An anastomotic leak is an unpredictable postoperative complication during recovery from colorectal surgery that may require a re-operation. Potentially pathogenic bacteria like Pseudomonas (and Enterococcus) contribute to the pathogenesis of an anastomotic leak through their capacity to degrade collagen and to activate tissue matrix metalloprotease-9 in host intestinal tissues. The microbiome, therefore, is the key to preventing an anastomotic leak after colorectal surgery. The aim of this trial was to investigate whether perioperative selective decontamination with a new mixture of locally acting antibiotics specially designed against Pseudomonas aeruginosa and Enterococcus faecalis can reduce or even stop early symptomatic leakage. All hospitalized patients in our University Clinic undergoing colorectal surgery with a left-sided anastomosis were included as two groups; patients in the intervention group received polymyxin B, gentamicin and vancomycin every six hours for five postoperative days and those in the control group did not receive such an intervention. An anastomotic leak was defined as a clinically obvious defect of the intestinal wall integrity at the colorectal anastomosis site (including suture) that leads to a communication between the intra- and extraluminal compartments, requiring a re-do surgery within seven postoperative days. Between February 2017 and May 2023, a total of 301 patients (median age of 63 years) were analyzed. An anastomotic leak was observed in 11 patients in the control group (n = 152), but in no patients in the intervention group (n = 149); this difference was highly significant. The antibiotic mixture (with polymyxin B, gentamicin and vancomycin) used for local decontamination in our study stopped the occurrence of anastomotic leaks completely. According to the definition of anastomotic leak, no further surgery was required after local perioperative decontamination.