graft. Third, create a temporary arteriovenous fistula using an interposition vein graft between the recipient artery and vein or create a recipient shunt by anastomosing a long length of a recipient vein end-to-side into the recipient artery. However, a recipient shunt should only be considered when the recipient vein is situated away from the recipient artery, continues distally into the limb away from the wound, and remains unscarred and of good caliber. This last solution is attractive not only when the selected flap pedicle is going to be too short but also when the selected recipient artery is relatively deep and inaccessible and when there are two experienced microsurgical teams. When the intended end-to-side arterial anastomosis is deep and somewhat inaccessible, the additional presence of the flap in the way will make it even more difficult to see and do. Use of a temporary arteriovenous or recipient shunt can improve visibility and move the two end-to-end anastomoses of the flap into a more accessible location. When there are two teams, the team dissecting the recipient site will always complete their task before the team raising the flap. The whole operation will be accelerated if the recipient team performs the more time-consuming end-to-side anastomosis of the interposition vein graft or recipient vein to the recipient artery.
With knowledge of the anatomy of the selected flap, a template of the defect can be positioned on the skin in such a way as to optimize the desired thickness of the flap and the necessary pedicle length. It is usually prudent to maximize the length of pedicle because the microsurgical anastomoses can be moved more proximally away from the wound if necessary (see Fig. 45.6A later). A loose pedicle will also facilitate end-to-side anastomoses. Only one flap margin should be incised initially, with an extension over the presumed course of the pedicle. The vascular pedicle can be visualized and dissected through this incision. In those rare cases in which the pedicle is found to be anomalous or absent, this initial incision can be closed without compromising the circulation to the skin.
Between the subcutaneous tissue, fascia, and muscle lie planes that are relatively avascular. It is in these planes that the flap should be elevated. For example, in harvesting a fascial flap, the surgeon must first incise the subdermal and subcutaneous plexus and then the fascia and elevate the fascial flap in the plane beneath the fascia but superficial to the muscle. To be one plane too superficial would result in division of the vessel supplying the flap. To be one plane too deep would result in a time-consuming, often bloody dissection.
Application of tension plays a major role in flap dissection. Once the marginal incision has been carried down to the correct plane for dissection, skin hooks should be applied to the corners of the flap. Firm upward traction away from the bed will display the plane for dissection. Often this plane only contains loose areolar tissue that can be stroked away with a knife. Tension on the flap lifts the pedicle off the floor of the secondary defect, on which the knife is dissecting, into the roof containing the vascular plexus. Care must be taken not to create a “cave” beneath the flap, that is, a central recess between areas of unincised skin, muscle, or fascia. When closed scissors are introduced into the tissues and then opened under gentle pressure to clear planes, they can be both precise and innocuous. They create and enter “caves,” defining structures to be preserved and those to be divided. When scissors are introduced open and closed to cut, they are less precise. It is therefore important when using scissors that the surgeon see both aspects of the structure to be divided so as to be confident about which tissues have been included between the jaws of the scissors.
Unrecognized damage to the vascular pedicle during dissection is one of the major causes of postoperative complications after free tissue transfer, often resulting in poor perfusion to the flap or culminating in thrombosis of an anastomosis. Any problem in the entire vascular system affecting either arterial inflow or venous outflow may eventually result in anastomotic thrombosis. Meticulous dissection of the pedicle therefore plays a paramount role in achieving a successful outcome. Hemostasis must be perfect during dissection of any flap because it is hazardous to achieve once the anastomoses have been completed. Although it has been shown that metal clips are not secure and that bipolar coagulation may harm the pedicle, both microclips and bipolar coagulation are routinely employed in dissecting the pedicle. The clips must be applied with sufficient force to ensure that they will not slip. If bipolar coagulation is being used, the pedicle should be insulated by using smooth forceps to hold the side branch between the pedicle and the point of coagulation.
After completion of the dissection of the pedicle, the flap should be allowed to perfuse after bathing the pedicle with either 2% lidocaine (Xylocaine) or papaverine, before dividing the pedicle and transferring the flap to the recipient site.
After the flap is placed into the defect, check to see that the end of the pedicle comes to lie approximately at the predicted site of the anastomoses. The flap can either be sutured into place definitively or be temporarily stapled into position. With definitive insetting, the completed anastomoses will be less likely to be at risk during final wound closure. Furthermore, the edema that will inevitably occur in the flap after reperfusion will be controlled and should not produce such swelling as to make complete closure of the flap impossible.
