Final results of grafting versus neurolysis in obstetrical brachial plexus palsy.

Background: The authors previously showed that neurolysis in obstetrical brachial plexus palsy resulted in improved function in some patients at 1 year's follow-up. In this study, the hypothesis that the long-term outcome of neuroma-in-continuity resection and nerve grafting yields better results than neurolysis was tested.

Methods: Obstetrical brachial plexus palsy patients treated with primary nerve surgery with a minimum follow-up of 4 years were studied. Patients were classified as undergoing neurolysis (n = 16) or resection and grafting (n = 92) and separated into Erb's or total palsy groups. The Active Movement Scale was used for patient evaluation. Changes in Active Movement Scale scores were analyzed using the Wilcoxon signed rank test. Fifteen movements were tested, and the proportion of patients in each group with scores deemed functionally useful (6 or 7) was compared using McNemar's exact test.

Results: After 4 years' follow-up, Erb's palsy neurolysis patients showed no improvement in function. Conversely, Erb's palsy grafting patients had improved function in seven movements. Total palsy neurolysis patients showed no improvement in function, whereas grafted patients showed improved function in 11 of 15 movements.

Conclusions: Early improvements in function produced by neurolysis in Erb's palsy were not sustained over time. Neuroma-in-continuity resection and nerve grafting for both Erb's and total palsy produced significant improvements in Active Movement Scores and in the proportion of patients demonstrating functionally useful scores. Neurolysis as a complete surgical treatment for obstetrical brachial plexus palsy should be abandoned in favor of neuroma resection and nerve grafting.