Acute Respiratory Distress Syndrome Following Pediatric Trauma: Application of Pediatric Acute Lung Injury Consensus Conference Criteria

Abstract

Objectives: To assess the incidence, severity, and outcomes of pediatric acute respiratory distress syndrome following trauma using Pediatric Acute Lung Injury Consensus Conference criteria.

Design: Retrospective cohort study.

Setting: Level 1 pediatric trauma center.

Patients: Trauma patients less than or equal to 17 years admitted to the ICU from 2009 to 2017.

Interventions: None.

Measurements and main results: We queried electronic health records to identify patients meeting pediatric acute respiratory distress syndrome oxygenation criteria for greater than or equal to 6 hours and determined whether patients met complete pediatric acute respiratory distress syndrome criteria via chart review. We estimated associations between pediatric acute respiratory distress syndrome and outcome using generalized linear Poisson regression adjusted for age, injury mechanism, Injury Severity Score, and serious brain and chest injuries. Of 2,470 critically injured children, 103 (4.2%) met pediatric acute respiratory distress syndrome criteria. Mortality was 34.0% among pediatric acute respiratory distress syndrome patients versus 1.7% among patients without pediatric acute respiratory distress syndrome (adjusted relative risk, 3.7; 95% CI, 2.0-6.9). Mortality was 50.0% for severe pediatric acute respiratory distress syndrome at onset, 33.3% for moderate, and 30.5% for mild. Cause of death was neurologic in 60.0% and multiple organ failure in 34.3% of pediatric acute respiratory distress syndrome nonsurvivors versus neurologic in 85.4% of nonsurvivors without pediatric acute respiratory distress syndrome (p = 0.001). Among survivors, 77.1% of pediatric acute respiratory distress syndrome patients had functional disability at discharge versus 30.7% of patients without pediatric acute respiratory distress syndrome patients (p < 0.001), and only 17.5% of pediatric acute respiratory distress syndrome patients discharged home without ongoing care versus 86.4% of patients without pediatric acute respiratory distress syndrome (adjusted relative risk, 1.5; 1.1-2.1).

Conclusions: Incidence and mortality associated with pediatric acute respiratory distress syndrome following traumatic injury are substantially higher than previously recognized, and pediatric acute respiratory distress syndrome development is associated with high risk of poor outcome even after adjustment for underlying injury type and severity.

Figure 1:. Progression of PARDS severity in the first 48 hours after onset, stratified by severity category at onset.

Patients meeting PARDS criteria while on non-invasive positive pressure ventilation (NIPPV) are combined with patients with mild PARDS (n=2 at PARDS onset; n=3 after extubation from mild or moderate PARDS).

Figure 2:. Mortality by PARDS severity at onset, 6 hours after onset, and peak severity within the first 48 hours after PARDS onset.

Patients meeting PARDS criteria while on non-invasive positive pressure ventilation (NIPPV) are combined with those who met criteria for mild PARDS.

Figure 3:. Duration of mechanical ventilation, ICU length of stay, and hospital length of stay among patients with and without PARDS who survived to hospital discharge.

Box-plots demonstrate median and interquartile range, with whiskers representing upper and lower adjacent values (1.5x the interquartile range). Black diamonds represent the mean number of additional days for PARDS patients compared to non-PARDS patients after adjustment for confounders, with the 95% confidence interval represented in parentheses.