Comparative effectiveness of two antimicrobial regimens for the treatment of pan-drug-resistant <i>Acinetobacter baumannii</i> infections: results from the DESPAIR study

Pan-drug-resistant <i>Acinetobacter baumannii</i> (PDR-AB) causes severe infections and constitutes a threat in several geographic regions. Little is known about the differential effectiveness of last-resort regimens, frequently used in this setting. We compared the effectiveness of two literature-proposed regimens against PDR-AB infections consisting of colistin, ampicillin-sulbactam, and either meropenem (regimen A) or tigecycline (regimen B). This is a retrospective analysis of prospectively collected data from 12 centers on adult patients with hospital-acquired pneumonia (HAP) or bloodstream infection (BSI), who had received definitive treatment with either regimen A or regimen B. The primary outcome was clinical failure, defined as any of the following occurring by day 14 from infection onset: death, initiation of salvage treatment, treatment withdrawal due to toxicity, persistent bacteremia for BSI patients and failure to improve oxygenation for HAP patients before and after propensity matching. Eighty-three patients were included in the primary analysis; 60 had received regimen A and 23 regimen B. Regimen B was significantly associated with clinical failure before and after propensity matching (odds ratios [OR]: 3.11; 95% confidence interval [CI]: 1.10-8.84 vs OR: 3.83; 95% CI: 1.26-11.63), respectively. Salvage therapy and treatment discontinuation due to toxicity were more frequent in patients treated with regimen B. In multivariable analysis, regimen B was independently associated with 28-day mortality before (hazard ratio [HR]: 2.53; 95% CI: 1.08-5.94) but not after propensity matching (HR: 2.64; 95% CI: [0.99-7.02]). Treatment with colistin, ampicillin-sulbactam, and meropenem against severe PDR-AB infections was associated with favorable outcomes compared to colistin, ampicillin-sulbactam, and tigecycline.