Content uploaded by Tommy Nai-Jen Chang
Author content
All content in this area was uploaded by Tommy Nai-Jen Chang on Jan 15, 2019
Content may be subject to copyright.
Vascularized Brachial Plexus
Allotransplantation—An Experimental Study in
Brown Norway and Lewis Rats
Tommy Nai-Jen Chang, MD,
1,2
Kuang-Te Chen, MD,
3
Tessa G o rde n, P h D ,
4
Bassem W. Daniel, MD,
2
Catherine Hernon, MD,
2
Mark Shafarenko, MS,
5
Yen-Lin Huang, MD,
6
Johhny Chuieng-Yi Lu, MD,
1,2
and David Chwei-Chin Chuang, MD
1,2
Background. Brachial plexus injuries are devastating. Current reconstructive treatments achieve limited partial functionality.
Vascularized brachial plexus allotransplantation could offer the best nerve graft fulfilling the like-with-like principle. In this experimen-
tal study, we assessed the feasibility of rat brachial plexus allotransplantation and analyzed its functional outcomes. Methods. A
free vascularized brachial plexus with a chimeric compound skin paddle flap based on the subclavian vessels was transplanted
from a Brown Norway rat to a Lewis rat. This study has 2 parts. Protocol I aimed to develop the vascularized brachial plexus
allotransplantation (VBP-allo) model. Four groups are compared: no reconstruction, VBP-allo with and without cyclosporine A
immunosuppression, VBP autotransplantation (VBP-auto). Protocol II compared the recovery of the biceps muscle and forearm
flexors when using all 5, 2 (C5 + C6) or 1 (isolated C6) spinal nerve as the donor nerves. The assessment was performed on week
16 and included muscle weight, functionality (grooming tests, muscle strength), electrophysiology and histomorphology of the
targeted muscles. Results. Protocol I showed, the VBP-allo with cyclosporine A immunosuppression was electrophysiologically
and functionally comparable to VBP-auto and significantly superior to negative controls and absent immunosuppression. In pro-
tocol II, all groups had a comparable functional recovery in the biceps muscle. Only with 5 donor nerves did the forearm show good
results compared with only 1 or 2 donor nerves. Conclusions. This study demonstrated a useful vascularized complete brachial
plexus allotransplantation rodent model with successful forelimb function restoration under immunosuppression. Only the allo-
transplantation including all 5 roots as donor nerves achieved a forearm recovery.
(Transplantation 2019;103: 149–159)
Brachial plexus injuries (BPI) remain a devastating disease
often leading to a complete loss of arm function. They
bear a profound social cost especially when young, healthy in-
dividuals are affected. Surgical reconstructive methods for BPI
remain valid.
1-6
Yet, lack of nerve graft resources is always one
of the biggest problems encountered when reconstructing
complicated acute BPI. Experimental studies still strive to find
the ideal nerve conduit for longer defects.
7-10
Long and large caliber nerve defects are required but are a
scarce source for peripheral nerve reconstruction. Autologous
small diameter nerve grafts (eg, nonvascularized sural nerve)
undergo rapid revascularization without long ischemia. Large
caliber nonvascularized nerve grafts are not efficiently revas-
cularized due to the occurrence of longer ischemia with central
necrosis. In experimental animal studies, vascularized nerve
grafts have indeed shown a better outcome than nonvascu-
larized nerve grafts. They prevent intraneural fibrosis with a
resulting higher number of regenerated axons. Vascularized
nerve grafts are superior, yet the donor sources are even
more limited.
11-17
Nerve allografts have been investigated for decades.
18-26
With improving immunosuppression regimens, they could
provide us with an unlimited amount of donor nerves for
Received 9 February 2018. Revision received 19 June 2018.
Accepted 20 June 2018.
1
Center for Vascularized Composite Allotransplantation, Department of Plastic and
Reconstructive Surgery, Chang Gung Memorial Hospital, and School of Medicine,
Chang Gung University, Taoyuan, Taiwan.
2
Divisionof Reconstructive Microsurgery, Department of Plastic and Reconstructive
Surgery, Chang Gung Memorial Hospital, Taoyuan, Taiwan.
3
Departmentof Plastic and Reconstructive Surgery,Saint Paul’s Hospital,Taoyuan,
Taiwan.
4
Divisionof Plastic and Reconstructive Surgery,Department of Surgery, The Hospi-
tal for Sick Children and The Hospitalfor Sick Children ResearchInstitute, Program in
Neuroscience and Mental Health, Toronto, Ontario, Canada.
5
Faculty of Medicine, Toronto, University of Toronto, Ontario, Canada.
6
Department of Anatomic Pathology, Chang Gung Memorial Hospital, and School
of Medicine, Chang Gung University, Taoyuan, Taiwan.
This study was supported by a grant from the Ministry of Science and Technology
Taiwan (NSC94-2314-B-182A-176) and the Chang Gung Memorial Hospital,
Linkou, Taiwan (CMRPG3A0441-3).
The authors declare no conflict of interest.
Correspondence: David Chwei-Chin Chuang, MD, Department of Plastic and
Reconstructive Surgery, Chang Gung Memorial Hospital, No. 5 Fu-Shing St.,
Taoyuan, Taiwan. (dccchuang@gmail.com).
