T h e n e w e ngl a nd j o u r na l o f m e dic i n e original article in Patients with Atrial Fibrillation The ACTIVE I Investigators* A bs tr ac t The members of the writing group (Salim Yusuf, M.B., B.S., D.Phil., Jeff S. Healey, M.D., Janice Pogue, M.Sc., Susan Chrolavicius, R.N., Marcus Flather, M.B., B.Ch., Robert G. Hart, M.D., Stefan H. Hohnloser, M.D., Campbell D. Joyner, M.D., Marc A. Pfeffer, M.D., and Stuart J. Connolly, M.D.), assume responsibility for the overall content and integrity of the article. Address reprint requests to Dr. Yusuf at the Population Health Research Institute, David Braley Cardiac, Vascular and Stroke Research Institute, Hamilton Health Sciences and McMaster University, 237 Barton St. East, Hamilton, ON L8L 2X2, Canada, or at yusufs@mcmaster.ca. * The Atrial Fibrillation Clopidogrel Trial with for Prevention of Vascular Events (ACTIVE I) investigators are listed in the Supplementary Appendix, available at NEJM.org. The affiliations of the writing group are listed in the Appendix. N Engl J Med 2011;364:928-38. Copyright 2011 Massachusetts Medical Society. Background The risk of cardiovascular events among patients with atrial fibrillation is high. We evaluated whether irbesartan, an angiotensin-receptor blocker, would reduce this risk. Methods We randomly assigned patients with a history of risk factors for stroke and a systolic blood pressure of at least 110 mm Hg to receive either irbesartan at a target dose of 300 mg once daily or double-blind placebo. These patients were already enrolled in one of two trials (of clopidogrel plus aspirin versus aspirin alone or versus oral anticoagulants). The first coprimary outcome was stroke, myocardial infarction, or death from vascular causes; the second was this composite outcome plus hospitalization for heart failure. Results A total of 9016 patients were enrolled and followed for a mean of 4.1 years. The mean reduction in systolic blood pressure was 2.9 mm Hg greater in the irbesartan group than in the placebo group, and the mean reduction in diastolic blood pressure was 1.9 mm Hg greater. The first coprimary outcome occurred at a rate of 5.4% per 100 person-years in both groups (hazard ratio with irbesartan, 0.99; 95% confidence interval [CI], 0.91 to 1.08; P = 0.85). The second coprimary outcome occurred at a rate of 7.3% per 100 person-years among patients receiving irbesartan and 7.7% per 100 person-years among patients receiving placebo (hazard ratio, 0.94; 95% CI, 0.87 to 1.02; P = 0.12). The rates of first hospitalization for heart failure (a prespecified secondary outcome) were 2.7% per 100 person-years among patients receiving irbesartan and 3.2% per 100 person-years among patients receiving placebo (hazard ratio, 0.86; 95% CI, 0.76 to 0.98). Among patients who were in sinus rhythm at baseline, there was no benefit of irbesartan in preventing hospitalization for atrial fibrillation or atrial fibrillation recorded on 12-lead electrocardiography, nor was there a benefit in a subgroup that underwent transtelephonic monitoring. More patients in the irbesartan group than in the placebo group had symptomatic hypotension (127 vs. 64) and renal dysfunction (43 vs. 24). Conclusions did not reduce cardiovascular events in patients with atrial fibrillation. (Funded by Bristol-Myers Squibb and Sanofi-Aventis; ClinicalTrials.gov number, NCT00249795.) 928
in Atrial Fibrillation The most common risk factor for the development of atrial fibrillation is hypertension, 1 and two common complications in atrial fibrillation are stroke and heart failure, both of which are related to elevated blood pressure. 2 Randomized trials of blood-pressure reduction have shown a reduced risk of stroke and heart failure in populations without atrial fibrillation. 3-6 However, a similar relationship has yet to be shown in patients with atrial fibrillation, and data from trials of blood-pressure reduction in patients with atrial fibrillation are lacking. Blockers of the renin angiotensin aldosterone system are hypothesized to have specific properties that beneficially alter atrial mechanical and electrical remodeling. 