Risk Factors for Ventilator-Associated Events in a PICU*

OBJECTIVE: The term ventilator-associated events includes ventilator-associated condition, infection-related ventilator-associated complication, and ventilator-associated pneumonia. We sought to identify potential new risk factors for ventilator-associated condition and infection-related ventilator-associated complication in the PICU population. DESIGN: Matched case control study. SETTING: Children's hospital at a tertiary care academic medical center. PATIENTS: During the study period, 606 patients were admitted to PICU and ventilated more than 48 hours; 70 children met ventilator-associated condition criteria. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: We applied the definition for ventilator-associated condition (i.e., a sustained increase in ventilator settings after a period of stable or decreasing support) to our database. Within ventilator-associated condition cases, 40 cases were infection-related ventilator-associated complication and 30 cases were noninfectious-related ventilator-associated condition. We identified 140 controls and matched to ventilator-associated condition cases with regard to age, immunocompromised status, and ventilator days to event. Patients with ventilator-associated condition had longer ICU stay versus controls; 24 days median (12-43 interquartile range) versus 7 days (4-14); (p < 0.01), respectively, and longer duration of ventilatory support 17 days (10-32) versus 6 days (3-10); p < 0.01, respectively. Mortality was 22.8% in the ventilator-associated condition versus 9% in the control group (p < 0.01). A multivariate regression analysis adjusted for Pediatric Index of Mortality 2 identified mean peak inspiratory pressure and acute kidney injury to be associated with ventilator-associated condition (odds ratio, 1.12 [95% CI, 1.02-1.22] and odds ratio, 2.85 [1.43-5.66], respectively). Acute kidney injury and neuromuscular blockade in a multivariate regression analysis adjusted for Pediatric Index of Mortality 2 were associated with infection-related ventilator-associated complication (odds ratio, 2.36 [1.03-5.40] and 3.19 [1.17-8.68], respectively). CONCLUSIONS: There is an association between ventilator-associated condition and infection-related ventilator-associated complication in critically ill children with acute kidney injury, ventilatory support, and neuromuscular blockade. Attention should be given by clinical practitioners to recognize these modifiable risk factors and to implement strategies to decrease the prevalence of ventilator-associated events.