Metronidazole and cefazolin vs cefazolin alone for surgical site infection prophylaxis in gynecologic surgery at a comprehensive cancer center

• Published in: American journal of obstetrics and gynecology 2024-09, Vol.231 (3), p.326.e1-326.e13
• Main Authors: Knisely, Anne, Iniesta, Maria D., Marten, Claire A., Chisholm, Gary, Schmeler, Kathleen M., Taylor, Jolyn S.
• Format: Article
• Language: English
• Subjects: cefazolin; gynecologic surgery; hysterectomy; metronidazole; quality improvement; surgical prophylaxis; surgical site infection
• ISSN: 0002-9378; 1097-6868; 1097-6868
• DOI: 10.1016/j.ajog.2024.03.043

Surgical site infection is one of the most common complications of gynecologic cancer surgery. Current guidelines recommend the administration of cefazolin preoperatively to reduce surgical site infection rates for patients undergoing clean-contaminated surgeries such as hysterectomy. To evaluate the impact of a quality improvement project adding metronidazole to cefazolin for antibiotic prophylaxis on surgical site infection rate for women undergoing gynecologic surgery at a comprehensive cancer center. This retrospective, single-center cohort study included patients who underwent surgery in the gynecologic oncology department from May 2017 to June 2023. Patients with penicillin allergies and those undergoing concomitant bowel resections and/or joint cases were excluded. The preintervention group patients had surgery from May 2017 to April 2022, and the postintervention group patients had surgery from April 2022 to June 2023. The primary outcome was a 30-day surgical site infection rate. Sensitivity analyses were performed to compare surgical site infection rates on the basis of actual antibiotics received and for those who had a hysterectomy. Factors independently associated with surgical site infection were identified using a multivariable logistic regression model adjusting for confounding variables. Of 3343 patients, 2572 (76.9%) and 771 (23.1%) were in the pre-post intervention groups, respectively. Most patients (74.7%) had a hysterectomy performed. Thirty-four percent of cases were for nononcologic (benign) indications. Preintervention patients were more likely to receive appropriate preoperative antibiotics (95.6% vs 90.7%; P