The first hand transplants generated much discussion not only among hand surgeons (1,2) and transplant surgeons, but also the public and became even more controversial after removal of the very first transplant. The reason for this is simply that a hand transplant is not a life-saving procedure, and its benefits must be carefully weighed against the potential risks of long-term immunosuppression. Encouraged by more recent results with experimental limb transplantation in large animal models and the two clinical cases performed in Lyon (3) and Louisville (4), we decided to perform a double-hand transplant in a man who had lost his hands during an attempt to deactivate a bomb and who already requested this transplant shortly after the accident in 1994. The transplant was eventually performed on March 7, 2000, after almost 1 year of waiting for a suitable donor. We here report on this case with specific emphasis on immunosuppression and its side effects as well as on functional outcome 1 year after transplantation.
Selection and Evaluation of Recipient
Both hands of the 47-year-old policeman were so severely traumatized when the bomb he was trying to defuse exploded that they had to be amputated at the wrist. Both eyes were injured, with a significant decrease in visual acuity of the left eye. Myoelectric prostheses enabled him to perform deskwork at a police station. Otherwise, the patient was in excellent condition with completely normal function of all vital organs and no signs of infection. After several discussions of every potential problem that might be associated with the procedure and, in particular, with pharmacologic immunosuppression and after having met the first patient in Lyon, our patient still expressed his ardent desire for a hand transplant. He was admitted in February 1999 for pretransplant work-up: his medical history was obtained and a physical examination performed, as were routine laboratory tests. He was negative for cytomegalovirus (CMV) and Epstein-Barr virus. His blood group was B, Rh positive, his HLA-type was A1, A2, B14, B35, DR: 11(5), 10, DR52, and he had 5% panel-reactive HLA antibodies. The outcome of magnetic resonance imaging, angiography, electromyography, testing of nerve conduction velocity, and thorough psychological testing made him a good candidate for a hand transplant. At the end of April 1999 he was finally put on our waiting list and registered with Eurotransplant.
Selection of Donor and Removal of Grafts
The donor was matched for gender, age, size, and blood group, but fully mismatched for HLA. Lymphocytotoxic cross-match, however, was negative. He was positive for CMV but negative for Epstein-Barr virus. Brain death had occurred owing to a cerebrovascular accident. Despite presumed-consent legislation in Austria, consent for hand donation was requested of the family. To keep cold ischemia as short as possible, donor and recipient operations were performed simultaneously in adjacent theaters. While one team dissected under tourniquet control all anatomic structures that were to be reconstructed, the other team freed tendons, nerves, and vascular structures in the recipient. When both teams had completed the dissection, the tourniquet in the donor was released and hemostasis carefully achieved. The forearm was then perfused with 500 ml of chilled University of Wisconsin solution through the brachial artery. Next, all structures were transected and both bones osteotomized at the mid-forearm to have enough length of all structures to be joined with the recipient. After revascularization of the right hand, the same procedure was applied to the left forearm. After wound closure the recipient’s cosmetic prostheses were fitted to the donor.
Donor and recipient operations were performed by members of the Department of Plastic and Reconstructive Surgery (H. Piza, H. Hussl, M. Ninkovic) and hand surgeons of the Department of Trauma Surgery (S. Pechlaner, M. Gabl). After trimming and fixation of both bones with seven-hole metal plates, the radial and ulnar artery were anastomosed with interrupted 8-0 Prolene sutures. After completing the cephalic and basilic vein anastomoses, the right graft was reperfused after a total cold ischemia time of 150 min. Next, all hand and finger flexors and extensors were repaired. The ulnar, median, and the superficial sensory branch of the radial nerve were then anastomosed with 9-0 nylon 12 cm above the wrist. The skin was closed, and an autologous split-thickness skin graft was used to cover a small defect at the palmar aspect of the left forearm. Essentially, the same procedure was thereafter performed on the left side. Because several muscles of the left forearm were defective, some tendons had to be repaired en masse or by transpositioning (Fig. 1). Total ischemia time for the left hand was 170 min. After dressing, both arms were placed on long-arm splints.
