In patients with persistent atrial fibrillation (AF), the sinus rhythm (SR) can be restored by direct current cardioversion (DCC), although the recurrence of AF after successful DCC is common. We examined whether transesophageal echocardiography (TEE)–guided early DCC, compared with the conventional approach of DCC after 3 weeks of anticoagulation with dabigatran-etexilat, reduces the recurrence of AF. A total of 126 consecutive patients with persistent AF were randomly assigned to a TEE followed by early DCC (n = 65) or to a conventional treatment with dabigatran-etexilat for 3 weeks followed by DCC (n = 61). None of the patients received any antiarrhythmic treatment other than β blockers, and all the DCCs were successful. Forty-eight–hour Holter monitoring was performed at 28 days and at 3, 6, and 12 months after the DCC. The primary outcome was AF recurrence lasting ≥30 seconds. The analysis was stratified by AF duration <60 (n = 62) or >60 days (n = 64) before DCC. We observed a significant reduction in the AF recurrence risk (p = 0.003) in patients with persistent AF <60 days who received early DCC, but there was no significant benefit of early DCC (p = 0.456) in patients with persistent AF lasting >60 days. The recurrence-free survival probability at 28 days in patients with persistent AF <60 days was 0.27 (95% confidence interval 0.14 to 0.51) in the conventional treatment group compared with 0.69 (95% confidence interval 0.54 to 0.87; p = 0.006) in the early DCC group. A benefit of early DCC persisted throughout 12 months of follow-up. In conclusion, TEE-guided early DCC in patients with persistent AF <60 days results in a significant reduction of AF recurrence.
Atrial fibrillation (AF) affects 1% to 2% of the population, with its prevalence increasing with age to 17% in those aged ≥80 years. The disorder is associated with a nearly twofold mortality risk and a fivefold to sevenfold ischemic stroke risk compared with the background population. In patients with an AF duration of <7 days, rhythm control treatment is common practice. Permanent patients with AF, however, have little, if any, chance of achieving a sinus rhythm (SR) because of irreversible remodeling and left atrium (LA) dilation, meaning that rate control and symptom relief is often the best strategy. The most appropriate treatment approach for patients with persistent AF, which is accountable for 25% to 30% of cases, is more complicated. We, therefore, set out to examine the benefits of the transesophageal echocardiography (TEE)–guided early direct current cardioversion (DCC) on AF recurrence in patients with the condition ≤4 months, dividing them into 2 subgroups of persistent AF lasting for <60 or >60 days. We hypothesized that the early DCC restoration of an SR in patients with persistent AF for <4 months would significantly reduce AF recurrence rates compared with treatment with dabigatran-etexilat for 3 weeks followed by DCC (conventional treatment). We also hypothesized that the benefits of early SR restoration would be more pronounced in patients with persistent AF for <60 days than in those with the condition for >60 days.
Methods
We conducted a randomized controlled study, enrolling 141 consecutive patients with persistent AF for a 2-year period (2011 to 2013). These patients were either admitted to the Department of Cardiology of Odense University Svendborg Hospital or referred to the outpatient clinic with a symptomatic (European Heart Rhythm Association score II to III) persistent AF duration of <120 days and an indication for DCC. Both groups were given dabigatran-etexilat (150 mg twice daily) immediately after randomization. In the early DCC group, TEE using a 3.5 MHz Vingmed System probe (GE Healthcare, Norway) was performed within 48 hours after randomization, and DCC was attempted after left atrial appendage thrombus exclusion. The conventional group received dabigatran-etexilat for 3 weeks before the DCC attempt. Both treatment groups continued with dabigatran-etexilat (150 mg twice daily) for a minimum of 4 weeks after DCC according to international guidelines. All patients also received 50 to 100 mg of metoprolol at the time of randomization as a rate control and for symptom relief. No other antiarrhythmic drug was used. After successful DCC, a maintenance dose of metoprolol was continued in both treatment groups. The DCCs were performed using a R Series ALS defibrillator (Zoll, Pittsburg, PA) and in accordance with the current guidelines. Before the DCC, a standard transthoracic echocardiography was performed. During the follow-up period of 12 months, all the patients had an electrocardiogram (ECG) and additional 48-hour Holter monitoring using a Modular Digital Holter recorder (Spacelabs Healthcare, Snoqualmie, WA) at 4 weeks and then after 3, 6, and 12 months after DCC. Analysis of the Holter recordings was conducted by trained and experienced technicians using the Sentinel software (Spacelabs Healthcare).
The primary outcome was ECG/Holter-documented AF recurrence lasting for ≥30 seconds. The study was approved by the local ethics committee and was carried out in accordance with the Declaration of Helsinki. Informed consent was obtained from the patients before their inclusion in the study, which is registered at ClinicalTrials.gov with the study ID number S-20110075.
The statistical methods used are as follows: the continuous variables are expressed as a mean ± SD, whereas the categorical variables are presented as counts and percentages of the 2 treatment groups. The categorical baseline values were compared with the chi-square test, whereas the continuous values were analyzed using the Wilcoxon rank-sum test. We also calculated nonparametric Kaplan-Meier survival plots and used these to predict the recurrence risk, using a cutoff of an AF duration of 60 days. We compared the recurrence-free survival at 28 days and at 3, 6, and 12 months, stratified by AF duration using the chi-square test.
