18-month outcomes of heterologous bilateral hand transplantation in a child: a case report

Articles

18-month outcomes of heterologous bilateral hand transplantation in a child: a case report Sandra Amaral, Sudha Kilaru Kessler, Todd J Levy, William Gaetz, Christine McAndrew, Benjamin Chang, Sonya Lopez, Emily Braham, Deborah Humpl, Michelle Hsia, Kelly A Ferry, Xiaowei Xu, David Elder, Debra Lefkowitz, Chris Feudtner, Stephanie Thibaudeau, Ines C Lin, Stephen J Kovach, Erin S Schwartz, David Bozentka, Robert Carrigan, David Steinberg, Suhail Kanchwala, Dan A Zlotolow, Scott Kozin, Frances E Jensen, Phillip R Bryant, Abraham Shaked, Matthew H Levine, L Scott Levin

Summary

Background Although heterologous vascular composite allotransplantation has become a burgeoning treatment option for adult amputees, there have been no successful cases previously reported in children. Here, we describe the surgical, immunological, and neurorehabilitation details with functional outcomes 18 months after heterologous bilateral hand and forearm transplantation in an 8-year-old child with quadrimembral amputations and a previous kidney transplant. Methods 2 years of extensive preparation by medical and surgical teams preceded the hand–forearm transplantation of this child. The initial immunosuppressive protocol included thymoglobulin, tacrolimus, prednisone, and mycophenolate mofetil. In July, 2015, our vascularised composite allotransplantation team did the first bilateral hand and forearm transplantation in a child, an 8-year-old boy with previous living-related kidney transplantation. The surgery included four teams working simultaneously on the donor and recipient limbs, aided by customised cutting guides that aimed to reduce ischaemia time. Following an extended length of time in hospital, skin biopsies and close monitoring of renal function and drug concentrations occurred weekly for the first 3 months and were slowly tapered to monthly, and then quarterly. Skin biopsies were also done when tissue rejection was suspected. Paediatric-specific rehabilitation techniques were applied to promote patient engagement during rehabilitation. Progress was assessed by monthly sensory and motor function tests during routine clinic visits and with serial functional brain imaging studies, including structural brain MRI, magnetoencephalography and transcranial magnetic stimulation. Findings The surgery lasted 10 h and 40 min. Vascular revision of the ulnar artery was required a few hours postoperatively. There were no further immediate postsurgical complications. Rejection episodes occurred throughout the first year but were reversed. An increase in serum creatinine led to the addition of sirolimus at 3 months after transplantation with concomitant reduction in tacrolimus targets. Sensibility to light touch was present by 6 months after transplantation. Intrinsic hand muscle innervation was present by 7–10 months after transplantation. At 18 months, the child had exceeded his previous adapted abilities. As of 18 months after transplantation surgery he is able to write and feed, toilet, and dress himself more independently and efficiently than he could do before transplantation. He remains on four immunosuppressive medications and functional neuroimaging studies have shown motor and somatosensory cortical reorganisation. Interpretation Hand transplantation in a child can be surgically, medically, and functionally successful under carefully considered circumstances. Long-term data on the functional trajectory, neurological recovery, psychological sequelae, and the potential late effect of immunosuppression are still needed to support broader implementation of paediatric vascular composite allotransplantation. Funding The Children’s Hospital of Philadelphia.

Introduction Since the first successful vascularised composite allo­ transplantation (VCA) of the hand in an adult in 1998 showed the feasibility of the surgical technique alongside immunosuppression, subsequent cases at multiple centres have led to further refinements, not only in medical and surgical management but also in patient selection, psycho­ social support, rehabilitation, and ethical considerations.1–5 Hand transplantation is not life-saving, but for many patients, the improvements in function and quality of life justify the commitment to lifelong immunosuppression and prolonged functional rehabilitation.3,6–8

In children, concerns underlying the risk–benefit balance of hand transplantation are more nuanced than in adults. Even with continually improving upper limb prosthetic technology, prosthetic abandonment rates remain as high as 45% and are higher in children than in adults, especially when prosthetics are fitted after 2 years of age.9–12 Children have longer potential lifespans to benefit from functioning VCA, but also a longer period of risk for adverse effects of immunosuppression, including infections after trans­ plantation and lympho­proliferative disease.13 Additionally, paediatric VCA requires robust microsurgical expertise, particularly in children, and is challenging because of the

www.thelancet.com/child-adolescent Published online July 18, 2017 http://dx.doi.org/10.1016/S2352-4642(17)30012-3

Lancet Child Adolesc Health 2017 Published Online July 18, 2017 http://dx.doi.org/10.1016/ S2352-4642(17)30012-3 See Online/Comment http://dx.doi.org/10.1016/ S2352-4642(17)30035-4 Division of Nephrology (S Amaral MD, S Lopez NP, E Braham MSW), Division of Neurology (S K Kessler MD), and Division of Rehabilitation Medicine (Prof P R Bryant DO), Department of Pediatrics; Division of Neuroradiology, Department of Radiology (W Gaetz PhD, E S Schwartz MD); Division of Plastic and Reconstructive Surgery, Department of Surgery (B Chang MD, I C Lin MD, S J Kovach MD, Prof L S Levin MD); and Department of Transplantation Surgery (Prof A Shaked MD, M H Levine MD), The Children’s Hospital of Philadelphia, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA; Center for Clinical Epidemiology and Biostatistics (S Amaral), Department of Neurology (S K Kessler, Prof F E Jensen MD), Department of Orthopaedic Surgery (C McAndrew PA, D Bozentka MD, D Steinberg MD, S Kanchwala MD, Prof L S Levin), Department of Pathology and Laboratory Medicine (X Xu MD, Prof D Elder MD), and Department of Psychiatry (D Lefkowitz PsyD), Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA (S Amaral); Department of Occupational Therapy, Center for Rehabilitation (T J Levy OT, D Humpl OT, M Hsia OT, K A Ferry OT), Department of Medical Ethics (Prof C Feudtner MD), and Department of Orthopaedic Surgery (R Carrigan MD, Prof L S Levin), The Children’s