7-9 Secondary analyses of data from trials of angiotensin-converting enzyme (ACE) inhibitors or angiotensin-receptor blockers (ARBs) in patients with heart failure or left ventricular dysfunction suggest that these agents prevent atrial fibrillation. 10 We hypothesized that an ARB may prevent cardiovascular events and enhance the maintenance of sinus rhythm in patients with intermittent atrial fibrillation by reducing blood pressure and by specific effects related to blockade of the renin angiotensin aldosterone system. 7-9 In the Atrial Fibrillation Clopidogrel Trial with for Prevention of Vascular Events (ACTIVE I), patients with atrial fibrillation who were already enrolled into two large parallel trials evaluating clopidogrel plus aspirin 11,12 underwent a second randomization, to irbesartan or placebo. The study involved the use of a partial factorial design to determine the effects of irbesartan on the risk of death from vascular causes, myocardial infarction, or stroke; on this composite outcome plus hospitalization for heart failure; and on recurrent atrial fibrillation in patients who were in sinus rhythm at study entry. Me thods Study Design and Protocol A steering committee was responsible for the design, conduct, and reporting of the study. The Population Health Research Institute designed the study, collected all data, adjudicated outcomes, and performed the analysis. Representatives of Bristol-Myers Squibb and Sanofi-Aventis, the study sponsors, were members of the steering committee, but the sponsors had no direct role in the collection, analysis, or interpretation of study data or in the writing of the manuscript. The members of the writing group had full access to all data and had final responsibility for publication decisions. The protocol was approved by the appropriate regulatory authorities and the ethics committee at each site. The protocol, including the statistical analysis plan, is available with the full text of this article at NEJM.org. All authors attest that the study was performed in accordance with the protocol and vouch for the accuracy and completeness of the reported data. All patients provided written informed consent. Patient Selection and Study Drugs Patients were eligible to enroll in the trial if they had permanent atrial fibrillation or had at least two episodes of intermittent atrial fibrillation in the previous 6 months. In addition, patients were required to have one of the following risk factors: an age of 75 years or older; treatment for hypertension; a history of stroke, transient ischemic attack, or non-central nervous system systemic embolism; a left ventricular ejection fraction of less than 45%; peripheral vascular disease; or an age of 55 to 74 years plus either diabetes mellitus or coronary artery disease. Patients were excluded if they required clopidogrel or oral anticoagulants, had received a diagnosis of peptic ulcer disease within the previous 6 months, had a history of intracerebral hemorrhage, or had thrombocytopenia or mitral stenosis. In one of the parallel trials, ACTIVE W, patients were randomly assigned to receive clopidogrel plus aspirin or oral anticoagulation. In the other parallel trial, ACTIVE A, patients received aspirin (75 to 100 mg per day) and were randomly assigned to either clopidogrel at a dose of 75 mg or placebo. Eligible patients from these two trials (who had a systolic systemic arterial pressure of at least 110 mm Hg and who were not receiving an ARB) were also eligible to be enrolled in this trial, ACTIVE I, and patients who provided consent were randomly assigned to receive irbesartan once daily (at a dose of 150 mg per day for 2 weeks, and 300 mg per day thereafter) or placebo. Assignment was performed with the use of a central randomization system, stratified according to center and enrollment into ACTIVE A or ACTIVE W. Study Outcomes The first coprimary outcome was the composite of stroke, myocardial infarction, or death from 929
