COMMENTARY

Dec 08, 2023 This Week in Cardiology Podcast

John M. Mandrola, MD

Disclosures

December 08, 2023

Please note that the text below is not a full transcript and has not been copyedited. For more insight and commentary on these stories, subscribe to the This Week in Cardiology podcast, download the Medscape app or subscribe on Apple Podcasts, Spotify, or your preferred podcast provider. This podcast is intended for healthcare professionals only.

In This Week’s Podcast

For the week ending December 08, 2023, John Mandrola, MD comments on the following news and features stories.

Brazilian EP Society Meeting

Before I start, I want to say a few words about the Brazilian Electrophysiology (EP) Society meeting in Rio. Wow. What a meeting. What a country.

Thank you to SOBRAC and Dr Eduardo Saad for the kind invitation and amazing hospitality.

I am struck by the strength of EP being done in Brazil. I was speaking with very young EP docs there and while I was going on about boring-old pulmonary vein isolation (PVI), they were telling me of their epicardial ventricular tachycardia (VT) ablations related to the prevalence of CHAGAS cardiomyopathy.

Also, a short note to listeners in Porto Alegre. Thanks for listening. If I make any flubs today, it is because I have done so many ablations this week. Yes, I do a podcast, but I also do hundreds of ablations and devices per year. We don’t have CHAGAS cardiomyopathy but we do have a lot of ischemic VT that requires ablation.

AF Guidelines – AF Ablation Gets Class 1A for HF

Two days before my debate last week with Luigi Di Biase on atrial fibrillation (AF) ablation in patients with heart failure (HF), the authors of the 2023 ACC/AHA guidelines published their new update. It’s a massive 156-page document.

There’s no way to cover it all, but let’s comment on some particulars and new recommendations.

First, the writers have upgraded to Class 1A the use of AF ablation in patients with HF with reduced ejection fraction (HFrEF). This is in section 9.2. Notably, the 2019 guidelines had AF ablation in HF as a Class 2b. Here is the statement in the colored box:

In appropriate patients with AF and HFrEF who are on GDMT (guideline directed medical therapy), and with reasonable expectation of procedural benefit, catheter ablation is beneficial to improve symptoms, QOL, ventricular function, and cardiovascular outcomes.

My Comments. Let’s do four chapters:

Chapter 1: I have content expertise here.

  • My approach to most patients with AF is as a psychologist. I reassure, remove fear, and make clear that we will help them, but we can go slow and often treatment does not require major procedures. That is most patients.

  • But there are occasional patients with AF who require prompt and aggressive action. These are the patients with AF and HFrEF. These are often men, often ambulatory from the HF clinic. They have a bad ventricle and are ok on GDMT. Then, AF hits, and they decline. In these cases, aggressive action to restore rhythm is appropriate. And AF ablation is often a core way to do that.

  • Another fraction of patients with HFrEF are those patients who present with AF of a month or so duration, and who have tachy-mediated cardiomyopathy. These patients, too, need rapid and aggressive therapy, first to stabilize with rate control (lots of digitalis), and then consideration of rhythm control. The thing is for every one of these patients, I see 10 patients with AF who can be treated slowly and with giving peace a chance.

Chapter 2:

  • Let’s talk about the modifiers in the recommendation. The first one is “appropriate” to modify patients. What does this mean? In my lecture in Brazil, I showed a number of studies that document co-morbidities in patients admitted with the diagnosis of HF. The take home is that the vast majority of patients with HF have many co-morbidities.

  • In one sample of American inpatient diagnosis, more than 80% of HF admissions had a high Charleson Co-morbidity Index. Contrast that with the Table 1s from the HF and AF ablation randomized controlled trials (RCTs).

  • In CASTLE AF, the mean age of patients was 62 years; same with CASTLE HTx and AATAC. Young male patients with minimal comorbidities. The point is that the supporting trials enrolled young ambulatory men. But that’s most of the HF I see.

  • Another modifier was “a reasonable chance of success.” I admit this is a grey area, but what I take from it is to try and emulate trials. We see a bunch of older patients with dilated left atriums, longer duration AF, fine fib waves on ECG, left ventricular (LV) scar.

  • My friends, resist the urge to ablate these patients. PVI with or without posterior wall isolation or roof lines is unlikely to work.

  • Another common group here in the southeastern United States is the morbidly obese. These patients also have such electrical and structural disease, PVI is unlikely to work.

