- Email: [email protected]
Hemi-contralateral C7 transfer to median nerve in the treatment of root avulsion brachial plexus injury
Hemi-Contralateral C7 Transfer to Median Nerve in the Treatment of Root Avulsion Brachial Plexus Injury Panupan Songcharoen, MD, Saichol Wongtrakul, MD, Banchong Mahaisavariya, MD, Bangkok, Thailand, Robert J. Spinner, MD, Rochester, MN Because of the poor clinical results in achieving hand function in patients with complete brachial plexus root avulsion with other nerve transfers, we evaluated 111 patients prospectively to evaluate the technique of the hemi-contralateral C7 transfer to the median nerve. The transfer was performed as a primary procedure in 62 patients and as a secondary procedure in additional 49 patients. Twenty-one of the 62 patients in the primary group had sufficient follow-up (at least 3 years) to assess the motor and sensory recovery in the median nerve. The adverse effects of the operation were also analyzed in all 111 patients. Six of the 21 (29%) patients obtained M3 and 4 (19%) experienced M2 recovery of the wrist and finger flexors. Ten (48%) patients obtained S3 and 7 (33%) had S2 recovery in the median nerve area. The rate of the advancing Tinel’s sign was markedly different between those achieving M3 function and the remaining patients. Although the age of the patient did not correlate with outcome, patients aged 18 and younger had the best motor recovery (ie, achieving M3 function in 3 of 6 cases). There was no correlation between the timing of the surgery after the initial injury, medical comorbidities, and clinical outcome. After surgery 108 of 111 (97%) patients experienced temporary paresthesia in the median nerve area, which resolved by an average of 2.8 months. Three (3%) patients had motor weakness of the donor limb; this resolved completely in 2 patients and left a mild deficit in wrist extension in 1 patient. (J Hand Surg 2001;26A: 1058 –1064. Copyright © 2001 by the American Society for Surgery of the Hand.) Key words: Contralateral C7, median nerve, root avulsion, brachial plexus injury, nerve transfer.
Total root avulsion brachial plexus injury has been a great challenge to peripheral nerve surgeons because it cannot be repaired and has no possibility of From the Department of Orthopaedic Surgery, Faculty of Medicine Siriraj Hospital, Mahido University, Bangkok, Thailand. Received for publication November 30, 2000; accepted in revised form June 27, 2001. No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. Reprint requests: Panupan Songcharoen, MD, Chief of Hand and Microsurgery Unit, Department of Orthopaedic Surgery, Faculty of Medicine Siriraj Hospital, Mahido University, Prannok Road, Bangkok 10700, Thailand. Copyright © 2001 by the American Society for Surgery of the Hand 0363-5023/01/26A06-0009$35.00/0 doi:10.1053/jhsu.2001.27764
1058 The Journal of Hand Surgery
spontaneous recovery. At present the most successful method to restore the affected limb function is nerve transfer.1 Shoulder and elbow functions have been successfully restored by spinal accessory, intercostal, and phrenic nerve transfer.2–5 After the previous success with the proximal portion of the limb, surgeons have attempted to innervate forearm muscles by using these donor nerves. Most of the results with nerve transfers were unsatisfactory until 1992 when Gu et al6 reported their success in the restoration of finger flexion and sensation in the palm by transfer of the normal C7 nerve root from the contralateral side to the affected median nerve. The purpose of this study was to report the clinical
The Journal of Hand Surgery / Vol. 26A No. 6 November 2001 1059
results of the use of one half of the contralateral C7 transfer to the median nerve to restore affected hand function as a primary procedure in patients with total root avulsion brachial plexus injury and to assess the adverse effects of the operation.
Materials and Methods A prospective study was performed during July 1993 to July 1999 in 111 patients with complete brachial plexus root avulsion who underwent contralateral C7 nerve root transfer to median nerve (via a vascularized pedicle interposition ulnar nerve graft) to restore hand function. The indication for this procedure was total root avulsion brachial plexus injury diagnosed with preoperative myelography and confirmed at surgery. The contralateral C7 transfer was a part of nerve transfer procedures in patients who also required spinal accessory to musculocutaneous and phrenic to suprascapular nerve transfers to restore elbow and shoulder function. The contralateral C7 transfer was performed as a primary procedure (at the same time as or within 2 weeks of the other nerve transfer procedures to restore shoulder and elbow function) in 62 patients. In 49 patients the contralateral C7 transfer was performed as a secondary procedure (secondary group). Patients in the secondary group will be reported separately because their outcomes might differ due to the longer period between the initial injury and surgery. Twenty-one patients in the primary group had adequate follow-up (at least 3 years) and were reviewed for functional recovery. There were 19 male and 2 female patients, ranging in age from 9 to 36 years (average, 25 years). The follow-up period ranged from 36 to 68 months with an average of 42 months. The elapsed time between injury and contralateral C7 transfer ranged from 1 to 8 months (average, 5 months). No patient had any medical comorbidity. The sensory and motor morbidity of the procedure was assessed in all 111 patients.
