Inappropriate Empirical Antibiotic Therapy and 28-Day Mortality in Critically Ill Patients: A Retrospective Cohort and Propensity Score Analysis in an Indonesian ICU

Abstract

Background. Inappropriate empirical antibiotic therapy (IEAT) is linked to higher mortality in sepsis, particularly where antimicrobial resistance is highly endemic, as in Southeast Asia.

Objective. To estimate the association between IEAT and 28-day mortality in Indonesian ICU patients while accounting for confounding by indication.

Methods. This retrospective cohort included 280 adults admitted to a tertiary academic ICU in Malang, Indonesia, from January 2022 to December 2023, with positive cultures drawn within 48 hours. Empirical regimens were classed by in-vitro susceptibility as appropriate (AEAT, n = 172) or inappropriate (IEAT, n = 108). Analyses used multivariable logistic regression and, as a second approach, 1:1 propensity score matching on seven clinical covariates.

Results. IEAT was given to 108 patients (38.6%). Overall 28-day mortality was 35.0% (98/280): 77.8% (84/108) with IEAT and 8.1% (14/172) with AEAT (p < 0.001). After adjustment for age, APACHE II score, and septic shock, IEAT had the largest adjusted odds ratio of the model (aOR 38.72; 95% CI 18.91–79.30; p < 0.001). In the matched cohort (n = 200; standardized mean differences < 0.10), the association persisted (OR 25.15; 95% CI 11.54–54.81; p < 0.001). Levofloxacin monotherapy was inappropriate in 76.4% of cases and meropenem in 42.5%. ESBL Klebsiella pneumoniae (20.0%) and XDR Acinetobacter baumannii (17.1%) were prevalent.

Conclusion. IEAT was associated with much higher 28-day mortality in this cohort. These findings support a review of local stewardship policies. Empirical fluoroquinolone use in tertiary intensive care should be restricted.