Fehmi Keçe

Chapter 6 130 6.2.4 Statistical Analysis Sample size for the prospective randomized groups was calculated based on the incidence of DC in our hospital with the first generation balloon (42%, expected to be comparable with the additional 90s group) and second generation balloon (5%, expected to be comparable with the additional 150s group). For the additional 120s group an expected incidence of 21% was used. A Cochran-Armitage test for linearity was performed to obtain a sample size for the three groups using STATA software, V.12 (Stata Corp, College Station, Texas, USA). With α=0.05 the necessary total sample size was 66 patients to detect a significant trend in DC with 80% power. The group size was therefore set at a total sample size of 75 patients. Baseline characteristics were compared between the randomized groups using one-way ANOVA for continuous variables and Chi-square tests or Fisher’s exact test for categorical variables. Predictors for reconnection/DC were tested using multivariate regression analysis. Multivariate Cox-regression was used to identify predictors for recurrence. Variables with a p<0.1 in univariate analyses were entered in the multivariate analysis using the enter method. The log-rank test was used to test differences for AF recurrences. A p-value of <0.05 was considered statistically significant. SPSS (version 23, SPSS Inc., Chicago, IL, USA) was used.