END-TO-SIDE NEUROTIZATION WITH DIFFERENT DONOR NERVES FOR TREATING BRACHIAL PLEXUS INJURY: AN EXPERIMENTAL STUDY IN A RAT MODEL WENGBO YANG, MD,1 JIANYUN YANG, MD,2 CONG YU, MD,2 and YUDONG GU, MD2 1 2
Department of Hand Surgery Nanjing First Hospital, Nanjing, China Department of Hand Surgery, Huashan Hospital, 12 WuLuMuQi Zhong Road, Shanghai, China 200040
Accepted 27 October 2013 ABSTRACT: Introduction: End-to-side neurotization is currently used to treat brachial plexus injury, but it is not clear which donor nerve yields the best outcome. We performed experiments to determine the optimal donor nerve. Methods: A total of 66 male Sprague-Dawley rats were assigned to 1 of 3 groups. Group A was the control group. In Group B, the phrenic nerve was used as the donor, while the ipsilateral C7 nerve root served as the donor in Group C. The epineurial window was used in end-to-side neurorrhaphy. Behavioral observations, histology, electrophysiology, and fluorescence retrotracing were performed postoperatively. Results: Fluorescence retrotracing confirmed nerve regeneration in both Groups B and C upon end-to-side neurotization. The outcome of Group B was superior to that of Group C. Conclusions: Use of the phrenic nerve as the donor nerve yielded a better outcome than use of the ipsilateral C7 nerve root. Muscle Nerve 50: 67–72, 2014
Brachial plexus injury remains a challenge to neurosurgeons. In some cases, the proximal stumps of nerve roots are inaccessible; the only treatment for this type of injury is nerve transfer or free muscle transfer. Advancements in microsurgical techniques have permitted the use of a variety of extraplexus and intraplexus donor nerves. The donor nerves include the phrenic nerve,1 the ipsilateral or contralateral C7 nerve root,2 and the accessory nerve, among others. However, the functions targeted by the donor nerves remain significantly impaired after nerve transfer. In the early 1990s, the technique of end-to-side neurorrhaphy was pioneered by Viterbo et al.3 Following this initial success, many studies have confirmed the effectiveness of this surgical approach in both animal models and clinical practice.4–6 For brachial plexus injury, the phrenic nerve and the C7 nerve root are the Additional Supporting Information may be found in the online version of this article. Abbreviations: CMAP, compound muscle action potential; DRG, dorsal root ganglia Key words: brachial plexus avulsion; end-to-side neurotization; ipsilateral C7 nerve root; phrenic nerve; rat model This research is funded by the Health Ministry of China (No. 2007-66). We received no payments or other benefits and did not make any commitment or agreement to provide such benefits from a commercial entity. No commercial entity paid or directed, or agreed to pay or direct, any benefits to any research fund, foundation, educational institution or other charitable or nonprofit organization with which we are affiliated or associated. Correspondence to: J. Yang; e-mail: [email protected] C 2013 Wiley Periodicals, Inc. V
Published online 6 November 2013 in Wiley Online Library (wileyonlinelibrary. com). DOI 10.1002/mus.24110
End-to-Side Neurotization
most frequently used donor nerves. Many studies have demonstrated the effectiveness of end-to-side neurotization with these donors7–9 but have not compared the outcome of end-to-side neurotization with different donor nerves. The aim of this experiment was to determine which of the 2 donor nerves yielded a superior outcome for the treatment of brachial plexus injury. MATERIALS AND METHODS Animals and Grouping. A total of 66 healthy adult male Sprague-Dawley rats weighing from 230 g to 250 g were assigned randomly to 1 of 3 groups. Group A, the control group, consisted of 18 rats with lesions of the C5 and C6 nerve roots (right side). Group B consisted of 24 rats with lesions of the C5 and C6 nerve roots (right side) followed by end-to-side neurotization of the phrenic nerve to the musculocutaneous nerve. Group C consisted of 24 rats with C5 and C6 nerve root lesions (right side) followed by end-to-side neurotization of the ipsilateral C7 nerve root to the musculocutaneous nerve. All rats in Group A and 18 of 24 rats in Groups B and C were assigned to 3 subgroups of 6 rats each for sacrifice at 4, 8, or 12 weeks after operation. The remaining 6 rats in Groups B and C were used for neuron localization by fluorescence retrotracing. All animals were maintained under standard conditions and given rodent food and water. All surgical procedures were in compliance with Chinese Animal Protection Guidelines. All investigations involving animals were approved by the Animal Research Committee, Shanghai Medical College, Fudan University. Surgery. After intraperitoneal injection of pentobarbital (40 mg/kg body weight), rats were prepped and draped. Surgery was performed on the right side in the supine position. All procedures were performed under a microscope at 103 magnification. The rats in Group A underwent the following procedure. A 2-cm, “L”-shaped cervical incision was made on the right side, and the sternocleidomastoid muscle and clavicle were retracted to expose the brachial plexus. The C5 and C6 nerve roots were cut off at the level of the intervertebral foramen to imitate complete C5 and C6 nerve root MUSCLE & NERVE
July 2014
67
injury. The proximal ends of the transected nerve roots drew back into the intervertebral foramen, and the distal ends were ligated and embedded into the soft tissue nearby to avoid nerve regeneration. The phrenic nerve was identified on the surface of the scalenus anterior muscle and freed distally. Another longitudinal incision was made on the medial side of the upper limb. The musculocutaneous nerve was identified, freed, and transected 10 mm proximal to the point at which it terminated in the biceps brachii. The proximal end was ligated and reflected into the pectoralis major to prevent nerve regeneration. Rats in Group B underwent the same procedure as those in Group A, with the addition of the following procedure. A 2-cm saphenous nerve graft was harvested from the anterior thigh through a 2cm incision. The distal end of the musculocutaneous nerve was sutured to the saphenous nerve graft with end-to-end neurorrhaphy with a 12/0 microstitch. The saphenous nerve graft was placed beneath the pectoralis major muscle, and the other end was pulled into the cervical incision. The epineurium of the phrenic nerve was opened and separated circumferentially to 4 mm in length. The end of the saphenous nerve graft was sutured to the phrenic nerve using the end-to-side technique with a helicoid 12/0 microstitch. The perineurium was not opened. The nerve graft was wrapped around the phrenic nerve with 2 wraps and sutured to the side of the phrenic nerve with 3 stitches. All rats in Group C underwent the same procedure as in Group A, with the following additional procedure. The saphenous nerve graft was harvested and sutured to the musculocutaneous nerve using the same method as in Group B. The saphenous nerve graft was placed beneath the pectoralis major muscle, and the other end was pulled into the cervical incision. The epineurium of the ipsilateral C7 nerve root was opened and separated circumferentially to 4 mm in length. The end of the saphenous nerve graft was sutured to the ipsilateral C7 nerve root using the end-to-side technique with a helicoid 12/0 microstitch. The nerve graft was wrapped around the ipsilateral C7 nerve root with 2 wraps and sutured to the side of the ipsilateral C7 nerve root with 3 stitches. Studies of Nerve Regeneration. Behavioral Assessment. Behavior was assessed before animals were sacrificed at 4, 8, and 12 weeks after the surgical procedure. We applied the Terzis grooming test to assess upper-limb behavior in the rats.10,11 The animals were tested by continuously squirting water over the animal’s snout to elicit a bilateral grooming response to wipe away the water. The function 68