Once the flap is in place, the microscope is introduced, the pedicle is approximated to the recipient vessels, and the site of the anastomoses is selected. If the venous repair lies at some distance from the arterial repair, this is the stage when the surgeon can most safely separate the vein and artery within the pedicle. The recipient vessels should be dissected circumferentially over a sufficient distance to allow easy rotation to bring their back walls forward. When an end-to-end anastomosis is to be performed, the recipient vessel should only be dissected over a length necessary to accommodate a microsurgical double approximator clamp. The techniques of microvascular anastomosis are described in the classic chapter Principles of Microvascular Surgery, available on ExpertConsult.com .
Both the arterial and venous anastomoses should be performed before removing the clamps from either vessel. The more difficult anastomosis is performed first. If a monitor such as a photoplethysmograph, laser Doppler, or tissue oximeter is to be used, the sterile probe should be sutured to the flap and connected before the microsurgical clamps are removed (see Fig. 45.18D later). In this way, reperfusion of the flap can be charted, as can any changes that occur during wound closure. If there is any persistent accumulation of blood in the wound that cannot be controlled, usually from scar or bone, a drain or drains should be inserted. Suction can be applied but only very carefully, because however far from the pedicle the drain lies, the two may come together with disastrous consequences.
No dressings should be applied to the flap surface; it is best for the flap to be easily observable at all times. The incisions can be covered with Neosporin ointment and a thin strip of nonadherent gauze. To protect the limb during sleep, a splint should be applied, leaving a generous window for inspection. A surgical glove is turned inside out to close the fingers and is packed with gauze sponges. The glove is laid over the flap, cast padding is applied to the whole limb, including the glove, and a plaster of Paris splint is made. Once the cast is dry, the prominence created by the packed glove is removed with a saw, similar to opening a boiled egg, and the glove is removed so that the flap is protected yet is always visible.
Postoperative care should be directed toward the comfort of both patient and flap and should be the responsibility of a team of trained, trusted nurses. It is important that the patient be up and about as soon as possible, preferably starting on the first postoperative day. All joints not included in the cast should be mobilized.
Several different anticoagulants such as intravenous dextran 40, intravenous heparin, subcutaneous low-dose heparin, and oral low-dose aspirin are used prophylactically by many surgeons to prevent thrombosis of the microsurgical anastomoses. There are no prospective controlled trials to confirm the efficacy of any of these anticoagulant regimens, and occasionally their disadvantages outweigh the theoretical benefits. One of the authors (NFJ) uses a 40-mL bolus of dextran 40 just before release of the microsurgical clamps, followed by a continuous intravenous infusion of 25 mL of dextran 40 per hour for 5 days (based on an adult patient who weighs 70 kg) together with aspirin, 81 mg orally daily. The other author (GDL) does not use any anticoagulant therapy after routine free tissue transfer.
Routine wound cleansing should be accomplished using hydrogen peroxide, followed by a thin smear of antibiotic ointment. This routine has many benefits: It eliminates a nutritious inoculum for bacteria; it ensures that color assessment is neither obscured by blood staining nor impaired by vivid contrast within the observer’s visual field; it permits assessment of the rate of bleeding from the wound, in itself a valuable index of venous insufficiency; and it has an inestimable effect on the morale of the patient who can see the healing wound.
The nurse should observe the flap regularly, usually every hour for the first 48 hours, to ensure its continued viability. Clinical evaluation of the flap is based on several criteria.
Although difficult to learn and impossible to convey adequately, color assessment still remains the mainstay of clinical evaluation. The lighting must obviously be good. The observer’s color perception can be neutralized by first looking at an area of normal skin for 10 seconds and then looking at the flap and comparing the colors.
A healthy free flap, especially a groin flap, is pale to a degree that makes the novice nervous. It is not pure white, when one compares it with a sheet of white paper, but rather it is a very pale pink. If arterial inflow is inadequate, a flap will become paler, with a faint blue-gray tinge. If venous drainage is compromised, the flap will at first become an angry red color and then progressively purple-red and purple-blue. The poorly perfused free flap at first assumes a waxy pallor that is white-tinged with yellow or brown, which differs only slightly from its healthy pink pallor and may be so subtle that only the experienced observer considers it pessimistic. The margins of the failing flap then become indurated and may blister, but doubts concerning the extent of loss may persist for several days.
