Fig. 7.1
(a) This transhumeral amputation with an associated elbow dislocation occurred as a result of an 11-year-old girl getting her sleeve caught in a manure spreader. (b, c) Despite the segmental injury and barnyard contamination, replantation was attempted as the patient’s age was favorable and the limb arrived to the replant center on ice within 4 h of the injury. (d) A crude x-ray of the contralateral humerus is shown to allow for appreciation of the necessary shortening. While plate fixation was performed for humeral fixation, a spanning external fixator was temporarily used to stabilize the elbow joint. A pedicled latissimus flap was turned down to reconstruct the missing biceps/brachialis and also provide soft tissue coverage of the hardware. Ultimately the patient went on functional elbow range of motion and assist hand function but required a number of debridements for soft tissue infection, even up to 12 months following replantation
As previously mentioned, patient’s presenting with proximal amputations are generally much more physiologically “sick” than those with distal amputations. Full ATLS guidelines must be followed, and other serious and life-threatening injuries must be ruled out or appropriately treated prior beginning replantation. As previously mentioned, blood loss is a significant concern, and it is not uncommon for the amputee to be in the advanced stages of hypovolemic shock [23]. Massive transfusions may be required.
General patient health and comorbidities must therefore be given careful consideration. Pediatric patients, for example, are almost always excellent candidates for major replantation given their low rates of baseline comorbidities that are not significantly altered or made life threatening in the presence of hypovolemic shock as a result of the injury and subsequent surgeries. By comparison, an adult patient with significant comorbidities such as diabetes, coronary or peripheral vascular disease, or renal or hepatic impairment may not have the physiological capacity to endure further blood loss and the multiple secondary surgeries associated with major limb replantation. Generally speaking, elderly patients are poor candidates for major replantation surgeries for all of the reasons listed as well as the poor regenerative potential within the replanted limb.
Age itself is not an absolute determinant of replantation though. Replantation may be considered in a healthy elderly patient with a distal forearm amputation with reasonable expectation for some return of extrinsic finger muscle/tendon function and therefore an assistive hand. As discussed previously, almost any amputation should be considered for replantation in a healthy child [24–28]. Functional outcomes are bolstered by excellent musculoskeletal healing and nerve regeneration capabilities. Epiphyseal growth is generally retained but may be slowed, following replantation unless the amputation or debridement encroaches upon the physis [29–32]. Given that 80 % of humeral growth occurs at the proximal physis and the majority of forearm length occurs through the distal radial and ulnar growth plates, even direct injury to the elbow growth plates will result in limited long-term limb length inequality. In patients less than 10 years of age, excellent neuroregenerative capabilities will usually result in reasonable recovery of sensation. Loss of motion or function in denervated muscle is common. In patients over the age of 10, sensory return is less predictable, and motor function in the hand is generally unexpected. Koul et al. presented the case of a 6-year-old girl who underwent bilateral proximal transhumeral replantations. At 23 months, she had regained sensation in both hands and only experienced denervation [33] of the left intrinsic requiring tendon transfers.
Ischemia Time
Ischemia time is especially critical when considering major limb replantation or revascularization. Of all the tissue found in the upper extremity, skeletal muscle is the least tolerant of ischemia. Experimental models have shown that skeletal muscle will tolerate 4 h of warm ischemia before myocyte death consistently begins [34–36]. At 6 h, near complete myocyte death is almost certain. The ability of muscle to survive to this extent seems to be linked to its ability to initially use glycogen and creatine phosphate as energy sources preferentially to adenosine triphosphate (ATP). However, once ATP depletion is initiated, myonecrosis occurs quickly [37]. Labbe et al. found that after 3, 4, and 5 h of ischemia, myonecrosis was 2, 30, and 90 %, respectively [38]. See Table 7.1 for ischemic tolerance by tissue type.
