Keywords
arm transplantation - vascularized composite allotransplantation - hand transplantation
Arm transplantations are performed less frequently than forearm and hand transplantations.
One of the main reasons for this is the great distance of nerve regeneration required
to reinnervate a large number of muscles, mainly the intrinsic musculature of the
hand.[1]
[2]
[3] Therefore, the patient requires intense and prolonged rehabilitation, as well as
a greater multidisciplinary support to obtain basic function of the limb. Another
factor that limits arm transplantation is the large amount of muscle mass present
in the transplant, which demands shorter periods of ischemia and carries the risk
of greater complications, including death.[4]
[5]
To the best of our knowledge, only eight arm transplants have been performed in seven
patients to date. Of these, one was at the level of the proximal third of the humerus,
five were transhumeral, and two were at the distal third of the humerus. Five patients
were reported in the medical literature, and the last two were reported in electronic
media and social networks.[1]
[2]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13] Functionally, elbow flexion and extension was satisfactory and recovered quickly.
Flexion and extension of the wrist and fingers was also obtained. However, function
of the intrinsic musculature was not reported. The reported sensitivity was of a protective
type with a distinction between thermal stimuli.[1]
[8]
[9]
[14]
The most proximal transplantation to date was performed in Germany by Höhnke et al.[6] The transplant was distal to the insertion of the deltoid muscle and had only one
functional report with a follow-up of 2 years.[14]
In Mexico, an upper limb transplantation protocol was implemented in 2007. After performing
our first successful case of bilateral forearm transplantation in 2012,[15] we performed an evaluation of 131 patients, with amputation of a segment of the
upper limb, in whom hand transplantation was considered a treatment option. Most of
these cases were arm amputations at varying levels.[16]
The final functional results of arm transplantation are so far unknown, mainly regarding
the reinnervation of intrinsic muscle;[9]
[14] however, due to the devastating disability that arm amputation causes, it is considered
beneficial to perform a bilateral arm transplantation in an ideal patient.[7] In the present case, one arm was transplanted at the level of the glenohumeral joint,
and the other at the level of the middle third of the humerus. As arm transplantation
is very rare, extensive knowledge of the upper limb anatomy is not common in any specialty.[17] Therefore, the aim of this article was to elucidate the surgical, microsurgical,
and anatomical considerations that enabled this bilateral arm transplantation.
Methods
A 51-year-old male patient suffered a high-voltage electric burn in January 2012 with
an entry point on the upper extremities and an anterior thorax exit point. This led
to the amputation of the upper right extremity at the level of the proximal epiphysis
of the humerus and amputation of the left arm at the transhumeral level. The patient
was unable to adapt to mechanical prostheses. His Disabilities of the Arm, Shoulder
and Hand (DASH) score was 75.83 points. The patient was healthy without psychiatric
history, personality disorders, anxious or narcissistic traits, or depressive or anxious
behavior secondary to traumatic episodes. He also had optimal support networks and
they understood the risks and consequences of the procedure. Hence, this patient was
considered an ideal candidate for upper limb transplantation.
The right stump had a stable cutaneous cover with an osteoporotic humeral head. There
was atrophy of the deltoid, supraspinatus, infraspinatus, and latissimus dorsi muscles.
The upper half of the pectoralis major was preserved. There was a palpable axillary
artery and positive Tinel's sign at the infraclavicular level ([Fig. 1]).
Fig. 1 Preoperative status. (A) Extensive scarring caused by exit point of energy, atrophy of the right pectoralis
major muscle, and redundancy of skin cover of the left stump. (B) Atrophy of the right supraspinatus, infraspinatus, and latissimus dorsi muscles.
(C) Remnant proximal epiphysis of the humerus with significant osteopenia and integrity
of the glenohumeral joint. (D) Important osteopenia of the distal third of the remaining humerus compared with
the proximal third. (E) The axillary artery is visualized until its third anatomical portion with presence
of the anterior and posterior humeral circumflex arteries (arrows). (F) Brachial artery in good condition in the proximal third of the left arm.
The left stump was at the level of the junction of the middle and distal thirds of
the humerus with a stable skin cover. The shoulder muscles were functional. There
was partial conservation of the origin of the two heads of the biceps brachii muscle
and the lateral and long heads of the triceps brachii muscle, as well as the upper
portion of the pectoralis major. The coracobrachialis muscle was healthy. There was
a positive Tinel's sign 5 cm proximal to the stump. The brachial artery was palpable
at the upper third of the stump ([Fig. 1]).
