Enhancing Risk Prediction After TAVI: The COPD-Augmented R₂CHA₂DS₂-VA Score and Its Prognostic Value

Highlight

– Development of the COPD-R9CHA2DS2-VA score integrating chronic obstructive pulmonary disease (COPD) and graded renal dysfunction for TAVI patients.
– Excellent discriminative accuracy for predicting one-year all-cause mortality with an internal bias-corrected AUC of 0.946.
– COPD’s addition significantly improved mortality prediction but not major adverse cardiovascular events (MACE).
– The tool requires external validation in diverse populations before routine clinical application.

Study Background

Transcatheter aortic valve implantation (TAVI) represents a transformative minimally invasive treatment for severe aortic stenosis, particularly valuable in patients at elevated surgical risk. Risk stratification in TAVI patients remains essential for prognostication and tailoring management strategies. Commonly used risk scores such as CHA2DS2-VASc, STS-PROM, and EuroSCORE II capture key comorbidities but inadequately quantify the prognostic impact of chronic obstructive pulmonary disease (COPD) and nuanced renal dysfunction. These comorbidities are prevalent and significantly influence post-TAVI outcomes, underscoring an unmet need for improved prediction models that systematically incorporate these variables.

Study Design

This retrospective cohort study examined 622 consecutive patients undergoing TAVI for severe aortic stenosis at a single center between 2018 and 2024. Researchers constructed the COPD-augmented R9CHA2DS2-VA score by adding a point for COPD presence and grading renal dysfunction quantitatively using estimated glomerular filtration rate (eGFR) thresholds (30-59 mL/min/1.73 m: 1 point; <30: 2 points) on top of the already established R9CHA2DS2-VA framework. The primary endpoint was one-year all-cause mortality; secondary endpoint was one-year major adverse cardiovascular events (MACE). Discrimination was evaluated by receiver operating characteristic (ROC) curve analysis with bootstrap internal validation (2,000 resamples), comparing the new score to established CHA2DS2-VASc-based scores. Multivariable logistic regression assessed independent prognostic associations of the continuous score.

Key Findings

The COPD-R9CHA2DS2-VA score demonstrated excellent discriminative ability for one-year all-cause mortality with an apparent area under the curve (AUC) of 0.946 (95% confidence interval [CI] 0.926-0.971). Internal bootstrap validation confirmed negligible optimism bias (mean optimism -0.00014), with the bias-corrected AUC identical to the apparent AUC, underscoring robust model performance within the cohort.

Adding COPD improved mortality prediction significantly over the base R9CHA2DS2-VA score, yielding an incremental AUC increase of +0.033 (p=0.0001). However, COPD’s addition did not meaningfully enhance discrimination for one-year MACE (AUC change +0.001; p=0.853), indicating limited prognostic impact regarding non-fatal cardiovascular events.

Multivariable logistic regression revealed each one-point increment in the COPD-R9CHA2DS2-VA score conferred a 3.52-fold increased odds of mortality (95% CI 2.45-5.06; p<0.001) and a 2.02-fold increased odds of MACE (95% CI 1.67-2.46; p<0.001), independent of other confounders.

The combined multivariable model exhibited exceptional discrimination for mortality (bias-corrected AUC 0.973, 95% CI 0.962-0.984) and good discrimination for MACE (bias-corrected AUC 0.795, 95% CI 0.752-0.841), further supporting the score’s predictive integrity.

Expert Commentary

The COPD-R9CHA2DS2-VA score represents a promising step toward refined risk stratification post-TAVI by integrating two clinically impactful yet often underweighted comorbidities: COPD and graded renal impairment. The internal validation results indicate strong model stability and predictive performance within this high-risk, comorbidity-rich cohort.

However, certain limitations merit consideration. The single-center retrospective design introduces potential selection bias and limits generalizability. Elevated comorbidity burden in the studied cohort may exaggerate discrimination metrics, positioning the reported AUC as an upper-bound rather than definitive performance estimate. Furthermore, the modest incremental benefit of COPD for predicting MACE suggests heterogeneous influence of COPD on distinct cardiovascular endpoints, warranting mechanistic investigations.

Expert consensus underscores the importance of external validation across multicenter, prospective cohorts with variable baseline risks to confirm transportability. Until such data emerge, clinicians should view the COPD-R9CHA2DS2-VA as a hypothesis-generating adjunct rather than a replacement of established risk instruments like STS-PROM or EuroSCORE II.

Conclusion

This study introduces the COPD-R9CHA2DS2-VA score, an innovative risk model systematically incorporating COPD and nuanced renal function grading to predict one-year all-cause mortality and MACE following TAVI. It demonstrated excellent internal discrimination for mortality and independent prognostic relevance. Nonetheless, external validation in diverse patient populations is urgently needed before clinical implementation. For now, the score aids conceptual understanding of comorbidity impacts but should complement, not supplant, existing clinically validated prognostic tools.

Future research should focus on validation, recalibration, and integration of the COPD-R9CHA2DS2-VA score within decision-making workflows, exploring mechanistic pathways linking COPD and renal dysfunction with adverse outcomes after TAVI, and assessing impact on therapeutic personalization to improve patient outcomes.

Reference

Boyaci F, Sahin MK, Akcay M, Yanik A, Yenercag M, Seker OO, Ucar M, Kokcu HI, Ozturk B, Kaya E, Caglioglu H, Yilmaz M, 5%ntuerk EF, Civici H. The COPD-augmented R9CHA2DS2-VA score: development and internal assessment for predicting one-year outcomes after transcatheter aortic valve implantation. BMC Cardiovasc Disord. 2026 Jun 27. doi: 10.1186/s12872-026-06158-0. Epub ahead of print. PMID: 42374241.