Table 7.1
Ischemic tolerance times by tissue
Muscle: 4 h |
Nerve: 8 h |
Fat: 13 h |
Skin: 13 h |
Bone: 4 days |
As a general rule, replantation of the forearm or arm should not be performed if the warm ischemia time is greater than 6 h. The extent of muscle necrosis beyond this point will preclude any reasonable expectation of functional recovery and, even more importantly, will expose the patient to the risk of metabolic and infectious complications implicit from the preservation of this nonviable tissue. Having an accurate time cutoff (as determined from the history) is vitally important because the clinical evaluation of ischemic necrosis can be unreliable as there is often no gross evidence of myonecrosis until several hours after myocyte death. Laboratory findings have found that muscle necrosis initially occurs centrally with minimal peripheral involvement or clinical evidence in response to ischemia [38]. If the replantation is performed outside of this 6 h window, the near complete muscle necrosis will become clinically apparent within 24 h of the replantation and becomes an unnecessary major source of infection and/or vascular thrombosis. Infections can be severe and even life threatening resulting in sepsis or gas gangrene.
Determining whether a limb is warm ischemic versus cold ischemic is still a source of variability when deciding to proceed with a major limb replantation. There have been reports of successful major limb replantations beyond 6 h after amputation but only when the amputated limb has been significantly cooled for the majority of the avascular period. Tantry et al. reported on a series of 14 upper extremity replantations (2 wrist level and 12 proximal) that underwent replantation at an average of 12 h. Of these, nine limbs retained viable vascular status. Importantly, all limbs had less than 2 h of warm ischemia [39].
Preservation of Amputated Parts
In addition to constant adequate external cooling as described previously, perfusion of an amputated limb with cold physiologic solution coupled with adequate external cooling likely represents the most effective means to temporarily cool the amputated part both pre- and intraoperatively during the debridement phase. This can be accomplished with injection of cold (4 °C) heparinized Ringer’s lactate or saline solution [18, 40] and external application of cold saline-soaked pads [40]. Perfusion with tissue preservation solutions such as University of Wisconsin (UW) or Euro-Collins (EC) solutions have also been used effectively both in animal experiments and clinically in preservation of amputated parts, although we have not had direct experience with these solutions [41–45]. When used, injection at 4 °C and 120 cm hydrostatic pressure has been found to be safe [42]. Once the part has been injected and cleaned and appropriately tagged, it may be placed in refrigerator at 4 °C or packed appropriately in an ice-filled cooler if additional time is required.
Mechanism/Extent of Injury (Fig. 7.2a–c)
Fig. 7.2
(a) A high humeral complete avulsion-type amputation that was sustained after the 55-year-old farmer’s arm was pulled into a hay bailer. Although the patient arrived between 2 and 3 h after amputation, the decision was not to replant given the patient’s age, purely avulsion mechanism, proximal level of injury, and gross organic contamination. (b, c) Note the distal bone injury with varying levels of injury to the muscle, nerves, vessels, and skin proximally over a segment greater than 20 cm
Given the high-energy level required to produce a major limb amputation, many extremities are so mangled by the injury itself that the possibility of replantation is excluded. The degree of injury to both the amputated part and the proximal stump has been found to be good indicators for immediate- and long-term success of major replantation [48]. However, full assessment of the injury including the status of the distal microvasculature is oftentimes not possible until the operating room. Internal damage to the vasculature, both due to the amputation as well as ischemic damage, may render it impossible to successfully reperfuse the distal segment despite what appears to be an acceptable anastomosis, the “no reflow phenomenon.”
Classically, there are three main mechanisms for amputation: guillotine, crush, and avulsion [4, 49]. Guillotine types are generally thought to be the most amenable to replantation given the typically smaller zone of injury and less extensive debridement required [13, 50, 51]. Most importantly, the nature of the injury does not typically avulse or stretch out the muscle units, nerves, and vessels during the course of the amputation. Unfortunately, most major proximal amputations are more frequently the result of a combination of all types, and often a predominance of the latter two mechanisms is typical. Given the heat involved with the machinery that causes many of these amputations, there is also a variable component of thermal injury. The resultant zone of thermal injury is often underappreciated in the early phase, much like burn surgery. This may result in significant loss of soft tissue coverage over vital structures a few days out from successful revascularization.
