A Population-based Study on Outcome in Relation to the Type of Resection in Low Rectal Cancer

PURPOSE: The aim of this population-based study was to assess treatment and outcome in patients resected for low rectal cancer, focusing on differences relating to the type of resection. METHODS: All patients in Stockholm with a rectal cancer within 6 cm of the anal verge, diagnosed from January 1995 to December 2003, and treated with abdominoperineal resection, anterior resection, or Hartmann procedure were included (n = 613). Clinical data, histopathology, and outcome were analyzed in relation to the type of surgery performed. RESULTS: Bowel perforations and incomplete tumor clearance were more common in the abdominoperineal group (12%) than in the anterior (4%) and Hartmann procedure groups (9%; P < .03). Incomplete tumor clearance was also more common in the abdominoperineal group (18%) than in the anterior (5%) or Hartmann groups (14%; P < .01). In multivariate analysis, local pelvic control was significantly associated with neoadjuvant radiotherapy and complete tumor resection, and survival was significantly associated with neoadjuvant radiotherapy, lower tumor stage, female gender, younger age, and complete tumor resection. Although local control and survival were poorer after abdominoperineal than after anterior resection, the type of operation was not an independent prognostic factor. CONCLUSION: Inadvertent bowel perforations and tumor-involved margins are more common after abdominoperineal than after anterior resection in low rectal cancer. Although previous studies have shown that abdominoperineal resection is also associated with poorer oncological outcomes, the present study could not confirm that the type of procedure per se is a significant determinant of prognosis.