The objectives were to review the evidence on whether intraoperative neurophysiological monitoring (IONM) sensitively and specifically detects intraoperative neurologic injury during spine surgery and whether it reduces neurological complications in spine surgery. A systematic literature review and meta-analysis were conducted using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, and studies spanning 1990–2022 were searched. The diagnostic accuracy of somatosensory evoked potentials (SSEPs) and motor-evoked potential (MEP) in predicting postoperative neurologic outcomes in spine surgery was assessed. We also analyzed clinical comparative studies that compared patients who underwent spine surgery with and without IONM to determine the rate of new neurologic events. Fifty studies that provided exact SSEP or MEP changes and postoperative neurological outcomes were analyzed. The features of SSEP changes included low sensitivity (67.7%), high specificity (94.1%), and strong negative predictive value (NPV) (92.1%). For the assessed alarm criteria of a 50% reduction, a 65% reduction, an 80% reduction in amplitude, and a total signal loss, the sensitivity and specificity values for MEP were 87.4%, 100%, 100%, and 100%, and 94.8%, 97%, 94.5%, and 90.5%, respectively. A random effects model was used to assess six research that compared neurological events with and without the use of IONM. The pooled Odds ratio was 0.5746 (confidence interval = 0.48; 0.67), z = 6.637, and the P value was 0.0001. The usefulness of IONM in detecting neurological events during spine surgery is well demonstrated by a growing body of research. However, prospective trials with high-level data are lacking to establish its efficacy in preventing new neurological deficits.