Efficacy of a Pneumonia Prevention Protocol in the Reduction of Ventilator-Associated Pneumonia in Trauma Patients

We hypothesized that implementing evidence-based ventilator-associated pneumonia (VAP) prevention (VAPP) strategies would decrease the incidence of VAP, and that VAP affects patient outcomes.A prospective study was performed with 331 consecutive ventilated trauma patients in a level one university teaching hospital. The VAPP protocol was modified to include elevation of the head of the bed more than 30 degrees , twice-daily chlorhexidine oral cleansing, a once-daily respiratory therapy-driven weaning attempt, and conversion from a nasogastric to an orogastric tube whenever possible. Ventilator days were compared with occurrences of nosocomial pneumonia, as defined by the U.S. Centers for Disease Control National Nosocomial Infection Surveillance criteria. Patients with and without VAP were compared to discern the effect VAP has on outcome.In 2003, there were 1,600 days of ventilator support with 11 occurrences of VAP (6.9/1,000 ventilator days). In 2004, there were two occurrences of VAP in 703 days of ventilation (2.8/1,000 ventilator days). In the analysis of outcomes of the patients with and without VAP, there was a statistically significant difference in total hospital days (38.7 +/- 26.2 vs. 13.3 +/- 15.5), ICU days (27.8 +/- 12.6 vs. 7.5 +/- 9.7), ventilator days (21.1 +/- 9.8 vs. 6.0 +/- 10.3), Functional Independence Measures (7.25 +/- 2.3 vs. 10.8 +/- 1.8), and hospital charges ($371,416.70 +/- $227,774.31 vs. $138,317.39 +/- $208,346.64)(p0.05 for all). The mortality rate did not decrease significantly (20% vs. 7.5%; p = NS). The difference in the mean Injury Severity Score in the two groups was not significant (21.9 +/- 9.6 vs. 16.7 +/- 11.4 points) and thus could not account for the differences in outcomes.These data suggest that a VAPP protocol may reduce VAP in trauma patients. Ventilator-associated pneumonia may result in more hospital, ICU, and ventilator days and higher patient charges.