MAUDE: Post-TAVR Leaflet Thrombosis Not Benign

Patrice Wendling

September 08, 2017

BARCELONA, SPAIN — Already on physicians' radar for a link with stroke or transient ischemic attack (TIA), a new study suggests leaflet thrombosis after transcatheter aortic-valve replacement (TAVR) is also associated with cardiogenic shock and death[1].

Among 5691 TAVR-related adverse events in the Food and Drug Administration's MAUDE database, researchers identified 30 cases of structural valve dysfunction due to leaflet thrombosis.

Despite a mix of pharmacologic and surgical interventions to address the thrombosis, nine patients died, three had stroke/TIA, and two developed cardiogenic shock.

Of the 30 thrombosis cases, seven had the diagnosis established at surgical valve explantation or autopsy.

"While these data do not provide the frequency of these events, and undoubtedly the clinical events often led to the recognition of leaflet thrombosis, they're useful in that they establish definitive evidence of its occurrence," study author Dr Ankur Kalra (University Hospitals Cleveland Medical Center, OH) said at the European Society of Cardiology (ESC) 2017 Congress. The report was published simultaneously in Structural Heart.

Concerns were raised in 2015 about subclinical leaflet thrombosis in bioprosthetic aortic valves but took on added steam after a recent analysis in 890 patients in the RESOLVE and SAVORY registries reported the phenomenon is far more common in transcatheter than surgical aortic valves (13.4% vs 3.6%) and in patients on dual antiplatelet therapy (DAPT) than on oral anticoagulation (14.9% vs 3.6%).

During a discussion of the results, Kalra observed that not all hypoattenuated leaflet thickening actually manifests as an increase in valve gradient or clinically as heart failure or stroke/TIA but that leaflet thrombosis remains a very important question for the field, particularly as TAVR moves to the intermediate- and lower-risk patient.

"Moving forward, I think it's an important question to answer with regard to how to manage patients post-TAVR; whether all-comers should be initiated on anticoagulation therapy or whether we should have risk-stratification models based on how the valve was deployed, patient characteristics, and certain procedural characteristics that can help us elucidate what would be the optimum management strategy for these patients," he said.

Commenting for theheart.org | Medscape Cardiology, Dr James McCabe (University of Washington Medical School, Seattle), coauthor of an accompanying editorial[2], said, "Until the incidence is known, it will be important not to throw the baby away with the bathwater," observing that the introduction and study of TAVR has been "exceptionally methodical and rigorous" with aggregate data that is really impressive.

"The totality of the data on subclinical and clinical valve thrombosis bears consideration to be sure, but I don't think this MAUDE analysis moves the needle too much other than to support the notion that careful, routine post-TAVR follow-up is needed to ensure that early signs of valve dysfunction—say an elevated gradient—are carefully assessed and/or managed before things take a turn for the worse," McCabe said.

"When it comes to post-TAVR anticoagulation or even DAPT, this is a space where randomized data will be particularly valuable, as observational data have significant issues with confounding by indication," he added.

Among the 5691 TAVR-related adverse events identified in the MAUDE database between January 2012 and October 2015, 156 events were listed as structural valve dysfunction and 30 of these were directly attributable to leaflet thrombosis.

Most cases presented as aortic stenosis (53.3%), regurgitation (23.3%), or both (13.3%), with the remaining three patients having stroke/TIA as the presenting symptom.

Notably, 60% of leaflet-thrombosis cases occurred in the first year following TAVR, but 40% of events occurred later, even 5 years after implantation.

Leaflet thrombosis was managed with escalation of antiplatelet or anticoagulant therapy in eight cases, valve-in-valve TAVR in three, and surgery in 14, but also with diuretics in one, thrombus aspiration in one, balloon aortic valvuloplasty in two, and no intervention in two.

"I still think that we are learning how to diagnose leaflet thrombosis but more importantly, we don't know how to manage them, and that was clearly evident from the database," Dr Abdallah Al-Mohammad (Sheffield Teaching Hospital, UK) commented in an interview. "Many people are being operated on, but we don't know why they're being operated on, and we need to look into those more carefully to see whether anticoagulation might help either before or when we diagnose it."

Kalra and colleagues acknowledge that the study has several limitations, including the potential with the MAUDE database for ascertainment bias, under- and overreporting of events, and a lack of information on management decisions between the initial adverse event and final intervention.

Still, "clinical or symptomatic leaflet thrombosis following TAVR should not be regarded casually as a benign event and needs to be addressed seriously," they conclude.

In their editorial, McCabe and Dr Mark Reisman (University of Washington Medical School) argue the registry is likely enriched for the most severe adverse events and go on to draw parallels between the attention garnered by leaflet thrombosis and the infection of 18 people with the H5N1 bird flu in Hong Kong in 1997.

The MAUDE analysis helps clinicians understand that clinically overt valve thrombosis exists and "is not simply a matter of thromboembolic complications but can present with valve dysfunction or even cardiogenic shock," they write. "Yet it remains unknown if leaflet thrombosis is the pandemic scourge by which TAVR may run afoul."

Kalra reports no relevant financial relationships. Disclosures for the coauthors are listed in the paper. McCabe and Reisman report no relevant financial relationships.

Follow Patrice Wendling on Twitter: @pwendl. For more from theheart.org, follow us on Twitter and Facebook.

Comments

3090D553-9492-4563-8681-AD288FA52ACE
Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as:

processing....