Background: Infections caused by Acinetobacter spp are often healthcare acquired and associated with high mortality. Extensively drug-resistant (XDR) Acinetobacter are nonsusceptible to at least 1 agent in all but 2 or fewer antibiotic classes. Few of the new antibiotics targeting multidrug-resistant gram-negative bacteria are effective against XDR Acinetobacter. Recent national guidelines for treatment of resistant gram-negative infections do not include Acinetobacter, leaving a knowledge gap in best practices. Methods: This retrospective cohort study included microbiology, clinical, and pharmacy data from all patients hospitalized between 2012 and 2018 at any Veterans’ Affairs medical center who had cultures that grew XDR Acinetobacter spp. Bivariate unadjusted analyses compared clinical outcomes by monotherapy versus combination therapy. Using mixed-effects ordinal logistic regression, propensity score–adjusted models accounting for severity of illness and other variables associated with treatment were fit to compare outcomes. Results: Of 11,546 patients with 15,364 cultures that grew Acinetobacter spp, 408 patients (3.5%) had 666 cultures (4.3%) with XDR Acinetobacter. Moreover, 276 of these patients (67.6%) had gram-negative targeted antibiotic treatment within −2 to +5 days from the culture. Furthermore, 118 patients (42.8%) received monotherapy, most commonly piperacillin-tazobactam (n = 54, 45.7%) or an anti-Pseudomonas cephalosporin (n = 21, 17.8%). Also, 158 (57.2%) patients received combination therapy, most commonly a carbapenem (n = 93, 58.9%) and/or polymyxin (n = 68, 43.0%). Moreover, 41 patients (25.9%) received both a carbapenem and polymyxin. In both unadjusted and adjusted analyses, there were no significant differences in the odds of 30-day mortality (aOR, 1.43; 95% CI, 0.86–2.38) or 1-year mortality (aOR, 1.04; 95% CI, 0.68–1.60) between combination therapy and monotherapy groups. Among 264 patients (96%) whose cultures occurred during an inpatient or long-term care admission, unadjusted analyses showed increased odds of in-hospital mortality (OR, 1.89; 95% CI, 1.08–3.29) and longer postculture length of stay in the combination therapy group: median, 23 days (IQR, 11–57) versus 14 days (IQR, 7–32) (P = .02). However, with propensity score adjustment, these associations were no longer significant. Furthermore, there was no significant difference in odds of 90-day readmission between groups in either unadjusted or adjusted analyses (aOR, 1.20; 95% CI, 0.74–1.95). Conclusions: In this large national cohort of patients with XDR Acinetobacter cultures, more patients received combination therapy than monotherapy, and carbapenems and polymyxins were the most-used classes. However, there were no significant differences in outcomes between patients receiving combination therapy and monotherapy, suggesting lack of clinical benefit to the common practice of treating XDR Acinetobacter infections with multiple antibiotics. Further research is needed to determine optimal treatment strategies for this pathogen.

Funding: No

Disclosures: None