Transcranial motor-evoked potential alert status and 2-year neurological outcomes after postoperative neurological deficits in spinal deformity surgery.

Abstract

To evaluate 2-year neurological outcomes after postoperative neurological deficits (PND) following spinal deformity surgery according to transcranial motor-evoked potential (TcMEP) alert status, and to examine delayed TcMEP deterioration during wound closure. We retrospectively reviewed 1472 patients who underwent spinal deformity surgery with intraoperative neurophysiological monitoring. The primary analysis included 49 patients with PND, stratified by TcMEP alert status. The primary outcome was residual neurological deficit at 2 years. PND occurred in 49/1472 patients (3.3%). Residual deficits were more frequent in alert-negative than alert-positive cases (8/15 [53.3%] vs 6/34 [17.6%], p = 0.017). Among alert-positive PND cases, no residual deficits were observed in responded cases, defined as full-responded or partial-responded cases (0/4 [0.0%]), whereas 6/30 non-responded cases (20.0%) had persistent deficits. In the cohort excluding adult spinal deformity (non-ASD), residual deficits were confined to alert-negative cases (6/7 [85.7%] vs 0/14 [0.0%], p < 0.001). In the exploratory cohort requiring temporary rod removal (n = 37), 13/37 TcMEP alerts (35.1%) occurred during wound closure, at a median of 19 min after deformity correction. Among patients with PND, those without intraoperative TcMEP alerts had poorer 2-year outcomes, particularly in the non-ASD cohort. Alert-negative events may reflect limited detection of isolated nerve root injuries, alert-threshold selection, or monitoring termination before evolving spinal cord compromise becomes detectable. Continued TcMEP monitoring through completion of skin closure may be considered in selected cases when delayed spinal cord compromise is suspected.