In recent years, several studies have compared TAVI with SAVR in patients at high and intermediate surgical risk, yielding favorable outcomes that have supported the expansion of TAVI in these populations. This progress prompted studies involving low-risk patients, such as the Evolut Low Risk (using self-expanding Corevalve® or Evolut® prostheses) and the PARTNER 3 study (using balloon-expandable SAPIEN 3® prostheses). 

The PARTNER 3 study included patients with severe symptomatic aortic stenosis who were at low surgical risk (defined by an STS-PROM score <4% and Heart Team consensus) and eligible for transfemoral access. Patients were randomized 1:1 to undergo TAVI with a balloon-expandable valve or SAVR. The primary endpoint was a composite of death, stroke, or rehospitalization (related to the procedure, valve, or heart failure), with a statistically significant reduction observed in the TAVI group at 1 and 2 years. The longer-term durability of these results remained unknown, leading to the extension of follow-up to five years. 

A total of 1,000 patients were randomized, with 503 in the TAVI group and 497 in the SAVR group, of whom 948 received the initially intended valve (496 TAVI and 454 SAVR). A notable disproportionate patient loss occurred in the surgical group. Through a telehealth-based tracking, vital status data were obtained for 66 of the 95 patients lost to follow-up. The mean patient age was 73 years, with 69% men and a mean STS-PROM score of 1.9%. The primary endpoint remained the same as in the original study, expressed as estimated percentages using the Kaplan-Meier method. Since patients could experience more than one event or hospitalization during the five-year period, another hierarchical composite primary endpoint was considered, including death, disabling or non-disabling stroke, and days of rehospitalization, expressed through a win ratio analysis. Secondary endpoints at five years included: death, stroke, new-onset atrial fibrillation (AF), reoperation on the aortic valve, endocarditis, prosthetic thrombosis, prosthetic dysfunction, functional status, and quality of life as measured by the Kansas City Cardiomyopathy Questionnaire (KCCQ-OS). A time-to-event analysis was conducted from baseline to 1 year, from 1 to 5 years, and from baseline to 5 years. 

The primary composite endpoint (death, stroke, or rehospitalization) from baseline to 5 years occurred in 22.8% of patients in the TAVI group and 27.2% in the SAVR group, with no statistically significant difference (p= 0.07). In an analysis of years 1 to 5, the primary endpoint occurred in 15.7% of TAVI patients and 13.7% of SAVR patients. The win ratio for the other hierarchical composite primary endpoint (death, disabling or non-disabling stroke, and days of rehospitalization) was 1.17, favoring TAVI, although it was not statistically significant (p=0.25). For other five-year outcomes, mortality was 10% for TAVI vs. 8.2% for SAVR, stroke 5.8% vs. 6.4%, rehospitalization 13.7% vs. 17.4%, new-onset AF 13.7% in TAVI vs. 42.2% in SAVR, and pacemaker implantation 13.5% in the TAVI group vs. 10.4% in the SAVR group. As for prosthetic complications, 2.5% of clinically significant prosthetic thrombosis cases occurred in the TAVI group versus 0.2% in the surgical group. Additionally, aortic insufficiency (AI) greater than mild was found in 24.5% of TAVI patients vs. 6.3% of SAVR patients. Prosthetic gradients were similar in both groups (12.8±6.5 mmHg in TAVI and 11.7±5.6 mmHg in surgery). At five years, 84% of patients in the TAVI group and 86% in the surgical group were alive and in NYHA class I-II. 

A significant aspect of this trial is the high number of patients lost to follow-up, particularly in the surgical group, which the authors themselves consider a potential source of bias. Another critique is the discrepancy between numerical figures present in tables and the text, with no explanation offered either in the article or supplementary material (for instance, in Table 1, the patients at risk from years 1 to 5, 490 for TAVI vs. 427 for SAVR, differ from the text on the following page, 453 TAVI vs. 372 SAVR, while primary endpoint percentages remain unchanged at 15.7% for TAVI vs. 13.7% for surgery). 

The authors conclude that among low-risk patients with severe, symptomatic aortic stenosis who underwent TAVI or surgery, there were no significant differences between groups in the two primary composite outcomes at five years, with the initial favorable trend for TAVI observed at 1 and 2 years reversing over time. 

COMMENTARY: 

The expansion of TAVI in patients with severe aortic stenosis at high and intermediate surgical risk has increased significantly in recent years. Current clinical practice guidelines from the European Society of Cardiology recommend TAVI for patients >75 years or at high surgical risk (STS-PROM or Euroscore II >8%) or deemed inoperable. For intermediate-risk patients, decision-making should occur within the Heart Team framework, considering anatomical, clinical characteristics, and patient preferences. However, for low-risk surgical patients, surgical valve replacement remains the treatment of choice. Given the favorable outcomes for TAVI in intermediate- and high-risk patients with limited follow-up due to survival constraints imposed by comorbidities, evaluating this technique in low-risk patients became appealing. As previously discussed, short-term findings were promising, but the five-year PARTNER 3 results cast doubt on the durability of TAVI. Likely, TAVI’s short-term benefits (lower mortality and stroke rates, shorter hospital stay, reduced rehospitalization, lower rates of AF and bleeding) diminish over time, with a notable increase in mortality and stroke rates in the TAVI group after the first year of follow-up. Additionally, prosthetic thrombosis, pacemaker implantation, and perivalvular insufficiency incidence were higher in the TAVI group. These complications could lead to worsened clinical and quality-of-life outcomes for patients and a potential need for reintervention, negating TAVI’s initial less invasive advantage. It seems prudent to continue evaluating each case within the Heart Team session and to inform low-risk patients of the benefits and risks of each technique, offering the best individualized treatment. 

REFERENCE: 

Mack MJ, Leon MB, Thourani VH, Pibarot P, Hahn RT, Genereux P,et al. Transcatheter Aortic-Valve Replacement in Low-Risk Patients at Five Years. N Engl J Med. 2023 Nov 23;389(21):1949-1960. doi: 10.1056/NEJMoa2307447.