Bilateral upper limb transplantation was performed in October 2015. The right limb
was transplanted at the glenohumeral joint and the left limb at the middle third of
the arm. The multiorgan donor was a 33-year-old man with brain death. The human leukocyte
antigen (HLA) donor/recipient mismatch was five of six, with a reactive panel of antibodies
against major histocompatibility complex class I (2%) and class II (7%) and negative
cross tests. The ABO and Rh groups of both donor and recipient were A positive. Serology
of donor/recipient was: cytomegalovirus −/+; rubella +/+; Toxoplasma −/−; and Epstein–Barr virus +/+. The donor and recipient were in separate hospitals
1 hour apart.
Recovery
Two surgical teams recovered the upper extremities simultaneously with the recovery
of the solid organs. Tourniquet control was not used.
Right arm
The deltoid muscle was disassembled from the clavicle. The axillary artery and vein
were identified and dissected distal to the subclavian artery and vein. The tendons
of the pectoralis major, latissimus dorsi, and teres major were identified and cut
3 cm from their insertion on the humerus. Disinsertion of the origin of the long portion
of the triceps brachii, long and short portion of the biceps brachii, and coracobrachialis
was done. The glenohumeral joint capsule was sectioned at the neck of the scapula
along with the tendons of the supraspinatus, subscapularis, infraspinatus, and teres
minor. The brachial plexus was identified at the lateral, posterior, and medial cords
and was tagged and sectioned. The axillary artery was cannulated at the level of its
junction with the subclavian artery. The axillary vein was sectioned distal to the
subclavian vein. The arm was perfused with 3 L of histidine–tryptophan–ketoglutarate
(HTK) solution at 4°C until clear drainage was obtained through the axillary vein.
The recovery time was 3 hours.
Left arm
A circumferential incision was made proximal to the insertion of the deltoid muscle.
The long and lateral heads of the triceps brachii, tendons of the short and long portion
of the biceps brachii, and coracobrachialis muscle were identified and released from
their origins. The cephalic vein was identified and dissected as were the brachial
artery and veins. Humeral osteotomy was performed distal to the insertion of the deltoid
muscle. The median, radial, ulnar, and musculocutaneous nerves were identified and
sectioned at their proximal thirds. The brachial artery was cannulated at its proximal
third in the arm and the veins were sectioned. The arm was perfused with the same
solution as the right arm until clear drainage was obtained through the veins. The
recovery time was 1 hour and 57 minutes.
Right Arm Implantation
In the residual stump, an infraclavicular incision was made with extension toward
the anterior border of the axilla and was continued along the previous surgical scars.
The tendon of the pectoralis major muscle was removed from the humerus. The axillary
artery and vein were identified and dissected from below the minor pectoralis muscle
up to the subclavian artery and vein. The median, musculocutaneous, ulnar, radial,
and axillary nerves were then identified and tagged. The residual tendons of the latissimus
dorsi and teres major muscles were tagged and released from the humerus. The residual
deltoid muscle was detached from the clavicle and the spine of the scapula. Subsequently,
the joint capsule of the shoulder was sectioned from the humeral head along with the
tendons of the supraspinatus, infraspinatus, teres minor, and subscapularis muscles.
The tendinous origin of the long portion of the biceps brachii was preserved. The
humeral head was resected, leaving the glenoid cavity and joint capsule.
The upper right extremity was removed from cold ischemia. The extremity was perfused
with a buffer solution consisting of 800 mL of Hartmann's formula and 80 g of human
albumin. The joint capsule of the glenoid cavity was joined to the humeral head joint
capsule with interrupted sutures of Nylon number 2. The donor axillary artery was
repaired end-to-end in its first anatomical portion with the third anatomical portion
of the recipient axillary artery, with 6–0 Nylon suture in a continuous pattern. The
arterial perfusion was immediate. The limb was allowed to bleed for 3 minutes with
a blood loss of approximately 400 mL. The axillary vein was repaired terminal–lateral
in its first anatomical portion using the parachute technique with continuous 6–0
Nylon suture. The total ischemia time was 3 hours and 48 minutes (3 hours and 11 minutes
of cold ischemia and 37 minutes of warm ischemia).