Surgical Technique The operation was performed under general anesthesia. The patient was placed on the operating table in the supine position with the upper part of the body slightly elevated. Both upper extremities, shoulders, neck, and chest were prepared. After the affected brachial plexus was explored, the nerve transfer procedures for restoration of shoulder and elbow functions were completed.
Preparation of the Donor C7 Nerve Root. The healthy contralateral plexus was explored through a transverse incision at the supraclavicular area. The incision began at the outer border of the sternocleidomastoid, 1 finger breath above the clavicle, and extended laterally to the distal third of the clavicle. After the skin and subcutaneous fat were retracted, the branches of the external jugular veins and the inferior belly of omohyoid muscle were identified and protected. The phrenic nerve on the anterior surface of the scalenus anterior was identified by electrical stimulation and preserved. The spinal nerve roots of C5 to T1 underneath the transverse cervical vessels were exposed. With the use of electrical stimulation (Aesculap Nerve Stimulator GN 24, Tuttlingen, Germany), the C7 root was confirmed by extension of the elbow. The nerve sheath of the C7 root was split longitudinally at the level of the root/middle trunk. Four to 6 nerve fascicles were normally found in the entire C7. Each fascicle was stimulated, and fascicles that mainly innervated the shoulder muscles were isolated for the nerve transfer procedure. In our experience most of the fibers innervating the shoulder muscles have been confined to the posterosuperior half of the C7 root. These isolated fascicles were dissected 4 cm from the surrounding tissue and severed for the later coaptation with the reversed vascularized pedicle ulnar nerve graft from the affected side, leaving the fascicles on the anteroinferior half of the C7 root that supply the elbow intact. A subcutaneous tunnel was prepared on the anterior chest wall from the skin incision to the opposite axilla. Preparation of the Recipient Median Nerve and the Ulnar Nerve Graft. The affected arm was then abducted 90° on an armrest. A longitudinal incision was made in the midaxillary line from the axilla to the back of the medial epicondyle. The ulnar nerve was identified in the groove behind the medial epicondyle. Meticulous dissection was performed to isolate and preserve the ulnar nerve along with its vascular pedicle, the superior ulnar collateral artery. Through the same skin incision, the median nerve in the proximal third of the affected arm was identified and isolated. The median and ulnar nerves were sectioned 2 cm proximal to the superior ulnar collateral arterial pedicle. The distal stumps of the nerves were coapted together with 8 – 0 nylon suture. The distal half of the ulnar nerve was isolated from the medial epicondyle to Guyon’s canal through a longitudinal incision on the ulnar side of the affected
1060 Songcharoen et al / Hemi-Contralateral C7 Transfer to Median Nerve
Figure 1. The diagram shows how the normal hemi-contralateral C7 was transferred to the median nerve of the affected side via a vascularized pedicle ulnar nerve graft.
forearm. The isolated ulnar nerve was sectioned distally at Guyon’s canal. The distal end of the harvested ulnar nerve based on the superior ulnar collateral vessels was reflected through the previously prepared subcutaneous pectoral tunnel (Fig. 1). The length of the ulnar graft was
about 36 cm. It was coapted with the isolated fibers of the contralateral C7 root with 8 – 0 nylon suture using ⫻4.5 loupe magnification (Figs. 1, 2). Hemostasis was achieved before closure of the incisions. After surgery the affected arm was immobilized in a Velpeau dressing for 4 weeks.
Figure 2. The distal part of ulnar nerve was reflected through a subcutaneous pectoral tunnel for coaptation with the prepared hemi-contralateral C7 stump.
The Journal of Hand Surgery / Vol. 26A No. 6 November 2001 1061
Physiotherapy began 4 weeks after surgery to maintain range of motion in all joints. Passive shoulder abduction beyond 90° was avoided after the nerve repair. A portable electrical stimulator (L-S Electrical Stimulator Model 2, Bangkok, Thailand) was prescribed for all patients for stimulation of the recovering muscle at home. Electrical stimulation was stopped when patients either obtained M3 motor recovery or at 4 years after surgery if there was no sign of motor recovery. Each patient was seen weekly for 2 weeks immediately after surgery, then 1 month later, then every 3 months thereafter. Beginning in the early postoperative period, a detailed neurologic assesment of the donor limb was performed. We assesed the morbidity of the procedure to the donor limb by interviewing the patients immediately after surgery regarding sensory complaints and by assessing sensory function. To differentiate any limitations from postoperative pain on motor testing, we first recorded the strength in the donor limb 2 weeks after surgery. At later follow-up examinations, the distal progression of Tinel’s sign and the recovery of autonomic, sensory, and motor function were recorded.
