Effect of Empirical Antimicrobial Therapy of Piperacillin-tazobactam Versus Carbapenems on Mortality for Patients with Sepsis from Nursing Homes

Objective Nursing home residents with a high risk of multidrug-resistant organism infection pose a complex challenge to broad-spectrum empirical antimicrobial therapy, particularly those infected with extended-spectrum β-lactamase-producing Enterobacteriaceae. The present study compared the efficacy of piperacillin-tazobactam and carbapenems as empirical antimicrobial treatments for patients with sepsis from nursing homes. Methods Using a nationwide inpatient database in Japan, we identified patients diagnosed with sepsis within two days of admission from nursing homes between 2018 and 2021. We selected patients who received intravenous piperacillin-tazobactam or carbapenems within two days of admission. In-hospital mortality was compared between the piperacillin-tazobactam and carbapenem groups using inverse probability of treatment weighting. Results We identified 8,025 eligible patients. Of these, 3,391 (42%) received piperacillin-tazobactam, and 4,634 (58%) received carbapenems within 2 days of admission. The inverse probability of treatment weighting analysis showed no significant difference in in-hospital mortality between the groups (31.6% in the piperacillin-tazobactam group and 32.8% in the carbapenem group; risk difference, 1.2%; 95% confidence interval, -3.2% to 0.9%). Conclusion Carbapenems and piperacillin-tazobactam as empirical antimicrobial therapy in patients with sepsis from nursing homes were associated with comparable in-hospital mortality rates. These findings highlight the importance of making decisions regarding broad-spectrum empirical antimicrobial therapy.