Advanced AFib Patients: Identification, Important Studies, and Ablation Strategies

A Fireside Chat

In this video, Randall Lee, MD, talks with Roderick Tung, MD, and Michael Riley, MD, about identification, important studies, and ablation strategies for patients with advanced atrial fibrillation (AFib). The recording took place at the 2023 Western Atrial Fibrillation Symposium in Park City, Utah. 

Randy Lee: Hi, I am Randy Lee. I am a cardiac electrophysiologist at the University of California San Francisco, and also a part-time medical director at AtriCure. I am here today with Rod Tung and Michael Riley. Can you both introduce yourselves and tell the audience where you are from?

Roderick Tung: Hi, I am Roderick Tung from the University of Arizona College of Medicine.

Michael Riley: I am Michael Riley. I am an electrophysiologist with the Wellstar Health System in the Atlanta area.

Lee: We are here at the Western AFib Symposium in Park City, Utah, and what we are going to do is have an impromptu discussion on some of the highlights from the meeting as well as your views on the posterior wall, including whether it should be ablated and how to ablate. So Rod and Michael, at this meeting there has been a lot of discussion on substrate vs length of AFib ablation. Why don't we start with that? What do you think is important—substrate or duration?

Riley: I think substrate clearly wins out over duration. It is pretty clear. We know that for persistent and long-standing persistent AFib patients, pulmonary vein isolation (PVI) is not enough and more of a role as AFib progresses from arrhythmogenic substrate, particularly posterior wall and other areas.

Tung: I think the STABLE-SR-II study with Minglong Chen very nicely showed that there are those patients with persistent AFib who do well with PVI alone. Everyone has been saying it, but we do not have enough evidence. Clearly, there is a lot of bias in this field and anecdotes of "This is what I have seen.” So it is nice to see with STABLE-SR-II that there is a population that does really well, which makes us question that when a patient is labeled as persistent, it does not mean that all persistents or all paroxysmals are the same. The real question is, how do you define the substrate? There is Utah I-IV, but that is based on magnetic resonance imaging and has not been largely reproducible at other centers. Electroanatomic mapping data is quite variable as well. We are not even sure which rhythm to be mapping it in—do you do the voltage map in AFib or during sinus rhythm? There are so many other questions.

Lee: DJ Lakkireddy presented a subset analysis of the aMAZE trial yesterday in which left atrial (LA) volume is used as a way to delineate patients with persistent AFib who might need more vs just PVI. Do you think that is a reasonable strategy?

Tung: I like it. We have known for so long that those patients who are not going to achieve sinus rhythm have a larger LA, that is even from cardioversion data. So when embarking on a sinus rhythm journey, you are not going to succeed with a really large LA, and there is probably going to be a U-shaped curve for the ones that are enlarged. But the ones that are not enlarged will not need advanced strategies. So it is about finding that sweet spot.

Riley: I agree. LA volume is probably a really good surrogate marker for the amount of substrate changes that develop with time. It is pretty clear that as the LA dilates, PVI oftentimes is not enough.

Lee: In the session that we just left, they were talking about the LAAOS III and aMAZE data. One question that came up is, is it triggers from the appendage, or is it a mass effect? What do you think about the data that DJ demonstrated yesterday, where if you do ligate the left atrial appendage (LAA) with PVI, there was a substantial difference?

Riley:I think data suggests that it is both triggers and probably a total bulk of substrate, but probably more of a role for substrate.

Tung: I agree with what Andrea [Natale] was saying, which is that what you see clinically is you do not need a lot of mass to have AFib. When you isolate a PV, that PV is still fibrillating. You can wall off the atrium, compartmentalize it, and see there is localized fibrillation in one area, and you do not need a lot. So, I think the critical mass theory is not supported by what we see clinically. In terms of it being a trigger, the most difficult thing to dissociate the LAA from is the ridge itself and the vein of Marshall (VOM). With ablating the LAA, Andrea has never been able to show that you need to isolate the whole appendage rather than do more work on the ridge. So how do we know that the VOM is not just more durable left-sided PVI? How do you know that ablating more is reinforcing the ridge and PVI? I think that is the challenge in our understanding.

Riley: Right.

Lee: I totally agree with you.

