Agreement Between Automated and Equation-Derived Mechanical Power in Pediatric Mechanical Ventilation

Background: Newer ventilators provide automated real-time mechanical power (MP auto ) measurements, but their accuracy in children across ventilation modes remains uncertain. We aimed to evaluate the agreement between the MP auto measured on the Mindray SV-600 ventilator and the equation-derived MP (MP calc ) across two tidal volume (V T ) settings, ventilation modes, and predefined age groups in ventilated children. Method: In this prospective observational physiology study, consecutive mechanically ventilated children aged 1 month to 18 years were enrolled between August 2024 and May 2025 in a 30-bed pediatric ICU. MP auto was compared with MP calc , using comprehensive and surrogate equations during continuous mandatory volume-control (VC-CMV) and pressure-control (PC-CMV) modes at V T of 6 and 8 mL/kg body weight. Subgroup analyses considered age and pediatric ARDS. In VC-CMV mode, MP was calculated with Gattinoni’s equation (MPVC comprehensive ) and Giosa’s equation (MPVC surrogate ). In PC-CMV mode, Van der Meijden’s equation (MPPC comprehensive ) and Becher’s equation (MPPC surrogate ) were used. Bland-Altman analyses evaluated agreement. Results: We included 147 children, with a median (interquartile range [IQR]) age of 49 (16–92) months, of whom 34 (23%) had pediatric ARDS. MP auto showed strong agreement with MP calc across ventilation modes and V T settings. The agreement was highest in VC-CMV, with minimal bias (−0.12 to +0.011 J/min; intra-class correlation [ICC] ≥0.95) and >88% of observations within clinical limits of agreement. In PC-CMV, MP auto demonstrated a systematic negative bias relative to MP calc (minimal bias −1.23 to −0.62 J/min; ICC ≥0.92), with similar findings in subjects with pediatric ARDS. Age influenced agreement between MP auto and MP calc (slope −0.008 to −0.005), especially in PC-CMV mode ( R 2 = 0.226–0.27). Pediatric ARDS status affected this relationship only in PC-CMV mode ( P = .02 for MPPC comprehensive and P = .006 for MPPC surrogate ). Conclusion: MP auto demonstrated strong agreement with equation-derived MP during VC-CMV mode across different age groups. However, MP auto showed a systematic negative bias relative to equation-derived MP in PC-CMV mode, particularly among younger children and those with pediatric ARDS.