Abstract

Background High-quality rectal cancer surgery is known to improve patient outcome. We aimed to assess the quality of colon cancer surgery by studying the extent of variation in the plane of surgical resection, the amount of tissue removed, and its association with survival. Methods All resections for primary colon adenocarcinoma done at Leeds General Infirmary (Leeds, UK) between Jan 1, 1997, and June 30, 2002, were identified. The specimens were photographed and graded according to the plane of mesocolic dissection. Tissue morphometry was done on 253 tumours. Univariate and multivariate models were used to ascertain whether there was an association with 5-year survival. The primary outcome measure was overall survival defined as death from any cause. Findings 521 cancers were identified, 122 were excluded because of either no photographic images or insufficient images to allow retrospective grading, leaving 399 specimens for analysis. There was marked variation in the proportion of each plane of surgery: muscularis propria in 95 of 399 (24%) specimens, intramesocolic in 177 of 399 (44%) specimens, and mesocolic in 127 of 399 (32%) specimens. Mean cross-sectional tissue area outside the muscularis propria was significantly higher with mesocolic plane surgery (mean 2181 [SD 895] mm2) compared with intramesocolic (mean 2109 [1273] mm2) and muscularis propria plane (mean 1447 [913] mm2) surgery (p=0·0003). There was also a significant increase in the distance from the muscularis propria to the mesocolic resection margin with mesocolic plane surgery (mean 44 [21] mm) compared with intramesocolic (mean 30 [16] mm) and muscularis propria plane (mean 21 [12] mm) surgery, which was independent of tumour site (all excisions p<0·0001). We noted a 15% (95% CI) overall survival advantage at 5 years with mesocolic plane surgery compared with surgery in the muscularis propria plane (HR 0·57 [0·38–0·85], p=0·006) in univariate analysis. However, this association was no longer significant in the multivariate model (HR 0·86 [95% CI 0·56–1·31], p=0·472), but was especially noted in patients with stage III cancers (HR 0·45 [95% CI 0·24–0·85], p=0·014; multivariate analysis). The plane of surgery and amount of mesocolon removed varied between the different sites with better planes in left-sided resections than right-sided ones, which were better than transverse resection (p<0·0001). Interpretation As previously shown in the rectum, we have now shown there is marked variability in the plane of surgery achieved in colon cancer. Improving the plane of dissection might improve survival, especially in patients with stage III disease. If confirmed by clinical trial data, such as from the ongoing National Cancer Research Institute Fluoropyrimidine, Oxaliplatin and Targeted Receptor pre-Operative Therapy for colon cancer (FOxTROT) trial of neoadjuvant chemotherapy in advanced resectable colon cancer, improvement of the plane of dissection might be a new cost-effective method of decreasing morbidity and mortality in patients with colon cancer. Funding National Institute for Health Research Academic Clinical Fellowship Programme, Experimental Cancer Medicine Centre programme (both Department of Health, London, UK), Yorkshire Cancer Research (Harrogate, UK), and the Pelican Trust (Basingstoke, UK).