Antithrombotic therapy is central to the management of atrial fibrillation. This analysis from the RHYTHM-atrial fibrillation (RHYTHM-AF) registry explored the appropriateness of antithrombotic therapy in relation to stroke risk and atrial fibrillation duration in patients with atrial fibrillation. RHYTHM-AF, a prospective multinational registry, enrolled consecutive adult patients with atrial fibrillation considered for cardioversion. We compared the type of antithrombotic therapy administered at the time of cardioversion and at discharge with stroke risk (“high stroke risk” defined by CHA 2 DS 2 -VASc >1) and duration of atrial fibrillation (≤48 vs >48 hours or unknown duration). Of 2,972 patients who were cardioverted (34.5% through pharmacologic cardioversion [PCV] and 65.5% through electrical cardioversion [ECV]), 65% were at high risk of stroke and 30% presented with atrial fibrillation of >48-hour or unknown duration. At the time of PCV and ECV, 36% (n = 242) and 84% (n = 1,075) of high-risk patients, respectively, were taking vitamin K antagonists or heparin. At discharge, these rates increased to 62% (n = 414) and 93% (n = 1,191), respectively. Of all low–stroke risk patients with short-duration atrial fibrillation undergoing PCV (n = 260) and ECV (n = 111), 7% (n = 17) and 30% (n = 33), respectively, were taking vitamin K antagonists or heparin at the time of cardioversion. At discharge, these rates increased to 19% (n = 50) and 40% (n = 44), respectively. In conclusion, ECV was frequently performed under appropriate antithrombotic therapy for most high-risk patients with atrial fibrillation, whereas PCV was frequently performed without appropriate antithrombotic therapy. To enhance pericardioversion stroke prevention, cardioversion algorithms should focus less on the type of conversion and more on stroke risk factors and atrial fibrillation duration.
Despite clear guidelines on anticoagulation, patients with paroxysmal atrial fibrillation appear to be less well anticoagulated relative to those with persistent atrial fibrillation, despite the presence of stroke risk factors. Current guidelines recommend therapeutic anticoagulation with warfarin (international normalized ratio 2.0 to 3.0) for a minimum of 3 weeks before cardioversion and 4 weeks after cardioversion. Lower rates of anticoagulation are associated with a greater risk of stroke in patients undergoing cardioversion. Although data signaling an association of thromboembolism and pharmacologic cardioversion (PCV) are limited, there is the perception that PCV is less thrombogenic than electrical cardioversion (ECV). This assumption is false as postcardioversion stunning depends on the duration of atrial fibrillation rather than the type or method of conversion per se. Yet, some drugs such as sotalol may increase postcardioversion stunning. Nonetheless, patients undergoing PCV appear to be less well anticoagulated than patients undergoing ECV, both at the time of cardioversion and long term thereafter. The aim of the present analysis from the RHYTHM-atrial fibrillation (RHYTHM-AF) study was to explore the types of antithrombotic therapy used in patients with atrial fibrillation and the appropriateness of antithrombotic therapy in relation to stroke risk, atrial fibrillation duration, and complication rates.
Methods
RHYTHM-AF, a prospective multinational registry, enrolled consecutive adult patients with atrial fibrillation considered for cardioversion. The details of the RHYTHM-AF study design have been published previously. In brief, 4,300 patients were enrolled from 177 participating hospitals and acute care centers in Australia, Brazil, France, Germany, Italy, the Netherlands, Poland, Spain, Sweden, and the United Kingdom from May 2010 to June 2011. Patients were eligible for inclusion in this registry if they were aged ≥18 years, had documented atrial fibrillation at the time of enrollment, had an atrial fibrillation diagnosis confirmed by electrocardiography, and had cardioversion (of atrial fibrillation) as a planned therapeutic option. Patients who were enrolled in another clinical trial or had a diagnosis of atrial flutter were excluded from enrollment in the study. Patient data (vital status, information on atrial fibrillation events, hospitalizations, and treatment) were collected at the time of atrial fibrillation presentation and at follow-up, conducted at 60 ± 10 days, by way of a remote Web-based data collection form. Major outcomes of the RHYTHM-AF study included the success rate of cardioversion and recurrence of atrial fibrillation.
RHYTHM-AF was conducted in accordance with the European Union Note for Guidance on Good Clinical Practice CPMP/ECH/135/95 and the Declaration of Helsinki. The study was only initiated at the site level after local and ethics approval requirements were obtained.
The present analysis from the RHYTHM-AF registry assessed: (1) the types of antithrombotic therapy used in patients with atrial fibrillation initially considered for rhythm control, that is, conversion to sinus rhythm; (2) the appropriateness of antithrombotic therapy in relation to stroke risk and atrial fibrillation duration at cardioversion and discharge; and (3) the complication rates in relation to appropriateness of antithrombotic therapy. Patients classified as having a high stroke risk who were being treated with antithrombotic therapy and those classified as having a low stroke risk not being treated with antithrombotic therapy were considered appropriately treated (i.e., inappropriate treatment included both patients who were over- and underanticoagulated). Patients with a CHA 2 DS 2 -VASc score >1 and/or with valvular atrial fibrillation were classified as “high stroke risk”; patients with a CHA 2 DS 2 -VASc score ≤1 without valvular atrial fibrillation were classified as “low stroke risk.” For the assessment of appropriateness of antithrombotic therapy, patients with CHA 2 DS 2 -VASc score 0 or 1 were also considered high stroke risk if atrial fibrillation was of >48-hour or unknown duration. Patients were considered to have valvular atrial fibrillation if they had any of the following: previous surgery for valvular heart disease; valvular heart disease as an indication for antithrombotic therapy before enrollment or at discharge; mitral stenosis under history of cardiovascular disease; or mitral stenosis (grade ≥2) as an abnormality found with transesophageal echocardiogram (TEE).
