Seven days versus extended duration antibiotic therapy for multidrug-resistant gram-negative bacterial infections in critically ill patients: A pooled analysis of the OPTIMISE and REGARD-VAP trials.

Multidrug-resistant Gram-negative bacteria (MDR-GNB) are increasingly encountered in intensive care units (ICUs), where prolonged antibiotic courses are often prescribed despite limited evidence guiding optimal duration. Determining whether shorter regimens are as effective as extended therapy is essential for stewardship and patient outcomes. We performed an aggregate data meta-analysis pooling two randomized controlled trials (RCTs): OPTIMISE (7 vs. 14 days) and REGARD-VAP (individualized 6-7 vs. 8-9 days) in MDR-GNB infections. The primary outcomes were composite outcome of mortality and reinfection/recurrence, mortality, and reinfection/recurrence. Random- effects models generated pooled odds ratios (OR) with 95 % confidence intervals (CI). Across 602 ICU patients, Klebsiella pneumoniae was the most frequent resistant pathogen, with Pseudomonas aeruginosa more prominent in OPTIMISE and Acinetobacter baumannii in REGARD-VAP. Short-course therapy did not significantly differ from extended therapy for mortality (OR 0.74, 95 % CI 0.45-1.23, I2 = 18 %) or clinical failure (OR 0.82, 95 % CI 0.50-1.36, I2 = 0 %). Relapse occurred more often with short-course therapy (OR 2.04, 95 % CI 0.85-4.92, I2 = 22 %) but was not statistically significant. Shorter regimens appear comparable to extended courses in critically ill patients, though potential relapse risk warrants further study. Future individual patient data meta-analyses may clarify pathogen- and host-specific factors guiding optimal duration.