The low-risk TAVR vs SAVR RCTs- PARTNER-3, Evolut Low-Risk, DEDICATE, and NOTION-2- demonstrate that TAVR is at least as effective as SAVR in terms of early mortality and disabling stroke, although outcomes varied across studies. The excellent outcomes observed in the PARTNER-3 trial reflect the highly selected patient population, while the NOTION-2 trial highlights the challenges posed by anatomical variations, such as bicuspid aortic valves (AV). These findings highlight the importance of a personalized approach guided by a multidisciplinary Heart Team, taking into account both clinical and anatomical aspects when choosing between TAVR and SAVR.
Trial registration number: ClinicalTrials.gov NCT02825134.
Five years following publication of the pivotal PARTNER-3 and Evolut Low-Risk randomized clinical trials (RCTs)—which assigned low-risk patients with severe symptomatic aortic stenosis (AS) to either transcatheter aortic valve replacement (TAVR) or surgical aortic valve replacement (SAVR) , -the results of 2 more low-risk TAVR vs SAVR trials have been reported recently. The DEDICATE trial was conducted at 38 sites in Germany and assigned low or intermediate surgical risk patients with severe AS to TAVR or SAVR, whilst the NOTION-2 trial was conducted at 9 Northern European sites and compared outcomes following TAVR or SAVR in low-risk AS patients aged 60 to 75 years. Previous meta-analyses based on the above-mentioned low-risk RCTs have reported a reduced or similar risk of all-cause mortality and/or stroke after TAVR when compared to SAVR, just assuming that the TAVR and SAVR cohorts in these RCTs were comparable. ,, Unfortunately, this is not necessarily the case, leading to meta-analyses whose rationale and significance is questionable. In this review, we reflect on the similarities shared by these 4 low-risk TAVR RCTs and on important differences in terms of trial design and outcomes.
Clinical outcome parity within trials but notable variability across studies
All 4 low-risk RCTs found TAVR to be comparable or superior to SAVR at 12-to-24-month follow-up according to their varying primary outcomes (PARTNER-3 and NOTION-2: death, stroke, or rehospitalization related to the valve, procedure, or heart failure; , DEDICATE: death or all stroke; Evolut Low-Risk: death or disabling stroke ). However, true comparison of the clinical outcomes reported in these trials is only possible by comparing the event rates of a hard and unambiguous endpoint—namely the composite of all-cause mortality and disabling stroke—which is currently only available at 12 months for all 4 RCTs. Figure 1 shows that the risk of all-cause mortality or disabling stroke following TAVR or SAVR was 2% to 5% at 12 months in most study cohorts, with some notable outliers as we shall discuss. Additionally, Figure 1 presents the one-year event rates for other significant secondary endpoints.
One-year clinical outcomes in the different low-risk TAVR vs SAVR trials. The top panel shows the reported one-year incidence rates of the combined endpoint death or disabling stroke in the TAVR ( left ) and SAVR ( right ) cohorts of the respective low-risk TAVR trials. The table below shows the one-year incidence rates of key secondary clinical outcomes in both treatment arms. SAVR, surgical aortic valve replacement; TAVR, transcatheter aortic valve replacement.
Comparison of TAVR cohorts
The rate of all-cause mortality or disabling stroke was lowest (1.0%) in the PARTNER-3 TAVR cohort, reflecting the extensive and strict selection criteria that excluded 25% of screened patients owing to strict anatomical considerations (e.g., left ventricular outflow tract or aortic root calcification, small aortic annular dimensions, etc.), resulting in the randomization of a highly selected low-risk AS population that may have favored TAVR outcomes. At the other extreme, the risk of all-cause mortality or disabling stroke was highest (6.1%) in the NOTION-2 bicuspid TAVR cohort, consistent with preceding retrospective data, demonstrating that the risk of devastating complications (annular rupture, sinus perforation, disabling stroke) is higher for TAVR in patients with bicuspid AS compared to those with tricuspid AS. Hereby, it is important to note that patients with bicuspid AS were excluded from all 3 other RCTs. However, a cautious interpretation of these data is necessary, since the NOTION-2 trial only enrolled a total of 100 bicuspid patients. Moreover, this trial did not include an assessment committee to verify anatomical suitability for TAVR. Additionally, 2 of the mortality events occurred in patients with a bicuspid Sievers type 2 morphology—although these are typically classified as unicuspid aortic valves, not bicuspid—and therefore should not have been enrolled in this trial. Finally, a single transcatheter heart valve (THV) was used in each of the PARTNER-3 and Evolut Low-Risk trials, whereas the use of a variety of different THVs was permitted in the more independent, investigator-initiated NOTION-2 and DEDICATE protocols. ,,,
Comparison of SAVR cohorts
Shifting focus to surgery, one of the most striking observations was the poor clinical outcome following SAVR in the German DEDICATE trial, with an 8.4% risk of all-cause mortality or disabling stroke at 12-month follow-up, compared with 1.6% and 2.9%, respectively, in the NOTION-2 and PARTNER-3 SAVR cohorts ( Figure 1 ). ,
Explanations for these widely varying results remain uncertain, but one possible explanation is the varying surgical risk profile of the study populations ( Table 1 ). The DEDICATE trial initially set out as an RCT enrolling intermediate-risk patients but evolved into a low-risk trial over time. Nevertheless, the DEDICATE trial enrolled slightly older patients with more co-morbidities and a higher surgical risk profile than the NOTION-2 trial, which truly enrolled young, robust patients with few co-morbidities (mean STS risk score 1.1%). , Another important observation is that concomitant surgery was most commonly performed in the PARTNER 3 trial (coronary artery bypass graft (CABG) 12.8%, mitral valve repair 1.6%) and the Evolut Low Risk trial (CABG 13.6%, mitral valve repair 0.6%), whereas this was performed to a lesser extent in the NOTION-2 trial (CABG 9.8%, mitral valve repair 0%) and DEDICATE trial (CABG 1.8%, mitral valve repair 0.2%). Unfortunately, detailed data on the type and size of surgical valves has not been reported systematically in these trials.
Table 1
Baseline characteristics in the different low-risk TAVR vs SAVR trials.
| PARTNER 3 | Evolut LR | DEDICATE | NOTION-2 | Tricuspid NOTION-2 | Bicuspid NOTION-2 | |
|---|---|---|---|---|---|---|
| Age, years | 73.5 | 73.9 | 74.5 | 71.0 | 71.5 | 69.8 |
| Gender, male | 69.3% | 65.1% | 56.7% | 62.7% | 65.2% | 56.0% |
| STS-score, % | 1.9 | 1.9 | 1.9 | 1.1 | 1.2 | 1.0 |
| Diabetes mellitus | 31% | 31% | 33% | 21% | 26% | 9% |
| Coronary artery disease | 28% | 21% | 36% | 12% | 14% | 4% |
| Atrial fibrillation | 17% | 16% | 28% | 16% | 16% | 16% |
| Bicuspid aortic valve | 0 | 0 | 0 | 26% | 0 | 100% |
| Aortic annulus, PD mean Ø, mm | 78.4 | NA | 78.6 | 79.2 | 79.8 | 79.0 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
