Mechanical Power and its Components vs Driving Pressure for Predicting Mortality in Acute Respiratory Distress Syndrome: A Prospective Observational Study.

Aim: Mechanical power (MP) has been proposed as a predictor of acute respiratory distress syndrome (ARDS) mortality, but evidence remains conflicting. We aimed to study its prognostic utility in predicting mortality in ARDS. Patients and methods: This was a single-center prospective observational study including 137 ARDS patients. The organ dysfunction scores, MP and its components (elastic static, elastic dynamic, and resistive power), driving pressure (DP), severity of acute kidney injury (AKI), lung ultrasound scores, pulmonary artery hypertension, days of intensive care unit (ICU) stay, and mortality outcomes were noted. Results: Out of 137 ARDS patients, there were 73 (53.3%) non-survivors. Mechanical power was significantly higher with median [interquartile range (IQR)] 29 (24.55-32) J/min in the mortality group and 24 (20-28.75) J/min in the survival group (p-value < 0.001, Mann-Whitney U-test). However, MP was not an independent predictor of mortality. Driving pressure is an independent predictor of mortality with DP >16 cm H2O, hazard ratio (HR) for mortality 2.925 [(95% confidence interval (CI) 1.778-4.810), Cox-proportional hazard p-value < 0.001]. Out of the three components of MP, dynamic elastic power component is the main contributor to high MP (=29 J/min). Conclusion: Although MP is significantly higher in ARDS non-survivors as compared to survivors, adjustments for confounders showed that it is not an independent predictor of mortality. Driving pressure is an independent predictor of mortality. Elastic dynamic power is the most important component of high MP. How to cite this article: Medhi PP, Chaudhuri S, Parampalli V, Devadiga S, Mareguddi AB, Karanth S, et al. Mechanical Power and its Components vs Driving Pressure for Predicting Mortality in Acute Respiratory Distress Syndrome: A Prospective Observational Study. Indian J Crit Care Med 2025;29(10):815-822.