Background The aim of our research was to explore the specific role of intermittent intraoperative neuromonitoring (IONM) versus visualization alone in preventing permanent nerve palsy in thyroid surgery. Methods A systematic review was conducted by searching electronic databases using specific keywords and completed by hand search. The article selection process was carried out by 2 independent investigators using well-defined inclusion and exclusion criteria. Articles evaluating the role of IONM versus visualization alone in preventing recurrent laryngeal nerve palsy were evaluated for inclusion. The relative risk with a confidence interval of 95% was estimated for nonrandomized studies, and risk difference was estimated for randomized control studies. Subgroup meta-analyses were carried out stratifying the studies on the basis of the design and the definition of permanent injury. The heterogeneity among the studies was evaluated with Higgins’ Index. Results A total of 14 articles were included; of these, 10 were nonrandomized studies and 4 were randomized control studies. With regard to the meta-analyses, including nonrandomized studies that defined as permanent an injury persisting for 6 months or 12 months after thyroid surgery, the overall relative risks were, respectively, 0.79 (confidence interval 95%, 0.60–1.05) and 0.75 (confidence interval 95%, 0.43–1.30). As for the meta-analysis including randomized control studies, the risk difference was 0.00 (confidence interval 95%, −0.01 to 0.00). No heterogeneity was found in any of the analyses conducted (Higgins’ Index = 0%). Conclusion Our results show that the use of IONM does not prevent permanent nerve palsy; no significant benefit of IONM over visualization alone in reducing the rate of recurrent laryngeal nerve injuries could be proven. In conclusion, IONM should not be considered the standard care in preventing recurrent laryngeal nerve palsy.