Impact of age versus comorbidities on outcomes after transcatheter aortic valve implantation: validating the guideline-recommended age cut-off

Abstract

Abstract Background Transcatheter aortic valve implantation (TAVI) has become the most common treatment for patients with severe aortic stenosis (AS) in Europe and North America. Current European guidelines recommend transfemoral TAVI for patients with severe, symptomatic AS aged ≥75 years. However, this cut-off is based solely on coincidentally existing patient cohort mean age of randomized controlled trials and individual risk factors might have a greater impact on outcomes after TAVI compared to age alone. Purpose This study aimed to validate the guideline-recommended age cut-off of 75 years compared to a comorbidity focused approach. Methods Using long-term prospective single-centre data, we analysed 4,278 patients with severe, symptomatic AS undergoing transfemoral TAVI between 2008-2022. A multivariable Cox regression model was fitted to assess predictors of all-cause mortality. Based on this model, a comorbid profile was defined as at least one prevalent comorbidity from the final model. Kaplan-Meier estimates were calculated for all-cause mortality after 3 years for young (<75 years) and elderly (≥75 years), and comorbid and non-comorbid patients. Results A total of 1,620 patients were included in the final analysis (<75 years: N=302 [18.6%], 70.8 [IQR 67.4-73.3] years; ≥75 years: N=1,318 [81.4%], 82.6 [IQR 79.5-85.6] years). In multivariable analysis, male gender (hazard ratio [HR] 1.37, 95%-confidence interval [CI] 1.06-1.77), atrial fibrillation (AF) (HR 1.78, 95%-CI 1.35-2.33), chronic obstructive pulmonary disease (COPD) (HR 1.44, 95%-CI 1.07-1.93) and chronic kidney disease (CKD) (eGFR <60ml/min/1.72m2) (HR 0.99, 95%-CI 0.98-0.99) were identified as predictors independently associated with all-cause mortality. Overall, 64% of patients were defined as comorbid by at least one risk factor emerging as an independent predictor: AF/COPD/CKD. While there was no difference in all-cause mortality between patients aged < or ≥75 years (27.5% vs. 23.9%, p=0.62) (Figure 1A), 3-year mortality rates were significantly higher in comorbid (30.4%) compared to non-comorbid subjects (14.5%, p<0.001) (Figure 1B). No difference was found comparing non-comorbid young to non-comorbid elderly (16.2% vs. 14.1%, p=0.69), but comorbid young showed higher mortality than non-comorbid elderly patients (34.0% vs. 14.1%, p<0.001) (Figure 1C). According to age-adjusted multivariable analysis, a comorbid risk profile was independently associated with all-cause mortality (HR 2.09, 95%-CI 1.49-2.93), whereas age was not predictive. Conclusion In a large all-comer TAVI patient cohort age was not predictive for mortality within a 3-year follow-up. This finding was true for both, non-comorbid as well as comorbid patients. Thus, outcome of patients with AS undergoing transfemoral TAVI depends on patients’ comorbidities rather than age. Our findings do not confirm the guideline recommended binary age cut-off currently recommended for patients with severe symptomatic AS.Figure 1