PURPOSE: Adequate lymph node analysis is critical for appropriate staging in colorectal cancer. The aim of this study is to determine whether 12 or more nodes recovered in stage III rectal cancer results in improved oncologic outcomes. METHODS: Two hundred and forty-eight patients with stage III rectal cancer from 1995 through 2004 were reviewed. They were categorized into 2 subgroups by the number of nodes retrieved (<12 and > or =12), and oncologic outcomes in terms of 5-year overall and disease-free survival were analyzed for all patients, patients with American Joint Committee on Cancer (AJCC) N1 disease (N=145), and those with AJCC N2 (N=103). RESULTS: Five-year overall and disease-free survival was 79.0% and 58.4%, respectively. There was no significant difference in clincopathologic features between <12 retrieval group and > or =12 group. Although there was significant difference in overall survival and disease-free survival between the number (<12 and > or =12) of lymph nodes removed in N2 disease (P=0.043; P=0.022) in univariate analysis, the total number of lymph nodes retrieved was not a prognostic factor affecting survival in multivariate analysis. The N2 stage and lateral margin involvement were prognostic factors affecting survival in multivariate analysis. CONCLUSION: This study showed that the total number of lymph nodes analyzed for stage III rectal cancer is not a prognostic factor on overall or disease-free survival in multivariate analysis.