Lymph node sampling in resectable hepatocellular carcinoma: national practice patterns and predictors of positive lymph nodes

Background and objectives: Routine lymphadenectomy (LND) for resectable hepatocellular carcinoma (HCC) remains controversial. We evaluated national LND trends to identify pre-operative factors associated with node-positive disease to determine which patients might benefit from LND. Methods: We identified HCC patients in the National Cancer Database (NCDB) treated with surgical resection between 2004 and 2015. Demographic, operative, pathologic, and survival data were compared. Multivariable regression was performed to determine preoperative predictors of pathologic nodal disease. Results: Of 8095 total resected patients, 1442 (17.8%) underwent hepatectomy with LND. Patients who received LND had higher preoperative clinical T (T3-T4: 20.0% vs 12.1%, p < 0.001) and N (N1: 3.3% vs 0.6%, p < 0.001) stages. The strongest independent predictor of pathologic nodal disease was clinical N stage (OR 106.54, CI 44.10–257.42). Survival was highest in patients whose surgeons omitted LND or were found with LND to be node-negative on final pathology (p < 0.001). Clinical node positivity had high negative predictive value (97.9%) but moderate positive predictive value (56.3%) in estimating pathologic nodal status. Conclusions: Defining preoperative clinical nodal status is imperative in HCC patients. Clinical node positivity was the strongest predictor of pathologic nodal disease and its associated worse prognosis. LND can be considered selectively in clinically node-positive patients.