ABSTRACT
INTRODUCTION@#Evidence supporting non-invasive ventilation (NIV) in paediatric acute respiratory distress syndrome (PARDS) remains sparse. We aimed to describe characteristics of patients with PARDS supported with NIV and risk factors for NIV failure.@*MATERIALS AND METHODS@#This is a multicentre retrospective study. Only patients supported on NIV with PARDS were included. Data on epidemiology and clinical outcomes were collected. Primary outcome was NIV failure which was defined as escalation to invasive mechanical ventilation within the first 7 days of PARDS. Patients in the NIV success and failure groups were compared.@*RESULTS@#There were 303 patients with PARDS; 53/303 (17.5%) patients were supported with NIV. The median age was 50.7 (interquartile range: 15.7-111.9) months. The Paediatric Logistic Organ Dysfunction score and oxygen saturation/fraction of inspired oxygen (SF) ratio were 2.0 (1.0-10.0) and 155.0 (119.4- 187.3), respectively. Indications for NIV use were increased work of breathing (26/53 [49.1%]) and hypoxia (22/53 [41.5%]). Overall NIV failure rate was 77.4% (41/53). All patients with sepsis who developed PARDS experienced NIV failure. NIV failure was associated with an increased median paediatric intensive care unit stay (15.0 [9.5-26.5] vs 4.5 [3.0-6.8] days; <0.001) and hospital length of stay (26.0 [17.0-39.0] days vs 10.5 [5.5-22.3] days; = 0.004). Overall mortality rate was 32.1% (17/53).@*CONCLUSION@#The use of NIV in children with PARDS was associated with high failure rate. As such, future studies should examine the optimal selection criteria for NIV use in these children.
ABSTRACT
OBJECTIVES: The Pediatric Acute Lung Injury Consensus Conference developed a pediatric specific definition for acute respiratory distress syndrome (PARDS). In this definition, severity of lung disease is stratified into mild, moderate, and severe groups. We aim to describe the epidemiology of patients with PARDS across Asia and evaluate whether the Pediatric Acute Lung Injury Consensus Conference risk stratification accurately predicts outcome in PARDS. DESIGN: A multicenter, retrospective, descriptive cohort study. SETTING: Ten multidisciplinary PICUs in Asia. PATIENTS: All mechanically ventilated children meeting the Pediatric Acute Lung Injury Consensus Conference criteria for PARDS between 2009 and 2015. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Data on epidemiology, ventilation, adjunct therapies, and clinical outcomes were collected. Patients were followed for 100 days post diagnosis of PARDS. A total of 373 patients were included. There were 89 (23.9%), 149 (39.9%), and 135 (36.2%) patients with mild, moderate, and severe PARDS, respectively. The most common risk factor for PARDS was pneumonia/lower respiratory tract infection (309 [82.8%]). Higher category of severity of PARDS was associated with lower ventilator-free days (22 [17-25], 16 [0-23], 6 [0-19]; p < 0.001 for mild, moderate, and severe, respectively) and PICU free days (19 [11-24], 15 [0-22], 5 [0-20]; p < 0.001 for mild, moderate, and severe, respectively). Overall PICU mortality for PARDS was 113 of 373 (30.3%), and 100-day mortality was 126 of 317 (39.7%). After adjusting for site, presence of comorbidities and severity of illness in the multivariate Cox proportional hazard regression model, patients with moderate (hazard ratio, 1.88 [95% CI, 1.03-3.45]; p = 0.039) and severe PARDS (hazard ratio, 3.18 [95% CI, 1.68, 6.02]; p < 0.001) had higher risk of mortality compared with those with mild PARDS. CONCLUSIONS: Mortality from PARDS is high in Asia. The Pediatric Acute Lung Injury Consensus Conference definition of PARDS is a useful tool for risk stratification.
