# Postoperative antibiotic treatment does not lower re-revision rate in presumed aseptic hip and knee revision arthroplasties with unexpected positive intraoperative cultures – a matched cohort study

## TL;DR

This study finds that giving antibiotics after hip or knee revision surgeries with unexpected positive cultures does not reduce the need for further surgery.

## Contribution

The study provides evidence that postoperative antibiotics do not lower re-revision rates in presumed aseptic revisions with unexpected positive cultures.

## Key findings

- Postoperative antibiotic treatment did not reduce re-revision rates (14.3% vs 15.7%) in patients with unexpected positive cultures.
- No confirmed periprosthetic joint infections were diagnosed in either group.
- Most unexpected positive cultures were from low-virulence organisms considered contaminants.

## Abstract

Aims: It remains unclear if postoperative antibiotic (AB) treatment is advantageous in presumed aseptic revision arthroplasties of the hip (rTHA) and knee (rTKA) with unexpected positive intraoperative cultures (UPIC). The aim of this study is to evaluate if there is a difference in the re-revision rate in patients with UPIC when treated with postoperative AB or when postoperative AB is withheld. Methods: In this retrospective matched cohort study we compared the re-revision rates in rTHA and rTKA with (AB group: 45 rTHA, 25 rTKA) and without (non-AB group: 45 rTHA, 25 rTKA) AB treatment in patients with UPIC. Baseline covariates for matching were the microorganism (likely or not likely to be a contaminant), patient demographics, joint, revision type, surgical site infection score, American Society of Anesthesiologists classification, serum C-reactive protein (CRP). Results: After a median follow-up of 4.1 (inter-quartile range, IQR: 2.9–5.5) years after rTHA and rTKA, the re-revision rate between the AB group and the non-AB group was 14.3 % versus 15.7 % (P=0.81). In the AB group, 4.3 % (3/70) of patients underwent revision due to septic complications compared to 5.7 % (4/70) in the non-AB group (P=0.69). None of the patients were diagnosed with a confirmed periprosthetic joint infection (PJI) according to the PJI diagnostic criteria of European Bone and Joint Infection Society (EBJIS). In 22/70 (31.4 %) of the patients in the AB group and in 15/70 (21.4 %) of the patients in the non-AB group, a diagnosis of “infection likely” was made according to the EBJIS criteria (P=0.18). All UPICs with low virulent microorganisms were considered to be contamination (coagulase-negative Staphylococci; Corynebacterium; anaerobic Gram-positive bacilli and cocci, e.g., Finegoldia magna, Cutibacterium acnes). Conclusion: Postoperative AB treatment did not result in a decreased re-revision rate in patients with UPIC in presumed aseptic rTHA and rTKA. Patients diagnosed with pathogens classified as a likely contaminant can be safely ignored.