Crush type amputation injuries result in large zones of injury and irregular wounds. An avulsion component is sometimes present within the soft tissues as well, and these injuries are sometimes referred to as crush-avulsion injuries. Due to the large area of injured tissue, extensive debridement and significant shortening are generally required. However, when adequate, debridement is performed, and clean margins can be obtained; a crush injury can be converted to a guillotine type and does not necessarily preclude replantation or portend a poor functional outcome [5].
Avulsion injuries typically carry the worst prognosis with regard to replantation [4, 52]. Their poor prognosis is attributable to many factors. Tissues in avulsion injuries are torn as opposed to cut. The individual structures are usually disrupted at different levels, and the zone of injury extends over a long distance. Tendons are often torn from their muscular junction rendering them next to impossible to repair. Chuang et al. proposed a classification for avulsion amputations based not on bony level but rather on neuromuscular level [53] See Table 7.2. Several studies have found this classification a useful predictor of functional level [53–55].
Table 7.2
Chuang classification for avulsion amputations based on neuromuscular level
Type I: Avulsion at or within the musculotendinous aponeurosis, remaining muscle functional and intact |
Type II: Avulsion within the muscle bellies and distal to the neuromuscular junction, proximal muscle remains innervated |
Type III: Avulsion within the muscle bellies and proximal to the neuromuscular junction, entire muscle is denervated |
Type IV: Avulsion through the joint |
What this classification fails to address is that the neurovascular damage is usually extensive, and it is difficult to clinically assess the healthy margins of the damaged nerve and vascular tissues in the acute phase. Consequently, even after shortening the nerves and vessels, damaged tissues can easily be coapted together. This renders repaired vessels prone to thrombosis, and neurologic recovery is often compromised. If the decision is made to proceed with a replant in the setting of an avulsion type of injury, long interposition vascular grafts are typically indicated to bypass damaged vessel segments. When these long vascular grafts are performed, it is quite likely that distal perfusion can be restored, and there is typically initial optimism with the results. However, the critical perfusion of the muscle bellies in the intervening segmental zone that has been literally bypassed can be further compromised by this long vein graft and can result in a large area of secondary muscle necrosis. In the setting of an avulsion-type amputation, especially in a contaminated environment, successful replantation may simply be impossible, and as the level of injury ascends to the proximal forearm and above, it is generally in the patient’s best interest to proceed directly with revision amputation and wound closure.
Level of Amputation
While the mechanism of amputation is the most important consideration when deciding whether to proceed with replantation, the level of amputation actually plays the most important role in determining functional prognosis. Generally, the more distal the level of the amputation, the better the anticipated functional outcomes after replantation [4, 17, 19, 52, 56]. In distal forearm amputations, outcomes are improved due to the ability to repair musculotendon units mostly through tendon, the level of injury occurring distal to the nerve end-plate innervation of the majority of the forearm musculature, a smaller distance required for nerve regeneration, the need to only repair the medial and ulnar nerves, and the injury occurring in an area of usually rapid bone healing [57, 58]. Successful reinnervation may be possible before end-organ atrophy has occurred.
The prognosis for proximal forearm amputations is more guarded primarily due to the loss of nervous innervation to an increasing number of musculotendon units [59]. Amputations through the elbow joint carry an even lower prognosis for return of clinical function to the hand. Elbow flexion and extension can be restored if the joint is stable, but injury to and potential subsequent degeneration of the joint compound the surrounding soft tissue injury. The result is frequently loss of range of motion. While revascularization of transhumeral amputations can be successfully accomplished, they carry an extremely guarded functional prognosis due to poor restoration of nerve function [16, 60]. Although the hand may regain protective sensation [16, 60], it is unusual that significant wrist and finger muscle function will return except in the setting of pediatric replantation [26]. However, if functional elbow flexion and extension can be achieved (with or without functional muscle transfers), there may be some utility if conversion to a below-elbow prosthesis can be achieved [18, 21]. Segmental amputations generally carry a poor prognosis and are relative contraindications to replantation. However, in select patients such as the pediatric population or those with clean, guillotine-type wounds, replantation may be considered [2, 61].
