Respiratory failure & ventilation
Subphenotypes of mechanically ventilated acute respiratory distress syndrome patients based on multi-dimensional pathophysiological parameters
Primary result · 28-day mortalityResult — go to the resultAdjusted hazard ratio 2.263 95% CI 1.206–4.245Authors' conclusion
Using unsupervised machine learning, this study identified two distinct ARDS subphenotypes characterized by divergent profiles in pulmonary edema, respiratory mechanics, and gas exchange. These subphenotypes were associated with significantly different clinical outcomes and exhibited a differential response to prone positioning therapy.
Study context
- Population
- Mechanically ventilated patients with acute respiratory distress syndrome (ARDS)derivation cohort
- Participants
- 111 participants
- Comparison
- Subphenotype 2 vs Subphenotype 1
- Primary endpoint
- 28-day mortality
- Follow-up
- 28 days
Methodological detailsunsupervised K-means clustering, derivation and independent validation cohorts
- Unsupervised K-means clustering
- Derivation and independent validation cohorts
Primary result
As reported in the abstract
Adjusted hazard ratio
2.263
95% CI 1.206–4.245
P = 0.011
By group
Subphenotype 2 vs Subphenotype 1
aHR 2.263 (95% CI 1.206–4.245); P = 0.011
Further findings
Interaction between subphenotype and prone positioning response for 28-day mortality
- Significant interaction (p-for-interaction)
- 0.015
Validation cohort (n = 55)
- Prognostic distinction and differential prone positioning response consistently replicated
Original publication
Subphenotypes of mechanically ventilated acute respiratory distress syndrome patients based on multi-dimensional pathophysiological parameters. Crit Care. Published 27 November 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →