BACKGROUND: Curve-specific risk factors for intraoperative neuromonitoring (IONM) alerts during surgery for adolescent idiopathic scoliosis (AIS) remain unclear.
METHODS: We identified 5,206 patients with AIS who underwent spinal fusion with multimodal IONM from a prospective database. Alerts were defined as ≥50% amplitude decrease in somatosensory evoked potentials (SSEPs) and/or transcranial motor evoked potentials (TcMEPs), or >10% SSEP latency increase. Demographics, radiographic parameters, surgical variables, triggering events, interventions, and outcomes were compared between the 252 patients with and 4,954 without an alert.
RESULTS: IONM alerts occurred in 5% of patients. Alert rates varied by Lenke classification, with Type-4 curves demonstrating the highest risk (8%; relative risk [RR] = 1.61, 95% confidence interval [CI], 1.15 to 2.24) and Type 5 demonstrating the lowest risk (2%; RR = 0.45, 95% CI, 0.25 to 0.82). Patients with an alert had a larger mean preoperative major Cobb angle (66.7° ± 15.0° versus 56.1° ± 12.1°, p < 0.001) and more frequent use of Ponte osteotomies (79% versus 66%, p < 0.001). Multivariable analysis demonstrated the major Cobb angle to be the leading independent risk factor for alerts (p < 0.001). Hypotension was the leading triggering event (23%), and raising blood pressure was the most common intervention overall (52%). Neurologic deficits occurred in 4% of patients with an alert versus <1% of those without an alert (p < 0.001). Only 12 patients (<1%) experienced neurologic deficits, with 2 having permanent deficits. At 2-year follow-up, the major Cobb angle correction did not differ significantly between the groups with and without alerts, alert status did not independently affect correction (p = 0.058), and the groups achieved similar outcomes (73% versus 71% correction, respectively).
CONCLUSIONS: IONM alerts occurred in 5% of patients with AIS and were independently associated with a larger major Cobb angle. Fewer than 1% of patients sustained a permanent neurologic injury, and alerts did not appear to compromise 2-year correction.
LEVEL OF EVIDENCE: Prognostic Level III. See Instructions for Authors for a complete description of levels of evidence.
Read the article: PubMed · Publisher (DOI)