Comparison of short versus long treatment duration of ESBL pneumonia in critically ill patients: a retrospective cohort study

Extended-spectrum β-lactamase-producing Enterobacterales (ESBL-E) are an increasing cause of pneumonia in critically ill patients. Current guidelines recommend 7 days of antibiotic therapy for hospital-acquired and ventilator-associated pneumonia, but evidence specific to ESBL-E pneumonia is limited. This study aims to compare the clinical outcomes of short versus long antibiotic treatment duration in critically ill patients with pneumonia caused by ESBL-E. This retrospective cohort study included adults admitted to an intensive care unit with pneumonia caused by ESBL-E between January 2019 and August 2024. Patients were categorised into short-duration (≤7 days) or long-duration (>7 days) antibiotic therapy groups. Logistic regression was used to evaluate the primary outcome of clinical cure. A total of 121 patients were included (64 short-duration, 57 long-duration). Clinical cure was achieved in 56.3% vs. 38.6% (p = 0.052) in unadjusted analysis. In the adjusted multivariable model, treatment duration was not significantly associated with clinical cure (OR 0.403; 95% CI 0.126–1.288; p = 0.125). Treatment failure occurred in 50.0% vs. 68.4% (p = 0.040), and microbiologic cure in 78.1% vs. 59.7% (p = 0.028). Median hospital length of stay was shorter with short-duration therapy (20.5 vs. 38.1 days, p < 0.001). Recurrence, all-cause mortality, ICU length of stay, mechanical ventilation, and antibiotic-free days were not significantly different. Multidrug-resistant organisms developed in 7.8% vs. 14.0% (p = 0.270). Short-duration antibiotic therapy for pneumonia caused by ESBL-E in critically ill patients showed no difference in clinical cure compared to long-duration therapy. Further studies are needed to determine the effect of antibiotic duration on clinical outcomes.