SESSION TITLE: Outcomes/Quality Control Posters
SESSION TYPE: Original Investigation Poster
PRESENTED ON: Wednesday, October 30, 2013 at 01:30 PM - 02:30 PM
PURPOSE: Identifying a valid and reproducible surveillance method for ventilator-associated pneumonia (VAP) has been difficult. Previous definitions lacked accuracy, objectivity, and consistency. In 2011, the CDC organized a working group to create a new surveillance algorithm for VAEs. The aim of this analysis is to prospectively determine the frequency of VAEs, the corresponding etiologies, and the preventability of each event in a mixed medical and surgical ICU population.
METHODS: All patients admitted to two medical ICUs (MICUs) and one surgical ICU (SICU) who are ventilated for ≥ 2 calendar days (CDs) are screened for VAEs. A ventilator-associated condition (VAC) is defined as an increase in the minimum daily FiO2 by ≥ 0.2 or an increase in the minimum daily PEEP by ≥ 3 cmH2O that persists for ≥ 2 CDs after a period of baseline stability. A VAC can be further characterized as an infection-related VAC (IVAC), possible VAP, and probable VAP based on clinical, laboratory, and microbiological criteria. Preventability is determined by the research and ICU team and a third party if there is disagreement.
RESULTS: In the first two months of an ongoing prospective study, 252 patients were ventilated for ≥ 2 CDs totaling 1610 ventilator days. 13 VACs were identified (5%) at a rate of 8 per 1000 ventilator days. 6 IVACs were identified at a rate of 4 per 1000 ventilator days including 2 possible VAPs and 3 probable VAPs. Other etiologies included pulmonary edema (2), acute lung injury (2), untreated pneumonia (1), and a combination of atelectasis and pulmonary edema (2). 3 (23%) VACs were thought to be preventable. 7 (54%) patients died.
CONCLUSIONS: In an ongoing prospective analysis, 5% of medical and surgical ICU patients developed a VAE. While 6 VAEs were IVACs, more than half were secondary to alternative etiologies considered non-preventable.
CLINICAL IMPLICATIONS: Hopefully, the findings from this study will assist in determining the accuracy of the NHSN surveillance algorithm to identify clinically meaningful and potentially preventable events.
DISCLOSURE: The following authors have nothing to disclose: Anthony Boyer, Marin Kollef, Hilary Babcock
No Product/Research Disclosure Information