1031. Nationwide Temporal Trends of Candidemia Incidence over 18 Years within the Veteran Health Administration System
Session: Poster Abstract Session: Bacteremia and Endocarditis
Friday, October 5, 2018
Room: S Poster Hall
  • ID week 2018 - 1031 H Suzuki.pdf (346.8 kB)
  • Background: Bloodstream infection due to Candida spp. is common and associated with significant mortality and morbidity. Previous population-based studies in 2000s and early 2010s have suggested that the incidence of candidemia might be increasing, presumably due to widespread use of central lines and broad-spectrum antibiotics. However, recent trends of candidemia incidence have not been not well described.

    Methods: We conducted a retrospective cohort study of all veterans cared for in the Veterans Health Administration (VHA) system from January 2000 to December 2017 to determine the incidence of candidemia. All patients who had positive blood cultures were identified using data available in the electronic medical record data warehouse, and the number of unique patients for each month was calculated. Patient-days was used as a denominator, and the incidence rate was expressed as the number of unique patients with candidemia per patient-days for each month. Temporal trends were analyzed by joinpoint regression models to identify statistically significant changes in trend.

    Results: Over the study period, 31,370 positive blood cultures for Candida spp. from 15,763 unique patients were identified. The mean monthly incidence rate was 22.5 per 100,000 patient-days (IQR: 15.6-28.4). Incidence rates were increasing in the early 2000s and relatively stable in the mid-2000s, followed by a sustained decline (Figure). Joinpoint regression analysis revealed there were two statistically significant changes in slope, one in 9/2003 (95% CI: 2/2002-1/2005) and another in 6/2007 (95% CI: 4/2006-3/2009).

    Conclusion: In the VHA system, there were significant changes in temporal trends of candidemia incidence rates over 18 years, including a substantial increase in the early 2000s followed by a sustained decline in later years. The incidence rates during 2016-2017 were nearly one-third of their peak in the mid-2000s. Possible explanations for the sustained decline include prevention efforts for healthcare-associated infections, such as central-line associated bloodstream infections. Further study is needed to investigate etiologies of these changes in temporal trends to identify potential effective prevention for candidemia.

    Hiroyuki Suzuki, MD1, Eli Perencevich, MD, MS, FIDSA, FSHEA2, Daniel Diekema, MD, FIDSA, FSHEA3, Daniel Livorsi, MD, MSc1, Marin Schweizer, PhD4, Rajeshwari Nair, PhD, MBBS, MPH1, Michael Ohl, MD MSPH1, Kelly Richardson, PhD2, Brice Beck, MA2, Bruce Alexander, PharmD2 and Michihiko Goto, MD, MSCI1, (1)Department of Internal Medicine, University of Iowa Carver College of Medicine, Iowa City, IA, (2)Iowa City VA Health Care System, Iowa City, IA, (3)University of Iowa Hospitals and Clinics, Iowa City, IA, (4)Department of Epidemiology, University of Iowa College of Public Health, Iowa City, IA


    H. Suzuki, None

    E. Perencevich, None

    D. Diekema, None

    D. Livorsi, None

    M. Schweizer, None

    R. Nair, None

    M. Ohl, Gilead Sciences, Inc: Grant Investigator , Research grant .

    K. Richardson, None

    B. Beck, None

    B. Alexander, None

    M. Goto, None

    Findings in the abstracts are embargoed until 12:01 a.m. PDT, Wednesday Oct. 3rd with the exception of research findings presented at the IDWeek press conferences.