Riley: The concern that I have as well is that while it is important to electrically isolate the appendage, if that is done without actually closing the appendage, then you lose atrial transport function and the risk of stroke is higher. So I think if you are going to isolate, the LAA should be closed. The data is pretty clear.

Lee: That is a very important point. Andrea’s and Karl-Heinz Kuck’s group clearly demonstrated there is over 20% thrombus formation even while on oral anticoagulation if you are able to isolate the LAA. But let's shift gears. There was a lot of discussion on the CAPLA, ERASE-AF, and CONVERGE trials. You have one, the CAPLA trial, being a negative study. The ERASE-AF trial did not just target the posterior wall, but it was a positive study, and then there was the CONVERGE trial. How do you bring those data sets in terms of your thought process and how you are going to treat your patients?

Riley: First of all, the CAPLA study looked at a very different patient population than the population studied in the CONVERGE trial. There were no patients with long-standing persistent AFib who were included in CAPLA. The average amount of time in AFib was about 5-6 months, with the longest duration of AFib being 11 months. So there were no long-standing patients included, whereas in the CONVERGE trial, 42% of patients were long-standing persistents. I do not think it is fair to look at a study like that doing endocardial posterior wall isolation and equate it with what was done in CONVERGE with epicardial ablation.

Tung: Yes, posterior wall is not posterior wall. Anatomically, you talk about the dome, which is more of the pulmonary venous posterior wall, and then you talk about the vestibule, which is the inferior pulmonary veins down to the coronary sinus. There is a difference in thickness. The convergent procedure mostly works from the vestibule, whereas the endocardial proceduralists work mostly on the dome, which we call the floor. I think that we are just comparing apples and oranges—they are not the same. But when you go through a convergent procedure, you are going to get both and you are going to try to achieve both. But I do not think we can really emulate that because of concerns of atrial-esophageal fistula, etc, on the endocardial side. The thing about CAPLA is it is not a standout in the sense that it is the first study to show it is negative. There was a Korean study and 3 negative ones prior to that, so there are 5 total. I think we can conclude that in all-comers there is no benefit, but that does not mean individual patients do not benefit. That leads us to more tailored therapy and that maybe we need to do it better and it needs to be transmural. Maybe the term “transmural insufficiency” might be the right way to describe a lot of our efforts in the posterior wall, or maybe pulsed field ablation (PFA) will save us. It is hard to know. But I think that to Andrea's point, and his vocal cords are probably bleeding by now, just because you do a box does not mean it is isolated. We know that maybe 30%-50% of these are not durably isolated. So, the intention to treat the gap, between intention to treat and non-treatment, is just too high with current radiofrequency (RF) technologies.

Riley: Right, and if you consider the ERASE-AF trial, the thing that really struck me is for the arm that got more extensive ablation, which underwent PVI and also extensive ablation of low-voltage areas, the recurrence rate was 35%. So where are those gaps? We talk about the LA being a very complex structure in terms of myofibers. I think with endocardial posterior wall ablation, we are missing a lot of the atrium. There is an important emphasis on epicardial-endocardial dissociation, where when you map the endocardial portion, it looks isolated, but then you have recurrences, gaps, and a large part of the epicardium that is not treated with endocardial methods.

Lee: Rod, you have nicely demonstrated the endo-epi asynchrony in terms of activity.

Tung: Yes, we know this has been seen in animal models, and we have seen that interoperatively in clamp models. But now that we have been putting these “McGriddle sandwiches” on for these very highly selected, 3-do, 4-do, 5-do cases, we definitely see evidence that the endo and the epi are often phase-shifted, and often one has more fibrillatory activity than the other. Sometimes we have even seen that you get epi-to-endo block or endo-to-epi block, and maybe you really need to go in and out, and it is intramural re-entry that is really perpetuating AFib.

Riley: Right.

Lee: When you looked at the CONVERGE data, you mentioned 2 things. One, it may be a more reliable lesion in terms of creating transmurality, but then also the vestibule, and I think that is what you were referring to with Andrea. So having said that the study was positive, what patients are you treating or when would you consider it?

Riley: I would say that at our institution, when we see persistent patients who we are going to recommend ablation to, most often we recommend the convergent approach.

Lee: Which patients are you referring to?

