Phrenic Nerve Transfer for Elbow Flexion and Intercostal Nerve Transfer for Elbow Extension

SCIENTIFIC ARTICLE

Phrenic Nerve Transfer for Elbow Flexion and Intercostal Nerve Transfer for Elbow Extension Mou-Xiong Zheng, MD, Wen-Dong Xu, MD, PhD, Yan-Qun Qiu, MD, Jian-Guang Xu, PhD, Yu-Dong Gu, PhD

Purpose To explore long-term recovery of elbow flexion and extension after transferring the phrenic nerve and intercostal nerves, respectively, in adults with global brachial plexus avulsion injuries. Methods Seven adults with global brachial plexus avulsion injuries had the phrenic nerve transferred to the musculocutaneous nerve (or to the anterior division of upper trunk) and intercostal nerves transferred to the triceps branch of the radial nerve at our hospital 7 to 12 years ago. The results of elbow motor strength testing using the Medical Research Council grading scale, and electrodiagnostic findings using electromyogram examinations, were studied retrospectively. Pulmonary function tests were also performed at final visits. Results Functional elbow flexion was obtained in most of the 7 cases (M2, 1; M3, 3; M4, 2; and M5, 1) but elbow extension was absent or insufficient in all subjects (M0, 1; M1, 3; and M2, 3). Electrical results showed successful biceps reinnervation in 6 patients and successful triceps reinnervation in 5. No patient experienced breathing problems, and pulmonary function results were within normal range. Conclusions In the long term, after brachial plexus avulsion injury in most patients who underwent both phrenic nerve and intercostal nerve transfer to achieve elbow flexion and extension eventually obtained satisfactory elbow flexion but poor elbow extension. We recommend against transferring the intercostal nerves to the triceps branch of radial nerve in conjunction with primary phrenic to musculocutaneous nerve transfer. (J Hand Surg 2010; 35A:1304–1309. Copyright © 2010 by the American Society for Surgery of the Hand. All rights reserved.) Type of study/level of evidence Therapeutic IV. Key words Brachial plexus injury, elbow function, long term, nerve transfer.

FromtheDepartmentofHandSurgery,Hua-ShanHospital,ShanghaiMedicalCollege;theInstituteof Hand Surgery; and the State Key Laboratory of Medical Neuroscience, Fudan University, Shanghai, China. Received for publication December 1, 2009; accepted in revised form April 5, 2010. Supported by the Chinese National Basic Research Program (Grant 2003CB515300); the “Dawn” Program of the Shanghai Education Commission, China (Grant 06SG04); and the Program for New Century Excellent Talents in University, China (Grant NCET-07-0209). No benefits in any form have been received or will be received related directly or indirectly to the subject of this article. Corresponding author: Wen-Dong Xu, MD, PhD, Department of Hand Surgery, Hua-Shan Hospital, Shanghai Medical College, Fudan University, 12 Wulumuqi Middle Road, Shanghai 200040, China; e-mail: [email protected]. 0363-5023/10/35A08-0014$36.00/0 doi:10.1016/j.jhsa.2010.04.006

1304 䉬 ©  ASSH 䉬 Published by Elsevier, Inc. All rights reserved.

avulsion injury (BPAI) is a devastating trauma of the upper extremity with prolonged recuperation and poor prognosis for injured young adults. Restoration of important proximal limb functions such as shoulder abduction and elbow flexion using nerve transfer and nerve grafting techniques provides satisfactory recovery in global BPAI patients.1 However, hand function is far less satisfactory, and most global BPAI patients are left with a permanent impairment.2,3 Thus, surgeons have been seeking advanced surgical techniques to recover coordinated and complex upper extremity functions.4 Active elbow extension is important in stabilizing the elbow so that patients can move their elbow without

G

LOBAL BRACHIAL PLEXUS

1305

AN-SSN; CC7MN; ICN-TCN AN-SSN; CC7MN; ICN-TCN AN-SSN; CC7MN; ICN-TCN AN-SSN; CC7-MN

AN-SSN; CC7-MN

AN-SSN; CC7-MN

AN-SSN; CC7-MN

20 7

JHS 䉬 Vol A, August 

SN, sural nerve; SRN, superficial radial nerve; AN, accessory nerve; SSN, suprascapular nerve; CC7, contralateral C7; MN, medium nerve.

