2262. Ceftazidime–Avibactam vs. Polymyxin B in the Treatment of Infections Due to Carbapenem-Resistant Enterobacteriaceae

Abstract

Background. Pharmacotherapy for carbapenem-resistant Enterobacteriaceae (CRE) infections is limited. There is a paucity of evidence to guide optimal management of CRE infections. Ceftazidime-avibactam, a novel cephalosporin/β-lactamase inhibitor, may be a reasonable alternative to colistin for CRE infections, but data on polymyxin B (PB) are lacking. Given the improved pharmacokinetic profile of PB compared with colistin, we sought to evaluate clinical and microbiological outcomes of patients treated with CAZ-AVI vs. PB for CRE infections. Methods. We conducted retrospective cohort study in adult patients treated with CAZ-AVI or PB for a CRE infection between June 2010 and August 2018. The primary outcome was all-cause mortality at 30 days. Secondary outcomes included clinical cure, microbiological cure, and development of resistance. Endpoints were analyzed using standard statistical measures. The influence of clinical variables other than antimicrobial therapy was assessed in a multivariable regression analysis. Results. The study included 117 patients, with 42 patients receiving CAZAVI and 75 receiving PB. Respiratory and urinary tract infections were most common, occurring in 37.6% and 20.5% of patients, respectively. Bloodstream infections occurred in 45 (35.9%) patients. In the CAZ-AVI group, there were 9 deaths (21.4%), compared with 19 deaths (25.3%) in the PB group (P = 0.653). No statistically significant differences were found in clinical cure or microbiologic cure between CAZ-AVI and PB. PB was associated with a higher incidence of nephrotoxicity (19% vs. 43%; P = 0.048). After adjustment for duration of therapy, combination therapy, and initial WBC, use of PB was not an independent predictor of mortality. Conclusion. No statistically significant differences between CAZ-AVI and PB were found in clinical or microbiologic outcomes in this cohort of patients treated for CRE infection. Further studies are necessary to confirm these preliminary findings to optimize clinical practice.