1067. Implementation of New Ventilator-associated Event Surveillance in a Community-based Academic Healthcare Center
Session: Poster Abstract Session: Surveillance of HAIs: Evaluating National Strategy
Friday, October 4, 2013
Room: The Moscone Center: Poster Hall C
  • 2013 ID week VAE poster.pdf (197.4 kB)
  • Background:   The National Healthcare Safety Network (NHSN) launched new surveillance definitions for ventilator-associated events (VAE) in Jan 2013. The correlation between these surveillance definitions and clinical syndromes is unclear.

    Methods:   Our 2-hospital, 1100-bed community-based academic healthcare system began VAE surveillance in 4 adult ICUs in Jan 2013 using NHSN definitions.  The laboratory reports semi-quantitative results for Gram stains: only “numerous” polymorphonuclear leukocytes (PMNs) meets the NHSN definition of ≥ 25 per high power field.  Patients defined as ventilator-associated complication (VAC), infection-related VAC (IVAC), and possible or probable ventilator-associated pneumonia (VAP) during Jan-Mar 2013 underwent chart review to assess clinical documentation and specific reasons why definitions were not met.  Administrative data were obtained to identify patients coded as VAP (ICD-9 code 997.31).

    Results:   During Jan-Mar 2013, 66 patients met surveillance definitions, including 41 VACs, 15 IVACs, 10 possible and 0 probable VAPs.  During all of 2012, 27 VAPs were identified in the same ICUs using the prior definitions.  Of all VAEs, discharge summaries documented only 1 (1.5%) as “VAP” and 31 (47%) as other types of pneumonia.  Of the 41 VACs, 14 (34%) did not meet IVAC criteria because of normal temperature and white blood cell count; 11 of these were not given antibiotics (ABx).  The remaining 27 VACs had either no new ABx started (14 [52%]), or ABx were started but continued < 4 days (13 [48%], including 2 patients who died <4 days from ABx start).  Of those who met IVAC criteria but were not possible VAP, Gram stain/cultures were negative in 9 (60%), positive but outside the window period in 3 (20%), an excluded organism in 2 (14%), and not done in 2 (14%).  Of the possible VAPs, the majority (80%) did not meet the probable VAP definition because they had moderate/many PMNs rather than “numerous.”  Only 3 patients had a VAP diagnosis code, 2 of whom had been identified as VAC, and 1 with no VAE.

    Conclusion:   In the first 3 months of VAE surveillance, our institution experienced a high rate of VACs, particularly compared to the prior VAP definition.  Half were identified as any type of pneumonia by clinicians. No probable VAP were identified, primarily because of the strict definition for purulent sputum.

    Marci Drees, MD, MS1,2, Benjamin Silverman, MD1, Nora Protokowicz, MSN, RN, NE-BC1, Lorraine Adkins, BSN, RN, CIC1, Leslie Freeman, RN, CIC1, Carol Briody, MT (ASCP), CIC1, Kathleen Wroten, BSN, RN, CIC1 and Michael Depietro, MD1, (1)Christiana Care Health System, Newark, DE, (2)Medicine, Jefferson Medical College, Thomas Jefferson University, Philadelphia, PA


    M. Drees, None

    B. Silverman, None

    N. Protokowicz, None

    L. Adkins, None

    L. Freeman, None

    C. Briody, None

    K. Wroten, None

    M. Depietro, None

    Findings in the abstracts are embargoed until 12:01 a.m. PST, Oct. 2nd with the exception of research findings presented at the IDWeek press conferences.