Riley: For persistent patients. Now, if they are early persistent patients, I think it is important to have that conversation. We talk about the possibility of doing PVI only for early persistent patients, but talk to them about the results of the CONVERGE trial. We let patients choose what they prefer, but I would say the majority of our persistent and long-standing persistent patients go on to have convergent ablation, so we select a hybrid approach. I would say the majority of our convergent ablations, at the time of the convergent procedure, also have LAA closure as part of that procedure.

Lee: So in your practice, how would you view it now with the results of ERASE-AF and CONVERGE trials? Which strategy would you choose, and why?

Tung: I think the jury is still out with low-voltage ablation. The DECAAF II trial did not show the same results of ablating low voltage. But again, the devil is in the details in terms of how you ablate low voltage. ERASE-AF did a better job in terms of doing it more extensively, so they would box things in and essentially try to isolate the low-voltage area. But remember, STABLE-SR and STABLE-SR-II were talking low voltage and they did not see any major differences in terms of improving success. I do think it is too elementary to believe that AFib is the same as scar-related ventricular tachycardia (VT) and that you can just do substrate modification and the whole thing is over. If you peel it back mechanistically, we do not say that if I modify scar, ventricular fibrillation goes away. And that is fibrillation. One is monomorphic VT, the other one is fib, and that is based on triggers. It is the same in the ventricle as it is in the atrium. But obviously what is perpetuating fib is hard to understand for both. ERASE-AF also has challenges because they did not have complete achievement of randomization, and the reason that is, is that 2 out of 3 patients did not have low voltage. But then when you look at the patients that just got PVI vs PVI and the randomization, there was a nearly statistically significant difference in terms of how they did from PVI on its own. So that is a challenge, because I think randomization, while not intentional, just did not pan out perfectly. I think that the trend is there at baseline for people with PVI vs PVI, those lines should be superimposed.

Riley: In thinking about my patient population, we know that paroxysmal patients do well with vein isolation alone, but when it is more persistent AFib, then we again tend to favor the convergent hybrid approach, if the patients are candidates.

Tung: We tend to see a lot of redo referrals for CONVERGE, which is completely data free. But clearly if you need to do more, then that makes sense. So I would say in our practice, we are seeing a lot of redo AFibs and hopefully we will be able to do a study one day.

Riley: At our institution, we have great collaboration with the surgeons, so we also have a lot of reverse referrals where patients are seen primarily by the surgeon for a surgical issue, like a valve repair or replacement, and that triggers an EP consult at our institution. So they get EPs involved very early in the care of these cardiac patients. If these patients have AFib, then they look to us for recommendations. Oftentimes, they will have posterior wall surgical ablation at the time of their valve surgery or their bypass. We follow the patient closely, see them back several weeks later, and then complete a convergent-like ablation where we go in and do endocardial mapping, complete PVI at that point, and clear up any gaps endocardially.

Lee: I think one thing that the CONVERGE trial clearly delineated is that posterior wall isolation, including the vestibule, does well for patients with long-standing persistent AFib. So let me ask you something. Do you think PFA will obviate the need to do endo-epi ablation? Do you think you will be able to achieve the results of the CONVERGE trial via just endocardial PFA?

Tung: The preclinical data is very promising, but the human data is unknown. That is actually my talk here at the symposium. I think the summary of evidence in the human studies, where we are able to look at the epi after endo PFA, is zero right now. So we do not know the answer. I still believe that the secret sauce to PFA is clearly the sequence, the intensity in terms of how many volts you are using. That special sauce is proprietary for different companies and they still need optimization of that, which is fine. But I think that we still need to look at durability before we even talk about transmurality. You might have endo durability, but not epi. The question is, do you even have durability, which is the first question that needs to be answered. When you look at the manifest data, compared to all of the very early studies, the outcomes are always going to be more sobering in the real world. But it also shows you there is that gap between what you are seeing in those studies. Why is that? Maybe it is a different substrate that is being chosen or maybe a different operator, but you would hope that PFA is the great equalizer between operators. We are not seeing that yet.

Riley: I totally agree. I think there is a lot of excitement about PFA, but there are too many unknowns. I was struck by some of the durability data that was presented here this weekend, where you get rapid isolation of veins, but oftentimes they recur when you look back. So are we going to get durable epicardial isolation with this? It is just too early to know. At the same time, we have an FDA-approved technology, convergent ablation, that treats these patients with known good outcomes.

