Piperacillin-tazobactam vs. carbapenems for treating hospitalized patients with ESBL-producing Enterobacterales bloodstream infections: A systematic review and meta-analysis

Abstract

Objectives: To meta-analyse the clinical efficacy of piperacillin-tazobactam vs. carbapenems for treating hospitalized patients affected by extended-spectrum beta-lactamase (ESBL)-producing Enterobacterales bloodstream infections (BSIs). Methods: Two authors independently searched PubMed-MEDLINE and Scopus database up to January 17, 2024, to retrieve randomized controlled trials (RCTs) or observational studies comparing piperacillin-tazobactam vs. carbapenems for the management of hospitalized patients with ESBL-BSIs. Data were independently extracted by the two authors, and the quality of included studies was independently assessed according to ROB 2.0 or ROBINS-I tools. Mortality rate was selected as primary outcome. Meta-analysis was performed by pooling odds ratios (ORs) retrieved from studies providing adjustment for confounders using a random-effects model with the inverse variance method. Results: After screening 3,418 articles, 10 studies were meta-analysed (one RCT and nine retrospective observational studies; N = 1,962). Mortality rate did not significantly differ between treatment with piperacillin-tazobactam vs. carbapenems (N = 6; OR: 1.41; 95% CI: 0.96–2.07; I² = 23.6%). The findings were consistent also in subgroup analyses assessing patients receiving empirical therapy (N = 5; OR: 1.36; 95% CI: 0.99–1.85), or patients having in ≥50% of cases urinary/biliary tract as the primary BSI source (N = 2; OR: 1.26; 95% CI: 0.84–1.89). Conversely, the mortality rate was significantly higher with piperacillin-tazobactam only among patients having in <50% of cases urinary/biliary tract as the primary source of BSI (N = 3; OR: 2.02; 95% CI: 1.00–4.07). Conclusions: This meta-analysis showed that, after performing appropriate adjustments for confounders, mortality and clinical outcome in patients having ESBL-producing Enterobacterales BSIs did not significantly differ among those receiving piperacillin-tazobactam compared to those receiving carbapenems.