Appendix 3

Summary Table

StudyAim of StudyStudy SizePatient Population/Inclusion and Exclusion CriteriaEndpoint(s)Statistical Analysis ReportedCI and/or p ValuesOR/HR/RR/OtherStudy Conclusion
ACTIVE A, Connolly et al. (10)To investigate whether the addition of clopidogrel to ASA would reduce risk of vascular events in patients with AF considered unsuitable for oral anticoagulation with warfarin.7,554Inclusion criteria: AF at enrollment or at least 2 episodes of intermittent AF in previous 6 mo and at least 1 of the following risk factors for stroke: age ≥75 y; systemic HTN during treatment; previous stroke, TIA, or non-CNS systemic embolism; LVEF <45%; PVD; or age 55–74 y and DM or CAD.
Exclusion criteria: Requirement of vitamin K antagonist or clopidogrel or the presence of any of the following risk factors for hemorrhage: documented peptic ulcer disease within previous 6 mo, history of intracerebral hemorrhage, significant thrombocytopenia (platelet count <50×109 per liter), or ongoing alcohol abuse.
Primary outcome was composite of stroke, MI, non-CNS systemic embolism, or death from vascular causes.Major vascular events occurred in 832 patients receiving ASA plus clopidogrel (6.8% per year) and in 924 patients receiving ASA plus placebo (7.6% per year).95% CI: 0.81 to 0.98; p=0.01RR: 0.89In AF patients considered unsuitable for warfarin, the addition of clopidogrel to ASA reduced risk of major vascular events, especially stroke, and increased risk of major hemorrhage.
Stroke occurred in 296 patients receiving ASA plus clopidogrel (2.4% per year) and 408 patients receiving ASA plus placebo (3.3% per year).95% CI: 0.62 to 0.83; p<0.001RR: 0.72
MI occurred in 90 patients receiving ASA plus clopidogrel (0.7% per year) and in 115 receiving ASA plus placebo (0.9% per year).95% CI: 0.59 to 1.03; p=0.08RR: 0.78
Major bleeding occurred in 251 patients receiving ASA plus clopidogrel (2.0% per year) and in 162 patients receiving ASA plus placebo (1.3% per year).95% CI: 1.29 to 1.92; p<0.001RR: 1.57
ACTIVE W, Connolly et al. (25)To determine if clopidogrel plus ASA was noninferior to oral anticoagulation therapy for prevention of vascular events in patients with AF.6,706Inclusion criteria: ECG evidence of AF; age ≥75 y; treatment for systemic HTN, previous stroke, TIA, or non-CNS systemic embolus; LV dysfunction with LVEF <45%; PAD; if age 55–74 y without 1 of the other inclusion criteria, then DM requiring drug therapy or previous CAD.
Exclusion criteria: Contraindication for clopidogrel or oral anticoagulant (ie, prosthetic mechanical heart valve), documented peptic ulcer disease within previous 6 mo, previous intracerebral hemorrhage, significant thrombocytopenia (platelet count <50×109 per liter) or mitral stenosis.
Primary outcome was first occurrence of stroke, non-CNS systemic embolism, MI, or vascular death.Composite of stroke, non-CNS embolus, MI, vascular death: 164 events in patients on oral anticoagulation (annual risk: 3.90%) and 234 events in patients on clopidogrel plus ASA (annual risk: 5.60%).95% CI: 1.18 to 1.76; p=0.0003RR: 1.44Oral anticoagulation with warfarin is superior to clopidogrel plus ASA in preventing vascular events, including stroke, in patients with AF.
Stroke (100 events for clopidogrel plus ASA; 59 events for oral anticoagulation).95% CI: 1.24 to 2.37; p=0.001RR: 1.72
Non-CNS embolism (18 events for clopidogrel plus ASA; 4 for oral anticoagulation).95% CI: 1.58 to 13.8; p=0.005RR: 4.66
Patients on oral anticoagulation who already received this treatment at study entry had a trend toward greater reduction in vascular events.95% CI: 1.19 to 1.80; p=0.0005RR: 1.50
And a lower risk of major bleeding on oral anticoagulation therapy.95% CI: 0.94 to 1.79; p=0.11RR: 1.30
Than patients not on oral anticoagulation therapy at entry.95% CI: 0.85 to 1.89; p=0.24RR: 1.27
ADONIS, Singh et al. (32)To investigate effect of dronedarone for maintenance of SR after electrical, pharmacologic, or spontaneous conversion from AF or atrial flutter.208 in placebo group and 417 in dronedarone groupInclusion criteria: Either sex, age at least 21 y, and at least 1 episode of AF (as seen on ECG) in preceding 3 mo and in SR for at least 1 h before randomization.
