Treatment of brachial plexus injury*

Treatment of brachial plexus injury*

JA.Orthop SciTreatment (1998) 3:71–80 Nagano: of brachial plexus injury 71 Instructional lecture Treatment of brachial plexus injury* Akira Nagano D...

999KB Sizes 0 Downloads 96 Views

JA.Orthop SciTreatment (1998) 3:71–80 Nagano: of brachial plexus injury

71

Instructional lecture Treatment of brachial plexus injury* Akira Nagano Department of Orthopaedic Surgery, The University of Tokyo Hospital Branch, 3-28-6 Mejirodai, Bunkyo-ku, Tokyo 112, Japan

Abstract: A brachial plexus injury is the most severe nerve injury of the extremities. To achieve good results from treatment, correct diagnosis and early nerve repair are mandatory. The brachial plexus should be explored as early as possible if there is an incised wound, if clinical findings or diagnostic imaging indicate that at least one root is avulsed, if there is damage to the subclavian artery, and if there is total-type injury. With an upper-type injury with no clinical signs of a preganglionic lesion, the patient should be treated conservatively for 3 months and if there are no signs of recovery, then the brachial plexus should be explored. During this exploration, recording of the spinal cord evoked potential (ESCP) or the somatosensory evoked potential (SEP) is mandatory to determine the site of injury. Nerve grafting is indicated for a rupture in the root demonstrating a positive ESCP or SEP potential, in the trunk or in the cord. Exploration of the brachial plexus should be extended distally as far as possible to achieve good results after nerve grafting; when this was done more than M3 (MRC grading) power of the infraspinatus, deltoid, and biceps was achieved in more than 70% of our 32, 30, 33 patients, respectively. Results of nerve grafting for the forearm muscles have been very poor. Intercostal nerve transfer is recommended to restore elbow flexion in root avulsion type of injury, with elbow flexion to more than M3 being regained in 70% of our 221 patients. The best results of intercostal nerve transfer were achieved in patients younger than 30 years who received the operation within 6 months after injury. Motor recovery of hand function after intercostal nerve transfer was poor but protective sensation was restored in fingers innervated by the median nerve. The recommended treatment for each type of injury is described according to the results achieved.

Introduction Of nerve injuries in the extremities, brachial plexus injury is the most severe, most being incurred through motorcycle accidents.12 Correct diagnosis19 and early nerve repair are mandatory. Treatment of a brachial plexus injury is either conservative or operative. The available operative procedures include neurolysis; nerve grafting; nerve transfer; reconstructive procedures such as tendon transfer and arthrodesis; and free muscle transplantation with nerve transfer. Indications for each procedure depend on such factors as the degree of damage, the site of injury, the type of involved roots, the time interval between the injury and the operation, and patient’s age, sex, and occupation. The degree of damage and the site of injury are the most important factors. Treatment is decided upon according to the following three steps: The first step is to decide whether the lesion can be treated conservatively or whether it should be treated operatively to explore the brachial plexus. If there is any brachial plexus exploration, the second step is to differentiate between preganglionic and postganglionic lesions, using intraoperative electrophysiological examinations. The third step is decide which procedure should be adopted for the particular patient, taking into account the above-mentioned indicative factors.

Key words: nerve grafting, nerve transfer, tendon transfer, shoulder arthrodesis, strategy

Indications for exploration of the brachial plexus

Offprint requests to: A. Nagano Received for publication on March 18, 1997; accepted on Aug. 7, 1997 * Presented at the 69th Annual Meeting of the Japanese Orthopaedic Association, Tokyo, April 12, 1996

Conservative treatment is indicated when the damage is non-degenerative, as ascertained by electrophysiological examinations. If the palsy was incurred through carrying a heavy rucksack or by malpositioning during an operation or during sleeping, conservative treatment is also indicated even if damage is degenerative.