Results Six of the 21 (29%) patients in the primary group obtained at least M3 recovery of the wrist and finger flexors (Fig. 3), and 4 (19%) patients had M2 recovery. The M3 recovery time ranged from 24 to 51 months (average, 34 months). At the early stage of the motor recovery, the wrist flexors and finger flexors recovered simultaneously. In all but 1 case the patients could not flex the fingers and wrist separately. Only 1 patient had independent wrist and finger flexion after a period of rehabilitation. Maximal motor function of the contralateral C7 transfer was best evaluated by having the patient extend the normal elbow and flex the fingers with the wrist in a neutral position. Ten (48%) patients had S3 recovery in the median nerve area, and 7 (33%) patients had S2 recovery. The S3 recovery time ranged from 26 to 48 months with an average of 33 months. Four (19%) patients had neither sensory nor motor recovery. The only statistically significant finding (p ⬍ .05) was the average time for Tinel’s sign to reach the elbow, ie, 7 months in the patients who recovered M3 function versus 9 months in the remaining patients. Even though the age of the patients at injury did not correlate with the surgical results, the best
outcomes were achieved in the group of patients aged 18 years and younger. The average age of patients in our study was 25 years compared with the average age of patients who had M3 recovery, 23 years (p ⫽ .44). Three of 6 patients aged 18 years and younger reached M3 function, and 4 patients achieved S3 function. The elapsed time between injury and operation and the health of our patients did not affect outcomes in this study. The average elapsed time between injury and operation was 5 months compared with 7 months in the group of patients with M3 motor recovery. Of 3 patients operated on 3 months or less after initial injury, 1 patient achieved M3 motor recovery. No patient had any chronic illness. Of the 111 patients, 108 (97%) patients experienced temporary paresthesia in the median nerve area of the donor hand after surgery; 63 (58%) patients had paresthesia on the index pulp (Fig. 4), 32 (30%) patients on the median innervated area of the palm, 8 (7%) patients on the anterolateral aspect of shoulder and upper arm, and 5 (5%) patients on the median innervated area of the palm and shoulder. The sensory change recovered spontaneously in all cases. The recovery time ranged from 2 weeks to 7 months with an average of 3 months. Weakness was found much less frequently than sensory change. Only 3 (3%) patients had weakness of the donor limb. Two patients had M4 weakness in the triceps that recovered completely in 2 months. One patient had M2 isolated weakness of the extensor digitorum communis that gradually improved to M4 over 3 years.
Discussion In root avulsion brachial plexus injury, nerve transfer has yielded a substantial degree of success. In the past decade elbow flexion and shoulder abduction could be regularly restored by this procedure; however, attempts to restore hand function by nerve transfer to the median nerve with intercostal and spinal accessory nerves have yielded poor results. The major causes of this failure are the discrepancy in the number of nerve fibers and functional type (motor and sensory) between donor and recipient. Considering that the normal C7 nerve root contains approximately 23,000 myelinated axons and the median nerve contains approximately 15,000 myelinated axons, the normal C7 (or only a half C7) is a theoretically reasonable donor for the median nerve.
1062 Songcharoen et al / Hemi-Contralateral C7 Transfer to Median Nerve
Figure 3. (A) M3 recovery of the wrist flexors was achieved in this patient. Clinical examination shows wrist flexion against gravity with the fingers extended. The wrist flexor tendons could be palpated strongly at the wrist. (B) In the same patient M3 motor recovery of the digital flexors of the recipient hand (L) was achieved. Finger flexion is performed with the wrist maintained in a neutral position. Note how this compares with the flexion of the normal donor hand (R).
In 1998 Gu et al7 and Chuang et al8 separately reported their results of the seventh cervical nerve root transfer from the contralateral healthy side to several recipient nerves for treatment of brachial plexus root avulsion with a considerable degree of motor and sensory recovery and a minimal degree of donor deficit. These investigators reported their results using the entire C7 nerve transfer elongated by a pedicled or free vascularized ulnar nerve graft. The normal contralateral C7 was shown to be a better alternative for transfer to the median nerve than the previously used donor nerves. The contralateral C7 transfer was first performed
in our center in 1993. Our technique uses only half the normal C7 and a vascularized pedicle interposition ulnar graft. The reasons for using only half of C7 nerve root are, first, to minimize the risk of downgrading the function of the only good upper limb. This is done by selecting the donor fascicles that supply mainly the shoulder muscles. Second, the median nerve generally contains only one half to two thirds of the number of myelinated axons of C7 root. The use of the whole C7 trunk exceeds the need of the recipient median nerve and causes unnecessary damage to the donor limb function. Finally, the cross-sectional area of the interposition ulnar nerve
The Journal of Hand Surgery / Vol. 26A No. 6 November 2001 1063
Figure 4. The area of temporary paresthesia on the index pulp of the donor hand.
