SNOWMASS, COLO. – stroke, major bleeding, or cardiac arrest, compared with rhythm and/or rate control drugs in a propensity score–weighted, retrospective, observational study.
Findings of the investigation, which included more than 183,000 real-world patients in routine clinical practice, were reported by, during the annual Cardiovascular Conference at Snowmass sponsored by the American College of Cardiology.
The results breathe new life into the controversy created by the previously reported CABANA trial (Catheter Ablation vs. Antiarrhythmic Drug Therapy for Atrial Fibrillation), a 10-country study in which 2,204 patients with atrial fibrillation (AFib) were randomized to catheter ablation or antiarrhythmic and/or rhythm control medications and followed for a mean of about 4 years. CABANA yielded a negative result (), with the prespecified intent-to-treat analysis indicating no significant between-group difference in the primary composite endpoint – the very same one that was positive in the large observational study.
However, CABANA was marred by major problems arising from protocol deviations: Nearly 28% of patients assigned to medical therapy crossed over to catheter ablation, typically because their antiarrhythmic drugs failed, and 10% of patients randomized to catheter ablation never got it. This muddies the waters when trying to identify a true stroke/mortality benefit for catheter ablation, if indeed any such benefit was actually present.
Here’s where the controversy arose: While CABANA must be called a negative trial based upon the disappointing results of the intent-to-treat analysis, a prespecified post hoc analysis of patients as actually treated showed a statistically significant 27% relative risk reduction for the primary composite endpoint in the catheter ablation group. That’s strikingly similar to the 30% relative risk reduction for catheter ablation seen in the huge observational study, where the CABANA-type primary outcome occurred in 22.5% of the medically managed patients and 16.8% of those who underwent catheter ablation, noted Dr. Noseworthy, professor of medicine and director of heart rhythm and physiology at the Mayo Clinic in Rochester, Minn.
He ought to know: He was both an investigator in CABANA and first author of the published observational study ().
In the observational study, Dr. Noseworthy and coinvestigators utilized a huge U.S. administrative health claims database in order to identify a nationally representative group of 183,760 AFib patients, 12,032 of whom were treated with catheter ablation and the rest with antiarrhythmic and/or rhythm control drugs during the same years the CABANA trial was enrolling patients. The two groups were balanced using propensity score weighting to adjust for baseline differences in 90 variables.
The investigators sought to learn if the CABANA study population was representative of real-world AFib patients, and whether the observational experience could help resolve the CABANA controversy. It turned out that most AFib patients seen in daily clinical practice were CABANA like; that is, 74% of them would have been eligible for the clinical trial because they were symptomatic, over age 65, or younger than 65 with at least one CHADS2 stroke risk factor. About 22% of the large real-world sample would have been excluded from CABANA because they’d failed on amiodarone and other antiarrhythmic agents or had previously undergone ablation. About 4% of patients failed to meet the CABANA inclusion criteria.
The risk reduction for the composite endpoint associated with catheter ablation in the large retrospective study was greatest in the CABANA-like patients, at 30%. It was less robust but still statistically significant at 15% in patients who met at least one of the exclusion criteria for the trial.
The sheer size of this study provides greater statistical power than in CABANA. Of course, a nonrandomized, propensity score–based comparison such as this is always susceptible to confounding, even after adjustment for 90 variables. But the observational study does offer “a glimmer of hope” that catheter ablation, done in the right patients, might confer a stroke risk reduction and mortality benefit, he said.
The 33% relative risk reduction in the small group of real-world patients who failed to meet the CABANA inclusion criteria, while numerically impressive, wasn’t close to statistical significance, probably because event rates in that population were so low.
“Even if you could reduce stroke risk with ablation in that low-risk group, it would be a very inefficient way to reduce the population burden of stroke,” Dr. Noseworthy observed.
Putting together the results of CABANA and the large observational study to sum up his view of where catheter ablation for AF[ib] stands today, Dr. Noseworthy commented, “Ablation is reasonable for symptom control in many patients, basically anyone who is either breaking through on drugs or doesn’t want to take the drugs and is highly symptomatic. And there may be a small stroke and/or mortality benefit for people who are in the sweet spot – and those are people who look a lot like the patients enrolled in CABANA.”
Patients who met the exclusion criteria for CABANA are too advanced in their AFib to be likely to derive a stroke or mortality benefit from catheter ablation. “It’s very hard to move the needle in these patients with either a drug or catheter ablation approach. I wouldn’t try to reduce the risk of stroke here with an expensive and invasive procedure,” the electrophysiologist concluded.
He reported having no financial conflicts regarding his presentation.