Regional lymph node sampling in lung carcinoma: a single institutional and national database comparison

Assessing regional lymph node metastasis is a key component of lung carcinoma staging and prognostication. Recent guidelines have suggested a quality metric of 10 total regional lymph nodes sampled with each stage I-II primary lung carcinoma resection. However, the extent of mediastinal lymph node sampling remains controversial. We assessed factors contributing to regional lymph node counts and effect on overall patient survival in an institutional cohort of 888 cases and the Surveillance, Epidemiology, and End Results national cancer registry (10 856 cases). The distribution of total lymph node counts in lobectomy and pneumonectomy cases was variable with a median of 10 and an interquartile range of 7 to 14. Multiple clinical and pathologic factors correlated with total regional node counts. Total lymph node counts of at least 10 in the institutional cohort did not correlate with significant differences in overall survival as compared with node counts of less than 10 (P = .38). In the Surveillance, Epidemiology, and End Results database, although 0 regional lymph nodes were correlated with reduced overall survival (hazard ratio, 1.47; P < .01), no significant difference was detected for 1 to 9 versus at least 10 nodes (P = .8). In conclusion, lymph node counts for primary lung carcinoma are driven by surgical, pathologic, and biologic variability. We find no evidence for a meaningful quality metric of 10 total regional lymph nodes at the institutional and national registry levels.