Tenorrhaphy of the long portion of the biceps brachii was performed distal to the
glenohumeral joint capsule; tenorrhaphy of the supraspinatus, infraspinatus, teres
minor, and subscapularis was also performed. The tendons of the latissimus dorsi,
teres major, and the long portion of the triceps brachii were repaired. The short
portion of the biceps brachii and the coracobrachialis muscle were reinserted into
the coracoid apophysis with Nylon number 1. Microsurgical epineural nerve repair of
the axillary, musculocutaneous, radial, ulnar, and lateral and medial branches of
the median nerves was performed using 9–0 Ethilon at magnification ×24. Prolene number
1 was used to reinsert the pectoralis major muscle on the humerus and the deltoid
muscle on the clavicle. Finally, we performed hemostasis, placement of Jackson–Pratt
type closed silicone drainage, and remodeling of the cutaneous flaps.
Left Arm Implantation
Incisions were made in the medial and lateral regions of the residual stump and connected
with a transverse incision. An anterior and posterior cutaneous flap was dissected
from distal to proximal up to the deltoid muscle insertion on the humerus. The brachial
artery and veins, as well as the median, ulnar, and radial nerves, and the residual
muscular masses of the biceps and triceps brachii were identified. Approximately 10
cm of all nerves were dissected up to the proximal third of the humerus. The recipient
humerus was cut distal to the insertion of the deltoid muscle.
The left extremity was removed from cold ischemia. A distal humeral osteotomy was
performed at the insertion of the deltoid. The osteosynthesis was performed with an
8.5 × 260 mm Fixion intramedullary nail. One brachial vein and then the brachial artery
were repaired via end-to-end anastomosis with 8–0 continuous suture. Revascularization
was immediate without any complications. Total ischemia time was 6 hours and 35 minutes
(5 hours and 57 minutes of cold ischemia and 38 minutes of warm ischemia).
Tenorrhaphy of the long portion of the biceps brachii and long portion of the triceps
brachii was performed with the Bunnell technique. The tendons of the short head of
the biceps brachii and the coracobrachialis muscle were inserted in the coracoid process.
Myorrhaphy was performed of the long and lateral portions of the triceps brachii,
as well as both heads of the biceps brachii and the coracobrachialis muscle to the
corresponding remnant muscles using the “piggyback” technique.
End-to-end epineural nerve repair of the median, ulnar, and radial nerves was performed
with 9–0 suture ([Fig. 2]). Neurorrhaphies were performed 7 cm proximal to the elbow joint. Finally, end–lateral
neurorrhaphy of the donor musculocutaneous nerve to the proximal third of the recipient
radial nerve was performed.
Fig. 2 Neurorrhaphies of the left arm. Circle A shows end-lateral neurorrhaphy of the donor
musculocutaneous nerve to the recipient radial nerve. DMCM, donor musculocutaneous
muscle; DMCN, donor musculocutaneous nerve; DMN, donor median nerve; DRN, donor radial
nerve; DUN, donor ulnar nerve; RMCM, recipient musculocutaneous muscle; RMN, recipient
median nerve; RRN, recipient radial nerve; RUN, recipient ulnar nerve. Black circles
show the sites of neurorrhaphy.
The total blood loss was 13,000 mL. Twenty-two globular packages, 14 plasma packages,
30 cryoprecipitate packages, and 3 platelet apheresis were transfused. The total surgery
time was 17 hours ([Fig. 3]).
Fig. 3 Immediate postoperative. Normal coloration, tissue perfusion, and temperature. Moderate
edema in both extremities with a stable patient in intensive care.
Secondary Surgeries
At postoperative day 12, a seroma was detected at the right axillary level that warranted
surgical evacuation. Intraoperatively, there was dehiscence of the insertion of the
short portion of the biceps brachii and the coracobrachialis muscle in the coracoid
process, as well as at the insertion of the pectoralis major on the humerus. Seroma
evacuation and reconstruction of the abovementioned tendons were performed with fascia
lata tendon allografts. The postoperative course was uneventful.
At postoperative day 306 and according to the Medical Research Council scale for muscle
strength, the patient presented as M1 in right elbow flexion and M2 in the left. Hence,
bilateral transposition of the muscular remnants of the pectoralis major was performed.
