Electrophysiology is on the brink of a possible disaster

A flawed study with positive results has sparked enthusiasm for implantation of left atrial appendage devices during AF ablation. it is a potential disaster.

The OPTION trial compared two strategies to reduce stroke and bleeding after catheter ablation of atrial fibrillation. The current strategy is to continue an anticoagulant medication. The alternative tested in OPTION is to implant a left atrial appendage device—abbreviated as LAAC (left atrial appendage closure).

AF ablation is one of the most common procedures in my field. We ablate left atrial heart cells in attempt to reduce or eliminate episodes of AF. The success of AF ablation is debatable—and not the issue for this post.

I write today to sound a warning about a wave of implantation of devices placed into the left atrium at the same time as the ablation. CMS has already issued a special billing code for extra payment of the device.

Here is a Tweet from Christopher Ellis at Vanderbilt describing an implant at the same time as ablation. “Even dropped a FLXPRO” which is another name for a LAAC.

---

As many readers know, I have long been skeptical regarding the evidence supporting this device. I’ve debated against LAAC on five continents. My main argument is that the device failed against warfarin in its seminal trials.

Proponents say, no worries, John, we will use LAAC in patients not eligible for anticoagulation. But these patients were excluded from clinical trials and we have near zero RCT-level evidence to support this procedure.

The thing that saddens me most is that hundreds of thousands of these devices have been implanted worldwide. And nearly no patient has been randomized.

---

Things got much worse this past weekend at the AHA meeting when the results of OPTION were released. The trial delivered positive results. Of course it did. I wrote days before that the trial was designed to deliver positive results.

Then I wrote yesterday about the four huge flaws of the trial. Briefly, the trialists chose an improper primary safety endpoint. Improper because it included nonprocedural major bleeding and nonmajor clinically-relevant bleeding. It is inappropriate to exclude procedural bleeding in the primary safety endpoint because procedural bleeding is the most common adverse effect of LAAC.

The second problem was an improper efficacy endpoint. The core treatment goal after AF ablation is stroke prevention. But the authors included all-cause death to the stroke/systemic embolism composite. This is a problem because efficacy was tested with a noninferiority design. Numerous studies show that neither anticoagulation nor appendage closure affects death. By adding an endpoint not affected by either treatment, you inflate the number of events (adding noise) and make noninferiority easier to reach.

Perhaps the most notable efficacy finding in OPTION was the very low stroke rates in either arm (1.2% and 1.3%). This is far lower than our simple CHADSVASC calculator would have it. To discern any signal of noninferiority or superiority of the device with stroke rates that low would require 5x more patients randomized. To sort out noninferiority in the DOAC vs warfarin trials required more than 10,000 patients per trial. OPTION enrolled only 1600.

The third problem was missing data. Nearly 9% of data was not collected. This is higher than either the safety and efficacy endpoint, respectively. This adds even more uncertainty to an already uninformative trial.

Finally, there is the issue of incomplete closure of the appendage with the device. One in five patients had incomplete closure. This is a worrisome outcome because peri-device leaks create turbulence and a higher incidence of clots and stroke.

---

I am afraid. Really afraid that my colleagues will begin implanting these devices at the time of AF ablation. We do hundreds of thousands of AF ablation procedures per year in the US.

All the forces are aligned to greatly expand the number of LAAC procedures. Doctors will earn extra money for these implants. High-volume procedural docs will gain status in their employed healthcare system as well as in the profession. Patients, too, will be excited when told that if they have the device they can stop anticoagulants. It’s easier to do the implant because you are already in the left atrium.

Yet the evidence from OPTION does not come close to supporting placing a permanent foreign body in the heart. We could easily be wrong.

Our profession should demand stronger evidence before doing something so aggressive. Be warned. This could be one of the worse chapters in modern cardiology.