PT - JOURNAL ARTICLE AU - Patrick Thornley AU - Nathan Evaniew AU - Michael Riediger AU - Mitchell Winemaker AU - Mohit Bhandari AU - Michelle Ghert TI - Postoperative antibiotic prophylaxis in total hip and knee arthroplasty: a systematic review and meta-analysis of randomized controlled trials AID - 10.9778/cmajo.20150012 DP - 2015 Jul 17 TA - CMAJ Open PG - E338--E343 VI - 3 IP - 3 4099 - http://www.cmajopen.ca/content/3/3/E338.short 4100 - http://www.cmajopen.ca/content/3/3/E338.full AB - Background: Postoperative antibiotic prophylaxis is currently the standard of care for patients undergoing total hip and knee arthroplasty. We evaluated the evidence for this practice in the reduction of surgical-site infections.Methods: We systematically searched MEDLINE, Embase and the Cochrane Library for randomized controlled trials (RCTs) published up to Aug. 15, 2014. We included all RCTs that compared postoperative antibiotic prophylaxis with postoperative placebo or no treatment in patients undergoing primary total hip or knee arthroplasty for osteoarthritis. We combined outcomes for surgical-site infection using a random-effects model and quantified heterogeneity using the χ2 test and the I2 statistic. We assessed the overall quality of the evidence according to the Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach.Results: We identified 4 RCTs (n = 4036) that met the inclusion criteria. Surgical-site infections occurred in 3.1% (63/2055) of patients in the prophylaxis group and 2.3% (45/1981) in the control group. Postoperative prophylaxis did not reduce the rate of surgical-site infections compared with placebo (risk difference 0.01, 95% confidence interval 0.00 to 0.02; I2 = 26%). This result was robust to sensitivity testing for losses to follow-up. According to the GRADE approach, the overall quality of evidence was very low.Interpretation: The available evidence did not show efficacy of postoperative antibiotic prophylaxis for the prevention of surgical-site infections in patients undergoing total hip or knee arthroplasty. Multicentred RCTs are likely to have an important impact on the confidence in the effect estimate and to change the estimate itself.