The purpose of this study was to determine the effect of a 3- versus 1-day antibiotic regimen on the rate of surgical site infection (SSI) in patients undergoing orthognathic surgery at a department of oral and maxillofacial surgery in Halifax, Nova Scotia, Canada. A prospective, randomized controlled trial was conducted. All patients received 1 day of intravenous antibiotics after surgery. Then, patients were randomly distributed into groups that received 2 days of additional antibiotics (group A) or placebo (group B). The primary outcome measured was the presence of SSI. The operating surgeon, concomitant extraction of teeth, surgical procedures performed, duration of intermaxillary fixation, and length of hospital stay were analyzed for an effect on SSI. Patients were followed for 1 year after surgery to identify SSIs that might have been diagnosed outside the hospital. The trial started with 288 patients, and 117 patients were lost to follow-up. Statistical analyses were ultimately performed on those 171 patients who were adherent to the study medication regimen. Group A (n = 86) and B (n = 85) SSI rates were 7.0 and 17.6% (number needed to treat = 10; P = .04), respectively. Mandibular bilateral sagittal split osteotomy (BSSO) was involved in 71% of SSIs. Intra- and postoperative surgical variables did not have a relevant effect on the SSI rate. Patients were followed for 1 year after surgery, and group A (n = 46) and group B (n = 44) had SSI rates of 4 and 25% (P < .05), respectively. Three days of postoperative cefazolin and cephalexin markedly decreases SSI rates compared with 1 day. However, the number needed to treat of 10 suggests that the benefits of the extended regimen might not outweigh the risks. The high prevalence of SSIs at the mandibular BSSO incisions might have been caused by contamination, with more saliva and reception of a lower blood supply, than maxillary Le Fort I incisions. Mandibular osteotomies could benefit from an extended antibiotic regimen to minimize SSIs and associated complications. Other surgical variables might not require special consideration for antibiotic therapy.