Exclusion criteria: Patients with permanent AF (ie, duration of at least 12 mo); women who could become pregnant and who were not using birth control; patients who had torsades de pointes; patients with persistent bradycardia of <50 bpm, PR interval of ≥0.28 s on ECG, second-degree (or higher) AVB, and clinically significant sinus-node disease without an implanted pacemaker; patients taking Class I or III antiarrhythmic agents; patients with NYHA class III or IV CHF; and patients with serum creatinine level ≥1.7 mg/dL (150 μmol/L), severe electrolyte abnormalities, and clinically significant hepatic, pulmonary, endocrine, or other disorders associated with AF.
Primary endpoint was time from randomization to first documented recurrence of AF. Secondary endpoints were symptoms and mean ventricular rate during first AF recurrence.Median times from randomization to documented recurrence of AF were 158 d in dronedarone group and 59 d in placebo group. At 12 mo, 61.1% of patients in dronedarone group and 72.8% of patients in placebo group had recurrence of AF.HR: 0.73Dronedarone was significantly more effective than placebo in maintaining SR.
Rhythm Control versus Rate Control for Atrial Fibrillation and Heart Failure (AF and CHF Investigators) (9)To investigate maintenance of SR (rhythm control) with ventricular rate control in patients with LVEF ≤35% and symptoms of CHF and history of AF.1,376 (682 in rhythm-control group and 694 in rate-control group)Inclusion criteria: LVEF ≤35% (measured by nuclear imaging, echocardiography, or cardiac angiography, with testing performed ≤6 mo before enrollment); history of CHF (defined as symptomatic NYHA class II or IV) within previous 6 mo, asymptomatic condition that patient had been hospitalized for HF during previous 6 mo, or LVEF ≤25%; history of AF (with ECG documentation) defined as 1 episode lasting for at least 6 h or requiring cardioversion within previous 6 mo or episode lasting for at least 10 min within previous 6 mo and previous electrical cardioversion for AF; and eligibility for long-term therapy in either of the 2 study groups.
Exclusion criteria: Persistent AF for >12 mo, reversible cause of AF or HF, decompensated HF within 48 h before intended randomization, use of antiarrhythmic drugs for other arrhythmias, second- or third-degree AVB (bradycardia of <50 bpm), history of long-QT syndrome, previous ablation of AV node, anticipated cardiac transplantation within 6 mo, renal failure requiring dialysis, lack of birth control in women of childbearing potential, estimated life expectancy <1 y, and age <18 y.
Primary outcome was time to death from CV causes.The primary outcome, death from CV causes, occurred in 182 patients (27%) in rhythm-control group and 175 patients (25%) in rate-control group.None of the secondary outcomes differed significantly between treatment groups.HR: 1.06The routine strategy of rhythm control does not reduce the rate of death from CV causes compared with a rate-control strategy in patients with AF and CHF.
95% CI: 0.86 to 1.30; p=0.53
Death from any cause (32% in rhythm-control group and 33% in rate-control group).95% CI: 0.80 to 1.17; p=0.73HR: 0.97
Ischemic or hemorrhagic stroke, 3% and 4%, respectively.95% CI: 0.40 to 1.35; p=0.32HR: 0.74
Worsening HF (defined as HF requiring hospitalization, administration of IV diuretic, or change in treatment strategy).95% CI: 0.72 to 1.06; p=0.17HR: 0.87
Composite outcome of death from CV causes, stroke, or worsening HF.95% CI: 0.77 to 1.06; p=0.20HR: 0.90
AFFIRM, Olshansky et al. (8)To evaluate and compare several drug classes for long-term ventricular rate control.2,027Inclusion criteria: (All criteria must have been met.) Episode of AF documented on ECG or rhythm strip within last 6 wk, age ≥65 y or <65 y plus ≥1 clinical risk factor for stroke (systemic HTN, DM, CHF, TIA, prior cerebral vascular accident, left atrium ≥50 mm on echocardiogram, fractional shortening <25% on echocardiogram [unless paced or LBBB present], or LVEF <40% (on radionuclide ventriculogram, contrast angiography, or quantitative echocardiography), duration of AF episodes in last 6 mo must total ≥6 h unless electrical and/or pharmacological cardioversion was performed before 6 h, duration of continuous AF must be <6 mo unless normal SR can be restored and maintained for ≥24 h in opinion of clinical investigator, patient (based on clinical and laboratory evaluation before randomization) must be eligible for both treatment groups based on history, patient must be eligible for ≥2 antiarrhythmic drugs (or 2 dose levels of amiodarone) and ≥2 rate-controlling drugs.