Copyright © 2018 Wolters Kluwer Health, Inc. All rights reserved.
ISSN: 004 1-1337/19/10301-01 49
DOI: 10.1097/TP.0000000000002387
Original Basic ScienceçGeneral
Transplantation ■January 2019 ■Volume 103 ■Number 1 www.transplantjournal.com 149
Copyright © 2018 Wolters Kluwer Health, Inc. All rights reserved.
the reconstruction of long nerve defects such as in BPI or sci-
atic nerve injuries. Additionally, it serves the like-with-like
principle. The use of vascularized nerve allografts under cy-
closporine A (CsA) and Tacrolimus (FK-506) treatment
27-29
has been reported. We expect that the use of nerve allografts
could shorten the operative time, reduce the donor site
morbidity, and achieve quicker functional recovery.
30,31
In the recently evolving hand allotransplantation series, the
coapted nerves, basically vascularized nerves allografts, were
successful. After 1-year posttransplantation, there was a re-
covery of sensation in the hand with return of the intrinsic
muscle function.
26,32-37
The purpose of this study is to develop a vascularized allo-
transplantation rat model for brachial plexus reconstruction
for the potential clinical use in the future. The feasibility and
outcome with and without immunosuppression are assessed.
Also, the influence of a number of donor nerves coaptated
within the allotransplanted brachial plexus is compared.
We hypothesize that vascularized brachial plexus allotrans-
plantation (VBP-allo) with an adequate immunosuppression is
feasible and comparable to the clinically hypothetical, optimal
VBP autotransplantation (VBP-auto). Furthermore, we hy-
pothesize that reconstructing all injured roots maximizes the
final outcome. One of most attractive thing is that the immu-
nosuppressant in the model can be designed for temporarily
use, not permanent, which can markedly reduce the morbid-
ity of the recipient.
MATERIALS AND METHODS
Animals
Forty recipient Lewis rats weighing 350 to 400 g were in-
cluded for both protocols. Vascularized allogenic brachial
plexuses were transplanted into Lewis rats. The brachial
plexus was unilaterally removed and donated from the same
number of Brown Norway rats with a similar body weight.
Allotransplantation of tissue from Brown Norway to Lewis
rats is known to constitute a major histocompatibility barrier.
All experimental conditions were approved by the Chang
Gung Memorial Hospital Animal Care Committee and by
Chang Gung University ethical guidelines.
We performed 2 study protocols. In protocol I, 24 rats re-
ceived a resection of the left brachial plexus from C5 to T1
(Figure 1A). They were then divided into 4 groups (n = 6/
group) (Table 1). In protocol II, 16 additional rats received a
resection and vascularized brachial plexus allotransplantation
and undergone an immunosuppression with CsA. A distal
coaptation to the musculocutaneous and median nerves was
performed. They were divided into 2 different groups and
compared later to group II from the protocol I (VBP-allo
+CsA) (Figures 1B and C): In group 1, C5 + C6 spinal nerves
were used as donor nerves, whereas in the second group, only
the C6 spinal nerve was used as the donor nerve (n = 8/group).
Subcutaneous CsA injections were made daily starting
24 hours preoperatively. During the first postoperative week,
the dose of CsA was 16 mg/kg/day, later it was reduced to
8 mg/kg/day at weeks 2 to 8, and then down titrated to
6 mg/kg per day for the maintenance of immunosuppression.
The rats wereweighedtwice a week. The dose of CsAwas ad-
justed based on body weight changes. Flaps created as de-
scribed below were monitored daily, and animals carefully
monitored for dehydration or drug toxicity.
Surgery
Harvest of Vascularized Brachial Plexus
For all animal groups, the rats were placed in a supine posi-
tion and anesthetized with isoflurane inhalation (Halocarbon
Laboratories). The left upper extremity and chest were shaved
and prepared in a sterile fashion. In the Brown Norway rats, a
21 cm elliptical area of skin was marked from the upper
arm down to the volar medial aspect of the elbow region
(Figure 2A). (1) The upper border of the marked skin flap,
used to monitor the perfusion and immune reaction, was
incised first and extended upward to the clavicle. (2) A
constant skin perforator was identified and retrograde
dissected up to the brachial artery and vein was performed.
The pectoralis major and minor muscles were divided to
expose the whole brachial plexus (Figure 2B). (3) The
brachial artery and vein were ligated below the perforators
at the level of the elbow. (4) Seven distal nerve branches
were identified and divided (suprascapular, subscapularis,
axillary, radial, musculocutaneous, median, and ulnar nerve).
Proximally, the 5 spinal nerves from C5 to T1 were identified
and divided as well. (5) The whole plexus around 3.5 cm in
length was harvested with the subclavian (brachial) artery,
vein, and an island pedicle monitoring skin flap (Figure 2C).
This neurocutaneous flap was only prepared when the Lewis
recipient rat was ready to receive the transplantation.
Recipient in Lewis Rats
In Lewis rats, a lazy S-shaped incision inferiorand roughly
parallel to the clavicle was made from the left midclavicle to
the proximal arm. The pectoralis major and minor muscles
were divided at their tendinous attachments and tagged with
a suture. About 3 cm length of the brachial plexus was dis-
sected and discarded. All of the proximal and distal stumps
of the transected nerves weretagged with 10-0 nylon sutures.
