Transfer of the spinal accessory nerve to the suprascapular nerve is a common procedure, performed to reestablish shoulder motion in patients with total brachial plexus palsy. However, the results of this procedure remain largely unknown.
Over an 11-year period (2002–2012), 257 patients with total brachial plexus palsy were operated upon in the authors' department by a single surgeon and had the spinal accessory nerve transferred to the suprascapular nerve. Among these, 110 had adequate follow-up and were included in this study. Their average age was 26 years (SD 8.4 years), and the mean interval between their injury and surgery was 5.2 months (SD 2.4 months). Prior to 2005, the suprascapular and spinal accessory nerves were dissected through a classic supraclavicular L-shape incision (n = 29). Afterward (n = 81), the spinal accessory and suprascapular nerves were dissected via an oblique incision, extending from the point at which the plexus crossed the clavicle to the anterior border of the trapezius muscle. In 17 of these patients, because of clavicle fractures or dislocation, scapular fractures or retroclavicular scarring, the incision was extended by detaching the trapezius from the clavicle to expose the suprascapular nerve at the suprascapular fossa. In all patients, the brachial plexus was explored and elbow flexion reconstructed by root grafting (n = 95), root grafting and phrenic nerve transfer (n = 6), phrenic nerve transfer (n = 1), or third, fourth, and fifth intercostal nerve transfer. Postoperatively, patients were followed for an average of 40 months (SD 13.7 months).
Failed recovery, meaning less than 30° abduction, was observed in 10 (9%) of the 110 patients. The failure rate was 25% between 2002 and 2004, but dropped to 5% after the staged/extended approach was introduced. The mean overall range of abduction recovery was 58.5° (SD 26°). Comparing before and after distal suprascapular nerve exploration (2005–2012), the range of abduction recovery was 45° (SD 25.1°) versus 62° (SD 25.3°), respectively (p = 0.002). In patients who recovered at least 30° of abduction, recovery of elbow flexion to at least an M3 level of strength increased the range of abduction by an average of 13° (p = 0.01). Before the extended approach, 2 (7%) of 29 patients recovered active external rotation of 20° and 120°. With the staged/extended approach, 32 (40%) of 81 recovered some degree of active external rotation. In these patients, the average range of motion measured from the thorax was 87° (SD 40.6°).
In total palsies of the brachial plexus, using the spinal accessory nerve for transfer to the suprascapular nerve is reliable and provides some recovery of abduction for a large majority of patients. In a few patients, a more extensive approach to access the suprascapular nerve, including, if necessary, dissection in the suprascapular fossa, may enhance outcomes.