Systematic review with meta-analysis of intraoperative neuromonitoring during thyroidectomy

Introduction: Intraoperative neuromonitoring (IONM) has been a commonly used technology during thyroid surgery aimed at reducing the incidence of recurrent laryngeal nerve palsy (RLNP), which is a severe complication and leads to significant impacts on a patient's life. In order to give a comprehensive assessment for potential benefits and disadvantage of IONM, this meta-analysis and systematic review discusses RLNP rate, predictive power, continuous intraoperative neuromonitoring (CIONM), and emphasises application during thyroid cancer surgeries. Methods: A literature search was performed in the following electronic databases: PubMed, Embase, and the Cochrane library from January 1, 2004 to July 30, 2016. After applying inclusion and exclusion criteria, 24 studies, including four prospective randomised trials, were selected. Heterogeneity of studies was checked by the Cochran Q test. Publication bias was assessed by funnel plots with Egger's linear regression test of asymmetry. Odds ratio (OR) was calculated by random effects model. Results: Overall, 9203 patients and 17,203 nerves at risk (NAR) were included. Incidence of overall, transient, and persistent RLNP in IONM group were, respectively, 3.15%, 1.82%, and 0.67%, whereas for the ID group, they were 4.37%, 2.58%, and 1.07%. The summary OR of overall, transient, and persistent RLNP compared using IONM and ID were, respectively, 0.81 (95% CI 0.66-0.99), 0.76 (95% CI 0.61-0.94), and 0.78 (95% CI 0.55-1.09). Conclusions: The presented data showed benefits of reducing RLNP rate by using IONM, but without statistical significance for persistent RLNP rate. For patients with thyroid cancer who undergo total thyroidectomy, using IONM may improve the outcome by reducing amount of residual thyroid tissue. However, no benefits were found for thyroid reoperation; visual identification and careful dissection remain standard for this challenging procedure. In addition, the relative low positive predictive power indicated intermittent intraoperative neuromonitoring (IIONM) may not be reliable; but CIONM was showed to be a more promising method, with prudent approach. (C) 2017 IJS Publishing Group Ltd. Published by Elsevier Ltd. All rights reserved.