1802. What is Best Available Therapy (BAT) for the Treatment of Carbapenem-resistant (CR) Gram-negative Infections in the US?
Session: Poster Abstract Session: Resistant Gram-Negative Infections: CRE Epidemiology
Saturday, October 10, 2015
Room: Poster Hall
  • IDWeek(#02)_poster 1802.pdf (4.3 MB)
  • Background: There are few antibiotics approved for the treatment of CR Gram-negative infections and their use is limited by drug toxicity and lack of evidence-based efficacy.  Published recommendations include a variety of combination regimens based on small observational studies, animal models or in vitro studies.

    Methods: Hospitalized patients with laboratory confirmed infection of E. coli, K. pneumoniae, P. aeruginosa and A. baumannii from specific infection sites (blood, respiratory, urine, other) were identified in electronic health records from Premier Healthcare Alliance database from 2009 to 2013. Two cohorts (Resistant and Susceptible) were identified based on antibiotic susceptibility. Antibiotic usage, including combination regimens, was classified into three time periods: Prior (before index date of infection); Empiric (from index date to day plus 3), or BAT (day 4-10). Treatments received during Empiric period and BAT periods by >=3% of patients for each bacterium and site of infection were described for each cohort.

    Results: Ten of 124 treatment patterns treated at least 3% of patients in BAT period and accounted for 24% of CR and 48.6% of CS A. baumannii blood stream infections (BSI) in the same period. Similarly, 9 of 207 treatment patterns with >=3% usage treated 22.5% of CR and 51.5% of CS A. baumannii respiratory infections. Usage of carbapenems in BAT period (17% to 38%) did not change  from its empiric period (17% to 38%) for CR A. baumannii and P. aeruginosa infections, but increased for CR of K. pneumoniae (by 11% to 44%) and E. coli infections (by 46% to 100%), and for all CS infections (by 20% to 108%). Colistin or polymyxin B (polymyxin) was used in 30% CR of A. baumannii BSI but 0% of CR E. coli BSI in BAT period. Polymyxin usage in BAT significantly increased from Empiric period for all CR infections (by 37% to 420%) except E. coli from blood or urine. Polymyxins were rarely used (0 to 2.8%) in CS infections. 

    Conclusion: CR was highly dependent on bacterial species and infection site. Antibiotic BAT was mostly combination treatment and highly variable but frequently included carbapenems. Polymyxin use was more common in A. baumannii BSI but not for other infections.  No standard antibiotic regimen exists for CR Gram-negative infections.

    Roger Echols, MD, FIDSA1, Bin Cai, MD, MPH, MS, PhD2, Eriko Ogura, MD3, Juan Camilo Arjona Ferreira, MD2, Mari Ariyasu, BA3, Takuko Sawada, BA3, Glenn Magee, BA, MBA4 and Tsutae Den Nagata, MD, PhD, FFPM3, (1)Clinical Development and Medical Affairs, SHIONOGI INC., Florham Park, NJ, (2)SHIONOGI INC., Florham Park, NJ, (3)SHIONOGI & CO., LTD., Osaka, Japan, (4)Premier Research Services, Premier Healthcare Alliance, Charlotte, NC


    R. Echols, SHIONOGI INC.: Consultant , Consulting fee

    B. Cai, SHIONOGI INC.: Employee , Salary

    E. Ogura, SHIONOGI & CO., LTD.: Employee , Salary

    J. C. Arjona Ferreira, SHIONOGI INC.: Employee , Salary

    M. Ariyasu, SHIONOGI & CO., LTD.: Employee , Salary

    T. Sawada, SHIONOGI & CO., LTD.: Employee , Salary

    G. Magee, None

    T. D. Nagata, SHIONOGI & CO., LTD.: Employee , Salary

    << Previous Abstract | Next Abstract

    Findings in the abstracts are embargoed until 12:01 a.m. PDT, Wednesday Oct. 7th with the exception of research findings presented at the IDWeek press conferences.