The Utility of Multi-Modal Intra-Operative Neurophysiological Monitor in Corrective Surgery for Thoracic Tuberculosis with Kyphosis

Abstract

Background This study assesses performance and utility of motor evoked potentials and somatosensory evoked potentials during corrective surgery to thoracic tuberculosis with kyphosis (TTK), as well as corresponding risk factors. Methods Patients diagnosed with TTK and underwent TTK corrective surgery from 2012 to 2018 were included. Relevant Data were retrospectively collected. Measures of intraoperative neurophysiological monitoring (IONM) performance were calculated. A receiver operating characteristic (ROC) curve and area under ROC curve (AUC) were deployed to identify the diagnostic accuracy of potential recovery. Univariate and multivariate analysis were performed to determine risk factors correlated with IONM alerting. Results 68 patients were included. The mean age was 31.7±20.3 years (3-78 years). IONM alerting occurred in 12 surgeries (12/68, 17.6%), of which 6 were somatosensory evoked potential (SSEP) alerting, 2 motor evoked potential (MEP) alerting, and 4 combinations of both SSEP and MEP. There were 10 posterior vertebral column resections (PVCRs) and 2 pedicle subtraction osteotomies (PSOs) with 1.83±1.19 vertebra resected. Four (4/68, 5.89%) patients were identified with presence of postoperative neurological deficits (PNDs). Calculated measures of performance of MEP and SSEP were as follows: sensitivity of 0.75, specificity of 0.86, positive predictive value (PPV) of 0.25, and negative predictive value (NPV) of 0.98 respectively. The AUC of evoked potential recovery in diagnosing a PND was 0.884. On logistic regression analysis, sever kyphosis was determined as the independent risk factor for IONM alerting. Conclusions Multimodal IONM can benefit 66.7% participants from neural impingement under appropriate intervention after IONM alerting. Corrective surgery in TTK is at risk of spinal injury, especially in severe kyphosis requiring three-column osteotomy.