Valvular & structural heart diseaseDiagnostic or prognostic study
Predicting Outcomes in Patients With Tricuspid Regurgitation Undergoing Transcatheter Edge-to-Edge Repair Using an Artificial Intelligence–Derived Risk Score
- Prognostic model development and validation study
- registry-based
- derivation and validation cohorts
- extreme gradient boosting (AI)
Authors' conclusion
The EuroTR score offers an easy-to-use, externally validated, accurate risk stratification tool for patients undergoing T-TEER.
Study context
- Population
- Patients with severe tricuspid regurgitation undergoing T-TEER, EuroTR registry
- Participants
- 1,826 patients
- Comparison
- EuroTR score vs established risk scores
- Derivationn = 1,225
- Validationn = 601
- Outcome
- 1-year mortality
- Follow-up
- 1 year
Performance
1-year mortality prediction by EuroTR score
Hazard ratio
4.26
95% CI 2.71–6.67
P < 0.001
By group
Validation
0.741 (95% CI: 0.699-0.783)
Harrell's C index
High-risk vs Low-risk
HR 4.26 (95% CI 2.71–6.67); P < 0.001
As reported in the abstract
Further findings
Combined endpoint: 1-year mortality, HF hospitalization or NYHA class ≥III dyspnea
1-year survival, entire cohort
- Overall 1-year survival
- 82.1% (95% CI: 80.1%-84.2%)
- No significant differences between derivation and validation cohorts
Subgroups (atrial vs nonatrial TR, TRILUMINATE-eligible vs noneligible, with vs without cardiac device leads)
- EuroTR score performance confirmed in several subgroups
Original publication
Predicting Outcomes in Patients With Tricuspid Regurgitation Undergoing Transcatheter Edge-to-Edge Repair Using an Artificial Intelligence–Derived Risk Score. JACC Cardiovasc Interv. Published 9 March 2026. doi:10.1016/j.jcin.2025.11.034
Summarised from the published abstract; every number is checked against it