Evidence of early bioprosthetic valve dysfunction after TAVR in aortic valve stenosis during one-year follow-up according to VARC-3

Abstract

Abstract Objectives Transcatheter aortic valve replacement (TAVR) has become a frequent treatment method for severe aortic stenosis and has increased significantly in younger patients and at lower surgical risk. Therefore, long-time considerations and the performance of TAVR devices are highly interesting. The Valve Academic Research Consortium 3 endpoint definitions (VARC-3) were released recently, offering updated and clear classifications of bioprosthetic valve dysfunction (BVD). In this retrospective multicenter study, we aimed to assess the prevalence of BVD during one-year follow-up and potential differences between self-expandable (SEV) and balloon-expandable valves (BEV). Methods Underlying conditions were evaluated by univariate analysis. Normality was tested using the Shapiro-Wilk normality test. The primary clinical endpoint was defined as immediate failure in device success, secondary endpoints were defined as prevalence of BVD during one year FU and composite endpoints according to VARC-3. Results The final study cohort consisted of 1478 patients from three tertiary care centers that underwent TAVR between 01/2015 and 12/2018. Mean Follow-up (FU) was 361 ± 26 days. During this time span overall BVD was documented in one-fourth of all patients (-BVD[-]: n=1090 (73.7%); BVD[+]: n=388; 26.3%). Further discrimination into the four main categories of BVD presented following results: Structural valve deterioration (SVD) n=142 (9.6%), Non-Structural valve deterioration (NSVD) n=235 (15.9%), hypo-attenuated leaflet thickening n=18 (1.2 %), and endocarditis n=31 (2.1%). Two-thirds of all events were identified within 30-days post implantation, accounting for new-onset of BVD during time course in one-third of all documented BVD. Device success was significantly lower in the +BVD cohort (80.6% vs. 70.1%, p<0.01*). When comparing SEV and BEV, there were significant differences in overall BVD at 30-days post implantation (13.8% vs 18.5%, p<0.01*), as well as in device success (74.1% vs. 81.5%, p<0.01*), early safety (63.8% vs. 69.8%, p=0.014) and clinical efficacy (94.1% vs. 89.9%, p<0.01*). Again, discrimination into the four main categories of BVD showed differences in NSVD at 30-days post implantation (9.8% vs. 14.3%, p<0.01*) as well as at 1-year FU (7.6% vs 14.8%, p<0.01*). Conclusion In this large retrospective multicenter study, we could observe that NSVD was the BVD category most frequently detected after TAVR in tricuspid aortic valves during 1-year FU. NSVD occurred mainly within 30 days post-implantation in patients with balloon-expandable valves, being one factor that significantly diminished device success, early safety, and clinical efficacy.