graft is generally smaller than the whole C7 trunk. Only half of the C7 root is needed to make an optimum coaptation between the donor C7 and the ulnar nerve graft. The use of a vascularized pedicle rather than free vascularized nerve shortens the operative procedure. We also believe that nerve transfer is preferential to free muscle transfer for patients treated early with complete brachial plexus injury. Although preliminary results with double free-muscle transfer have been encouraging,9 we believe that this technique should be reserved for late cases or those that have failed attempts at nerve reconstruction. In comparing our results of hemi C7 transfer with the results of Gu et al7 of whole C7 transfer, the percentage of M3 motor recovery of wrist and finger flexors in their series was 2 times higher and the percentage of S2 sensory recovery in the median nerve area was comparable (Table 1). In comparing the functional deficit of the donor limb after hemi C7 transection in our series with the series by Chuang et al8 of whole C7 transection, their study showed a much lower percentage of temporary sensory change (Table 2). The area of dysesthesia was confined only to the hand, whereas the area of sensory alteration was found in the palm and shoulder in our results. Motor weakness of the donor limb was much lower in our series, although the weakness occurred in the same muscle groups. In our opinion,
the much lower rate of donor limb weakness in our series was mainly attributable to the use of only half of C7 in the transfer procedure. In addition, intraoperative electrical stimulation of the harvested half of C7 was performed to confirm that it was the portion that most innervated the shoulder muscles. These muscles in general were supplied mainly by the C5 and C6 roots. Therefore, the possibility of downgrading the motor function of the donor limb was reduced. However, it was still possible that the rate of transient motor weakness was higher immediately after surgery. Because of the relatively common postoperative pain in the donor limb, patients often
Table 1. Comparison of Functional Recovery of the Injured Limb Gu et al7 Donor No. of patients Average age (yr) of patients Elapsed time (mo) between injury and C7 transfer ⱖM3 recovery (%) ⱖS2 recovery (%)
C7 8
Songcharoen et al ⁄ C7 21
12
26*
24.7
12.5* 63 87
4.9 29 81
*These numbers include data of all patients treated with contralateral C7 transfer to different recipients. The median nerve comprised 8 of 20 cases in this series.
1064 Songcharoen et al / Hemi-Contralateral C7 Transfer to Median Nerve
Table 2. Comparison of Donor Limb Morbidity Donor No. of patients Sensory deficit (%) Area of deficit Motor deficit (triceps, extensor digitorum communis) (%)
Chuang et al8
Songcharoen et al
C7 21 52 Palm 19
⁄ C7 111 97 Palm and shoulder 3 12
had give-way weakness that could not be distinguished accurately from true weakness. Even though the result of contralateral C7 transfer to median nerve is better than other donor nerves, the success rate of this procedure is still far too low. The most important obstacle to a higher motor recovery rate is the atrophy of the recipient muscles. Attempts to improve the success rate of this procedure should include reducing the interval between injury and the contralateral C7 transfer procedure, minimizing the length of the interposition nerve graft, and prolonging the irreversible atrophic time of the recipient muscles. In our opinion, the contralateral C7 nerve root should be used only to transfer to a recipient nerve with complicated function, eg, median nerve. Because the loss of the C7 nerve root carries some risks of downgrading the donor limb function, no matter how trivial the risk is, function in the only healthy upper limb should be preserved unless there is no other alternative. The indications, selection of
patients, and techniques for the contralateral C7 transfer need better definition.
