Methods: We performed a retrospective cohort study including all patients aged ≥18 years who underwent orthopedic surgery from January 2016 through December 2017 in a tertiary care hospital in Seoul, South Korea.
Results: During the study period, 7,369 episodes of orthopedic surgery were identified. The prevalence of C. difficile infection was 7.7 cases per 1,000 surgical procedures (95% confidence interval, 6.0-10.0). The risk of CDI was the highest among patients who underwent spine surgery (33.8 cases per 1,000 surgical procedures), followed by hip/femur surgery (12.4), knee (3.8), and extremity (3.2). The risk of CDI increased according to the increase in duration of proton pump inhibitor: 0.1% (no use), 0.3% (1-7 days), and 2.7% (>7 days, P < 0.001). The independent risk factors associated with postoperative CDI were age (odds ratio [OR] per 1-year increase, 1.04; P < 0.001), charlson comorbidity index score (OR per 1-point increase, 1.26; P < 0.001), duration of proton pump inhibitor (OR per 1-day increase, 1.02; P < 0.001), and operation time (OR per 1-hour increase, 1.30; P = 0.003). Of 6724 episodes of surgical procedure for which patients received exclusively perioperative antibacterial prophylaxis, 22 episodes of postoperative CDI occurred (3.2 cases per 1,000 surgical procedures). Among this subgroup, the risk of CDI increased according to increase in duration of antibacterial prophylaxis: 0% (<24 hour), 0.28% (1-7 days), 1.27 (>7 days; P < 0.001). After adjusting confounding factors, duration of perioperative antibacterial prophylaxis remained a significant risk factor for postoperative CDI (OR per 1-day increase, 1.11; P < 0.001). Patients with CDI had a higher rate of postoperative mortality (10.5% vs. 0.6%; P < 0.001) and an increased length of hospital stay (mean 42 vs. 10 days; P < 0.001).
Conclusion: Judicious use of proton pump inhibitor and avoiding of extension of prophylactic antibiotics can reduce postoperative CDI after orthopedic surgery.
D. Y. Kim,
Y. J. Kim, None
M. Lee, None
H. J. Lee, None
K. H. Park, None