Leaflet Thrombosis Common After Aortic Valve Implant, More So in TAVR Than in Surgery

WASHINGTON, DC—Subclinical leaflet thrombosis occurs “frequently” in patients who received bioprosthetic valves for the treatment of aortic stenosis, with the problem occurring more commonly in patients who undergo transcatheter versus surgical aortic valve replacement (SAVR), according to data presented today at the American College of Cardiology (ACC) 2017 Scientific Session.

Of 890 patients who underwent computed tomography (CT) imaging and had interpretable scans in the RESOLVE and SAVORY registries, 12% had subclinical leaflet thrombosis, including five of 138 patients (3.6%) with thrombosis of surgically implanted valves and 101 of 752 patients (13.4%) who underwent TAVR.

Overall, the rate of stroke was not significantly different among patients with and without reduced leaflet motion, but subclinical leaflet thrombosis was associated with a significantly increased risk of transient ischemic attack compared with no CT-identified leaflet thrombosis (6% vs 1%; P = 0.0005). 

For Raj Makkar, MD (Cedars Sinai Medical Center, Los Angeles, CA), the senior researcher who presented the results during the late-breaking clinical trial session, the data should challenge the current standard of care with dual antiplatelet therapy (DAPT) given that anticoagulation, not DAPT, was protective. The prevalence of reduced leaflet motion was significantly lower among patients receiving oral anticoagulation than in patients who received DAPT (3.6% vs 14.9%; P<0.001), with the non-vitamin K oral anticoagulants (NOACs) equally effective as warfarin for the prevention of leaflet thrombosis.

“We have to be careful how we interpret these imaging findings and apply them to everyday clinical situations,” Makkar told TCTMD. “What I take away is that dual antiplatelet therapy is not very effective. So why bother giving it to someone who is old and fragile if it is not effective in reducing subclinical leaflet thrombosis? For people who are at an increased risk of bleeding, perhaps we could do away with one antiplatelet agent.”

On the opposite end of the spectrum, in younger patients, which is where TAVR is expanding, “perhaps it is more important to consider giving oral anticoagulation,” said Makkar.   

The results of the study were published online today in the Lancet to coincide with the ACC presentation.

Reduced Leaflet Motion Identified in 2015

The issue of subclinical leaflet thrombosis first emerged in 2015 when Makkar published pooled data from RESOLVE and SAVORY, as well as data from the Portico Resheathable Transcatheter Aortic Valve System US investigational device exemption (PORTICO-IDE) study, in the New England Journal of Medicine. That early data identified an increased risk of leaflet thickening in 22 of 55 patients who underwent TAVR in the PORTICO-IDE study and in 17 of 132 patients who underwent TAVR in RESOLVE/SAVORY (and in 2 of 27 valves implanted surgically).

In this expanded analysis of 931 patients, CT scans were performed in 84% of patients with transcatheter valves and 16% of surgically treated patients. The median time from valve replacement to CT scan in the entire cohort was 83 days, with the surgical patients undergoing CT after 163 days and the TAVR group after 58 days. In the registry, a wide range of TAVR technologies were used, including Sapien 3 (Edwards Lifesciences), CoreValve and Evolut R (Medtronic), Lotus (Boston Scientific), and Portico (St. Jude Medical), among others.

Among 58 patients with reduced leaflet motion who had follow-up CT imaging, the impaired motion was resolved in all 36 patients who received anticoagulation for 3 months. In contrast, subclinical leaflet thrombosis persisted or progressed in more than 90% of patients not receiving oral anticoagulation. Among eight patients who had leaflet function restored following anticoagulation, but who stopped treatment after 3 months, follow-up CT scans revealed that reduced leaflet motion reoccurred in four patients.   

For Danny Dvir, MD (University of Washington, Seattle), who was not involved in the analysis, the latest update helps physicians better understand the magnitude of the issue, but it still leaves many questions unanswered, not the least of which is whether all patients should be treated with anticoagulation after valve implantation.