Exclusion criteria: Not presented based on judgment that certain therapies are contraindicated or inclusion would confound the result. Criteria included cardiac, other medical, and nonmedical.
Overall rate control with various drugs (average follow-up 3.5±1.3 y).Overall rate control was met in 70% of patients given beta blockers as the first drug (with or without digoxin) versus 54% with calcium channel blockers (with or without digoxin) and 58% with digoxin alone.
Multivariate analysis revealed a significant association between first drug class and several clinical variables, including gender, history of CAD, pulmonary disease, CHF, HTN, qualifying episode being first episode of AF, and baseline heart rate.
Rate control is possible in the majority of patients with AF. In the AFFIRM follow-up study, beta blockers were most effective. The authors noted frequent medication changes and drug combinations were needed.
ANDROMEDA, Kober et al. (30)To evaluate efficacy of dronedarone in reducing hospitalization due to CHF in patients with symptomatic HF.627Inclusion criteria: Patients age ≥18 y hospitalized with new or worsening HF and who had at least 1 episode of SOB on minimal exertion or at rest (NYHA class III or IV) or paroxysmal nocturnal dyspnea within 1 mo before admission and wall-motion index of no more than 1.2 (approximating EF of no more than 35%).
Exclusion criteria: Acute MI within 7 d before screening, heart rate <50 bpm, PR interval >0.28 s, sinoatrial block or second- or third-degree AVB not treated with pacemaker, history of torsades de pointes, corrected QT interval >500 ms, serum potassium level <3.5 mmol/L, use of Class I or III antiarrhythmic drugs, drugs known to cause torsades de pointes, or potent inhibitors of P450 CYP3A4 cytochrome system, other serious disease, acute myocarditis, constrictive pericarditis, planned or recent (within preceding month) cardiac surgery or angioplasty, clinically significant obstructive heart disease, acute pulmonary edema within 12 h before randomization, pregnancy or lactation, expected poor compliance, or participation in another clinical trial and previous treatment with dronedarone.
The primary endpoint was composite of death from any cause or hospitalization for HF.After inclusion of 627 patients, the trial was prematurely terminated for safety reasons. At a median follow-up of 2 mo, death had occurred in 8.1% of the dronedarone group and 3.8% of the placebo group.95% CI: 1.07 to 4.25; p=0.03HR: 2.13Dronedarone increased early mortality in patients recently hospitalized with symptomatic HF and depressed LV function. 96% of deaths were attributed to CV causes, predominantly progressive HF and arrhythmias.
After an additional 6 mo, 42 patients in the dronedarone group (13.5%) and 39 patients in the placebo group (12.3%) died.95% CI: 0.73 to 1.74; p=0.60HR: 1.13
The primary endpoint did not differ significantly between the 2 groups; there were 53 events in the dronedarone group (17.1%) and 40 events in the placebo group (12.6%).95% CI: 0.92 to 2.09; p=0.12HR: 1.38
ATHENA, Hohnloser et al. (29)To determine if dronedarone would reduce rate of composite outcome of hospitalization due to CV events or death in patients with AF.4,628Inclusion criteria: Patients with paroxysmal or persistent AF or atrial flutter with at least 1 of the following: age at least 70 y, arterial HTN, DM, previous stroke, TIA, systemic embolism, LA diameter ≥50 mm, and LVEF ≤40%.
