Elastic Ventilatory Energy Predicts ARDS Severity and Mortality: A Bayesian Retrospective Analysis

Résumé fourni par la source

Background: Mechanical power integrates the energy delivered to the respiratory system during mechanical ventilation but may not reflect the mechanical stress applied to lung tissue. Metrics focused on the elastic component of ventilatory energy may provide a more physiologically relevant estimate of parenchymal stress. We evaluated elastic static power and elastic static power normalized to predicted body weight (PBW) and their association with ARDS severity and mortality Methods: We conducted a single-center retrospective cohort study in an ICU, analyzed using Bayesian modeling. Adult subjects ( N = 137) receiving volume-controlled invasive mechanical ventilation for COVID-19 pneumonia were included, encompassing individuals without ARDS and with mild, moderate, or severe ARDS. Elastic static power and PBW-normalized elastic static power were derived from ventilatory parameters recorded within the first 24 h after initiation of mechanical ventilation. Bayesian regression models were used to evaluate associations between these variables and ARDS severity, ICU mortality, and 28-day mortality. Results: Posterior distributions for elastic load metrics favored greater ARDS severity and increased mortality. Elastic static power favored moderate-to-severe ARDS (relative risk 1.32, 95% credible interval [CrI] 1.10–1.57), ICU mortality (hazard ratio [HR] 1.36, 95% CrI 1.16–1.60), and 28-day mortality (HR: 1.31, 95% CrI: 1.17–1.48). Similar patterns were observed after normalization to PBW. Conclusions: Posterior distributions for elastic ventilatory energy, in both its nonnormalized and PBW-normalized forms, consistently favored greater ARDS severity and mortality. These findings support the use of elastic load metrics as physiologically relevant indicators of ventilatory stress.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.