The patient was cared for at the Department of Surgery, Division of Transplant Surgery. An oxygen probe was used for the first week for graft monitoring, and a heparin drip was commenced shortly after surgery to maintain prothrombin time at approximately 50 sec; it was replaced after 1 week with oral acetylsalicylic acid (100 mg/d), which has been maintained until now. Skin biopsies were performed at regular intervals and whenever clinically indicated.
Immunosuppressive Therapy and Infection Prophylaxis
Induction therapy with antithymocyte globulin (ATG; Fresenius Medical Care, Bad Homburg, Germany) at a dosage of 2.5 mg/kg was begun during surgery and continued until day 3. Bronchospasm and hypotension occurred soon after ATG was started and were managed promptly. Before revascularization of the first graft the patient received 500 mg of methylprednisolone i.v., 250 mg on day 1 and 125 mg on day 2. Thereafter, steroids were switched to oral prednisolone, which was rapidly tapered to 25 mg on day 8. Prednisolone was further gradually reduced to 15 mg by day 15 and 10 mg by day 200 and was subsequently reduced to 7.5 mg after the first postoperative year. Tacrolimus (Fujisawa, Munich, Germany) was commenced at 0.20 mg/kg body weight in two daily oral doses and then adjusted to maintain trough levels, as measured with Microparticle Enzyme Immunoassay (Abbott Laboratories, Abbott Park, IL), of 15 ng/ml during the first postoperative month, 12 mg/ml between 2 and 6 months, and 10 mg/ml thereafter. In addition, he was given 2×1 g mycophenolate mofetil (Roche, Basel, Switzerland).
Because bone marrow may be transplanted together with these composite-tissue grafts, recipient blood was tested for donor cells using the gene print STR Systems–vWA (Promega Medicine, Madison, WI) at 12 months.
Inasmuch as the donor was positive and the recipient negative for CMV, ganciclovir prophylaxis was started intravenously on day 1 at 2×200 mg per day for 6 days and then switched to 3×1 g orally. For prophylaxis against Pneumocystis carinii infection co-trimoxazole was given orally at a dosage of 960 mg three times per week for 1 year.
A special rehabilitation program was designed based on early protective joint motion (EPM) and cognitive therapy after Perfetti (5). Passive EPM was begun on day 3, and shortly thereafter cognitive exercises including sensory reeducation and cortical reintegration were started. Nerve regeneration was regularly assessed by the presence of Tinel’s sign. Assisted active exercises and electrical stimulation of the thenar and hypothenar muscle groups were added 3 weeks after transplantation for the right hand and, because of the more complex muscle reconstruction of the left hand, with a delay of 1 week for the left. Active finger movements and occupational therapy were initiated 1 week later. The patient was provided with thermoplastic splints and trained in all basic and profession–related activities of daily life. At postoperative week 9, electromyographic biofeedback training was introduced. The whole rehabilitation program was performed for 6 hours a day, 5 days a week, for 12 months, with cognitive therapy accounting for more than 80%.
Overall graft function with regard to motility was determined by comparing active movements of every joint to the normal range of motion (6).
Except for the skin graft on the left forearm, which became necrotic and required a 3- to 4-cm autologous skin graft, wound and bone healing was uneventful. At 7 months bone healing was complete as shown by x-ray and bone scanning. Six months after transplantation the patient developed swelling of the left forearm and hand. Multiple arteriovenous fistulas were identified as being the cause and were ligated on November 3.
Fifty-five days after transplantation the patient began to develop disseminated erythema, which soon affected major parts of both grafts but not the dermal autograft (Fig. 2). Although histology demonstrated superficial and deep perivascular mononuclear cell infiltrates, skin lesions were initially misinterpreted as contact dermatitis. We soon realized that in this setting histologic changes were consistent with acute rejection (Fig. 3).
