Results of Grafting the Anterior and Posterior Divisions of the Upper Trunk in Complete Palsies of the Brachial Plexus

Purpose In most complete brachial plexus injuries, at least I root still is available for grafting. We report on the results obtained with reconstruction of the brachial plexus using short sural nerve grafts that connect nonavulsed roots to the anterior, posterior, or both divisions of the upper trunk. Methods We prospectively studied 22 young adults with complete brachial plexus palsy who had surgical repair an average of 5 months after trauma. Sural nerve grafts connected the C5 root to the anterior division and the C6 root to the posterior division of the upper trunk. When the C6 root was not available, the posterior division of the upper trunk was repaired by means of a nerve transfer. In all cases except one, the suprascapular nerve was repaired via a nerve transfer. Outcomes were assessed an average of 35 months after surgery, focusing on recovery of muscle strength, categorized using the Medical Research Council scale. We compared the results obtained after a single root graft, either C5 (n = 1, 1) or C6 (n = 1), with those observed after double root grafting (i.e., C5 + C6; n 9). The single case of 3 roots available for grafting was excluded for this comparative study. Results With grafting of the anterior division of the upper trunk, 17 of the 22 patients (n = 15) regained useful pectoralis major and biceps function of at least M3. Grafting the anterior and the posterior divisions of the upper trunk resulted in 18 of the 22 patients (n = 18) recovering shoulder abduction-adduction and either elbow flexion or extension. In only 5 cases (5 of 22 patients), however, was shoulder abduction-adduction achieved with concomitant recovery of both elbow flexion and extension. Grafting the posterior division of the upper trunk did not enhance the recovery of shoulder abduction, but it did restore elbow extension in approximately 6 of the 9 patients. In terms of muscle strength, an average of 2.3 muscles scored M3 or M4 in the single-root group, compared with 3.1 in the C5/C6 group (p < .05). The relative probability of recovering elbow flexion and shoulder adduction did not differ between patients with 1 versus 2 root grafts. The results of nerve transfers to the posterior division and of forearm muscle reinnervation were poor. Conclusions Grafting the divisions of the brachial plexus ensured multiple function reconstruction in 18 of the 22 patients (n = 18). However, only 5 of 22 patients (n = 4) experienced restoration of elbow flexion and extension. (J Hand Surg 2008;33A:1529-1540. Copyright (C) 2008 by the American Society for Surgery of the Hand. All rights reserved.)