SNOWMASS, Colorado (EGMN) – In-hospital mortality due to left atrial catheter ablation for atrial fibrillation is markedly increased at lower-volume centers, according to a recent analysis of all patients in the U. S. government health insurance database undergoing the increasingly popular procedure.
“This is a study [that] I think is very sobering. I find the data believable. There are a moderate number of deaths that you hear about through the grapevine,” Dr. Roger A. Winkle said at a conference sponsored by the American College of Cardiology.
Somehow these deaths often don’t find their way into recent reports from large multicenter registries and meta-analyses, which are based on voluntary reporting of safety and efficacy data, typically from selected centers of excellence.
In contrast, the Medicare database includes all Medicare beneficiaries undergoing catheter ablation for atrial fibrillation (AF) at U.S. hospitals. That’s what makes the Medicare study uniquely compelling – and disturbing – according to Dr. Winkle, director of electrophysiology at Sequoia Hospital, Redwood City, California.
Dr. Ethan R. Ellis of Beth Israel Deaconess Medical Center, Boston, and his coworkers evaluated radiofrequency catheter ablation procedures for AF done in 8,288 Medicare patients during 2001-2006. Dr. Winkle noted that the investigators excluded from their analysis 2,223 of these patients who underwent ablation at 562 hospitals that do fewer than two such procedures per year.
In-hospital death as a procedural complication occurred in 0.1% of patients at hospitals in the top quartile in terms of volume, with more than 104 ablations annually. The mortality rate was 0.3% at hospitals in the second quartile, with 50-103 ablations annually, and jumped significantly to 0.65% in the 50% of hospitals in the bottom two quartiles, with 12-49 cases annually (Heart Rhythm 2009;6:1267-73).
Dr. Winkle noted that “6.5 per 1,000 is 1 death in every 150 patients, versus 1 in 1,000 at high-volume centers. I think that’s a little bit scary.”
In contrast, a recent meta-analysis of 63 radiofrequency ablation and 34 antiarrhythmia drug studies, typically done at centers of excellence and with voluntary data reporting, found zero procedure-related deaths and no left atrial esophageal fistulae in more than 5,000 ablation procedures (Circ. Arrhythm. Electrophysiol. 2009;2:349-61).
“I don’t believe this. It can’t be true. But that’s what they showed,” he commented.
In the Medicare study, the overall complication rate increased from 6.7% in 2001 to 10.1% in 2006 as the annual number of ablation procedures performed nationally climbed sixfold. The rates of pneumothorax and stroke/transient ischemic attack were higher at low-volume hospitals. In contrast, perforation, tamponade, and overall vascular complication rates didn’t vary by procedural volume.
Results of a recent worldwide survey of 182 high-volume catheter ablation centers provide a contemporary snapshot of procedural efficacy in the hands of experienced operators. The survey included nearly 21,000 ablation procedures done on 16,309 AF patients in 2003-2006. Overall, 70% of patients have remained asymptomatic with no need for further antiarrhythmic drugs during intermediate follow-up averaging 18 months after a mean of 1.3 procedures. Another 10% of treated patients became asymptomatic with the addition of previously ineffective antiarrhythmic agents (Circ. Arrhythm. Electrophysiol. 2009 Dec. 7 [doi:10.1161/CIRCEP.109.859116]).
In-hospital mortality occurred in 1 in 650 patients in this large cohort, a figure that “seems believable,” according to Dr. Winkle.
At Sequoia Hospital, where Dr. Winkle and his colleagues have performed more than 1,000 catheter ablations since 2003, the cure rate following a single procedure averaging 2 hours in duration is 75% for paroxysmal AF, 60% for persistent AF, and 53% for long-standing persistent AF. After a second and occasionally a third ablation procedure, these rates rise to 88%, 75%, and 71%, respectively. There have been no deaths or permanent neurologic deficits at this center, no atrial esophageal fistulae, and one case of pulmonary vein stenosis that did not require intervention.
“Every hospital now wants to have an A-fib ablation program. We have 11 hospitals within 50 miles of our hospital doing A-fib ablations. I’m just not sure that’s going to be the future if the Medicare database is correct – and I actually think it may be. It might be that this is a procedure that has to be done only in high-volume centers,” the electrophysiologist commented.
Dr. Winkle argued that the wide range in mortality between high- and low-volume ablation centers as shown in the Medicare database coupled with the sometimes unrealistically rosy picture that emerges from voluntary reporting constitutes a powerful argument for creation of a mandatory independent registry of long-term outcomes and procedural complication rates in all ablation centers.
“We desperately need in this field a Society of Thoracic Surgeons–type database where everyone is monitored by external forces. Right now everybody tells their patients some numbers they read in the literature, but very few people actually sit down and tell patients what their own complication and success rates are,” Dr. Winkle said.
The real-world numbers are all over the map, commented Dr. Andrew S. Wechsler, who has kept a close eye on the state of catheter ablation while working to develop minimally invasive surgical means of treating AF.
“I think catheter-based ablation of atrial fibrillation is just like complex cardiac surgery, where the results are so critically dependent on the numbers of [procedures] that are done in a given center. They’re very system based as well as individual operator based,” observed Dr. Wechsler, professor and chairman of the department of surgery at Drexel University, Philadelphia.
Disclosures: Dr. Winkle indicated he has no financial interests relevant to his presentation. Dr. Wechsler is the medical director at Estech Corp., which makes equipment that surgeons use in treating AF.