The thoracoacromial artery and subclavian vein were dis-
sected and prepared for the anastomoses (Figure 3A).
The Vascularized Brachial Plexus Allotransplantation
From Brown Norway to Lewis Rats
After the donor brachial plexus and its accompanying
monitoring skin flap removed from the donor Brown
Norway (Figure 3B), the neurocutaneous flap transplanted
to the recipient Lewis rat base on the vascular pedicle. The
Brown Norway rats were sacrificed after allograft harvest.
The donor subclavian artery and vein were anastomosed to
the recipient thoracoacromial artery and subclavian vein in
end-to-end fashion to preserve the blood flow to the lower
arm (Figure 3C). The perfusion of the flap was confirmed
by the monitoring skin flap. The coaptation of nerves
according to the protocols was done by using 11-0 nylon
epineural sutures. The pectoralis major muscle was sutured
back, and the monitored skin flap was sutured for wound
closure with interrupted nylon sutures (Figure 3D).
Postoperatively, the rats were placed under a heat lamp
and monitored carefully until they awoke from general anes-
thesia. Each rat was housed individually, kept on a 12-hour
light/dark cycle, and allowed free access to food and water.
The skin flaps were monitored daily for 16 weeks for signs
of circulatory compromise, wound infection, and immune
balance. The transplantation of the donor of the Brown
150 Transplantation ■January 2019 ■Volume 103 ■Number 1 www.transplantjournal.com
Copyright © 2018 Wolters Kluwer Health, Inc. All rights reserved.
Norway rats black skin regrew hair over time. Functional
evaluation started at 4 weeks postoperatively.
Outcome Measurement
Grooming Test
The grooming test for the assessment of shoulder abduc-
tion and elbow flexion by removing drops of water placed
onto their heads.
38,39
This was documented every 2 weeks
for 4 weeks postoperatively and was performed every 2 weeks
thereafter till the animals' sacrifice. The test is graded on a
5-point scale: scored 5 if the paw reaches behind the ear, 3
if the paw passes the snout but does not reach the eye, and
1 if the paw moves but does not reach the snout. Multiple as-
sessments were performed, and the best score was recorded.
Electrophysiological Study
At 16 weeks postoperatively, the Lewis rats were rean-
esthetized with isoflurane inhalation to explore the surgical
site. Two reference electrodes were positioned subcutaneously
and bilaterally in the lateral chest region. A stimulating electrode
was placed and the recording electrode inserted in the corre-
sponding target muscles, including the musculocutaneous
nerve, median nerve (protocol I and II), radial nerve, median
nerve, ulnar nerve (protocol I only) in turn, with the corre-
sponding muscles are the biceps, forearm flexors (included all
flexors of the wrist and digits) (protocol I and II), triceps, fore-
arm extensor group (including the extensor muscles of the wrist
and digits), respectively (protocol I only). Median and ulnar
nerves were stimulated simultaneously in assessing the forearm
flexor group. Stimulating rectangular pulses of 0.5 ms duration
FIGURE 1. The surgical model of VBP-allo. A,All 5 C5 to T1 spinal nerves are used as donor nerves (in groups I to IV in protocols I; and in the
control group of protocol II). B, Two spinal nerves, C5 and C6, are used as donor nerves (in group I of protocol II). C, Single spinal nerve C6 is
used as donor nerve (in group II of protocol II). *The control group of protocol II is group II from protocol I.
TABLE 1.
Study groups in the 2 protocols
Protocol Group (abbreviation) Description Number
I I (DEN) No reconstruction after denervation as the negative control group 6
II (VBPA-allo + CsA) Vascularized brachial plexus allotransplantation with CsA treatment 6
III (VBP-allo −CsA) Vascularized brachial plexus allotransplantation without CsA treatment. 6
IV (VBP-auto) Vascularized brachial plexus autotransplantation as the positive control group 6
II I VBP-allo + CsA with 2 donor nerves (C5 and C6 spinal nerves 8
II VBP-allo + CsA with only 1 donor nerve (C5 spinal nerve) 8
n, number of rats.
© 2018 Wolters Kluwer Chang et al 151
Copyright © 2018 Wolters Kluwer Health, Inc. All rights reserved.
and 50 to 100 V in amplitude were delivered. The test was per-
formed bilaterally so that each rat had its control data.
Muscle Strength Test
After electrophysiological testing, the isometric contrac-
ture forces of the target muscles (biceps, triceps, forearm ex-
tensor group, and forearm flexor group in protocol I,
biceps and forearm flexors in protocol II) was measured
using a force-displacement transducer and digital recording
software. Initially, the resting muscle length of the tendons
was determined. The distal tendons of the biceps and triceps
were severed at the insertion at the elbow. The wrist and dig-
ital flexors were isolated, sutured together, and detached dis-
tally at the wrist. The same procedures were performed for
wrist and digital extensor muscles. Sequentially, each muscle
group was attached to the force transducer. The shoulder, el-
bow, and wrist joints were immobilized on the operating
platform using fixation pins. Before nerve stimulation, the
FIGURE 2. A, Location of the monitoring skin flap on the Brown Norway rat before starting the surgery. B, The septocutaneous perforator
vessels were always identified and included in the skin flap before completing the back cut. C, The left brachial plexus based on the subclavian
vessels including a monitoring skin flap was isolated and prepared for transplantation.