72

A. Nagano: Treatment of brachial plexus injury

The brachial plexus should be explored as early as possible if there is an incised wound, as the brachial plexus is ruptured in most such instances; if clinical findings or diagnostic imagings indicate that at least one root is avulsed, since the other roots are likely to be also severely damaged; or when there is damage to the subclavian artery,19 because in almost all such instances the plexus is ruptured in the infraclavicular region and several roots are often simultaneously avulsed. If the lesion is suspected to be postganglionic, based on clinical findings and a myelogram, the timing of the exploration is decided according to the extent of paralysis. Prior to 1978, we treated such patients conservatively.13 Of 50 patients with total-type injury (Table 1), 8 responded fully to conservative treatment and all muscles recovered. A further 5 patients showed useful recovery in all muscles except the extensor digiti communis and/or the intrinsic muscles. Overall, 13 (26%) of the 50 patients regained useful function through conservative treatment and did not need surgical intervention. The other 37 patients had a poor recovery, which suggested that nerves in these patients had been ruptured or avulsed and surgical repair should have been attempted. Of the 32 patients with an uppertype injury treated conservatively (Table 1), almost complete recovery was achieved in 13 patients (41%). In the other 19, recovery of the spinati and the deltoid was poor. This indicated that conservative treatment could lead to useful recovery in 40% of patients with an upper-type injury. In the light of all these results, we recommend exploration of the brachial plexus as early as possible in total-type injury. Patients with an uppertype injury with no clinical signs of a preganglionic lesion should be treated conservatively for 3 months, and if there are no signs of recovery, then the brachial plexus should be explored. Exploration is not indicated for a patient with lower root damage and normal upper roots. It is also not indicated for patients older than 50 years, or for patients injured more than 1 year previously, as the results after nerve repair in such instances have generally been poor.

Table 1. Patients with brachial plexus injury who received conservative treatment Type of injury

Age at operation (years) Follow-up period (years)

Total (50 patients)

Upper (32 patients)

4–44 (23) 7–31 (6)

4–66 (29) 1–17 (6)

Mean values are given in parentheses

Differentiation between preganglionic and postganglionic lesions during operation Lesions in the brachial plexus are classified as either preganglionic or postganglionic. For a postganglionic lesion, good results can be achieved by conservative treatment or by nerve grafting, but for a preganglionic lesion no recovery can be expected by such means and nerve transfer or another reconstructive procedure should be considered without delay. Preganglionic and postganglionic lesions are subclassified into six categories (Fig. 1). “Zone II (1I)” appears to be a root lesion, but is thought to be a combination of intraforaminal root avulsion or damage to the root most proximal. “Apparent continuity” is intraforaminal avulsion. “Vacant sheath” means that the root is avulsed but is connected to the foramen by fibrous tissue and not by neural elements. When the brachial plexus is explored, the level of injury is at first ascertained directly by simple inspection. Root avulsion and vacant sheath are easily diag-

Fig. 1. Classification of lesions of brachial plexus injuries

Table 2. Intraoperative diagnosis of level of injury to the brachial plexus 1. Macroscopic diagnosis 2. Histological diagnosis Acetylcholinesterase stain (Karnovsky) Choline acetyltransferase stain (Engel) 3. Electrophysiological diagnosis Somatosensory evoked potential recording Evoked spinal cord potential recording

A. Nagano: Treatment of brachial plexus injury

73

nosed by inspection or palpation. However, apparent continuity and zone II (1I) cannot be distinguished, as ordinary exploration of the brachial plexus can only ascertain the condition of the root as far as the tip of the transverse process. Laminectomy reveals the states of the root and rootlets directly, but is extremely invasive. Instead of laminectomy, one or more intraoperative procedures (listed in Table 2) can be employed for a definitive diagnosis of nerve rupture, zone II (1I), and apparent continuity. We employ intraoperative electrophysiological examinations such as recording the somatosensory evoked potential (SEP)29 and the evoked spinal cord potential (ESCP)11 to confirm the site of injury (Fig. 2). Of 334 roots

submitted to intraoperative electrophysiological examinations, 12 were diagnosed as being in apparent continuity; 38 (44.2%) of 86 roots suspected to have postganglionic lesion by inspection were determined as zone II (1I) (Table 3).14 We have grafted nerve to roots with absent SEP, but the results were very poor. So SEP recording is mandatory to determine the site of injury. The disadvantage of SEP recording is that it is strongly affected by the depth of anesthesia. When the level of anesthesia is deep, SEP sometimes cannot be recorded even from a normal root. On the other hand, ESCP is easily recorded at any depth of anesthesia and the amplitude of ESCP is larger than that of SEP. So we now mainly record ESCP, inserting an electrode into the epidural space at the C1–C2 level.

