1257. Assessing risk factors for an outbreak of Burkholderia cenocepacia in non-cystic fibrosis (CF) patients
Session: Poster Abstract Session: Healthcare Epidemiology: Outbreaks
Friday, October 5, 2018
Room: S Poster Hall
  • ACH_burkholderiaIDweekposter_V4_9-13-18_AH.pdf (391.6 kB)
  • Background: Burkholderia spp. have been associated with outbreaks of healthcare associated infections in non-CF patients, mostly attributable to point sources of contaminated solutions or medications. Fewer non-point source outbreaks have been described.

    Methods: We conducted a matched case:control (1:3) study to assess risk factors for B. cenocepacia during an outbreak that occurred in 2017 in a 738-bed university-affiliated hospital involving patients hospitalized on several ICUs and non-ICUs. Clinical isolates identified as B. cepacia complex were speciated using sequencing of the recA allele and genotyped by pulsed field gel electrophoresis (PFGE). Case subjects were patients with a positive culture for the B. cenocepacia outbreak strain (PFGE pattern 17-A, recA 365) from 6/1-12/31/2017. Control subjects had negative respiratory cultures for Burkholderia spp. within 10 days of respective cases’ culture dates and were hospitalized on the same unit at the same time as respective cases. Potential risk factors including procedures, devices, and medications (previously linked to point source outbreaks) were examined. A 5-day exposure window was studied for procedures and devices as this was the shortest interval noted between a case subject’s negative and first positive culture. Exact conditional logistic regression was used to analyze risk factors; Mann Whitney U and Fisher’s exact tests were used to compare demographic and clinic characteristics of case and control subjects.

    Results: Seventeen cases (all with positive respiratory tract cultures) and 41 unit-matched controls were studied. Case and control subjects had similar demographic characteristics, illness severity, and comorbidities. No point source was identified. Only exposure to invasive mechanical ventilation was associated with case status (OR: 10.5, [CI95 1.9, ∞), p=0.0083). Cases had longer hospital lengths of stay (52 v 33 days, p=0.02) than controls, but similar in-hospital mortality (24% v 12%, p=0.43).

    Conclusion: These findings suggest sub-optimal infection prevention and control practices related to respiratory interventions, including cleaning and disinfection of ventilators, may have contributed to the outbreak. Reinforcement of best practices helped reduce transmission of the outbreak clone.

    Alexandra Hill-Ricciuti, MPH1, William Greendyke, MD2,3, Matthew Oberhardt, PhD4, Fann Wu, PhD5, Daniel Green, MD5, Susan Whittier, PhD5, Elaine Larson, RN PhD6, Lisa Saiman, MD, MPH1,2 and E. Yoko Furuya, MD, MS7,8, (1)Pediatrics, Columbia University Medical Center, New York, NY, (2)Infection Prevention and Control, NewYork-Presbyterian Hospital, New York, NY, (3)Department of Medicine, Columbia University Medical Center, New York, NY, (4)Value Institute, NewYork-Presbyterian Hospital, New York, NY, (5)Pathology, Columbia University Medical Center, New York, NY, (6)School of Nursing, Columbia University, New York, NY, (7)Medicine, Columbia University Medical Center, New York, NY, (8)NewYork-Presbyterian Hospital, New York, NY


    A. Hill-Ricciuti, None

    W. Greendyke, None

    M. Oberhardt, None

    F. Wu, None

    D. Green, None

    S. Whittier, None

    E. Larson, None

    L. Saiman, None

    E. Y. Furuya, 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.