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

Summary

Using MIMIC-IV longitudinal data and group-based trajectory modeling, three mechanical power trajectories (low, medium, high) were identified in ARDS. High mechanical power trajectory independently associated with higher 28-day mortality (adjusted OR 1.33), offering a dynamic risk metric beyond single timepoints.

Key Findings

• Identified three distinct mechanical power trajectories (low, medium, high) in 1,439 ARDS patients.
• High mechanical power trajectory was associated with increased 28-day mortality (adjusted OR 1.33; 95% CI 1.05–1.68; P=0.017).
• Higher mechanical power correlated with increased PaO2, PaCO2, ventilation parameters (VE, VT, Pplat, PEEP, FiO2, peak pressure) and labs (WBC, creatinine, BUN).

Clinical Implications

Monitoring and minimizing mechanical power over time, not just at single snapshots, may reduce mortality risk; trajectory classification could inform alerts and ventilator titration.

Limitations

• Retrospective single-database analysis with potential residual confounding
• Effect size is modest and physiologic drivers (e.g., effort, esophageal pressure) were not directly measured

Future Directions

Prospective trials testing ventilator strategies to lower mechanical power trajectories; integration into real-time decision support to maintain patients on low-power paths.