1

Articles

Hospital of Philadelphia, Philadelphia, PA, USA; Department of Plastic Surgery, McGill University, Montreal, QC, Canada (S Thibaudeau MD); and Pediatric Hand and Upper Extremity, Shriners Hospital for Children, Philadelphia, PA, USA (D A Zlotolow MD, S Kozin MD) Correspondence to: Dr Sandra Amaral, Division of Nephrology, Department of Pediatrics, The Children’s Hospital of Philadelphia, University of Pennsylvania, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA 19104, USA [email protected]

Research in context Evidence before this study Before embarking on this procedure, we searched PubMed for articles about hand–forearm transplantation, using a combination of the keywords “vascular composite allograft”, “children”, “adults”, “immunosuppression”, “outcomes”, “complications”, “neurological”, “psychological”, “quality of life”, and “hand transplant”. Our literature review began in January, 2014, and continued until the procedure in July, 2015. Previous research primarily entailed case reports and case series in adults, with only one identifiable case in an adolescent who died postoperatively, one case of conjoined twins with a limb transplanted from one twin to another, and one case of a 1-month-old infant who received an arm and hand transplant from a deceased identical twin sibling. Existing adult data suggested varying degrees of functional and psychosocial benefits but overall low risk of rejection with consistent recipient adherence. The scarce paediatric data suggested surgical feasibility but did not provide sufficient information on immunological aspects. Long-term data were also limited in these patients and there were no previous reports of successful hand–forearm transplantation in children who received heterologous transplantations. Added value of this study This clinical report provides evidence that heterologous hand–forearm transplantation is surgically and medically

small size of vessels and nerves. Two paediatric cases14,15 of limb reconstruction and trans­ plantation in identical twin infants were previously reported that provided information about feasibility and surgical technique, but not immunological issues. One previously reported attempt at VCA with a non-biologically identical donor to an adolescent resulted in severe complications and death in the immediate postoperative period.16 Brain development is a unique consideration in the assessment of paediatric hand transplantation candidates. In an adult experiencing amputation and later transplantation, the goal is reactivation of already mature cerebral motor control of hand function.17 By contrast, in the developing brain of a child, the cortical mapping of the region corresponding to limb motor control18 is still being refined during crucial periods of neuronal and synaptic plasticity.19,20 This enhanced ability to dynamically modify neuronal connections and maps in response to environmental influences might present an advantage in children to more optimally recover function of a transplanted limb. However, the prolonged rehabilitation required before functional improvement necessitates sustained attention and a degree of delayed gratification, which can be difficult for younger children to manage. Here, we describe the first bilateral hand–forearm transplantation in a child, with emphasis on surgical and medical care, rehabilitation, and functional outcomes. 2

feasible in children. Tailored, child-centred therapies are ongoing and place substantial burdens on the patient and caregiver; however, short-term functional outcomes are positive. The patient presented in this Article provides new information regarding the potential for cortical recovery of hand motor and primary sensory representation, in the context of amputation that occurred during a crucial childhood developmental period of rich fine motor development. Importantly, the patient has surpassed previous functional abilities and skin allograft rejection has been easily reversed and has not appeared to hinder functional progress. Implications of all the available evidence Taken together, the information learned from this case shows that heterologous hand–forearm transplantation can be successfully achieved in children, with positive short-term outcomes. However, the case also highlights areas for further research. This patient was previously immunosuppressed, having received a kidney transplantation previously. The risk–benefit of performing this procedure in a non-immunosuppressed child must be considered differently. More data are also needed to better inform paediatric candidate selection, including greater knowledge regarding timing of limb loss and crucial windows for neurological recovery.

Methods

The patient The recipient was an 8-year-old African-American boy who had contracted staphylococcal sepsis with systemic ischaemic injury at the age of 2 years, which led to quadrimembral amputation and kidney failure. The right upper limb amputation was at the radiocarpal joint. The left upper limb amputation was at the level of the distal radius, with the radiocarpal joint preserved with some wrist extension and flexion. From 2 to 8 years of age, the child had developed integrated bilateral coordination skills with residual limbs. Upper limb function included bimanual distal forearm use for tasks such as eating with a fork, and unilateral distal residual limb use for tasks such as pushing a door lever. After 2 years of peritoneal dialysis, he received a kidney allograft from his mother at 4 years of age. After thymoglobulin (1·5 mg/kg) and methylprednisolone (15 mg/kg) induction, he was maintained on a steroid-free protocol comprised of mycophenolate mofetil (270 mg/m² per dose) and tacrolimus (trough levels 4–6 ng/mL) with normal renal function (serum creatinine 0·4 mg/dL). He had no episodes of rejection after kidney transplantation, and no hypertension, hyperlipidaemia, or proteinuria. At 6 years of age, he presented to the Shriners Hospital for Children in Philadelphia seeking lower extremity prosthetics. Previous attempts at using upper extremity

www.thelancet.com/child-adolescent Published online July 18, 2017 http://dx.doi.org/10.1016/S2352-4642(17)30012-3

