Lymph node ratio is a significant predictor of disease-specific mortality in patients undergoing esophagectomy for cancer

Abstract

The seventh edition of the American Joint Committee on Cancer esophageal cancer staging system classifies nodal status by the number of malignant nodes (LNMs) found. This may be confounded by variations in lymphadenectomy and specimen review. The ratio of lymph nodes containing metastases to the total nodes excised (LNR) has been suggested as an alternative. We seek to validate the use of LNR for staging and determine the effect of the total lymph node yield (LNY) on its accuracy. A review of our prospective esophageal database identified 94 patients who underwent esophagectomy for cancer at out institution from 1992 until 2010. Univariate and multivariate analyses were performed. The mean age of our patients was 59.4 years. Transthoracic esophagectomy was performed in all but three instances. The majority of tumors were adenocarcinoma, 76 per cent. Overall survival at 2 and 5 years was 52 and 29 per cent, respectively. LNY correlated with LNM (r = 0.302, P = 0.001) but not LNR (r = 0.012, P = 0.912). Using Kaplan-Meier analysis, LNR had no effect on disease-specific (DS) survival (P = 0.803). However, a Cox proportional hazards regression model showed LNR to be a significant predictor of DS mortality (hazard ratio, 9.47; P = 0.049). The lack of correlation between LNR and LNY suggests that LNR may be a more robust staging method when LNY is low. Furthermore, LNR was found to be a significant predictor of DS mortality when controlling for other factors influencing survival. However, neither a staging system based on LNR nor its efficacy compared with the current system could be determined from these data.