Possible Clots on Bioprosthetic Aortic Valves: Potential Roadblock for Structural Cardiology

John Mandrola


October 09, 2015

A provocative study[1] published in the New England Journal of Medicine this week infused a dose of caution into the field of structural cardiology.

"Extremely important," is how two experts described the finding that subclinical thrombosis may cause decreased mobility of bioprosthetic aortic-valve leaflets in a not-so-small number of patients.[2]

The NEJM Study:

Detailed coverage of the study can be found on heartwire on Medscape. Here is the short story:

Dr Raj Makkar (Cedars–Sinai Heart Institute, Los Angeles) and colleagues from two other groups performed 4-D CT scans on a series of patients implanted with various models of bioprosthetic valves. Their study cohort included 55 patients from the PORTICO IDE trial and 132 patients from two physician-initiated registries with surgical or transcatheter aortic valves.

The advanced imaging technique provided investigators a far more detailed look at valve function than has been available before.

They observed decreased leaflet mobility in 22 patients (40%) from the PORTICO trial and 17 patients (13%) from the registries. Transthoracic echocardiograms (performed in a subgroup of patients) did not detect any differences in gradients. Leaflet-mobility issues were seen in multiple prosthesis types. Patients with therapeutic anticoagulation had a far lower prevalence of leaflet mobility, and anticoagulation mostly resolved the leaflet-mobility problem.

These findings, along with the presence of abnormal opacities on the corresponding leaflet, led researchers to the logical assumption that subclinical thrombosis was the cause of the problem.

As for outcomes, there were no between-group differences in patients from the clinical trial, but in the pooled registries, patients with reduced mobility had a higher incidence of stroke and transient ischemic attack than those with normal leaflet motion (three of 17 [18%] vs one of 115 [1%], P=0.007). The low number of events make it impossible to know whether this is clinically significant.


This could be big news.

Medical dogma has long held that tissue valves—because they are tissue and not metal—do not require anticoagulant therapy. This is the essence of their advantage over metal valves. Patients faced with valve replacement make the trade-off of less durability with tissue valves because they are told they won't need warfarin. Add to that, tissue valves can now be placed via a catheter.

Let's also think structurally: decreased leaflet mobility partially obstructs blood flow, which then creates turbulence. (Think of putting your thumb on a garden hose.) Turbulent jets may lead to faster degeneration of the leaflets. It's worth remembering an aortic valve opens and closes 100,000 times each day.

The best word to describe these findings is a signal—as in a hint or a clue. This is a small study with three different data sets. It's hard to dispute the FDA perspective,[3] which was written alongside the study and editorial in the journal, that the "benefits of using these devices for the currently approved indications continue to outweigh the risks."

An important facet of this study was the imaging. Wall Street Journal reporter Tom Burton asked on Twitter: "How did docs put thousands of pig and cow heart valves in people for 30 years and not know clot risk?" That's easy to answer. We could not have known because we didn't have 4D imaging from CT scans. All we had were echocardiograms, and in this study, standard echo showed no problems. I asked the chest radiologist at our hospital whether we do 4D CT scanning. She said, no way, that's a research tool and is not routinely available.

The imaging issue is key. Namely, are these aberrations on software-created images clinically meaningful? We've had a 3-decade experience with tissue valves, and I don't know of any increase in stroke risk with bioprosthetic valves—and the incidence of transcatheter-valve thrombosis is rare.

In their editorial, Dr David Holmes (Mayo Clinical, Rochester, MN) and Dr Michael Mack (Baylor Scott and White Health, Plano, TX) emphasize the uncertainty that arises from these novel observations. They say more data are urgently needed and the uncertainty will be best resolved with scientific rigor.

You can bet there will be intense investigation of this matter. That's a good thing, because it's not hard to feel the momentum building toward using transcutaneous valves in lower-risk patients.

I recently spoke with a young trainee from a structural training program. "If I needed an aortic valve, I'd take a [transcatheter aortic-value replacement] TAVR, and when it wears out, just do a valve-in-valve second TAVR." His tone, confident, matter-of-fact and fearless, fit the we-can-fix-anything cardiology phenotype perfectly.

I can't help but see it differently. These new findings would surely figure in my decision. After looking at those images, my first thought would be how much longer I could live with my own valve. What is the risk of not having a pig, cow, or metallic valve put in my body?

Then I would consider the issue of anticoagulation. If future studies show clots impair bioprosthetic valves, I might favor the longer-duration mechanical valve.

Stay tuned. This is only the beginning. Until we know more, this seems like a great time to introduce slow medicine to the field of structural cardiology.

Let's close on a positive note. I've been a skeptic of overimaging and overquantifying the human condition, but in this case, it's likely that knowledge gained from technology will lead us to better patient care.



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.