Articles

prosthetics had been unsuccessful, possibly because the patient found it easier to use residual limbs for most tasks. Because of his pre-existing immunosuppression, strong family support, and shown adherence to medical treatment, he was referred to the VCA programme at the Children’s Hospital of Philadelphia for assessment as an upper extremity transplant candidate. Pretransplantation assessment visits spanned approx­ imately 18 months and included repeated visits with orthopaedic surgery, plastic surgery, transplantation medicine, and occupational and physical therapy. Pre­ operative occupational therapy assessment encompassed reviewing the caregiver’s goals for independence with toileting, clothing fasteners, brushing teeth, and cutting food; and the candidate’s goals for climbing monkey bars and gripping a baseball bat. The candidate was independent with dressing without fasteners, feeding, grooming, and bathing with adapted strategies. General upper extremity strength and sensation of intact and residual muscles were within normal limits for a child of his age. A non-standardised version of the Box and Block test21,22 was used preoperatively, since the candidate had to rely on adducting forearms to transfer blocks. In reviewing the adult VCA literature, the popular functional tests did not appear applicable for the paediatric population, and most paediatric assessments focus on fine motor development and not self-care. For these reasons, the patient was tested with the Goal Oriented Assessment of Life Skills,23 since it incorporated daily living skills and school tasks. This assessment also provided the therapists with crucial secondary information regarding cognitive endurance, ability to complete tasks with multi-step directions, problem solving, and frustration tolerance. A child psychologist, paediatric transplantation pharmacist, and social worker assessed psychosocial readiness to undertake surgery and a prolonged rehabilitation period, history of medical adherence, and the family’s ability to provide social and logistical support. The child and his mother, his primary caregiver, had shown resilience through his initial critical illness, peritoneal dialysis, and kidney transplantation, and no psychosocial contraindications to transplantation were identified. Before hand transplantation, he had no panel reactive antibodies (PRA) and tested positive for previous Epstein-Barr virus and cytomegalovirus exposures. The ethics committee and hospital surgical and medical leadership deemed the relative increase in immuno­ suppression following VCA to be an acceptable risk given the potential benefits. The Institutional Review Board deemed this procedure novel clinical care and exempt as a research study. Informed consent and assent discussions with the patient and caregiver were on the basis of best available assessment of risk and benefit, using outcome data from long-term paediatric organ transplantation, and data from adults with hand transplants,1–3,13 with clear communication with the patient’s mother regarding the

large number of unknowns involved. Multiple extensive conversations occurred for over a year with the patient and caregiver, delineating specific risks including fores­ hortened kidney allograft survival, sensitisation, and potential loss, chronic rejection, and non-function of the hand transplants. The caregiver understood the risk of immunosuppression and rejection or graft loss, the risk of surgery, and the need for prolonged rehabilitation for the child. After this process, the child was able to express a desire for hands and the knowledge that it was a surgery and might not work.

The donor The local organ procurement organisation, Gift of Life, assisted in planning organ procurement and transport. A suitable donor became available in early July, 2015, within 3 months of listing. The patient had been listed in the United Network for Organ Sharing (UNOS) deceased donor transplantation registry in late April, 2015. Parameters for the vascularised composite allograft excluded shared maternal human leucocyte antigens (HLA) to reduce the risk of inducing collateral rejection to the kidney allograft. Other parameters included blood type, HLA compatibility, size, skin tone, and 2-h air travel distance to reduce ischaemic time. Before listing, several surgical rehearsals with full teams of surgeons, anaesthesiologists, and operating room staff were done to promote efficiency and effectiveness during the complex procedure.

Procedure The donor limbs were procured and perfused with Belzer UW cold storage solution. Cold ischaemia time was 6 h. Operating room preparation of the recipient included bilateral upper extremity indwelling axillary nerve block catheter placement, large bore lines for vascular access, and thymoglobulin induction with intravenous methylprednisolone. Four teams worked simultaneously on the donor and recipient limbs, identifying and affixing sterile labels to major peripheral nerves, tendons, and vessels to facilitate repair following bony fixation. Osteosynthesis was done with Materialise custom cutting guides applied to the donor and recipient radius and ulna, on the basis of preoperative CT, a method that saved approximately 1 h of ischaemia time (appendix).24 Following rigid osteo­ synthesis with plates and screws, microsurgical repair of the radial and ulnar arteries was followed by multiple vein repairs. Tenorrhaphy of the flexor and extensor tendons of the wrist, thumb, and digits was followed by repair of the median, superficial radial, and ulnar nerves. Interdigitating skin flaps were closed with excess donor skin included for accommodation of swelling and subsequent biopsies. Despite anticoagulation with heparin infusion, 2 h after the operation the right hand was noted to be poorly perfused. Because of postoperative oedema and redundancy of the ulnar artery repair, kinking occurred a few hours after the completion of the

www.thelancet.com/child-adolescent Published online July 18, 2017 http://dx.doi.org/10.1016/S2352-4642(17)30012-3

See Online for appendix

3

Articles

Right hand Left hand

Bimanual

Sensory Pain sensibility

2 months

2 months

··

Deep pressure sensibility

3 months

3 months

··

Emerging light touch sensibility

7 months

7 months

··

Monofilament threshold 4·08 on palmar surface index finger*

8 months

12 months

··

Vibration sensation threshold present but impaired

17 months

17 months

··

Finger naming present, but without mastery

17 months

17 months

··

Motor Nearing lateral pinch

1 month

1 month

··

Forming hand shape for gross grasp

1 month

1 month

··

Grasp–carry 2·5 cm³ cube

3 months

8 months

··

Intrinsic muscles supporting prehension

8 months

10 months

··

Self-feeding with hands

6 months

12 months

Bimanual coordination for scissor and crayon skills

··

··

Functional

Wiping after toileting

··

9 months

Swinging a bat

··

··

Dynamic grasp for writing

6 months

··

Dressing with hands

··

··

·· 8 months ·· 12 months ·· 12 months

Timepoints specified according to first observations in follow-up visits. *The monofilament threshold of 4·08 corresponds to protective sensation.

