Objective: Elevated dead space has been consistently associated with increased mortality in adults with respiratory failure. In children, the evidence for this association is more limited. We sought to investigate the association between dead space and mortality in mechanically ventilated children.
Design: Single-center retrospective review.
Setting: Tertiary care pediatric critical care unit.
Patients: Seven hundred twelve mechanically ventilated children with an arterial catheter.
Interventions: None.
Measurements and main results: The end-tidal alveolar dead space fraction ((PaCO2-PETCO2)/PaCO2), a dead space marker, was calculated with each arterial blood gas. The initial end-tidal alveolar dead space fraction (first arterial blood gas after intubation) (per 0.1 unit increase: odds ratio, 1.59; 95% CI, 1.40-1.81) and day 1 mean end-tidal alveolar dead space fraction (odds ratio, 1.95; 95% CI, 1.66-2.30) were associated with mortality. The relationship between both initial and day 1 mean end-tidal alveolar dead space fraction and mortality held in multivariate modeling after controlling for any of the following individually: PaO2/FIO2, oxygenation index, 24-hour maximal inotrope score, and Pediatric Risk of Mortality III (all p<0.01), although end-tidal alveolar dead space fraction was no longer significant after controlling for the combination of oxygenation index, 24-hour maximal inotrope score, and Pediatric Risk of Mortality III. In 217 children with acute hypoxemic respiratory failure, initial end-tidal alveolar dead space fraction (per 0.1 unit increase odds ratio, 1.38; 95% CI, 1.14-1.67) and day 1 mean end-tidal alveolar dead space fraction (per 0.1 unit increase odds ratio, 1.60; 95% CI, 1.27-2.0) were associated with mortality. Day 1 mean end-tidal alveolar dead space fraction remained associated with mortality after controlling individually for any of the following in multivariate models: PaO2/FIO2, oxygenation index, and 24-hour maximal inotrope score (p≤0.02), although end-tidal alveolar dead space fraction was no longer significant after controlling for the combination of oxygenation index, 24-hour maximal inotrope score, and Pediatric Risk of Mortality III.
Conclusions: Increased dead space is associated with higher mortality in critically ill children, although it is no longer independently associated with mortality after controlling for severity of oxygenation defect, inotrope use, and severity of illness. However, because end-tidal alveolar dead space fraction is easy to calculate at the bedside, it may be useful for risk stratification and severity-of-illness scores.