Acute cardiac care & resuscitation
Enhanced Early Risk Stratification in Cardiogenic Shock Using Machine Learning
- multicenter retrospective registry
- hierarchical agglomerative clustering
+1 more
- hierarchical agglomerative clustering
Authors' conclusion
Hierarchical clustering reproduced previously identified CS phenotypes that were consistent across the first 3 days of ICU stay, providing an enhanced framework for early risk stratification. Prospective validation is required for clinical implementation.
Study context
- Population
- Adults with cardiogenic shock (CS) admitted to intensive care unit (ICUs) between 2016 and 2022adults with CS
- Participants
- 2,906 adults
- Exposure
- Machine learning (ML)-derived CS phenotypes across ICU days 1 to 3 vs SCAI staging for mortality stratification
Main finding
As reported in the abstract
Mortality by day 1 phenotype · day 1
Observational data: associations, not effects.
Further findings
Phenotype persistence · day 2 and day 3
- Similar phenotypes persisted on day 2 and day 3
Mortality stratification within each SCAI stage · days 1 to 3
- ML-derived phenotypes enhanced mortality stratification within each SCAI stage
Original publication
Enhanced Early Risk Stratification in Cardiogenic Shock Using Machine Learning. JACC Adv. Published 13 October 2026. doi:10.
Summarised from the published abstract; every number is checked against it