Table: Post-transplantation sensory, motor, and functional milestones

surgery. This was recognised immediately and the patient was returned to the operating room for vascular revision, which included shortening the ulnar artery repair and performing an additional venous anastomosis of the vena comitans of the ulnar artery. There were no further vascular events. Post-transplantation surveillance included daily physical examination of the limbs as well as pulse oximetry to monitor tissue perfusion. Weekly skin biopsies were done and graded by the 2008 Banff criteria.25 The patient received standard antimicrobial prophylaxis with valganciclovir and trimethoprim–sulfamethoxazole. He transitioned from heparin to aspirin by day 3 after the operation. Brain MRI images were obtained for the patient on a 3·0 Tesla Siemens Verio (TM) scanner using a 32-channel receive-only head radiofrequency (RF) coil. A 3D Magnetization-Prepared Rapid Acquisition Gradient-Echo (MPRAGE) scan was obtained with an axial orientation and a field of view of 256 × 256 × 192 and a matrix of 256 × 256 × 192 to yield 1 mm isotropic voxel resolution (repetition time/echo time=1900/2·87 ms; inversion time=1100 ms; flip angle=9 degrees). TMS motor mapping (Nexstim NBS, Helsinki, Finland) was done using a 70 mm figure-of-eight coil with single pulses given at 110% of motor threshold over the anatomic hand knob and surrounding cortex. 4

Somatosensory evoked responses were measured using a 275-channel magnetoencephalography (MEG; VSM MedTech Inc, Coquitlam, BC, Canada). Tactile stimulation of the left and right index fingers was done separately by use of pneumatic pulses of compressed air delivered via clip-on balloon diaphragms. A pressure level (30 psi) was optimised to achieve non-painful stimulation and a robust brain response from the postcentral somatosensory cortex. A stimulus duration of 35 ms was used to accommodate mechanical diaphragm elasticity and air-flow dispersion along the air tube from the compressed air source. The interstimulus interval (ISI) was jittered between 0·5 s and 0·7 s. Data were obtained in epochs of 0·4 s (–0·1 s to 0·3 s) for a total of 500 trials. Somatosensory responses were averaged and then filtered between 1 and 40 Hz and the direct current offset was removed using the pre-trigger 100 ms time period. The first cortical P50m response was then localised and coregistered to the patient’s structural MRI using single equivalent current dipole methods. The Box and Block test of unilateral gross motor dexterity is comprised of a grasp–carry–release sequence to transfer 2·5 cm³ blocks from one side of a wall to another. After transplantation surgery, procedures were standardised as described by Jongbloed-Pereboom and colleagues.22 Before transplantation, the patient was permitted to compensate by using both of his residual limbs to complete the test bilaterally. The nine-hole peg test is a standardised measure of manual dexterity involving precision placement of small pegs. The patient was seated at a desk with a pegboard situated on the midline at abdominal level. Standardised procedures were followed as described by Wang and colleagues,26 with the following exception. One single trial was administered at each timepoint because of extensive interdisciplinary follow-up testing, whereas Wang and colleagues recorded the best of two trials. Weekly skin biopsies to monitor graft rejection were reduced after 3 months to once every 2 weeks, then monthly, and finally every 2–3 months at 1 year after transplantation. Laboratory tests to monitor kidney function, immuno­ suppressive drug concentration, and complete blood counts were done with biopsies. The patient also received routine screening for Epstein-Barr virus, cytomegalovirus, and polyomavirus, as well as testing for the development of donor-specific antibodies. At 6 weeks, 3 months, 7 months, 11 months, and 14 months after transplantation, the patient had structural brain MRI and MEG to record neural correlates of sensory responses and hand move­ ment, and motor cortex mapping with transcranial magnetic stimulation (TMS) motor evoked potentials.

Occupational therapy Occupational therapy after transplantation focused on integration of the hands into the patient’s body schema

www.thelancet.com/child-adolescent Published online July 18, 2017 http://dx.doi.org/10.1016/S2352-4642(17)30012-3

Articles

and on transitioning from pretransplantation adaptive bilateral residual limb patterns to using hand movements to engage in daily activities. Therapies were tailored to serve the attention span, motivation, activities, and emotions of a child. Although the Goal Oriented A

Assessment of Life Skills appeared to be ideal before the operation, when used postoperatively, the test items did not reflect quantitative or qualitative changes over time, but the clinical observations of bimanual coordination were evident. B

Figure 1: Photos of an 8-year-old boy who had hand–forearm allotransplantation Pictures before (A) and 9 months after (B) surgery. Biceps

0 μV 0·0 ms

Flexor mass

106 μV 11·7 ms

EDC

102 μV 14·3 ms

APB

944 μV 20·3 ms

FDI

0 μV 0·0 ms

ADM

808 μV 19·0 ms

Figure 2: Motor evoked potentials after transcranial magnetic stimulation Cortical representation of intrinsic right hand muscles at 10 months after transplantation. To the left, white pins represent stimulation loci that produced a motor evoked potential of 200 µV or greater in one or more intrinsic hand muscles. To the right, a representative tracing from a single stimulus showing motor evoked potentials in hand muscles, with peak to peak amplified specified in µV and response latency in ms. EDC=extensor digitorum communi. APB=adductor pollicis brevis. FDI=first dorsal interosseous. ADM=adductor digiti minimi.

www.thelancet.com/child-adolescent Published online July 18, 2017 http://dx.doi.org/10.1016/S2352-4642(17)30012-3

5

Articles

A

therapy facilitated compensatory movement strategies to promote a sense of success and satisfaction.

Box and Block test Right hand

60

Left hand

Normative reference

Psychological care

Number of blocks

50 40 30 20 10

ys 13 s,

13

th

s,

on

th

m

on

17

m 12

8

da

ys da

ys

ay 1d

29

s,

s, th

on

m 8

m

on

th

s, th on m 6

da

s ay 5d

ay

ys

3d

da

s,

13

on m

4

B

*

th

s, th on 3m

* s

*

0

Time to complete nine-hole peg test

Role of the funding source

180

The funder had no role in the study design, data collection, data analysis, data interpretation, or writing of the report. The corresponding author had full access to all the data in the study and had final responsibility for the decision to submit for publication.

