Supplemental Perioperative Oxygen and the Risk of Surgical Wound Infection: A Randomized Controlled Trial

Abstract

ABSTRACT Three hundred patients were enrolled in a double-blind, randomized, controlled study designed to investigate the ability of supplemental perioperative oxygen to reduce wound infection. Participants were age 18 to 80 years, had no coexisting serious medical conditions, and were scheduled to undergo elective colorectal surgery at one of 14 participating hospitals in Spain. No patients undergoing minor or laparoscopic surgery were included. Anesthesia and antibiotic prophylaxis were standardized for the study. Patients were randomized by computer-generated codes to receive an oxygen/air mixture of 30% or 80% fraction of inspired oxygen (F102) intraoperatively and postoperatively for 6 hours. After 6 hours, oxygen was given only in amounts needed to maintain 92% saturation. Postoperative care was determined by the attending surgeon who was unaware of the patient's oxygen group. Wounds were inspected daily and surgical site infections (SSI) were diagnosed according to the definitions of the Centers for Disease Control and Prevention. Infections occurring during the first 14 days were considered for analysis. Nine patients did not meet inclusion criteria. Of the remaining 291 patients, 143 received 30% and 148 received 80% oxygen. The 2 groups were similar in clinical characteristics, including preoperative laboratory studies and risk of infection score. Surgical characteristics, including length of operative procedure, blood loss, and transfusion rate, were also similar between the 2 groups. Fifty-seven patients (19.3%) developed a wound infection. The incidence of wound infection was 35 of 143 (24%) in the 30% F102 group and 22 of 148 (14.9%) in the 80% F102 patients (P = 0.04). The risk of SSI was 39% lower in the high oxygen group compared with those who received less oxygen (relative risk, 0.61; 95% confidence interval [CI], 0.38–0.98). Other measures of surgical outcome, including return of bowel function, ability to tolerate solid food, ambulation, suture removal, and duration of hospitalization, were not significantly different for the 2 treatment groups. When patients who developed wound infections were compared with those without infection, the group with SSI had a longer time to ambulation (mean, 4.9 vs 3.9 days; P = 0.008), a longer time to staple removal (11.6 vs 10.1 days; P = 0.007), and were in the hospital longer (15.1 vs 10.7 days; P = 0.001). Two patients, both in the 30% oxygen group, died of sepsis during the study period. After multivariate analysis of possible confounding variables, the relative risk of wound infection in patients who received 80% oxygen was 0.46 (95% CI, 0.22–0.95; P = 0.04) compared with those who received 30% oxygen.