FIGURE 3. A, Dissected neurocutaneous donor flap in a Brown Norway rat. B, Recipient site in a Lewis rat after preparation and removal of a
brachial plexus from C5 to T1. C, Monitoring skin flap with the brachial plexus (C5-T1), the subclavian artery and veins. D, The skin paddle
showed adequate circulation after the anastomosis of artery, vein, and all the nerve coaptations.
152 Transplantation ■January 2019 ■Volume 103 ■Number 1 www.transplantjournal.com
Copyright © 2018 Wolters Kluwer Health, Inc. All rights reserved.
muscle length was adjusted to match its resting length, and
therefore the resting tension was recorded before the stimula-
tion of the nerve. The musculocutaneous nerve was stimu-
lated for activation of the biceps muscle; the radial nerve
for triceps and forearm extensor muscle and median and ul-
nar nerve simultaneously for the forearm flexor muscle
group. Each nerve stimulated at a threshold, 2 times, and
10 times threshold stimulus voltage. Five recordings made
at each stimulus level. The outcome was recorded as the
weight for each threshold stimulated. The right forelimb
served as the control, whereas the left forelimb was the
experimental limb.
Muscle and Nerve Histomorphometry
On completion of electromyography and muscle force stud-
ies, each muscle group was harvested and weighed. Nerve
specimens were obtained proximal to the coaptation site and
cut in 1 mm distance. The nerves then were stained with hema-
toxylin and eosin and 2% toluidine blue. Image-pro plus 6.0
software was used to evaluate the histological sections.
Axon Count and Diameter
Nerve specimens were embedded in epoxy resin and cut
into 1-μm-thick sections and stained with 2% toluidine blue.
The selected sections were photographed under light micro-
scope at 400magnification and enlarged digitally to 1000.
The number of axons was counted in randomly selected areas
within each specimen. Axon count, axon diameter, and fiber di-
ameter were measured with the help of Image-Pro Premier soft-
ware (Media Cybernetics, Inc., Rockville, MD).
Statistical Analysis
The statistical analysis was performed using One-way
analysis of variance with Tukey post hoc test and unpaired
ttest for all comparisons. All analyses were performed
with GraphPad Prism (GraphPad Software Inc., La Jolla,
Calif.). For all analyses, a Pvalue less than 0.05 was
considered significant.
RESULTS
Protocol I: Proof of Principle of the VBP-allo
With Immunosuppression
All of the outcomes were evaluated 16 weeks postopera-
tively. Two rats died in the DEN group, 1 in the VBP-allo
+CsA group, and 1 in the VBP-auto group duringthe postop-
erative course. The rats died at the immediate postoperative
course (within 1 day) due to hypothermia, blood loss or
deep anesthesia.
Effective Allotransplantation and Nerve Regeneration
With Immunosuppression
All skin flaps remained viable in group II (VBP-allo+CsA)
and group IV (VBP-auto) of the Lewis rats, and, in VBP-allo
transplant group, the black hair on the transplanted skin
grew normally (Figure 4). In group III (VBP-allo-CsA), the
skin flaps showed signs of rejection between 5 and 7 days.
The skin flaps shrank and sloughed off between 19 and
22 days postoperatively.
Nerve regeneration proceeded in the rats in all groups as
demonstrated in the median nerve specimen taken 5 mm dis-
tal the site of surgical insertion between the proximal and dis-
tal nerve stumps of the transected brachial plexus. Compared
with the control group (Figure 5A and 6A), the VBP-allo
+CsA group showed well-vascularized nerve tissue with
many myelinated nerve fibers presented (Figures 5B and
6B). In contrast, the longitudinal sections of the nerves in
VBP-allo-CsA group were infiltrated with fat tissue and
FIGURE 4. Postoperative monitoring in VBP-allo+CsA group. The skin paddle was used to monitor the immune tolerance, wound infection
status and microcirculation. A, Immediately postoperatively, (B) 7 days postoperatively, (C) 14 days postoperatively, (D) 60days postoperatively
showed complete wound healing.
© 2018 Wolters Kluwer Chang et al 153
Copyright © 2018 Wolters Kluwer Health, Inc. All rights reserved.
showed scarce axon count and less nerve regeneration; only
few small diameter and poorly myelinated regenerated
nerve fibers could be identified (Figures 5C and 6C).
The nerve regeneration of the allotransplantation group
with immunosuppression (VBP-allo+CsA) compared reasonably
well with the autotransplanted group IV (VBP-auto). The
FIGURE 5. H&E stains of median nerve (400) at the level of 1 cm proximal to the neuromuscular junction. A, Normal median nerve charac-
terized by well-oriented axon fibers with spindle-shaped nuclei. B, Group II (VBP-allo+CsA): axon fibers have increased cellularity but with mild
degree of degenerative changes manifested by vacuolization. C, Group III (VBP-allo-CsA): Extensive degenerative vacuolization of axon fibers
indicating severe rejection. Some fibers lack nuclei and have swelling cytoplasm with intracytoplasmic eosinophilic debris, D, Group IV (VBP-
auto): well-oriented axon fibers with increased cellularity.
