From the Authors:

We thank Dr. Mathur and Mr. Bullock for their interest in our study of outcome in pediatric acute hypoxemic respiratory failure (AHRF) (1). Ventilation strategy in these patients aimed at minimizing peak airway pressures and oxygen toxicity, allowing for low tidal volumes, permissive hypercapnia, and lower oxygen partial pressures (2). No bedside compliance measurements were made; instead, positive end-expiratory pressures were continuously adjusted to optimize tidal volumes, with the aim of reducing peak airway pressures and fraction of inspired oxygen. As stated, the institution of high-frequency oscillatory ventilation (HFOV) was done at the attending physician's discretion, and four patients dying on the ventilator succumbed while on HFOV.

The pediatric risk of mortality (PRISM) score assessed within the 12 hours of mechanical ventilation was a significant predictor of outcome, its mean differing between survivors and nonsurvivors (16 ± 7 vs. 22 ± 11, p < 0.001). As expected, no clear threshold could be identified that discriminated between survival and death, reflecting the heterogeneous pattern as to how pediatric AHRF evolves over time. Our study focused on the impact of the severity of oxygenation failure on outcome; this proved to independently affect prognosis. As discussed in the article, multiorgan failure certainly played an important role both in those dying with Pa_O2 equal to or less than 50 mm Hg and those with Pa_O2 values greater than 50 mm Hg. Ventilatory parameters over the last 24 hours before death were not statistically different between these two groups, leading us to state that any threshold of Pa_O2 incompatible with survival is arbitrary and likely differs independently of overall organ integrity. The pediatric logistic organ dysfunction (PELOD) score was not validated until 2003 (3), when data collection for this study had been completed. Improved insight into the role of multiorgan dysfunction might be expected if such a scoring is being performed serially during the entire observational period.