References 1. Narakas AO, Hentz VR. Neurotization in brachial plexus injuries: indication and results. Clin Orthop 1988;237:43– 56. 2. Nagano A, Tsuyama N, Ochiai N, Hara T, Takahashi M. Direct nerve crossing with the intercostal nerve to treat avulsion injuries of the brachial plexus. J Hand Surg 1989; 14A:980 –985. 3. Allieu Y, Privat JM, Bonnel F. Paralysis in root avulsion of the brachial plexus: neurotization by spinal accessory nerve. Clin Plast Surg 1984;11:133–136. 4. Songcharoen P. Brachial plexus injury in Thailand: a report of 520 cases. Microsurgery 1995;16:35–39. 5. Songcharoen P, Mahaisavariya B, Chotigavanich C. Spinal accessory neurotization for restoration of elbow flexion in avulsion injuries of brachial plexus. J Hand Surg 1996;21A: 387–390. 6. Gu Y-D, Zhang G-M, Chen D-S, Yan J-G, Cheng X-M, Chen L. Seventh cervical nerve root transfer from the contralateral healthy side for treatment of brachial plexus root avulsions. J Hand Surg 1992;17B:518 –521. 7. Gu Y-D, Chen D-S, Zhang G-M, Cheng X-M, Xu J-G, Zhang L-Y, et al. Long-term functional results of contralateral C7 transfer. J Reconstr Microsurg 1998;14:57–59. 8. Chuang DC-C, Cheng S-L, Wei F-C, Wu C-L, Ho Y-S. Clinical evaluation of C7 spinal nerve transection: 21 patients with at least 2 years’ follow-up. Br J Plast Surg 1998;51:285–290. 9. Doi K, Muramatsu K, Hattori Y, Otsuka K, Tan S-H, Nanda V, Watanabe M. Restoration of prehension with the double free muscle technique following complete avulsion of the brachial plexus: indications and long-term results. J Bone Joint Surg 2000;82A:652– 666.
Total root avulsion brachial plexus injury has been a great challenge to peripheral nerve surgeons because it cannot be repaired and has no possibility of From the Department of Orthopaedic Surgery, Faculty of Medicine Siriraj Hospital, Mahido University, Bangkok, Thailand. Received for publication November 30, 2000; accepted in revised form June 27, 2001. No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. Reprint requests: Panupan Songcharoen, MD, Chief of Hand and Microsurgery Unit, Department of Orthopaedic Surgery, Faculty of Medicine Siriraj Hospital, Mahido University, Prannok Road, Bangkok 10700, Thailand. Copyright © 2001 by the American Society for Surgery of the Hand 0363-5023/01/26A06-0009$35.00/0 doi:10.1053/jhsu.2001.27764
1058 The Journal of Hand Surgery
spontaneous recovery. At present the most successful method to restore the affected limb function is nerve transfer.1 Shoulder and elbow functions have been successfully restored by spinal accessory, intercostal, and phrenic nerve transfer.2–5 After the previous success with the proximal portion of the limb, surgeons have attempted to innervate forearm muscles by using these donor nerves. Most of the results with nerve transfers were unsatisfactory until 1992 when Gu et al6 reported their success in the restoration of finger flexion and sensation in the palm by transfer of the normal C7 nerve root from the contralateral side to the affected median nerve. The purpose of this study was to report the clinical
The Journal of Hand Surgery / Vol. 26A No. 6 November 2001 1059
results of the use of one half of the contralateral C7 transfer to the median nerve to restore affected hand function as a primary procedure in patients with total root avulsion brachial plexus injury and to assess the adverse effects of the operation.
Materials and Methods A prospective study was performed during July 1993 to July 1999 in 111 patients with complete brachial plexus root avulsion who underwent contralateral C7 nerve root transfer to median nerve (via a vascularized pedicle interposition ulnar nerve graft) to restore hand function. The indication for this procedure was total root avulsion brachial plexus injury diagnosed with preoperative myelography and confirmed at surgery. The contralateral C7 transfer was a part of nerve transfer procedures in patients who also required spinal accessory to musculocutaneous and phrenic to suprascapular nerve transfers to restore elbow and shoulder function. The contralateral C7 transfer was performed as a primary procedure (at the same time as or within 2 weeks of the other nerve transfer procedures to restore shoulder and elbow function) in 62 patients. In 49 patients the contralateral C7 transfer was performed as a secondary procedure (secondary group). Patients in the secondary group will be reported separately because their outcomes might differ due to the longer period between the initial injury and surgery. Twenty-one patients in the primary group had adequate follow-up (at least 3 years) and were reviewed for functional recovery. There were 19 male and 2 female patients, ranging in age from 9 to 36 years (average, 25 years). The follow-up period ranged from 36 to 68 months with an average of 42 months. The elapsed time between injury and contralateral C7 transfer ranged from 1 to 8 months (average, 5 months). No patient had any medical comorbidity. The sensory and motor morbidity of the procedure was assessed in all 111 patients.
Surgical Technique The operation was performed under general anesthesia. The patient was placed on the operating table in the supine position with the upper part of the body slightly elevated. Both upper extremities, shoulders, neck, and chest were prepared. After the affected brachial plexus was explored, the nerve transfer procedures for restoration of shoulder and elbow functions were completed.