T h e n e w e ngl a nd j o u r na l o f m e dic i n e vascular causes, and the second coprimary outcome was this composite outcome plus hospitalization for heart failure. (For definitions, see Appendix A in the Supplementary Appendix, available at NEJM.org). All components of the coprimary outcomes were reviewed by an adjudication committee; members of the committee were unaware of the study-drug assignments. Other outcomes were recurrence of atrial fibrillation on 12-lead electrocardiography performed at 2 years and at the end of the study, the rate of rehospitalization for atrial fibrillation, and recurrence of atrial fibrillation with the use of transtelephonic monitoring. Transtelephonic Monitoring Substudy Patients from selected centers who were in sinus rhythm at the time of randomization participated in the substudy on the recurrence of atrial fibrillation. The patients were instructed how to use transtelephonic event monitors (ER300, Braemar) Table 1. Baseline Characteristics of the Patients.* Variable (N = 4518) (N = 4498) Age yr 69.5±9.7 69.6±9.7 Blood pressure mm Hg Systolic 138.3±17.6 138.2±17.2 Diastolic 82.6±11.5 82.2±11.1 Heart rate beats/min 75.3±14.4 74.9±14.4 Body-mass index 28.9±4.90 28.7±4.9 CHADS 2 score 2.0±1.1 2.0±1.07 Creatinine clearance ml/min 75.5±31.6 74.5±31.4 Left ventricular ejection fraction % 55±13 55±13 Male sex no. (%) 2745 (60.8) 2730 (60.7) Type of atrial fibrillation no. (%) Permanent 2982 (66.0) 2898 (64.4) Paroxysmal 886 (19.6) 921 (20.5) Persistent 644 (14.3) 668 (14.9) Missing data 6 (0.1) 11 (0.2) Markers of high risk no. (%) Age 75 yr 1551 (34.3) 1576 (35.0) Hypertension 3982 (88.1) 3947 (87.8) Stroke, transient ischemic attack, or systemic embolism 613 (13.6) 599 (13.3) Left ventricular dysfunction 568 (12.6) 561 (12.5) Peripheral artery disease 103 (2.3) 133 (3.0) Age 55 74 yr plus diabetes or coronary artery disease 1008 (22.3) 970 (21.6) Heart failure 1460 (32.3) 1421 (31.6) Diabetes mellitus 906 (20.1) 881 (19.6) Previous cardioversion no. (%) 1669 (36.9) 1704 (37.9) Ethnic group no. (%) European 3394 (75.1) 3375 (75.0) Non-European 1124 (24.9) 1123 (25.0) ECG findings at randomization no. (%) Sinus rhythm 847 (18.7) 883 (19.6) Atrial fibrillation 3457 (76.5) 3390 (75.4) Atrial flutter 37 (0.8) 42 (0.9) Other rhythm 177 (3.8) 183 (4.1) ECG evidence of left ventricular hypertrophy 765 (16.9) 749 (16.7) Intraventricular conduction delay 622 (13.8) 597 (13.3) 930
in Atrial Fibrillation Table 1. (Continued.) Variable (N = 4518) (N = 4498) Smoking status no. (%) Current smoker 331 (7.3) 367 (8.2) Former smoker 1910 (42.3) 1897 (42.2) Alcohol consumption >1 drink/day no. (%) 362 (8.0) 374 (8.3) Use of medication at baseline no. (%) Aspirin 2652 (58.7) 2666 (59.3) Thienopyridine 146 (3.2) 144 (3.2) Oral anticoagulant 1721 (38.1) 1692 (37.6) Angiotensin-converting enzyme inhibitor 2720 (60.2) 2724 (60.6) Angiotensin-receptor blocker 232 (5.1) 211 (4.7) Beta-blocker 2458 (54.4) 2455 (54.6) Digoxin 1586 (35.1) 1562 (34.7) Calcium-channel blocker 1220 (27.0) 1225 (27.2) Diuretic 2453 (54.3) 2432 (54.1) Vasodilator 537 (11.9) 498 (11.1) Lipid-lowering drug 1405 (31.1) 1459 (32.4) Antiarrhythmic drug 1025 (22.7) 1041 (23.1) * Plus minus values are means ±SD. The body-mass index is the weight in kilograms divided by the square of the height in meters. ECG denotes electrocardiographic. The CHADS 2 score, an index of the risk of stroke in patients with atrial fibrillation, ranges from 1 to 6, with higher scores indicating an increased risk of stroke. A total of 910 patients (10.1%) 454 (10.0%) in the irbesartan group and 456 (10.1%) in the placebo group were receiving drugs to slow the heart rate (verapamil or diltiazem). for two monitoring periods (between randomization and 4 months and between 18 and 20 months). Patients were asked to transmit a recording and to complete a questionnaire if they had symptoms of atrial fibrillation. They were also asked to transmit weekly recordings taken while they were asymptomatic to detect asymptomatic atrial fibrillation. All transtelephonic monitoring transmissions were analyzed at a core laboratory in Hamilton by investigators who were unaware of the study-drug assignments. Statistical Analysis We originally anticipated that enrollment of 9000 patients and follow-up for a mean of 3 years would result in 89% statistical power to detect a reduction of 15% in the risk of the first coprimary outcome (at an alpha level of 0.045) if the event rate per year in the control group was 7%, and 92% power to detect a 15% risk reduction in the second coprimary outcome (at an alpha level of 0.01), assuming an event rate of 11% per year. Because overall event rates were lower than expected (5.4% and 7.7%, respectively), follow-up was extended by 1.1 years (mean, 4.1 years), with a power of 80% to detect risk reductions of 15.5% and 15.8% in the two coprimary outcomes, respectively. An independent data and safety monitoring board monitored the accumulating data during the trial according to prespecified guidelines. 