A Trial That Changes What a Cardiologist Can Promise an Ablation Patient

Cardiologists at hospitals in Germany and Poland asked people with atrial fibrillation to agree to something unusual: an operation that might turn out to be a performance. Of the 262 who consented, 89 were taken into the catheter laboratory, given the same puncture in the groin as everyone else, and returned to the ward without a catheter ever having been threaded up to the heart. Neither those patients nor the staff who cared for them afterward were told who had had which. The trial was built to answer one question: how much of the relief people report after an ablation actually comes from the ablation.

Atrial fibrillation is the most common sustained disturbance of heart rhythm. The upper chambers of the heart quiver rather than beat, which can leave a person with palpitations, breathlessness and exhaustion, and raises the risk of stroke. Catheter ablation treats it by threading a wire up a vein in the groin and destroying small patches of tissue around the mouths of the pulmonary veins, where the stray electrical signals usually start (more than 500,000 such procedures are performed worldwide each year).

The answer, published in The Lancet and presented on Aug. 30 at the European Society of Cardiology congress in Munich, is that a good deal of it may not. Six months after their procedure, patients in both arms scored far higher on the standard questionnaire that measures how much atrial fibrillation interferes with daily life. The gap between the two arms was 2.6 points. Catheter ablation "did not demonstrate superiority over a sham procedure" for quality of life at six months, write Rolf Wachter of University Hospital Leipzig, Nikolaos Dagres of the German Heart Centre at Charité in Berlin and their colleagues.

That is a narrower statement than it may read as, and the uncertainty around the 2.6 points is the reason. The range runs from slightly below zero to a gain of about eight points, which is more than the roughly five that research on this questionnaire suggests a patient must gain before the change is one they would notice. Failing to show that ablation is better is not the same as showing the two are equivalent. What the trial establishes is that any benefit is smaller than two earlier sham-controlled trials reported, and small enough that a trial this size could not resolve it.

Procedures are almost never tested this way, and the consent form explains why. Asking someone to swallow a capsule that may be a placebo is one thing; asking them to be sedated, punctured and sent home without knowing whether anything was done to their heart is another. Recruiting for this trial took four years.

Two earlier fake operations pointed the other way

This is the third time an ablation for atrial fibrillation has been tested against a sham, and the first two came out the other way. SHAM-PVI, run at British centers and published in JAMA in 2024, found an advantage of 18.4 points for ablation on the same questionnaire at six months. PFA-SHAM, a Prague-led trial published in Circulation in June, compared pulsed-field ablation with a sham and found a gap of 32.6 points. The Leipzig trial, PVI-SHAM-AF, enrolled more patients than both of those put together, and found nothing.

Why three careful trials can disagree

None of the three looks careless, and the patients they enrolled offer one explanation. PFA-SHAM admitted only people whose questionnaire scores were below 50, meaning those whose atrial fibrillation was making them thoroughly miserable and who had a long way to climb. The Leipzig trial set no such floor, and its patients started well above it. Less room to improve means less improvement to find. That is a possible reconciliation rather than a settled one; nobody has compared the trials head to head.

The shams themselves also differed, which matters when the sham is the control. SHAM-PVI's version included pacing the phrenic nerve, which runs to the diaphragm; Leipzig's was the groin puncture and nothing more. The two trials also asked different questions. In SHAM-PVI, quality of life was a secondary outcome: that trial was designed around how much atrial fibrillation the patients actually had. PVI-SHAM-AF was designed around how they felt.

The rhythm result still stands

On the rhythm itself, the three trials agree. According to the University of Leipzig's announcement of the results, 73 percent of the patients who received a real ablation were free of atrial fibrillation half a year later, against 52 percent of those who received the sham. Nothing in this trial argues that the procedure fails to do the mechanical job it was designed for.

What is in question is the other benefit ablation is meant to deliver. In the university's statement, Wachter said the data show that the greater part of the improvement patients report appears not to come from the ablation itself, but from something like a placebo effect or from medication adjusted at the same time. Dagres drew out the consequence for practice: guidelines recommend the procedure partly to improve quality of life, and this trial finds that the improvement cannot be sufficiently explained by what the ablation does.

Harm was uncommon and fell on both sides. One patient in each group died, and neither death was attributed to the procedure. Serious adverse events judged related or possibly related to what was done occurred in six patients in the ablation group and four in the sham group; one of those was an ischemic stroke, and it happened to a patient in the sham arm. The trial was paid for by Helios Gesundheit, a private hospital operator whose own centers perform ablations and whose commercial interest therefore runs against the result.

These are the six-month figures, and the 12-month follow-up is still running. Patrick Haag, the trial's medical coordinator at the Heart Centre Leipzig, said the results change nothing about how well ablation prevents atrial fibrillation from returning, and a good deal about what patients should be led to expect for their own sense of well-being. The team says it will now go back through the data looking for the patients in whom the procedure does both.