FIGURE 6. Toluidine blue-stained transverse section of median nerve (400) at 1 cm proximal to the neuromuscular junction. A, Normal me-
dian nerve characterized by well-oriented, uniform, well-myelinated axons. B, Group II (VBP-allo+CsA): a well vascularized nerve allograft with
well-myelinated axon fibers of variable sizes. There is no significant scarring but minimal degenerative vacuolization is noted. C, Group III (VBP-
allo-CsA): severe degeneration and immunological rejection. There is extensive fibrosis surrounding the axon fibers of variable sizes with or with-
out vacuolization. D, Group IV (VBP-auto): A well-vascularized nerve autograft with large well-myelinated axon fibers demonstrates no evidence
of significant scarring.
154 Transplantation ■January 2019 ■Volume 103 ■Number 1 www.transplantjournal.com
Copyright © 2018 Wolters Kluwer Health, Inc. All rights reserved.
fibrotic scars were less apparent, and the fascicles were well
oriented (Figures 5D and 6D).
The significant positive effect of the immunosuppression with
CsA on allografts when compared without CsA is reflected within
the axon count in Figure 7 and axon diameter in Figure 8.
Furthermore, the myelinated axon count in the arm
(Figures 7A and B) and forearm (Figures 7C and D) was
not statistically different between group II (VBP-allo+CsA)
and IV (VBP-auto), but the distal nerves (median and ulnar)
showed more discrepancy between the groups. Also, there
was no difference when comparing the groups II and IV with
the contralateral healthy side. The nerve fiber diameters were
also similar between groups II and IV but relatively smaller
than in the control group, which is believed to be due to the
axonal sprouting (Figures 8A-D).
Behavior
Grooming test was assessed from 4 weeks after surgery
(Figure 9). The rats in group II (VBP-allo+CsA) achieved a
mean grooming test score of 3.5 which was not significantly
different from the mean of 3.75 in group IV (VBP-auto).
In contrast, the group III (VBP-allo-CsA) was significantly
worse but the recovery was still better than group I (DEN).
Muscle Electromyography and Isometric
Contractile Force
The electrophysiological study (muscle action potentials)
in group II and IV was similar in all examined target muscles
(triceps, biceps, forearm flexors, and forearm extensors). In
group III, the proximal muscles (triceps and biceps) also
demonstrated similar outcomes than in group II and IV,
but the distal musculature (forearm flexors and extensors)
was poorer (Figure 10). The muscle strength in groups II
and IV rats also demonstrated similar muscle strength in all
examined muscles, while in group III both proximal and
the distal muscles were poor. The muscle weight in group I
showed the most significant atrophy in all proximal and dis-
tal muscle groups. Groups II and IV demonstrated compara-
ble result in muscle weight and similar degrees of atrophy.
Rats in group III also showed atrophy especially in the distal
muscle groups, but the level of atrophy was not as severe as
those in group I (Figure 11).
Protocol II (Comparison of the Efficacy of Different
Donor Nerve Count)
In this protocol, we compared the effectiveness of 5 (C5-
T1), 2 (C5 and C6), and 1 (isolated C6) donor nerve to the bi-
ceps and forearm flexors. In all groups, the immune-tolerance
and the nerve regeneration were achieved. The biceps muscle
weight among all groups showed no difference (Figure 12A).
The muscle contraction in 5 donor nerves was not significantly
different when compared with 2 donor nerves, but better
than 1 donor nerve (Figure 12B). Still, all groups were
powerful enough in the biceps muscle. The motor action
potential on the other hand was superior in the 5 donor
nerve groups when compared to 2 and 1 donor nerve,
respectively (Figure 12C).
As for the forearm flexors, the muscle weight, and muscle
contraction force, the 5-donor nerves group was significantly
superior to 2 and 1 donor nerve groups (Figures 12A-B).
Interestingly, there was no significant difference as for the
FIGURE 7. A-D, Comparison of axon count of different target nerves. Marked decrease in musculocutaneous, radial, median, and ulnar
nerves axon count was noted in the allotransplantation group without immunosuppression (group III: VBP-allo-CsA) compared with the
CsA-treated recipients (group II) and autografts (group IV).
© 2018 Wolters Kluwer Chang et al 155
Copyright © 2018 Wolters Kluwer Health, Inc. All rights reserved.
motor action potential (Figure 12C). The comparison of 5,
2, and 1 donor nerves showed a sufficient proximal muscle
innervation, whereas the distal muscles achieved an
insufficient reinnervation in the groups with 1 or 2 spinal
donor nerves (Figure 12A-C).
DISCUSSION
Nerve Allografts in Animal Model
Brachial plexus reconstruction with nerve allografts cur-
rently requires immunosuppressants.
18-25,40
In the 1970 to
1980s, the development of microsurgical techniques and
immunosuppression regimens made nerve allografts a pos-
sible reconstructive option. Zalewski and Gulati
20
and
Muramatsu et al
30
demonstrated that nonvascularized
nerve allotransplantation in rats have similar functional
outcomes when compared to autologous nerve grafts un-
der appropriate immunosuppression.
