Intravenous antibiotic duration in the treatment of prosthetic joint infection: systematic review and meta-analysis.

Introduction: Long antibiotic courses, including intravenous (IV) and oral administrations, are utilized in prosthetic joint infection (PJI) treatment. This meta-analysis examines the non-inferiority of short courses ( $<$  4 weeks) of IV antibiotics compared to long courses in treating PJI. Critical review of IV treatment is necessary due to the clinical, physical, and financial burden associated with it and its continued prolonged use in the US without much evidence to support the practice. Methods: Following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA), databases were searched using predefined medical subject headings (MeSH). Results: The nine included studies reported 521 total hip arthroplasties (THAs) and 530 total knee arthroplasties (TKAs). There was no significant difference in the overall success rate in short- vs. long-duration IV antibiotics for PJI treatment: odds ratio (OR) of 1.65, 95 % confidence interval (CI) of 0.78-3.46, and $p=0$ .18. However, due to the moderate to high heterogeneity ( $I^2=68$  %, $p$   $<$  0.01) amongst studies, an adjusted success rate was calculated after the exclusion of two studies. This showed a statistically significant difference between both groups (OR of 2.45, 95 % CI of 1.21-4.96, $p$   $<$  0.001) favoring a short course of antibiotics and reflecting a more homogenous population ( $I^2=51$  %, $p=0$ .06). Conclusion: This study highlights the limited data available for evaluating IV antibiotic duration in the setting of PJI. We found that a shorter duration of IV antibiotics was non-inferior to a longer duration, with an improved OR of 2.45 for treatment success, likely shortening inpatient stay as well as lessening side effects and antimicrobial resistance with a lower cost to patients and overall healthcare.