Exclusion criteria: Permanent AF; unstable hemodynamic condition (ie, decompensated HF within previous 4 wk); NYHA class IV CHF; planned major surgery; acute myocarditis; bradycardia with a heart rate of <50 bpm or PR interval >0.28 s or previous clinically significant sinus-node disease; severe noncardiac illness limiting life expectancy; pregnancy, breast-feeding, or lack of adequate birth control among women of childbearing potential; calculated glomerular filtration rate at baseline <10 mL/min, potassium level <3.5 mmol/L if not currently being corrected, and requirement for concomitant medication that was prohibited.
Primary outcome was death or first hospitalization due to CV events. Secondary outcomes were death from any cause, death from CV causes, and hospitalization due to CV events.Primary outcome occurred in 734 patients (31.9%) in the dronedarone group and in 917 patients (39.4%) in the placebo group.95% CI: 0.69 to 0.84; p<0.001HR: 0.76Dronedarone reduced risk of hospitalization or death in patients with paroxysmal or persistent AF or atrial flutter, which was largely due to a reduction in hospitalization for AF. Death from any cause was not reduced. Adverse effects that were more common with dronedarone than placebo were bradycardia, prolonged QT, diarrhea, nausea, rash, and increase in serum creatinine.
116 deaths (5%) in the dronedarone group and 139 (6%) in the placebo group.95% CI: 0.66 to 1.08; p=0.18HR: 0.84
63 deaths from CV causes (2.7%) in the dronedarone group and 90 (3.9%) in the placebo group.95% CI: 0.51 to 0.98; p=0.03HR: 0.71
675 (29.3%) first hospitalizations due to CV events in the dronedarone group and 859 (36.9%) in the placebo group. A first hospitalization for AF occurred in 14.6% of the dronedarone group and 21.9% of the placebo group.95% CI: 0.67 to 0.82; p<0.001HR: 0.74
26 (1.1%) deaths from cardiac arrhythmia in the dronedarone group and 48 (2.1%) in the placebo group.95% CI: 0.34 to 0.88; p=0.01HR: 0.55
Analysis of stroke in ATHENA, Connolly et al. (37)To assess efficacy of dronedarone 400 mg bid for prevention of CV hospitalization or death from any cause in patients with AF/atrial flutter.4,628Inclusion criteria: Paroxysmal or persistent AF or atrial flutter and at least 1 additional risk factor for CV events, including age ≥75 y or age 70 y with ≥1 of the following: HTN, DM, prior stroke or TIA, LA enlargement (≥50 mm Hg), or depressed LVEF (<40%).
Exclusion criteria: Permanent AF, unstable hemodynamic situation, and NYHA class IV HF. Patients must have had both SR and AF or atrial flutter documented in 6 mo before enrollment.
Primary endpoint was first occurrence of CV hospitalization or death due to any cause. Analysis of stroke posthoc and not prespecified.Risk of stroke decreased from 1.8% per year to 1.2% per year.95% CI: 0.46 to 0.96; p=0.027HR: 0.66Fewer strokes occurred in the dronedarone group, but this finding was not anticipated and was not prespecified. Whether it was a chance finding or due to a beneficial effect of the drug is not certain.
DAFNE, Touboul et al. (33)To determine most appropriate dose of dronedarone for prevention of AF after cardioversion.474Inclusion criteria: Either sex, age 21–85 y, with persistent AF (72-h and 12-mo duration) in which cardioversion and antiarrhythmic treatment are warranted. AF either lone or associated with ischemic or hypertensive heart disease or DCM.
Exclusion criteria: More than 2 cardioversions in last 6 mo, acute reversible cause; atrial flutter as presenting arrhythmia; unstable angina or recent MI; QT interval >500 ms or history of torsades de pointes; severe bradycardia; advanced AVB; treatment with other antiarrhythmic drugs, NYHA class III or IV CHF; LVEF <35%; Wolff-Parkinson-White syndrome; ICD.
Primary endpoint was time to first documented AF recurrence (AF defined as episode lasting for at least 10 min and documented by 2 distinct ECGs separated by same time duration).Increased time to AF relapse with 800 mg of dronedarone (effect less apparent at higher doses). Median time to first AF recurrence was 5.3 d in placebo group and at 60 d in the 800-mg dronedarone group. At 6 mo 35% of patients treated with 800-mg dronedarone remained in SR versus 10% of placebo group.95% CI: 28 to 72; p=0.001RR reduction: 55%Dronedarone 800 mg qd appeared to be safe and effective for prevention of AF relapses after cardioversion.