Preparation of the Donor C7 Nerve Root. The healthy contralateral plexus was explored through a transverse incision at the supraclavicular area. The incision began at the outer border of the sternocleidomastoid, 1 finger breath above the clavicle, and extended laterally to the distal third of the clavicle. After the skin and subcutaneous fat were retracted, the branches of the external jugular veins and the inferior belly of omohyoid muscle were identified and protected. The phrenic nerve on the anterior surface of the scalenus anterior was identified by electrical stimulation and preserved. The spinal nerve roots of C5 to T1 underneath the transverse cervical vessels were exposed. With the use of electrical stimulation (Aesculap Nerve Stimulator GN 24, Tuttlingen, Germany), the C7 root was confirmed by extension of the elbow. The nerve sheath of the C7 root was split longitudinally at the level of the root/middle trunk. Four to 6 nerve fascicles were normally found in the entire C7. Each fascicle was stimulated, and fascicles that mainly innervated the shoulder muscles were isolated for the nerve transfer procedure. In our experience most of the fibers innervating the shoulder muscles have been confined to the posterosuperior half of the C7 root. These isolated fascicles were dissected 4 cm from the surrounding tissue and severed for the later coaptation with the reversed vascularized pedicle ulnar nerve graft from the affected side, leaving the fascicles on the anteroinferior half of the C7 root that supply the elbow intact. A subcutaneous tunnel was prepared on the anterior chest wall from the skin incision to the opposite axilla. Preparation of the Recipient Median Nerve and the Ulnar Nerve Graft. The affected arm was then abducted 90° on an armrest. A longitudinal incision was made in the midaxillary line from the axilla to the back of the medial epicondyle. The ulnar nerve was identified in the groove behind the medial epicondyle. Meticulous dissection was performed to isolate and preserve the ulnar nerve along with its vascular pedicle, the superior ulnar collateral artery. Through the same skin incision, the median nerve in the proximal third of the affected arm was identified and isolated. The median and ulnar nerves were sectioned 2 cm proximal to the superior ulnar collateral arterial pedicle. The distal stumps of the nerves were coapted together with 8 – 0 nylon suture. The distal half of the ulnar nerve was isolated from the medial epicondyle to Guyon’s canal through a longitudinal incision on the ulnar side of the affected
1060 Songcharoen et al / Hemi-Contralateral C7 Transfer to Median Nerve
Figure 1. The diagram shows how the normal hemi-contralateral C7 was transferred to the median nerve of the affected side via a vascularized pedicle ulnar nerve graft.
forearm. The isolated ulnar nerve was sectioned distally at Guyon’s canal. The distal end of the harvested ulnar nerve based on the superior ulnar collateral vessels was reflected through the previously prepared subcutaneous pectoral tunnel (Fig. 1). The length of the ulnar graft was
about 36 cm. It was coapted with the isolated fibers of the contralateral C7 root with 8 – 0 nylon suture using ⫻4.5 loupe magnification (Figs. 1, 2). Hemostasis was achieved before closure of the incisions. After surgery the affected arm was immobilized in a Velpeau dressing for 4 weeks.
Figure 2. The distal part of ulnar nerve was reflected through a subcutaneous pectoral tunnel for coaptation with the prepared hemi-contralateral C7 stump.
The Journal of Hand Surgery / Vol. 26A No. 6 November 2001 1061
Physiotherapy began 4 weeks after surgery to maintain range of motion in all joints. Passive shoulder abduction beyond 90° was avoided after the nerve repair. A portable electrical stimulator (L-S Electrical Stimulator Model 2, Bangkok, Thailand) was prescribed for all patients for stimulation of the recovering muscle at home. Electrical stimulation was stopped when patients either obtained M3 motor recovery or at 4 years after surgery if there was no sign of motor recovery. Each patient was seen weekly for 2 weeks immediately after surgery, then 1 month later, then every 3 months thereafter. Beginning in the early postoperative period, a detailed neurologic assesment of the donor limb was performed. We assesed the morbidity of the procedure to the donor limb by interviewing the patients immediately after surgery regarding sensory complaints and by assessing sensory function. To differentiate any limitations from postoperative pain on motor testing, we first recorded the strength in the donor limb 2 weeks after surgery. At later follow-up examinations, the distal progression of Tinel’s sign and the recovery of autonomic, sensory, and motor function were recorded.
