Results
Of the 259 patients (mean age, 60.6 ± 11.4 years; 50.6% male), 127 (49%) had ERAT during the 3-month blanking periods. Among the patients with ERAT, the timing of the ERAT was during the first month post ablation in 65 patients (51.2%), during the second month in 14 (11.0%), and during the third month in 48 (37.8%). Baseline demographic and clinical characteristics of the study population are detailed in Table 1. Patients with ERAT were older (mean, 63.1 ± 9.5 and 58.2 ± 12.5 years, respectively; p <0.001) and had larger left atrial size (mean, 56.1 ± 12.1 and 52.3 ± 8.9 mm, respectively; p = 0.004) than those without ERAT. In addition, a significantly greater proportion of patients with ERAT had a history of persistent AF (92.1% and 74.2%, respectively; p<0.001).
The following operative and postoperative risk factors were more commonly recorded in patients with ERAT than in those without ERAT (Table 2): longer cardiopulmonary bypass (CPB) time (181.3 ± 63.3 and 154.8 ± 59.2 minutes, respectively; p =0.001) and aortic cross-clamp (ACC) time (128.1 ± 53.6 and 108.2 ± 45.3 minutes, respectively; p =0.001) and more postoperative pulmonary complications and reoperations for bleeding (11.0% and 3.0%; p= 0.022). Patients with ERAT received bi-atrial surgical AF ablation more frequently (80.3% and 62.1%, P = 0.002) and used more amiodarone during blanking periods (82.7% and 21.2%, P<0.001).
The factors independently associated with ERAT after multivariate logistic analyses are shown in Table 3. Older age (per year: odds ratio [OR], 1.03; 95% CI, 1.00-1.06; p =0.012), history of coronary artery disease (CAD) (OR, 6.98; 95% CI, 1.38-35.36; p =0.019), history of persistent AF (OR, 3.18; 95% CI 1.36-7.43; p =0.008), larger left atrial size (OR, 1.03; 95% CI, 1.00-1.06; p =0.022), longer CPB time (OR, 1.00; 95% CI, 1.00-1.01; p =0.033), and reoperation for bleeding after cardiac surgery (OR, 3.00; 95% CI, 1.41-6.37; p =0.004) were associated with the occurrence of ERAT during the three-month blanking periods.
At the 12-month follow-up, 74 of 127 patients (58.3%) with ERAT were free from late AF recurrence compared with 129 of 132 patients (97.8%) without ERAT (p <0.001; Figure 1A). AF-free survival rate was 95.4%, 64.3%, and 8.3% among those who had ERAT in the first, second, and third months, respectively (p <0.001; Figure 1B). To evaluate the role of ERAT as a predictor contributing to LR, the univariate Cox regression model was performed according to the occurrence and timing of ERAT (classified by month) after surgical ablation. After adjustment for the occurrence of ERAT, age (per year: HR, 1.04; 95% CI, 1.00-1.07; p =0.025), male gender (HR 1.77; 95% CI, 1.02-3.09; p =0.043), larger LA size (HR, 1.04; 95% CI, 1.02-1.06; p <0.001), and the occurrence of ERAT (HR, 17.73; 95% CI 5.48-57.34; p <0.001) were shown to be independent risk factors for LR. In addition, adjusted analyses were conducted using the timing of ERAT within the blanking periods. The occurrences of ERAT during the second (HR, 16.70; 95% CI, 3.98-70.22;p =0.001) and third months (HR, 119.75; 95% CI, 36.25-395.59;p <0.001) were the most powerful independent predictors of LR (Table 4).
The ROC curve that determined the cut-off value of the blanking period is shown in Figure 2. The AUC was 0.938 (95% CI: 0.893 to 0.983,p < 0.001), showing excellent discrimination. The ideal cut-off value for the blanking period was 58 days. The occurrence of ERAT beyond 58 days predicted LR with a sensitivity of 93.2% and specificity of 86.8%