A Prospective Study of Ventilator-Associated Pneumonia in Children

Abstract

OBJECTIVE. We conducted a prospective, observational study in a tertiary care pediatric center to determine risk factors for the development of and outcomes from ventilator-associated pneumonia. METHODS. From November 2004 to June 2005, all NICU and PICU patients mechanically ventilated for >24 hours were eligible for enrollment after parental consent. The primary outcome measure was the development of ventilator-associated pneumonia, which was defined by both Centers for Disease Control and Prevention/National Nosocomial Infections Surveillance criteria and clinician diagnosis. Secondary outcome measures were length of mechanical ventilation, hospital and ICU length of stay, hospital cost, and death. RESULTS. Fifty-eight patients were enrolled. The median age was 6 months, and 57% were boys. The most common ventilator-associated pneumonia organisms identified were Gram-negative bacteria (42%), Staphylococcus aureus (22%), and Haemophilus influenzae (11%). On multivariate analysis, female gender, postsurgical admission diagnosis, presence of enteral feeds, and use of narcotic medications were associated with ventilator-associated pneumonia. Patients with ventilator-associated pneumonia had greater need for mechanical ventilation (12 vs 22 median ventilator-free days), longer ICU length of stay (6 vs 13 median ICU-free days), higher total median hospital costs ($308534 vs $252652), and increased absolute hospital mortality (10.5% vs 2.4%) than those without ventilator-associated pneumonia. CONCLUSIONS. In mechanically ventilated, critically ill children, those with ventilator-associated pneumonia had a prolonged need for mechanical ventilation, a longer ICU stay, and a higher mortality rate. Female gender, postsurgical diagnosis, the use of narcotics, and the use of enteral feeds were associated with an increased risk of developing ventilator-associated pneumonia in these patients.