PURPOSE: To investigate the influence of the introduction of total mesorectal excision (TME) on local recurrence rate and survival in patients with rectal cancer. METHODS: A total of 171 consecutive patients underwent anterior or abdominoperineal resection for primary rectal cancer. When the TME technique was introduced, the clinical setting, including the surgeons, remained the same. Group 1 (1993-95, n =53) underwent conventional surgery and group 2 (1995-2001, n = 118) underwent TME. All patients were followed for 7 years or until death. RESULTS: Between the two groups, no statistically significant differences were present with regards to patient-, treatment-, or tumor-related characteristics apart from the time point of radiotherapy. The total local recurrence rates were 11 of 53 (20.8%) in group 1 and 7 of 118 (5.9%) in group 2, and the rates of isolated local recurrences were 6 of 53 (11.3%) in group 1 and 2 of 118 (1.7%) in group 2. Both differences were highly statistically significant. The disease-free survival in groups 1 and 2 was 60.4 and 65.3% at 5 years, and 58.5 and 65.3% at 7 years, respectively. Excluding patients with synchronous or metachronous distant metastasis from the analysis, both the disease-free survival and the cancer-specific survival were statistically significantly better in group 2 than in group 1. No statistically significant difference between the two groups was detected regarding the overall survival. CONCLUSIONS: The introduction of TME led to an impressive reduction of the local recurrence rate. Survival is mainly determined by the occurrence of distant metastasis, but TME seems to improve survival in patients without systemic disease.