Portal de Periódicos da CAPES

Preliminary Results of Bilateral Arm Transplantation

2009; Wolters Kluwer; Volume: 88; Issue: 5 Linguagem: Inglês

10.1097/tp.0b013e3181b3921a

1534-6080

Luis Landín, Pedro C. Cavadas, Peter Nthumba, Javier Maupoey, Francisco Vera‐Sempere,

Reconstructive Surgery and Microvascular Techniques

Amputees who suffer bilateral amputations above the elbow adapt poorly to prostheses and are usually dependent on others for care (1). The reinnervation of forearm allografts has been shown to be feasible (2), and the bilateral transplantation of hands has given encouraging functional results (3). On the basis of these findings, the bilateral transplantation of arms was proposed as an experimental reconstructive procedure in a 29-year-old man who had lost both forearms as a result of an electrical burn injury but was otherwise healthy. Because of the extent of the injury, neural recovery was expected to take at least 24 months before it would permit independence to be regained. We report observations obtained during the first 5 months after the operation. Early withdrawal of steroid therapy was planned given that induction therapy with alemtuzumab can facilitate steroid-free immunosuppression (4). The recipient was right handed and had undergone amputations at the level of the right distal third of the humerus and the left proximal radio-ulnar joint 2 years previously. No functional remnants of the forearm muscles were present. The left arm required a muscle transfer of the latissimus dorsi to allow flexion of the elbow, and this function was restored 6 months before transplantation. Owing to the risk of bone nonunion, instability, and, eventually, the need for arthrolysis, the left elbow was to be discarded. Tinel’s signs for the radial, ulnar, and median nerves, in addition to brachial pulses, were present in the distal portion of the arm. An institutional review board approved the procedure, and the patient signed a written consent form. The donor was a brain dead, 25-year-old man who was matched for weight and height with the recipient (±10%). The blood groups of the donor and recipient were A and AB, respectively, and human leukocyte antigens were fully mismatched. The recipient was positive for cytomegalovirus and Epstein-Barr virus antibodies. The induction protocol consisted of alemtuzumab (30 mg administered before revascularization) and methylprednisolone (500 and 250 mg on the first and second postoperative days, respectively). Subsequently, tacrolimus (initial trough level 15 ng/mL tapered to 10 ng/mL after 3 months) and mycophenolate mofetil (2 g/day tapered to 1.5 g/day after 3 months to minimize the risk of lymphoproliferative disease after transplantation [5]) were given. Follow-up examinations and prophylaxis against opportunistic infections were performed according to published protocols (6–8). Allografts from below the deltoid were recovered before crossclamping under tourniquet ischemia. The superficial veins were ligated and the brachial artery was transected proximal to the branches of the distal half of the brachialis, biceps, and triceps muscles, along with the median, ulnar, and radial nerves, whereas the nerves of the forearm were discarded. A transverse osteotomy was performed on each humerus using an oscillating saw irrigated by saline. Subsequently, the brachial artery was cannulated and each allograft was perfused with 2 L of cold reconstituted Belzer-University of Wisconsin solution until the basilic vein showed a clear return. After the surgical procedure had been completed, prosthetic limbs were applied. The allografts were immersed in the perfusion solution and covered on ice for air transportation. An artery-last sequence of implantation was performed synchronously by two teams to prevent excessive blood loss, allow accurate repair of nonvascular structures, and minimize the duration of ischemia (9). The humeri were synthesized using 4.5-mm locking compression plates (Synthes, Oberdorf, Switzerland). Epineural neurorrhaphies were performed on the median, ulnar, and radial nerves 8 cm above the humero-radial joint (HRJ), where healthy tissues were found. Myotenorrhaphies were performed on the triceps, brachialis, and biceps brachii muscles to join them to their equivalents on the right side and to the triceps and transferred latissimus dorsi muscles on the left side. Arteriovenous long saphenous vein grafts were interposed side-to-end bilaterally from the brachial artery to the basilic vein of the recipient during dissection of the stump. This was necessitated by the presence of heavy scarring. End-to-end anastomosis of the basilic