Functional restoration of elbow flexion in nonobstetric brachial plexus injuries: A meta-analysis of nerve transfers versus grafts.

Abstract

This systematic literature review was performed according to the PRISMA guidelines. A random-effects model meta-analysis was conducted, and the I-square was used to assess heterogeneity. The Medical Research Scale (MRC) score was used to assess the efficacy of the procedures. Nine studies comprising 490 patients overall were identified. In the pooled analysis, functional recovery of elbow flexion defined as MRC ≥ M3, was superior in the transfer (N = 272/350, 77.7%) compared to the graft group (N = 99/140, 70.7%); however statistical significance was not reached (OR: 1.95; 95%CI: 0.79-4.83; I2 : 58.8%). However, the odds for successful restoration of elbow flexion (MRC≥M3) were significantly higher when the ulnar (OR:12.20; 95%CI:3.05-48.80; I2 :0%) or pectoral nerves (OR: 9.69; 95% CI: 1.83-51.25; I2 : 0%) were used as healthy donors for the transfer compared to the graft procedures. Results between the two groups were similar when the intercostal, spinal accessory, thoracodorsal, contralateral C7 and phrenic nerves were used as donors for the transfer procedures. The ulnar or pectoral nerve transfer to musculocutaneous is associated with statistically significant superior rates of elbow flexion recovery as compared to graft. No differences were identified in the pooled analysis or the subgroups of other donors used in nerve transfers. Future randomized studies or prospective cohorts are needed to validate our results.