Trajectory patterns of mechanical power and prognosis in ARDS: a longitudinal analysis using group-based trajectory...

Mechanical power has been identified as a predictor of prognosis in ARDS; however, previous studies based on cross-sectional data may fail to capture the dynamic pathophysiological changes and progression of pulmonary conditions. This study aimed to investigate the association between mechanical power trajectories and 28-day mortality using longitudinal data.

Data were extracted from the Medical Information Mart for Intensive Care IV (MIMIC-IV, version 2.2) database. Mechanical power was divided into quartiles to compare distribution characteristics and temporal trends across groups. Group-based trajectory modeling (GBTM) identified distinct mechanical power trajectories. Multivariate logistic regression analyzed the association between trajectory groups and 28-day mortality.

A total of 1,439 eligible patients were included. Stratification by mechanical power quartiles showed that higher mechanical power was associated with increases in arterial oxygen partial pressure, carbon dioxide, serum creatinine, blood urea nitrogen, potassium, hemoglobin, white blood cell count, respiratory rate, minute ventilation, tidal volume, plateau pressure, positive end-expiratory pressure, FiO₂, and peak airway pressure (all P for trend < 0.001). GBTM identified three trajectories classified as low, medium, and high mechanical power groups. Multivariate logistic regression revealed that compared to the low-power group, the high-power group had a significantly higher risk of 28-day mortality after full adjustment (P = 0.017; OR 1.33; 95% CI 1.05-1.68).

Distinct mechanical power trajectories in ARDS patients were identified and shown to be associated with 28-day mortality, providing practical insights for ventilator management optimization.