End-to-Side Neurotization
of the upper limb was graded as follows: 0, no response; 1, flexion at elbow, not reaching the snout; 2, flexion reaching the snout; 3, reaching below the eyes; 4, reaching to the eyes; 5, reaching to the ears and beyond. The behavior grades were assessed and recorded by an observer who was blinded to the experimental group. Electrophysiology. After behavioral observation, rats in Groups A, B, and C were anesthetized before electrophysiological evaluation (DantecNeuromatic 2000, Italy). In Group B, the stimulating electrode (Dantec-Neuromatic 2000, Italy) was placed on the phrenic nerve proximal to the endto-side neurorrhaphy site. In Group C, the stimulating electrode was placed on the ipsilateral C7 nerve root proximal to the end-to-side neurorrhaphy site. In Group A, the stimulating electrode was placed on the upper trunk. The recording electrode (Dantec-Neuromatic 2000, Italy) was placed over the mid-point of the right biceps brachii muscle in all groups. An electrical stimulus of 2.5-mA amplitude and 0.04-ms duration was applied, and the latency and maximum amplitude of the compound muscle action potential (CMAP) were recorded. The same recording was repeated on the left side. The latency delay rate and the amplitude recovery rate of the CMAP on the right side were expressed as percentages of those obtained on the left side. Room temperature was maintained at 25 C. Muscle Weights. After electrophysiological examination, the biceps brachii muscles from both sides were detached from the bone at their origin and terminus. The wet muscle weights were measured with an electronic scale with a precision of 0.0001 g. The recovery rate of the wet weight of the right biceps muscle was expressed as a percentage of that on the left side. Muscle Histology. After recording muscle weights, all biceps muscles were fixed in 10% paraformaldehyde and washed in buffer. After dehydration with alcohol, all samples were embedded in paraffin. The muscles were cross-sectioned at 5 lm at the midpoint of the muscle belly. The slices were stained with hematoxylin-eosin and examined under a light microscope (Leica DWLB2, Germany). A computerized image analysis system (QWin histomorphometry system, Leica, Germany) was applied to measure and calculate the mean muscle fiber cross-sectional area. The recovery rate of the cross-sectional area was expressed as a percentage of that on the contralateral side. Nerve Histology and Nerve Fiber Counting. Segments of the musculocutaneous nerve from all groups were excised for histological study and nerve fiber counting. The segments were taken from an area 5 mm distal to the coaptation site. MUSCLE & NERVE
July 2014
FIGURE 1. Grooming Test Grades of the groups at different intervals after operation. The vertical axis shows the mean grooming test scores of each group. The horizontal axis shows the time after the operation.
The specimens were fixed in 2.5% glutaraldehyde buffered with cacodylate, washed in sodium cacodylate buffer (0.2 mol/L, pH 7.4), and postfixed in 1% osmium tetroxide. The specimens were then dehydrated and embedded in Epon. Cross-sections were cut 0.5 lm thick from the middle of the nerve samples and stained with toluidine blue. All cross-sections were examined under a light microscope. The total number of myelinated fibers was calculated for each specimen at a final magnification of 4003. Neuron Localization by Fluorescence Retrotracing. The 12 rats that were designated for neuron localization staining underwent another surgery 12 weeks after the first procedure. The two nerve coaptation sites were again exposed in Groups B and C. The musculocutaneous nerve was transected at a site close to the biceps brachii muscle, and the proximal end was placed in a small plastic bottle filled with 0.2% True Blue (Sigma-Aldrich, St. Louis, Missouri) for 60 min. The phrenic nerve in Group B and the ipsilateral C7 nerve root in Group C were freed and transected 10 mm distal to the coaptation site. The proximal ends were placed into small plastic bottles filled with 0.2% Diamidino Yellow (Sigma-Aldrich) for 60 minutes. At this stage, the ends of the nerves were all ligated and embedded into the proximal soft tissue. Two animals from each group were perfused with 4% paraformaldehyde at 3, 7, or 14 days after the second operation. The C2–T2 dorsal root ganglia (DRG) and corresponding spinal cord were harvested and, after
dehydration, embedded in Neg-50 frozen section medium. Slices with a thickness of 30 lm were made with a freezing microtome (Microm HM505E, Germany) and observed under a fluorescence microscope (Leica DWLB2, Germany) to confirm the origin of the regenerated nerves. We used STATA 7.0 statistical software to compare groups by one-way analysis of variance and least significant difference test. All results are stated as the mean 6 standard deviation; P < 0.05 was considered significant.
Data Analysis.
RESULTS Behavioral Assessment.
No flexion of the elbow was observed in Groups A or C at 4 weeks after the operation. However, flexion of the elbow with the same rhythm as animal respiration was identified in Group B. Slight elbow flexion was identified in Group C at 8 weeks after the operation. The range of elbow flexion increased gradually over time in Groups B and C. There was no return of function in Group A at 12 weeks after the operation. The biceps movement that was in synchrony with respiration in Group B never disappeared as the animal retrained the muscle. At the end of 12 weeks, the flexion function of the elbow in Group B (see Supp. Video S1, which demonstrates the grooming test in Group B—available online) was much better than that in Group C (see Supp. Video S2, which demonstrates the grooming test in Group C—available online). The Grooming Test also showed the same result (Fig. 1), and there was a statistically
Table 1. Amplitude recovery rate and CMAP latency delay of the right biceps muscle at different postoperative intervals (% 6 SD). Amplitude recovery rate of CMAP (% of contralateral side) Groups A B C
CMAP latency delay (% of contralateral side)
4 Week
8 Week
12 Week
4 Week
8 Week
12 Week
0 8.59 6 1.96† 0
0* 24.48 6 7.03 19.09 6 7.07
0* 51.01 6 9.46† 31.22 6 8.01
0 510.91 6 105.82† 0
0* 202.48 6 34.43 206.7437 6 17.07
0* 195.72 6 12.63 195.35 6 18.48
*P < 0.05 between Groups A and B, C. †
P < 0.05 between Groups B and C.
CMAP indicates compound muscle action potential.
