Risk stratification is essential for selecting patients with severe tricuspid regurgitation undergoing tricuspid valve transcatheter edge-to-edge repair (T-TEER). A study published in JACC: Cardiovascular Interventions developed and externally validated an artificial intelligence (AI)–based risk model, the EuroTR score, to estimate 1-year mortality after T-TEER.

The EuroTR score was derived from the European Registry of Transcatheter Repair for Tricuspid Regurgitation, including 1,225 patients in the derivation cohort and 601 in the validation cohort (N = 1,826). The model incorporated 18 clinical, laboratory, echocardiographic, and hemodynamic variables and was generated using an extreme gradient boosting algorithm, with performance evaluated against established risk scores.

Across the study population, 1-year survival was 82.1% (95% CI 80.1–84.2), with similar results in derivation and validation cohorts. The EuroTR score stratified patients into low- and high-risk groups for 1-year mortality (HR 4.26; 95% CI 2.71–6.67; P < 0.001), with discrimination in the validation cohort showing a Harrell’s C-index of 0.741 (95% CI 0.699–0.783).

Higher EuroTR scores were also associated with increased likelihood of a combined endpoint of 1-year mortality, heart failure hospitalization, or persistent dyspnea corresponding to New York Heart Association (NYHA) functional class ≥III. The probability of this endpoint increased from 30.6% in patients with EuroTR risk rank <5% to 85.5% in those with risk rank ≥95%.

Model performance remained consistent across subgroups, including atrial vs nonatrial tricuspid regurgitation, TRILUMINATE-eligible vs noneligible patients, and individuals with or without cardiac implantable electronic device (CIED) leads. Overall, the EuroTR score provides an externally validated tool for risk stratification in patients undergoing T-TEER.