Treatment for a lesion in continuity Table 3. Level of injury, ascertained by intraoperative electrophysiological examinations, for 334 roots injured in the supraclavicular region Number of roots Preganglionic lesion Avulsion Vacant sheath Apparent continuity Postganglionic lesion Zone II (1I) Rupture See Fig. 1, for definitions of terms

248 203 33 12 86 38 48

Lesions in continuity, of degenerative type, are subclassified into three categories, as second; third; or fourth-degree injury, according to Sunderland. For lesions in continuity, treatment involves longitudinal incisions made into the fibrous layers of the epineurium under microscopic guidance. If the fascicular pattern is undamaged and the fascicles are soft, the lesion is subclassified as a second-degree injury and should be treated conservatively. When the fascicular pattern is undamaged but fascicles are fibrotic, the lesion is subclassified as a third-degree injury. From a third-

Fig. 2. Intraoperative electrophysiological examinations to differentiate postganglionic from preganglionic lesions. For a postganglionic lesion, both somatosensory evoked potential (SEP) and evoked spinal cord potential (ESCP) are recorded, but not the nerve action potential (NAP), which is absent. For a preganglionic lesion, SEP and ESCP are both absent and NAP can be recorded. St., Stimulation

74

A. Nagano: Treatment of brachial plexus injury

Table 4. Patients in whom nerve grafting was performed for brachial plexus lesions Root/Trunk lesion (37 patients)

Cord lesion (13 patients)

Ax 1 SS (21 patients)

15–42 (22)

15– 49 (30)

15–38 (21)

Age at operation (years) Time interval between injury and operation (months) Follow-up period (months)

1– 8 (3)

1–9 (4)

24– 60 (30)

1– 6 (3)

24–120 (43)

24–60 (31)

Mean values are given in parentheses Ax 1 SS, Combined injury to the axillary and suprascapular nerves

Table 5. Results of nerve grafting in root or trunk lesion (MRC grading)

Table 7. Results of nerve grafting in suprascapular nerve lesions (MRC grading)

Number of patients

Infraspinatus Deltoid Biceps Triceps

M3

M2

M1

M0

Total

1 2 12

3 5 8 4

1 3 3 3

4 7

2 4 2 3

11 21 25 11

Table 6. Results of nerve grafting in cord lesion (MRC grading) Number of patients

Deltoid Biceps Triceps

and

Number of patients

M4

1

axillary

M4

M3

M2

M1

M0

Total

3 2 1

3 5 2

1 1 1

1

1

9 8 4

degree injury, there would be some recovery through conservative treatment but useful regeneration is not expected. If a lesion in the upper roots or upper trunk, in the suprascapular, axillary, or musculocutaneous nerve is a third-degree injury, it is better treated by nerve grafting. If the fascicular pattern is lost, the lesion is subclassified as a fourth-degree injury; in such instances, the nerve is diagnosed as ruptured and nerve grafting should be performed. Neurolysis has been reported to be effective. However, for a traction lesion it is very difficult to estimate the efficacy of the neurolysis. Narakas21 has reported that neurolysis was effective in patients with longstanding injury only when scar tissue was observed around the nerve or inside the epineurium between the fascicles, preventing recovery or causing pain. However, the results seen after neurolysis may have been brought about by spontaneous recovery, and may not be related to neurolysis.

Infraspinatus Deltoid

M5

M4

M3

M2

M1

Total

1 3

7 4

6 7

5 5

2 2

21 21

Nerve grafting Nerve grafting is indicated for ruptures in the root with positive ESCP or SEP potential, in the trunk and in the cord. Between 1982 and 1992, we performed nerve grafting in 71 patients who were followed for more than 2 years after the operation. The root or trunk was ruptured in 37 patients, the cord in 13 patients, and the axillary and suprascapular nerves in 21 patients (Table 4). The results are shown in Tables 5, 6, and 7. After nerve grafting for the biceps, the same results were obtained for a root/trunk lesion as for a cord lesion. However, the results for combined injury in the axillary and the suprascapular nerves were better than those for a root/trunk lesion, both for the infraspinatus (P , 0.05) and the deltoid (P , 0.01) (Table 8). This difference is attributed to the condition of the part of the nerve distal from the injured site. A brachial plexus injury is usually caused by a traction trauma, in which the neural tissues are torn out along the course of the nerve. Double- or triple-level lesions are encountered frequently in the suprascapular nerve in combined ruptures in the axillary and the suprascapular nerves (Fig. 3).24 The musculocutaneous nerve can be explored easily along its whole course, so the injured site is easily determined, but exploration of the whole course of the axillary and the suprascapular nerves is not so easy. In patients with poor results for the infraspinatus or deltoid, the suprascapular and the axillary nerves are likely to have been ruptured in the distal portion; this portion is not

A. Nagano: Treatment of brachial plexus injury

75

Table 8. Results after nerve grafting (MRC grading) Lesion Root/Trunk Number of patients $M3 Infraspinatus Deltoid Biceps Triceps