An Achilles tendon allograft was used to firm up their insertion in both biceps brachii
tendons.
Results
The patient was healthy with a blood pressure of 131/87 mm Hg and was placed on a
triple maintenance immunosuppression scheme (tacrolimus, mycophenolate mofetil, and
prednisone) with corporal integrity. The skin color of both extremities was similar
to the recipient. The right upper extremity was 1 inch longer than the contralateral
side. Arterial pulses and palpable veins were present. Temperature, capillary refill,
and nail growth were normal. There was decreased hair growth without sweating.
The right shoulder had a similar shape and volume to the contralateral side. There
was abduction (deltoid, supraspinatus) of 90 degrees and M4 and flexion (deltoid,
coracobrachialis, and supraspinatus) of 100 degrees and M4. Internal and external
rotation was M1, elbow flexion (biceps brachii, pectoralis major) was 120 degrees
and M3, elbow extension was M5, pronosupination was M2, and wrist extension was M2.
There was no mobility of the fingers. There was no pain during active mobility. There
was thermal sensitivity allowing discrimination of cold and heat.
The left limb was transplanted with redundant skin at the level of the union with
the graft. Total elbow flexion and extension was M5, pronosupination was M2, wrist
extension was M4, and finger flexion was M2. There was thermal sensitivity allowing
discrimination of cold and heat with residual deep pressure ([Figs. 4], [5], and [6]).
Fig. 4 Twelve postoperative months. (A) Angiotomography showing anterior circumflex humeral artery (arrows). (B) Angiotomography showing posterior circumflex humeral artery (arrows). (C) Humeral head with joint integrity and increased articular space, humeral head descended.
(D) Bone consolidation of left humerus.
Fig. 5 Functionality at 15 months. (A) Recovered body integrity with good skin color match. (B) Elbow flexion. (C) Total flexion of right shoulder. (D) Nearly complete abduction of right shoulder.
Fig. 6 Total left elbow flexion.
The patient regained his corporal integrity and improved his self-esteem and sense
of well-being within a social group. The functions obtained so far allowed him to
eat by himself, prepare food, go to the bathroom alone during night, open doors, carry
bags, swim, switch on and off the lights, and play with his family.
The patient stated, “Until now, the function obtained has justified the risks involved
in transplantation and immunosuppression.”
Discussion
The functional disability caused by transhumeral amputation is greater than the disability
produced by transradial amputation. Although the overall function of a transhumeral
transplant may be less than that of a distal forearm transplant, the magnitude of
the potential improvement is greater due to the greater preoperative disability.[2] Thus, the risk–benefit ratio favors transplantation.[7]
The surgical technique for hand and forearm transplantation has been extensively described;[1]
[18]
[19]
[20]
[21]
[22]
[23]
[24] however, the technique for arm transplantation has been insufficiently reported,
as only few cases of arm transplantation have been performed.[1]
[2]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13] Hence, we have described the surgical technique and some microsurgical concepts
of arm transplantation at the level of glenohumeral joint and a midhumeral transplant,
as well as the clinical results.
Use of tourniquet is recommended for recovering the upper extremity. This avoids hemodynamic
destabilization of the donor and allows fast recovery, usually in less than an hour.[2]
[7]
[17]
[22]
[25] However, when the limb to be recovered has a large amount of muscle mass (e.g.,
an arm), it is necessary to decrease the ischemia time. Hence, we do not advice the
use of a tourniquet. The right arm recovery was complex, meticulous, and caused bleeding.
Therefore, we planned to perform the back table procedure during the recovery time
while continually maintaining the arterial circulation of the arm.
All arm muscles were recovered in their entirety, including the deltoid muscle. The
right blood vessels were recovered up to the origin of the subclavian vessels to preserve
the anterior and posterior circumflex humeral arteries and thus maintain the circulation
of the humeral head. This caused a total recovery procurement time of 3 hours but
without ischemia and without hemodynamically destabilizing the donor.
The left arm was also recovered without the use of a tourniquet, and all the muscles
to be transplanted were kept intact. The time of left arm recovery was also prolonged
but without ischemia time.
The total ischemia time reported in arm transplantation ranges from 1.5 to 9 hours,
the most frequent being 5 to 6 hours of total ischemia.[2]
[7]
[8]
[26] In our case, the total ischemia time in the right arm was 3 hours and 48 minutes
and in the left arm was 6 hours and 35 minutes, which is in accordance with previously
reported ischemia times.