End-to-Side Neurotization
MUSCLE & NERVE
July 2014
69
Table 2. Recovery rate of wet muscle weight and muscle fiber cross-sectional area of the right biceps muscle at different postoperative intervals (% 6 SD). Recovery rate of wet weight (% of contralateral side) Groups A B C
Recovery rate of muscle fiber cross-sectional area (% of contralateral side)
4 Week
8 Week
12 Week
4 Week
8 Week
12 Week
41.87 6 6.62 49.52 6 4.72 42.34 6 6.13
31.82 6 2.79 68.93 6 7.88† 47.99 6 8.76
30.67 6 5.32* 71.37 6 4.89† 60.18 6 5.63
67.83 6 2.97* 74.24 6 9.82 72.30 6 6.11
62.86 6 2.39* 90.20 6 6.88 89.66 6 4.32
56.29 6 2.36* 93.51 6 3.87 91.22 6 8.70
*P < 0.05 between Groups A and B, C. †
P < 0.05 between Groups B and C.
significant difference (P < 0.001).
between
the
2
groups
Electrophysiology. No CMAP was recorded in Group A at 12 weeks after the operation. No CMAP was recorded from the right biceps muscle in Group C at 4 weeks after the operation, but a CMAP appeared at 8 weeks after the operation. A CMAP was recordable in group B as early as 4 weeks after the operation. The CMAP values of all groups are listed in Table 1. The CMAP amplitude increased with time in Groups B and C, and the latency of the CMAP gradually shortened. At 12 weeks after the operation, there was a significant difference in the CMAP amplitude between Groups B and C (P 5 0.01) but not the latency (P 5 0.97). Muscle Weight and Muscle Histology. The recovery rate of wet weight of the right biceps muscle and the recovery rate of muscle fiber cross-sectional area across postoperative intervals are shown in Table 2. The recovery rates of muscle weight and muscle fiber cross-sectional area increased gradually with time in Groups B and C. At 12 weeks after the operation, differences were observed between the groups (P < 0.05), with Group B exhibiting the best results. The same tendency in recovery was recorded in muscle fiber cross-sectional area at different times after the operation. Statistically significant differences in the recovery rate of muscle fiber cross-sectional area were observed among Groups B, C, and A (P < 0.05) at 12 weeks postoperation. Although Group B achieved the best recovery, there was no statistically significant difference between Groups B and C (P 5 0.65). Histological analysis of the biceps muscle revealed decreased muscle fiber area and increased fibrosis at 4 weeks postoperation in all 3 groups; Group B displayed the least fibrosis. The muscle fiber area increased in Groups B and C with time but decreased in Group A. Nerve
Histology
and
Nerve
Fiber
Counting.
Histological analysis of the musculocutaneous nerve revealed new nerve fibers at 4 weeks after the operation in Groups B and C. The number and size of the regenerated axons increased with 70
End-to-Side Neurotization
time in both groups. At 12 weeks after the operation, no regenerated nerve fibers were observed in Group A. Although more regenerated nerve fibers were observed in Group B, there was no statistically significant difference between Groups B and C (P 5 0.44) (Table 3). Neuron Localization by Fluorescence Retrotracing.
Some fluorescently labeled neurons were visible 3 days after administration of the tracer. Most stained cells were only labeled with True Blue. In Group B, the positive neurons were centered in the C3–C5 DRG. In Group C, however, the positive neurons were centered at the C7 level. Diamidino Yellow-labeled or double-labeled neurons were visualized 7 days after administration of the tracer. In group B, the neurons single-labeled with True Blue or Diamidino Yellow and the double-labeled neurons appeared in the C3–C5 anterior horn (Fig. 2a and b). However, the double-labeled and single-labeled neurons only appeared in the C7 anterior horn in Group C. The double-labeled neurons were fewer than the Diamidino Yellowlabeled neurons in both groups. In addition, only a few True Blue-labeled neurons are identified. Fourteen days after administration of the tracer, the labeled neurons could not be distinguished clearly due to diffusion of the fluorescence. DISCUSSION
We found that the outcome of the group using the phrenic nerve as the donor was superior across measures compared with that of the group using the ipsilateral C7 nerve root. Nerve transfer is used Table 3. Summary of myelinated axon counts in the musculocutaneous nerve at different postoperative intervals. No. of regenerated nerve fibers (Mean 6 SD) Groups A B C
4 Week
8 Week
12 Week
0* 140.5 6 77.3† 6.83 6 5.8
0* 802.3 6 290.6 583.7 6 187.4
0* 1337.3 6 367.9 1187.3 6 190.7
*P < 0.05 between Groups A and B, C. †
P < 0.05 between Groups B and C.
MUSCLE & NERVE
July 2014
FIGURE 2. (a) Neurons labeled with True Blue or Diamidino Yellow at the level of the C4 anterior horn in Group B. Image was taken 7 days after administration of tracer. The magnification is 4003. (b) Double-labeled neuron (arrow) was identified at the C4 level of the anterior horn in Group B, indicating that the regenerated nerve fiber of the musculocutaneous nerve came from the neuron of the phrenic nerve with end-to-side neurorrhaphy. Image was taken 7 days after administration of tracer. The magnification is 4003.