11 21 25 11

36% 33 80 36

Cord Number of patients $M3 9 8 4

67% 88 75

AX 1 SS Number of patients $M3 21 21

67%* 67**

* P , 0.05; ** P , 0.01 (Mann-Whitney U-test), compared to root/trunk lesion Ax 1 SS, Combined injury to the axillary and suprascapular nerves

Table 9. Elbow flexion after nerve transfer to the musculocutaneous nerve (MRC grading)

Intercostal nerve15 Spinal accessory nerve28 Cervical plexus2 Phrenic nerve5 Contralateral C76 Ulnar nerve23

Fig. 3. Rupture sites of the suprascapular nerve in 19 patients with combined rupture of the axillary and suprascapular nerves. Notch, the suprascapular notch

revealed by ordinary exploration of the brachial plexus. Accordingly, to obtain better results, exploration of the brachial plexus should be extended distally as far as possible. A distal rupture should be suspected when there is a displaced fracture in the humerus or in the glenoid, or when there is bruising along the course of a neurovascular bundle. Generally, the results from shorter nerve grafts are better than those from longer nerve grafts.9 However, in our series, the length of the grafted nerve to restore the function of the deltoid or biceps bore no relation to the results.26 In our experience even with grafts as long as

Number of patients

More than M3

221 184 12 40 3 8

72% 75 50 80 67 75

25 cm, the function of the biceps was restored to more than M3 (MRC grading). Nerve trunks or cutaneous nerves can be used as vascularized nerve grafts for nerve grafting. Better results had been expected with the vascularized procedure, but the results were reported to be similar to those with free sural nerve grafts.1 A free sural nerve thus appears to be quite sufficient for nerve grafting. Results after nerve grafting for the forearm muscles have been very poor.26

Nerve transfer Nerve transfer or tendon transfer is defined as a procedure to restore upper extremity function in patients with a preganglionic lesion. Carlstedt and Noren3 recently repaired an intraspinal rupture in the roots by nerve grafting, gaining considerable functional improvement. However, the efficacy of their procedure has not yet been established. If there is no suitable donor for tendon transfer, then nerve transfer is the only method of treatment. Donor nerves available for nerve transfer include the intercostal nerve,15 spinal accessory nerve,28 cervical plexus,2 phrenic nerve,5 contralateral C7 root,6 and the ulnar nerve.23 The recovery rate of elbow flexion to

76

A. Nagano: Treatment of brachial plexus injury

more than M3 for each of these six donors has been similar (except for cervical nerve donors), with 70%– 80% of the patients recovering more than M3 power of elbow flexion (Table 9). The selection of the donor nerve for the restoration of elbow flexion must take into account the disadvantages of each procedure. Cervical plexus nerve transfer involves a high risk of unstable scapula, especially if used in combination with spinal accessory nerve transfer. If the serratus anterior function is already poor or nil, then any scapulohumeral function recovered by neurotization is largely redundant.22 The phrenic nerve is a pure motor nerve, but transfer of this nerve puts at risk otherwise normal respiratory functions. Gu et al.5 and Yamamoto et al.30 have reported that postoperative pulmonary function after phrenic nerve transfer was maintained within the normal range. However, until a long-term follow-up study has been completed, it is not yet clear whether the respiratory function would be maintained as the patients got older. Therefore, at present, transfer of this nerve should be avoided. The contralateral C7 root is used only when other donor nerves are not available. It is usually employed as a

donor nerve for free muscle transplantation to restore hand function. The ulnar nerve is indicated only for a C5-C6 root avulsion with normal C7-T1. Use of the spinal accessory nerve has the advantage that it is a pure motor nerve with functional characteristics similar to those of the musculocutaneous and the suprascapular nerves, which therefore facilitates postoperative motor rehabilitation.22 There are some disadvantages with the use of the spinal accessory nerve. Sometimes the middle and lower parts of the trapezius become paralyzed by this transfer. If the serratus anterior is already paralyzed, the stability of the scapula is reduced, so that even when spinati and deltoid function recover, useful shoulder function cannot be fully regained, since the stability of the scapula remains weakened. Intercostal nerve transfer has no known disadvantage, but it cannot be employed in patients with multiple rib fractures. Overall, intercostal nerve transfer is recommended to restore elbow flexion. We sutured two intercostal nerves to the musculocutaneous nerve in each of 247 patients aged more than 16 years, and 221 of these patients were followed-up for more than 2 years after the operation (Table 10). The

Table 10. Patients in whom intercostal nerve transfer was performed for brachial plexus injury Recipient nerve (number of patients) Mc (n 5 221) Age at operation (years) Time interval between injury and operation (months) Follow-up period (months)