12.1 2.3 Fracture of humerus Traction injury by machine 54 161 Right

20 6

Right

9.7 4.0 None Car accident 85 178 Right

20 5

Left

9.4 3.0 None Car accident 85 178 Right

Motorcycle accident 75 180 Right Left 22 4

Right

9.0 3.0

8.8 1.9

Fracture of tibia and fibula Fracture of clavicle Motorcycle accident 63 168 Right 23 3

Left

39 2

Right

Right

170

78

Motorcycle accident

Fracture of ulna

1.5

7.7

I: PN-ADUT; II: ICN-TBRN I: PN-MCN(SN); II: ICN-TBRN I: PN-ADUT; II: ICN-LHT I: PN-MCN(RSN); II: ICN-TBRN I: PN-MCN(SN); II: ICN-TBRN I: PN-MCN(SN); II: ICN-TBRN I: PN-MCN(RSN); II: ICN-TBRN 7.3 4.6 None Motorcycle accident 65 167 Right 34 1

Right

Other Surgeries Surgical Procedures Follow-Up Duration (y) Delay Before Surgery (mo) Concomitant Injuries Injury Mechanism Weight (kg) Height (cm) Dominant Hand Injured Side Current Age

MATERIALS AND METHODS Patients We recruited 7 right-handed men with global BPAI for a retrospective follow-up study. They were initially diagnosed with BPAI with absent active elbow flexion and extension, and underwent surgery in our hospital between 1997 and 2001. All patients underwent 2-stage nerve transfer surgery to restore elbow flexion and extension. Demographic data were as follows: mean age (⫾ standard deviation [SD]) at the time of injury was 26 ⫾ 8 years (median, 22 y); mean interval (⫾SD) between trauma and first surgery was 2.8 ⫾ 1 months (median, 3 mo); and mean duration (⫾SD) between the first surgery and the last visit was 9 ⫾ 2 years (median, 8 y). Table 1 lists the characteristics of these patients, including height, weight, injury mechanisms, dominant hand, and surgery delay. In each patient, the PN was transferred to the musculocutaneous nerve (MCN) or anterior division of up-

Patient

the help of the contralateral hand, and elbow stability improves the results of grasp reconstruction surgery.5–7 Therefore, increasing attention has been paid to triceps recovery. Recent techniques for elbow extension involve nerve transfer and free muscle transfer.5,8 In previous studies, authors have suggested that neurotization of both triceps and biceps with intercostal nerves (ICNs) would yield crippling co-contraction in adults, and thus should be avoided.6 On the other hand, insufficient donor nerves are available in global BPAI patients9; thus, it is necessary to investigate restoration of both elbow flexion and extension with limited motor origins, especially recovery of elbow extension. Although the phrenic nerve (PN) and ICNs both control muscles that assist respiration, they are dominated by different nuclei in the central nerve system. Studies of cerebral plasticity indicate that a new network in the central nerve system may be established after nerve transfer.10,11 Given this previous research, it was hypothesized that transferring the PN and ICNs to reinnervate biceps and triceps, respectively, may provide an acceptable functional outcome. However, the question of synergistic donors supplying antagonistic recipients remains unanswered. In this study, we explored the outcome of triceps and biceps reconstruction with PN and ICNs, respectively. We were particularly concerned with triceps recovery. Patients were evaluated at a minimum of 7 years after surgery for long-term assessment of motor strength and electrodiagnostic study, which ensured sufficient establishment of functional restoration and cerebral plasticity.

