Introduction: The benefits of mechanical bowel preparation for elective colorectal surgery have been challenged by multi-center, multi-surgeon randomized trials. However, the procedure is an established practice. Heterogeneity in study design may have prevented surgeons from translating “scientific evidence” to their individual clinical practice. We designed a prospective single-blinded randomized trial to investigate whether preoperative mechanical bowel preparation influences the incidence of surgical site infection and anastomotic failure after elective colorectal surgery by a single surgeon. Patients and Methods: All patients scheduled to undergo an elective colon or proximal rectal resection with a primary anastomosis by the same surgeon from October 2001 were enrolled and randomized to receive either oral polyethylene glycol lavage solution (Group A) or no mechanical bowel preparation whatsoever (Group B). Dietary restrictions were limited to 12 hours prior to surgery. A standard intravenous antibiotic prophylaxis scheme was used. Exclusion criteria included immunosupression, preoperative chemoradiotherapy, diverting stoma and perforated and/or obstructing tumor. Patients were followed by an independent observer for wound infection, intrabdominal sepsis and anastomotic failure within 30 days after surgery. Results: Until July 2005, one hundred and ten patients had been enrolled. Two patients (2%) were preoperatively excluded because of active immunosupression. One hundred and eight patients were randomized but 11 of them (10%) were excluded from analysis (diverting stoma in 9 cases, contained perforation in 1 patient and unresectable tumor in 1 patient). Of the remaining 97 patients, 48 were assigned to Group A and 49 to Group B. The mean age was 66.5 ± 12 years in Group A and 67.9 ± 15 years in Group B (NS). There was no difference in sex distribution between groups. The most frequent indication for surgery was colorectal adenocarcinoma (71%). Overall, twelve patients (12.4%) developed postoperative wound infection, six in each group (Group A = 12.2% and Group B= 12.5%; NS). There were 3 cases of intrabdominal sepsis and all of them occurred in Group A (6.3%). The overall rate of anastomotic failure was 6.3% (n = 6), 4 patients in Group A (8.3%) vs. 2 patients in Group B (4.1%) (NS). The overall complication rate in Group A was 27.1% vs, 16.3% in Group B. The NNH was 9.3 Conclusion: Although underpowered, our intermediate analysis suggests that a surgeon may have worse outcomes in terms of surgical site infection and anastomotic failure rates if preoperative mechanical bowel preparation with poliethylenglycol is routinely used.