Clinical Application
Clinical nerve allotransplantation has been performed
by using cadaveric nerve allografts or as a component of
vascularized composite allotransplantations (hands, face).
Mackinnon pioneered the reconstruction of large periph-
eral nerve defects in the extremities by using cadaveric allo-
grafts.
11
In 2011, Elkwood et al
33
reported another clinical
series in which brachial plexus reconstruction was
FIGURE 9. Grooming test in protocol I. FIGURE 10. Electrophysiological tests of protocol I.
FIGURE 8. A-D, Comparison of axon diameter of different target nerves. The mean axon diameter was larger in CsA-treated recipients (group II)
and autografts (group IV) than in recipients without immunosuppression (group III) in the radial, musculocutaneous, median, and ulnar nerves.
156 Transplantation ■January 2019 ■Volume 103 ■Number 1 www.transplantjournal.com
Copyright © 2018 Wolters Kluwer Health, Inc. All rights reserved.
performed by both cadaveric and living donor allografts.
In both series, acceptable clinical results were shown and
immunosuppressive therapy was discontinued after nerve
regeneration with an average of 12 to 18 months. The
nerve allograft probably serves as a temporary conduit
allowing axons to regenerate, whereas the donor Schwann
cells are replaced by the recipient.
23
Follow-upinhandallo-
transplantation trials has shown that nerve regeneration into
the transplanted extremity allows for some extrinsic and in-
trinsic muscle regeneration as well as sensory recovery within
1 year postoperatively.
26,33-35
Similar conclusions were also
made in the face transplant trials.
36,37
The results were en-
couraging and showed that nerve regeneration in composite
tissue allotransplantation with immunosuppressants to be
feasible with functional recovery.
The choice of immunosuppressants and their optimal dos-
age is still under debate. FK-506 and CsA might even have a
neuroprotective antioxidative effect.
41
Nerve regeneration in rats is quicker than in humans. The
CsA-negative groups (Figure 7A) have a reasonable axon
count in the musculocutaneous and radial nerve correlating
to a relatively low, but present force (Figure 11) of the biceps/
triceps muscles, respectively, their weight (Figures 12A-C).
Obviously, some axons can pass the nerve allograft or
possibly show lateral sprouting into target muscles.
Vascularized Brachial Plexus Allotransplantation
In Levy's human cadaveric study, 4 angiosomes of the
human brachial plexus were identified while almost entirely
being nourished by the subclavian system.
42
An allogenic ca-
daveric brachial plexus is therefore considered a potential so-
lution for specific BPI including (1) total plexus palsy with at
least 1 spinal nerve and avulsions and/or scars in the remain-
ing nerves, (2) large benign neoplastic lesions involving the
brachial plexus, such as plexiform neurofibroma, and(3) irra-
diated brachial plexus neuritis. The use of a vascularized bra-
chial plexus allograft offers several advantages: (1) it allows
total reconstruction of the plexus nerves, (2) it provides more
axons during nerve regeneration, (3) and it permits the utili-
zation of larger “trunk”grafts without the problem of central
necrosis. The biceps muscle in protocol II was chosen to com-
pare the shorter nerve regeneration potential in the proximal
muscles versus the longer nerve regeneration of the forearm
FIGURE 11. Muscle strength test of protocol I.
FIGURE 12. A-C, Comparison of biceps and forearm flexors between 5, 2, and 1 spinal nerves as donor nerves. This shows that the reinner-
vation in the downstream muscles was achieved. However, 1 or 2 spinal nerves as donor nerves can only selectively reinnervate. Functional
recovery may occur in the proximal muscles, but less in the distal muscles.
© 2018 Wolters Kluwer Chang et al 157
Copyright © 2018 Wolters Kluwer Health, Inc. All rights reserved.
flexors. Increasing the number of donor nerves showed in
Figure 12C a proportional superior preservation of intramus-
cular action potentials of the biceps, whereas the muscle
weight and strength did not show an apparent gross atrophy.
This might reflect a relatively mild neuromuscular injury. On
the contrary, in distal flexors with a longer time and distance
required to reinnervate, 5 donor nerves seem be more power-
ful to the regeneration than 2 and 1 donor nerve.
Animal Model for Brachial Plexus Allotransplantation
In this study, we demonstrated the use of a composite chi-
meric neurocutaneous flap consisting for nerve reconstruc-
tion. The skin has the strongest antigenicity. Therefore, the
inclusion of the skin flap not only allowed for the monitoring
of the circulation, but also for assessing the immunologic reac-
tion. Our study showed that group II (VBP-allo+CsA) had
similar efficacy to group IV (VBP-auto), although vascularized
nerve allografts would be more antigenic than nonvascu-
larized nerve allografts. Cyclosporine A was an effective immu-
nosuppressant that permitted nerve regeneration, although
recently the attention has been shifted to FK-506, which was re-
cently proven to promote the nerve regeneration in addition to
its immunomodulatory effects.
