The diagnosis of a lymph node-negative colorectal carcinoma should imply a good prognosis; however, the outcomes for TNM stage II patients remain variable. Few studies have examined the relationship of the number of lymph nodes examined to the prognosis of this stage. The aim of this study was to determine whether the number of lymph nodes examined has an effect on prognosis of a relatively large sample of patients undergoing curative surgery for stage II colorectal cancer at a single institution. Data on patients who underwent surgery for colorectal cancer between January 1980 and April 2000 were prospectively collected in a database. Patients with TNM stage II or stage III tumours who were treated with curative intent were removed. Patients over 80 years of age were excluded from the survival analysis. Survival comparisons were made using Kaplan-Meier curves and the log-rank test. Multivariate analysis was performed using a Cox regression model. A total of 625 cases of TNM stage II cases and, for comparison purposes, 415 stage III cases, were analysed. Lymph node retrieval in stage II cases was affected by the patient's age (P=0.04) and gender (P=0.02), tumour grade (P<0.0001), tumour site (P<0.0001), and necessity to carry out extended resection (P<0.0001). In stage III cases, lymph node retrieval was affected by patient age (P<0.0001), tumour grade (P=0.02), and tumour site (P=0.002). Decreased lymph node detection was associated with increasing hazard ratios among the 480 TNM stage II patients under 80 years of age, but not among the 345 patients with TNM stage III tumours. Five year survival rate for patients with stage III tumours with only 1-3 positive lymph nodes (52.6%) was similar to that of patients with stage II tumour who had nine or fewer lymph nodes examined (51.3%). These results demonstrate that the prognosis of TNM stage II colorectal cancer is dependent on the number of lymph nodes examined. Patients with few nodes examined have a poorer prognosis. It is possible that a smaller number of lymph nodes examined reflects a diminished immune response. It can be presumed that those patients with stage II tumour with only a few nodes examined should be offered postoperative chemotherapy on a routine basis.