Diagnostic or prognostic study
Artificial intelligence-derived risk score for mortality in secondary mitral regurgitation treated by transcatheter edge-to-edge repair: the EuroSMR risk score
- Risk prediction model development and validation study
- artificial intelligence-derived
- derivation and validation cohorts
Authors' conclusion
The EuroSMR risk score may allow for improved prognostication in heart failure patients with severe SMR, who are considered for a M-TEER procedure.
Study context
- Population
- Patients with severe secondary mitral regurgitation (SMR) undergoing M-TEER, EuroSMR cohort
- Participants
- 4,172 vs 428
- Comparison
- EuroSMR risk score vs established risk models
- Derivationn = 4,172
- Validationn = 428
- Outcome
- Survival or survival + clinical improvement
- Follow-up
- 1 year
Performance
Discrimination of 1-year mortality
Hazard ratio
4.3
95% CI 3.7–5.0
P < 0.001
EuroSMR risk score vs established risk models
HR 4.3 (95% CI 3.7–5.0); P < 0.001
As reported in the abstract
- AUC for 1-year mortality
- 0.789 (95% CI 0.737-0.842)
- Predicted 1-year mortality range
- <5% to >70%
Further findings
Extreme-risk population (2.6% of entire cohort), not surviving beyond 1 year
| Comparison | HR (95% CI) | P value | |
|---|---|---|---|
| Extreme-risk population | 6.5 (3.0–14) | P < 0.001 | |
| 0.050.21520Hazard ratio (log scale) |
Top 5% EuroSMR risk scores: event rates · 1-year follow-up
- Mortality
- 72.7%
- Mortality or lack of clinical improvement
- 83.2%
Original publication
Artificial intelligence-derived risk score for mortality in secondary mitral regurgitation treated by transcatheter edge-to-edge repair: the EuroSMR risk score. Eur Heart J. Published 19 January 2024. doi:10.1093/eurheartj/ehad871
Summarised from the published abstract; every number is checked against it