AT ACC 17
WASHINGTON (FRONTLINE MEDICAL NEWS) – Subclinical leaflet thrombosis, as evidenced by reduced leaflet motion and corresponding hypoattenuating lesions on high-resolution CT, occurs frequently in bioprosthetic aortic valves implanted either by the transcatheter route or surgically, according to a report presented at the annual meeting of the American College of Cardiology and simultaneously published March 19 in the Lancet.
In an observational cohort study involving 890 patients enrolled in two single-center registries, reduced leaflet motion signaling the presence of thrombosis was detected in 106 patients (12%) by four-dimensional, volume-rendered CT. The thrombosis was hemodynamically silent and undetected by transthoracic echocardiograms done in a large subgroup of the study participants. Mean aortic valve gradients were numerically higher in affected patients, but still fell within the normal range in most of them, said Raj R. Makkar, MD, of Cedars-Sinai Heart Institute, Los Angeles.
Dr. Makkar and his associates studied this issue by analyzing data in the RESOLVE (Assessment of Transcatheter and Surgical Aortic Bioprosthetic Valve Thrombosis and its Treatment with Anticoagulation) registry at Cedars-Sinai and the SAVORY (Subclinical Aortic Valve Biosprosthesis Thrombosis Assessed with Four-Dimensional CT) registry at the Rigshospitalet Heart Center in Copenhagen. Their study focused on enrolled patients who underwent either TAVR (626 patients) or SAVR (264 patients) and then had high-resolution CT scans done at various times after implantation. All the CT scans were analyzed in blinded fashion, and the study participants were followed for a mean of 540 days.
The prevalence of subclinical leaflet thrombosis was significantly higher in TAVR valves (13%) than in SAVR valves (4%). The severity of the thrombosis also was significantly higher in TAVR valves, when measured by the extent of leaflet motion restriction (71.0% vs. 56.9%).
One possible reason for this discrepancy is that traumatic injury to the pericardial leaflets, which theoretically could predispose to thrombus formation, may be more likely to occur during crimping and deployment of balloon-expandable and self-expanding stent valves in TAVR. Another possibility is that resection of the calcified native aortic valves during SAVR might improve flow dynamics after valve replacement, compared with leaving native aortic valve cusps in situ during TAVR. A third possibility is that inadvertent incomplete expansion or overexpansion of transcatheter valves might alter mechanical stress on the leaflets, predisposing them to thrombus formation, compared with the uniform expansion of valves placed surgically, the investigator said.
“Nevertheless, these findings should be interpreted in the context of findings from multiple randomized controlled trials showing similar mortality and stroke rates, better hemodynamics, and equivalent durability of transcatheter aortic valves at 5 years, compared with surgical valves,” he noted.
Patients taking novel oral anticoagulants (NOACs) or warfarin for anticoagulation before valve replacement were less likely to develop subclinical leaflet thrombosis than patients not taking those anticoagulants. And the thrombosis appeared to resolve in all patients treated with NOACs or warfarin for 30 days after it was found on CT, but it persisted or progressed in those not treated with anticoagulation. Thus, “anticoagulation with NOACs or warfarin was effective in prevention and treatment of reduced leaflet motion, but dual-antiplatelet therapy, which is the standard of care, was not,” the investigators said (Lancet 2017 Mar 19. doi: 10.1016/S0140-6736(17)30757-2 .
“Our study challenges the American College of Cardiology and American Heart Association and European Society of Cardiology and European Association for Cardio-Thoracic Surgery guidelines, which recommend dual-antiplatelet therapy after TAVR and do not recommend routine anticoagulation after TAVR,” he noted.
However, routine anticoagulation for all TAVR and SAVR patients, or even for all who develop leaflet thrombosis, cannot be recommended on the basis of results from a single observational study. The benefit of anticoagulation may not offset the risk of bleeding in this predominantly elderly population with multiple comorbidities, he said.
There were no significant differences between patients who developed subclinical leaflet thrombosis and those who did not regarding rates of death, MI, or stroke. However, “reduced leaflet wall motion was significantly associated with increased rates of all TIAs [transient ischemic attacks], nonprocedural TIAs, and post-CT TIAs.” Dr. Makkar said.
Based on his new findings, Dr. Makkar suggested that clinicians keep possible leaflet thrombosis on a replaced aortic valve in mind if they see clinical evidence for it, such as an increase in the valve’s pressure gradient or development of a TIA or stroke. These events should trigger further investigation of the valve with CT angiography, and if thrombosis is then confirmed, the patient should receive treatment with an anticoagulant, although the appropriate duration of anticoagulant treatment remains unclear, he said in a video interview . Dr. Makkar also said that trials are needed to determine whether routine CT angiography assessment or routine prophylactic treatment with an anticoagulant is warranted because the antiplatelet therapy that patients currently receive after aortic valve replacement seems unable to prevent leaflet thrombosis.
Since this was an observational study based on data from nonrandomized registries, the findings cannot prove causality but only an association between leaflet thrombosis and TIA. They must be substantiated in further studies, including “the current Food and Drug Administration–mandated imaging substudies in randomized clinical trials.”
Dr. Makkar has received consultant fees and/or honoraria from Abbott Vascular, Medtronic, Cordis, and Entourage Medical, and research grants from Edwards Lifesciences and St. Jude Medical.
Mitchel L. Zoler contributed to this report.