  • Now, my friend, Luigi Di Biase will say you have to do more than PVI. To this, I say, yes, I understand, but can you show me any data that refutes STAR AF 2? Can you tell me where I should do said extra ablation? The answer is that Luigi can tell me about olive oil and Primitivo, but I don’t think he (or anyone) can say where else to ablate outside the PVs for patients with advanced AF.

Chapter 3: In this chapter I want to argue with the writers about this level of recommendation. The highest level.

  • The writers will support a class 1A because their guideline recommendation says it is specifically to improve symptoms, quality of life (QOL), ventricular function, and cardiovascular (CV) outcomes. They will say there are multiple RCTs and three meta-analyses supporting the recommendation. Which is true, but…..

  • CASTLE AF dominates most of the meta-analyses, and it was a positive trial. It found a 38% reduction in the primary endpoint of hospitalizations for HF and mortality. Yet, as much as I love Nassir Marrouche, the trial has limitations.

  • It was unblinded, which raises the possibility of performance bias.

  • It was underpowered due to early termination.

  • There was substantial missing data. And it was uneven, with more patients lost to follow-up in the active arm vs control arm.

  • The primary analysis was based on a small number of events.

  • CASTLE AF investigators screened more than 3000 patients to enroll 360 patients. So, in this highly selected population, there are many reasons to be cautious when we are translating the evidence at the bedside.

  • Supporting a cautious view of CASTLE AF were the non-significant results of the RAFT-AF trial. RAFT was a Canadian trial of about 400 patients with HF and AF; AF ablation-based rhythm control vs rate control, which could include AV node ablation (AVNA) and pacing.

  • Nearly one in three patients in the rate control arm had AVNA and pacing—and this was before conduction system pacing (CSP).

  • This trial was terminated early due to futility, and the end result was nonsignificant for the primary endpoint of death or HF events. Though it did get close to significance, there was only a difference in 14 primary outcome events in an unblinded trial.

Another problem with the meta-analyses, is that when you apply the Cochrane tool (for individual studies) and GRADE Framework (overall body of evidence) to assess risk of bias to the trials, you find that all the trials are at high risk of bias, due primarily to their unblinded nature.

So, in the end, even though there are three meta-analyses, a neutral assessment of the evidence would lead to an overall low confidence.

In sum, I know why the authors choseClass  1A, but I think the components of the evidence warrants more caution than the highest-level recommendation. A 1A-recommendation may tempt over-enthusiastic ablation-docs to ignore the important modifiers “appropriate patients” and “reasonable chance of success.” My friends in EP, please don’t go overboard with AF ablation.

Chapter 4. I want to close this discussion with a note to my HF colleagues. One thing that I sense, and see sometimes, is this barrier between HF and EP. I am often referred patients with HF and AF late in the course. There is this great push to put every every patient with HF in the we-must-get-them-on-the-big-four-drugs category. Well, that is fine, but

  • Sometimes the problem with HF is AF.

  • Sinus rhythm is the best HF therapy.

  • We should always remember that most of the HF trials had either no or few patients with AF, and the Kotecha meta-analysis that found beta-blocker therapy reduced mortality in patients with HF who were in sinus rhythm but had no effect for those in AF.

So please, to my friends in HF, get us EP docs involved early when there is AF (or left bundle branch block). Sinus rhythm is sometimes the best treatment for HF, far better than beta-blockers and sacubitril/valsartan.

AF Guidelines – EARLY Rhythm Control Gets Emphasis

The number 5 Top Take Home messages reads:

 With the emergence of new and consistent evidence, this guideline emphasizes the importance of early and continued management of patients with AF that should focus on maintaining sinus rhythm and minimizing AF burden.

This is based on the EARLY AF ablation trials with cryoballoon (CB) and EAST AFNET 6. I’ve discussed these many times, so I will be brief. The STOP AF and EARLY AF ablation trials were both CB vs antiarrhythmic drugs (AAD) in patients with new AF.

  • Both trials showed CB is better than drugs. But the endpoint was AF episodes, not outcomes.

  • I strongly believe we should be going much slower in patients with relatively new AF. Reassure, wait, treat risk factors. Lots of new and short duration AF will resolve over time and removal of triggers.

  • I am not saying wait years, and I am not saying wait for paroxysmal AF to become persistent AF, just resist the urge to go to early AF ablation. Give peace a chance.