28,43,44
The design of protocol II was more clinically oriented,
since donor nerves are usually limited clinically. We found
that when only 1 or 2 donor nerves were available, the func-
tional recovery was more prominent in the proximal muscles
than in the distal muscle groups. This is congruent to our clin-
ical experience from over 30 years.
41
In the groups of protocol II with fewer reconstructed spi-
nal nerves we chose C5/C6 and not C8/T1. According to
Coggeshall et al,
45
the number of myelinated axons from
C5-T1 was 1295, 1484, 1575, 1734, and 1213, respectively,
and therefore minimally different. In our extensive clinical
experience, induction exercises and postoperative physio-
therapy is required to improve the outcome; however, in
the rats, this is not possible. Moreover, the main reason for
choosing C5/C6 over C8/T1 is the easier access and exposure
of the more superficial C5/C6.
CONCLUSIONS
In this study, we have developed a new rodent model that
allows for the assessment of a vascularized allogeneic bra-
chial plexus transplantation, which is a potent utility for the
treatment of extensive and challenging acute BPIs. In addi-
tion, we emphasize that 1 or 2 spinal nerves as donor nerves
cannot sufficiently innervate all the terminal branches com-
pared with 5 spinal nerves. Functional recovery may occur
in the proximal muscles but is less in the distal muscles.
In conclusion, vascularized brachial plexus allotransplan-
tation with all 5 donor nerves in combination with suitable
immunosuppression shows potential for clinical application
in the future.
REFERENCES
1. Naraka’s A. Surgical treatment of avulsion type injuries of the brachial
plexus. In: Brunelli G, editor. Textbook of Microsurgery.Masson:Milan;
1988:781–787.
2. Millesi H. Update on the treatment of adult brachial plexus injuries. In:
Gilbert A, editor. Brachial Plexus Injuries. Martin Dunitz: London; 2001:
77–90.
3. Terzis JK, Papakonstantinou KC. The surgical treatment of brachialplexus
injuries in adults. Plast Reconstr Surg. 2000;106: 1097–1122; quiz
1123-1094.
4. Alnot J. Traumatic paralysis of the brachial plexus: preoperative problems
and therapeutic indications. In: Terzis JK. Microreconstruction of Nerve
Injuries. Saunders; 1987:325–345.
5. Chuang DC. Brachial plexus injury: nerve reconstruction and functioning
muscle transplantation. Semin Plast Surg. 2010;24:57–66.
6. Chuang DC. Adult brachial plexus reconstruction with the levelof injury: re-
view and personal experience. Plast Reconstr Surg. 2009;124(Suppl 6):
e359–e369.
7. Keilhoff G, Pratsch F, Wolf G, et al. Bridging extra large defects of periph-
eral nerves: possibilities and limitations of alternative biological grafts from
acellular muscle and Schwann cells. Tissue Eng. 2005;11:1004–1014.
8. Fansa H, Keilhoff G, Forster G, et al. Acellular muscle with Schwann-cell
implantation: an alternative biologic nerve conduit. J Reconstr Microsurg.
1999;15:531–537.
9. Hu J, Zhu QT, Liu XL, et al. Repair of extended peripheral nerve lesions in
rhesus monkeys using acellular allogenic nerve grafts implanted with au-
tologous mesenchymal stem cells. Exp Neurol. 2007;204:658–666.
10. Sachanandani NF, Pothula A, Tung TH. Nerve gaps. Plast Reconstr Surg.
2014;133:313–319.
11. Best TJ, Mackinnon SE, Evans PJ, et al. Peripheral nerve revascularization:
histomorphometric study of small- and large-caliber grafts. J Reconstr
Microsurg.1999;15:183–190.
12. Mackinnon SE, Kelly L, Hunter DA. Comparison of regeneration across a
vascularized versus conventional nerve graft: case report. Microsurgery.
1988;9:226–234.
13. Shibata M, Tsai TM, Firrell J, et al. Experimental comparison of
vascularized and nonvascularized nerve grafting. JHandSurg.1988;13:
358–365.
14. Seckel BR, Ryan SE, Simons JE, et al. Vascularized versus non-
vascularized nerve grafts: an experimental structural comparison. Plast
Reconstr Surg. 1986;78:211–220.
15. Gilbert A. Vascularized sural nerve graft. Clin Plast Surg. 1984;11:73–77.
16. Breidenbach WC, Terzis JK. Vascularized nerve grafts: an experimental
and clinical review. Ann Plast Surg. 1987;18:137–146.
17. Zhu Y, Liu S, Zhou S, et al. Vascularized versus nonvascularized facial
nerve grafts using a new rabbit model. Plast Reconstr Surg. 2015;135:
331e–339e.
18. Bain JR, Mackinnon SE, Hudson AR, et al. The peripheral nerve allograft:
an assessment of regenerationacross nerve allografts in rats immunosup-
pressed with cyclosporin a. Plast Reconstr Surg. 1988;82:1052–1066.
19. Bain JR. Peripheral nerve and neuromuscular allotransplantation: current
status. Microsurgery. 2000;20:384–388.
20. Zalewski AA, Gulati AK. Survival of nerve allografts in sensitized rats
treated with cyclosporin a. JNeurosurg. 1984;60:828–834.
