Pediatric ARDS biomarkers: missing the random forest for the trees

Excerpt

Acute respiratory distress syndrome (ARDS) is characterized by acute onset of diffuse bilateral pulmonary edema and severe hypoxemia not fully explained by cardiac dysfunction [1]. Primarily defined for adults, ARDS affects 10% of mechanically ventilated children in pediatric intensive care units (PICUs) [2], with a mortality rate of 20% in modern cohorts [3, 4]. ARDS is heterogeneous, with patients having distinct co-morbidities and inciting etiologies (pneumonia, non-pulmonary sepsis). This heterogeneity has contributed to negative trial results in adults and pediatrics, as therapies effective in some patients are ineffective in others [5]. Methods to reduce heterogeneity including sub-phenotyping using protein and mRNA biomarkers have been proposed for improving patient selection for future clinical trials [6]. Biomarkers have also been proposed to predict development of, accurately diagnose, and prognosticate ARDS. Biomarkers may also provide insight into ARDS pathophysiology, which remains remarkably imprecise despite 50 years of research. …