Statistical Analysis
Continuous variables were expressed as median and interquartile ranges
and compared using Mann-Whitney U test. Patients were compared based on
recurrence status (yes or no recurrence) and AF type, distinguishing
between those with paroxysmal AF and persistent AF. Categorical
variables were expressed as counts and percentages and were compared
using the Chi-square test or exact Fisher test, as warranted.
Comparative effectiveness of ablation strategies was evaluated with
multivariable Cox proportional hazards models. Patients were followed
from the index date of the first repeat AF ablation. Models were
adjusted for age, sex, BMI, AF pattern (e.g. paroxysmal, persistent),
diabetes, coronary artery disease, and hypertension. Cox models compared
individual ablation strategies, number of ablation strategies performed,
and presence of persistent AF. Kaplan-Meier estimates were utilized to
generate time-to-first-arrhythmia recurrence curves, which were
subsequently compared employing the log-rank test. In our models, we
analyzed the effects of all strategies, irrespective of whether they
were used alone or in combination with others. Interaction analyses
between the primary ablative strategies were conducted to ensure the
comprehensive evaluation of their impact. Results are presented as
hazard ratios (HR) along with their corresponding 95% confidence
intervals (CI). Effects estimates with 95% CI were used to determine
statistical significance together with P value < 0.05.
Statistical analyses were conducted using STATA version 18 (StataCorp
LLC, College Station, TX, USA).