Objective: We investigated the prognostic value of the lymph node ratio (LNR), determined different cutpoints in cohorts with examined lymph nodes (eLNs) ≤15 or eLNs >15, and compared hypothetical pNr with 7th edition UICC pN stage. Methods: A total of 1,772 gastric cancer patients undergoing D2 resection were enrolled. All patients were divided into two sub-cohorts according to eLNs ≤15 or eLNs >15. The optimal LNR cut-point was calculated using decision tree method. Homogeneity, discriminatory ability, and monotonicity of gradients of the pNr and UICC pNstages were compared using linear trend χ2, likelihood ratio χ2 statistics, and Akaike information criterion (AIC) calculations. Results: The optimal cut-points were 0, 20, 35, 65 in eLNs >15 cohort and 0, 35, 70 in eLNs ≤15 cohort. pNr stage was an independent factor for gastric cancer prognosis. The pNr stage had higher linear trend and likelihood ratio χ2 scores and lower AIC values compared with those for UICC pN stage. Conclusion: The LNR cutpoint should be different according to the numbers of eLNs. pNr can predict survival more accurately than UICC pN stage in both the ≤15 and the >15 eLNs cohorts.