Surgical Treatment
Initial surgical treatment of the major upper limb replantations involves a thorough cleaning and debridement. When the patient arrives in the operating room, the affected arm and at least one leg should be prepped into the operative field. At the same time a critical inspection of the tissue is carried out and a rapid decision made as to whether replantation is a viable option. Meticulous debridement is critical and should be carried out under loupe magnification. If possible, this should be performed under an appropriate tourniquet to minimize blood loss and facilitate the speed of the process. All devitalized, avascular muscle should be removed back to the tendinous junction, and fascial planes between the skin and muscle as well as between major muscle groups should be aggressively inspected for debris. Muscle adjacent to the amputation site is often severely injured and must be trimmed back a minimum of 1–2 cm. Bone shortening of an amputation is inevitable, and even with a guillotine-like wound, a minimum of 3 cm of total shortening is routine.
The importance of the debridement phase cannot be overstated. Major limb amputations are high-energy injuries that commonly result from highly contaminating mechanisms, and farm and industrial accidents are typical. Given the inherent risk of myonecrosis, infection is the most common and potentially devastating complication of major limb replantation. Replantation is aborted if the part cannot be debrided to a clean wound, the extent of damage is too severe, the ischemia time is prolonged, or the amputated part is not perfusable due to microvascular damage [62, 63]. Once the wound is converted to a clean, guillotine-type wound, tagging of neurovascular structures can be performed, and tendon ends can be sutured in preparation for repair. The bone ends are then approximated for appropriate fixation. During this process, it is advisable to perfuse the amputated part with 1 L or more of cold heparinized saline or ringer’s lactate to flush out stationary blood and prime the vascular system. While UW solution can also be used for a complete amputation, it must not be used to perfuse an incomplete amputation as it is highly systemically toxic. UW solution must be completely flushed out of an amputated part prior to venous repair. Throughout the debridement, the intraoperative hypothermia should be maintained with bags of ice placed on the amputated part.
The foundation of major upper limb replantation is early revascularization. For this reason, once the decision to proceed with replantation is made, prompt vascular shunting (even prior to bone fixation) can be performed [64–67]. This can be performed directly from the proximal stump or with the use of the contralateral arm or femoral artery [68, 69]. If the amputation is incomplete and adequate venous drainage exists, then the vascular dilemma is less problematic as arterial shunting is all that is required. In complete amputations or incomplete injuries with inadequate venous return, venous shunting (after a delay of 10–15 min to allow for drainage of toxic metabolites) can be performed to help minimize blood loss [66]. Shunting may be performed with a large intravascular (IV) catheter, ventriculoperitoneal (VP) shunt, or Sundt’s carotid shunt. Our experience is that this approach should be considered if the ischemia time is approaching 6 h for forearm amputations or 4 h for upper arm amputations.
If system limitations or geographic restrictions predict that the transport time of an injury of this type to a replant center will be prolonged (greater than 6 h), initial judicious debridement and vascular shunting may be best performed by the referring institution prior to transport if a vascular surgeon is available [63]. While temporary shunting may be the only way to get a major upper limb amputation to a replantation center within the ischemic window, great care must be taken when transporting after shunting. The shunts must be secured firmly, and the patient should be sent with several units of packed red blood cells for transport as continuous bleeding is expected and life-threatening exsanguination is a real possibility. Other authors have favored perfusing the limb with heparinized arterial blood when delay in transport is expected [18].
Once the amputation has been successfully debrided and the injury is deemed to be replantable, the sequence of repair of a major upper extremity amputation differs from digital- and hand-level replantations. There is no one definitive sequence of events that must be followed, but generally speaking, rapid reestablishment of arterial inflow is paramount. Once a decision is made whether to shunt based upon the guidelines set out above, a general, reasonable order for a replantation protocol is as follows:
1.
Rapid bone shortening and stable fixation
2.
Arterial repair
3.
Venous repair
4.
