Background: The current American Joint Committee on Cancer (AJCC) staging system for esophageal cancer does not define the minimum number of lymph nodes (LNs) necessary for accurate nodal staging. This study aimed to seek the minimum number of LNs examined for adequate nodal staging of patients with node-negative esophageal cancer.
Methods: We conducted a retrospective review of 592 patients undergoing R0 resection with node-negative esophageal carcinoma between December 1996 and December 2004. The relationship between the total number of examined LNs and death from esophageal cancer was investigated by means of a scatterplot of this variable versus Martingale residuals from a Cox proportional hazard regression model without the variable of interest. A smoothed line fit of the scatterplot was applied to detect the reasonable cutoff point.
Results: The patients were classified into four categories according to the number of examined LNs: < or =5, 6 to 9, 10 to 17, and > or =18. A reduced hazard ratio of death was observed with an increasing number of LNs examined. The 5-year cancer-specific survival rate was 42.8% among patients with < or =5 LNs examined, compared with 52.6, 56.8, and 75% for those with 6-9 LNs, 10-17 LNs, and > or =18 LNs, respectively. Multivariate Cox regression analysis suggested that female sex, lower grade of cell differentiation, lower T category and increasing number of examined LNs were independent factors favoring cancer-specific survival.
Conclusions: At least 18 LNs should be resected for accurate staging of operable esophageal carcinoma. However, a validation from other institute is warranted.