21. Zalewski AA, Gulati AK. Failure of cyclosporin-A to induce immunological
unresponsiveness to nerve allografts. Exp Neurol. 1984;83 :659–663.
22. Evans PJ, Midha R, Mackinnon SE. The peripheral nerve allograft: a com-
prehensive review of regeneration and neuroimmunology. Prog Neurobiol.
1994;43:187–233.
23. Mackinnon SE, Hudson AR, Bain JR, et al. The peripheral nerve allograft:
an assessment of regeneration in the immunosuppressed host. Plast
Reconstr Surg. 1987;79:436–446.
24. Lassner F, Schaller E, Steinhoff G, et al. Cellular mechanisms of rejection
and regeneration in peripheral nerve allografts. Transplantation. 1989;
48:386–392.
25. Midha R, Mackinnon SE, Becker LE. The fate of Schwann cells in periph-
eral nerve allografts. J Neuropathol Exp Neurol. 1994;53:3 16–322.
26. Dubernard JM, Petruzzo P, Lanzetta M, et al. Functional results of the first
human double-hand transplantation. Ann Surg. 2003; 238:128–136.
27. Auba C, Hontanilla B, Arcocha J, et al. Peripheral nerve regeneration
through allografts compared with autografts in FK506-treated monkeys.
JNeurosurg. 2006;105:602–609.
28. Tajdaran K, Shoichet MS, Gordon T, et al. A novel polymeric drug delivery
system for localized and sustained release of tacrolimus (FK506).
Biotechnol Bioeng. 2015;112:1948–1953.
29. Udina E, Gold BG, Navarro X. Comparison of continuous and discontinu-
ous FK506 administration on autograft or allograft repair of sciatic nerve
resection. Muscle Nerve. 2004;29:8 12–822.
30. Muramatsu K, Doi K, Kawai S. Vascularized allogeneic nerve transplanta-
tion with cyclosporine immunosuppression. Ann Plast Surg. 1994;33:
507–516; discussion 516-508.
31. Muramatsu K, Doi K, Kawai S. Vascularized allogeneic joint, muscle, and
peripheral nerve transplantation. Clin Orthop Relat Res. 1995;194–204.
158 Transplantation ■January 2019 ■Volume 103 ■Number 1 www.transplantjournal.com
Copyright © 2018 Wolters Kluwer Health, Inc. All rights reserved.
32. Mackinnon SE, Doolabh VB, Novak CB, et al. Clinical outcome follow-
ing nerve allograft transplantation. Plast Reconstr Surg. 2001;107:
1419–1429.
33. Elkwood AI, Holland NR, Arbes SM, et al. Nerve allograft transplantation
for functional restoration of the upper extremity: case series. JSpinalCord
Med. 2011;34 :241–247.
34. Francois CG, Breidenbach WC, Maldonado C, et al. Handtransplantation:
comparisons and observations of the first four clinical cases. Microsur-
gery. 2000;20:360–371.
35. Shores JT, Brandacher G, Lee WP. Hand and upper extremity transplan-
tation: an update of outcomes in the worldwide experience. Plast
Reconstr Surg. 2015; 135: 351e–360e.
36. Siemionow M, Gharb BB, Rampazzo A. Pathways of sensoryrecovery af-
ter face transplantation. Plast Reconstr Surg. 2011;127:1875–1889.
37. Khalifian S, Brazio PS, Mohan R, et al. Facial transplantation: the first
9 years. Lancet. 2014;384:21 53–2163.
38. Bertelli JA, Mira JC. Behavioral evaluating methods in the objective clinical
assessment of motor function after experimental brachial plexus recon-
struction in the rat. J Neurosci Methods. 1993;46:2 03–208.
39. Inciong JG, Marrocco WC, Terzis JK. Efficacy of intervention strategies in a
brachial plexus global avulsion model in the rat. Plast Reconstr Surg.
2000;105:2059–2071.
40. Grochowitz P, Hettlage P, Schatzl M, et al. Immunosuppression in nerve
allografting: analysis of revascularization and cellular infiltrates. Transplant
Proc. 1985;675.
41. Bulatova N, Yousef AM, Al-Khayyat G, et al. Adverse effects of tacrolimus
in renal transplant patients from living donors. Curr Drug Saf.2011;6:
3–11.
42. LevySM, Taylor GI, Baudet J, et al. Angiosomes of the brachial plexus: an
anatomical study. Plast Reconstr Surg. 2003;112:17 99–1806.
43. Tulaci KG, Tuzuner A, Karadas Emir H, et al. The effect of tacrolimus on
facial nerve injury: histopathological findings in a rabbit model. Am J
Orolaryngol. 2016;37:393–397.
44. Shahraki M, Mohammadi R, Najafpour A. Influence of Tacrolimus (FK506)
on nerve regeneration using allografts: a rat sciatic nerve model. JOral
Maxillofac Surg. 2015;73:1438 .e1431–1438 .e1439.
45. Coggeshall RE, EmeryDG, Ito H, et al. Unmyelinated and small myelinated
axons in rat ventral roots. J Comp Neurol. 1977;173:175–184.
© 2018 Wolters Kluwer Chang et al 159
Copyright © 2018 Wolters Kluwer Health, Inc. All rights reserved.
