Retrospective Evaluation of the Association Between Mechanical Power and Outcome in Mechanically Ventilated Dogs from a Multicenter Database (2019–2025): 76 Cases

OBJECTIVE: To assess the association between mechanical power (MP) and survival in mechanically ventilated dogs with primary pulmonary parenchymal disease and to compare its prognostic performance against individual ventilator parameters and oxygenation indices. DESIGN: Retrospective, multicenter observational cohort study. SETTING: Three veterinary medical teaching institutions and one private veterinary referral center. ANIMALS: Seventy-six client-owned dogs mechanically ventilated for primary pulmonary parenchymal disease between March 2019 and March 2025. INTERVENTIONS: None. METHODS: MP (J/min/kg) was calculated using a simplified equation for pressure-controlled (PC) ventilation: 0.098 × respiratory rate × tidal volume × (dynamic driving pressure + positive end-expiratory pressure). Median values were calculated per patient for each of the first 3 ventilation days. Multivariate logistic regression was used to assess associations between MP, its components, and survival. Area under the receiver operating characteristic curve was calculated to evaluate predictive performance. RESULTS: Survivors had significantly lower MP values than nonsurvivors across all 3 initial days of ventilation (Day 1: 0.30 vs. 0.57 J/min/kg; Day 2: 0.33 vs. 0.55 J/min/kg; Day 3: 0.31 vs. 0.47 J/min/kg; p = 0.008 to <0.001). A 0.05-J/min/kg increase in MP was associated with a 19%, 26%, and 34% greater likelihood of mortality on Days 1, 2, and 3, respectively, with good discriminative performance (area under the receiver operating characteristic curve = 0.80, 0.84, and 0.84). Dynamic variables (respiratory rate, dynamic driving pressure, peak inspiratory pressure) were more strongly associated with survival than static variables (tidal volume, positive end-expiratory pressure). CONCLUSIONS: MP is associated with outcome in dogs undergoing mechanical ventilation and demonstrates stronger predictive value than its individual components. These findings align with human data and support considering MP during ventilator management. Further studies are needed to define thresholds and evaluate whether reducing MP improves outcomes.