Risk factors for mortality in children with pneumonia admitted to the pediatric intensive care unit

Aims: To describe the epidemiology of children with severe pneumonia and identify risk factors for poor outcomes. Methods: We conducted a retrospective study of children admitted to pediatric intensive care unit (PICU) from 2010 to 2014 with a diagnosis of pneumonia. Clinical microbiological, ventilation and other pertinent PICU data were collected. Primary outcome was PICU mortality. Univariate and multivariate logistic regression model were used to identify risk factors for mortality. Results: Severe pneumonia consisted of 237/3539 (6.7%) of PICU admissions. Of these, 162/237 (68.4%) required mechanical ventilation. 32/237 (13.5%) patients died. The majority of patients had no organisms identified 82/237 (34.6%). A sole bacterial or viral pathogen was identified in 48/237 (20.1%) and 41/237 (17.9%) patients, respectively. Patients with viral pneumonias were more likely to develop acute respiratory distress syndrome compared to other etiologies (7/41 [17.1%] vs 8/196 [4.0%]; P = 0.006). Bacterial pneumonias were associated with lung abscess (4/ 48 [8.3%] vs 2/189 [1.5%]; P = 0.016) and necrotizing pneumonia (18/48 [37.5%] vs 15/189 [7.9%]; P < 0.001) compared to other etiologies. Co-detections (> 1 respiratory pathogens isolated) occurred in 62/237 (26.2%) patients and were associated a higher rate of mechanical ventilation, and decreased ventilator and PICU free days. After adjusting for severity of illness, risk factors formortality were: hospital acquired pneumonia (HAP) (aOR: 2.92 [95% CI 1.15, 7.40]; P = 0.024) and bacteremia (aOR: 5.03 [95% CI 1.77, 14.35]; P = 0.003). Conclusions: Severe pediatric pneumonia accounts for a significant number of PICU admissions and is associated with significant mortality risk. The presence of co-morbidities, HAP and bacteremia were early prognostic variables independently associated with poor clinical outcomes.