The COPD-augmented R₂CHA₂DS₂-VA score: development and internal assessment for predicting one-year outcomes after transcatheter aortic valve implantation.

BACKGROUND: Existing CHA₂DS₂-VASc-based scores incompletely capture the prognostic impact of chronic obstructive pulmonary disease (COPD) and of graded renal dysfunction in patients undergoing transcatheter aortic valve implantation (TAVI). We acknowledge that the score's development in a high-risk, high-comorbidity cohort may produce optimistic performance estimates that require attenuation through external validation. We aimed to develop a novel COPD-augmented R₂CHA₂DS₂-VA (COPD-R₂CHA₂DS₂-VA) score that systematically integrates quantitative renal function and COPD, and to internally assess its discriminative performance for one-year all-cause mortality and major adverse cardiovascular events (MACE).

METHODS: In this single-center, retrospective cohort study, 622 consecutive patients undergoing TAVI for severe aortic stenosis (2018-2024) were analyzed. The COPD-R₂CHA₂DS₂-VA score was constructed as a prespecified clinical tool by augmenting the R₂CHA₂DS₂-VA framework with COPD (1 point) and refining renal dysfunction into two graded estimated glomerular filtration rate strata (eGFR 30-59 mL/min/1.73 m²: 1 point; eGFR < 30: 2 points). Discriminative performance was evaluated using receiver operating characteristic curve analysis with bootstrap internal validation (2,000 resamples) and compared with established CHA₂DS₂-VASc-based scores. Multivariable logistic regression assessed the independent prognostic value of the score treated as a continuous variable.

RESULTS: The COPD-R₂CHA₂DS₂-VA score demonstrated encouraging discriminative ability for one-year all-cause mortality (apparent area under the curve [AUC] 0.946, 95% CI 0.926-0.971). As a prespecified tool with fixed component weights, bootstrap internal validation confirmed negligible optimism (mean optimism - 0.00014), with bias-corrected AUC identical to the apparent value. The addition of COPD significantly improved mortality discrimination over the R₂CHA₂DS₂-VA score (ΔAUC + 0.033, p = 0.0001). For one-year MACE, the apparent AUC was 0.806 (95% CI 0.765-0.852); the addition of COPD did not significantly enhance MACE discrimination (ΔAUC + 0.001, p = 0.853). In multivariable analysis, each one‑point increase in the COPD-R₂CHA₂DS₂-VA score was independently associated with 3.52-fold higher odds of one-year mortality (95% CI 2.45-5.06, p < 0.001) and 2.02-fold higher odds of MACE (95% CI 1.67-2.46, p < 0.001). The bias-corrected AUC of the full multivariable model was 0.973 (95% CI 0.962-0.984) for mortality and 0.795 (95% CI 0.752-0.841) for MACE.

CONCLUSIONS: The COPD-R₂CHA₂DS₂-VA score demonstrated encouraging discriminative ability for one-year all-cause mortality after TAVI in this single-center development cohort (apparent AUC 0.946). However, this high discriminative performance is likely influenced by the elevated comorbidity burden in our cohort and the single-center design, and should be interpreted as an upper-bound estimate. The addition of COPD provides a statistically significant but clinically modest incremental improvement for mortality (ΔAUC + 0.033, p = 0.0001), with no significant enhancement for MACE (ΔAUC + 0.001, p = 0.853). These findings are preliminary and hypothesis-generating; urgent external validation in diverse, prospective cohorts is essential before any clinical application can be considered. At present, this score should be regarded as a development-phase, hypothesis-generating tool that complements rather than replaces guideline-recommended instruments such as STS-PROM and EuroSCORE II. The score's clinical utility remains unproven.