Descriptive statistics are presented, and data are expressed as n (%) or mean ± SD. Comparisons were performed by chi-square test or Wilcoxon rank sum test, as appropriate. A p value of <0.05 was considered to be statistically significant.
Results
Recruitment continued locally to meet the target sample size of 4,300 patients; in this study, we report on a smaller sample that was analyzed based on globally applied data cut times (n = 3,940). Baseline demographics and clinical characteristics for the cohort of patients considered for rhythm control, subdivided into those who subsequently underwent ECV (n = 1,946), PCV (n = 1,026), or no cardioversion (n = 968), are presented in Table 1 . The overall mean age was 66 years, and 38% of the patients were women. Overall, 62% of patients had hypertension (left ventricular hypertrophy at echocardiography present in 8.0%) and 15% had heart failure (with depressed left ventricular ejection fraction in 10%). The average CHA 2 DS 2 -VASc score was 3, and 69% of the total population had CHA 2 DS 2 -VASc score >1.
| Variable | Total, n = 3,940 ∗ (%) | ECV, n = 1,946 ∗ (%) | PCV, n = 1,026 ∗ (%) | No CV, n = 968 ∗ (%) |
|---|---|---|---|---|
| Age (yrs) † | 66 ± 12 | 66 ± 11 | 66 ± 14 | 66 ± 13 |
| Women | 38 | 33 | 46 | 38 |
| Hypertension | 62 | 62 | 62 | 63 |
| Previous MI | 11 | 12 | 9 | 12 |
| Diabetes mellitus | 17 | 17 | 16 | 20 |
| Previous stroke or TIA | 5 | 5 | 5 | 5 |
| Heart failure | 15 | 17 | 11 | 16 |
| COPD | 8 | 6 | 11 | 8 |
| CHA 2 DS 2 -VASc † | 3 ± 2 | 3 ± 2 | 3 ± 2 | 3 ± 2 |
| AF ≤48 h | 35 | 15 | 75 | 33 |
| Heart rate (beats/min † ) | 105 ± 31 | 94 ± 26 | 123 ± 29 | 105 ± 31 |
| Systolic BP (mm Hg † ) | 132 ± 20 | 131 ± 19 | 133 ± 23 | 131 ± 20 |
| Diastolic BP (mm Hg † ) | 80 ± 13 | 80 ± 13 | 80 ± 15 | 79 ± 12 |
| LVH present § | 8 | 7 | 11 | 8 |
| TEE performed | 24 | 34 | 6 | 24 |
∗ Sample size, unless otherwise noted.
§ Sample sizes for total, ECV, PCV, and no CV groups were n = 3,799, 1,854, 1,020, and 925, respectively.
At cardioversion, 58% of the 1,946 subjects undergoing ECV had an international normalized ratio of 2.0 to 4.5. Of note, international normalized ratio was <2.0 in 18% of those undergoing ECV and 43% of the 1026 patients undergoing PCV ( Table 2 ).
| INR | Total (n = 3,940) | ECV (n = 1,946) | PCV (n = 1,026) | |||
|---|---|---|---|---|---|---|
| VKA, n = 2,206 (%) | No VKA, n = 1,723 (%) | VKA, n = 1,514 (%) | No VKA, n = 432 (%) | VKA, n = 230 (%) | No VKA, n = 796 (%) | |
| <2.0 | 9 (206/2,206) | 34 (582/1,723) | 10 (149/1,514) | 45 (196/432) | 25 (57/230) | 48 (386/796) |
| >2.0 to ≤4.5 | 54 (1,182/2,206) | 3 (51/1,723) | 72 (1,091/1,514) | 10 (44/432) | 40 (91/230) | 1 (7/796) |
| >4.5 | 4 (87/2,206) | 1 (20/1,723) | 5 (76/1,514) | 2 (9/432) | 5 (11/230) | 1 (11/796) |
| Not applicable | 5 (115/2,206) | 25 (430/1,723) | 5 (81/1,514) | 36 (157/432) | 15 (34/230) | 34 (273/796) |
| Unknown | 28 (616/2,206) | 37 (640/1,723) | 8 (117/1,514) | 6 (26/432) | 16 (37/230) | 15 (119/796) |
Antithrombotic treatment according to stroke risk and atrial fibrillation duration in all patients who underwent cardioversion is shown in Figure 1 . For patients who presented with atrial fibrillation duration of ≤48 hours, vitamin K antagonist (VKA) or heparin was used in 13% of those at low risk of stroke compared with 39% of those at high risk of stroke and/or valvular atrial fibrillation. VKA was used in ≥90% of patients who presented with atrial fibrillation of >48-hour or unknown duration.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
