Earlier meta-analyses of small randomized trials suggested that mechanical bowel preparation (MBP) should be omitted before colorectal surgery because it does not affect complication rates 0 mortality and may be even harmful; however, more recent large randomized trials suggested an increased occurrence of pelvic abscesses in the absence of MBP. Therefore, an updated large meta-analysis was conducted to re-evaluate the role of MBP in colorectal surgery. Furthermore, the influence of different kind of MBP regimes on infectious outcomes was examined. The meta-analysis was conducted according to the QUOROM statement; the inclusion criteria were randomized clinical trials comparing MBP with no MBP before colorectal surgery. The primary outcome was anastomotic leakage; secondary outcomes were other septic complications. Fourteen trials were included with a total number of 4859 patients: 2452 in the MBP group and 2407 in the no MBP group. We found no statistical difference between the groups for anastomotic leakage [OR = 1.12 (0.82-1.53), P = 0.46], pelvic or abdominal abscess (P = 0.75), and wound sepsis (P = 0.11). When all surgical site infections were considered, the meta-analysis favored no MBP [OR = 1.40 (1.05-1.87), P = 0.02].Sensitivity analyses showed similar results for all subgroups but when poor or small trials were excluded, there was a slightly higher risk of deep abdominal abscesses with no MBP, however, the number needed to harm was as high as 333 patients, suggesting this difference to be not clinically relevant. The use of different MBP regimes did not influence primary and secondary outcomes. The main limitation concerned rectal surgery for which the limited data preclude any interpretation. Although it did not confirm the harmful effect of mechanical bowel preparation (suggested by previous meta-analyses), this meta-analysis including almost 5000 patients, demonstrates with a high level of evidence that any kind of mechanical bowel preparation should be omitted before colonic surgery.