Introduction We aimed to determine whether maximal exercise capacity, left ventricular function, and quality of life (QoL) scores are sustained in patients with atrial fibrillation (AF) after termination of training following exercise-based cardiac rehabilitation (ECR). Methods In a prospective study, 58 patients with AF (age, 62 ± 7 years) were randomized into an exercise training group for 12 months (ECR, n = 13), a group with 6-month detraining after a 6-month exercise training (DT, n = 15), and a medical treatment only group (MT, n = 30). For exercise training, cycling on a bicycle ergometer was performed in the DT and ECR groups 3 times a week for 6 and 12 months, respectively. Each session started with a 10-minute warm-up at 60%–70% of the maximal heart rate (HRpeak), followed by four 4-minute intervals at 80%–90% of the HRpeak, with 3 minutes of active recovery at 60%–70% of the HRpeak between intervals, ending with a 5-minute cool-down period. Peak exercise oxygen consumption (Vo2), left ventricular function, plasma lipid level, N-terminal pro B-type natriuretic peptide level, and QoL score were measured at baseline, after 6 months of training, and after additional 6 months of continuous training or detraining follow-up assessments. Results Six months of exercise training increased the peak Vo2 (ECR: 28.2 ± 7.0 ml/[kg·min] and DT: 28.0 ± 8.3 ml/[kg·min] vs. MT: 23.0 ± 5.2 ml/[kg·min], p= 0.047) and QoL scores (36-Item Short-form Health Survey) and decreased the total cholesterol level (ECR: 146.8 ± 43.8 mg/dl and DT: 140.6 ± 22.4 vs. MT: 182 ± 44.2 mg/dl, p = 0.008). Detraining resulted in an increase in QoL score and a decrease in total cholesterol level (ECR: 142.8 ± 21.4 mg/dl and DT: 151.8 ± 39.8 vs. MT: 176.8 ± 34.2 mg/dl, p = 0.017). However, the exercise training-induced increase in peak Vo2 reverted to the baseline level after detraining (ECR: 27.1 ± 5.1 ml/kg/min vs. DT: 24.4 ± 7.2 ml/[kg·min] and MT: 21.5 ± 4.86 ml/[kg·min], p = 0.017). The resting left ventricular systolic and diastolic functions were not significantly different, with no inter-group difference after 6 months of training and 6 months of continuous training or detraining. Conclusions As the QoL scores are maintained despite peak Vo2 being lowered by detraining after an exercise training period in patients with AF, physicians should encourage patients’ participation in a continuous exercise program to sustain the improvement in both the QoL score and peak Vo2 with exercise training. Further studies with larger sample sizes are needed to observe the long-term effects of exercise training and detraining.

Introduction Exercise training (ET) improves endothelial function and arterial stiffness in patients with cardiovascular disease. However, whether ET improves endothelial function and arterial stiffness in patients with atrial fibrillation (AF) is unclear. If it does, for how long the effects are sustained remains to be determined. Methods In a prospective study, 58 patients with AF (age, 62 ± 7 years) were randomized into an ET group for 12 months (CT, n = 13), a group with 6-month detraining after a 6-month ET (DT, n = 15), and a medical treatment only group (MT, n = 30). For ET, cycling on a bicycle ergometer was performed in the DT and CT groups 3 times a week for 6 and 12 months, respectively. Each session started with a 10-minute warmup at 60% to 70% of the maximal heart rate (HRpeak), followed by four 4-minute intervals at 80% to 90% of the HRpeak, with 3 minutes of active recovery at 60% to 70% of the HRpeak between intervals, ending with a 5-minute cooldown period. Peak exercise oxygen consumption (Vo2), intimal-medial thickness (IMT) of the carotid artery measured on high-resolution ultrasonography, and left ventricular function were measured at baseline, after 6 months of training, and after additional 6 months of continuous training or detraining follow-up assessments. In addition, plasma von Willebrand factor (vWF), endothelin-1, nitric oxide, tumour necrosis factor alpha, interleukin-1 beta, interleukin-6, and interleukin-10 levels were measured as indices of endothelial function. Results The 6-month ET increased peak Vo2 (CT: 29.0 ± 6.5 ml/[kg·min] and DT: 26.0 ± 8.2 ml/[kg·min] vs. MT: 23.0 ± 5.2 ml/[kg·min], p = 0.04) and decreased plasma vWF levels (CT: 103.7 ± 30.7 IU/dL and DT: 106.0 ± 31.2 IU/dL vs. MT: 145.0 ± 47.7 IU/dL, p = 0.01). Detraining significantly reduced the ET-induced increase in Vo2 and decreased the vWF to baseline level, although continuous ET maintained changes in Vo2 (CT: 28.1 ± 5.3 ml/[kg·min] and DT: 22.4 ± 6.6 ml/[kg·min] vs. MT: 21.5 ± 4.8 ml/[kg·min], p = 0.043) and in vWF (CT: 84.3 ± 39.1 IU/dL vs. DT: 122.2 ± 27.5 IU/dL and MT: 135.9 ± 50.4 IU/dL, p = 0.014). However, carotid arterial IMT, and resting left ventricular systolic and diastolic functions showed no significant changes, with no inter-group differences after 6 months of training and 6 months of continuous training or detraining. Conclusions The decreased vWF level suggested that ET can be a strong non-pharmacologic option to improve endothelial function in patients with AF. However, it can rapidly lose its effects after detraining. Therefore, physicians should encourage their patients to participate in a continuous exercise program to sustain its benefits in terms of improved exercise capacity and endothelial function in patients with AF.