veins and then the brachial arteries of the allografts to the grafts was performed. The implantation procedure was completed 6 hr after recovery of the allografts. Fasciotomies of the forearm and hand were performed to prevent edema of the largetissue mass, and these were healed by second intention. This permitted mobile-wad biopsies of the skin and muscle to be performed 1 week after the transplantation procedure. Rehabilitation of the hands and elbows was initiated after 3 days and 3 weeks, respectively. Five months after the surgery, good psychological adaptation was observed. In both allografts, M3 power flexion and extension of the elbow (10) (with a range of motion of 0°-140°), and progression of the Tinel’s sign to 7 cm below the HRJs were observed. Examination using a needle electrode (Medelec Synergy, VIASYS, Dublin, OH) showed polyphasic voluntary motor unit potentials of 1.8 mV in 18.15 ms and 0.9 mV in 22.40 ms in the flexor carpi radialis and flexor digitorum superficialis muscles, respectively. The left index, long, and ring fingers became M2 active to allow flexion of the metacarpophalangeal joints by 25° on postoperative day 129, whereas the right wrist was M2 active to allow 30° extension and 10° radial deviation by postoperative day 155. Immunoglobulin G antibodies against human leukocyte antigen B17 were observed using flow cytometry (One Lambda, Canoga Park, CA); this antigen was absent from the allografts. Biopsies revealed infiltration of the skin with Banff I CD3+CD20− cells, but the muscle was generally spared. All the biopsy samples stained positive for C4d deposition, but without evidence of injury on histologic examination (Fig. 1). Clinical rejection, infection, metabolic complications, or peripheral blood chimerism (AmpF/STR, Applied Biosystems, Foster City, CA) were not observed.FIGURE 1.: Histologic findings in biopsy specimens obtained 1 week after transplantation. The left column shows skin, the right column shows muscle. (Top, left) Appearance of the skin on staining with hematoxylin-eosin (×200 magnification). The microanatomy of the skin was indistinguishable from that of normal epidermis and papillary dermis. (Middle, left) Staining for CD3+ cells revealed a few lymphocytes in the papillary dermis, with no exocytosis to the epidermis, which confirmed a pathological score of Banff grade I (×200). (Bottom, left) C4d staining of identifiable capillaries (×200). (Top, right) Perimysium and myofibrils are shown, and abundant nuclei of myocytes can be observed in the endomysium that surrounds each myofibril (hematoxylin-eosin; ×200). (Middle, right) Myocyte nuclei should not be mistaken for lymphocytes. Staining for CD3 confirms that only a few lymphocytes are close to a capillary, and the perimysium and endomysium are free of infiltration (×200). (Bottom, right) Deposition of C4d in the perimysial capillaries (×200).This preliminary report demonstrates the feasibility of bilateral transplantation of arm allografts, and their potential reinnervation, at a level considerably more proximal than those reported previously (11). Significant scarring of the arm required muscle transfer and arteriovenous grafts; this is a combination that has never been reported previously in transplantation of the hand. We can confirm the findings of the composite tissue transplant group of Lyon (France), who showed that infiltration of muscle with lymphocytes does not exceed that of the skin (12). The deposition of C4d was not accompanied by injury to the allograft in the absence of donor-specific antibodies. The combination of lymphocyte depletion caused by administration of alemtuzumab and a regimen of early withdrawal of steroids resulted in no signs of clinical rejection during the 5-month follow-up period. This observation provides evidence to support the suggestion that large allografts are less likely to be rejected (13, 14). Although the administration of tacrolimus may have prompted the rapid progression of reinnervation and the incipient recovery of motor function (15), we will consider this transplant to have been successful only if the patient regains the ability to care for himself within the next 2 to 3 years. Luis Landin Pedro C. Cavadas Clinica Cavadas, Plastic and Reconstructive Surgery Division “La Fe” University Hospital Valencia, Spain Peter Nthumba Clinica Cavadas Pedro Cavadas Foundation, “La Fe” University Hospital Valencia, Spain Javier Ibañez Clinica Cavadas, Plastic and Reconstructive Surgery Division “La Fe” University Hospital Valencia, Spain Francisco Vera-Sempere Pathology Division, Hand Transplantation Team “La Fe” University Hospital Valencia, Spain