[Untitled]

Introduction: Acute respiratory distress syndrome (ARDS) represents the most severe form of acute lung injury. The epidemiology of ARDS in children in Singapore has not been previously described. Our overall objective was to describe the epidemiology and resource utilization in children with ARDS; and to identify risk factors for poor outcomes in these critically ill children. Methods: We reviewed the medical records of patients who were admitted to our hospital with the discharge diagnosis of ARDS between January 2009 and December 2012. We utilized the American-European Consensus Conference's definition for ARDS. Demographic, clinical and radiographic data were extracted through chart review. Pertinent ventilator data and arterial blood gas results were collected on diagnosis, day one, three and seven of ARDS. We considered the requirement of invasive ventilation support as ventilator-free days with the maximum cut-off duration of 28 days. Primary outcome of interest was survival to pediatric intensive care unit (PICU) discharge. We summarized continuous and categorical variables as median with ranges and absolute numbers with percentages respectively. Wilcoxon rank-sum test and Fisher exact test were used for analysis. Results: Twenty children were identified to have the diagnosis of ARDS. Median age of these children was 8.3 years (4.0 days – 16.0 years). Of these children, 12 (60%) had significant comorbidities (three cardiovascular, five hematologic, one metabolic and three genetic syndromes). Median admission PRISM II score was 4.1 (0.7 – 86.3). On diagnosis of ARDS, the median arterial partial pressure of oxygen to fraction of inspired oxygen (PaO2:FiO2) ratio and oxygenation index (OI) were 93.6 (36.2 - 186.7) and 27.4 (9.4 – 101.1) respectively. The most common modes of ventilator strategy on day of diagnosis of ARDS were high frequency ventilation (HFV) [n=7(35%)] and airway pressure release ventilation (APRV) [n =4 (20%)]. The following adjunct therapies were used for ARDS: inhaled nitric oxide [2(10%)], prone position [8(40%)] and steroids [9(45%)]. 17 (85%) patients required inotrope support during their PICU stay. 15 (75%) patients had sepsis/pneumonia as risk factor for ARDS. Eight (40%) patients had evidence of air leaks. Median 28-day ventilation-free duration was 0 (0 – 23) days. PICU mortality from ARDS was 12/20 (60%). There was no difference between the median OI at diagnosis of ARDS between survivors and non-survivors [25.3 (13.9 – 46.4) vs. 34.1 (9.4 – 101.1), p=0.56]. However, there was a statistical difference in OI at 24 hours [21.2 (5.8 – 44.0) vs. 65.0 (5.0 – 131.0), p=0.032], day three [13.6 (5.8 – 34) vs. 61.6 (4.0 – 117.4), p=0.038] and day seven [8.4 (5.4 – 31.8) vs. 49.1 (31.8-57.4), p=0.017] between survivors and non survivors respectively. We did not find any association between lactate levels, need for continuous renal replacement therapy and transfusion requirements with mortality. Conclusions: In our center, we supported more than half of our children with ARDS with open lung ventilation strategies (HFV and APRV). There was no association between OI at diagnosis and mortality from ARDS. However, we did find an association with subsequent OIs and mortality.