DIONYSOS, Le Heuzey et al. (36)To compare efficacy and safety of amiodarone and dronedarone in patients with persistent AF.504 (249 dronedarone 400 mg bid; 255 amiodarone 600 mg qd for 28 d, then 200 mg qd).Inclusion criteria: Age ≥21 y, documented AF for >72 h in patients for whom cardioversion and antiarrhythmic treatment were indicated and who were receiving oral anticoagulants.
Exclusion criteria: Previous chronic treatment with amiodarone, hypo- or hyperthyroidism or other contraindications to amiodarone, corrected QT interval ≥500 ms, paroxysmal AF, atrial flutter, severe NYHA class III or IV CHF, severe bradycardia, or high-degree AVB. Patients in whom contraindicated concomitant treatment was mandatory were excluded (including Vaughan Williams Class I and III antiarrhythmic drugs; drugs that cause torsades de pointes; potent inhibitors of cytochrome P[CYP] 3A4; and substrates of CYP3A4 with narrow therapeutic margin).
Primary composite endpoint was recurrence of AF (including unsuccessful electrical cardioversion, no spontaneous conversion, and no electrical cardioversion) or premature discontinuation of study. MSE was occurrence of thyroid, hepatic, pulmonary, neurologic, dermatologic, ocular, or gastrointestinal-specific events or premature discontinuation of study drug after adverse event.Dronedarone 75.1%; amiodarone 58.8%
AF recurrence after successful cardioversion: 36.5% with dronedarone and 24.3% with amiodarone.
Premature discontinuation of drug tended to be less frequent with dronedarone (10.4% versus 13.3%).
95% CI: 1.28 to 1.98; p<0.0001HR: 1.59Dronedarone was less effective than amiodarone in decreasing AF recurrence; however, it had a better safety profile.
MSE was 39.3% with dronedarone and 44.5% with amiodarone at 12 mo, mainly driven by fewer thyroid, neurologic, dermatologic, and ocular events in dronedarone group.95% CI: 0.60 to 1.07; p=0.129HR: 0.80
EURIDIS, Singh et al. (32)To evaluate dronedarone compared with placebo for maintenance of SR after electrical, pharmacologic, or spontaneous conversion from AF or atrial flutter612Inclusion criteria: Either sex, age at least 21 y, and at least 1 episode of AF (as seen on ECG) in preceding 3 mo and in SR for at least 1 h before randomization.
Exclusion criteria: Patients with permanent AF (ie, duration of at least 12 mo); women who could become pregnant and who were not using birth control; patients who had torsades de pointes; patients with persistent bradycardia of <50 bpm, a PR interval of ≥0.28 on ECG, second-degree (or higher) AVB, and clinically significant sinus-node disease without an implanted pacemaker; patients taking Class I or III antiarrhythmic agents; patients with NYHA class III or IV CHF; and patients with serum creatinine level ≥1.7 mg/dL (150 μmol/L), severe electrolyte abnormalities, and clinically significant hepatic, pulmonary, endocrine, or other disorders associated with AF.
Primary endpoint was time from randomization to first documented recurrence of AF. Secondary endpoints were symptoms and mean ventricular rate during first AF recurrence.Median times from randomization to documentated recurrence of AF were 96 d in dronedarone group and 41 d in placebo group.
At 12 mo 67.1% of patients in dronedarone group and 77.5% of placebo group had recurrence of AF.
95% CI: 0.64 to 0.96; p=0.01HR: 0.78Dronedarone was significantly more effective than placebo in maintaining SR.
RACE II, Van Gelder et al. (3)To investigate if lenient rate control is not inferior to strict control for preventing CV morbidity and mortality in patients with permanent AF.614Inclusion criteria: Permanent AF up to 12 mo, age ≤80 y, mean resting heart rate >80 bpm, and current use of oral anticoagulation therapy (or ASA if no risk factors for thromboembolic complications present).
Exclusion Criteria: Paroxysmal AF; contraindications for either strict or lenient rate control (eg, previous adverse effects on negative chronotrophic drugs); unstable HF defined as NYHA class IV HF or HF necessitating hospital admission <3 mo before inclusion; cardiac surgery <3 mo ago; any stroke; current or foreseen pacemaker, ICD, and/or cardiac resynchronization therapy; signs of sick sinus syndrome or AV conduction disturbances (ie, symptomatic bradycardia or asystole >3 s or escape rate <40 bpm in awake symptom-free patients; untreated hyperthyroidism or <3 mo euthyroidism; inability to walk or ride a bike.