Despite these prolonged ischemia times and the large amount of muscle mass involved,
our patient remained stable hemodynamically and biochemically after the revascularization
of each arm, as well as at the end of surgery, without symptoms of reperfusion syndrome.
Reperfusion syndrome can occur with a total ischemia time of less than 4 hours and
can be fatal in cases of arm transplantation.[27] To avoid this in the right arm, arteriorrhaphy was initially performed, and the
venous return was allowed to drain without phleborrhaphy. Thus, we avoided the flow
of any lysis products (myoglobin) and cytokines to the recipient. The disadvantage
was the significant bleeding, which led to a requirement for massive transfusion with
its potential risks.
The right arm was transplanted with an intact humerus, as the residual humeral head
had marked osteopenia. This is consistent with previously reported amputations, in
which, regardless of the mechanism of amputation, the residual bones have shown marked
osteopenia due to disuse or injury.[17] Additionally, fractures of the humeral neck carry the risk of inadequate consolidation,
even with the application of osteosynthesis material, due to disruption of circulation
of the humeral head.[28] Furthermore, the mechanism of amputation in the present case was a high-voltage
electric burn, which undoubtedly affected the microcirculation of the humeral head.
The recovery of intact muscles as a functional unit in both arms allowed us to reinsert
them into their corresponding origins, thereby, avoiding midlevel muscle repair in
the left arm, diminishing the scar, and allowing us to correctly manipulate the tendon
and muscle tension. In this left arm, insertion and innervation of the muscular stumps
of both the biceps and triceps brachii were preserved. These were repaired with the
“piggyback” technique to increase the chances of obtaining adequate elbow flexion.
These principles and technique have been previously described by Carlsen and Schnneberger.[17]
[18]
Although end-to-side neural repair is more experimental than clinical,[29]
[30] the musculocutaneous nerve of the donor was repaired end–lateral to the recipient
radial nerve in the left arm at its proximal third. This was because the biceps brachii
muscle stumps, although atrophied, presented with good contraction, and we chose not
to lose its function when performing end-to-end repair between musculocutaneous nerves.
It was also surgically more accessible to perform this neurorrhaphy on the upper third
of the radial nerve. The possibility of not obtaining reinnervation of the brachial
biceps donor was contemplated. In this situation, the elbow flexion would be recovered
with the function of the recipient stumps of the brachial biceps and the transference
of pectoralis major muscle.
The reinnervation in the present case has been clinically evaluated for the presence
of mobility. The results of electrophysiological studies will be reported later. However,
we consider important to report that the motor unit action potential on the left donor
biceps brachii was registered at 5 postoperative months and increased its amplitude
gradually until reaching 500 mV at 14 months posttransplant, compared with the right
biceps brachii that reached 826 mV at the same time.
Although at postoperative day 306, there was clinical and electrophysiological activity
in the transplanted biceps, we decided to perform the pectoralis major transfer to
avoid edema and subsequent fibrosis in the hands.
The first movement recovered was flexion and extension of the left elbow, which is
currently at M5. This movement is due to reinnervation of the transplanted biceps
and triceps brachii. The second movement recovered was the abduction of the right
shoulder, which is currently at M4. This movement is due to the repaired suprascapular
tendon and the transplanted deltoid muscle. The flexion of the right elbow was reinforced
with the transfer of the pectoralis major and is currently at M4. Left wrist extension
is at M4, and this movement is due to the transplanted wrist extensor muscles. Digital
flexion and pronosupination were incipient in both extremities. In addition, the patient
had thermal discrimination and sensitivity to deep pressure.
These clinical results at postoperative 15 months are similar to those reported in
previous arm transplantation cases. The patient will continue intensive rehabilitation
to improve mobility.
Conclusion
Clinical results at postoperative 15 months are similar to those reported in previous
arm transplantation cases. Basic concepts in arm transplantation include decreasing
total ischemia time to prevent the flow of cytokines and lysis products from the graft
to the recipient and transplantation of total muscle units. Surgical salvage procedures
such as end-to-side neurorrhaphy were successfully applied in the present case. The
description of the surgical technique and its relationship with the clinical results
will enable better results to be obtained in future arm transplantations.
