Time-to-event and survival curves were estimated by using the Kaplan-Meier approach and are displayed as descriptive graph (11). For other dichotomous variables, the chi-square test and the Fisher exact test were used as appropriate. Continuous variable were compared using the selleck chemicals llc Student t test when normal distribution was confirmed; otherwise the Wilcoxon rank test was used. To identify potential predictors of inappropriate shocks, univariate and multivariate (using logistic model) analyses were carried out. Only predictors with p values <0.10 (on univariate analysis) were added in the multivariate model. All endpoint analyses were carried out on the basis of the intention-to-treat principle. Patients

with missing outcome data were considered in the analysis as follows: censored at the time of last follow-up for survival analysis or assuming none experienced the outcome of interest for dichotomous variables. The statistical software used for the analyses was SAS version 9.2 (SAS Institute Inc., Cary, North Carolina). A total of 462 patients were included in the OPTION trial, and 453 received study devices. The dual-chamber setting group consisted of 230 patients, and 223 patients were assigned to the single-chamber setting group (Figure 1). The clinical characteristics of the 2 groups

at baseline are given in Table 1. The average follow-up duration was 23.4 ± 7.9 months. During the trial, a BKM120 chemical structure total of 47 patients crossed over from one treatment group to the other: 39 crossed over from the single-chamber setting group to the dual-chamber setting group and 8 from the dual-chamber setting group to the single-chamber setting group. Known reasons for crossover from the single-chamber setting to dual-chamber setting arm included the

occurrence of inappropriate therapies Parvulin in 13 patients, clinical causes in 5 patients, and programming errors in 8 patients. The switch from the dual-chamber setting arm to the single-chamber setting arm was explained by lead issues in 2 patients and programming errors in 3 patients. The time to first inappropriate shock was significantly longer in the dual-chamber setting group compared with the single-chamber setting group (p = 0.012, log-rank test) (Figure 2A). The hazard ratio was 2.5 (95% confidence interval [CI]: 1.2 to 5.3) in favor of dual-chamber setting therapy. The endpoint of all-cause death or cardiovascular hospitalizations occurred in 46 patients (20.0%) in the dual-chamber setting group and 50 (22.4%) in the single-chamber setting group (Table 2). The pre-specified equivalence analysis with a margin of 17% confirmed the equivalence of dual-chamber setting therapy to single-chamber setting therapy (p < 0.001). Figure 2B illustrates the occurrence of the events over time. A total of 88 patients (19.9%) received at least 1 ICD shock: 37 (16.1%) in the dual-chamber setting group and 51 (22.9%) in the single-chamber setting group (Table 3).