Capillary refill may be assessed after blanching by fingertip pressure or by running a blunt point across the flap. If the refill after blanching is very slow and the flap is pale, arterial insufficiency is the likely cause. If capillary refill is too swift and the flap has a bluish hue, venous outflow is impaired.
Although the color and capillary refill of a flap can be assessed on a regular protocol, bleeding from the flap should only be evaluated if there is any suspicion of impending compromised perfusion to the flap. The flap and adjacent tissue can be punctured with an 18-gauge needle or a No. 11 scalpel blade and the resultant bleeding compared. A healthy flap should produce bright red blood of the same color as the control and should bleed just a little longer. If no bleeding occurs or if the flap only bleeds very briefly, arterial inflow is inadequate. If the blood is dark red or purple and bleeding persists much longer, venous congestion should be suspected. Persistent dark red bleeding from around the margins of a swollen flap is almost pathognomonic of venous occlusion.
Several techniques for postoperative monitoring of free tissue transfers have been developed and evaluated in experimental free flap models. Several of these techniques have been used in large clinical series. There have been no prospective clinical trials of any monitoring technique, and there is still no consensus as to which will eventually become the standard accepted method. , ,
An ideal monitoring technique should provide a continuous recording of flap perfusion, or flap metabolism, with immediate detection of arterial or venous occlusion. It is important that the criteria indicative of arterial or venous occlusion be easily interpreted by nursing personnel or junior medical staff. Finally, an ideal system should allow monitoring of both visible and “buried” types of free tissue transfers. , ,
Several techniques have been described or are currently being used:
Photoplethysmography
Differential surface temperature monitoring
Doppler surface monitoring ,
Laser Doppler monitoring
Implantable Doppler probes ,
Implantable pH electrodes
Implantable oxygen electrodes
Pulse oximetry
Tissue oximetry ,
Photoplethysmography has been shown to indicate early signs of compromised flap perfusion, owing to excessive tension during skin closure. Photoplethysmographic waveforms can even be transmitted by telephone to a remote monitoring station to allow the surgeon at home to interpret any abnormal waveforms so that the nursing personnel can be advised whether the waveform is normal or whether the flap should undergo reexploration. Tissue oximetry based on near-infrared spectroscopy is showing promise in providing continuous postoperative monitoring of tissue oxygen saturation in free flaps.
A surface temperature probe can be used to compare the relative difference in temperature between the flap and a second control temperature probe on adjacent normal skin. Differential surface temperature monitoring has been evaluated retrospectively in a large series of 600 free tissue transfers. Differential surface temperature measurements greater than 1.8°C at two time points has been shown to be diagnostic of arterial or venous occlusion.
The characteristic arterial pulsation of either a free skin flap or a free muscle flap resurfaced with a split-thickness skin graft can be heard with a handheld pencil 8-MHz Doppler probe. The best arterial Doppler signal on the surface of the flap is marked by a suture and the arterial signal monitored both during closure and every hour postoperatively for the first 48 hours. Augmentation of the venous “hum” can also be produced by compression of the flap tissue, thereby confirming the patency of venous outflow. The laser Doppler flowmeter has been used for postoperative monitoring of free skin flaps and musculocutaneous flaps and free muscle flaps resurfaced with split-thickness skin grafts.
An implantable 20-MHz ultrasonic Doppler probe can be used for continuous monitoring of the patency of the arterial anastomosis and will provide instantaneous detection of an arterial occlusion. , However, this technology also requires further refinement, both with regard to positioning the probe either on the artery or the vein and recognition of a venous occlusion.
Paradoxically, because of the ever-increasing success rates of free tissue transfers, it will become more and more difficult to document statistically the value of any postoperative monitoring technique. One of the authors (NFJ) uses Doppler surface monitoring of the arterial pulsation with a handheld Doppler probe every hour for postoperative monitoring of free skin, muscle, and osteocutaneous flaps and a pulse oximeter for postoperative monitoring of toe-to-hand transfers. An audible alarm can be set to detect the lowest acceptable level of oxygen saturation or the loss of the arterial pulse rate, and this obviously reduces the frequency of nursing vigilance. Loss of the arterial pulse is indicative of arterial occlusion in a toe transfer, whereas an oxygen saturation level falling below 90% is suggestive of venous occlusion.
