Acute respiratory distress syndrome (ARDS) complicating severe respiratory syncytial virus (RSV) infection has been described in only a few infants. In contrast to the low mortality rates usually associated with RSV infections (< 5%), mortality rates in the range of 40-70% have been reported in pediatric patients with ARDS. However, studies on patients with ARDS are usually lumped with respect to causation, and the disease course of RSV-induced ARDS has not been previously studied. We examined the pulmonary function abnormalities of 37 infants with RSV-induced respiratory failure who were admitted to our pediatric intensive care unit for assisted ventilation. Measurements included respiratory mechanics, maximum expiratory flow-volume curves, and lung volumes. These allowed the calculation of a Murray lung injury score (modified for pediatric use) in which radiographic findings, ventilator settings, lung compliance, and blood gas results were considered. We identified ten infants with severe restrictive lung disease who fulfilled the clinical criteria for classification as ARDS. All had lung injury scores above 2.5, compatible with a diagnosis of ARDS. Twenty-seven infants had obstructive patterns of lung function consistent with a clinical diagnosis of RSV bronchiolitis. The patients with RSV-induced ARDS were significantly younger, and had a longer time on assisted ventilation (P < 0.05) and a higher proportion of predisposing illnesses (P < 0.05, odds ratio = 6.67, two-tailed Fisher's exact test) when compared with the patients who had obstructive disease. Only one patient (who had immunodeficiency) died, and all others were successfully managed on conventional mechanical ventilation. We conclude that RSV-induced respiratory failure represents a relatively benign cause of ARDS in pediatric patients. Our observations support the notion of differentiating ARDS with respect to causation, especially when novel and experimental therapy is considered and mortality rates are analyzed.