to treat brachial plexus avulsion or injuries in which the brachial plexus nerve roots rupture close to the vertebral foramina, an injury that prevents nerve regeneration. A challenging problem for nerve transfer is that the number of donor nerve fibers is far fewer than that of injured nerve fibers. End-to-side neurorrhaphy provides a new supply of donor nerve fibers and can preserve function of the donor nerve. Although lateral sprouting is widely accepted as the basis for reinnervation, the underlying mechanism remains unclear. Many factors contribute to the final outcome, including the selection of donor nerves.5,12–14 Mennen reported that the best results were achieved with proximal motor and distal sensory reinnervation.15 Millesi and Schmidhammer showed that the use of pure motor nerves as the donor yielded a better outcome than the use of mixed sensorimotor nerves.16 In addition, the effectiveness of end-to-side neurotization in a helicoid manner for treatment of brachial plexus injury has been confirmed by many authors.8,9,17 Although the number of phrenic nerve fibers is significantly fewer than that of the ipsilateral C7 nerve root, the numbers of the regenerated nerve fibers showed no statistically significant difference between the 2 groups in our experiment. In addition, because the ipsilateral C7 nerve root has mixed motor and sensory fibers, the number of regenerated motor nerve fibers is a percentage of the mixed total reported above for Group C. The number of regenerated nerve fibers suggests that the phrenic nerve is more robust than the ipsilateral C7 nerve root. Two properties are believed to underlie the robust regenerative ability of the phrenic nerve. First, most phrenic nerve fibers are motor fibers, whereas the C7 nerve root has both End-to-Side Neurotization
motor and sensory fibers. Its motor fibers may regenerate into the sensory fibers of the recipient nerve and vice versa. Second, the phrenic nerve gives off spontaneous impulses18,19 which can stimulate nerve regeneration, but the C7 nerve root does not. In our experiment, the transport speed of True Blue was much quicker than that of Diamidino Yellow. Thus, 3 days after administration of the tracers, we only observed True Blue-labeled neurons. However, double-labeled neurons were observed 7 days after administration, and we considered this the best timing to evaluate both tracers. In Group B, the appearance of double-labeled neurons at the C3–C5 level indicates that some phrenic nerve fibers regenerated into the musculocutaneous nerve with end-to-side neurotization. The visualization of double-labeled neurons at the C7 level in Group C demonstrates regeneration of the ipsilateral C7 nerve root with end-to-side neurotization. In our experiment, the fluorescence retrotracing test is only used for qualitative purposes, and a limitation is that we cannot provide accurate numbers of the 3 kinds of labeled neurons. We hypothesize that only some of the donor nerve fibers regenerated into the recipient nerve with end-to-side neurotization. This is the reason that double-labeled neurons are fewer than Diamidino Yellow-labeled neurons. Inevitably, a few donor nerve fibers were injured when we sutured the nerves, leading to a few True Blue-labeled neurons. However, these True Blue-labeled neurons are far fewer than the double-labeled neurons, demonstrating that most of the regenerated nerve fibers come from the end-to-side neurotization. In the behavioral assessments, we determined that elbow flexion in Group B was superior to that MUSCLE & NERVE
July 2014
71
in Group C. In addition to nerve generation itself, stimulation of the biceps muscle is an important factor for regeneration. In Group B, after the phrenic nerve reinnervated the biceps muscle, spontaneous impulses constantly stimulated the muscle. In Group C, the biceps received stimulation and contracted only when the rats performed the functions that were innervated by the C7 nerve root. One confound of the rat model is the difficulty of recapitulating the rehabilitation process as it occurs in humans; it is relatively simple to instruct a human patient to flex the elbow with shoulder adduction, the main rehabilitation method after ipsilateral C7 nerve transfer. Even if this method could be applied in rats, stimulation cannot be achieved 24 h per day. Finally, we note additional limitations of our experiments in animal models, which may differ in the degree and mechanism of injury compared with human patients. In our study, we cut the C5 and C6 nerve root; in a clinical setting, nerve root avulsion and rupture are 2 different types of nerve injury. However, the muscles innervated by the injured nerve root undergo atrophy in both injures. Although our study is based on nerve rupture, the treatment method for avulsion is the same as for nerve root rupture close to the vertebral foramen. In addition, the phrenic nerve and the ipsilateral C7 nerve root could be injured with the C5 and C6 nerve roots. In these patients, end-to-side neurotization using these 2 donor nerves is not possible. The differences in recovery outcome between these 2 donor nerves in humans should be investigated further. REFERENCES 1. Gu YD, Wu MM, Zhen YL, Zhao JA, Zhang GM, Chen DS, et al. Phrenic nerve transfer for brachial plexus motor neurotization. Microsurgery 1989;10:287–289.
72
End-to-Side Neurotization
2. Gu YD, Cai PQ, Xu F, Peng F, Chen L. Clinical application of ipsilateral C7 nerve root transfer for treatment C5 and C6 avulsion of brachial plexus. Microsurgery 2003;23:105–108. 3. Viterbo F, Trindade JC, Hoshino K, Mazzoni Neto A. End-to-side neurorrhaphy with removal of the epineurial sheath: an experimental study in rats. Plast Reconstr Surg 1994;94:1038–1047. 4. Amr SM, Moharram AN. Repair of brachial plexus lesions by end-toside side-to-end grafting neurorrhaphy: experience based on 11 cases. Microsurgery 2005;25:126–146. 5. Lutz BS, Chuang DC, Hsu JC, Ma SF, Wei FC. Selection of donor nerves - an important factor in end-to-side neurorrhphy. Br J Plast Surg 2000;53:149–154. 6. Mennen U, van der Westhuizen MJ, Eggers IM. Re-innervation of M. Biceps by end-to-side nerve suture. Hand Surg 2003;8:25–31. 7. Stamatoukou A, Papadogeorgou E, Zhang Z, Pavlakis K, Zoubos AB, Soucacos PN. Phrenic nerve neurotization of the musculocutaneous nerve with end-to-side neurorrhaphy: a short report in a rabbit model. Microsurgery 2006;26:268–272. 8. Wang M, Xu W, Zheng M, Teng F, Xu J, Gu Y. Phrenic nerve end-toside neurotization in treating brachial plexus avulsion: an experimental study in rats. Ann Plast Surg 2011;66:370–376. 9. Yan YH, Yan JG, Matloub HS, Zhang LL, Hettinger P, Sanger J, et al. Helicoid end-to-side and oblique attachment technique in repair of the musculocutaneous nerve injury with the phrenic nerve as a donor: an experimental study in rats. Microsurgery 2011;31:122–129. 10. Bertelli JA, Mira JC. Behavioral evaluating methods in the objective clinical assessment of motor function after experimental brachial plexus reconstructionin the rat. J Neurosci Methods 1993;46:203–208. 11. Inciong JG, Marrocco WC, Terzis JK. Efficacy of intervention strategies in a brachial plexus globalavulsion model in the rat. Plast Reconstr Surg 2000;105:2059–2071. 12. McCallister WV, Tang P, Smith J, Trumble TE. Axonal regeneration stimulated by the combination of nerve growth factor and ciliary neurotrophic factor in an end-to-side model. J Hand Surg Am 2001;26: 478–488. 13. Schmidhammer R, Redl H, Hopf R, van der Nest DG, Millesi H. Synergistic terminal motor end-to-side nerve graft repair: investigation in a non-human primate model. Acta Neurochir Suppl 2007;100:97–101. 14. Walker JC, Brenner MJ, Mackinnon SE, Winograd JM, Hunter DA. Effect of perineurial window size on nerve regeneration, blood-nerve barrier integrity, and functional recovery. J Neurotrauma 2004;21:217–227. 15. Mennen U. End-to-side suture in clinical practice. Hand Surg 2003;8: 33–42. 16. Millesi H, Schmidhammer R. End-to-side coaptation–controversial research issue or important tool in human patients. Acta Neurochir Suppl 2007;100:103–106. 17. Yan JG, Matloub HS, Sanger JR, Zhang LL, Riley DA, Jaradeh SS. A modified end-to-side method for peripheral nerve repair: large epineurial window helicoid technique versus small epineurial window standard end-to-side technique. J Hand Surg Am 2002;27:484–492. 18. Gu YD, Ma MK. Use of the phrenic nerve for brachial plexus reconstruction. Clin Orthop Relat Res 1996;323:119–121. 19. Zhang CG, Ma JJ, Terenghi G, Mantovani C, Wiberg M. Phrenic nerve transfer in the treatment of brachial plexus avulsion: an experimental study of nerve regeneration and muscle morphology in rats. Microsurgery 2004;24:232–240.
MUSCLE & NERVE
July 2014
Department of Hand Surgery Nanjing First Hospital, Nanjing, China Department of Hand Surgery, Huashan Hospital, 12 WuLuMuQi Zhong Road, Shanghai, China 200040
Accepted 27 October 2013 ABSTRACT: Introduction: End-to-side neurotization is currently used to treat brachial plexus injury, but it is not clear which donor nerve yields the best outcome. We performed experiments to determine the optimal donor nerve. Methods: A total of 66 male Sprague-Dawley rats were assigned to 1 of 3 groups. Group A was the control group. In Group B, the phrenic nerve was used as the donor, while the ipsilateral C7 nerve root served as the donor in Group C. The epineurial window was used in end-to-side neurorrhaphy. Behavioral observations, histology, electrophysiology, and fluorescence retrotracing were performed postoperatively. Results: Fluorescence retrotracing confirmed nerve regeneration in both Groups B and C upon end-to-side neurotization. The outcome of Group B was superior to that of Group C. Conclusions: Use of the phrenic nerve as the donor nerve yielded a better outcome than use of the ipsilateral C7 nerve root. Muscle Nerve 50: 67–72, 2014
Brachial plexus injury remains a challenge to neurosurgeons. In some cases, the proximal stumps of nerve roots are inaccessible; the only treatment for this type of injury is nerve transfer or free muscle transfer. Advancements in microsurgical techniques have permitted the use of a variety of extraplexus and intraplexus donor nerves. The donor nerves include the phrenic nerve,1 the ipsilateral or contralateral C7 nerve root,2 and the accessory nerve, among others. However, the functions targeted by the donor nerves remain significantly impaired after nerve transfer. In the early 1990s, the technique of end-to-side neurorrhaphy was pioneered by Viterbo et al.3 Following this initial success, many studies have confirmed the effectiveness of this surgical approach in both animal models and clinical practice.4–6 For brachial plexus injury, the phrenic nerve and the C7 nerve root are the Additional Supporting Information may be found in the online version of this article. Abbreviations: CMAP, compound muscle action potential; DRG, dorsal root ganglia Key words: brachial plexus avulsion; end-to-side neurotization; ipsilateral C7 nerve root; phrenic nerve; rat model This research is funded by the Health Ministry of China (No. 2007-66). We received no payments or other benefits and did not make any commitment or agreement to provide such benefits from a commercial entity. No commercial entity paid or directed, or agreed to pay or direct, any benefits to any research fund, foundation, educational institution or other charitable or nonprofit organization with which we are affiliated or associated. Correspondence to: J. Yang; e-mail: [email protected] C 2013 Wiley Periodicals, Inc. V
Published online 6 November 2013 in Wiley Online Library (wileyonlinelibrary. com). DOI 10.1002/mus.24110
End-to-Side Neurotization
most frequently used donor nerves. Many studies have demonstrated the effectiveness of end-to-side neurotization with these donors7–9 but have not compared the outcome of end-to-side neurotization with different donor nerves. The aim of this experiment was to determine which of the 2 donor nerves yielded a superior outcome for the treatment of brachial plexus injury. MATERIALS AND METHODS Animals and Grouping. A total of 66 healthy adult male Sprague-Dawley rats weighing from 230 g to 250 g were assigned randomly to 1 of 3 groups. Group A, the control group, consisted of 18 rats with lesions of the C5 and C6 nerve roots (right side). Group B consisted of 24 rats with lesions of the C5 and C6 nerve roots (right side) followed by end-to-side neurotization of the phrenic nerve to the musculocutaneous nerve. Group C consisted of 24 rats with C5 and C6 nerve root lesions (right side) followed by end-to-side neurotization of the ipsilateral C7 nerve root to the musculocutaneous nerve. All rats in Group A and 18 of 24 rats in Groups B and C were assigned to 3 subgroups of 6 rats each for sacrifice at 4, 8, or 12 weeks after operation. The remaining 6 rats in Groups B and C were used for neuron localization by fluorescence retrotracing. All animals were maintained under standard conditions and given rodent food and water. All surgical procedures were in compliance with Chinese Animal Protection Guidelines. All investigations involving animals were approved by the Animal Research Committee, Shanghai Medical College, Fudan University. Surgery. After intraperitoneal injection of pentobarbital (40 mg/kg body weight), rats were prepped and draped. Surgery was performed on the right side in the supine position. All procedures were performed under a microscope at 103 magnification. The rats in Group A underwent the following procedure. A 2-cm, “L”-shaped cervical incision was made on the right side, and the sternocleidomastoid muscle and clavicle were retracted to expose the brachial plexus. The C5 and C6 nerve roots were cut off at the level of the intervertebral foramen to imitate complete C5 and C6 nerve root MUSCLE & NERVE
July 2014
67
injury. The proximal ends of the transected nerve roots drew back into the intervertebral foramen, and the distal ends were ligated and embedded into the soft tissue nearby to avoid nerve regeneration. The phrenic nerve was identified on the surface of the scalenus anterior muscle and freed distally. Another longitudinal incision was made on the medial side of the upper limb. The musculocutaneous nerve was identified, freed, and transected 10 mm proximal to the point at which it terminated in the biceps brachii. The proximal end was ligated and reflected into the pectoralis major to prevent nerve regeneration. Rats in Group B underwent the same procedure as those in Group A, with the addition of the following procedure. A 2-cm saphenous nerve graft was harvested from the anterior thigh through a 2cm incision. The distal end of the musculocutaneous nerve was sutured to the saphenous nerve graft with end-to-end neurorrhaphy with a 12/0 microstitch. The saphenous nerve graft was placed beneath the pectoralis major muscle, and the other end was pulled into the cervical incision. The epineurium of the phrenic nerve was opened and separated circumferentially to 4 mm in length. The end of the saphenous nerve graft was sutured to the phrenic nerve using the end-to-side technique with a helicoid 12/0 microstitch. The perineurium was not opened. The nerve graft was wrapped around the phrenic nerve with 2 wraps and sutured to the side of the phrenic nerve with 3 stitches. All rats in Group C underwent the same procedure as in Group A, with the following additional procedure. The saphenous nerve graft was harvested and sutured to the musculocutaneous nerve using the same method as in Group B. The saphenous nerve graft was placed beneath the pectoralis major muscle, and the other end was pulled into the cervical incision. The epineurium of the ipsilateral C7 nerve root was opened and separated circumferentially to 4 mm in length. The end of the saphenous nerve graft was sutured to the ipsilateral C7 nerve root using the end-to-side technique with a helicoid 12/0 microstitch. The nerve graft was wrapped around the ipsilateral C7 nerve root with 2 wraps and sutured to the side of the ipsilateral C7 nerve root with 3 stitches. Studies of Nerve Regeneration. Behavioral Assessment. Behavior was assessed before animals were sacrificed at 4, 8, and 12 weeks after the surgical procedure. We applied the Terzis grooming test to assess upper-limb behavior in the rats.10,11 The animals were tested by continuously squirting water over the animal’s snout to elicit a bilateral grooming response to wipe away the water. The function 68
End-to-Side Neurotization
of the upper limb was graded as follows: 0, no response; 1, flexion at elbow, not reaching the snout; 2, flexion reaching the snout; 3, reaching below the eyes; 4, reaching to the eyes; 5, reaching to the ears and beyond. The behavior grades were assessed and recorded by an observer who was blinded to the experimental group. Electrophysiology. After behavioral observation, rats in Groups A, B, and C were anesthetized before electrophysiological evaluation (DantecNeuromatic 2000, Italy). In Group B, the stimulating electrode (Dantec-Neuromatic 2000, Italy) was placed on the phrenic nerve proximal to the endto-side neurorrhaphy site. In Group C, the stimulating electrode was placed on the ipsilateral C7 nerve root proximal to the end-to-side neurorrhaphy site. In Group A, the stimulating electrode was placed on the upper trunk. The recording electrode (Dantec-Neuromatic 2000, Italy) was placed over the mid-point of the right biceps brachii muscle in all groups. An electrical stimulus of 2.5-mA amplitude and 0.04-ms duration was applied, and the latency and maximum amplitude of the compound muscle action potential (CMAP) were recorded. The same recording was repeated on the left side. The latency delay rate and the amplitude recovery rate of the CMAP on the right side were expressed as percentages of those obtained on the left side. Room temperature was maintained at 25 C. Muscle Weights. After electrophysiological examination, the biceps brachii muscles from both sides were detached from the bone at their origin and terminus. The wet muscle weights were measured with an electronic scale with a precision of 0.0001 g. The recovery rate of the wet weight of the right biceps muscle was expressed as a percentage of that on the left side. Muscle Histology. After recording muscle weights, all biceps muscles were fixed in 10% paraformaldehyde and washed in buffer. After dehydration with alcohol, all samples were embedded in paraffin. The muscles were cross-sectioned at 5 lm at the midpoint of the muscle belly. The slices were stained with hematoxylin-eosin and examined under a light microscope (Leica DWLB2, Germany). A computerized image analysis system (QWin histomorphometry system, Leica, Germany) was applied to measure and calculate the mean muscle fiber cross-sectional area. The recovery rate of the cross-sectional area was expressed as a percentage of that on the contralateral side. Nerve Histology and Nerve Fiber Counting. Segments of the musculocutaneous nerve from all groups were excised for histological study and nerve fiber counting. The segments were taken from an area 5 mm distal to the coaptation site. MUSCLE & NERVE
July 2014
FIGURE 1. Grooming Test Grades of the groups at different intervals after operation. The vertical axis shows the mean grooming test scores of each group. The horizontal axis shows the time after the operation.
The specimens were fixed in 2.5% glutaraldehyde buffered with cacodylate, washed in sodium cacodylate buffer (0.2 mol/L, pH 7.4), and postfixed in 1% osmium tetroxide. The specimens were then dehydrated and embedded in Epon. Cross-sections were cut 0.5 lm thick from the middle of the nerve samples and stained with toluidine blue. All cross-sections were examined under a light microscope. The total number of myelinated fibers was calculated for each specimen at a final magnification of 4003. Neuron Localization by Fluorescence Retrotracing. The 12 rats that were designated for neuron localization staining underwent another surgery 12 weeks after the first procedure. The two nerve coaptation sites were again exposed in Groups B and C. The musculocutaneous nerve was transected at a site close to the biceps brachii muscle, and the proximal end was placed in a small plastic bottle filled with 0.2% True Blue (Sigma-Aldrich, St. Louis, Missouri) for 60 min. The phrenic nerve in Group B and the ipsilateral C7 nerve root in Group C were freed and transected 10 mm distal to the coaptation site. The proximal ends were placed into small plastic bottles filled with 0.2% Diamidino Yellow (Sigma-Aldrich) for 60 minutes. At this stage, the ends of the nerves were all ligated and embedded into the proximal soft tissue. Two animals from each group were perfused with 4% paraformaldehyde at 3, 7, or 14 days after the second operation. The C2–T2 dorsal root ganglia (DRG) and corresponding spinal cord were harvested and, after
dehydration, embedded in Neg-50 frozen section medium. Slices with a thickness of 30 lm were made with a freezing microtome (Microm HM505E, Germany) and observed under a fluorescence microscope (Leica DWLB2, Germany) to confirm the origin of the regenerated nerves. We used STATA 7.0 statistical software to compare groups by one-way analysis of variance and least significant difference test. All results are stated as the mean 6 standard deviation; P < 0.05 was considered significant.
Data Analysis.
RESULTS Behavioral Assessment.
No flexion of the elbow was observed in Groups A or C at 4 weeks after the operation. However, flexion of the elbow with the same rhythm as animal respiration was identified in Group B. Slight elbow flexion was identified in Group C at 8 weeks after the operation. The range of elbow flexion increased gradually over time in Groups B and C. There was no return of function in Group A at 12 weeks after the operation. The biceps movement that was in synchrony with respiration in Group B never disappeared as the animal retrained the muscle. At the end of 12 weeks, the flexion function of the elbow in Group B (see Supp. Video S1, which demonstrates the grooming test in Group B—available online) was much better than that in Group C (see Supp. Video S2, which demonstrates the grooming test in Group C—available online). The Grooming Test also showed the same result (Fig. 1), and there was a statistically
Table 1. Amplitude recovery rate and CMAP latency delay of the right biceps muscle at different postoperative intervals (% 6 SD). Amplitude recovery rate of CMAP (% of contralateral side) Groups A B C
CMAP latency delay (% of contralateral side)
4 Week
8 Week
12 Week
4 Week
8 Week
12 Week
0 8.59 6 1.96† 0
0* 24.48 6 7.03 19.09 6 7.07
0* 51.01 6 9.46† 31.22 6 8.01
0 510.91 6 105.82† 0
0* 202.48 6 34.43 206.7437 6 17.07
0* 195.72 6 12.63 195.35 6 18.48
*P < 0.05 between Groups A and B, C. †
P < 0.05 between Groups B and C.
CMAP indicates compound muscle action potential.