TABLE 1. Characteristics of Patients With BPAI Who Underwent PN and ICN Transfer to Reconstruct Elbow Flexion and Extension, Respectively

ELBOW FUNCTION AFTER NERVE TRANSFER

1306

ELBOW FUNCTION AFTER NERVE TRANSFER

per trunk (ADUT) in the first exploratory surgery. ICN transfer was performed at a delayed stage at an interval of approximately 2 months, to avoid possible damage to respiratory function. In all cases, we performed a preoperative nerve conduction study and chest x-ray to confirm the integrity of the bilateral PNs. Each patient provided informed consent to participate in the research. The institution ethics committee granted study approval. Surgical procedure PN-MCN or PN-ADUT: The patient was given general anesthesia without muscle-paralyzing agents so that intraoperative nerve stimulation studies could be performed to assess nerve function. Then the patient was laid supine with the head turned to the healthy side, and a pillow was placed under the shoulder. An incision 7 cm long was made 2 cm parallel to and above the clavicle. The external jugular vein, transverse cervical vessels, and omohyoid muscle were ligated. The brachial plexus was exposed, with the 3 trunks of the plexus and their divisions identified. Nerve transfer was indicated when we determined from the emptiness of the intervertebral foramen that the trunks had been avulsed from the spinal cord; this diagnosis of global root avulsion was confirmed by intrasurgical EMG. The PN was identified on the surface of the anterior scalenus muscle; its function was tested using a nerve stimulator, and it was isolated and transected as distally as possible. As described by Dong et al.,12 the PN was transferred to the anterior division of the upper trunk in the cases in which the division level of the brachial plexus was relatively intact. The upper trunk was dissected at the trunkdivision level, and the proximal end of the PN was coapted to the ADUT directly using 8-0 polypropylene sutures. When the distal end of the ADUT was unavailable or extensively scarred, the PN was transferred to the MCN instead. We identified and divided the MCN and coapted the proximal end of the PN via a sural nerve graft to the distal end of the MCN using 8-0 polypropylene sutures. Other nerve transfers were performed simultaneously or at a delayed stage when appropriate: for example, accessory-to-suprascapular nerve transfer (Table 1). A head–shoulder spica bandage immobilized the surgically treated upper extremity for 4 to 6 weeks after surgery. ICN to triceps branch of radial nerve (TBRN) transfer (ICN-TBRN): The patient was similarly given general anesthesia and placed supine with a pillow under the shoulder, with the head turned to the uninjured side. A continuous thoracobrachial incision was performed with an axillary zplasty followed by a long incision along the anterior

axillary line to the inferior margin of the fifth rib. The pectoralis major muscle was reflected medially and superiorly, and slips of the serratus were bluntly divided to expose each rib. We used electrocautery to open the periosteal sleeve, and a periosteal elevator to dissect the periosteum from the internal surface of the rib. Sharp dissection through the periosteum identified the ICN and intimately associated artery; extreme care was taken to avoid perforating the pleura, which lies immediately deep to the nerves. We harvested a total of 2 to 3 ICNs for transfer according to the cross-section size of TBRN. The third, fourth, and fifth ICNs were most commonly used. The TBRN was found and traced onto the radial nerve. The ICN extremities were then sutured to the TBRN using 8-0 polypropylene sutures. Postoperatively, the entire upper extremity was placed in thoracobrachial immobilization with the elbow held in 90° of flexion for 4 to 6 weeks. Function evaluations: We evaluated long-term results of elbow flexion and extension in all 7 patients. The power of elbow flexion and extension were graded according to the Medical Research Council muscle strength grading system. Motor function was evaluated on a scale ranging from M0 to M5. Good or useful recovery was defined as M3 or higher; poor recovery was M2 or less.13,14 Meanwhile, elbow movement was also observed at rest or during deep breathing. Needle electromyography was used to examine the compound muscle action potential of biceps and triceps when respectively stimulating phrenic nerve and intercostal nerves before anastomotic stoma. Respiratory function assessment: We investigated clinical respiratory symptoms in all patients on their follow-up visit. No patient reported breathing problems in resting or moderate activities. We obtained pulmonary function parameters and used the best of 3 technically satisfactory readings. The same trained technician assessed forced vital capacity (FVC) in sitting positions, forced expiratory volume in 1 second (FEV1), and total lung capacity (TLC) using Master Screen Diffusion Equipment (Jaeger, Bavaria, Germany). The technician was blinded to surgical procedures. Measured lung function results were compared with age-, gender-, height-, and weight-matched normal reference values obtained from healthy Chinese populations. Values of 80% predicted levels were set as the lower limit of normal range.15 RESULTS According to the Medical Research Council scoring scale, postoperative active biceps strength and number of patients were M0, 0; M1, 0; M2, 1; M3, 3; M4, 2; and M5,