Ax (n 5 7)

Rad (n 5 15)

Med (n 5 10)

16–50 (22)

19–25 (30)

2–25 (18)

5–34 (18)

1–25 (4)

2–7 (4)

1–11 (5)

1– 6 (4)

24–38 (27)

24– 84 (44)

26– 40 (34)

24– 458 (47)

Mean values are given in parentheses Mc, musculocutaneous nerve; ax, axillary nerve; rad, radial nerve; med, median nerve

Table 11. Results of intercostal nerve transfer to the musculocutaneous nerve to restore elbow flexion with respect to age at operation in those older than 16 years (MRC grading) Number of patients Age (years)

M4

M3

M2–M0

Total

16–20 21–30 31–40 41–50 Total

36 21 3

58 36 4 2 100

34 18 6 3 61

128 75 13 5 221

60

Table 12. Results of intercostal nerve transfer to the musculocutaneous nerve to restore elbow flexion compared between those aged 30 years or younger and those older than 30 years (MRC grading) Number of patients Age (years)

M4

M3

M2–M0

Total

16–30 31–50 Total

57 3 60

94 6 100

52 9 61

203 18 221

A. Nagano: Treatment of brachial plexus injury

77

power of elbow flexion was restored to more than M4 in 60 patients, to M3 in 100, to M2 or M1 in 46, and to M0 in 15. Clinically, 160 (72.4%) of the 221 patients who underwent the operation could voluntarily flex the elbow joint against gravity. The results were analyzed in relation to two factors — age at operation and time interval between the injury and the operation. There were no significant differences among the results for each 10-year age-group (Table 11), but the results for those patients younger than 30 years were better than those of patients older than 31 years (P , 0.05, MannWhitney U-test) (Table 12). With respect to the time interval between the injury and the operation (Table 13), the results for patients operated within 3 months after injury were best, and the shorter the interval, the better the results (P , 0.0001, Spearman rank correlation). The results for patients operated within 6 months of injury were better than those for patients operated later than 6 months after injury (P , 0.0001, MannWhitney U-test) (Table 14). These results showed that the best indications for intercostal nerve transfer to restore elbow flexion were in patients who were younger than 30 years and operated on within 6 months after injury. For patients older than 31 years, it may be better

Table 13. Results of intercostal nerve transfer to the musculocutaneous nerve to restore elbow flexion with respect to time interval between injury and operation in those aged 16 years or older (MRC grading) Number of patients Time interval (months)

M4

M3

M2–M0

Total

0–3 4–6 7–9 10–12 13–25 Total

33 22 5

45 45 6 2 1 99

16 23 10 7 4 60

94 90 21 9 5 219

60

Table 14. Results of intercostal nerve transfer to the musculocutaneous nerve to restore elbow flexion (MRC grading): Operations performed within 6 months of injury versus those performed later than 6 months

to use three intercostal nerves, and for patients in whom the time interval between the injury and operation is more than 9 months, we have employed a free muscle transplantation with intercostal nerve transfer. Hara et al.8 used the rectus femoris for this procedure in 17 patients. The power of elbow flexion was restored to M4 in 10 patients, and to M3 in 4. Overall, 14 (83%) of the 17 patients recovered voluntary flexion of the elbow joint against gravity. To restore shoulder function, spinal accessory nerve transfer21 and cervical plexus nerve transfer2 to the suprascapular nerve have each shown rather good results. Narakas22 performed spinal accessory nerve transfer in 23 patients. Abduction of more than 50° was recovered in 39% of the patients, and external rotation of more than 30°, starting with the forearm against the chest, was recovered in 30%. We have performed intercostal nerve transfer to the axillary nerve in combination with phrenic nerve transfer to the suprascapular nerve in 7 patients (Table 10), and 6 (88%) of them regained more than 50° abduction and more than 30° external rotation, starting with the forearm against the chest.20 These results were better than those from either spinal accessory nerve transfer alone or from cervical plexus nerve transfer alone, although the number of our patients was small. The better results were probably because we achieved neurotization in both the axillary and the suprascapular nerves. To restore shoulder function, both the suprascapular and axillary nerves should be neurotized. If the serratus anterior is also paralyzed, the long thoracic nerve should be neurotized, using cervical plexus nerve transfer to stabilize the scapula. To restore hand function, we have performed intercostal nerve transfer to the radial or the median nerve (Table 10).20 In 15 patients, two intercostal nerves were sutured to the radial nerve. The power of the extensor carpi radialis and the extensor digiti communis were each restored to M3 in 2 patients, while the power of the extensor pollicis longus was M0 in all 15. In a further 10 patients, two intercostal nerves were sutured to the

