Related Experiment Video
Updated: Aug 6, 2026
Surfactant Depletion Combined with Injurious Ventilation Results in a Reproducible Model of the Acute Respiratory Distress Syndrome (ARDS)
Published on: April 7, 2021
Data-driven subphenotyping of severe ARDS patients requiring VV-ECMO
Micha Landoll1,2,3, Stephan Strassmann1, Wolfram Windisch1,4
1Department of Pneumology and Critical Care Medicine, ARDS and ECMO Centre, Cologne-Merheim Hospital, Cologne, Germany.
Identifying distinct acute respiratory distress syndrome (ARDS) subphenotypes in veno-venous extracorporeal membrane oxygenation (VV-ECMO) patients using electronic health records is crucial. Renal, hepatic, and inflammatory dysfunctions significantly impact survival rates in these critical patients.
Area of Science:
- Critical Care Medicine
- Pulmonology
- Biomedical Informatics
Background:
- Acute Respiratory Distress Syndrome (ARDS) is a heterogeneous condition complicating management during Veno-Venous Extracorporeal Membrane Oxygenation (VV-ECMO).
- Risk stratification and personalized treatment are challenging due to ARDS heterogeneity.
- Identifying distinct ARDS subphenotypes is essential for improving patient outcomes.
Purpose of the Study:
- To identify discrete ARDS subphenotypes in VV-ECMO patients.
- To utilize high-resolution electronic health record (EHR) data for subphenotype identification.
- To assess clinical outcome differences between identified ARDS subphenotypes.
Main Methods:
- Analysis of 26 clinical parameters from 598 adult ARDS patients on VV-ECMO.
- K-means clustering applied to EHR data including inflammation, coagulation, kidney/liver function, and mechanical ventilation parameters.
- Shapley Additive Explanations (SHAP) used to identify key differentiating parameters and survival factors.
Main Results:
- Distinct ARDS subphenotypes were identified, primarily driven by inflammation, kidney/liver function, coagulation, and mechanical ventilation parameters.
- Significant variations in survival rates observed between subphenotypes, with kidney/liver dysfunction clusters showing lower survival (32% and 21%).
- Longer intensive care unit (ICU) length of stay correlated with subphenotypes exhibiting multi-organ dysfunction.
Conclusions:
- Data-driven clustering of EHR parameters effectively identifies clinically meaningful ARDS subphenotypes in VV-ECMO patients.
- Renal, hepatic, and inflammatory dysfunctions are critical determinants of survival in this population.
- Subphenotype-based stratification holds promise for refining risk assessment and management strategies for severe ARDS patients on VV-ECMO.
