AB130. SOH22ABS114. Intra-operative nerve monitoring and recurrent laryngeal nerve injury during thyroid surgery: a Bayesian network meta-analysis of prospective studies

Background: Recurrent laryngeal nerve (RLN) injury leading to vocal cord palsy is a dreaded complication of thyroid surgery occurring in 1–5% of cases. The main approaches to RLN preservation are: RLN visualization without nerve monitoring (No-NM), intermittent intraoperative nerve monitoring (I-IONM) and continuous intraoperative nerve monitoring (C-IONM). At present it is unclear which of these strategies should be the preferred method of RLN preservation. We performed a systematic review and network meta-analysis (NMA) to define the optimal method of RLN preservation during thyroid surgery using only the highest levels of available evidence.

Methods: A systematic review and NMA was performed according to PRISMA guidelines. NMA was conducted using R packages netmeta. Only prospective studies were included.

Results: Eighteen studies met inclusion and exclusion criteria from 973 studies identified, including 22,080 patients and 40,642 nerves at risk (NAR). The mean age was 53.1 years (range, 28.2–57.5 years) and 15,708 (71.9%) of patients underwent total thyroidectomy. Combining direct and indirect evidence identified C-IONM as having the lowest incidence of RLN injury following thyroid surgery [C-IONM vs. I-IONM, odds ratio (OR) 0.24, 95% CrI: 0.02–1.60; C-IONM vs. No-NM, OR 0.19, 95% CrI: 0.02–1.30]. I-IONM demonstrated a protective effect over No-NM (OR 0.79, 95% CrI: 0.63–1.02).

Conclusions: This NMA incorporating only the highest level of available evidence demonstrated a benefit of IONM (both C-IONM and I-IONM) in reducing the incidence of RLN injury following thyroid surgery. C-IONM use may confer additional benefit over I-IONM but further suitably powered randomized trials are necessary to support this finding.

Keywords: Lobectomy; nerve monitoring; recurrent laryngeal nerve (RLN); thyroidectomy; vocal cord palsy