Results Six of the 21 (29%) patients in the primary group obtained at least M3 recovery of the wrist and finger flexors (Fig. 3), and 4 (19%) patients had M2 recovery. The M3 recovery time ranged from 24 to 51 months (average, 34 months). At the early stage of the motor recovery, the wrist flexors and finger flexors recovered simultaneously. In all but 1 case the patients could not flex the fingers and wrist separately. Only 1 patient had independent wrist and finger flexion after a period of rehabilitation. Maximal motor function of the contralateral C7 transfer was best evaluated by having the patient extend the normal elbow and flex the fingers with the wrist in a neutral position. Ten (48%) patients had S3 recovery in the median nerve area, and 7 (33%) patients had S2 recovery. The S3 recovery time ranged from 26 to 48 months with an average of 33 months. Four (19%) patients had neither sensory nor motor recovery. The only statistically significant finding (p ⬍ .05) was the average time for Tinel’s sign to reach the elbow, ie, 7 months in the patients who recovered M3 function versus 9 months in the remaining patients. Even though the age of the patients at injury did not correlate with the surgical results, the best
outcomes were achieved in the group of patients aged 18 years and younger. The average age of patients in our study was 25 years compared with the average age of patients who had M3 recovery, 23 years (p ⫽ .44). Three of 6 patients aged 18 years and younger reached M3 function, and 4 patients achieved S3 function. The elapsed time between injury and operation and the health of our patients did not affect outcomes in this study. The average elapsed time between injury and operation was 5 months compared with 7 months in the group of patients with M3 motor recovery. Of 3 patients operated on 3 months or less after initial injury, 1 patient achieved M3 motor recovery. No patient had any chronic illness. Of the 111 patients, 108 (97%) patients experienced temporary paresthesia in the median nerve area of the donor hand after surgery; 63 (58%) patients had paresthesia on the index pulp (Fig. 4), 32 (30%) patients on the median innervated area of the palm, 8 (7%) patients on the anterolateral aspect of shoulder and upper arm, and 5 (5%) patients on the median innervated area of the palm and shoulder. The sensory change recovered spontaneously in all cases. The recovery time ranged from 2 weeks to 7 months with an average of 3 months. Weakness was found much less frequently than sensory change. Only 3 (3%) patients had weakness of the donor limb. Two patients had M4 weakness in the triceps that recovered completely in 2 months. One patient had M2 isolated weakness of the extensor digitorum communis that gradually improved to M4 over 3 years.
Discussion In root avulsion brachial plexus injury, nerve transfer has yielded a substantial degree of success. In the past decade elbow flexion and shoulder abduction could be regularly restored by this procedure; however, attempts to restore hand function by nerve transfer to the median nerve with intercostal and spinal accessory nerves have yielded poor results. The major causes of this failure are the discrepancy in the number of nerve fibers and functional type (motor and sensory) between donor and recipient. Considering that the normal C7 nerve root contains approximately 23,000 myelinated axons and the median nerve contains approximately 15,000 myelinated axons, the normal C7 (or only a half C7) is a theoretically reasonable donor for the median nerve.
1062 Songcharoen et al / Hemi-Contralateral C7 Transfer to Median Nerve
Figure 3. (A) M3 recovery of the wrist flexors was achieved in this patient. Clinical examination shows wrist flexion against gravity with the fingers extended. The wrist flexor tendons could be palpated strongly at the wrist. (B) In the same patient M3 motor recovery of the digital flexors of the recipient hand (L) was achieved. Finger flexion is performed with the wrist maintained in a neutral position. Note how this compares with the flexion of the normal donor hand (R).
In 1998 Gu et al7 and Chuang et al8 separately reported their results of the seventh cervical nerve root transfer from the contralateral healthy side to several recipient nerves for treatment of brachial plexus root avulsion with a considerable degree of motor and sensory recovery and a minimal degree of donor deficit. These investigators reported their results using the entire C7 nerve transfer elongated by a pedicled or free vascularized ulnar nerve graft. The normal contralateral C7 was shown to be a better alternative for transfer to the median nerve than the previously used donor nerves. The contralateral C7 transfer was first performed
in our center in 1993. Our technique uses only half the normal C7 and a vascularized pedicle interposition ulnar graft. The reasons for using only half of C7 nerve root are, first, to minimize the risk of downgrading the function of the only good upper limb. This is done by selecting the donor fascicles that supply mainly the shoulder muscles. Second, the median nerve generally contains only one half to two thirds of the number of myelinated axons of C7 root. The use of the whole C7 trunk exceeds the need of the recipient median nerve and causes unnecessary damage to the donor limb function. Finally, the cross-sectional area of the interposition ulnar nerve
The Journal of Hand Surgery / Vol. 26A No. 6 November 2001 1063
Figure 4. The area of temporary paresthesia on the index pulp of the donor hand.
