Thyroid surgery with intraoperative neuromonitoring of the recurrent laryngeal nerve: a prospective multicentre study in Germany with 7617 patients

Abstract

Abstract Background Injury to the recurrent laryngeal nerve (RLN) is a frequently noted and serious complication in thyroid surgery. Several studies have established that identification of the RLN may reduce the incidence of RLN palsy. To date, no prospective study has evaluated whether the RLN palsy rate can be further reduced by intraoperative neuromonitoring of the RLN. Methods Between 1 January and 31 December 1998, surgery for benign and malignant goitre was performed on 7607 patients in 45 hospitals. Data were collected prospectively by questionnaire. RLN function was evaluated before and after operation in all patients. RLN palsy was defined as permanent when there was no evidence of recovery within 6 months after surgery. The RLN palsy rate was adjusted to nerves at risk in the three groups ‘no identification of the RLN’, ‘intraoperative identification of the RLN without neuromonitoring’ and ‘identification of the RLN with neuromonitoring’. Statistical analysis was by χ2 test. Results Mean patient age was 51·8 years; 72·7 per cent were women. Some 7256 patients were treated for benign goitre (multinodular goitre 74·0 per cent, uninodular goitre 16·6 per cent, recurrent goitre 6·1 per cent, Graves' disease 3·3 per cent, Hashimoto and De Quervain thyroiditis less than 0·1 per cent) and 351 patients for thyroid carcinoma, with a negative selection of recurrent goitre and thyroid carcinoma for the two groups with intraoperative RLN identification (P < 0·0001). With respect to the extent of resection, in cases of lobectomy the rate of permanent RLN palsy was 1·5 per cent with intraoperative neuromonitoring which was significantly lower than rate with intraoperative RLN identification without neuromonitoring (2·6 per cent) or no RLN identification (6·3 per cent) (P < 0·0001). Detailed analysis of the complete spectrum of extent of resection and indications for surgery showed a positive trend for intraoperative neuromonitoring in comparison to the group with RLN identification without the use of neuromonitoring, but this trend was not statistically significant because of the small number of permanent RLN palsies. Conclusion Intraoperative neuromonitoring of the RLN has a beneficial effect for high-risk patients with recurrent goitre and thyroid carcinoma in whom the RLN is at greatest risk in cases of lobectomy or total thyroidectomy.