150

Time (s)

120 90

Results

60 Normative reference

30

ys da 13 s, 17

m

on

th

th on m 12

5m

on

th

s,

s,

8

13

da

da

ys

ys

†

0

Figure 3: Objective tests of dexterity over time after hand and forearm transplantation Normative data are mean with error bars representing twice SD, but data from this study are from a single trial at each timepoint. Normative reference represents the performance of the dominant hand of typically developing peers. (A) Box and Block standardised test of unilateral gross motor dexterity. Before the transplant, the patient compensated by using both of his residual limbs to complete the test (depicted as a grey line). Performance after transplantation was unilateral according to standard procedures. Each hand improved over time. *For the first three postoperative left hand measures, the patient was unable to complete the task. (B) The nine-hole peg test is a standardised measure of manual dexterity performance. Before transplantation, the patient was unable to complete the test. Performance using each hand improved over time post-transplant, following the course of sensorimotor return. †For the first postoperative left hand measure, the patient was unable to perform the task. See Online for video

6

The patient and his mother also regularly met with the psychologist and social worker to assess and support coping with transplantation and rehabilitation, and to plan for school and social re-integration. Child and parental stress associated with the rehabilitation requirements were routinely assessed, with tailored interventions provided as needed. Likewise, given the patient’s extended absence from his community school setting, preparation was focused on his transitioning to less intensive rehabilitation, maintaining performance despite receiving less individualised attention in the school setting, and managing potential peer reactions to his changed physical appearance.

Starting 6 days after transplantation, the child received daily inpatient occupational therapy for 5 weeks then inpatient rehabilitation for 2 weeks. He was discharged 7 weeks after transplantation. Outpatient occupational therapy continued throughout the first year at a day hospital programme closer to the patient’s home. A patient-centred paediatric occu­ pational therapy approach was used to maintain patient engagement, including the use of biofeedback video games, ageappropriate motor imagery, and exercises using finger lights and puppets. As functional grasping progressed, therapy involved the use of writing implements and utensils for feeding. Scheduling naps and leisure time was also necessary. Occasionally, occupational

Postoperative therapy initially included passive gliding and encouraging activation of extrinsic hand muscles, splinting to protect the flexor and extensor tendon sutures, oedema management, and psychological acceptance of the new limbs. Digit movement was present within days of transplantation because the patient’s extrinsic hand muscles were connected to the tendons of the transplanted hands (video). Short bursts of small amplitude and easily fatigable movements progressed to movements supportive of self-care and leisure skills over the first year of occupational therapy (table; figure 1). Transcranial magnetic stimulation elicited motor evoked potentials from the abductor pollicis brevis and first dorsal interosseous, both intrinsic hand muscles, could be evoked in the right hand at 7 months and in the left hand by 10 months (figure 2). Movements clearly attributable to intrinsic hand muscle function, such as lumbrical flexion, began to emerge at 8 months in the right hand and 10 months in the left hand. Repeated administration of two objective tests of dexterity (Box and Block test21,22 and nine-hole peg test26) showed postoperative improvements of hand skills, and improved efficiency compared with preoperative compensatory strategies (figure 3). Gradual recovery of sensory function occurred over the first year. Within weeks, the patient adapted to the presence of the new hands, adjusting his perception of limb length and peri-personal space around the hands. By 12 months, tactile sensation over the dorsal and palmar surfaces of both hands was sensitive to a 4·08 or smaller Semmes Weinstein monofilament, above the protective sensation threshold. Corresponding to this, MEG

www.thelancet.com/child-adolescent Published online July 18, 2017 http://dx.doi.org/10.1016/S2352-4642(17)30012-3

Articles

A

B

Pneumatic tactile station

C

MEG somatosensory fields

Week 33: RD2 tactile stimulation

P50m

50·004 fT 0·0074 s

–Stim (0 ms)

Week 33: LD2 tactile stimulation

50·004 fT 0·0074 s

–Stim (0 ms)

Dipole source localisation

62·5 ms

RT

Central sulcus

P50m

57·0 ms

Central sulcus

RT

Figure 4: Somatosensory responses evoked by magnetoencephalography (A) Evoked responses to tactile stimulation of the right and left index fingers were recorded separately by use of MEG. (B) Somatosensory evoked fields from the averaged evoked response showed large amplitude dipolar cortical responses at about 50 ms, termed the P50m somatosensory evoked response. The leftmost line shows the time of stimulus onset, and the rightmost line indicates the timepoint of the P50m peak (and associated topographical field patterns). (C) Left panels show axial view and right panels show saggital view. Dipole analysis of the P50m response was localised to the posterior bank of the postcentral sulcus of the contralateral primary somatosensory cortex, proximal to the hand–motor knob. These robust responses are of typical cortical location, latency, and dipole orientation. These results show that the median nerve can govern the conduction of tactile mechanical stimulation from the transplanted fingers and produce robust, somaesthetic information from the approximate typical regions of somatosensory cortex. MEG=magnetoencephalography. LD2=left digit 2. RD2=right digit 2. fT=femtotesla. RT=right hemisphere. Tacrolimus Sirolimus

18

Creatinine Rejection episodes

5

16 4

12 3 10 8 2

Creatinine (mg/dL)

Tacrolimus and sirolimus (ng/mL)

14

6 4

1

2 0

0 0

20

40 Weeks after operation

60

80

Figure 5: Serum creatinine, immunosuppressive drug concentrations, and rejection episodes over time

measures of somatosensory responses to tactile stimulation of the digits showed large amplitude somatosensory evoked fields with typical latencies and orthotopic source localisation to the primary somato­ sensory area in the postcentral gyrus (figure 4). Serial radiographs showed evidence of bone healing and remodelling, with anatomic alignment of the osseous structures. Physes remained open.