graft is generally smaller than the whole C7 trunk. Only half of the C7 root is needed to make an optimum coaptation between the donor C7 and the ulnar nerve graft. The use of a vascularized pedicle rather than free vascularized nerve shortens the operative procedure. We also believe that nerve transfer is preferential to free muscle transfer for patients treated early with complete brachial plexus injury. Although preliminary results with double free-muscle transfer have been encouraging,9 we believe that this technique should be reserved for late cases or those that have failed attempts at nerve reconstruction. In comparing our results of hemi C7 transfer with the results of Gu et al7 of whole C7 transfer, the percentage of M3 motor recovery of wrist and finger flexors in their series was 2 times higher and the percentage of S2 sensory recovery in the median nerve area was comparable (Table 1). In comparing the functional deficit of the donor limb after hemi C7 transection in our series with the series by Chuang et al8 of whole C7 transection, their study showed a much lower percentage of temporary sensory change (Table 2). The area of dysesthesia was confined only to the hand, whereas the area of sensory alteration was found in the palm and shoulder in our results. Motor weakness of the donor limb was much lower in our series, although the weakness occurred in the same muscle groups. In our opinion,
the much lower rate of donor limb weakness in our series was mainly attributable to the use of only half of C7 in the transfer procedure. In addition, intraoperative electrical stimulation of the harvested half of C7 was performed to confirm that it was the portion that most innervated the shoulder muscles. These muscles in general were supplied mainly by the C5 and C6 roots. Therefore, the possibility of downgrading the motor function of the donor limb was reduced. However, it was still possible that the rate of transient motor weakness was higher immediately after surgery. Because of the relatively common postoperative pain in the donor limb, patients often
Table 1. Comparison of Functional Recovery of the Injured Limb Gu et al7 Donor No. of patients Average age (yr) of patients Elapsed time (mo) between injury and C7 transfer ⱖM3 recovery (%) ⱖS2 recovery (%)
C7 8
Songcharoen et al ⁄ C7 21
12
26*
24.7
12.5* 63 87
4.9 29 81
*These numbers include data of all patients treated with contralateral C7 transfer to different recipients. The median nerve comprised 8 of 20 cases in this series.
1064 Songcharoen et al / Hemi-Contralateral C7 Transfer to Median Nerve
Table 2. Comparison of Donor Limb Morbidity Donor No. of patients Sensory deficit (%) Area of deficit Motor deficit (triceps, extensor digitorum communis) (%)
Chuang et al8
Songcharoen et al
C7 21 52 Palm 19
⁄ C7 111 97 Palm and shoulder 3 12
had give-way weakness that could not be distinguished accurately from true weakness. Even though the result of contralateral C7 transfer to median nerve is better than other donor nerves, the success rate of this procedure is still far too low. The most important obstacle to a higher motor recovery rate is the atrophy of the recipient muscles. Attempts to improve the success rate of this procedure should include reducing the interval between injury and the contralateral C7 transfer procedure, minimizing the length of the interposition nerve graft, and prolonging the irreversible atrophic time of the recipient muscles. In our opinion, the contralateral C7 nerve root should be used only to transfer to a recipient nerve with complicated function, eg, median nerve. Because the loss of the C7 nerve root carries some risks of downgrading the donor limb function, no matter how trivial the risk is, function in the only healthy upper limb should be preserved unless there is no other alternative. The indications, selection of
patients, and techniques for the contralateral C7 transfer need better definition.
References 1. Narakas AO, Hentz VR. Neurotization in brachial plexus injuries: indication and results. Clin Orthop 1988;237:43– 56. 2. Nagano A, Tsuyama N, Ochiai N, Hara T, Takahashi M. Direct nerve crossing with the intercostal nerve to treat avulsion injuries of the brachial plexus. J Hand Surg 1989; 14A:980 –985. 3. Allieu Y, Privat JM, Bonnel F. Paralysis in root avulsion of the brachial plexus: neurotization by spinal accessory nerve. Clin Plast Surg 1984;11:133–136. 4. Songcharoen P. Brachial plexus injury in Thailand: a report of 520 cases. Microsurgery 1995;16:35–39. 5. Songcharoen P, Mahaisavariya B, Chotigavanich C. Spinal accessory neurotization for restoration of elbow flexion in avulsion injuries of brachial plexus. J Hand Surg 1996;21A: 387–390. 6. Gu Y-D, Zhang G-M, Chen D-S, Yan J-G, Cheng X-M, Chen L. Seventh cervical nerve root transfer from the contralateral healthy side for treatment of brachial plexus root avulsions. J Hand Surg 1992;17B:518 –521. 7. Gu Y-D, Chen D-S, Zhang G-M, Cheng X-M, Xu J-G, Zhang L-Y, et al. Long-term functional results of contralateral C7 transfer. J Reconstr Microsurg 1998;14:57–59. 8. Chuang DC-C, Cheng S-L, Wei F-C, Wu C-L, Ho Y-S. Clinical evaluation of C7 spinal nerve transection: 21 patients with at least 2 years’ follow-up. Br J Plast Surg 1998;51:285–290. 9. Doi K, Muramatsu K, Hattori Y, Otsuka K, Tan S-H, Nanda V, Watanabe M. Restoration of prehension with the double free muscle technique following complete avulsion of the brachial plexus: indications and long-term results. J Bone Joint Surg 2000;82A:652– 666.