The Safety of Catheter Ablation for Atrial Fibrillation

By Michael H. Crawford, MD, Editor

SYNOPSIS: A meta-analysis of randomized, controlled studies of first catheter ablation for atrial fibrillation showed complication rates are low and have declined over time. Mortality rates are very low and have remain unchanged over the same period.

SOURCE: Benali K, Khairy P, Hammache N, et al. Procedure-related complications of catheter ablation for atrial fibrillation. J Am Coll Cardiol 2023;81:2089-2099.

Catheter ablation (CA) is superior to drug therapy for maintaining sinus rhythm in patients with atrial fibrillation (AF) and is one of the fastest-growing procedures in cardiovascular medicine. Reported procedural complication rates are inconsistent, and there is little information about recent trends in this evolving technique.

Benali et al performed a systematic review and pooled analysis of patients’ first CA for either paroxysmal or persistent AF in randomized, controlled trials (RCTs) that were conducted from 2013 to 2022 using either radiofrequency or cryoballoon. In each RCT, there was at least one year of follow-up. The primary outcome was the rate of procedural complications in the first year. Secondary outcomes involved specific complications and the CA method. The authors analyzed 89 RCTs that included 15,701 patients.

The overall complication rate was 4.5%, and 2.4% of these were severe complications. Vascular complications were the most numerous (1.3%; of which serious vascular complications were 0.3%), followed by pericardial effusion/tamponade (0.8%; of which 0.4% required fluid drainage), and stroke in 0.2%. Less frequently occurring events were phrenic nerve injury (0.1%), pulmonary vein stenosis (0.05%) and atrio-esophageal fistula (0.05%). Procedure-related deaths occurred in 0.06%.

Researchers evaluated temporal trends by dividing the population into two groups: those from 2013 to 2017 and those between 2018 and 2022. Investigators noted a significant decline in complication rates over time (5.3% vs. 3.8%; P < 0.05), which was driven mainly by a drop in severe complication rates (3.1% vs. 1.9%; P = 0.001). However, the procedural one-year mortality rate was not different between the two periods (0.06% vs. 0.05%; P = not significant). Demographic characteristics, such as sex, and procedural factors, such as CA type, did not show any difference in complications. The authors concluded the complication rates associated with CA are low and have declined over the last five years. The mortality rates were very low (< 0.1%) and unchanged over time.

COMMENTARY

The electrophysiologists (EPs) in my group will see patients with AF only if they are referred by a general cardiologist. Thus, I frequently must describe the potential risks and benefits of CA to patients for them to consider visiting with my EP colleagues. Consequently, this study was of interest.

I used to say the overall complication rate was about 5%, which was correct five years ago. Benali et al found CA complication rates have declined about 30% over the last five years to 3.8% total and 1.9% for serious complications (I say 4% and 2% to make it simple for the patient). This is remarkable considering today’s CA patients are older and experience heart failure caused by reduced left ventricular function more often.

This reduction in complication rates could be attributed to technical improvements, such as uninterrupted anticoagulation, ultrasound imaging, esophageal temperature monitoring, and contact force sensing catheters, rather than patient selection. Specifically, in the Benali et al study, there was no population characteristic such as sex, age, or type of AF that was related to the complication rate. Also, there was no difference based on the type of CA (radiofrequency or cryoballoon). Mortality rates have remained very low at 0.05%. Most deaths occurred within 30 days of the procedure and usually happened because of atrio-esophageal fistulas (75%) or strokes (25%).

The major strength of the Benali et al research is that they confined the studies included in the meta-analysis to RCTs, which alleviates the biases associated with retrospective observational studies. However, since RCTs usually are performed in large, high-volume centers, complication rates might have been underestimated compared to lower-volume centers. Also, only radiofrequency and cryoballoon ablation studies were included. Newer techniques might further lower complication rates. In addition, individual patient-level data were not available, so other characteristics such as race, ethnicity, and socioeconomic status could not be considered. Despite these limitations, this study will inform my patient discussions about their need to see an EP specialist for CA consideration.