Postoperatively, the patient received thymoglobulin induction with intravenous methylprednisolone (10 mg/kg), myco­phenolate mofetil (600 mg/m² per dose), and tacrolimus (0·03% ointment). Goal trough tacrolimus concentrations of 10–12 ng/mL were difficult to achieve. In this setting, at week 3, the child experienced his first rejection episode, grade I–II bilaterally, which improved with intravenous methylprednisolone and topical

www.thelancet.com/child-adolescent Published online July 18, 2017 http://dx.doi.org/10.1016/S2352-4642(17)30012-3

7

Articles

betamethasone dipropionate (0·05% ointment) and topical tacrolimus. Fluconazole (3 mg/kg once a day) was added to boost systemic tacrolimus concentrations. When tacrolimus concentrations increased to 12–15 ng/mL, the rejection resolved. 1 month later, bilateral grade I rejection recurred but responded to topical betamethasone. Concomitantly, the serum creatinine concentration increased from 0·5 g/dL to 0·7 g/dL. The creatinine increase was presumed to be secondary to tacrolimus toxic effects. In response, at 3 months after transplantation, sirolimus was added to the treatment regimen to enable reduction in tacrolimus (figure 5). Serious rejection episodes occurred in month 4 (grade III) and month 7 (grade II–III), presenting as erythematous rash with oedema of both hands. Both episodes were successfully treated with topical beta­ methasone, topical tacrolimus, and intravenous methyl­ prednisolone for 3 days. Oral prednisone was tapered gradually over 2–4 weeks depending on the timing of the resolution of the rejection episode. After month 8, several episodes of grade I rejection occurred but cleared with topical treatment, and adjustments to tacrolimus or sirolimus doses. As reported in adult cases,27 all rejections had been T-cell mediated, with no evidence of B-cell or antibody-mediated rejection. The patient did not develop panel reactive antibody or donor-specific antibody to the kidney or limb allografts. As of January, 2017, 18 months after transplantation, he was stable on myophenolate mofetil (205 mg/m² per dose), sirolimus (target trough 50–80 ng/mL), tacrolimus (target trough 5–8 ng/mL), and prednisone (5 mg daily with a serum creatinine of 0·9 mg/dL and cystatin C 0·93 mg/L; estimated glomerular filtration rate 57 mL/min per 1·73 m²; panel).28 To date, the patient has not been weaned off Panel: Comparison of immunosuppression before and after hand transplantation Kidney transplantation (Before) Immunosuppression • Thymoglobulin 1·5 mg/kg for 5 days • Methylprednisolone 10 mg/kg for 5 days Maintenance at 4 years • Mycophenolate mofetil 270 mg/m² per dose • Tacrolimus target 4–6 ng/mL Hand transplantation (After) Immunosuppression • Thymoglobulin 1·5 mg/kg for 4 days • Methylprednisolone 15 mg/kg for 1 day followed by 1·5 mg/kg for 3 days Maintenance at 18 months • Mycophenolate mofetil 205 mg/m² per dose • Tacrolimus target 5–8 ng/mL • Prednisone 5 mg per day • Sirolimus target 50–80 ng/mL

8

immunosuppression because of intermittent episodes of hand rejection. Of note, there have been no observed declines in functional recovery associated with hand transplant rejection. Adverse events in the first year included a urinary tract infection, rhinovirus, two episodes of acute kidney injury associated with dehydration, and neutropenia responsive to granulocyte-colony stimulating factor. Related to the sirolimus, the patient experienced mouth ulcers, responsive to triamcinolone paste, and hyperlipidaemia, treated with a statin. There was no proteinuria or hyper­ tension. Viral studies for cytomegalovirus and Epstein-Barr virus have been persistently negative. A brief spike in polyomavirus viral load to 13 707 copies per mL occurred at 6 months after transplantation but resolved with no intervention. The rehabilitation phase of the transplantation placed considerable demands on the patient and his mother. For the first 6–8 months after transplantation, the patient was more dependent on others for self-care and was required to engage in uniquely challenging daily therapies. Child life, social work, and psychology services were provided to support the patient and his mother throughout the process. The patient tolerated the rehabilitation process well, and required a level of encouragement, rest breaks, and engagement strategies that would be expected of a child his age. He often expressed pride in his accomplishments during therapy and in his functional gains. The patient’s expressed distress primarily stemmed from the length of hospitalisation (5 weeks of inpatient hospital stay followed by 2 weeks of inpatient rehabilitation), frequency of hospital visits, and resultant time away from home. Distress was treated by the multidisciplinary team by first trying to ameliorate sources of distress (ie, reducing the length or frequency of visits when possible), and if that was unable to be achieved, then to provide additional coping support for the patient and family from the team psychologist and social worker. Tailored cognitive-behavioural family interventions (goal setting, problem-solving, stress reduction) were used by the family for identified areas of distress.

Discussion Our report provides proof-of-principle that with effective planning and preparation by VCA surgical teams, a transplant medicine team, occupational therapy or rehabilitation teams, social work, and psychology, as well as an independent team to weigh ethical considerations, hand transplantation in a child can be done successfully. 18 months after transplantation, functional outcomes have exceeded preoperative function. The patient’s course was complicated by multiple episodes of graft rejection, minor systemic infections, moderate renal transplant functional impairment, hyperlipidaemia, the need for chronic anticoagulation with low-dose aspirin, the need for more immunosuppression than he required

www.thelancet.com/child-adolescent Published online July 18, 2017 http://dx.doi.org/10.1016/S2352-4642(17)30012-3

