Radiofrequency Ablation, Cryotherapy Ablation, or Pulsed-Field Ablation to Treat Paroxysmal Atrial Fibrillation Unresponsive to Pharmacological Treatments: Interpreting Efficacy Through Reconstruction of Individual Patient Data From Randomized Trials.

Three techniques of catheter ablation (CA; radiofrequency, cryoballoon, and pulsed-field ablation) are available to treat patients with paroxysmal atrial fibrillation (PAF) who do not adequately respond to pharmacological treatments. Our study was aimed at comparing these techniques based on the data of randomized studies because these are considered the best sources of efficacy data. After selecting pertinent trials, our analysis studied the time-to-event data published for these three techniques. An artificial intelligence method was used that reconstructs individual patient data from the Kaplan-Meier curves. The endpoint was an arrhythmia recurrence. A preliminary heterogeneity analysis was performed. Then, our main analysis was based on individual patient data reconstructed from Kaplan-Meier graphs. The hazard ratio (HR) was its main parameter. Three randomized trials were included. Our heterogeneity analysis confirmed an acceptable level of between-trial heterogeneity that allowed us to pool the curves from the different trials; however, cryoballoon ablation with a two-minute duration fared worse than the other techniques. Then, our main analysis estimated the following values of HR: pulsed-field ablation versus radiofrequency ablation, 0.549 (95%CI, 0.413-0.730; p<0.001); pulsed-field ablation versus cryoballoon ablation, 0.478 (95%CI, 0.364-0.633); radiofrequency ablation versus cryoballoon ablation, HR=0.871 (95%CI, 0.585-1.295; p=0.506). In conclusion, radiofrequency ablation and cryoballoon ablation showed similar effectiveness (except for the two-minute cryoballoon ablation, which fared worse). Our results showing the superiority of pulsed-field ablation versus thermal ablation must be interpreted with caution because the patients given pulsed-field ablation were limited, and their follow-up was shorter than that of patients receiving thermal ablation.