End-to-Side Neurotization
MUSCLE & NERVE
July 2014
69
Table 2. Recovery rate of wet muscle weight and muscle fiber cross-sectional area of the right biceps muscle at different postoperative intervals (% 6 SD). Recovery rate of wet weight (% of contralateral side) Groups A B C
Recovery rate of muscle fiber cross-sectional area (% of contralateral side)
4 Week
8 Week
12 Week
4 Week
8 Week
12 Week
41.87 6 6.62 49.52 6 4.72 42.34 6 6.13
31.82 6 2.79 68.93 6 7.88† 47.99 6 8.76
30.67 6 5.32* 71.37 6 4.89† 60.18 6 5.63
67.83 6 2.97* 74.24 6 9.82 72.30 6 6.11
62.86 6 2.39* 90.20 6 6.88 89.66 6 4.32
56.29 6 2.36* 93.51 6 3.87 91.22 6 8.70
*P < 0.05 between Groups A and B, C. †
P < 0.05 between Groups B and C.
significant difference (P < 0.001).
between
the
2
groups
Electrophysiology. No CMAP was recorded in Group A at 12 weeks after the operation. No CMAP was recorded from the right biceps muscle in Group C at 4 weeks after the operation, but a CMAP appeared at 8 weeks after the operation. A CMAP was recordable in group B as early as 4 weeks after the operation. The CMAP values of all groups are listed in Table 1. The CMAP amplitude increased with time in Groups B and C, and the latency of the CMAP gradually shortened. At 12 weeks after the operation, there was a significant difference in the CMAP amplitude between Groups B and C (P 5 0.01) but not the latency (P 5 0.97). Muscle Weight and Muscle Histology. The recovery rate of wet weight of the right biceps muscle and the recovery rate of muscle fiber cross-sectional area across postoperative intervals are shown in Table 2. The recovery rates of muscle weight and muscle fiber cross-sectional area increased gradually with time in Groups B and C. At 12 weeks after the operation, differences were observed between the groups (P < 0.05), with Group B exhibiting the best results. The same tendency in recovery was recorded in muscle fiber cross-sectional area at different times after the operation. Statistically significant differences in the recovery rate of muscle fiber cross-sectional area were observed among Groups B, C, and A (P < 0.05) at 12 weeks postoperation. Although Group B achieved the best recovery, there was no statistically significant difference between Groups B and C (P 5 0.65). Histological analysis of the biceps muscle revealed decreased muscle fiber area and increased fibrosis at 4 weeks postoperation in all 3 groups; Group B displayed the least fibrosis. The muscle fiber area increased in Groups B and C with time but decreased in Group A. Nerve
Histology
and
Nerve
Fiber
Counting.
Histological analysis of the musculocutaneous nerve revealed new nerve fibers at 4 weeks after the operation in Groups B and C. The number and size of the regenerated axons increased with 70
End-to-Side Neurotization
time in both groups. At 12 weeks after the operation, no regenerated nerve fibers were observed in Group A. Although more regenerated nerve fibers were observed in Group B, there was no statistically significant difference between Groups B and C (P 5 0.44) (Table 3). Neuron Localization by Fluorescence Retrotracing.
Some fluorescently labeled neurons were visible 3 days after administration of the tracer. Most stained cells were only labeled with True Blue. In Group B, the positive neurons were centered in the C3–C5 DRG. In Group C, however, the positive neurons were centered at the C7 level. Diamidino Yellow-labeled or double-labeled neurons were visualized 7 days after administration of the tracer. In group B, the neurons single-labeled with True Blue or Diamidino Yellow and the double-labeled neurons appeared in the C3–C5 anterior horn (Fig. 2a and b). However, the double-labeled and single-labeled neurons only appeared in the C7 anterior horn in Group C. The double-labeled neurons were fewer than the Diamidino Yellowlabeled neurons in both groups. In addition, only a few True Blue-labeled neurons are identified. Fourteen days after administration of the tracer, the labeled neurons could not be distinguished clearly due to diffusion of the fluorescence. DISCUSSION
We found that the outcome of the group using the phrenic nerve as the donor was superior across measures compared with that of the group using the ipsilateral C7 nerve root. Nerve transfer is used Table 3. Summary of myelinated axon counts in the musculocutaneous nerve at different postoperative intervals. No. of regenerated nerve fibers (Mean 6 SD) Groups A B C
4 Week
8 Week
12 Week
0* 140.5 6 77.3† 6.83 6 5.8
0* 802.3 6 290.6 583.7 6 187.4
0* 1337.3 6 367.9 1187.3 6 190.7
*P < 0.05 between Groups A and B, C. †
P < 0.05 between Groups B and C.
MUSCLE & NERVE
July 2014
FIGURE 2. (a) Neurons labeled with True Blue or Diamidino Yellow at the level of the C4 anterior horn in Group B. Image was taken 7 days after administration of tracer. The magnification is 4003. (b) Double-labeled neuron (arrow) was identified at the C4 level of the anterior horn in Group B, indicating that the regenerated nerve fiber of the musculocutaneous nerve came from the neuron of the phrenic nerve with end-to-side neurorrhaphy. Image was taken 7 days after administration of tracer. The magnification is 4003.