JHS 䉬 Vol A, August 

1307

ELBOW FUNCTION AFTER NERVE TRANSFER

TABLE 2. Functional Results and Respiratory Function of Patients With BPAI Who Underwent PN and ICN Transfer to Reconstruct Elbow Flexion and Extension, Respectively

Patient

Biceps Power

MCN Lat

MCN Amp

MCN-Biceps CMAP

Triceps Power

TBRN Lat

TBRN Amp

TBRN-Triceps CMAP

Respiratory Symptoms

FVC %

FEV1 %

TLC %

1 2 3 4 5 6 7

M3 M2 M4 M3 M4 M5 M3

9.6 9.8 7.4 7.2 9.8 4.9 5.8

5 0.5 8.4 8.8 6.8 11 3.1

Mono-mix phase Small amount Mono-mix phase Mono-mix phase Mono-mix phase Mono-mix phase Mono-mix phase

M2 M1 M2 M2 M0 M2 M1

6.6 N/A 5.8 8.3 N/A 11.2 8.8

4.2 N/A 1.6 0.3 N/A 1 0.7

Monophase Small amount Mono-mix phase Mono-mix phase None Monophase Monophase

Good Fair Good Fair Fair Good Good

73.2 73.7 77.8 75 92.6 104.1 76.9

74.6 77.6 77.2 78.2 84.5 108.7 74.8

78.6 79.1 76.6 74.1 93 90 73.3

Lat, latency; Amp, amplitude; CMAP, compound muscle action potential; N/A, not available; FVC, forced vital capacity; FEV1, forced expiratory volume in one second; TLC, total lung capacity. Patients 1, 3, 4, 6, and 7 had biceps and triceps successfully reinnervated with good biceps power but poor triceps power. Patient 2 had both the muscle of biceps and triceps poorly reinnervated with disappointing function as well. Patient 5 had biceps successfully reinnervated with good function, but no recovery in triceps was found. None of these 7 patients reported respiratory problems. The FVC, FEV1, and TLC results were recorded as a percentage of normal predicted values.

FIGURE 1: Comparison of functional recovery between biceps and triceps in 7 patients under long-term observation. The muscle power of biceps and triceps according to the Medical Research Council scoring scale and corresponding number of patients are presented. Six of 7 patients had good recovery of elbow flexion, whereas none obtained useful elbow extension.

1. Six patients regained useful elbow flexion. Postoperative active triceps strength and number of patients were M0, 1; M1, 2; M2, 4; M3, 0; M4, 0; and M5, 0. No one regained functional elbow extension (Table 2, Fig. 1) In the EMG study, patient 2 had poor reinnervation of both biceps and triceps. Patient 5 had poorly reinnervated triceps. The rest of the target muscles showed electromyographic reinnervation (Table 2). We also looked at the relationship between deep inspiration and elbow flexion and found that the 6 patients who obtained satisfactory biceps function could flex their elbows independently without assistance of deep inspiration. Patient 2 regained M2 of biceps power, but he could only flex the elbow horizontally with the help of deep breathing. None of these 7 patients regained functional elbow extension, even with the assistance of deep inspiration. We discovered that