Lee: That does bring up a point, which is that the CONVERGE trial did lead to FDA approval of the EPi-Sense System (AtriCure) for long-standing persistent AFib, which is the only approval in that class of patients. Going back to the manifest AFib studies, the early results on paroxysmal, persistent, and long-standing persistents seem to mirror what we see with RF and cryo. So I think it sort of alludes to what you are discussing, which is that early operator use as well as algorithms for PFA improve. It is not like ablation catheters, where RF is RF, it is that each company will have a different secret sauce. So I do not think we can generalize it.

Tung: But it will be nice if we want to have comparative studies with vestibular ablation and management, that you could then probably go at it with PFA, because you are not worried about the esophagus, etc. Again, that vestibule is thick. If you reach somewhere in the anatomy, it can be up to maybe 10, 11, 12, and you get close to the annulus. The other thing that we do not talk about is that it takes a higher voltage, and therefore, it might be nerve-sparing at some levels, but maybe the success is also predicated upon some cardiac neuromodulation. When you are ablating the vestibule and doing it epicardially, you are getting all those inferior ganglionated plexi (GPs) as well. We know GPs are controversial, but some studies show that it is very beneficial. In my opinion, there is no question that there is some benefit to some autonomic modulation when you are doing LA ablation.

Riley: There was some data that was a little bit concerning that we heard earlier in the conference about the fact that they are seeing some aortic injuries with PFA and some inflammation in the aorta itself from ablation. So I think there are just some unknowns right now.

Tung: Right.

Lee: Well, I think this covers a lot of the salient points of what has been covered in this meeting. So hopefully next year we can have an update and I look forward to seeing both of you then!

US Indication for EPi-Sense® Coagulation System/EPi-Sense ST™ Coagulation Device

Indications: The EPi-Sense Coagulation System/EPi-Sense ST™ Coagulation Device is intended for the treatment of symptomatic long-standing persistent atrial fibrillation (continuous atrial fibrillation greater than 12 months duration) when augmented in a hybrid procedure with an endocardial catheter listed in the instructions for use, in patients (1) who are refractory or intolerant to at least one Class I and/or III antiarrhythmic drug (AAD); and (2) in whom the expected benefit from rhythm control outweighs the potential known risks associated with a hybrid procedure such as delayed post-procedure inflammatory pericardial effusions. Contraindications include patients with Barrett’s Esophagitis, left atrial thrombus, a systemic infection, active endocarditis, or a localized infection at the surgical site at the time of surgery. Adverse Events: Reported adverse events associated with epicardial ablation procedure may include, but are not limited to, the following: pericardial effusion/cardiac tamponade, pericarditis, excessive bleeding, phrenic nerve injury, stroke/TIA/neurologic complication. Warnings: Physicians should consider post-operative anti-inflammatory medication to decrease the potential for post-operative pericarditis. and/or delayed post-procedure inflammatory pericardial effusions. Physicians should consider post-procedural imaging (i.e. 1-3 weeks post-procedure) for detection of post-procedure inflammatory pericardial effusions. Precautions: Precautionary measures should be taken prior to considering treatment of patients: (1) Deemed to be high risk and who may not tolerate a potential delayed post-procedure inflammatory pericardial effusion. (2) Who may not be compliant with needed follow-ups to identify potential safety risks. To ensure patients undergoing treatment with the EPi-Sense/EPi-Sense ST device are well informed, the benefits, potential risks and procedural outcomes associated with the EPi-Sense/EPi-Sense Hybrid Convergent procedure should be discussed with the patient. Physicians should document accordingly in the medical record. Qualified operators are physicians authorized by their institution to perform surgical sub-xyphoid pericardial access. The coagulation devices should be used by physicians trained in the techniques of minimally invasive endoscopic surgical procedures and in the specific approach to be used. Operators should undergo training on the use of EPi-Sense/EPi-Sense ST device before performing the procedure. Safety and effectiveness of concomitant left atrial appendage closure was not evaluated in the CONVERGE study. Follow-up should be conducted at approximately 30 days postprocedure to monitor for signs of delayed onset pericarditis or pericardial effusion. Rx Only