Incidence and risk factors of ventilator-associated pneumonia: a one-year prospective survey

Objective: To determine the incidence of ventilator-associated pneumonia (VAP) and to analyse predisposing factors. Design: A one-year prospective cohort survey. Setting: An eight-bed medical intensive care unit (ICU) of a teaching hospital located in Tunisia. Subjects: Seventy-three consecutive patients, mechanically ventilated for a period exceeding 48 hours. Interventions: None. Measurements: All included patients were prospectively followed in order to diagnose pneumonia on the basis of culture of at least one bacterial species at a significant concentration (>103 CFU/ml) from specimens obtained with a protected telescoping catheter. The crude incidence was estimated and predisposing factors analysed by multivariate analysis. Results: Ventilator-associated pneumonia developed in 38% of mechanically ventilated patients in our ICU, yielding an incidence of 46 episodes per 1,000 patient-ventilator-days. From the multivariate analysis, factors significantly associated with an increased risk for developing nosocomial pneumonia (NP) were: multiple intubations during mechanical ventilation (odds ratio [OR], 10.92; 95% confidence interval [CI] 2.5-46.7), enteral feeding (OR, 40.3; CI, 3.3-294.8) and depressed level of consciousness (OR, 40.3; 95% CI 3.3-423.1). Overall mortality rate in infected patients was 55% and was not different from that observed in non-infected patients. Conclusion: The incidence of VAP in our ICU is high. Because some of the factors identified as influencing the risk of NP in ventilated patients could be avoided by medical interventions, a preventative strategy taking into account these identified risk factors might lower the incidence of VAP in our ICU.