Articles

before hand transplantation, months of intensive rehabilitation, and a prolonged period of time until functional recovery to a pretransplantation level. As in all cases of hand transplantation, the patient will continue daily therapy to optimise hand function. We will continue to follow up functional progress, neuroimaging, immuno­ suppressive drug concentrations, and skin biopsies when indicated to better inform risk–benefit balance for future candidates. Additionally, we will continue to provide psychosocial support, with particular attention to how the child and parent are coping with the ongoing demands of therapy and treatments, and ongoing pharmacist and psychologist interventions as needed to optimise medication adherence. Unsurprisingly, experience from the care of this child has raised many questions for future work in paediatric hand transplantation. Given that hand–forearm trans­ plantation is not a life-saving procedure, substantial caution must be taken in assessing the potential harms and benefits. This child, who had a previous livingrelated kidney transplantion, was already exposed to immuno­ suppression and the associated risks of systemic complications, such as diabetes, infections, and malignancy. For this case, these potential risks were initially thought to be marginally increased by the immunosuppression for VCA, which was expected to be similar to his kidney transplant immunosuppression albeit with the addition of low-dose alternate day prednisone treatment in the long term. Certainly, the surgery itself was not without risk.16 For our patient, a unique risk lay in the effect of the surgery on the underlying renal allograft function. Our patient experienced a doubling of creatinine after hand transplantation and it is unclear how much this allograft injury will foreshorten the renal allograft survival. This increase in creatinine prompted a lowering of calcineurin inhibitor dosing and initiation of sirolimus as an additional treatment to attempt to limit nephro­toxicity; however, the patient is likely to have incurred some degree of chronic renal injury with persistent increase in creatinine to 0·9 mg/dL and resultant reduced glomerular filtration rate. Additionally, immuno­ suppression necessitated intensification from a steroidfree dual regimen (mycophenolate mofetil and tacrolimus) to quadruple immunosuppression, including steroids. The long-term adverse effects of steroid therapy, including impaired growth and bone health, are well known.29 Because steroid-free immuno­suppressive regimens are associated with higher rates of rejection in adults, we do not anticipate withdrawing steroids completely at this time but we do plan to taper to the use of alternate-day steroids to optimise the patient’s growth.30 To date, immunosuppression has not been able to be weaned due to ongoing, intermittent episodes of hand rejection. Greater understanding of the clinical significance of low-grade rejection is needed as are novel therapeutics to control the robust immunological

response observed in this patient and in previous adults who have had VCA. Given the robust immunosuppression required to minimise rejection in this patient and the high demands of therapy, the consideration of lower limb trans­ plantation in children should also be approached with caution. Since children and adults function very well with lower extremity prosthetics, imparting almost normal function to the amputee, the risks of immuno­ suppression might not warrant the potential benefits of lower limb transplantation. By contrast, upper extremity trans­plantation features and sensation play particularly important roles in connecting socially and physically with others and the environment (ie, gestures, selfesteem, and touch). Moving forward, we must also consider the balance in assessing risk–benefit for children who have not been previously immunosuppressed. One would speculate that a child without kidney allograft with the degree of rejection that this patient has experienced would require the same immunosuppressive regimen of four agents (prednisone, tacrolimus, sirolimus, and mycophenolate) to control hand rejection. There is also long-term risk of chronic rejection of the hands with subsequent indications for amputation. Whether or not having functional hand transplants for a time-limited period of life warrants incurring the risks (eg, immunosuppression, need for additional extensive rehabilitation after reamputation) is likely to be an individual ethical choice; we would recommend that both the patient and parents should be able to make the decision to have the procedure, with ethical approval given on a case-by-case basis at this early stage of child vascular allograft transplantation. In paediatrics, this choice also necessitates weighing the informed choice of the parents with the assent of the child. Ongoing public reporting of outcomes, both positive and negative, will be needed to advance the process of informed consent. Additionally, paediatric-specific metrics of quality of life should be assessed in future paediatric hand transplant recipients to elucidate the benefits to a child’s daily lived experience. One of the most promising findings to emerge from this case is that cortical recovery of hand motor and primary sensory representation occurred, including intrinsic muscle re-innervation, despite the absence of hands during a developmental period of rich fine motor development between the ages of 2 and 8 years. Childhood brain development appears to be marked by windows of time that are optimised for the acquisition of certain skills.19,20 The loss of opportunity to lay down pathways during crucial periods of development might lead to permanent deficits even when primary inputs are restored—a process that has been best characterised in the visual cortex. Whether there are crucial periods for motor and somatosensory development, particularly of the hand, is less clear, but this issue is a key consideration for the future of paediatric hand transplantation and

www.thelancet.com/child-adolescent Published online July 18, 2017 http://dx.doi.org/10.1016/S2352-4642(17)30012-3

9

Articles

paediatric candidate selection. On the basis of our early findings, we would strongly suggest that future paediatric hand transplantation candidates have extensive neuroimaging and functional testing first to better understand potential for recovery and to inform targeted therapies. In summary, the first case of bilateral heterologous hand–forearm transplantation in a child shows that with multidisciplinary, collaborative, and planned manage­ ment, VCA in children is surgically and medically possible with positive short-term outcomes. More data are needed to advance VCA in children, including greater knowledge of long-term implications to better provide informed consent to patients and caregivers. Contributors SA, SKK, TJL, WG, MHL, AS, and LSL did the literature search, compiled the figures, designed the study, collected, analysed, and interpreted data, and wrote the first draft of the manuscript. CM, BC, SL, EB, DH, MH, KAF, XX, DE, and DL provided clinical, multidisciplinary care to the patient, assisted with data interpretation, and reviewed and contributed to the manuscript content. CF provided ethical oversight and guidance to the VCA programme and reviewed and contributed to the manuscript content. ST, ICL, SJK, DB, RC, DS, SKa, DAZ, and SKo provided surgical care to the patient, assisted with data interpretation, and reviewed and contributed to the manuscript content. ESS, FEJ, and PRB contributed to neuroimaging and neurorehabilitation expertise for data analysis and interpretation, and reviewed and contributed to the the manuscript content. Declaration of interests FEJ declares investigator-initiated research funding from Eisai Pharma, outside the submitted work. MHL declares non-financial support from Pfizer outside the submitted work. LSL declares grants from Hansjorg Wyss fund and the US Department of Defense during the completion of the study. All other authors declare no competing interests. Acknowledgments SA is supported by the National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases (K23DK083529, R03DK099486 and R01DK110749). SKK currently receives research support from the State of Pennsylvania and the Freidreich Ataxia Research Alliance. MHL is supported by the National Institutes of Health (R01DK106243, K08DK092282) and the US Department of Defense (MR120023P3, RT150071, RT150100).LSL is supported by The Hansjorg Wyss Fund and the Department of Defense. References 1 Dubernard JM, Owen E, Herzberg G, et al. Human hand allograft: report on first 6 months. Lancet 1999; 353: 1315–20. 2 Petruzzo P, Gazarian A, Kanitakis J, et al. Outcomes after bilateral hand allotransplantation: a risk/benefit ratio analysis. Ann Surg 2015; 261: 213–20. 3 Bernardon L, Gazarian A, Petruzzo P, et al. Bilateral hand transplantation: functional benefits assessment in five patients with a mean follow-up of 7·6 years (range 4–13 years). J Plast Reconstr Aesthet Surg 2015; 68: 1171–83. 4 Kumnig M, Jowsey SG, Moreno E, Brandacher G, Azari K, Rumpold G. Case series on defense mechanisms in patients for reconstructive hand transplantation: consideration on transplant defense concept. Ann Transplant 2014; 19: 233–40. 5 Jabłecki J. World experience after more than a decade of clinical hand transplantation: update on the Polish program. Hand Clin 2011; 27: 433–42. 6 Thaunat O, Badet L, Dubois V, Kanitakis J, Petruzzo P, Morelon E. Immunopathology of rejection: do the rules of solid organ apply to vascularized composite allotransplantation? Curr Opin Organ Transplant 2015; 20: 596–601. 7 Hautz T, Engelhardt TO, Weissenbacher A, et al. World experience after more than a decade of clinical hand transplantation: update on the Innsbruck program. Hand Clin 2011; 27: 423–31.