Table 15. Results after intercostal nerve transfer to the median nerve (MRC grading) Number of patients

Number of patients

Operation within 6 months of injury Operation more than 6 months after injury

M4

M3

M2–M0

Total

55

90

39

184

5

9

21

35

M3 M2 M1 M0

FCR

FPL

FDP

FDS

APB

3 0 3 4

1 1 0 8

2 1 2 5

1 2 1 6

0 1 0 9

FCR, Flexor carpi radialis; FPL, flexor pollicis longus; FDP, flexor digitorum profundus; FDS, flexor digitorum superficialis; APB, abductor pollicis brevis

78

median nerve to restore wrist and finger flexion and sensation in the fingers. Motor recovery was poor (Table 15). With respect to sensory recovery, protective sensation was restored in fingers innervated by the median nerve in 90% of the 10 patients. Hara et al.8 have performed free muscle transplantation in combination with intercostal nerve transfer to restore wrist extension. They used the gracilis for the transplanted muscle in 51 patients. The power of the wrist extension was restored to M4 in 12 patients, and to M3 in another 17 patients. Overall, 29 (57%) of their patients were able to extend the wrist joint against gravity. With the aid of a flexor hinge splint or of tenodesis of the flexor digitorum profundus onto the radius, patients were able to grasp or pick up small objects using an opponens splint and recovered the use of the hands for activities of daily living. However, several problems remained, e.g., grasping or pinching power was weak and control of the elbow joint was poor. Also, elbow flexion and wrist extension could not be separated, since elbow flexion was restored by the same intercostal nerve transfer. To restore more useful function, restoration of elbow extension and finger flexion are necessary. Doi et al.4 developed a new procedure to restore finger flexion and extension, as well as elbow flexion and extension, through a double free-muscle transfer and multiple nerve transfers using the spinal accessory and intercostal nerves. Elbow flexion was restored to more than M3 in all ten patients, elbow extension in two, finger flexion in six, and finger extension in five patients. With respect to functional recovery, five patients were able to grasp light objects and three of these patients were each able to lift a box weighing from 5 to 13 kg with both hands.

Timing of tendon transfer The courses of recovery of the deltoid, biceps, and triceps after conservative treatment have been similar.13 In patients in whom conservative treatment restored muscular strength to at least M1 by 6 months after injury or to at least M2 by 9 months, final recovery was more than M4. In nearly all patients in whom strength was M0 at 12 months or M2 at 18 months, final recovery was less than M2. The strength measured at 24 months was similar to the final outcome. All the forearm muscles showed a recovery course very similar to that mentioned above. We could not determine any prognostic index less than 6 months after injury. Therefore, we believe that reconstructive surgery, such as tendon transfer, is indicated when the muscular power is only M0 12 months after injury or only M2 18 months after injury.

A. Nagano: Treatment of brachial plexus injury

Reconstructive surgery When neither nerve grafting nor nerve transfer are indicated or when the results of either are poor, function can be restored by tendon transfer or arthrodesis. Procedures to restore shoulder function include either multiple muscle transfer27 or shoulder arthrodesis.7,10,16 Multiple muscle transfer is possible only for patients with C5-C6 root injury and normal C7-T1. Our results for multiple muscle transfer were good in children, but in adults, more than 90° elevation was seldom achieved and power was weak. Shoulder arthrodesis is able to restore powerful elevation, but has the disadvantage of causing a larger restriction in shoulder joint rotation. Accordingly, multiple muscle transfer is indicated for children and desk workers with a C5-C6 root injury and normal C7-T1, while shoulder arthrodesis is indicated for manual laborers with a C5C6 root injury and normal C7-T1. Shoulder arthrodesis is also indicated for other types of injury except in children. Various procedures have been reported for shoulder arthrodesis. These are categorized as: (1) screws and plaster casts,7 (2) plates,10 and (3) external fixation.16 External fixation has a considerable advantage in that it allows postoperative adjustment of joint position, which cannot be offered by internal fixation with a screw or plate, as it is almost impossible to confirm the optimum joint position intraoperatively. In an upper-type brachial plexus injury, the angle is set at 40° abduction, 30° flexion, and sufficient rotation for the hand to reach the mouth when the elbow is flexed. In total-avulsion brachial plexus injury the aim of joint fusion is to enable the patient to hold an object between the chest and the upper arm, so the joint is fixed at 20° abduction, 15° flexion, and 50° internal rotation. To restore elbow flexion, the donor muscles include the forearm flexor-pronator group, the latissimus dorsi, and the pectoralis major.18 Latissimus dorsi transfer is indicated only for a C5-C6 lesion with normal C7-T1, but in this type of injury the latissimus dorsi is usually used to restore external rotation of the shoulder joint. Since there is a marked operation scar after pectoralis major transfer, we recommend forearm flexor-pronator group transfer to restore elbow flexion. To restore hand function, surgical reconstruction is necessary for a C5-C8 avulsion with normal T1, and for any residual dysfunction after conservative treatment or nerve repair. In patients with a C5-C8 avulsion with normal T1, wrist and finger extension are restored by tenodesis of the extensor digiti communis onto the radius (Fig. 4).25 In patients with residual dysfunction, hand function is restored by an ordinary tendon transfer.

