1414: RISK FACTORS ASSOCIATED WITH VENTILATOR-ASSOCIATED PNEUMONIA IN PEDIATRIC TRAUMATIC BRAIN INJURY

Learning Objectives: Mechanical ventilation in patients who suffer a traumatic brain injury is known to increase the incidence of pneumonia when the course of mechanical ventilatory support is prolonged. The high incidence of this serious extracranial complication warrants the need to determine risk factors for VAP which place patients at increased risk. Most available literature about VAPs in TBI patients is in the adult population, or focuses more broadly on pediatric VAP but does not specifically investigate the pediatric TBI population. The purpose of this study was to determine modifiable and nonmodifiable risk factors for development of VAP in pediatric TBI population. Secondary purpose was to evaluate if patients in which a VAP bundle was completed with good compliance had a decreased rate of VAP. Methods: The patient population was derived from the Nemours Trauma Database from December 2007 through July 2017. 63 patients met criteria of pediatric intensive care unit admission and intubation. VAP was defined using the 2016 National Healthcare Safety Network pedVAP criteria. Patients were selected for VAP surveillance when ventilated via endotracheal tube for greater than two calendar days. Clinically defined pneumonia was determined by review of ventilator settings (daily minimum PEEP and FiO2), microbiology cultures from respiratory secretions, chest radiographs, fever assessments, and lung sounds. Results: Of 63 mechanically ventilated children, 16 developed VAP (25.4%). Length of hospital stay (95% CI -1.11-0.24, p=0.003), ICU days (95% CI -1.00-0.29, p=0.0007), and prolonged ventilator days (95% CI -1.06-0.28, p=0.001) were predictive of VAP. Higher index of severity score was predictive of VAP (95% CI -16.92-2.38, p=0.011). There was no indication of relationship between those who received good VAP bundle compliance (p=1). More cases of VAP were seen in those who received oral care, though oral care as a risk factor itself was not significant (p=0.107). Older children with larger endotracheal tubes had more cases of VAP though not statistically significant (p=0.068). Conclusions: Longer ventilation days, length of stay, ICU days, and injury severity appear to be the features most important in predicting VAP. Patients receiving VAP bundle compliance were not at a decreased risk of VAP.