ObjectiveTo explore effect of lymph node ratio (LNR) on prognosis for patients with stage Ⅲ colorectal cancer. MethodsThe clinicopathologic data of 1 424 stage Ⅲ patients who underwent curative resection in this hospital from January 2009 to December 2013 were analyzed retrospectively. These patients were divided into low LNR group (LNR<optimal cut-off value) and high LNR group (LNR≥optimal cut-off value) according to the optimal cut-off value by receiver operating characteristic curve. The correlation between prognosis and clinicopathologic features was analyzed by using univariate analysis and Cox's proportional hazards regression model. The 5-year cumulative disease-free survival (DFS) and overall survival (OS) were calculated by using Kaplan-Meier method. ResultsThe area under the receiver operating characteristic curve was 0.639, the optimal cut-off value of LNR was 0.267. The 5-year cumulative DFS and OS of the low LNR group were significantly higher than those of the high LNR group (DFS:53.8% versus 34.4%, P<0.001; OS:60.5% versus 44.5%, P<0.001). The results of univariate analysis showed that the preoperative carcinoembryonic antigen (CEA) level, degree of differentiation, TNM stage, pT stage, pN stage, vascular invasion, tumor diameter, and LNR were associated with the DFS and OS (P<0.05), but the number of lymph nodes retrieved was only associated with the DFS (P<0.05). The results of Cox analysis showed that the preoperative CEA level, degree of differentiation, TNM stage, and LNR were the independent prognostic factors for the DFS and OS (P<0.05), but the number of lymph nodes retrieved and the vascular invasion were only the independent prognostic factors for the DFS (P<0.05). ConclusionLNR is a prognostic factor in stage Ⅲ colorectal cancer, and it might be as a supplement for TNM staging system.