Transcatheter aortic valve replacement (TAVR) continued to show superiority over surgical replacement in terms of the primary composite endpoint in low-surgical-risk patients at 2 years of follow-up in the landmark randomized PARTNER 3 trial, but the between-group differences favoring the transcatheter procedure in some key outcomes have narrowed considerably, Michael J. Mack, MD, reported in a video presentation of his research during the joint scientific sessions of the American College of Cardiology and the World Heart Federation, which was presented online this year. ACC organizers chose to present parts of the meeting virtually after COVID-19 concerns caused them to cancel the meeting.
Dr. Michael J. Mack
“On the basis of 1-year data, many physicians were counseling patients that TAVR outcomes were better than surgery. Now we see that the outcomes are roughly the same at 2 years,” said Dr. Mack, who is medical director of cardiothoracic surgery and chairman of the Baylor Scott & White The Heart Hospital – Plano (Tex.) Research Center.
PARTNER 3 randomized 1,000 patients with severe symptomatic aortic stenosis with a tricuspid valve and a very low mean Society of Thoracic Surgeons risk score of 1.9% to TAVR with the Sapien 3 valve or surgical aortic valve replacement (SAVR). The 1-year results presented at ACC 2019 caused a huge stir, with the primary composite outcome of death, stroke, or cardiovascular rehospitalization occurring in 8.5% of TAVR patients and 15.6% of the SAVR group, representing a 48% relative risk reduction and a resounding win for TAVR (N Engl J Med. 2019 May 2;380:1695-705). At 2 years, the difference in the composite outcome remained statistically significant, but the gap had closed: 11.5% with TAVR and 17.4% with SAVR for a 37% relative risk reduction.
Moreover, the between-group difference in stroke, which at 1 year was significantly in favor of TAVR at 1.2% versus 3.3%, was no longer significant at 2 years, with rates of 2.4% versus 3.6%. Nor was the difference in mortality significant: 2.4% with TAVR, 3.2% with SAVR.
What was a statistically significant between-group difference at 2 years – and an eye-catching one at that – involved the cumulative incidence of valve thrombosis confirmed by CT or echocardiography: 2.6% in the TAVR arm, compared with 0.7% with SAVR, with most of these unwanted events coming in year 2.
The good news was there was no echocardiographic evidence of deterioration in valve structure or function in either study arm at 2 years. The mean gradients and aortic valve areas remained unchanged in both arms between 1 and 2 years, as did the frequency of mild or moderate paravalvular leak. Prospective follow-up will continue annually out to 10 years.
“I think it’s way too early to expect to see a signal, but I think it’s somewhat comforting at this point that there’s no signal of early structural valve deterioration,” Dr. Mack said.
Discussant Howard C. Hermann, MD, commented: “I guess the biggest concern in looking at the data is the increase in stroke and valve thrombosis, both numerically and relative to SAVR, between years 1 and 2.”
Dr. Mack offered a note of reassurance regarding the valve thrombosis findings: The rates he presented were based upon the now-outdated second Valve Academic Research Consortium (VARC-2) definition, per study protocol. When he and his coinvestigators recalculated the valve thrombosis rates using the contemporary VARC-3 definition of valve deterioration and bioprosthetic valve failure, the incidence was very low and not significantly different in the two study arms, at roughly 1%.
Dr. Hermann, professor of medicine and director of the cardiac catheterization laboratories at the University of Pennsylvania, Philadelphia, had a question: As a clinician taking care of TAVR patients, what clinical or hemodynamic findings should prompt an imaging study looking for valve thrombus or deterioration that might prompt initiating oral anticoagulation?
“If there’s a change in hemodynamics, an increasing valve gradient, if there’s increasing paravalvular leak, or if there’s a change in symptoms, that should prompt an imaging study. Only with confirmation of valve thrombosis on an imaging study should anticoagulation be considered. Oral anticoagulation is not benign: Of the six clinical events associated with valve thrombosis in the study, two were related to anticoagulation,” Dr. Mack replied.
“Regarding whether patients should receive warfarin or a novel anticoagulant, I don’t think we have evidence that there’s benefit to anything other than warfarin at the current time,” he added.
Dr. Mack reported receiving research support from Edwards Lifesciences, the sponsor of PARTNER 3, as well as from Abbott, Gore, and Medtronic.