Mortality After Piperacillin-Tazobactam–Resistant vs –Susceptible Bloodstream Infection

Importance: Use of piperacillin-tazobactam (TZP) as treatment for bloodstream infections (BSIs) has increased over time in Europe, as has antimicrobial resistance to TZP in some European countries. Objective: To evaluate mortality associated with TZP-resistant BSIs in the Danish health care system. Design, Setting, and Participants: In this cohort study, data collected in near real time were obtained from Danish national health registers from November 15, 2018, through November 21, 2024. Participants were patients aged 18 years or older with monomicrobial Escherichia coli or Klebsiella pneumoniae BSI, restricted to the first positive (index) blood culture with a TZP-susceptibility test result during the follow-up period. Exposures: E coli or K pneumoniae BSI and a TZP susceptibility test classified as TZP resistant or TZP susceptible. Main Outcome and Measures: Thirty-day mortality hazard ratios (HRs) for BSIs with TZP-resistant vs TZP-susceptible bacteria were estimated from the index date (blood culture) until death, emigration, or end of the observation period. Mortality was drawn from Danish national health registers, with no restriction on in-hospital or postdischarge mortality. Univariable and multivariable models adjusted a priori for self-reported gender, age, comorbidity, year, length of hospital stay until BSI, total length of inpatient stays in the prior year, and BSI in the prior 6 months with species other than that in the index culture. Results: The study population included 34 379 patients (median overall age, 76 years [IQR, 66-83 years]; 51.5% male). Among 28 649 patients (81.9%) with E coli BSI, 1689 (5.9%) had a TZP-resistant BSI; among 6314 patients (18.1%) with K pneumoniae BSI, 597 (9.5%) had a TZP-resistant BSI. There was no loss to follow-up, and about 1% of participants were right-censored by the end of the observation period (E coli: 343 [1.2%]; K pneumoniae, 88 [1.4%]). Mortality was similar for patients with TZP-resistant and TZP-susceptible BSI after adjusting for baseline risk factors (E coli: crude HR, 1.22 [95% CI, 1.08-1.38]; adjusted HR [AHR], 1.09 [95% CI, 0.94-1.27]; K pneumoniae: crude HR, 1.27 [95% CI, 1.05-1.52]; AHR, 1.15 [95% CI, 0.94-1.41]). Conclusions and Relevance: In this cohort study of patients with monomicrobial E coli or K pneumoniae BSI, there was no significant difference in mortality for patients with BSI caused by TZP-resistant vs TZP-susceptible bacteria after adjustment for baseline risk factors. The result is important given increasing TZP resistance in Denmark and other countries.