Experience Rules: AF Ablation Complication Risk Higher at Lower-Volume Centers

May 09, 2019

SAN FRANCISCO — Should all catheter ablations for atrial fibrillation (AF) be performed at very experienced hospitals, maybe by setting minimum volume requirements, or even by restricting the procedure to appointed centers of excellence?

Food for thought about the wisdom of such measures comes from an analysis covering the years 2010 to 2014 in United States, consistent with data from earlier periods and for other cardiac procedures, suggesting that for AF ablation, experience counts.

The comorbidity-adjusted risk for serious complications went up significantly at hospitals performing 20 or fewer AF ablations annually, compared with those doing at least 53 per year, in the analysis comparing hospital case-volume by tertiles over the 5 years.

Those in the lowest-volume, compared with the highest-volume tertile, showed more than twofold greater risks for stroke, in-hospital death, or any complication, and more than five times the risk for cardiac perforations.

Perforations were also significantly more common at middle-tertile centers compared with those in the highest tertile.

Jim Cheung

"It's pretty safe to say that procedural volume is an independent predictor in the analysis," Jim W. Cheung, MD, Weill Cornell Medical College, New York City, told theheart.org | Medscape Cardiology.

"Regardless of comorbidities, it is associated with worse outcomes," he said. "We don't know if it's causal, but it's an association that makes sense."

Moreover, he noted, most ablations performed over the 5 years — overwhelmingly — were done at the least experienced centers by annual case volume. Lowest-volume hospitals together accounted for almost 80% of the 1738 centers doing AF ablations during that period.

Those in the highest-volume tertile did about 6% of all the ablations, including the 2% of all procedures done at centers with annual volumes of more than 100.

The number of hospitals in the lowest-volume tertile climbed significantly over the 5 years (P < .001 for trend), especially beginning in 2012.

The increase was accompanied by significant declines in number of hospitals in the middle (P = .017) and top (P = .009) tertiles for annual volume, said Cheung, who presented the analysis here at the Heart Rhythm Society 2019 Scientific Sessions.

It's yet to be determined whether the field should establish guidelines for maintaining a certain AF-ablation volume threshold for institutions or operators, for example, or perhaps restrict AF ablations to a limited number of centers that will be sure to maintain high volumes, he observed.

"Those are questions that need to be debated, but it's not hard to argue whether it's good to have the procedures at the lowest-of-the-lowest-volume centers."

In the analysis based on 1738 centers with admissions for AF ablation from 2010 through 2014 in the Nationwide Readmissions Database (NRD), 79.3% of centers did ablations at the rate of 20 or fewer per year. Far and away most of the centers (63.2% of the total) did 10 or fewer per year.

Also of the total, 14.8% of centers did 21 to 50 procedures per year, and 5.9% did 51 or more per year; only 2% did more than 100 annually.

Most serious complications in the analysis went up at lower volumes. But interestingly, Cheung pointed out, neither tertile 1 nor 2 showed an increased risk for readmission compared with tertile 3, whether for any cause or for AF or atrial tachycardia, or for cardiac or noncardiac causes considered separately.

Adjusted Hazard Ratio (HR) for Complications, Hospital Annual Volume Tertiles 1 and 2 vs 3, 1738 Centers 2010–2014*
Complication Event Tertile 1 vs 3, HR (95% CI) P Tertile 2 vs 3, HR (95% CI) P
Cardiac perforation 5.11 (3.70–7.05) <.001 2.52 (1.80–3.54) <.001
Vascular complications 1.49 (1.20–1.84) <.001 1.19 (0.96–1.47) .105
Stroke 2.37 (1.14–4.93) .021 1.57 (0.79–3.12) .202
Death at index admission 2.18 (0.81–5.82) .125 1.75 (0.66–4.62) .259
Any complication 2.06 (1.73–2.45) <.001 1.39 ( 1.16–1.67) <.001
*Tertile 1: 1–20 per year; tertile 2: 21–52 per year; tertile 3: ≥53 per year.
Adjustments included age and sex; history of heart failure, coronary disease, coronary revascularization, valvular disease, implantation with cardiac rhythm device, diabetes, obesity, pulmonary disease, renal disease, and anemia; length of hospitalization; whether teaching hospital; and hospital size by number of beds.

In another finding, he said, "across the board, the patients who were treated at lower-volume centers were older and sicker — they had more coronary artery disease, they had a lot more congestive heart failure." They also had significantly more valvular disease, diabetes, and renal and pulmonary disease, and stayed in the hospital longer.

"It's almost like adding insult to injury." Not only are patients at lower-volume hospitals sicker, Cheung said, their risk for serious complications could be increased because of the lower case volume.

The analysis attempted to control for patient comorbidities and history of other cardiovascular procedures, he noted, "and even then, the complication rates were significantly higher at the lower-volume centers."

So if comorbidities and clinical history aren't big contributors to the excess complication risks at lower-volume centers, what factors do play a big role?

Cheung suggested that a low number of ablations at some centers may be a marker for limited overall clinical capabilities, including fewer specialists available on short notice and a dearth of other resources that other centers may have in more abundance.

It may be that the richer the center in terms of that overall hospital character, the lower their risk for complications from AF ablation, Cheung speculated. If true, he proposed, it would support a preference to have ablations performed at larger centers of excellence.

Cheung discloses consulting for Abbott and Biotronik and receiving fellowship support from Biosense Webster, Biotronik, Boston Scientific, Medtronic, and St. Jude Medical.

Heart Rhythm Society (HRS) Scientific Sessions 2019: Abstract S-PO01-123. Presented May 8, 2019.

Follow Steve Stiles on Twitter: @SteveStiles2. For more from theheart.org | Medscape Cardiology, follow us on Twitter and Facebook.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.