Impact and feasibility of an emergency department-based ventilator-associated pneumonia bundle for patients intubated in an academic emergency department.

DeLuca, Lawrence A; Walsh, Paul; Davidson, Donald D; Stoneking, Lisa R; Yang, Laurel M; Grall, Kristi J H; Gonzaga, M Jessica; Larson, Wanda J; Stolz, Uwe; Sabb, Dylan M; Denninghoff, Kurt R

BACKGROUND Ventilator-associated pneumonia (VAP) has been linked to emergency department (ED) intubation and length of stay (LOS). We assessed VAP prevalence in ED intubated patients, feasibility of ED VAP prevention, and effect on VAP rates. METHODS This was a quality improvement initiative using a pre/post design. Phase 1 (PRE1) comprised patients before intensive care unit (ICU) bundle deployment. Phase 2 (PRE2) occurred after ICU but before ED deployment. Phase 3 (POST) included patients received VAP prevention starting at ED intubation. Log-rank test for equality and Cox regression using a Breslow method for ties were performed. Bundle compliance was reported as percentages. Number needed to treat (NNT) was calculated by ventilator day. RESULTS PRE1, PRE2, and POST groups were composed of 195, 192, and 153 patients, respectively, with VAP rates of 22 (11.3%), 11 (5.7%), and 6 (3.9%). Log-rank test showed significant reduction in VAP (χ2 = 9.16, P = .0103). The Cox regression hazard ratio was 1.38 for the Clinical Pulmonary Infection Score (P = .001), and the hazard ratio was 0.26 for the VAP bundle (P = .005). Bundle compliance >50% for head-of-bed elevation, oral care, subglottic suctioning, and titrated sedation improved significantly with introduction of a registered nurse champion. NNT varied from 7 to 11. CONCLUSIONS VAP was common for ED intubated patients. ED-based VAP prevention is feasible. We demonstrate significant reduction in VAP rates, which should be replicated in a multicenter study.