A. Nagano: Treatment of brachial plexus injury

79

c

a

Fig. 4a–c. Tenodesis of the extensor digitorum communis to the radius in avulsion of C5-C8 with normal T1 to restore wrist and finger extension. a Preoperative finger flexion and extension; b operative procedure; c postoperative finger flexion and extension

b

Strategy for treatment of root-avulsion injury In patients with total root avulsion, elbow flexion is restored by intercostal nerve transfer, and shoulder function is restored by shoulder arthrodesis. Hand function is restored by an intercostal nerve transfer to the median nerve, or by free muscle transplantation in combination with nerve transfer.17 In patients with a C5 rupture with avulsion in C6-T1, elbow flexion is restored by an intercostal nerve transfer

onto the musculocutaneous nerve. Shoulder function is restored by nerve grafting between C5 and the suprascapular and axillary nerves. Hand function is restored as in total root avulsion. In patients with avulsion in C5-C8 with normal T1, elbow flexion is restored by an intercostal nerve transfer. Shoulder function is restored by multiple nerve transfers involving a spinal accessory nerve transfer onto the suprascapular nerve, an intercostal nerve transfer onto the axillary nerve, and a cervical plexus

80

nerve transfer onto the long thoracic nerve, or by shoulder arthrodesis with a cervical plexus nerve transfer onto the long thoracic nerve. Wrist and finger extension are restored by tenodesis of the extensor digiti communis to the radius. In patients with avulsion of C5-C7 with normal C8-T1, elbow flexion is restored by an intercostal nerve transfer, and shoulder function is restored as for an avulsion in C5-C8 with normal T1. Wrist and finger extension are restored by a tendon transfer, if necessary. In patients with avulsion in C5-C6 with normal C7T1, elbow flexion is restored by a tendon transfer, and shoulder function is restored by double nerve transfers involving a spinal accessory nerve transfer onto the suprascapular nerve and an intercostal nerve transfer onto the axillary nerve, or by multiple muscle transfers. In patients in whom the interval from the injury to the operation is more than 12 months, function may be restored by a tendon transfer or by arthrodesis. Otherwise, function may be restored by free muscle transplantation in combination with nerve transfers.

References 1. Birch R, Dunkerton M, Bonney G, et al. Experience with the free vascularised ulnar nerve graft in repair of supraclavicular lesions of the brachial plexus. Clin Orthop 1988;237:96–104. 2. Brunelli G. Neurotization of avulsed roots of the brachial plexus by means of anterior nerves of the cervical plexus. In: Terzis JK, editor. Microreconstruction of nerve injuries. Philadelphia: WB Saunders, 1987:435–45. 3. Carlstedt T, Noren G. Repair of ruptured spinal nerve roots in a brachial plexus lesion. Case report. J Neurosurg 1995;82:661–3. 4. Doi K, Sakai K, Kuwata N, et al. Double free-muscle transfer to restore prehension following complete brachial plexus avulsion. J Hand Surg Am 1995;20:408–14. 5. Gu Y-D, Wu M-M, Zhen Y-L, et al. Phrenic nerve transfer for brachial plexus motor neurotization. Microsurgery 1989;10:287–9. 6. Gu Y-D, Zhang D-S, Chen J-G, et al. Seventh cervical nerve root transfer from the contralateral healthy side for treatment of brachial plexus root avulsion. J Hand Surg Br 1992;17:518–21. 7. Hara T, Tsuyama N. Shoulder arthrodesis (in Japanese). Shujyutsu (Operation) 1979;33:865–75. 8. Hara T, Takahasi M, Akasaka Y, et al. Combined surgery of free muscle transplantation and intercostal nerve crossing as a reconstructive procedure for elbow flexion and wrist extension in brachial plexus injuries (in Japanese). J Jpn Soc Surg Hand 1986;3:238–42. 9. Hentz VR, Narakas A. The results of microneurosurgical reconstruction in complete brachial plexus palsy: Assessing outcome and predicting results. Orthop Clin North Am 1988;129:107–14.

