Characteristics and impact of postoperative surgical site infection on increased antibiotic duration among patients with laryngocarcinoma: a retrospective cohort study.

Abstract

Whether increased antibiotic duration is necessary for surgical site infection (SSI) in patients after neck surgery is unclear. We investigated the characteristics of SSI, and the impact of SSI on increased antibiotic duration among patients with laryngocarcinoma (LC). A retrospective cohort study including consecutive LC patients ≥18 years, undergoing surgery without remote metastasis was conducted from October 2015 to February 2022 in the Department of Otolaryngology-Head and Neck Surgery, Chengdu Second People's Hospital. SSI was defined according to current guidelines. Patients were stratified into 3 groups including no-infection, lower respiratory tract infection (LRTI) and SSI. Patient characteristics was recorded. Patients were followed up until discharge. A multiple linear regression model including SSI and other factors including age, sex, comorbidity and antibiotic treatments was performed to explore the impact of SSI on increased antibiotic duration among LC patients with postoperative infection. A total of 88 patients were included, with 26 (29.5%) in no-infection group, 38 (43.2%) in LRTI group, and 24 (27.3%) in SSI group. Laryngocutaneous fistula occurred in 8 (33.3%) patients with SSI. Thirteen (34.2%) patients in LRTI group and 9 (37.5%) patients in SSI group experienced postoperative infection within 2 days after surgery, and antibiotic susceptibility tests were performed in 18 (47.4%) and 12 (50.0%) patients in LRTI and SSI group, respectively (P>0.05 for both). Levofloxacin and cefoperazone-sulbactam were the most commonly used antibiotics for postoperative infection in both LRTI and SSI groups (P>0.05 for both), irrespective of antibiotic susceptibility tests or not. The postoperative antibiotic duration in SSI group was significantly longer than that in LRTI group (13.62±4.28 days in SSI vs. 11.22±3.64 days in LRTI, P=0.021). A multiple linear regression analysis including SSI, age, sex, diabetes, antibiotic susceptibility test and hypoalbuminemia showed that, SSI was independently associated with increased antibiotic duration with LRTI as the reference among LC patients diagnosed [regression coefficient β=3.02, 95% confidence interval (CI): 1.01-5.03, P=0.004], whereas antibiotic susceptibility test was not (P=0.467). SSI may be independently associated with increased postoperative antibiotic duration in patients with LC with or without antibiotic susceptibility test.