Association of the Estimated Dead Space Fraction and the Ventilatory Ratio with Mortality in Patients with Acute Respiratory Distress Syndrome

Introduction: Estimated methods for impaired ventilation such the estimated dead space fraction calculation and the ventilatory ratio (VR) are feasible to do at the bedside and correlate well with direct dead space fraction measurements.

Hypothesis: We hypothesize that estimations of ventilation impairment at the onset of ARDS have independent predictive validity for 30-day mortality. The second hypothesis was that the predictive validity increases when these estimations are redone after 24 hours.

Objective: To compare the methods for dead space estimation measurements through different equations, and the VR in patients with ARDS, and to determine their prognostic capacities for death at day 30.

Material and Methods: Secondary analysis of 637 patients with ARDS from the “Molecular Diagnosis and Risk Stratification of Sepsis” (MARS) project in mixed ICUs of two teaching hospitals. Different formulas were used to calculate dead space fraction (unadjusted Harris-Benedict VD/VTHB, Penn State VD/VTPS, and direct estimate VD/VTdirect and VR.

Results: VD/VT estimations and VR was significantly higher among non-survivors for all estimating equations at both days 1 and 2. Each estimate of dead-space fraction and VR were statistically independent predictors of death at 30 days. The addition of each one of the predictors improved the predictive accuracy of the baseline model.

Conclusion: Estimations of impaired ventilation are independently associated with mortality at 3o days in the early course of ARDS and added predictive accuracy to indexes of oxygenation and respiratory system mechanics.