Short- versus long-course antibiotic therapy in mechanically ventilated sepsis patients with pneumonia: a real-world cohort analysis

Abstract

Objective

To compare outcomes of short-course versus longer-course antibiotic regimens in mechanically ventilated adult patients with sepsis due to pneumonia.

Methods

The authors performed a retrospective cohort study using the MIMIC-IV database (2008–2022), identifying 2590 adults meeting Sepsis-3 criteria with a primary diagnosis of pneumonia requiring invasive mechanical ventilation. Patients were stratified by antibiotic duration (short: 2‒7 days; long: 8‒14 days). Demographics, comorbidities, severity (SOFA score, PaO₂/FiO₂ ratio), and microbiology were extracted. The primary outcome was 90-day all-cause mortality. The authors used Kaplan-Meier analysis, multivariable Cox proportional hazards models, and subgroup analyses to assess the effect of treatment duration.

Results

The study found that overall 90-day mortality did not differ significantly (short-course ∼41.3% vs. long-course ∼41.2%; p = 0.98). Subgroup analyses indicated that in patients > 65-years, those with severe hypoxemia (low PaO₂/FiO₂), on RRT, or infected with Staphylococcus aureus or Pseudomonas aeruginosa, shorter therapy was associated with higher mortality rates (p < 0.05). Conversely, younger or less severely ill patients showed no significant effect of duration. These findings may reflect residual confounding.

Conclusions

In this observational cohort, shorter antibiotic courses were not associated with increased 90‑day mortality after adjustment. Given potential confounding and limited recurrence data, subgroup signals are hypothesis‑generating and require confirmation in randomized trials.