“Although the paper supports the possible need for treating patients with anticoagulation, we cannot start treating most of our frail, severe aortic stenosis patients with anticoagulation until we have clear answers from randomized trials, such as GALILEO and ATLANTIS,” Dvir told TCTMD. “We must remember that the vast majority of patients with reduced leaflet thickening did not have any clinical events and their transvalvular gradients were within normal limits. If we start treating many of our patients with anticoagulation without a solid indication we may do harm to some of them by causing serious bleeding events.”

That said, Dvir predicts that in selected patient populations, such as those with a low risk of bleeding, use of oral anticoagulation will increase.

Makkar agreed that data from RESOLVE/SAVORY should stimulate debate but said that clinical recommendations for antiplatelet and/or anticoagulant therapy must come from randomized clinical trials. He pointed out, though, that trials testing NOACs for stroke prevention in TAVR include higher-risk/older patients, likely leaving a gap in knowledge about anticoagulation in younger populations.

“I would say, you can go patient by patient,” said Makkar, referring to anticoagulation therapy. “If I was 65 years old and getting TAVR, I would either want to have routine CT to make sure I don’t have [subclinical leaflet thrombosis] or would go ahead and take a short course of oral anticoagulation, although that’s not based on solid clinical trial data.”

Valve Durability?   

For Dvir, what is also not known is whether reduced leaflet motion observed on CT might result in an accelerated degeneration of the valve over time and if TAVR is truly associated with a larger risk of subclinical thrombosis than SAVR. “We will need to wait several years until the large trials gather true long-term follow-up and can answer these questions,” he said.

To TCTMD, Makkar agreed that whether or not subclinical leaflet thrombosis affects durability or structural integrity of the valve long-term remains to be seen. In their observational study, there was a small, but significant, increase in the mean aortic valve gradient of patients with reduced leaflet motion.

Additionally, 14% of patients with subclinical leaflet thrombosis had aortic valve gradients greater than 20 mm Hg and a new increase in the aortic valve gradient of more than 10 mm Hg. Turning this around, Makkar said that approximately 85% of patients with leaflet thickening had no change in their aortic valve gradient. This suggests that impaired leaflet motion is hemodynamically silent in the vast majority of patients and can be missed by transthoracic echocardiography.

The change in aortic-valve gradient among certain patients with subclinical leaflet thrombosis might represent a high-risk subset who could be vulnerable to valve deterioration over time, said Makkar. “I think we can speculate that [leaflet thickening] might have an impact on long-term durability, but I think it’s going to be very hard to prove it,” he said.   

Gregg Stone, MD (Columbia University Medical Center, New York, NY), who wrote an editorial along with Jeroen Bax, MD, PhD (Leiden University Medical Center, the Netherlands), said the identification of subclinical leaflet thrombosis is a relatively new phenomenon and “we don’t know what it means.” Numerous studies, registries, and comparative randomized trials are currently looking at the issue to determine if there are true differences in rates of leaflet thrombosis between SAVR- and TAVR-treated patients as well as the clinical implications of the findings.

“This is the best study to date looking at this issue, because it’s the largest and because it includes two separate registries,” said Stone. “I think the findings are intriguing and interesting, but they certainly need validation.”

To TCTMD, Stone noted that early leaflet thrombosis might resolve without permanent clinical sequelae. In the analysis of RESOLVE and SAVORY, there was only a “very weak link” between subclinical leaflet thrombosis and transient ischemic attack, which are clinical events physicians want to avoid but aren’t associated with any long-term, permanent adverse outcomes for the patient.

“Right now, we have good randomized data of TAVR and SAVR with careful echocardiographic follow-up of patients through to 5 years, and we don’t see any difference in hemodynamic structural deterioration of the valve,” said Stone. “We need longer-term follow-up from comparative randomized trials, though.”