Better Late Than Never: SHAM-PVI Shows AF Ablation Works

This transcript has been edited for clarity. 
John M. Mandrola, MD: Hi, everyone. This is John Mandrola, from theheart.org | Medscape Cardiology. I’m very excited to have the first author and senior author of the SHAM-PVI trial, Dr Dulai and Dr Veasey. I’ll let them introduce themselves. 
Rick A. Veasey, MBBS, MD: Hello. I’m Rick Veasey. I’m a consultant cardiologist and electrophysiologist at Eastbourne General Hospital. I was the principal investigator and lead author for the SHAM-PVI study. 
Rajdip Dulai, MBBS: My name’s Dr Dulai, and I am the lead investigator and first author of the SHAM-PVI study. I’m currently an electrophysiologist at the Essex Cardiothoracic Centre. 
Mandrola: Thank you for joining us. Why don’t one of you give us the lead results of the SHAM-PVI trial, the first proper sham-controlled trial of atrial fibrillation (AF) ablation versus a placebo procedure. I have been calling for this for 10 years, and I’m really excited to talk to you both.
Dulai: The SHAM-PVI study was conducted at two centers in the UK: Eastbourne District General Hospital and the Essex Cardiothoracic Centre. The main result of the study was that we saw a significant reduction in AF burden with AF fibrillation compared with a sham procedure. This was accompanied with substantial improvements in symptoms and overall quality of life. 
Mandrola: Many of our colleagues have said such a trial is not necessary because clearly AF ablation works. Tell us what you were thinking about and how the conversations went in the planning of this.
Veasey: This was a study I’d had in my head for 10 years or more, perhaps like yourself, thinking that it needed to be done. We’ve all seen multiple patients who have had AF ablations where they come back to clinic feeling fantastic, but still clearly in AF. The benefits from a placebo procedure are well documented throughout the decades for medicine.
It’s really the gold standard test of an intervention, a procedure, a therapy that we’re able to offer people to do a placebo-controlled study. I think this is something we need to be doing more and more in medicine so that we properly see the effects of what we’re offering to patients, not only because these are treatments that come with risks, but they’re also treatments that come with quite considerable cost implications for healthcare services. 
It was a study that was a bit of a slow burner for us. I had the idea of this study, but I didn’t have a research fellow myself, and Raj’s primary study that he was employed to do fell through. We ended up with a study with no research fellow and a research fellow with no study to do, and the pair married together very well. 
It was 2018 or 2019 that we got the ball rolling and went through the ethics process. We started the study about 6 months prior to the COVID pandemic. We’d put the first 10-12 patients through the study but then had to stop for, I think it was at least 18 months, and overcome that obstacle before we were able to get it finished. 
There’s been a large amount of debate about the primary endpoint of the study measuring AF burden vs symptoms, but 5-10 years ago, that was still how AF studies were designed. It was all about measuring the AF burden; that was the gold standard for these studies.
We’ve also previously done studies trying to assess symptoms and symptom improvements with various treatments in AF and really struggled with power calculations, because it’s harder to do and harder to estimate the effect of a placebo intervention. Going forward, we’ve actually now got a good estimation of what sort of symptom improvement you might expect against a placebo procedure. 
Mandrola: One group gets cryoballoon ablation, which is a typical way to do AF ablation. Then you decided for the placebo procedure to do vascular access, sedation, and phrenic nerve pacing. 
Was there a discussion about how adequate that was for a placebo procedure? Some people have suggested that a transeptal placebo procedure might be a better placebo. Would you comment on the, choice of the procedure and the placebo?
Veasey: We spoke to patients who had had catheter ablation procedures before and surveyed what are the things that they remember from their ablation procedure. The phrenic nerve pacing was the number one thing that stands out in terms of a sensation and an awareness that patients have had something done.
We went for the phrenic nerve pacing in both groups, and as part of the placebo procedure. We didn’t particularly look at doing the transeptal puncture. We really wanted to keep the placebo intervention as low risk as possible. We reviewed that with several cardiologists and electrophysiologists and felt it was a reasonable sham to offer people to compare AF ablation against.