all patients had elbow flexion during sleep. Some even reported subconscious elbow flexion in forced deep inspiration, but they could independently control it. We also investigated the relationship between muscle power and EMG results and found that 6 of 7 patients achieved both satisfactory biceps reinnervation and elbow flexion (M3 or more), whereas the remaining one obtained poor electrical and physical results in biceps. EMG study showed successful reinnervation of triceps in 5 patients; however, only 3 patients could extend his or her elbow horizontally, when eliminating gravity, whereas the other 3 patients could not complete elbow extension and one patient obtained no triceps contraction. With respect to different recipients (ADUT vs MCN) for elbow flexion, we found that 2 patients (patients 1 and 3) with PN-ADUT achieved M3 and M4 biceps power. When MCN was neurotized by PN with an interpositional graft, 4 of 5 patients achieved M3 or more biceps power and one of 5 achieved only M2. Finally, we evaluated postoperative respiratory function and found that all patients could tolerate this defect of unilateral PN and 2 to 3 ipsilateral ICNs. Clinically, no patient experienced breathing problems in daily life and none experienced restrictions in breathing during physical activities. FVC, FEV1 and TLC were 82 ⫾ 12%, 82 ⫾ 12%, and 81 ⫾ 8% of predictive values, respectively. These average values were above the lower limit of normal range.15 DISCUSSION Global brachial plexus root avulsion is a severe injury that causes complete dysfunction in the upper extremity. Nerve transfer provides the best results for reconstruction of BPAI, but to regain more complicated

JHS 䉬 Vol A, August 

1308

ELBOW FUNCTION AFTER NERVE TRANSFER

functions, more donor nerves are needed.6,9,16 According to previous studies, restoration of elbow flexion is the first priority in brachial plexus reconstruction, and satisfactory results have been achieved during recent decades.17,18 Most frequently, we used the PN as a donor, because it contains a large number of motor axons.6 The more axons in the donor nerve, the better the chance for functional recovery.19,20 In our clinical practice, we have achieved great success in elbow flexion reconstruction using PN transfer. Gu and Ma previously reported that 85% of patients regained biceps power to M3 or more after PN transfer.21 Dong et al. reported surgical results of PN transfer to the ADUT; the overall effective rate of this procedure was 83% (MRC ⱖ 3).12 Those authors preferred PN-ADUT transfer to PCN-MCN. Functional outcomes of these 2 methods were similar, but the former one does not require a long nerve graft. Studies also revealed that unilateral PN or unilateral ICNs could be sacrificed without disturbing respiratory function.21–24 In this study, no one reported severe respiratory problems in daily life. Average FVC, FEV1, and TLC were within normal range. Considering that no patient experienced breathing problems in resting and moderate exercise, we believe that this combination of PN and ICN transfer would not cause notable respiratory problems for our patients during long-term observation. These patients could tolerate this unilateral PN and ICN defect. Donor nerves that contain more motor fibers should preferentially be considered for reconstruction of elbow function because of its high priority. However, there are limited donor nerves in patients with global BPAI. In addition, paralytic shoulders and hands also need motor nerve transfers. We therefore considered using PN and ICNs to reinnervate biceps and triceps, respectively. Yet both PN and ICNs originally acted synergistically in respiratory movement. It was generally unknown whether synergistic donors could supply functional recovery for a pair of antagonistic recipients. Previous studies of these nerve transfers have provided some clinical and theoretical clues for this combined strategy. In clinical observation, elbow flexion must be initiated by deep breathing when contraction is first generated and biceps contraction should be assisted by forced breathing for a period of time thereafter. Eventually, independent elbow flexion could be established without the help of respiration. Similarly in a study of contralateral C7 nerve transfer, contraction of re-innervated muscle was necessarily initiated by simultaneous muscle contraction of the contralateral side, but ultimately volitional contraction of target muscle could

