TAVR Embolic Protection: A Safety Net in Need of More Data

Michael J. Mack, MD; Sanjit S. Jolly, MD, MSc


November 11, 2016

This feature requires the newest version of Flash. You can download it here.

Michael J. Mack, MD: Good afternoon. My name is Mike Mack. I am a cardiac surgeon at Baylor Scott & White Health in Dallas.

Sanjit S. Jolly, MD, MSc: I am Sanjit Jolly, an interventional cardiologist and associate professor at McMaster University in Canada.

Dr Mack: We are here at Transcatheter Cardiovascular Therapeutics (TCT) 2016, where there have been some exciting late-breaking clinical trials. One was the SENTINEL trial,[1] which looked at the role of neuroprotection in patients undergoing transcatheter aortic valve replacement (TAVR). Before we talk about its results, Sanjit, could you tell us about current practice in Canada of TAVR vis-à-vis surgical aortic valve replacement (SAVR) and how decisions are made?

Dr Jolly: Our institute has a heart team approach with interventional cardiologists, cardiac surgeons, and sometimes geriatricians. We look at the whole patient, including frailty and surgical risk, and decide who is best suited for TAVR. One of the challenges, of course, is that it is expensive, and in the public healthcare system there are only a certain number of cases that are funded per year. We want to offer it to the patients who really need it; we cannot offer it to everybody.

Stroke Risk With TAVR

Dr Mack: Are you concerned about stroke in patients undergoing TAVR? Do you think there is any difference between SAVR and TAVR for stroke risk?

Dr Jolly: There was a concern in the initial PARTNER trial[2] that TAVR may be disadvantaged regarding periprocedural stroke and earlier up-front risk. We have not seen many strokes at our institute. The thought is that as valves get better, perhaps there is less periprocedural risk. The question is: What are strokes being caused by further down the line? Is it the vascular risk of the patient, or are we seeing thrombus on the leaflets? These have been picked up asymptomatically on CT in some recent publications.[3]

Dr Mack: I totally agree. At the current time, we have to say that the risk for stroke is probably equivalent between SAVR and TAVR, but very low in both. I think it is directly related to how closely you look afterwards, whether it is a neurologist examining the patient or whether it is you or me examining the patient. The more you look, the more subtle neurologic findings you find.

Despite the clinical rate being very low, we know that if you do diffusion-weighted MRI on all patients, the overwhelming majority have evidence of embolization to the brain. If you can decrease the number of embolic lesions to the brain as detected by MRI, then perhaps there would be better clinical outcomes. It is a bit of a clinical conundrum when most patients do not have stroke symptoms but have imaging findings. Does it really correlate with clinical outcomes? Do you have any sense of that?

Dr Jolly: It would make sense that if you had a large enough trial, you [might] prevent those [strokes detected by] diffusion-weighted MRI. You look at the overall cost of the therapy. If some type of cerebral protection device were simple to deploy, fairly cost-effective, and, say, it costs what a drug-eluting stent costs these days (which is not very much), then this may be an important therapy. It is kind of like the seatbelt on the car: You do not need it very often, but when you do, it can make a difference.


Dr Mack: Right. That is a great analogy.

The SENTINEL trial was a randomized trial of cerebral protection using the Claret Sentinel™ capture device (Claret Medical; Santa Rosa, California) which is put in via the right brachial artery. It has two filters: one which sits in the innominate artery, and the other which sits in the left carotid artery. It does not protect the whole brain. The vascular distribution of the left vertebral artery is not protected. The trial was randomized 2:1 for protection versus no protection. Do you want to give a summary of the results that were presented this morning?

Dr Jolly: Absolutely. The primary outcome was diffusion-weighted MRI volume of defects. There was not a significant reduction [of defects], so the study did not meet its primary outcome. When the investigators looked at clinical stroke rates, there was a numerical difference. In the control arm, it was somewhere between 9% and 10%. That was not statistically significantly different. This trial was not powered for clinical outcomes.

They did adjust for a multivariable model. What was interesting is that they looked at the stroke rates based on the different valves: Sapien S3 (Edwards Lifesciences; Irvine, California) versus the CoreValve® Evolut® R (Medtronic; Minneapolis, Minnesota) versus Sapien XT. They actually showed that there was significant heterogeneity. It may be that stroke rates are much lower with the newest valve, making this less of an issue. What is your perspective?

Dr Mack: There is certainly some validity to it. As I have been involved in this field and involved in the neuro ARC (Academic Research Consortium) initiative, which establishes standardized endpoints for clinical trials regarding neurologic findings, I have been increasingly impressed by how much is not known in this field. For example, we know that we have diffusion-weighted lesions by imaging findings. How significant are they? We know that they cannot be good for you. How bad are they and how do you demonstrate how bad they are? Do we measure volume of lesions or number of lesions? Or number of lesions over a certain size?

We also learned that there is significant heterogeneity in terms of detection. Which scanner is being used: 1.5T or 3T? This study used a 3T scanner. Maybe it was too sensitive and picked up very small lesions, therefore obscuring a benefit. There seemed to be a trend toward benefit from clinical stroke, although it was not significant. I think further information will be gleaned from this, including neurocognitive testing and [risk for] delirium afterwards.

I also think that these patients received general anesthesia. We have now gone to a minimalist procedure where patients get conscious sedation. We know that anesthesia has an effect on neurocognitive outcomes. It is going to be interesting [evaluating] that. Although it is somewhat disappointing that SENTINEL did not reach its primary endpoint, some information indicates that there may be a clinical benefit to it. And other trials are ongoing right now. What are your thoughts?

Dr Jolly: I agree with you. I would like to see a significantly larger trial. It would be a very important advance if we were able to reduce [stroke risk]. They reduced rate of stroke by about 50%. If that is a true finding, that is very important. Obviously the trial was too small to show that. If it is positive, we should use the therapy.

Dr Mack: This was another extremely interesting trial here at TCT 2016. As a recent publication in the New England Journal of Medicine[4] explains, just because you have a negative trial does not mean that a lot of valid information will not come from it. There will be a lot more forthcoming information from this.

Thank you for joining us. We hope you found this TCT 2016 session informative.


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.