to treat brachial plexus avulsion or injuries in which the brachial plexus nerve roots rupture close to the vertebral foramina, an injury that prevents nerve regeneration. A challenging problem for nerve transfer is that the number of donor nerve fibers is far fewer than that of injured nerve fibers. End-to-side neurorrhaphy provides a new supply of donor nerve fibers and can preserve function of the donor nerve. Although lateral sprouting is widely accepted as the basis for reinnervation, the underlying mechanism remains unclear. Many factors contribute to the final outcome, including the selection of donor nerves.5,12–14 Mennen reported that the best results were achieved with proximal motor and distal sensory reinnervation.15 Millesi and Schmidhammer showed that the use of pure motor nerves as the donor yielded a better outcome than the use of mixed sensorimotor nerves.16 In addition, the effectiveness of end-to-side neurotization in a helicoid manner for treatment of brachial plexus injury has been confirmed by many authors.8,9,17 Although the number of phrenic nerve fibers is significantly fewer than that of the ipsilateral C7 nerve root, the numbers of the regenerated nerve fibers showed no statistically significant difference between the 2 groups in our experiment. In addition, because the ipsilateral C7 nerve root has mixed motor and sensory fibers, the number of regenerated motor nerve fibers is a percentage of the mixed total reported above for Group C. The number of regenerated nerve fibers suggests that the phrenic nerve is more robust than the ipsilateral C7 nerve root. Two properties are believed to underlie the robust regenerative ability of the phrenic nerve. First, most phrenic nerve fibers are motor fibers, whereas the C7 nerve root has both End-to-Side Neurotization
motor and sensory fibers. Its motor fibers may regenerate into the sensory fibers of the recipient nerve and vice versa. Second, the phrenic nerve gives off spontaneous impulses18,19 which can stimulate nerve regeneration, but the C7 nerve root does not. In our experiment, the transport speed of True Blue was much quicker than that of Diamidino Yellow. Thus, 3 days after administration of the tracers, we only observed True Blue-labeled neurons. However, double-labeled neurons were observed 7 days after administration, and we considered this the best timing to evaluate both tracers. In Group B, the appearance of double-labeled neurons at the C3–C5 level indicates that some phrenic nerve fibers regenerated into the musculocutaneous nerve with end-to-side neurotization. The visualization of double-labeled neurons at the C7 level in Group C demonstrates regeneration of the ipsilateral C7 nerve root with end-to-side neurotization. In our experiment, the fluorescence retrotracing test is only used for qualitative purposes, and a limitation is that we cannot provide accurate numbers of the 3 kinds of labeled neurons. We hypothesize that only some of the donor nerve fibers regenerated into the recipient nerve with end-to-side neurotization. This is the reason that double-labeled neurons are fewer than Diamidino Yellow-labeled neurons. Inevitably, a few donor nerve fibers were injured when we sutured the nerves, leading to a few True Blue-labeled neurons. However, these True Blue-labeled neurons are far fewer than the double-labeled neurons, demonstrating that most of the regenerated nerve fibers come from the end-to-side neurotization. In the behavioral assessments, we determined that elbow flexion in Group B was superior to that MUSCLE & NERVE
July 2014
71
in Group C. In addition to nerve generation itself, stimulation of the biceps muscle is an important factor for regeneration. In Group B, after the phrenic nerve reinnervated the biceps muscle, spontaneous impulses constantly stimulated the muscle. In Group C, the biceps received stimulation and contracted only when the rats performed the functions that were innervated by the C7 nerve root. One confound of the rat model is the difficulty of recapitulating the rehabilitation process as it occurs in humans; it is relatively simple to instruct a human patient to flex the elbow with shoulder adduction, the main rehabilitation method after ipsilateral C7 nerve transfer. Even if this method could be applied in rats, stimulation cannot be achieved 24 h per day. Finally, we note additional limitations of our experiments in animal models, which may differ in the degree and mechanism of injury compared with human patients. In our study, we cut the C5 and C6 nerve root; in a clinical setting, nerve root avulsion and rupture are 2 different types of nerve injury. However, the muscles innervated by the injured nerve root undergo atrophy in both injures. Although our study is based on nerve rupture, the treatment method for avulsion is the same as for nerve root rupture close to the vertebral foramen. In addition, the phrenic nerve and the ipsilateral C7 nerve root could be injured with the C5 and C6 nerve roots. In these patients, end-to-side neurotization using these 2 donor nerves is not possible. The differences in recovery outcome between these 2 donor nerves in humans should be investigated further. REFERENCES 1. Gu YD, Wu MM, Zhen YL, Zhao JA, Zhang GM, Chen DS, et al. Phrenic nerve transfer for brachial plexus motor neurotization. Microsurgery 1989;10:287–289.
72
End-to-Side Neurotization
2. Gu YD, Cai PQ, Xu F, Peng F, Chen L. Clinical application of ipsilateral C7 nerve root transfer for treatment C5 and C6 avulsion of brachial plexus. Microsurgery 2003;23:105–108. 3. Viterbo F, Trindade JC, Hoshino K, Mazzoni Neto A. End-to-side neurorrhaphy with removal of the epineurial sheath: an experimental study in rats. Plast Reconstr Surg 1994;94:1038–1047. 4. Amr SM, Moharram AN. Repair of brachial plexus lesions by end-toside side-to-end grafting neurorrhaphy: experience based on 11 cases. Microsurgery 2005;25:126–146. 5. Lutz BS, Chuang DC, Hsu JC, Ma SF, Wei FC. Selection of donor nerves - an important factor in end-to-side neurorrhphy. Br J Plast Surg 2000;53:149–154. 6. Mennen U, van der Westhuizen MJ, Eggers IM. Re-innervation of M. Biceps by end-to-side nerve suture. Hand Surg 2003;8:25–31. 7. Stamatoukou A, Papadogeorgou E, Zhang Z, Pavlakis K, Zoubos AB, Soucacos PN. Phrenic nerve neurotization of the musculocutaneous nerve with end-to-side neurorrhaphy: a short report in a rabbit model. Microsurgery 2006;26:268–272. 8. Wang M, Xu W, Zheng M, Teng F, Xu J, Gu Y. Phrenic nerve end-toside neurotization in treating brachial plexus avulsion: an experimental study in rats. Ann Plast Surg 2011;66:370–376. 9. Yan YH, Yan JG, Matloub HS, Zhang LL, Hettinger P, Sanger J, et al. Helicoid end-to-side and oblique attachment technique in repair of the musculocutaneous nerve injury with the phrenic nerve as a donor: an experimental study in rats. Microsurgery 2011;31:122–129. 10. Bertelli JA, Mira JC. Behavioral evaluating methods in the objective clinical assessment of motor function after experimental brachial plexus reconstructionin the rat. J Neurosci Methods 1993;46:203–208. 11. Inciong JG, Marrocco WC, Terzis JK. Efficacy of intervention strategies in a brachial plexus globalavulsion model in the rat. Plast Reconstr Surg 2000;105:2059–2071. 12. McCallister WV, Tang P, Smith J, Trumble TE. Axonal regeneration stimulated by the combination of nerve growth factor and ciliary neurotrophic factor in an end-to-side model. J Hand Surg Am 2001;26: 478–488. 13. Schmidhammer R, Redl H, Hopf R, van der Nest DG, Millesi H. Synergistic terminal motor end-to-side nerve graft repair: investigation in a non-human primate model. Acta Neurochir Suppl 2007;100:97–101. 14. Walker JC, Brenner MJ, Mackinnon SE, Winograd JM, Hunter DA. Effect of perineurial window size on nerve regeneration, blood-nerve barrier integrity, and functional recovery. J Neurotrauma 2004;21:217–227. 15. Mennen U. End-to-side suture in clinical practice. Hand Surg 2003;8: 33–42. 16. Millesi H, Schmidhammer R. End-to-side coaptation–controversial research issue or important tool in human patients. Acta Neurochir Suppl 2007;100:103–106. 17. Yan JG, Matloub HS, Sanger JR, Zhang LL, Riley DA, Jaradeh SS. A modified end-to-side method for peripheral nerve repair: large epineurial window helicoid technique versus small epineurial window standard end-to-side technique. J Hand Surg Am 2002;27:484–492. 18. Gu YD, Ma MK. Use of the phrenic nerve for brachial plexus reconstruction. Clin Orthop Relat Res 1996;323:119–121. 19. Zhang CG, Ma JJ, Terenghi G, Mantovani C, Wiberg M. Phrenic nerve transfer in the treatment of brachial plexus avulsion: an experimental study of nerve regeneration and muscle morphology in rats. Microsurgery 2004;24:232–240.
MUSCLE & NERVE
July 2014