be obtained after long-term re-education.11 Functional magnetic resonance imaging (fMRI) studies of cerebral plastic changes after ICNs and contralateral C7 transfer also provided evidence for the occurrence of reorganization in the pathway between primary cerebral motor cortex and newly reinnervated end organ.10 These studies verified that cerebral plasticity actually took place after nerve transfer. Thus, we recruited 7 patients who had PN and ICN transferred to reinnervate the biceps and triceps, respectively. Data were collected after 7 to 12 years of follow-up in which a long-term period was required to ensure sufficient establishment of cerebral plasticity. In this study, patients 1, 3, 4, 6, and 7 had both biceps and triceps successfully reinnervated, with good biceps power but poor triceps recovery. Patient 2 had neither biceps nor triceps well reinnervated; the patient experienced disappointing elbow flexion and extension recovery. Patient 5 had the biceps successfully reinnervated with good function, but no recovery in triceps was found. With respect to elbow flexion, both patients with PN-ADUT achieved functional elbow flexion, whereas 4 of 5 patients with PN-MCN obtained M3 or more biceps power. In this study we found that both PNADUT and PN-MCN resulted in satisfactory functional outcomes. However, the differences between these 2 procedures still need further investigation with more cases. The EMG studies showed that all 6 of 7 patients with functional return of elbow flexion also achieved success reinnervation of the MCN as well. On the other hand, 4 patients regained triceps reinnervation but none acquired functional elbow extension. Three of these 4 patients could only extend their elbows with gravity eliminated. The remaining patient only regained M1 triceps strength. Triceps recovery was poor in all 7 patients in this study, which appears to disagree with previous studies. Doi et al. reported on 24 patients with intercostal to triceps transfer.5 They found a better return of elbow extension, with 25% of patients achieving M3 or M4 and 33% achieving M2. These better results may reflect the fact that this was the sole transfer performed in these patients. We found that when PNMCN/ADUT was combined with ICN-TBRN, it seemed impossible to regain elbow extension to M3 or more, even when the triceps were successfully reinnervated on EMG. This may be attributed to failure in cerebral plasticity after transferring 2 nerves that innervate muscles of respiration to a pair of antagonistic muscles. Previous studies of cerebral plasticity after ICNs or

JHS 䉬 Vol A, August 

ELBOW FUNCTION AFTER NERVE TRANSFER

contralateral C7 transfer demonstrated that a new pathway between cerebral cortex and target muscle could be finally established after long-term re-education.10,11 Areas in the cerebral cortex that originally controlled ICN and cross C7 could eventually control new target muscles volitionally. As we observed, although reinnervated muscles could be controlled independently, involuntary elbow flexion still occurred in subconscious conditions. This phenomenon indicated that reflex action that was monitored by lower levels of the central nerve system might accidentally occur when a higher level of the central nervous system— cerebral cortex— lost its control. We inferred that although PN and ICNs were dominated by different nuclei in the central nerve system, they both monitored respiratory movement. Originally, they acted synergistically in respiration. While deep inspiration was initiated, the impulse was sent out from the cerebral cortex to diaphragmatic muscle and intercostal muscles simultaneously to enlarge the volume of the thoracic cage. However, MCN (or ADUT) and TBRN performed antagonistic functions. After nerve transfer, cerebral plasticity may be prone to recover one of the antagonistic muscles and the other would otherwise be repressed, even when a new peripheral nerve pathway had been re-established in both MCN (or ADUT) and TBRN. It is believed that elbow flexion is more frequently used in daily life and thus plays a more important part in survival for human beings; embryology may explain this paradox.25 In this sense, it is acceptable that elbow flexion regained better functional recovery than elbow extension. Our attempt to restore both elbow flexion and extension by transferring PNs and ICNs brought about a far less satisfactory outcome in elbow extension than flexion. In our opinion,transfer of ICNs to the TBRNs should be avoided after transferring PN to MCN (or ADUT) nerve primarily. Reconstructing elbow flexion and wrist extension with PN and ICNs may result in more acceptable results in adults with complete BPAI. However, because of the relatively young age and small sample of our enrolled patients, one should be conservative in extrapolating results to different patients. REFERENCES 1. Merrell GA, Barrie KB, Katz DL, Wolfe SW, Haven H. Results of nerve transfer techniques for restoration of shoulder and elbow function in the context of a meta-analysis of the English literature. J Hand Surg 2001;26A:303–314. 2. Songcharoen P. Management of brachial plexus injury in adults. Scand J Surg 2008;97:317–323.

