PURPOSE:The aim of this article was to examine local recurrence after curative resection for carcinoma of the rectum in which the surgical technique of total mesorectal excision was not performed.METHODS:A single surgeon managed the patients and the data collected prospectively. Total excision of the distal mesorectum was not performed in the upper third or mid rectum.RESULTS:From 1969 to 1993 curative resections were performed in 549 patients, of which 17 died postoperatively, leaving 532 for analysis. Sphincter‐saving resection was performed in 468 patients (88 percent) and abdominoperineal excision in 58 (10.9 percent). The pathology stages (Dukes) were A, 158 (29.7 percent); B, 184 (34.7 percent); and C, 190 (35.7 percent). Five hundred seventeen patients (97.2 percent) were followed up for a minimum of five years. The median period of follow‐up was 82 months. Local recurrence confined to the pelvis occurred in 17 patients, and local recurrence associated with distant metastases occurred in 24 patients. The total five‐year local recurrence rate was 7.6 percent. Local recurrence was increased in Stage C tumors (P=<0.0001). Diathermy dissection in the pelvis was associated with a decreased local recurrence rate (P=0.023). The five‐year survival rate in curative resections was 72.5 percent.CONCLUSIONS:It is essential that articles presenting local recurrence rates should include both local recurrence in isolation and that which occurs with distant metastases. Although total mesorectal excision for rectal cancer was not performed in this study, the local recurrence rate is not materially different from that in several articles where total mesorectal excision has been used. Whether the distal mesorectum needs to be pursued in mid‐rectal cancer is not yet proven.