Results
We screened 199 patients and excluded those with an AF duration of >120 days (n = 24) and those with reversible causes of the condition (thyrotoxicosis, n = 4; infection, n = 14; acute coronary syndrome, n = 3; and valvular AF, n = 13). Of 141 suitable patients, we excluded a further 11 because of spontaneous conversion and 4 because of the presence of thrombi in the left atrial appendage. This left 126 patients who were suitable for DCC. These 126 participants with persistent AF underwent a successful DCC either early (n = 65) or after conventional treatment with dabigatran-etexilat for 3 weeks (n = 61). The baseline characteristics in the 2 treatment groups were very similar. However, although the patients were randomly assigned to the conventional or early DCC groups, the mean ages did differ significantly Table 1 .
Variable | Conventional N = 61 | TEE N = 65 | P-value |
---|---|---|---|
Sex ( Female) | 13 (21%) | 7 (11%) | 0.17 |
Age (Years) | 69.1 ± 8.48 | 66.3 ± 8.54 | 0.043 |
AF time (Days) | 63.4 ± 28.7 | 63.0 ± 36.6 | 0.71 |
Diabetes | 7 (12%) | 7 (11%) | 1 |
Hypertension | 31 (52%) | 30 (46%) | 0.66 |
Dyslipidemia | 16 (27%) | 16 (25%) | 0.95 |
Body mass index | 29.2 ± 4.60 | 29.4 ± 4.67 | 0.94 |
EHRA score (I/II/III/IV) | 1/44/16/0 | 0/22/43/0 | 3.39*10 -5 |
CHA2DS2-VASc total | 1.97 ± 1.38 | 1,41 ± 1.00 | 0.25 |
HAS-BLED total | 1.21 ± 0.78 | 1.02 ± 0.75 | 0.12 |
Angiotensin receptor inhibitor | 8 (13%) | 7 (11%) | 0.87 |
Angiotensin II receptor blocker | 13 (21%) | 4 (6%) | 0.026 |
Statins | 17 (28%) | 12 (18%) | 0.30 |
Dabigatran etexilat | 61 (100%) | 65 (100%) | 0.72 |
Beta blocker | 58 (95%) | 60 (93%) | 0.20 |
Calcium channel blockers | 13 (21%) | 8 (12%) | 0.26 |
Digitalis | 7 (11%) | 9 (14%) | 0.90 |
Baseline Echocardiography : | |||
Left atrial volume (ml) | 72.7 ± 10.5 | 71.8 ± 8.86 | 0.83 |
Left ventricular ejection fraction (%) | 54.2 ± 7.90 | 49.3 ± 8.03 | 0.0041 |
Left ventricular volume (ml) | 103.7± 15.4 | 106.2 ± 14.9 | 0.28 |
E/a (post DC) | 1.11 ± 0.427 | 1.12 ± 0.463 | 0.66 |
E/E’ | 8.00 ± 2.29 | 7.94 ± 2.40 | 0.87 |
Left ventricular septum (cm) | 1.06 ± 0.116 | 1.09 ± 0.137 | 0.45 |
We observed a significant reduction in AF recurrence at 28 days and at 3, 6, and 12 months, in patients with AF persisting for <60 days who received early DCC. The recurrence-free survival probability follow-up at 28 days in the patients with AF for <60 days was 0.27 (95% confidence interval 0.14 to 0.51) in the conventional group versus 0.69 (95% confidence interval 0.54 to 0.87; p = 0.006) in the early DCC group. The statistically significant benefit of early DCC persisted throughout the follow-up period and was still present at 12 months after cardioversion ( Table 2 ). The impact of early DCC is also visualized in the Kaplan-Meier plot ( Figure 1 ), which shows no significant effect on recurrence-free survival if AF duration before DCC is disregarded. However, if stratified by AF duration >60 or <60 days before DCC, the Kaplan-Meier plot clearly illustrates the benefit of early DCC on recurrence-free survival in those with AF for <60 days ( Figure 2 ). During the follow-up, 94 (96.9%) of the recurrences were type-persistent AF, and only 3 (3.1%) were episodes of paroxysmal AF lasting for >30 seconds.
Variable | 28 days | 3 months | 6 months | 12 months | ||||
---|---|---|---|---|---|---|---|---|
Recurrence free survival probability | CI (95%) | Recurrence free survival probability | CI (95%) | Recurrence free survival probability | CI (95%) | Recurrence free survival probability | CI (95%) | |
Conventional AF < 60 days | 0.27 | 0.14-0.51 | 0.23 | 0.11-0.47 | 0.19 | 0.09-0.89 | 0.08 | 0.02-0.29 |
Early DCC AF < 60 days | 0.69 | 0.54-0.87 | 0.56 | 0.41-0.76 | 0.47 | 0.32-0.68 | 0.38 | 0.24-0.59 |
P-value for difference | 0.0017 | 0.0059 | 0.010 | 0.0027 | ||||
Conventional AF > 60 days | 0.40 | 0.27-0.60 | 0.37 | 0.24-0.57 | 0.34 | 0.22-0.54 | 0.25 | 0.15-0.45 |
Early DCC AF > 60 days | 0.30 | 0.18-0.50 | 0.30 | 0.18-0.50 | 0.27 | 0.16-0.48 | 0.18 | 0.08-0.38 |
P-value for difference | 0.41 | 0.55 | 0.53 | 0.46 |