New TAVR Insights: Survival Advantage of TAVR Over SAVR Not Sustained At Five Years, Severe Prosthesis-Patient Mismatch Not Benign Following TAVR

The 5-year outcomes of a trial conducted by Gleason et al. published in JACC demonstrated a similar safety profile, functional recovery, and freedom from severe structural valve deterioration (SVD) for both TAVR and SAVR, consistent with earlier-term reports. These outcomes supported TAVR as a reasonable alternative to SAVR in the high-risk population and its current Class I indication.

The CoreValve US Pivotal High-Risk Trial was the first randomized trial to show superior 1-year mortality of transcatheter aortic valve replacement (TAVR) compared with surgical aortic valve replacement (SAVR) among high operative mortality–risk patients. The current paper by Gleason et al. reported long(er) term (5-year) outcomes following self-expanding transcatheter aortic valve replacement (TAVR) with CoreValve (Medtronic, Minneapolis, Minnesota) versus surgical aortic valve replacement (SAVR) as an as-treated (AT) analysis of the CoreValve U.S. Pivotal High-Risk trial. The investigators sought to compare TAVR to SAVR for mid-term 5-year outcomes of safety, performance, and durability. Surgical high-risk patients were randomized (1:1) to TAVR with the self-expanding bioprosthesis or SAVR. VARC-1 (Valve Academic Research Consortium I) definitions were applied. Severe hemodynamic structural valve deterioration was defined as a mean gradient ≥ 40 mm Hg or a change in gradient ≥ 20 mm Hg or new severe aortic regurgitation. Five-year follow-up was planned.

In this trial, Gleason and his colleagues from the University of Pittsburgh School of Medicine randomized a total of 797 patients at 45 U.S. centers, of whom 750 underwent an attempted implant (TAVR ¼ 391, SAVR ¼ 359). The overall mean age was 83 years, and the STS score was 7.4%. The findings of the study depicted all-cause mortality rates at 5 years of 55.3% for TAVR and 55.4% for SAVR. Subgroup analysis showed no differences in mortality. Major stroke rates were 12.3% for TAVR and 13.2% for SAVR. Mean aortic valve gradients were 7.1 ± 3.6 mm Hg for TAVR and 10.9 ± 5.7 mm Hg for SAVR. No clinically significant valve thrombosis was observed. Freedom from severe SVD was 99.2% for TAVR and 98.3% for SAVR (p ¼ 0.32), and freedom from valve reintervention was 97.0% for TAVR and 98.9% for SAVR (p ¼ 0.04). A permanent pacemaker was implanted in 33.0% of TAVR and 19.8% of SAVR patients at 5 years.

Therefore, the study demonstrated similar mid-term survival and stroke rates in high-risk patients following TAVR or SAVR. There appeared to be no difference in overall stroke, but there was an excess of stroke in SAVR versus TAVR in those undergoing nonfemoral access (18.4% for TAVR and 33.5% for SAVR; p ¼ 0.02).  Indeed, TAVR with the self-expanding valve in high-risk patients at 5 years had a similar safety profile as SAVR. Functional outcomes were good in both groups. Significant SVD was very unusual in both groups. There was a greater need for pacemaker implantation and a small excess of valve reintervention with TAVR but a significant excess of moderate structural valve deterioration with SAVR.

TAVR was a reasonable alternative to SAVR in the elderly high-risk group. However, the investigators believed that long-term data regarding structural valve deterioration and the impact of stent frames of transcatheter aortic valves were still necessary to establish TAVR’s utility and appropriateness in patients with extended life expectancy. Commenting on the findings of the study in an accompanying editorial, Dr. Hasan Jilaihawi, NYU, stated, “Although there was no difference between TAVR and SAVR in 5-year mortality, the principal emphasis should not be on long-term survival in an elderly population, many of whom would be likely to die over that time period in the absence of disease, and with evidence of disparate late data capture between groups. Even with a first-generation self-expanding TAVR device, in comparison with surgery, patients were discharged sooner, with lower rates of stroke in those with extreme vasculopathy, and stayed out of the hospital longer and, interestingly in the face of durability questions with TAVR, had less structural valve deterioration.” Speaking of the clinical implications of the study, he added, “Undoubtedly, there are anatomic subsets where the “TAVR risk” may outweigh the “SAVR risk,” but this has become the exception rather than the rule. The tables have turned and, at least in those with greater than low surgical risk, TAVR has become the first choice for patients, cardiologists, and nontechnically biased surgeons alike; in such patients, rather than performing TAVR for only those with prohibitive surgical risk, we are now reserving SAVR for cases with elevated TAVR risk.”

In another elegant study that constituted the largest analysis on the incidence and clinical impact of Prosthesis-patient mismatch (PPM) following AVR, Herrmann et al. reported that severe PPM was not benign following TAVR. Severe PPM was indeed associated with a 15% to 20% increase in the risk of 1-year mortality and heart failure rehospitalization. However, several factors may have contributed to underestimating the impact of PPM in this study, including the short follow-up and the potential overestimation of the presence and severity of PPM due to obesity, low-flow state, and pressure recovery. Further studies are needed to assess the impact of PPM, defined with the use of the predicted EOAi, on mortality, rehospitalization, valve reintervention, and quality of life.