What actions can be taken to salvage a failing free flap?
The patient’s general condition (e.g., hypotension, cardiopulmonary impairment, or a low circulating volume, manifested by a low urinary output) may all adversely affect perfusion of the flap. A cold vasoconstricted extremity may be due to circumstances as grave as advanced shock or as simple as a low room temperature. Both should be evaluated and corrected.
Look for any tight dressings or sutures. Remove the splint and all the dressings down to the flap. Cut any tight sutures.
Reposition the patient’s arm; in particular, determine whether elevation or dependency of the arm improves flap perfusion. If it does, make the patient comfortable in this new position.
Look for signs of hematoma and evacuate the hematoma. If the flap is swollen or edematous and discolored and there is persistent serosanguineous fluid oozing from the flap margins, a hematoma should be strongly suspected. A flap hematoma may cause loss of a flap; its evacuation may salvage the flap. A few sutures along the flap margin opposite the entrance of the vascular pedicle into the flap should be cut immediately. This may allow some dissipation of the pressure induced by the hematoma, but formal evacuation of the hematoma should be done in the operating room as soon as possible after the hematoma is detected.
If none of these maneuvers rapidly improves the appearance of the flap, the patient should be returned to the operating room immediately for exploration of the pedicle; if this is done quickly, it can be very successful. The situation is almost always worse than it appears. Chen and colleagues reported a 9.9% reexploration rate in 1142 free flaps and an overall salvage rate of 63.7%. Of the failing flaps, 96% presented within 72 hours and 85% of them could be salvaged.
Initially, only the incision that provides access to the pedicle is reopened. If all the sutures around the flap are removed, the edematous flap will be very difficult to control during inspection of the pedicle and it will be impossible to replace the flap back into the primary defect with the precision achieved at the first surgery.
Any hematoma should be gently removed with warm irrigation and the gloved finger. Suction should only be used with extreme care.
The pedicle and recipient vessels proximal to the anastomoses should be examined for obvious kinks, occlusions, or thrombosis. The point of occlusion in a vein will usually be obvious, the upstream segment being turgid and the downstream segment collapsed. Arterial occlusion is usually less evident and can be detected by lack of a pulse on palpation or with a sterile Doppler probe. Testing of arterial flow by the “milking” or Acland test should not be done until a microsurgical clamp has been placed across the artery downstream distal to a side branch, through which emboli, which would otherwise wash into the flap microcirculation, can be safely diverted. After placement of the clamp, the side branch can be opened to determine the presence or absence of flow across the anastomosis.
Do not hesitate to open an anastomosis between clamps to assess flow and if necessary redo the anastomosis. If a clot is present in a patient with a normal clotting mechanism, remember that blood flow only stops if arterial inflow is poor, if “runoff” through the flap is impeded, or if the anastomosis is technically imperfect. If the anastomosis looks good, inspect the downstream segment for an unrecognized injury to the pedicle. If none is found and inflow through the recipient artery is strong, excise the anastomosis and do it again. There is usually enough redundancy in the recipient artery and arterial pedicle to allow a second end-to-end anastomosis. If not, a short interposition vein graft may be necessary. Revision of an end-to-side anastomosis is more difficult unless a “T”-graft , has been used or can be harvested. With a “T”-graft, three cuts and three anastomoses can save the situation.
Do not hesitate to excise a redundant segment and perform another anastomosis.
The patient should be kept in the operating room for a period of time after reexploration and revision of the anastomoses to ensure that further thrombosis does not occur.
There are as many publications in the literature concerning experimental methods of enhancing flap survival as there are of monitoring, possibly more, but there are no prospective clinical trials documenting the efficacy of any pharmacologic agent to enhance flap survival. However, medicinal leeches and various thrombolytic agents may play an occasional role in the salvage of a failing free flap, either as an adjunct intraoperatively, during reexploration of the microsurgical anastomoses, or if the patient or the flap is not deemed suitable for surgical reexploration.
It is well established that leeches can be used to salvage a failing replanted digit if venous outflow is compromised or if a suitable vein is not available for anastomosis. In an extension of this technique, leeches have been used to relieve venous congestion in pedicled flaps and free flaps. The success of flap salvage after leech therapy has been reported to be as high as 60% to 70%. Just as in the replant situation, there is one report of successful salvage of a free flap performed with only an arterial anastomosis in which leeches provided venous drainage of the flap for 8 days until neovascularization was reestablished.
