Intraoperative Neural Monitoring for Early Vocal Cord Function Assessment After Thyroid Surgery: A Systematic Review and Meta-Analysis

Objective This study evaluated the diagnostic accuracies of various forms of intraoperative neural monitoring (IONM) in terms of predicting vocal cord palsy after thyroidectomy. Methods Two authors independently reviewed the six databases (PubMed, the Cochrane database, Embase, the Web of Science, SCOPUS, and Google Scholar) from their dates of inception to March 2021. Intraoperative electromyographic neuromonitoring (IONM) was compared with laryngoscopic detection (the reference method). True-positive, true-negative, false-positive, and false-negative data were extracted from each study. Methodological quality was evaluated using the Quality Assessment of Diagnostic Accuracy Studies ver. 2 tool. Results Twenty-seven prospective or retrospective studies were included. The diagnostic odds ratio for IONM was 152.9623 ([95% confidence interval [75.4249; 310.2085]. The area under the summary receiver operating characteristic curve was 0.966. The sensitivity, specificity, negative predictive value, and positive predictive value were 0.8219 ([0.6862; 0.9069]), 0.9783 ([0.9659; 0.9863]), 0.9943 ([0.9880; 0.9973]), and 0.5523 ([0.4458; 0.6542]), respectively. The correlation between sensitivity and false-positive rate was 0.200, indicating the absence of heterogeneity. Subgroup analysis showed that the diagnostic accuracies of the continuous IONM was higher than those of intermittent IONM, and recent publications (> 2011) was higher than early publication (< 2010). Conclusions As the technology and knowledge of IONM have been accumulated and progressed over the past decades, the predictive value of IONM in postoperative vocal cord palsy has also improved. Moreover, the advances of continuous IONM technology could make a breakthrough in vocal cord evaluation after thyroid surgery.