Collateral Damage: A Patient, a New Procedure and the Learning Curve

“We don’t know anything. We’re really pushing the technology. Just because you can, doesn’t mean you should. We don’t know anything about what lesions we really need, anything  about what locations you need, whether they need to be transmural or not. We don’t know anything about it at all, so far, as far as I am concerned.”

“The Cat is Out of the Bag”

In the long hot summer days at the end July 1998, the FDA’s Circulatory System Devices Panel held an important meeting in order to deal with a recurring problem.

An invasive medical procedure had once again become popular with the public, and the FDA had been left behind. Standards and guidelines were needed for cardiac catheter ablation procedures, so that de jure could be in synch with de facto, and new developments could be somewhat regulated. Such a big crowd was expected at the Gaithersburg Holiday Inn to discuss the matter that they opened up the Grand Ballroom, which is what you get when you remove the temporary wall that divides the Goshen Room from the Potomac Room.

So. How to design a trial to evaluate a procedure which has, for better or worse, already become the standard of care? How to provide marketing approval for procedures and devices that already dominate the marketplace? Catheter ablation for more simple arrhythmias such as Wolfe Parkinson White and Atrial Flutter had been proven to be safe and effective within traditional FDA definitions. But now, there was pressure from industry and medical consumers to do the same for conditions that were far more complex. The panel had to figure out a way to evaluate catheter ablation for atrial fibrillation.

When someone suggested looking at the literature and then holding trials to sort of codify the current state of affairs, GW’s Cindy Tracy cut to the chase: “If you are talking about going and evaluating what we would now consider a standard radiofrequency treatment for atrial flutter, I don’t think that is going to fly in any sense because the cat is out of the bag with that, and that is clinical practice at this point.”

Tracy could see what was coming and wanted to get ahead of the curve for once.

DR. PORTNOY: Cindy, what if the study device is one that is currently approved for SVT ablation and, right now, we know that those devices are also used for off-label ablation so there is no adequate current.control under those circumstances because the study device, itself, is the one that is being used today to do off-label A-flutter ablation.

DR. STUHLMULLER: Can I interrupt for one second. I think part of what the panel is struggling with and I
the issues that came up yesterday is when you are doing a study, you can’t compare an investigational device to an investigational device. I think what the panel is asking for clarification on here today from the agency is what do you do when the off-label use of an approved device is considered the standard of care and how can you factor that into a study design potentially against an investigational device.

DR. TRACY: That’s exactly it.

*****

We’ve got to get this right it’s out there ….  , cabg?/stents

DR. VETROVEC: There have been a number of trials in all kinds of things where the first three procedures done in each investigator’s institution are done a specific way. In this case, it would right-only. Then, after that was demonstrated to have no complications and given that the data coming in at that point showed that the centers were all having a high incidence of recurrence, then it would allow all the investigators to go forward and do right and left at the same time.

DR. ROSS: My name is Michael Ross. I am from industry, a company called Atrionics, working specifically
in catheter ablation of atria1 arrhythmias. I remain a little confused on this right-sided, left-sided, debate.
The reasons are as follows. If you look at the results of right-sided lesioning over the past couple of years, at best, I think the companies that have released their results are operating at the margins and, at worst, I would say that the data that I am seeing from these studies would probably never pass FDA
scrutiny. I am wondering as we move from the question which is, will right-sided lesions work, to, are they needed at all, and they probably are– but the more important statement is that left-sided lesions are almost certainly indicated to cure this disease. So it begs the question. How do you consent a
patient for a right-sided-only procedure and is it ethical do to? If we are trying to cure this disease and not change the results with drug therapy, how do we go to these patients and tell them we are going to do a procedure on the right side, it is probably not going to work but we just need to get this data.” http://www.fda.gov/OHRMS/DOCKETS/AC/98/transcpt/3442t2b.pdf p 89

“To submit a patient to a right-sided ablation just to get practice is not going to happen. It is not going to happen.” Dr. Simmons http://www.fda.gov/OHRMS/DOCKETS/AC/98/transcpt/3442t2b.pdf p 90

In 1999, as Principal Investigator for a clinical trial evaluating the safety and efficacy of the Guidant Heart Rhythm Technologies Linear Ablation System, Hugh Calkins reported early trial results to The American Journal of Cardiology: http://tinyurl.com/Guidant-NewAblationSytem-1999

“To date, 15 patients have been enrolled, and the procedure was acutely effective in 14 of 15 patients with no complications. Atrial fibrillation has recurred during short-term follow-up in 12 of 15 patients, a not surprising result, because this initial phase of testing involved only right-sided ablation.”