11,12 Analyses were based on the intention-to-treat principle, and the time to a first occurrence of an outcome in the two study groups was compared with the use of Cox proportional-hazards regression models, stratified according to enrollment in either ACTIVE A or ACTIVE W. Subgroup analyses used tests of interactions in the same models. Modeling of proportional rate and mean functions was used to compare the time to recurrent outcomes in the two groups. 13 The total number of hospital admissions in the two study groups was compared with the use of Poisson regression. 14 Differences in blood pressure were compared with the use of analysis of variance. The prevalence of atrial fibrillation on followup electrocardiography was compared between study groups with the use of Fisher s exact test, 931
T h e n e w e ngl a nd j o u r na l o f m e dic i n e and the treatment effect was expressed as relative risks. The time to outcomes such as cardioversion or hospitalization for atrial fibrillation was assessed with the use of the Kaplan-Meier method, and differences between the study groups were compared with the use of the log-rank test. In the transtelephonic monitoring substudy, the time to a first episode of recurrent atrial fibrillation in each group was examined with the use of a plot of conditional probability. A second analysis was conducted with the use of the recurrent-events (proportional-means) model 13 in order to include each individual episode of recurrent atrial fibrillation. R esult s Study Population Table 1 lists the baseline characteristics of the patients. The mean age was 69.6 years, 60.7% of the patients were men, and the mean CHADS 2 score was 2.0 (the CHADS 2 score, an index of the risk of stroke in patients with atrial fibrillation, ranges from 1 to 6, with higher scores indicating an increased risk of stroke). Among patients who were enrolled in both ACTIVE W and ACTIVE I, a total of 2048 (51.0%) were randomly assigned to receive an oral anticoagulant, and 1970 (49.0%) were randomly assigned to receive clopidogrel plus aspirin. (After termination of ACTIVE W, 16.7% of patients received aspirin, and 86.6% of patients were treated with nonstudy oral anticoagulants.) Among patients enrolled in both ACTIVE A and ACTIVE I, a total of 2494 patients (49.9%) were randomly assigned to receive clopidogrel plus aspirin and 2504 patients (50.1%) were assigned to aspirin alone. A total of 19.2% of patients were in sinus rhythm at enrollment. Commonly used medications included rate-control drugs (betablockers in 54.5% of patients, digoxin in 34.9%, and calcium-channel blockers that slow heart rate in 10.1%) and antiarrhythmic drugs (in 22.9%). Follow-up and Adherence to the Study-Drug Regimens The mean follow-up was 4.1 years (median, 4.5 years) and was complete for vital status in 8976 of 9016 patients (99.6%). The rates of permanent discontinuation of study drugs were 14.8% with irbesartan and 13.7% with placebo at 1 year, 20.6% and 19.4% at 2 years, and 30.3% and 30.3% at 4 years, respectively. The most common reason for discontinuation of the study medication was withdrawal of consent (accounting for about half the cases of discontinuation), and the incidence of symptomatic hypotension was higher among patients who received irbesartan than among patients who received placebo (2.8% [127 patients] vs. 1.4% [64 patients], P<0.001). Blood Pressure and Antihypertensive Therapy At study entry, the mean blood pressure was 138.3 mg Hg systolic and 82.6 mm Hg diastolic in the irbesartan group and 138.2 mm Hg systolic and 82.2 mm Hg diastolic in the placebo group. The mean reduction in blood pressure was 6.8 mm Hg systolic and 4.5 mm Hg diastolic in patients who received irbesartan, as compared with 3.9 mm Hg