Association of compliance-normalized airway and transpulmonary mechanical power with mortality in PARDS.

The prognostic significance of mechanical power (MP) and the optimal normalization strategy in pediatric acute respiratory distress syndrome (PARDS) remains uncertain. We evaluated the association between airway and transpulmonary mechanical power normalized using different strategies and 28-day mortality in children with PARDS. In this multicenter retrospective observational study, 128 invasively ventilated subjects with PARDS were included from four tertiary pediatric intensive care units between June 2020 and December 2025. Airway MP and transpulmonary mechanical power (MPL) were calculated during pressure-controlled ventilation and normalized to predicted body weight (PBW), body surface area (BSA), and static respiratory system compliance (Cstat). The primary outcome was 28-day mortality. Associations were examined using multivariable logistic regression adjusted for age, sex, ARDS severity (PALICC-2), and Pediatric Index of Mortality-3 (PIM-3). Discriminative performance was assessed using receiver operating characteristic analysis with DeLong comparisons, and optimal thresholds were identified using the Youden index. The 28-day mortality rate was 20.3%. Absolute airway MP was not associated with mortality. In adjusted analyses, only compliance-normalized airway MP (MP/Cstat) was associated with mortality (adjusted OR 6.8, 95% CI 1.0-44.8; p = 0.044). Compliance-normalized transpulmonary MP (MPL/Cstat) was also associated with mortality (adjusted OR 30.6, 95% CI 1.5-615.7; p = 0.03). PBW- and BSA-normalized indices were not significant. Compliance-normalized indices showed numerically higher AUC values (MP/Cstat 0.68; MPL/Cstat 0.70), although the differences were not significant by DeLong testing. Mortality increased stepwise across MP/Cstat tertiles (p for trend = 0.01). Among MP components, only the compliance-normalized elastic component was significantly associated with mortality.

conclusionsMechanical power, normalized to static respiratory system compliance rather than anthropometric parameters, was associated with mortality and showed a graded dose-response relationship in PARDS. Compliance-based normalization may provide additional physiological insight into mechanical power assessment in PARDS. WHAT IS KNOWN: • Mechanical power integrates multiple ventilatory variables and has been associated with outcomes in adult ARDS. • Different normalization approaches, including anthropometric and physiologic methods, have been proposed to account for variability in body size and lung mechanics. The clinical relevance of these normalization strategies in PARDS remains uncertain. WHAT IS NEW: • In a multicenter cohort of subjects with PARDS, mechanical power normalized to static respiratory system compliance was independently associated with 28-day mortality for both airway and transpulmonary measurements. In contrast, normalization to predicted body weight or body surface area was not independently associated with mortality after adjustment. • Mortality increased progressively across tertiles of compliance-normalized airway mechanical power. Among mechanical power components, only the compliance-normalized elastic component was independently associated with mortality.