Association of Duration and Type of Surgical Prophylaxis With Antimicrobial-Associated Adverse Events

IMPORTANCE The benefits of antimicrobial prophylaxis are limited to the first 24 hours postoperatively. Little is known about the harms associated with continuing antimicrobial prophylaxis after skin closure. OBJECTIVE To characterize the association of type and duration of prophylaxis with surgical site infection (SSI), acute kidney injury (AKI), and Clostridium difficile infection. DESIGN, SETTING, AND PARTICIPANTS In this multicenter, national retrospective cohort study, all patients within the national Veterans Affairs health care system who underwent cardiac, orthopedic total joint replacement, colorectal, and vascular procedures and who received planned manual review by a trained nurse reviewer for type and duration of surgical prophylaxis and for SSI from October 1, 2008, to September 30, 2013, were included. Data were analyzed using multivariable logistic regression, with adjustments for covariates determined a priori to be associated with the outcomes of interest. Data were analyzed from December 2016 to December 2018. EXPOSURES Duration of postoperative antimicrobial prophylaxis (< 24 hours, 24-< 48 hours, 48-< 72 hours, and >= 72 hours). MAIN OUTCOMES AND MEASURES Surgical site infection, AKI, and C difficile infection. RESULTS Of the 79 058 included patients, 76 109 (96.3%) were men, and the mean (SD) age was 64.8 (9.4) years. Among 79 058 surgical procedures in the cohort, all had SSI and C difficile outcome data available; 71 344 (90.2%) had AKI outcome data. After stratification by type of surgery and adjustment for age, sex, race, diabetes, smoking, American Society of Anesthesiologists score greater than 2, methicillin-resistant Staphylococcus aureus colonization, mupirocin, type of prophylaxis, and facility factors, SSI was not associated with duration of prophylaxis. Adjusted odds of AKI increased with each additional day of prophylaxis (cardiac procedure: 24-< 48 hours: adjusted odds ratio [aOR], 1.03; 95% CI, 0.95-1.12; 48-< 72 hours: aOR, 1.22; 95% CI, 1.08-1.39; >= 72 hours: aOR, 1.82; 95% CI, 1.54-2.16; noncardiac procedure: 24-< 48 hours: aOR, 1.31; 95% CI, 1.21-1.42; 48-< 72 hours: aOR, 1.72; 95% CI, 1.47-2.01; >= 72 hours: aOR, 1.79; 95% CI, 1.27-2.53). The risk of postoperative C difficile infection demonstrated a similar duration-dependent association (24-< 48 hours: aOR 1.08; 95% CI, 0.89-1.31; 48-< 72 hours: aOR, 2.43; 95% CI, 1.80-3.27; >= 72 hours: aOR, 3.65;95% CI, 2.40-5.53). The unadjusted numbers needed to harm for AKI after 24 to less than 48 hours, 48 to less than 72 hours, and 72 hours or more of postoperative prophylaxis were 9, 6, and 4, respectively; and 2000, 90, and 50 for C difficile infection, respectively. Vancomycin receipt was also a significant risk factor for AKI (cardiac procedure: aOR, 1.17; 95% CI, 1.10-1.25; noncardiac procedure: aOR, 1.21; 95% CI, 1.13-1.30). CONCLUSIONS AND RELEVANCE Increasing duration of antimicrobial prophylaxiswas associated with higher odds of AKI and C difficile infection in a duration-dependent fashion; extended duration did not lead to additional SSI reduction. These findings highlight the notion that every day matters and suggest that stewardship efforts to limit duration of prophylaxis have the potential to reduce adverse events without increasing SSI.