  • The EAST AFNET trial has important limitations. This showed that an early rhythm control (ERC) strategy vs a conservative mostly rate control strategy in patients with new AF led to lower CV outcomes.

  • Of importance, EAST AFNET was not an ablation trial. Less than 20% of the ERC strategy had ablation. While EAST AFNET was positive, there is a serious concern for performance bias because there were far more interactions in the ERC group.

  • What’s more, 60% of the conservative arm was in sinus rhythm at one year vs 80% in ERC. How, my friends, does a delta of only 20% more patients in sinus rhythm drive a reduction in CV outcomes? Answer: it can’t. If it was 90/10 or 80/20, maybe but that’s almost no difference in sinus rhythm.

AF Guidelines – Dubious Upgrade of Percutaneous LAA Closure/Occlusion

In point 8 of the top 10 take-homes, the authors say this:

Left atrial appendage occlusion (LAAO) devices receive a higher level Class of Recommendation: In view of additional data on safety and efficacy of left atrial appendage occlusion devices, the Class of Recommendation has been upgraded to 2a compared with the 2019 AF Focused Update for use of these devices in patients with long-term contra indications to anticoagulation.

Let me go to their text in section 6.5.1. They write that,

RCTs have demonstrated percutaneous LAAO to be noninferior to warfarin and DOACs for stroke and systemic embolism with a reduced risk of major bleeding. Citations 1-4

This reads as if it is obvious and true. But it is not exactly true. And I want to be specific on the citations.

  • Citation 1 is the PROTECT AF trial. Yes, it found LAAO was non-inferior to warfarin. But it did not pass US Food and Drug Administration (FDA) muster because of many internal validity issues. FDA mandated a second trial, called PREVAIL.

  • Citation 2 was PREVAIL: and I will read you directly from their 5-year paper so you don’t think it is me being biased. “The 5-year outcomes of the PREVAIL trial revealed that: 1) for the first coprimary composite endpoint of stroke, systemic embolism, or cardiovascular/unexplained death, LAAC did not meet the noninferiority endpoint.”

  • While it is true that the second co-primary endpoint of stroke and systemic embolism excluding the 7 days after the procedure met noninferiority (NI), it was dubious because, a) what patient ever can exclude the 7 days after a procedure? And, b) it met NI only on risk difference, not rate ratio (RR). And the upper bound of the RR was nearly 5, meaning you would have to allow that LAAO was 5 times worse than warfarin.

  • Citation 3 was the combined PROTECT and PREVAIL meta-analysis. This showed no difference in stroke or bleeding compared with warfarin, but of course, meta-analyses are not NI trials, and the results in PROTECT and PREVAIL were widely divergent with many more ischemic strokes in PREVAIL.

  • Citation 4 was the PRAGUE 17, LAAO vs direct oral anticoagulants (DOACs). Yes, it found it LAAO NI to DOACs, but the authors combined stroke and systemic embolism (efficacy endpoints) with bleeding (safety endpoints) and when you combine endpoints expected to go in opposite directions, they cancel each other out and NI is reached. This is not how NI trials are supposed to work.

  • I know I have said this before, but it’s worth re-iterating that the entire foundation of this procedure rests on shaky grounds.

Now to the “new” upgraded recommendation of 2A. They say this is because of additional data.

In the discussion of this point, they cite an EWOLUTION registry study. This is deeply flawed because EWOLUTION was a voluntary registry wherein patients were included at the discretion of the investigators. My gosh, do you think investigators include their worst cases?

They also cite the great new PINNACLE FLX study. This was an investigation device exemption study looking at the results of the newest Watchman Flx device.

  • In this study of 400 patients, 12 strokes occurred in the first year. That’s 3% incidence. Compare that to 1.7% per year in ROCKET AF, a similar CHADSVASC cohort.

  • Also, the rate of major bleeding was 7.9% at 12 months. Contrast that with a rate of 3.6% per year in ROCKET AF with rivaroxaban. My friends, I am not convinced that Watchman Flx is the answer.

  • The authors have not cited the recent 12-month data from the SWISS APERO study which I mentioned previously.

  • This Amulet vs Watchman study found that 50% of both groups had patency of the appendage by CT scan. Here is the translation: Half of the devices put in, in experienced centers, failed to completely close the appendage. Peri-device leaks are terrible because they increase the risk of stroke. And you wonder why ischemic strokes were so much higher in PREVAIL.