Composite of death from CV causes, hospitalization for HF, and stroke, systemic embolism, bleeding, and life-threatening arrhythmic events. Follow-up duration 2 y, with maximum 3 y.Primary outcome incidence at 3 y was 12.9% in lenient-control group and 14.9% in strict-control group. Absolute difference with respect to lenient-control group of −2.0%.90% CI: 0.58 to 1.21; p=0.001HR: 0.84Lenient rate control is as effective as strict rate control and easier to achieve in patients with permanent AF.
Absolute difference −2.0%
Absolute difference, 90% CI: −7.6 to 3.5; p<0.001
More patients in lenient-control group met heart rate target or targets (304 [97.7%] versus 203 [67.0%] in strict-control group).
Frequencies of symptoms and adverse events were similar in the 2 groups.
p<0.001
ThermoCool, Wilber et al. (51)To investigate catheter ablation with ADT in patients with symptomatic AF.167Inclusion criteria: Enrollment required at least 3 episodes of symptomatic AF (≥1 episode verified by ECG) within 6 mo before randomization and not responding to at least 1 antiarrhythmic drug (class I, class III, or AV nodal blocker).
Exclusion criteria: AF >30 d, <18 y, EF <40%, previous ablation for AF, documented LA thrombus, amiodarone therapy in previous 6 mo, NYHA class III or IV, MI within previous 2 mo, CABG within previous 12 mo, thromboembolic event in previous 12 mo, severe pulmonary disease, prior valvular cardiac surgical procedure, presence of ICD, contraindication to antiarrhythmic or anticoagulation medications, life expectancy <12 mo, and LA size of at least 50 mm in parasternal long axis.
Primary endpoint was freedom from protocol-defined treatment failure, which included documented symptomatic paroxysmal AF during effectiveness evaluation period.66% of patients in catheter ablation group remained free from protocol-defined treatment failure versus 16% of patients treated with ADT.95% CI: 0.19 to 0.47; p<0.001HR: 0.30Catheter ablation is more effective than medical therapy alone in preventing recurrent symptoms of paroxysmal AF in patients who have already failed treatment with 1 antiarrhythmic drug. Ideal candidates for catheter ablation are younger patients with minimal structural abnormalities and multiple symptomatic episodes of paroxysmal AF over time despite appropriate pharmacological therapy.
70% of patients treated by catheter ablation remained free of symptomatic recurrent atrial arrhythmia versus 19% of patients treated with ADT.95% CI: 0.15 to 0.39; p>0.001HR: 0.24
63% of patients treated by catheter ablation were free of recurrent atrial arrhythmia versus 17% of patients treated with ADT.95% CI: 0.18 to 0.45; p<0.001HR: 0.29

ADT indicates antiarrhythmic drug therapy; AF, atrial fibrillation; ASA, aspirin; AV, atrioventricular; AVB, atrioventricular block; bid, twice a day; bpm, beats per minute; CABG, coronary artery bypass graft surgery; CAD, coronary artery disease; CHF, congestive heart failure; CI: confidence interval; CNS, central nervous system; CYP, cytochrome P; CV, cardiovascular; d, day; DCM, dilated cardiomyopathy; DM, diabetes mellitus; ECG, electrocardiogram; EF, ejection fraction; h, hour; HF, heart failure; HR, hazard ratio; HTN, hypertension; ICD, implantable cardioverter-defibrillator; IV, intravenous; LA, left atrial; LBBB, left bundle-branch block; LV, left ventricular; LVEF, left ventricular ejection fraction; MI, myocardial infarction; mm, millimeter; mo, month; ms, milliseconds; MSE, main safety endpoint; NYHA, New York Heart Association; PAD, peripheral arterial disease; PR interval, interval between onset of P wave and onset of QRS complex on an ECG; PVD, peripheral vascular disease; qd, once per day; RR, relative risk; s, seconds; SD, standard deviation; SOB, short of breath; SR, sinus rhythm; TIA, transient ischemic attack; wk, week; and y, year.