Hirudo medicinalis may ingest an amount of blood almost 10 times its own weight; however, the therapeutic effect of leeches is not due to the volume of blood ingested but due to continued bleeding from the bite after detachment of the leech. Several pharmacologically active substances are secreted by the leech, including hyaluronidase, an antihistamine, a vasodilator, and hirudin, which is the most potent natural anticoagulant known. Hirudin is a 65-amino acid polypeptide with a molecular weight of 9000 daltons, and it inhibits the conversion of fibrin to fibrinogen.
The most significant risk of leech therapy is infection with Aeromonas hydrophila, which may cause a major necrotizing soft tissue or muscle infection with an estimated incidence between 7% and 20%. Consequently, it may be prudent not to use leeches for a failing free muscle flap. A. hydrophila infections may be prevented or treated secondarily with a second- or third-generation cephalosporin or an aminoglycoside antibiotic. The second major risk of prolonged leech therapy is blood loss and the eventual need for blood transfusion. Leeches must also be prevented from “wandering” or attaching to normal better perfused tissues.
Several thrombolytic agents have been used for the treatment of various arterial and venous thromboses, including myocardial infarction, pulmonary embolism, deep venous thrombosis, and thrombotic-embolic problems in the upper and lower extremities. However, there are only a few reports of thrombolytic therapy for salvage of a failing free flap.
Heparin may prevent anastomotic thrombosis by blocking the conversion of prothrombin to thrombin and by decreasing platelet adhesiveness. Systemic intravenous infusion of heparin is sometimes used by some hand surgeons to prevent thrombosis of the arterial and venous anastomoses in replanted or revascularized digits, especially when these injuries have been associated with a crushing mechanism or after multiple revisions of the anastomoses. However, systemic heparin may be complicated by bleeding and the development of hematomas. Local infusion of heparin through a tiny silicone catheter inserted into one of the venae comitantes of a free gracilis muscle flap was successful in preventing thrombosis of a revised end-to-side anastomosis of the other vena comitans. Heparin was infused at a rate of 250 U/hour for 6 days, and the flap was successfully salvaged after revision of the venous anastomosis was performed without inducing systemic anticoagulation.
Streptokinase is an enzyme isolated from group C β-hemolytic streptococci that converts plasminogen into the enzyme plasmin, which in turn sequentially degrades fibrin in a thrombus. Successful salvage of a failing free flap due to arterial or venous thrombosis has been reported, either by infusion of streptokinase into the totally isolated free flap or by selective arterial infusion of streptokinase into the flap by direct arterial puncture proximal to the arterial anastomosis using a fine 30-gauge needle. By avoiding return of streptokinase to the systemic circulation, very high doses of intraarterial streptokinase infusion can be used, up to 60,000 U in 1 hour.
Urokinase is a naturally occurring enzyme derived from human kidney cells and probably has significant advantages over streptokinase. Unlike streptokinase, urokinase does not cause allergic reactions and can be used at very high concentrations with less risk of hemorrhage and without inactivating plasminogen. Serletti and associates reported the successful salvage of five free flaps that were failing due to thrombosis of the venous anastomosis by infusion of 250,000 U of urokinase over 30 minutes through a 25-gauge catheter inserted into the recipient artery proximal to the arterial anastomosis. One of the authors (NFJ) has successfully salvaged two radial forearm flaps and one toe transfer that were failing due to venous thrombosis or combined arterial and venous thromboses by expedient reexploration and intraarterial infusion of urokinase into the recipient artery proximal to the arterial anastomosis.
There are two case reports of the use of an intravenous infusion of tissue plasminogen activator (t-PA) in conjunction with an intravenous infusion of heparin to successfully salvage a free flap compromised by a presumed arterial thrombosis. The optimal dose of t-PA to achieve therapeutic thrombolysis has not been determined. Intravenous infusion of t-PA was begun at 12 mg/hour but was discontinued after 15 minutes because of hemorrhage from the flap donor site.
In summary, reexploration of the microsurgical anastomoses is the best solution to salvage a failing free flap, but leeches and the newer thrombolytic agents may occasionally be indicated either before reexploration or in conjunction with surgical reexploration.
Become a Clinical Tree membership for Full access and enjoy Unlimited articles
If you are a member. Log in here