Muscle/tendon repair
5.
Nerve repair if possible
6.
Skin coverage and/or soft tissue coverage reconstruction
Bone Fixation/Shortening (Fig. 7.3a–e)
Fig. 7.3
(a) This 66-year-old male sustained a distal forearm amputation as a result of a radial arm saw with a resultant sharp mechanism and narrow zone of injury – primarily thermal. (b, c) Skeletal shortening and fixation were accomplished with volar plating of the distal radius and Darrach-type resection of the small portion of distal ulnar head that remained with the distal segment. (d, e) The replantation was successful in the short term, and the radius went on to unite uneventfully. The patient eventually achieved very strong flexion and extension of the fingers with a secondary tenolysis performed 6 months after index procedure. Sensory return to the hand was limited, and intrinsic motor function was absent. An extensor indicis opponensplasty was performed at the time of tenolysis to improve thumb function even though this tendon had been repaired at the time of amputation
Bone should be shortened to remove contamination and comminution as well as allow for direct well-approximated bony apposition that allows for ease of fixation and reliable bone healing. Rapid but stable fixation that allows for early motion is key [70, 71]. We prefer to use plates when possible, and in adults, 3.5 mm dynamic compression plates (DCP) are preferred for the forearm and 4.5 mm DCP plates for the humerus. Pediatric patients will need more size appropriate fixation – and any combination of 2.0–4.5 mm compression plates can be appropriate for either forearm or humerus depending upon the size of the child. If the bone is tightly apposed transversely, then standard compression plating techniques are sufficient, whereas locking screws may be indicated for spanning type constructs if length preservation is required. In forearm amputations, resection of the distal ulna can be considered if the injury is too distal for shortening or too comminuted for repair. Creation of a one-bone forearm should also be considered if bone comminution precludes rapid and stable fixation of both forearm bones. In the very uncommon instance where there is an associated unstable elbow not amenable to internal fixation and ligament repair, a spanning external fixator could be used.
The amount of shortening should be judged critically before final fixation performed, and one should err on the side of slightly too much shortening rather than not enough. Proper shortening enables adequate debridement of the zone of injury and facilitates reapproximation of the musculotendinous units and neurovascular structures and sets up a higher likelihood of primary soft tissue approximation.
Arterial Repair
Arteries must be debrided to healthy ends with no apparent intimal damage via inspection under magnification. This is best judged by testing the adherence of the intima to the muscularis layer. The microscope facilitates examination of the vessels, particularly below the elbow. Tensionless direct repair is preferable but often not possible, and interposition vein grafting should be performed as required. If two surgical teams are present, one team can rapidly harvest the saphenous vein, while the other team stabilizes the bone and freshens the arterial inflow. Saphenous vein harvested from above the knee works well as a size match for grafting of the brachial artery, while vein from the same system below the knee works well for the radial and/or ulnar arteries. Judging the appropriate length of a vein graft to inset into the arm can be tricky, especially when the graft crosses the elbow joint. As the arm and elbow are usually laying in extension during replantation, the arterial repair is usually performed in extension. This can result in the graft being subjected to kinking when the elbow is later flexed and the graft accordions on itself. In forearm amputations, one or both arteries may be repaired depending upon the quality and size of both vessels.
Venous Repair
Veins are repaired after a 10–15-min period of time has passed after arterial repair to allow for physiologic washout of the vascular system. At least two veins should be repaired, and three or more are ideal. Large cutaneous veins are generally easier to dissect and repair than the deep veins, which tend to be harder to dissect as they are thinner walled and more adherent to the surrounding deep muscle due to frequent branches. However, in some crush or avulsion injuries with skin compromise and/or loss, the venae comitantes may be the only repairable veins. Proximal to the wrist, these veins are usually of an adequate diameter to be repaired and do provide adequate drainage despite their smaller stature.
