Intraoperative neuromonitoring during reverse shoulder arthroplasty.

Background: The aim of this study was to evaluate the risk of nerve injury with neuromonitoring during reverse total shoulder arthroplasty.

Materials: This study included 15 shoulders of 15 patients (11 females and 4 males) who underwent reverse total shoulder arthroplasty. The mean age was 74.8 ± 4.4 years. Nine shoulders had cuff tear arthropathy, 4 had massive rotator cuff tears, 2 had osteoarthritis, and 1 had rheumatoid arthritis. The somatosensory evoked potentials of the median nerve, transcranial motor evoked potentials, and free-electromyograms from 6 upper-extremity muscles were measured intraoperatively. We defined a nerve alert as 50% amplitude attenuation or 10% latency prolongation of the somatosensory evoked potentials and transcranial motor evoked potentials and sustained neurotonic discharge on free-electromyogram.

Results: Thirty-one alerts were recorded in 11 patients. The axillary nerve was associated with 17 alerts. Eleven alerts occurred during the glenoid procedure and 5 alerts occurred during the humeral procedure. One patient who did not recover from the alert of the axillary nerve had clinically incomplete paralysis of the deltoid muscle.

Conclusion: The present findings suggest that the axillary nerve was the nerve most frequently exposed to the risk of injury, especially during glenoid and humeral implantation.