One final comment on the second recommendation in the LAAO section. They write:

In patients with AF and a moderate to high risk of stroke and a high risk of major bleeding on oral anticoagulation, percutaneous LAAO (pLAAO) may be a reasonable alternative to oral anticoagulation based on patient preference, with careful consideration of procedural risk and with the understanding that the evidence for oral anticoagulation is more extensive.

I don’t know what to say about this. The two additional citations include a non-controlled registry from 2013 and EWOLUTION, which I’ve explained is deeply flawed.

  • The idea of telling a patient that pLAAO is a reasonable alternative to OACs is nothing short of egregious. Or outstandingly bad. There is no convincing data for that.

  • Using patient preference sounds nice, but too often I am sent cases, or hear directly from patients, that patient preference is a coercion technique.

  • I worry that our field is going down a precarious moral road here. I am not religiously opposed to LAAO. I have referred patients who have electrically isolated LAA and had a stroke on OAC for closure.

  • I am not against studying this technique. Though that should have come before we have done hundreds of thousands of implants.

  • I am impressed by surgical LAAO as shown in LAAOS 3, They show that the appendage can be an appropriate target. Though recall that LAAOS 3 included continuation of OACs.

  • I am concerned that there is way too much enthusiasm for this preventive procedure. It’s preventive. A preventive procedure with a 3% to 8% major complication rate.

I dare you to look critically at the evidence. It’s not strong. And If you convince a patient to have percutanoeus closure rather than OAC, I think you are on very shaky grounds.

One final comment on the matter of absolute contra-indication to OAC. You think this is a great indication for LAAO. Okay. What happens to this patient who cannot take OACs because of bleeding when he or she develops a device-related thrombus? If that happens, say 3% to 7% of the time, you have harmed that patient.

Postural Orthostatic Tachycardia Syndrome (POTS)

There’s a group at the University of Oklahoma interested in autonomic innervation of the heart. Stavros Stavrakis and his team have done work on the use of transcutaneous vagus nerve stimulation (tVNS) for treatments of AF and HF preserved EF (HFpEF).

My friends, this sounds crazy, but it is a clip on the tragus of the ear. That’s it. They use below pain threshold stimulation, which exerts anti-andrenergic and anti-inflammatory effects.

Their most recent study involves treating patients with POTS. Perhaps our most vexing condition in EP/cardiology. JACC -Electrophysiology published the RCT.

  • This was a simple experiment. the active arm is tVNS, the sham is an ear clip on the ear lobe instead of the tragus.

  • Stimulation was for 1 hour over 2 months.

  • Endpoints were postural tachycardia, heart rhythm variability (HRV), and serum cytokines and anti-autonomic antibodies.

  • Mean age 34 years. All female. 26 patients total. All with well documented POTS.

  • Postural tachycardia was significantly less in the active arm compared with the sham arm at 2 months (mean postural increase in heart rate 17.6 ± 9.9 beats/min vs 31.7 ± 14.4 beats/min; P = 0.01).

  • Antiadrenergic autoantibodies and inflammatory cytokines were lower in the active arm compared with the sham arm at 2 months (P < 0.05).

  • HRV was better in the active arm. No device-related side effects were observed.

  • Symptoms assessed using a 31-point questionnaire, however, were not significantly improved.

They found for the first time that the intermittent tVNS of the auricular branch of the vagus nerve at the tragus, over a 2-month period, attenuated the orthostatic tachycardia.

Their concluding statement:

Our results support the emerging paradigm of noninvasive neuromodulation to treat POTS. Mechanistically, this effect appears to be related to reduction of antiautonomic autoantibodies and inflammatory cytokines, and improvement in autonomic tone. Further studies are warranted.

My Comments. This is a nice study. There are hints of potential for patients with POTS, a difficult condition that seems surely related to autonomic dysmodulation. The treatment – tVNS — is non-invasive, well-tolerated, and it seemed to work.

The headwinds here are that it’s a small study; there was no improvement in symptoms, and the heart rate changes were modest. But this is promising, and I agree with the authors that more study is warranted.

This seems like how medical science should work. You do a small study. You don’t over interpret it. You see signals. And if there are enough signals, you do bigger studies and see what happens.

POTS is difficult to treat, so anything positive is welcome. Kudos to the University of Oklahoma team.

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.

processing....