10

8

9 10 11 12 13 14 15 16 17 18 19 20 21

22 23 24

25 26 27 28 29

30

Salminger S, Sturma A, Roche AD, et al. Functional and psychosocial outcomes of hand transplantation compared with prosthetic fitting in below-elbow amputees: a multicenter cohort study. PLoS One 2016; 11: 1–13. Biddiss EA, Chau TT. Upper limb prosthesis use and abandonment: a survey of the last 25 years. Prosthet Orthot Int 2007; 31: 236–57. Biddiss EA, Chau TT. Upper-limb prosthetics: critical factors in device abandonment. Am J Phys Med Rehabil 2007; 86: 977–87. Toda M, Chin T, Shibata Y, Mizobe F. Use of powered prosthesis for children with upper limb deficiency at Hyogo Rehabilitation Center. PLoS One 2015; 10: e0131746. Scotland TR, Galway HR. A long-term review of children with congenital and acquired upper limb deficiency. J Bone Joint Surg Br 1983; 65: 346–49. Kim JJ, Marks SD. Long-term outcomes of children after solid organ transplantation. Clinics (Sao Paulo) 2014; 69 (suppl 1): 28–38. Zuker RM, Redett R, Alman B, Coles JG, Timoney N, Ein SH. First successful lower-extremity transplantation: technique and functional result. J Reconstr Microsurg 2006; 22: 239–44. Petruzzo P, Lanzetta M, Dubernard JM, et al. International Registry on Hand and Composite Tissue Transplantation.Transplantation 2010; 90: 1590–94. Iglesias M, Leal P, Butrón P, et al. Severe complications after bilateral upper extremity transplantation. Transplantation 2014; 98: e16–17. Vargas CD, Aballéa A, Rodrigues EC, et al. Re-emergence of hand-muscle representations in human motor cortex after hand allograft. Proc Natl Acad Sci USA 2009; 106: 7197–202. Chakrabarty S, Martin JH. Postnatal development of the motor representation in primary motor cortex. J Neurophysiol 2000; 84: 2582–94. Ismail FY, Fatemi A, Johnston MV. Cerebral plasticity: windows of opportunity in the developing brain. Eur J Paediatr Neurol 2016; 21: 23–48. Knudsen EI. Sensitive periods in the development of the brain and behavior. J Cogn Neurosci 2004; 16: 1412–25. Platz T, Pinkowski C, van Wijck F, Kim I-H, di Bella P, Johnson G. Reliability and validity of arm function assessment with standardized guidelines for the Fugl-Meyer Test, Action Research Arm Test and Box and Block Test: a multicentre study. Clin Rehabil 2005; 19: 404–11. Jongbloed-Pereboom M, Nijhuis-van der Sanden MWG, Steenbergen B. Norm scores of the box and block test for children ages 3–10 years. Am J Occup Ther 2013; 67: 312–18. Miller LJ, Oakland T. Goal oriented assessment of life skills. 2013. http://www.wpspublish.com/store/p/2787/goal-goal-orientedassessment-of-lifeskills (accessed June 2, 2017). Mendenhall SD, Schmucker RW, De la Garza M, Lutfy J, Levin LS, Neumeister MW. Osteosynthesis in forearm transplantation using a novel ulnar-shortening osteotomy system for simultaneous both bone fixation. Vascular Compos Allotransplant 2015; 2: 53–59. Cendales LC, Kanitakis J, Schneeberger S, et al. The Banff 2007 working classification of skin-containing composite tissue allograft pathology. Am J Transplant 2008; 8: 1396–400. Wang YC, Bohannon RW, Kapellusch J, Garg A, Gershon RC. Dexterity as measured with the 9-Hole Peg Test (9-HPT) across the age span. J Hand Ther 2015; 28: 53–60. Morelon E, Kanitakis J, Petruzzo P. Immunological issues in clinical composite tissue allotransplantation: where do we stand today? Transplantation 2012; 93: 855–59. Schwartz GJ, Work DF. Measurement and estimation of GFR in children and adolescents. Clin J Am Soc Nephrol 2009; 4: 1832–43. Tsampalieros A, Knoll GA, Molnar AO, Fergusson N, Fergusson DA. Corticosteroid use and growth after pediatric solid organ transplantation: a systematic review and meta-analysis. Transplantation 2016; 101: 694–703. Kanitakis J, Petruzzo P, Badet L, et al. Chronic rejection in human vascularized composite allotransplantation (hand and face recipients): an update. Transplantation 2016; 100: 2053–61.

www.thelancet.com/child-adolescent Published online July 18, 2017 http://dx.doi.org/10.1016/S2352-4642(17)30012-3