Prophylactic level II neck dissection guided by frozen section for clinically node-negative papillary thyroid carcinoma: Is it useful?

Background: Prophylactic lateral neck dissection (PLND) is generally not performed for papillary thyroid carcinoma (PTC). When performed, occult metastases are found in up to 50 % of patients, although the incidence of occult level II nodes seems low. Our aim was to evaluate frozen section analysis-oriented elective level II PLND in patients with clinically node-negative (cN0) PTC. Methods: This retrospective study included patients with cN0 PTC treated with total thyroidectomy and prophylactic bilateral central and lateral neck dissection of ipsilateral levels III and IV. Frozen section analysis of PLND III and IV was performed. If positive, the PLND was extended to level II. We measured the accuracy of frozen section analysis, the incidence of occult level II metastasis, and oncologic outcomes. Results: A total of 295 patients were included. For frozen section analysis, the sensitivity was 71.0 %, specificity 99.6 %, positive predictive value 97.8 %, negative predictive value 92.4 %, overall accuracy 93.2 %. Definitive analysis found lateral node metastases in 63 of the 295 (21 %) patients. Extension to level II was performed in 27 of 46 cases (59 %). Level II contained metastatic nodes in 12 of 27 (44 %) patients. There was no difference in total doses of ¹³¹I administered to patients with or without level II disease. Even when extension of PLND to level II was not performed, no cases of recurrent or persistent disease in level II occurred. Conclusions: Frozen section analysis was highly accurate. The rate of occult metastases in level II was low. Detection of additional metastases in level II did not modify subsequent treatment or the rate of recurrence and is not useful for routine application.