SAN FRANCISCO — Subclinical leaflet thrombosis rates early on favoring surgical over transcatheter aortic valve replacement (SAVR/TAVR) level out by 1 year, although the long-term impact on clinical outcomes and valve durability remain unclear, a new PARTNER 3 CT substudy suggests.
Overall, 10% of patients had hypoattenuated leaflet thickening (HALT) and reduced leaflet motion at 30 days, which resolved spontaneously in half, even without anticoagulation. At the same time, new HALT appeared in 20% of patients, bringing the 1-year incidence to 24%.
There were no deaths or myocardial infarctions in patients with HALT, although pooled rates of death, stroke, transient ischemic attack (TIA), and thromboembolic events were numerically higher than among those without HALT (day 7 to 365: 8.6% vs 2.9%; P = .11).
“However, given the low total number of events, the data are inconclusive and only hypothesis-generating. Longer-term follow-up and larger datasets will further clarify the impact on clinical outcomes,” study author Raj Makkar, MD, Cedars-Sinai Medical Center, Los Angeles, said at Transcatheter Cardiovascular Therapeutics 2019.
Subclinical reduced leaflet motion in bioprosthetic valves was first identified by the investigator in a 2015 analysis of the PORTICO-IDE study and raised concern after being linked in TAVR patients with higher rates of stroke or TIA in pooled data from the RESOLVE and SAVORY trials and higher rates of cardiogenic shock and death in the MAUDE database.
“The overall rate of clinical events is very, very low, so that’s reassuring,” Makkar told theheart.org | Medscape Cardiology. “But the issues with regard to structural valve deterioration are important for patients because you want to make sure whatever valve you put in is durable. With that in mind, we just have to follow these patients closely.”
The issue of valve deterioration, however, is becoming more controversial as the use of TAVR increases in low-risk patients, who may be younger at implantation.
“It is tempting and attractive to make that connection, especially because this finding reduces leaflet motion, and we all know there are some patients who fail their valve early on and there doesn’t seem to be a plausible explanation, but once again it is speculation at best,” he said.
“The overall message is that even though it is present so frequently, because there is no clear-cut association with life-threatening clinical events, there is no argument for routine anticoagulation,” Makkar said. “Similarly, no argument, no recommendation can be made for routine CTs in patients.”
HALT by the Numbers
The substudy included 408 patients at low surgical risk, in whom 346 CTs were evaluable at 30 days and 312 were evaluable at 1 year. CTs were analyzed by a corelab blinded to patient information and time of CT, with investigators blinded to the results. Clinical events were adjudicated by committee.
Thrombus on bioprosthetic valves can present as HALT with relatively normal leaflet motion, HALT with reduced leaflet motion but normal valve gradients, and clinical valve thrombosis with elevated gradients, Makkar said.
In the per protocol population, the incidence of HALT was significantly higher after TAVR than SAVR at 30 days (13.3% vs 5%; P = .03) but not at 1 year (27.5% vs 20.2%; P = .19). Reduced leaflet motion was present in all patients with HALT, although not all CTs were evaluable for leaflet mobility, he noted.
In patients with reduced leaflet motion compared with those without, mean aortic gradients were significantly higher at 30 days (13.3 vs 11.4 mm Hg; P = .04) but not at 1 year (13.6 vs 12.5 mm Hg; P = .23). Mean gradients were similar between patients with and without HALT at both time points.
As for the clinical consequences, there were four deaths and three MIs by 1 year — all in patients without HALT at 30 days. Clinical valve thrombosis developed in three patients with HALT and one patient without HALT at 30 days, Makkar reported.
Among the 25 patients with HALT at 30 days, HALT resolved in 14 at 1 year without anticoagulation.
Anticoagulation in the Crosshairs
During a press conference, Robert Bonow, MD, Northwestern University, Chicago, questioned the need for anticoagulation after TAVR, noting that there is a class IIa indication for anticoagulation for 3 to 6 months in patients receiving a bioprosthetic surgical valve.
“If we were to move that now into the TAVR realm, to say it’s reasonable to consider anticoagulation for 3 months to 6 months, that might explain not only the early findings but maybe even the change that occurs with time,” he said.
Ajay Kirtane, MD, SM, New York-Presbyterian/Columbia University Irving Medical Center, New York City, said with HALT the “anticoagulation issue is tough because the patients who had it, none of them were on anticoagulation and it resolves. But on the other hand, you could argue it may be more important early than later.”
Makkar countered: “We are looking at an imaging finding and you could give a drug that could eliminate this but at what expense? So the risk/benefit ratio of a therapy is best studied in a clinical trial dedicated to that particular question and not guided by imaging findings.”
He suggested that microdamage may be occurring to the leaflet during crimping and TAVR valve preparation and that the procedure may also alter flow dynamics in the aortic root.
Makkar also pointed out that the GALILEO trial was stopped early last year after a higher risk for death, thromboembolic events, and bleeding was identified in older adults treated with rivaroxaban (Xarelto, Bayer/Janssen) rather than antiplatelet therapy after TAVR.
Fellow PARTNER 3 investigator Michael Mack, MD, The Heart Hospital Baylor, Plano, Texas, said the substudy was “mandated by the FDA to answer this question about valve thrombosis, and I think it’s gone a long way toward doing that, but not completely answered it.”
Mack said he wasn’t surprised by the higher incidence of valve thrombosis with TAVR or the low incidence of clinical events, but was struck by the percent of HALT that resolved without anticoagulation. He argued that routine oral anticoagulation isn’t warranted after TAVR and took it even one step further.
“Now that we have routine surveillance of surgical valves, for the first time I think it calls into question the IIa recommendation for 3 months of anticoagulation,” Mack said. “It’s what we’ve always done and will probably stop doing on the basis of this.”
Commenting further, Mack said: “The other shoe that hasn’t dropped is its long-term effect on structural valve deterioration. I do think that early HALT does explain premature structural valve deterioration. Patients in both surgery and TAVR who come back 2, 3, 4 years later with a valve that isn’t working, we always question why. I think this is the explanation of why, but again it doesn’t justify routine anticoagulation.”
Makkar reports consulting for and research grants from Edwards Lifesciences, Abbott, Medtronic, and Boston Scientific.
Transcatheter Cardiovascular Therapeutics (TCT) 2019. Presented September 27, 2019.