Excluded From Pivotal Trials, Low-risk Complex AS Patients Warrant Caution
Death and stroke outcomes in most patients with multivalvular or CV disease were as good as or better than those with AS alone.
A significant proportion of real-world patients with severe aortic stenosis and low surgical risk have an additional valvular or CV condition that was never accounted for in pivotal trials of TAVI versus SAVR and that requires a serious discussion by the heart team, Canadian research shows.
In the study of primarily low-risk patients who underwent SAVR between 2000 and 2019, 40.8% had at least one of the following conditions: non-tricuspid aortic valve (NTAV), severe coronary artery disease (SevCAD), concomitant mitral/tricuspid valve (CMTV), or concomitant ascending aorta replacement (CAAR). All four of those patient subgroups were routinely excluded from RCTs of TAVI compared with SAVR.
The researchers, led by Alberto Alperi, MD (Laval University, Quebec, Canada), found that for the most part, SAVR resulted in similar 30-day and 1-year outcomes among patients in these subgroups compared with those who had straightforward aortic stenosis. Overall, observed mortality was lower than the estimates from EuroSCORE II and the STS score. The exception was the subgroup with CMTV, for whom 30-day and 1-year mortality were higher than in the non-CMTV patients.
“These results should be considered in the expansion of TAVR toward the treatment of low-risk patients and may inform future randomized trials according to specific clinical variables, particularly in the presence of NTAV, CMTV, and SevCAD,” Alperi and colleagues write in the paper, published this week in the Journal of the American College of Cardiology. While the PARTNER 3 and Evolut low-risk trials indicated that TAVI may be an acceptable option in many low-risk patients, the authors stress that more data are needed to understand the prevalence and clinical impact of these excluded subgroups.
Commenting for TCTMD, cardiovascular surgeon Gilbert Tang, MD (Mount Sinai Health System, New York, NY), said the Canadian findings, though limited by their nonrandomized, single-center design, suggest that SAVR is very safe in patients who fall into these subgroup categories that make them more complex than other low-risk severe aortic stenosis patients.
“I think it shows that patients with multivalvular disease, or aortic stenosis plus other cardiac conditions that would benefit from combined interventions, should have a surgical consultation and a heart team discussion about the optimal management strategy,” he said.
Clinical Outcomes Seen
The study by Alperi and colleagues included 6,772 patients. Of those who had at least one of the clinical variables of interest, NTAV (mainly bicuspid aortic valves) was the most common at 27.6%, followed by CAAR at 10.5%, and SevCAD and CMTV at 5.8% each. The mean STS score was 1.94% and the mean EuroSCORE II was 3.10%.
In the NTAV subgroup, the 30-day mortality rate was 0.9%, which was lower than the risk-adjusted mortality ratio (RAMR) estimated by EuroSCORE II (RAMR 0.32; 95% CI: 0.19-0.55) or STS (RAMR 0.59; 95% CI: 0.34-1.02). The 30-day stroke and TIA rates were 1.6% and 0.3%, respectively. Compared with the rest of the cohort, the NTAV subgroup had a lower rate of 30-day mortality (adjusted OR 0.42; 95% CI: 0.22-0.81) and similar stroke and TIA rates. Mortality at 1 year also was lower (adjusted HR 0.59; 95% CI: 0.38-0.93). Further analysis by surgical date indicated a lower observed than expected mortality rate for the late versus early NTAV patients.
Alperi and colleagues note that the positive outcomes in the NTAV group add to a recent analysis of the Society of Thoracic Surgeons/American College of Cardiology TVT Registry, which included low-risk patients with bicuspid aortic valve. That study also found comparable early clinical outcomes, including mortality and disabling stroke, following TAVI compared with patients who had trileaflet stenosis.
In the SevCAD subgroup, the 30-day mortality was 2.6%, which was similar to the RAMR estimated by EuroSCORE II and STS. The 30-day stroke and TIA rates were 3.3% and 0.3%, respectively. Compared with the rest of the cohort, the SevCAD subgroup had similar rates of 30-day mortality, stroke, TIA, and 1-year mortality. In both the early and late SevCAD patients, mortality rates were similar to those expected by STS, with some differences according to left main involvement.
In the CMTV subgroup, the 30-day mortality was 5.9%, which was higher than that estimated by STS (RAMR 2.27; 95% CI 1.41-3.70) and similar to EuroSCORE II. The 30-day stroke rate was 2.3%, and the TIA rate was 0.98%. Compared with the rest of the cohort, the CMTV subgroup had higher 30-day mortality (adjusted OR 2.61; 95% CI 1.51-4.50), but no differences in rates of stroke or TIA. The rate of 1-year mortality was higher than that observed for patients without CMTV (adjusted HR 2.50; 95% CI 1.65-3.80).
Finally, in the CAAR subgroup, the observed 30-day mortality was 2.1%, which was similar to that estimated by STS and lower than that estimated by EuroSCORE II. The 30-day stroke rate was 2.3%, and there were no TIA events. Compared with the rest of the cohort, the CAAR group had no significant differences in 30-day mortality, stroke, or 1-year mortality.
Caution, Holistic Approach Needed
According to Alperi and colleagues, bicuspid aortic valve “represented one of the largest subgroups excluded from trials with a plausible option of being treated with TAVR: 945 out of 5,310 patients exhibited a bicuspid aortic valve as the only characteristic precluding trial participation, thus representing approximately one-fifth of the entire severe aortic stenosis low-risk population.”
For TAVI to be considered in this particular subset, they add, the results “must be outstanding to mimic those obtained by SAVR in this important group of patients.”
To TCTMD, Tang agreed that the caution from Alperi and colleagues about the bicuspid subgroup is important in light of registry data suggesting that bicuspid TAVI could result in a higher risk of stroke.
“For younger, low-risk patients that's not ideal, because even though mortality is a hard endpoint, most patients and families would say that stroke may be even worse than death, especially disabling stroke,” Tang added. “There needs to be a holistic approach to managing these patients, and patients should have a say as well [regarding] what they want in terms of care and with what they feel comfortable.” He also noted that particularly for patients in the more uncommon subgroups, referral to high-volume centers with more experience in complex interventions may be warranted.
In an accompanying editorial, Poonam Velagapudi, MD, MS (University of Nebraska Medical Center, Omaha), and colleagues suggest an algorithm to help with decision-making in a low-risk patient with any of the four complex disease characteristics identified in the study. Along the way, there are clear indications pointing to surgery, or to TAVI, but in specific scenarios where patient wishes are at odds with anatomical considerations, the algorithm recommends a heart team discussion. Treatment decisions, the editorialists add, are on the horizon, with clinical trial designs having been proposed for some of the subgroups, including SevCAD and NTAV.
“Until then, low-risk patients undergoing TAVR must be carefully selected by the heart team based on patient and anatomic factors with a thorough discussion of risks versus benefits; SAVR remains the treatment of choice in certain clinical situations,” Velagapudi et al conclude.
Alperi A, Voisine P, Kalavrouziotis D, et al. Aortic valve replacement in low-risk patients with severe aortic stenosis outside randomized trials. J Am Coll Cardiol. 2021;77:111-123.
Velagapudi P, Bapat V, Kodali S. When excluded from randomized clinical trials: to “OR” or “‘TAVR”? J Am Coll Cardiol. 2021;77:124-127.
- Alperi was supported by a research grant from Martín Escudero foundation.
- Velagapudi reports no relevant conflicts of interest.
- Tang is a consultant for Abbott, Medtronic and W.L. Gore & Associates; and serves as a physician proctor for Medtronic.