The artery can be unclamped during venous repair. Some surgeons will administer a systemic bolus of heparin at this point to help prevent thrombosis of the arterial repair, typically 2,500–3,000 units, although the necessity of this is unclear, and anastomosis patency seems more related to surgical technique [72, 73]. Additionally, if ischemia time has been prolonged or if extensive muscle mass is present, a bolus of sodium bicarbonate prior to unclamping of the vein and return of acidic byproducts to the system circulation can be considered. This is typically given at a dose of 1 meq/kg body weight (usually 1.5–2 ampules) and is run in slowly over roughly 5 min prior to unclamping of the repaired vein.
Musculotendinous Repair
Repair is significantly easier in distal forearm guillotine-type amputations where the flexor and extensor tendons can be directly repaired. Repair is performed with the surgeons preferred method but should allow for early mobilization; we prefer a six-strand cruciate-type repair.
The flexor pollicis longus tendon is often the most compromised because of its very distal belly. If it is nonviable or not repairable, then transfer of the index profundus at the time of replant should be performed to restore this critical thumb function. The index profundus tendon can be sutured side to side to the middle finger in this scenario.
Amputations through the muscle bellies or at the musculotendinous junction more proximally in the forearm present a greater challenge. In this setting, the tendon is typically woven through any remaining proximal tendon and the muscle belly and secured with nonabsorbable suture. When this is not possible, the epimysium is reapproximated to try and reduce the gap in the muscle fibers, and a layered repair from deep to superficial is sometimes helpful to better align fibers and eliminate dead space that serves as a nidus for hematoma and/or infection. In some amputations where little to no functional return would be expected because of tendon or muscle loss in a given muscle or muscle group, tendon grafts or tendon transfers may be used for primary reconstruction, although one must be certain of the viability and preserved innervations of both the tendon to be transferred as well as the synergistic tendons remaining.
The biceps and triceps are generally easily approximated for amputations at the level of the elbow, but when the injury is more than 4–5 cm above the distal humerus, the tendons may not be repairable. In the higher-energy injuries typical of transhumeral level, significant loss of the muscle can represent both a functional as well as a wound closure problem. Oftentimes, a rotational muscle transfer utilizing the latissimus dorsi provides a healthy wound coverage solution, and the transferred tendon can even be inset and tensioned such that it can be functional to help restore elbow flexion or extension [74–77].
Nerve Repair
If a tensionless primary repair can be performed after bone shortening and trimming of the nerve ends back to what appears to be healthy margins, a primary epineural repair may be performed. After the nerves are trimmed back under the microscope, it is not uncommon that the gap between the nerves is not amenable to direct repair, and nerve grafts are required. Grafting is generally performed primarily with sural nerve grafts if the replantation appears stable and the wounds are able to be reapproximated over the graft. If the replant viability is questionable or the soft tissue coverage is uncertain, grafts can be performed at a later date. Delayed nerve grafting has the advantage of allowing one to determine that the limb can be successfully revascularized, decreasing the morbidity of an unnecessary nerve graft harvest. Additionally, a delay provides a situation wherein a healthy margin of the nerve can be better determined and one can ensure that nerve grafts can be inset against healthy fascicles proximal to the zone of injury. If delayed nerve repair is chosen, nerve ends should be tagged to facilitate identification at a later time.
Skin Coverage
Like muscles, skin edges should be debrided until healthy, bleeding edges are obtained. In extensive crush-avulsion injuries, large deficits may be the result, and in the early hours after the injury, definitive margins may be very difficult to determine. If skin edges cannot be directly apposed, the goal is for early, but not necessarily immediate, coverage. A combination of rotational or transposition skin flaps, pedicled muscle flaps, and skin grafts can be used to close wounds. Wounds should be allowed to drain and initially may either be left open or covered with a negative pressure device as long as neurovascular structures, especially vein grafts, are not exposed. In some instances, grafts may need to be placed or tunneled into extra-anatomic positions to facilitate soft tissue coverage. While, split-thickness skin grafts may be used to cover native vessels or nerve grafts, in no instances should skin graft be placed directly over vein grafts due to the risk of desiccation, thrombosis, or even rupture. Immediate rotational tissue transfer with the latissimus may be useful, and, occasionally, free tissue transfer is required.
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
