Optimal total mesorectal excision for rectal cancer is by dissection in front of Denonvilliers' fascia (Br J Surg 2004; 91: 121–123)

Sir We have a couple of queries regarding the design and conclusions of this study. First, the authors’ power calculation determined that 38 patients were needed in each group for statistical analysis. Their actual conclusions, however, are based upon fewer than half of that number in each group. In addition, 54 patients in total were randomized, but results were reported on only 34. Underpowering, coupled with inadequate follow-up, reduces the statistical accuracy of the paper. Furthermore, only one patient underwent emergency surgery in the mesh arm of the study; patients with intra-abdominal contamination are most at risk of developing complications following mesh insertion. G. S. Ranger, B. Lovett, D. S. J. Collier, A.K. Salih Department of Surgery, Basildon Hospital, Essex SS16 5NL, UK DOI: 10.1002/bjs.4768 Optimal total mesorectal excision for rectal cancer is by dissection in front of Denonvilliers’ fascia (Br J Surg 2004; 91: 121–123)