Dulai: When we were designing the study, it was designed with patient input — what they felt was the phrenic nerve pacing. We went along the route of trying to simulate that cryoablation procedure while minimizing risk. I would add that if you look at the data, there are two patients in the ablation group who did not have an ablation because of issues with the transeptal puncture. We felt that that risk to the placebo group would not be warranted. 
If you look at blinding immediately after the procedure, it was near perfect. Patients were unable to guess which procedure they had. 
Mandrola: For listeners, I’ll just say that phrenic nerve pacing is used in cryoballoon ablation as a standard to study for phrenic nerve injury. It causes jumping of the diaphragm, and patients feel that. 
I understand why a transeptal puncture wasn’t done for safety reasons. Tell us about the patient population. One of the interesting things with this trial is that a large number of patients improved in the placebo arm. I think 80% of patients had persistent AF and had cardioversion. Would you comment about patient selection and improvement in the placebo arm? 
Dulai: The inclusion criteria stipulated patients with paroxysmal or persistent AF despite [use of] a class I or III antiarrhythmic, including beta-blocker use.
We had some exclusion criteria, including long-term persistent AF in patients with severe systolic dysfunction. The patients recruited are reflective of our patient population. When people are referred to us for ablation, it turns out that the majority have persistent AF. And around 80% of the study population had persistent AF. 
The average age was in the late 60s and the majority of patients had tried a class I or class III antiarrhythmic before. When we designed the study, one of the most important things was that, when they leave the cath lab, they had to be in sinus rhythm, so a normal heart rate. Otherwise, everyone would have just guessed if they’ve had an ablation or not, and you wouldn’t have a blinded study. 
The way we designed the study was that we did a stratified randomization according to type of AF, so that the number of persistent AF patients and paroxysmal AF patients were the same in each group.
Those patients who were in AF at the time of the procedure— so the persistent AF patients — were cardioverted as well. Both groups had an equal number of cardioversions, and then they went on to have their cryoablation or sham procedure. 
We did see a reduction, as you mentioned, in AF burden in the sham group. We think that was mostly due to the intervention of the cardioversion. 
Mandrola: SHAM-PVI, ORBITA, and ORBITA-2: Why do these very bold, brave, placebo-controlled trials come from the UK? I’m not sure that US patients would go for this. Perhaps it’s the way we discuss things with patients. How does this happen in the UK?
Veasey: We generally find, certainly in Eastbourne, that we have quite a high uptake of patients into our research studies. They particularly liked getting the loop recorders implanted. They like the degree of attention that comes with that. We don’t advertise this as a benefit of taking part in research; we explain what the study is. Throughout the years, we’ve done a lot of research in Eastbourne and we’ve had very good uptake.
I think the difficulty in other healthcare settings is the funding. Particularly in the US, you’re paying for a procedure. Who’s going to pay to have a placebo procedure? Who’s going to fund the hospital? We had access to funding to cover that ourselves.
I think the difference is in the health systems. They facilitate research in the UK quite well. 
Mandrola: Can I ask you to comment on a conclusion, at least at the European Society of Cardiology (ESC) Congress, that this completely ruled out any placebo effect. Do you think it completely rules out a placebo effect?
Veasey: I agree that it doesn’t completely rule out a placebo effect, because you’d then need a placebo procedure against nothing and against intervention, but it clearly shows that there are benefits way above a placebo effect for patients who are referred for catheter ablation. There’s clearly something different from our placebo to our intervention that gets rid of your AF and improves your symptoms. 
Dulai: I would say this is one of the few trials that has a hugely positive primary outcome. If you look at all secondary outcomes — whether it’s the Atrial Fibrillation Effect on Quality of Life (AFEQT) questionnaire, the Mayo AF-Specific Symptom Inventory, the 36-item Short Form Health Survey, or time to AF — every single secondary outcome is in favor and correlates with the reduction of AF burden that we see in the primary outcome.
Mandrola: I agree. I really appreciate that you took the time to come on and talk with us about it. I really congratulate you on planning this, getting it done, and showing everybody that there is a placebo-resistant effect to AF ablation. Thank you. 
Veasey: Pleasure. Thank you. 
Dulai: Thank you.