1309

3. Rohde R, Wolfe SW. Nerve transfers for adult traumatic brachial plexus palsy. Bull Hosp for Spec Surg 2007;3:77– 82. 4. Millesi H. Coordinated function oriented movements after multiple root avulsion. Acta Neurochir Suppl 2007;100:117–119. 5. Doi K, Shigetomi M, Kaneko K, Soo Heong T, Hiura Y, Yasunory YH. Significance of elbow extension in reconstruction of prehension with reinnervated free-muscle transfer following complete brachial plexus avulsion. Plast Reconstr Surg 1997;100:364 –372. 6. Terzis JK, Kostopoulos VK. The surgical treatment of brachial plexus injuries in adults. Plast Reconstr Surg 2007;119:73e–92e. 7. Chuang DCC, Hattori Y, Ma HS, Chen HC. The reconstructive strategy for improving elbow function in late obstetric brachial palsy. Plast Reconstr Surg 2002;109:116 –126. 8. Moberg E. Surgical treatment for absent single-hand grip and elbow extension in quadriplegia. J Bone Joint Surg 1975;57A:196 –206. 9. Midha R. Nerve transfers for severe brachial plexus injuries: a review. Neurosurg Focus 2004;16:1–10. 10. Malessy MJ, Bakker D, Dekker AJ, Van Duk JG, Thomeer RT. Functional magnetic resonance imaging and control over the biceps muscle after intercostal-musculocutaneous nerve transfer. J Neurosurg 2003;98:261–268. 11. Beaulieu JY, Blustajn J, Teboul F, Baud P, De Schonen S, Thiebaud JB, et al. Cerebral plasticity in crossed C7 grafts of the brachial plexus: an fMRI study. Microsurgery 2006;26:303–310. 12. Dong Z, Zhang CG, Gu YD. Surgical outcome of phrenic nerve transfer to the anterior division of the upper trunk in treating brachial plexus avulsion. J Neurosurg 2010;112:383–385. 13. MacAvoy MC, Green DP. Critical reappraisal of medical research council muscle testing for elbow flexion. J Hand Surg 2007;32A: 149 –153. 14. James MA. Use of the medical research council muscle strength grading system in the upper extremity. J Hand Surg 2007;32A:154 – 156. 15. Zheng JP, Zhong NS. Normative values of pulmonary function testing in Chinese adults. Chin Med J (Engl) 2002;115:50 –54. 16. Millesi H. Surgical management of brachial plexus injuries. J Hand Surg 1977;21A:367–378. 17. Chuang DC, Epstein MD, Yeh MC, Wei FC. Functional restoration of elbow flexion in brachial plexus injuries: results in 167 patients (excluding obstetrical brachial plexus injury). J Hand Surg 1993; 18A:285–291. 18. Vekris MD, Beris AE, Johnson EO, Korobilias AV, Pafilas D, Vekris AD, et al. Musculocutaneous neurotization to restore elbow flexion in brachial plexus paralysis. Microsurgery 2006;26:325–329. 19. Gu YD, Ma MK. Nerve transfer for treatment of root avulsion of the brachial plexus: experimental studies in a rat model. J Reconstr Microsurg 1991;7:15–22. 20. Chen L, Gu YD. An experimental study of contralateral C7 root transfer with vascularized nerve grafting to treat brachial plexus avulsion. J Hand Surg 1994;19B:60 – 66. 21. 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. 22. Gu YD, Ma MK. Use of the phrenic nerve for brachial plexus reconstruction. Clin Orthop 1996;323:119 –121. 23. Waikakul S, Wongtragul S, Vanadurongwan V. Restoration of elbow flexion in brachial plexus avulsion injury: comparing spinal accessory nerve transfer with intercostal nerve transfer. J Hand Surg 1999;24A:571–577. 24. Xu WD, Gu YD, Liu JB, Yu C, Zhang CG, Xu JG. Pulmonary function after complete unilateral phrenic nerve transection. J Neurosurg 2005;103:464 – 467. 25. Narakas AO, Hentz VR. Neurotization in brachial plexus injuries: indications and results. Clin Orthop 1988;237:43–56.

JHS 䉬 Vol A, August 