A. Nagano: Treatment of brachial plexus injury 10. Müller ME, Allgower M, Schneider R, et al. Manual of internal fixation. Techniques recommended by the AO group. 2nd ed. Berlin Heidelberg Tokyo New York: Springer, 1979:384–5. 11. Murase T, Kawai H, Ohta I, et al. Intraoperative diagnosis in brachial plexus palsy using evoked spinal cord potentials (in Japanese). Rinshou-Seikeigeka (Clin Orthop Surg) 1992;27:1237–45. 12. Nagano A, Tsuyama N, Hara T. Diagnosis and treatment of brachial plexus injuries. In: Itami Y, Nishio Y, editors. Seikeigeka Mook 19, Peripheral nerve injury (in Japanese). Tokyo: Kanehara Shuppan, 1981:187–202. 13. Nagano A, Tsuyama N, Hara T, et al. Brachial plexus injuries. Prognosis of postganglionic lesions. Arch Orthop Trauma Surg 1984;102:172–8. 14. Nagano A. Brachial plexus injuries — diagnosis (in Japanese). J Jpn Soc Surg Hand 1988;4:839–43. 15. Nagano A, Tsuyama N, Ochiai N, et al. Direct nerve crossing with the intercostal nerve to treat avulsion injuries of the brachial plexus. J Hand Surg Am 1989;14:980–5. 16. Nagano A, Okinaga S, Ochiai N, et al. Shoulder arthrodesis by external fixation. Clin Orthop 1989;247:97–100. 17. Nagano A. Treatment of brachial plexus injuries (in Japanese). Shujyutsu (Operation) 1989;43:1287–98. 18. Nagano A, Ochiai N, Okinaga S. Restoration of elbow flexion in root lesions of brachial plexus injuries. J Hand Surg Am 1992;17:815–21. 19. Nagano A. Diagnosis and treatment of brachial plexus injuries — diagnosis (in Japanese). J Jpn Soc Surg Hand 1994;10:1022–30. 20. Nagano A, Yamamoto S, Mikami Y. Intercostal nerve transfer to restore upper extremity functions after brachial plexus injury. Ann Acad Med Singapore 1995;24(Suppl):42–5. 21. Narakas A. Brachial plexus surgery. Orthop Clin North Am 1981;12:303–23. 22. Narakas, A. Brachial plexus lesions. In: Leung PC, editor. Microsurgery in orthopaedic practice. Singapore: World Scientific, 1995:188–254. 23. Oberlin C, Beal D, Leechavengvongs S, et al. Nerve transfer to biceps muscle using a part of the ulnar nerve for C5-C6 avulsion of the brachial plexus: Anatomical study and report of four cases. J Hand Surg Am 1994;19:232–7. 24. Ochiai N, Nagano A, Okinaga S, et al. Brachial plexus injuries. Surgical treatment of combined injuries of the axillary and suprascapular nerves. (in Japanese). J Jpn Soc Surg Hand 1988;5:151–5. 25. Ochiai N, Nagano A, Yamamoto S, et al. A tenodesis of extensor digitorum in treatment of brachial plexus injuries involving C5, 6, 7 and 8 nerve roots. J Hand Surg Br 1995; 20:671–4. 26. Ochiai N, Nagano A, Sugioka H, et al. Nerve grafting in brachial plexus injuries. Results of free grafts in 90 patients. J Bone Joint Surg Br 1996;78:754–8. 27. Saha AK. Surgery of the paralyzed and flail shoulder. Acta Orthop Scand (Suppl). 1967;97:5–90. 28. Songcharoen P, Mahaisavariya B, Chotigavanich C, Spinal accessory neurotization for restoration of elbow flexion in avulsion injuries of the brachial plexus. J Hand Surg Am 1996;21:387–90. 29. Sugioka H, Tsuyama N, Hara T, et al. Investigation of brachial plexus injuries by intraoperative cortical somatosensory evoked potentials. Arch Orthop Traumat Surg 1982;99:143–51. 30. Yamamoto S, Nagano A, Ochiai N, et al. Respiratory functions after intercostal nerve and/or phrenic nerve transfer (in Japanese). J Jpn Soc Surg Hand 1993;9:884–6.