Survival after resection of colorectal cancer based on anatomical segment of involvement.

ABSTRACT

PURPOSE: To determine survival differences for patients undergoing colonic or rectal resection for cancer on the basis of the specific anatomical location of primary tumor. METHODS: A total of 143,747 patients undergoing segmental colectomy, hemicolectomy, anterior resection, or abdominoperineal resection (APER) for adenocarcinoma from 1995 to 2009 were identified from 13 Surveillance, Epidemiology, and End Results regions. The primary end point was overall survival determined by adjusted hazard ratios (HRs); the secondary end point was lymph node yield. RESULTS: Total lymph node yield significantly decreased from proximal to distal resected segment in stage 0-II cancer, but not in stage III cancer. Lymph node ratio increased from cecum to hepatic flexure and then decreased distally (p < 0.001). Adjusted HRs revealed that survival after right colonic resection for ascending hepatic flexure and transverse colon cancer was not significantly different from cecal cancer. Survival after left colonic resection for descending colon cancer was not different from splenic flexure cancer, but sigmoid colectomy carried improved survival (HR 0.95, p = 0.027). APER carried worse survival compared to anterior resection (HR 1.28, p < 0.001) or right colonic resection for cecal cancer (HR 1.61, p < 0.001). CONCLUSIONS: Survival after resection from colorectal cancer depends on specific anatomical segment and not just the division between colon and rectum, or left and right colon. This may be related to inherent differences in the anatomical characteristics of the particular colorectal segment, with varying lymph node yields contributing to understaging. This supports an individualized approach to colorectal cancer, with particular attention to surgical technique, leading to survival improvement.