Use of Completion Lymph Node Dissection for Sentinel Lymph Node-Positive Melanoma

BACKGROUND: For patients with sentinel node-positive melanoma (SNPM), randomized trials, first reported in 2015, found no benefit for routine completion lymph node dissection (CLND) in selected patients. This study examines time trends in CLND and explores institutional and clinical factors associated with CLND. STUDY DESIGN: The National Cancer Database was queried for patients older than 18 years from 2012 to 2016 with SNPM. A high-volume center was defined as >80th percentile for number of sentinel node procedures. Poisson regression assessed temporal trends and identified patient, pathologic, and institutional characteristics associated with CLND. RESULTS: From 2012 to 2016, we identified 7,146 patients with SNPM. The proportion of patients undergoing CLND was steady in 2012 to 2014 (61% to 63%), but decreased to 57% in 2015 and 50% in 2016 (p < 0.0001). The proportion of patients with SNPM who underwent CLND decreased over time for both high- (66% to 52%; p < 0.0001) and lower-volume centers (55% to 45%; p = 0.06). Female sex (relative risk [RR] 0.97; p < 0.001) and increasing age (RR 0.98; p < 0.0001) were associated with lower likelihood of CLND. Increased Breslow depth (RR 1.015; p = 0.006), ulceration (RR 1.067; p = 0.02), and high-volume centers (RR 1.180; p < 0.0001) were associated with higher likelihood of CLND. Regional differences in likelihood of CLND were also present (p < 0.0001). CONCLUSIONS: Completion lymph node dissection in SNPM decreased over time, with the greatest change in 2016. Several patient, pathologic, and institutional characteristics were associated with likelihood of CLND. As evidence supports close observation for selected patients, efforts should be undertaken to improve and standardize patient selection for CLND across institutions caring for patients with melanoma. ((C) 2020 by the American College of Surgeons. Published by Elsevier Inc. All rights reserved.)