Lung Injury Etiology and Other Factors Influencing the Relationship Between Dead-Space Fraction and Mortality in ARDS

In ARDS, elevated pulmonary dead-space fraction (V /V ) is a particularly strong indicator of mortality risk. Whether the magnitude of V /V is modified by the underlying etiology of ARDS and whether this influences the strength of its association with mortality remains unknown. We sought to elucidate the impact of ARDS etiology on V /V and also to determine whether ARDS severity, as classified by the Berlin definition, has correspondence with changes in V /V . This single-center, retrospective, observational study (2010-2016) measured V /V in 685 subjects with ARDS as part of clinical management with lung-protective ventilation. Volumetric capnography was used to measure V /V with 99% of measurements occurring within 48 h of ARDS onset. Demographic information as well as illness severity scores and pulmonary mechanics data also were collected. Multivariate logistic regression modeling was done to assess the strength of association between V /V and mortality. V /V was elevated across etiologies, with aspiration and pneumonia having significantly higher V /V than non-pulmonary sepsis or trauma. Differences in the magnitude of V /V across etiologies did not necessarily correspond with mortality between etiologies. However, within each etiology grouping, V /V was significantly elevated in non-survivors versus survivors. The same results were found in both moderate and severe (but not mild) ARDS using the Berlin definition. In the final adjusted model, the strongest mortality risk was V /V , wherein the risk of death increased by 22% for every 0.05 increase in V /V . V /V magnitude varies by ARDS etiology, as does mortality. Only in mild ARDS does V /V fail to distinguish non-survivors from survivors. Nonetheless, V /V has the strongest association with mortality risk in those with ARDS.