Wound site infections increase costs, hospital stay, morbidity, and mortality. Techniques used for wounds management after laparotomy are primary, delayed primary, and vacuum-assisted closures. The objective of this study is to compare infection rates between those techniques in contaminated and dirty/infected wounds. Eighty-one laparotomized patients with Class III or IV surgical wounds were enrolled in a three-arm randomized prospective study. Patients were allocated to each group with the software Research Randomizer® (Urbaniak, G. C., & Plous, S., Version 4.0). Presence of infection was determined by a certified board physician according to Centers for Disease Control's Criteria for Defining a Surgical Site Infection. Twenty-seven patients received primary closure, 29 delayed primary closure, and 25 vacuum-assisted closure, with no exclusions for analysis. Surgical site infection was present in 10 (37%) patients treated with primary closure, 5 (17%) with primary delayed closure, and 0 (0%) patients receiving vacuum-assisted closure. Statistical significance was found between infection rates of the vacuum-assisted group and the other two groups. No significant difference was found between the primary and primary delayed closure groups. The infection rate in contaminated/dirty-infected laparotomy wounds decreases from 37 and 17 per cent with a primary and delayed primary closures, respectively, to 0 per cent with vacuum-assisted systems.