Statistical analysis
The sample size was estimated by the time to VA elimination since ablation and ablation applications.7,10,13 The average IA successful rate (𝜋) was assumed 85% and the difference (𝜹) was assumed 20%. For a power of 90% (Z𝛽=1.28) at a two-sided alpha level of 0.05 (Z𝛼=1.96), 134 patients were needed. Assuming a dropout rate of 20%, 160 patients were planned to be enrolled. Due to the significant subgroup effect, after reviewing the data, the ethics committee and the leadership discontinued the trial in July 2020.
The primary and secondary outcome was analyzed using the intention-to-treat (ITT) principle. The logistic regression model using frontward selection (Wald) was applied to test the multiplicative interaction between the EVA sites and strategies. Their addictive interaction of absolute excess risk due to interaction (AERI) was calculated.14 Continuous variables were expressed as mean ± standard deviation or median (interquartile range). Categorical variables were expressed as number (frequency %). For comparison of patient characteristics, differences of continuous variables were tested using Student’s t-test if normally distributed; otherwise, Mann-Whitney U test was used. Categorical variables were compared using Fisher exact test. The VA recurrence was compared based on a time-to-first-event analysis using the log-rank test. A two-tailed P value of less than 0.05 was considered statistically significant. For multiple comparisons of 6 contrasts among subgroups based on the EVA sites and IA strategies, a two-tailed P value of less than 0.0083 was considered statistically significant using Bonferroni correction. All statistical analysis was performed using SPSS 26.