Statistics
Continuous variables were expressed as mean and standard deviation (SD) when normally distributed (positive Kolmogorov–Smirnoff test). Categorical variables were reported as frequencies and percentage. Parametric variables were compared by means of paired Student's t-test and nonparametric variables by the Wilcoxon test or chi-square test. Comparison of more than 2 groups was performed with ANOVA analysis and post hoc Bonferoni test for parametric variables and with the Kruskal–Wallis test for nonparametric variables. Cohen's κ coefficient was used as a statistical measure of inter-rater agreement and a good level was defined as κ > 0.61.
To determine independent associations with TE events, univariate logistic regression analysis was performed. Variables with a P-value ≤0.1 were then included in the multivariate analysis for the determination of odds ratio (OR) and its 95% confidence intervals (CI). With the observed 25 events and a 1:3 case–control ratio, a 35% difference in LAA morphology distribution (as reported previously) was detectable with 75% power and P-value <0.05. A 2-tailed P ≤0.05 was considered statistically significant. Analysis was performed with SPSS v20.0 (SPSS Inc., Chicago, IL, USA).