Prognostic Implications and Efficacy of Catheter Ablation by Atrial Fibrillation Type

Figure 1. Cumulative incidence of composite outcomes in patients with PAF and PersAF. Cumulative incidences of composite outcomes (ie, all‐cause death, hospitalization for HF, stroke, and bleeding) of patients with PAF and PersAF. The shaded areas represent the 95% CI. Patients with PersAF showed a significantly higher composite outcome incidence at 2 years than those with PAF (log‐rank P<0.001). HF indicates heart failure; PAF, paroxysmal atrial fibrillation; and PersAF, persistent atrial fibrillation.

Figure 2. Subgroup analysis of association between AF type and composite outcomes using Cox proportional hazards model. Cox proportional hazards regression adjusted for baseline characteristics was used to assess the risk of adverse outcomes. Dependent variables were the composite of all‐cause death, hospitalization for HF, stroke, and bleeding within the 2‐year follow‐up. Dots and bars indicate the adjusted HR and 95% CI, respectively. PersAF was an independent predictor of the composite outcome among various subgroups, excluding women, those with HF, and those without left atrial enlargement. AF indicates atrial fibrillation; CHF, congestive heart failure; HF, heart failure; HR, hazard ratio; LAD, left atrial dimension; LVEF, left ventricular ejection fraction; and PersAF, persistent atrial fibrillation.

Figure 3. Changes in AFEQT scores of patients with PAF and PersAF. The radar graph shows the AFEQT scores at baseline and 1 year after registration by the type of AF. The dots indicate the mean overall summary and individual scores of the 4 domains. The blue and red lines indicate the AFEQT scores of patients with PAF and PersAF, respectively. AFEQT scores at baseline (left). AFEQT scores 1 year after registration (right). AF indicates atrial fibrillation; AFEQT, Atrial Fibrillation Effect on Quality‐of‐life questionnaire; PAF, paroxysmal atrial fibrillation; and PersAF, persistent atrial fibrillation.

Figure 4. Subgroup analysis of the association between AF type and 1‐year changes in the AFEQT‐OS score using linear regression analysis. Linear regression model adjusted for baseline characteristics and AFEQT‐OS score was used to assess the association between AF type and the change in AFEQT‐OS score. The dependent variable was the 1‐year change in the AFEQT‐OS score with respect to the baseline. Dots and bars indicate adjusted differences and 95% CI, respectively. AF indicates atrial fibrillation; AFEQT‐OS, Atrial Fibrillation Effect on Quality‐of‐life questionnnaire overall summary; CHF, congestive heart failure; LAD, left atrial dimension; LVEF, left ventricular ejection fraction; PAF, paroxysmal atrial fibrillation; and PersAF, persistent atrial fibrillation.

Figure 5. Comparison of cumulative incidence of composite outcomes in propensity score‐matched patients receiving CA and medical therapy. The figure shows the cumulative incidence of composite outcomes (ie, all‐cause death, hospitalization for HF, stroke, and bleeding) between the propensity score‐matched patients undergoing CA and medical therapy in each AF type. Undergoing CA was associated with a lower incidence of composite outcomes in those with PAF (A), and the trend, which did not reach statistical significance, was consistent in those with PersAF (B). AF indicates atrial fibrillation; CA, catheter ablation; HF, heart failure; PAF, paroxysmal atrial fibrillation; and PersAF, persistent atrial fibrillation.