P02. Carbapenem-resistant Enterobacterales acquisition following piperacillin-tazobactam versus meropenem treatment: a propensity-matched cohort study

Abstract Objectives We aimed to quantify the ecological consequences of carbapenems versus piperacillin-tazobactam treatment on the risk of carbapenem-resistant Enterobacterales (CRE) acquisition. Methods We conducted a retrospective cohort study with propensity-score matching. The study was conducted in a single large hospital in Israel between 2014–2023. We included all adult patients treated with piperacillin-tazobactam (PTZ) or a carbapenem for at least 5 days. We excluded patients with known carriage of CRE at treatment initiation and patients in the PTZ group who received a carbapenem. The outcome assessed was CRE acquisition through surveillance or clinical samples in the 90 days after end of treatment. The propensity score for treatment selection was derived to match patients in the treatment groups using 1:1 nearest-neighbor matching, with a caliper of 0.01. A sensitivity analysis of the full cohort was performed through a regression model of CRE acquisition, adjusting the analysis using inverse probability of treatment weighting. Results We included 7731 patients of whom 6.2% (483/7731) acquired CRE: 9.3% (264/2844) of patients treated with a carbapenem acquired CRE versus 4.5% (219/4887) of patients treated with PTZ (unadjusted odds ratio [OR] 0.46, 95% confidence interval [CI] 0.38–0.55). In the matched cohort, CRE acquisition occurred in 6.2% (122/1980) of patients treated with PTZ versus 7.9% (157/1980) of patients treated with carbapenems (PS-adjusted OR 0.76. 95% CI 0.6–0.97). Independent risk factors for CRE acquisition included hospitalization in the hemato-oncology or neurosurgical wards, longer antibiotic exposure and hospitalization, bloodstream infections, low albumin and higher glucose. Adjusted to these and weighted by the inverse probability of treatment, PTZ was associated with CRE acquisition with an OR of 0.81, 95% CI (0.67–0.99). Conclusions Carbapenems are significantly associated with CRE acquisition compared to PTZ in a CRE endemic settings. This consideration should be factored into guidance on carbapenem versus PTZ treatment.