Skin lesions disappeared after treatment with 750 mg, 500 mg, and 500 mg methylprednisolone on three consecutive days and topical therapy with tacrolimus and steroids. On day 188 a very few red spots suspicious of incipient rejection responded immediately to local tacrolimus therapy. Since then no further signs of rejection were observed.
Interestingly, 1 year after transplantation no donor cells were detectable in the recipient’s blood.
No infectious complications whatsoever occurred throughout the entire observation period. Despite prophylaxis and proven ganciclovir sensitivity, virus replication was demonstrated on day 34 by pp-65 antigenemia assay and CMV DNA hybrid capture assay. The asymptomatic patient was therefore switched to Foscavir (100 mg i.v. for 8 days). After he became negative for CMV, he was put back on oral ganciclovir. Virus replication recurred on day 78. The patient remained negative after three cycles of cidofovir (435 mg/14 days i.v.) and subsequent oral ganciclovir administered until now.
Nerve regeneration had reached the center of the palm (approximately 17 cm from the anastomoses) at 3 months and the finger tips at 6 months. At 1 year the texture of both grafts was that of normal hands. Apart from hair regrowth, sweating became demonstrable. Electromyography demonstrated reinnervation of intrinsic muscles of both hands. Temperature, pain, and pressure sensation were present in both hands and all fingers. Results of the moving two-point discrimination test are shown in Table 1. The Semmes-Weinstein Calibrated Monofilament Test showed normal response to D 2.83 in the right hand and thumb as well as the index finger of the left hand, but a response to filament F 3.61 of the three ulnar fingers (long, ring, and small fingers) as a sign of diminished light touch.
The active range of motion of the forearm, wrist, and fingers without any stabilization is shown in Tables 2–4. As compared with the normal range of motion, it is 61.1% for the right wrist joint and 49.3% for the left wrist joint, 69.9% for the finger joints of the right hand and 59.2% for those of the left hand. Taken all together the average active range of motion in our patient in both hands is 60.1%. He demonstrated a grip strength of 14.0 kg in his right hand and 8.5 kg in his left, a key grip of 1.3 kg right and 1.2 kg left, with a lateral pinch strength of 0.4 kg only in the right hand, as tested with a baseline hand dynamometer. He was able to perform tip pinch and pulp pinch, as well as chuck pinch and key pinch on both sides. The patient is able to turn the pages of a newspaper, write with a pen or pencil, use a computer, dial and use a telephone, eat with a knife and fork, drink from a cup or glass, take care of his personal hygiene, and dress himself except for buttoning a shirt (Figs. 4 and 5). Not only has the patient returned to work, but 15 months after the transplant he even passed the examination for a motorcycle driver’s license. He is in excellent condition with normal renal function (serum creatinine, 0.9 mg%), normal liver function, normal carbohydrate metabolism, and normal blood pressure without medication. He is negative for CMV and Epstein-Barr virus. More importantly, he is extremely pleased with both the functional and cosmetic outcome; he has fully integrated both grafts in his appearance and regards them as his own. The patient continues to have regular physiotherapy and is still making progress in both sensitivity and motility and particularly in movement of intrinsic muscles.
The objective of a hand transplant must be to enable the recipient to feel and perform more functional activities than would be possible with myoelectric prostheses. We are therefore convinced that only the loss of both hands can indicate this kind of therapy. For the time being we would not even accept the loss of the dominant hand as an indication. Furthermore, because after a single-hand transplant the patient will certainly compare the function of the graft with his own hand, the outcome may be regarded as unsatisfactory.
As for the anatomic prerequisites, we agree with the groups in Lyon and Louisville that the level of amputation should be at the wrist or not far above, and the forearm musculature should be intact (1,2). Even more important for recipient selection is his or her psychological stability and ability to understand the complexity of the therapeutic modality as well as to cope with the associated stress. The presence of a sound sociofamilial background is also essential. In addition, he or she must accept the risk of the procedure and, even more so, of chronic immunosuppression.
