327. Trends in Gram-negative Bloodstream Isolates with Limited High-efficacyLow-toxicity Antibiotic Options among Inpatients at 180 US Hospitals
Session: Poster Abstract Session: HAI: Multi Drug Resistant Gram Negatives
Thursday, October 27, 2016
Room: Poster Hall

Background: Gram-negative bacterial bloodstream isolates (GNBSIs) resistant to all available high-efficacy/low-toxicity (HELT) antibiotics (quinolones, penicillins, cephalosporins, carbapenems) exhibit difficult to treat resistance (DTR), which limits options for empiric and targeted therapy. The wrong empiric antibiotic may be selected even when GNBSIs remain susceptible to 1 or 2 HELT agents.  Analyzing DTR trends allows assessment of the potential burden of inappropriate therapy and impact of antimicrobial resistance control interventions.

Methods: Inpatient encounters (2009-2013) with ≥ 1 positive GNBSI (E. coli, Enterobacter spp., Klebsiella spp., P. aeruginosa or A. baumannii) were analyzed using PremierTM Database linked patient data from 180 US hospitals. Laboratory-reported categorical susceptibility breakpoints were used to categorize GNBSIs into respective DTR strata (non-DTR, DTR-2, DTR-1 and full DTR, denoting >2, 2, 1 and 0 active HELT antibiotic classes, respectively). Annual prevalence of DTR/GNBSIs was calculated. Overall and species-specific trends were evaluated using Poisson regression to estimate the significance (p<0.05) of annual percent change (APC).

Results: Of 46,498 GNBSI encounters, 2,911 (6.3%) were DTR (468 Full DTR; 1,030 DTR-1; 1,413 DTR-2) (Table 1).  The GNBSI species most frequently exhibiting Full DTR was A. baumannii (39%); E. coli was the most common species with DTR-1 (45%) and DTR-2 (59%). For most DTR-1 or 2 (69.6%), a carbapenem was the only active agent (Figure 1). The prevalence of DTR-1 or 2 increased over time (APC= 1.14, p=0.0002), while Full DTR remained stable (APC= 0.94, p=0.07) (Figure 2); none of the species-specific trends among those with DTR-1 or 2 demonstrated a significant APC.

Conclusion: We identified a significant increase in GNBSIs susceptible only to 1 or 2 HELT antibiotic classes, although overall prevalence remained low. Most often, only carbapenems remained effective. Regardless of the contribution of changing susceptibility breakpoint standards over time, the trend reflects the expanding burden of challenging resistance patterns faced by providers. How these observed trends affect empiric antibiotic selections over time requires further investigation.


 

Alicen B. Spaulding, PhD, MPH1, Yi Ling Lai, MPH1, Jennifer Adjemian, PhD2, D. Rebecca Prevots, PhD, MPH3, Tara Palmore, MD4, Chanu Rhee, MD, MPH5, Michael Klompas, MD, MPH, FRCPC, FIDSA5, John P. Dekker, M.D., Ph.D.6, Anthony Suffredini, MD7, David C. Hooper, MD8, Scott Fridkin, MD, FIDSA9, Robert L. Danner, MD7 and Sameer S. Kadri, MD, MS.10, (1)NIH/NIAID, Bethesda, MD, (2)NIH/NIAD, Bethesda, MD, (3)Epidemiology Unit, Laboratory of Clinical Infectious Diseases, National Institute of Allergy and Infectious Diseases (NIAID), National Institutes of Health, Bethesda, MD, (4)NIH/CC, Bethesda, MD, (5)Department of Population Medicine, Harvard Medical School and Harvard Pilgrim Health Care Institute, Boston, MA, (6)Department of Laboratory Medicine, NIH Clinical Center, NIH, Bethesda, MD, (7)Critical Care Medicine, National Institutes of Health, Bethesda, MD, (8)Harvard Medical School and Massachusetts General Hospital, Boston, MA, (9)Division of Healthcare Quality Promotion (DHQP), Centers for Disease Control and Prevention, Atlanta, GA, (10)Critical Care Medicine Department, National Institutes of Health Clinical Center, Bethesda, MD

Disclosures:

A. B. Spaulding, None

Y. L. Lai, None

J. Adjemian, None

D. R. Prevots, None

T. Palmore, None

C. Rhee, None

M. Klompas, None

J. P. Dekker, None

A. Suffredini, None

D. C. Hooper, None

S. Fridkin, None

R. L. Danner, None

S. S. Kadri, None

Findings in the abstracts are embargoed until 12:01 a.m. CDT, Wednesday Oct. 26th with the exception of research findings presented at the IDWeek press conferences.