OBJECTIVE: We analyzed 3 previously identified cut-off values of lymph node ratios (0.17, 0.41, and 0.69) in a large population-based collective of patients with colorectal cancer for their prognostic value. SUMMARY BACKGROUND DATA: The lymph node ratio (LNR) (relation of tumor-infiltrated to total examined lymph nodes) has a high prognostic impact, but the relevant cut-off values are not determined. METHODS: Patients (N = 27,803) with a primary colorectal cancer diagnosed and operated in the Munich region between 1991 and 2006 were registered in the Munich Cancer Registry. Lymph node numbers and survival data were available for 17,309 patients with a mean follow-up of 5.9 years. RESULTS: The mean number (+/-SD) of resected lymph nodes was 16.8 +/- 8.4. Twelve or more lymph nodes were resected in 76.8%. Estimated 5-year overall survival decreased significantly with increasing LNR: LNR = 0 in 71.4%, LNR 0.01 to 0.17 in 52.4%, LNR 0.18 to 0.41 in 33.3%, LNR 0.42 to 0.69 in 19.8%, and LNR > or = 0.70 in 8.3% (P < 0.001). Multivariable survival analyses identified separately both LNR and pN- category, as well as number of resected lymph nodes, patient's age, tumor location, pT-category, pM-status, R-status, tumor grade, and year of operation as independent prognostic factors. CONCLUSION:: The 3 cut-off values of LNRs had strong independent prognostic value in a population-based collective of patients with colorectal cancer. The LNR should be routinely reported and included in the American Joint Committee on Cancer staging system. Nevertheless, the benefit of lymphadenectomy on survival is still unclear.