What determines the outcome after total mesorectal excision for rectal carcinoma—15 years experience of a specialist surgical unit

Abstract

OBJECTIVES: Outcome after surgery for rectal cancer is often correlated with surgical technique, without detailed pathological examination of the specimen. Measurement of the number of lymph nodes by a fat clearance technique, mesorectal spread and involvement of the circumferential margins together with clinical findings at operation are correlated with outcome in a consecutive series of patients treated by a specialist surgeon. PATIENTS AND METHODS: The effect of prognostic factors on patients' survival was analysed using Cox's proportional hazards regression models. Survival curves and 5-year survival rates were calculated using Kaplan-Meier analysis. RESULTS: After curative procedures, 78.6% (95% confidence interval (CI) 72.8-84.3) of patients survived compared with 6.0% (95% CI 0.7-11.4) after palliative procedures; the hazard of dying was 1.5 times higher for patients with Dukes' B tumours and 6.4 times higher for those with Dukes' C compared with those with Dukes' A (P < 0.0001). Patients with four or more lymph nodes involved had a two-fold increase in the hazard of dying compared with those with three or less involved lymph nodes. CONCLUSION: The strongest predictor of 5-year survival for patients with rectal carcinoma was whether the operation was deemed curative or palliative. Neither the extent of mesorectal spread nor the involvement of the circumferential resection margin had a significant effect on survival, after surgery in a specialist unit, compared with published series after conventional surgery.