New insights on low-risk TAVR vs. surgery: PARTNER 3 at 2 years
In a population of patients with severe symptomatic aortic stenosis at low surgical risk, TAVR using a balloon-expandable valve compared with surgery at 2 years yielded a 37% reduction in death, stroke or CV rehospitalization, but more death, stroke and increased valve thrombosis events occurred from 1 to 2 years among patients who underwent TAVR, according to new data from the PARTNER 3 trial.
As Healio previously reported, PARTNER 3, which enrolled 1,000 patients with severe aortic stenosis and a mean Society of Thoracic Surgeons Predicted Risk Of Mortality score of 1.9%, was presented at the 2019 American College of Cardiology Scientific Session, demonstrating that TAVR with a balloon-expandable valve (Sapien 3, Edwards Lifesciences) was comparable to, and in some aspects superior to, surgery in terms of safety and efficacy.
New 2-year results, reported in a virtual presentation at the 2020 ACC Scientific Session by Michael J. Mack, MD, FACC, medical director of cardiovascular surgery at Baylor Scott and White Health, now show:
- At 2 years, the primary composite endpoint of death, stroke or rehospitalization was reported in 17.4% of patients assigned surgery vs. 11.5% assigned TAVR (HR = 0.63; 95% CI, 0.45-0.88). At 1 year, this outcome occurred in 15.6% vs. 8.5%, respectively (P < .001).
- At 2 years, all-cause death was reported in 3.2% of patients assigned surgery vs. 2.4% assigned TAVR (HR = 0.75; 95% CI, 0.35-1.63). At 1 year, death was reported in 2.5% vs. 1% (P = .08).
- At 2 years, stroke was reported in 3.6% of patients assigned surgery vs. 2.4% assigned TAVR (HR = 0.66; 95% CI, 0.31-1.4). At 1 year, stroke was reported in 3.3% vs. 1.2%, respectively (P = .03).
- At 2 years, death or disabling stroke was reported in 3.8% of patients assigned surgery vs. 3% assigned TAVR (HR = 0.77; 95% CI, 0.39-1.55). At 1 year, death/disabling stroke was reported in 3.1% vs. 1%, respectively (P = .02).
- At 2 years, rehospitalization was reported in 12.5% of patients assigned surgery vs. 8.5% assigned TAVR (HR = 0.67; 95% CI, 0.45-1). At 1 year, rehospitalization was reported in 11.3% vs. 7.3%, respectively (P = .04). The primary cause of rehospitalization between years 1 and 2 was congestive HF in both groups, Mack said.
- At 2 years, valve thrombosis according to the VARC 2 definition was reported in 0.7% of patients assigned surgery vs. 2.6% assigned TAVR (P = .02). At 1 year, valve thrombosis was reported in 0.2% vs. 1%, respectively (P = .13). Most of the events occurred between 1 and 2 years, Mack said.
Regarding the valve thrombosis finding, Mack noted that the valve thrombosis definitions by VARC 2 criteria are outdated and may be exaggerated by recent CT imaging leaflet thickening studies.
The researchers reported no significant deterioration in the functioning of the valve between years 1 and 2 in either group. Also, hemodynamic improvements and frequency of moderate or mild paravalvular regurgitation were unchanged from 1 to 2 years, Mack said.
The PARTNER 3 investigators will continue to follow patients and track outcomes for 10 years. As this trial was conducted in low-risk patients who have a longer life expectancy, “long term assessment of structural valve deterioration is required,” he said.
Reference:
Mack MJ, et al. Joint American College of Cardiology/New England Journal of Medicine Late-Breaking Clinical Trials. Presented at: American College of Cardiology Scientific Session; March 28-30, 2020 (virtual meeting).
Disclosures: The PARTNER 3 trial was funded by Edwards Lifesciences. Mack reports he has received research support from Abbott, Edwards Lifesciences, Gore and Medtronic and is a trial co-principal investigator or study chair for Abbott, Edwards Lifesciences and Medtronic.
Perspective
Back to Top
The new data suggest to me that while the primary outcome still holds at 2 years — ie, TAVR with the Sapien 3 valve is still noninferior and maybe superior to surgery with respect to all-cause mortality, stroke and rehospitalization — there is new concern, or at least a potential signal related to the higher stroke rate and higher valve thrombosis rate at 2 years. This warrants more careful consideration and longer-term follow-up to see if it is just a signal between year 1 and 2 or if it is something that truly pans out as the PARTNER 3 participants are followed for 10 years. This all goes back to working collaboratively with the heart team and personalizing decision making for patients.
Perspective
Back to Top
The PARTNER 3 study comparing TAVR vs. surgical AVR in low-risk patients with severe symptomatic aortic stenosis showed superior mortality, stroke and rehospitalization in TAVR as compared to surgical AVR. At 2-year follow-up, there was some catch up in events with TAVR as compared to surgical AVR. There appears to be higher risk for valve thrombosis, and this has clinical consequences. Patients will still prefer the less-invasive approach to AVR when all other things are equal. But, if the curves now diverge over 10-year follow-up and TAVR becomes worse, particularly in durability, this may lead to reemergence of surgical AVR in the lower-risk population.
I don’t think these data will change practice now, but it will give some support to surgeons who feel surgical AVR is the primary approach for low-risk patients until more long-term data are available. TAVR will remain the standard of care in patients at higher risk for surgery, particularly in older patients, in whom early outcomes matter more than very late outcomes. Results of the long-term follow-up for the low-risk population are of critical importance and will be watched with substantial interest.