Catheters were snaked into the hearts of fifteen people. A hot wire was dragged  back and forth inside their hearts to make scars. Fourteen of the 15 people did not get injured in the process.

It didn’t help anybody, but hardly anyone got hurt. Mission accomplished.

It was considered to be a successful study because Hugh Calkins knew it would turn out that way.

*******

Cindy Tracy2a p26 You are strictly dealing on the right side, at least for typical flutter, so you are not exposing them to the risk of a trans-septal and left-sided energy delivery. We expect at least a 90 percent acute success rate. It has been about a 1 to a 5 percent recurrence rate, at least from what I have seen so far.

***

Cindy Tracy 2b p74: DR. WHARTON: The thing about that is if we render a person noninducible, the assumption is going to be made by the practicing physician that they can stop Coumadin. That is one of the big issues about this whole thing about symptomatic A-fib anyway. We can render most people
asymptomatic of drugs or HIS ablation, if you want to make them asymptomatic.

The bigger issue is, can I do something that would allow me to take Coumadin off with all of the sort of
associated morbidity with that and the cost of monitoring anticoagulation. So I think that, as we look at these procedures, as we start looking at these procedures as curative procedures and not palliative procedures, we are going to have to look very closely at what we are doing to atrial-transport function and emboli risk.

DR. TRACY: I couldn’t agree more. We really have to understand why we are doing this in the first place.

Right now, there is such an unknown thing to me–I am a real skeptic. I don’t know why we are talking about this whole thing in the first place, anyway. Still, you are talking about an entity where the prognosis is defined by the underlying cardiac condition so anything we have been doing so far for patients with atrial fibrillation has been to make them symptom free.

We are opening up the question now of are we going to make them better — are we going to reduce their risk of stroke further by doing an ablation or a maze procedure to make them better–are we going to reduce their risk of stroke by adequately and appropriately anticoagulating patients at risk for stroke in
the first place.

We would have to go a long way before I would be convinced that we are achieving better than what we can do with Coumadin therapy. So why are we doing this in the first place? I don’t know. But I sure know that I don’t want this to be done and to have anybody made worse, to increase their risk of stroke by not knowing about the silent episodes of atrial fibrillation, by inappropriately discontinuing Coumadin therapy.”

Curtis p 78
Let’s go to NO. 14. “Is it appropriate to begin an A-fib study in the right heart only in order to characterize the safety of the device in a lower-risk environment or can patients be treated in the left heart with a new ablation system without any right heart experience?”

I don’t know that doing it in the right heart, by itself, is going to tell me something about–if it is 100 percent safe in the right heart and nothing ever happens there, it still doesn’t tell me I am not going to have a stroke when I do it on the left side. So I don’t think that is the reason why we would thinking of it.

The reasons why investigators have thought about right-sided-only versus right-and-left is it is easier,
shorter. If you can get an adequate success rate on the right side, you avoid having to go on the left side which we think is likely to have a higher complication rate.

So I don’t think the way the question, as posed–I am not sure that is the right question to ask.

DR.,SIMMONS: I agree. It implies you are doing a lesser procedure just to find out what the risks are.
DR. CURTIS: Yes.
DR. WHARTON: Can I make one other comment that I think goes unnoticed with regard to this subject? There is a huge learning curve with the investigator with any new catheter. I think, in terms of safety, it is probably a better for them to learn in the right atrium where there is less they can hurt before we start sticking–

Tracy p80: DR. TRACY: I appreciate what you are saying but if it is not likely to be– if the success is not likely to be very high with right-sided lesions only, then it doesn’t make sense to limit a study to right-sided lesions only.
DR. WHARTON: Can’t argue that.
DR. CURTIS: Then you are using the patient to get
your learning curve in without–

Tracy p93: DR. TRACY: We are struggling to figure out exactly what it is that needs to be done. We don’t even know. So I think it makes designing a study very, very difficult because we don’t know very much about even what it is that we are trying to accomplish.