In contrast to most replants after crush or avulsion injuries, anatomic structures in a hand recipient can be expected to be in good shape, as are those of the graft, which makes the surgical procedure technically less demanding than for a replant. In our patient, reconstruction of all essential structures was possible for the right hand, whereas on the left side some tendons had to be repaired en masse or by transpositioning. Skin and bone healing seemed not to differ much from that of replants. Nerve regeneration, however, was faster than usually seen in replantation. This effect has been ascribed to the use of tacrolimus, which has been shown to trigger the synthesis of axotomy-induced growth-associated protein (GAP-43) (7,8). The short ischemia time may also have contributed to the fast regeneration. Using defined criteria to assess the functional outcome, the overall achieved motor function is 60% of normal, which together with the degree of sensibility without cold intolerance is seen by the patient himself as more than he expected and enables him to pursue activities he could not with his myoelectric prostheses. The envisaged goal has thus been achieved. But what price has the patient paid so far? Apart from surgery and the stress of the perioperative period, it was the burden of an extremely painstaking rehabilitation program. We are convinced, however, that the intensity and duration of the program were crucial for our patient to achieve nearly normal hand function.
Three categories of adverse events are primarily associated with long-term immunosuppression: toxicity of the individual immunosuppressive drugs, opportunistic infections, and malignancy. The first dose reaction to ATG administered during surgery was easily controlled. Nephrotoxicity, neurotoxicity, and diabetogenicity are the most common side effects observed with tacrolimus. Tacrolimus levels were regularly measured and kept below 16 ng/ml during the first month and below 12 ng/ml thereafter. Even the concomitant administration of Foscavir and cidofovir under careful hydration did not cause detectable renal damage. Serum creatinine at 1 year was the same as before transplantation, as was carbohydrate metabolism. Not only were no neurotoxic side effects observed, but even a positive effect was seen with regard to nerve regeneration. Mycophenolate mofetil was well tolerated without any gastrointestinal disturbances. An increase in weight may be caused at least in part by steroids. Despite ganciclovir prophylaxis at the usual dosage we were not able to prevent CMV infection. This suggests that high-risk patients may require the addition of more-potent anti-viral agents to the prophylactic regimen. Because there is increasing evidence that CMV infection may trigger immunologic reactions and therefore have a negative impact on graft survival, we had a major interest in returning the patient to CMV-negative status (9,10). To meet this goal it was necessary to change therapy twice. There were no other infection problems. In particular, the patient remained negative for Epstein-Barr virus, which may be important with regard to the development of lymphoma as it is well known that induction therapy with antilymphocyte antibodies significantly increases the risk of lymphoma (11,12). Overall immunosuppression in our patient, however, was kept at a level comparable to that for recipients of a pancreas or heart transplant. With this type of immunosuppression we were able to prevent rejection, except for one episode that was easy to reverse. Interestingly, CMV replication actually coincided with the first and only rejection episode. Muscle has been shown to elicit a greater cell-mediated immune response than that generated by the other components of a composite-tissue graft (13). Therefore, even though no muscle biopsy was performed we strongly believe on clinical grounds that this single rejection episode was sufficiently treated. Thus, the long-term prognosis for our bilateral hand transplant should be good. As no donor cells were detectable in the patient’s blood at 12 months after transplantation, chimerism is obviously not responsible for this favorable immunologic situation.
As of November 2001 eight unilateral hand transplants (Lyon, France, 1, Louisville, KY, 2, Guangzhou, China, 2, Kuala Lumpur, Malaysia, 1, and Monza, Italy, 2) have been performed, and a total of four bilateral hand transplants (Lyon, France, 1, Innsbruck, Austria, 1, Guangzhou, China, 1, and Harping, China, 1) (14). Remarkably enough, so far only the very first hand transplant had to be removed 28 months after transplantation. This patient, however, was reported to be uncooperative and to have completely stopped taking his immunosuppressive medication, which led to progressive rejection. Histology of the amputated hand showed signs of acute and chronic rejection of the skin but no histologic signs of rejection of the nerves, muscles, tendons, or bone (15). All other grafts are in place and show satisfactory function. A 100% graft survival at 6 months has so far not been achieved in any other field of transplantation and in particular not at the beginning of the program. This success must be ascribed to excellent surgery and careful postoperative management including pharmacologic immunosuppression as well as physical rehabilitation. Because clearly defined variables for the assessment of composite-tissue graft function do not exist and all hand transplants were not evaluated in the same way, it is not easy to compare them with regard to outcome. It would seem, however, that our patient compares very favorably with the others.
In summary, hand transplantation can be performed successfully at this time with an acceptable risk. Careful weighing of the potential benefits against the enormous effort associated with such a complex procedure and, even more so, against the early risk of infection and the late risk of malignancy, shows that this therapy should be reserved for select patients. Moreover, composite-tissue transplantation offers an enormous potential for reconstructive surgery in general.
The authors thank Maria Barbach, Carla Kaiser and Susanne Wopfner (physiotherapy), as well as Dunja Estermann (occupational therapy).
1. Foucher G. Prospects for hand transplantation. Lancet 1999; 353: 1286.
2. Lee WPA, Mathes DW. Hand transplantation: pertinent data and future outlook. J Hand Surg 1999; 24A: 906.
3. Dubernard JM, Owen E, Herzberg G, et al. Human hand allograft: report on first 6 months. Lancet 1999; 353: 1315.
4. Jones JW, Gruber SA, Barker JH, Breidenbach WC. Successful hand transplantation. N Engl J Med 2000; 343: 468.
5. Perfetti C. Der hemiplegische Patient—kognitiv therapeutische Übungen. In: Munich, Pflaum Verlag, 1997; 23.
6. Hunter JM, Mackin EJ, Callahan AD. Rehabilitation of the hand: surgery and therapy. Mosby, 1995.
7. Wang MS, Zeleny-Pooley M, Gold BG. Comparative dose-dependence study of FK506 and cyclosporin A on the rate of axonal regeneration in the rat sciatic nerve. J Pharmacol Exp Ther 1997; 282: 1084.
8. Gold BG, Yew JY, Zeleny-Pooley M. The immunosuppressant FK506 increases GAP-43 mRNA levels in axotomized sensory neurons. Neurosci Lett 1998; 241: 25.
9. Dickenmann MJ, Cathomas G, Steiger J, Mihatsch MJ, Thiel G, Tamm M. Cytomegalovirus infection and graft rejection in renal transplantation. Transplantation 2001; 71: 764.
10. Koskinen PK, Kallio EA, Tikkanen JM, Sihvola RK, Häyry PJ, Lemström KB. Cytomegalovirus infection and cardiac allograft vasculopathy. Transplant Infect Dis 1999; 1: 115.
11. Opelz G, Henderson R. Incidence of non-Hodgkin lymphoma in kidney and heart transplant recipients. Lancet 1993; 342: 1514.
12. Swinnen LJ, Costanzo-Nordin MR, Fisher SG, et al. Increased incidence of lymphoproliferative disorder after immunosuppression with the monoclonal antibody OKT3 in cardiac-transplant recipients. N Engl J Med 1990; 323: 1723.
13. Lee WPA, Yaremchuk MJ, Pan YC. Relative anti-genicity of components of a vascularized limb allograft. Plast Reconstr Surg 1991; 87: 401.
14. Composite tissue allografts: 3rd International Symposium, Lyon, 29–30 November 2001. Dubernard JM, ed. Paris: John Libbey Eurotext.
© 2002 Lippincott Williams & Wilkins, Inc.
15. Petruzzo P, Dubernard JM. Hand transplantation: Lyon experience. In: Composite tissue allografts: 3rd International Symposium, Lyon, 29–30 November 2001. Dubernard JM, ed. Paris: John Libbey Eurotext 2001; 63.