systolic and 2.6 mm Hg diastolic in patients who received placebo (mean difference during the trial, 2.9 mm Hg systolic and 1.9 mm Hg diastolic). At 2 years, 10.7% of patients receiving irbesartan and 7.7% of patients receiving placebo were not receiving any bloodpressure lowering agent; 55.3% of patients receiving irbesartan and 53.3% of patients receiving placebo were receiving one or two nonstudy drugs to reduce blood pressure, and 34.1% and 39.1%, respectively, were receiving three or more nonstudy drugs to reduce blood pressure. At 2 years, the mean number of such drugs per patient was 2.00 among patients who received irbesartan as compared with 2.15 among patients who received placebo (P<0.001). Coprimary Outcomes As shown in Table 2, the first coprimary outcome occurred in 5.4% of patients per 100 patient-years in both study groups (hazard ratio with irbesartan, 0.99; 95% confidence interval [CI], 0.91 to 1.08; P = 0.85) (Fig. 1A), with no significant difference in the second coprimary outcome (7.3% per 100 patient-years with irbesartan and 7.7% per 100 patient-years with placebo; hazard ratio, 0.94; 95% CI, 0.87 to 1.02; P = 0.12) (Fig. 1B). Analysis of recurrent events, which accounts for multiple events in the same patient, showed little difference in the first coprimary outcome (1100 vs. 1123 events; hazard ratio, 0.97; 95% CI, 0.89 to 1.07; P = 0.58) but a significant reduction in the second coprimary outcome (1791 vs. 1993; hazard ratio, 0.89; 95% CI, 0.82 to 0.98; P = 0.02). The only component of the coprimary outcome that showed a nominally significant reduction with 932
in Atrial Fibrillation Table 2. Relative Risks of Primary and Key Secondary Outcomes. Outcome Coprimary outcomes Stroke, myocardial infarction, or death from vascular causes Stroke, myocardial infarction, hospitalization for heart failure, or death from vascular causes Recurrent events Stroke, myocardial infarction, or death from vascular causes Stroke, myocardial infarction, hospitalization for heart failure, or death from vascular causes Cerebrovascular and systemic embolic events (N = 4518) (N = 4498) no. of patients with first events (%/100 patient-yr of follow-up) Relative Risk (95% CI) P Value 963 (5.4) 963 (5.4) 0.99 (0.91 1.08) 0.85 1236 (7.3) 1291 (7.7) 0.94 (0.87 1.02) 0.12 1100 (24.3) 1123 (25.0) 0.97 (0.89 1.07) 0.58 1791 (39.6) 1993 (44.3) 0.89 (0.82 0.98) 0.02 Stroke 379 (2.1) 411 (2.3) 0.91 (0.79 1.05) 0.20 Transient ischemic attack 130 (0.7) 150 (0.8) 0.86 (0.68 1.09) 0.21 Non central nervous system systemic embolism 47 (0.3) 64 (0.3) 0.72 (0.50 1.05) 0.09 Stroke, transient ischemic attack, or non central nervous system embolism Types of stroke 515 (2.9) 584 (3.3) 0.87 (0.77 0.98) 0.02 Ischemic 308 (1.7) 329 (1.8) 0.93 (0.79 1.08) 0.35 Primary hemorrhagic 33 (0.2) 46 (0.2) 0.71 (0.46 1.11) 0.14 Hemorrhagic transformation of ischemic stroke 6 (<0.1) 19 (0.1) 0.31 (0.13 0.78) 0.01 Primary hemorrhagic or secondary transformation 39 (0.2) 65 (0.4) 0.60 (0.40 0.89) 0.01 Stroke type uncertain 46 (0.2) 55 (0.3) 0.83 (0.56 1.23) 0.35 Myocardial infarction 143 (0.8) 135 (0.7) 1.05 (0.83 1.33) 0.67 Hospitalization for heart failure 482 (2.7) 551 (3.2) 0.86 (0.76 0.98) 0.02 Death from vascular causes 666 (3.6) 646 (3.5) 1.02 (0.92 1.14) 0.67 Death from any cause 949 (5.1) 929 (5.0) 1.01 (0.93 1.11) 0.75 irbesartan was a first hospitalization for heart failure (482 patients with events in the irbesartan group vs. 551 in the placebo group; hazard ratio, 0.86; 95% CI, 0.76 to 0.98; P = 0.02). Cerebrovascular Events and Systemic Embolism There were trends toward fewer strokes with irbesartan than with placebo (379 vs. 411), fewer transient ischemic attacks (130 vs. 150), and fewer cases of systemic emboli (47 vs. 64). Thus, a post hoc analysis of the composite of any of these cerebrovascular and embolic events was nominally significant (515 vs. 584; hazard ratio, 0.87; 95% CI, 0.77 to 0.98; P = 0.02). Recurrence of Atrial Fibrillation Of the 9005 patients who underwent a baseline electrocardiographic examination, 6926 (76.8%) were in atrial fibrillation, 1730 (19.2%) were in sinus rhythm, and 349 (3.9%) had a paced or other rhythm at baseline. Among patients who were in atrial fibrillation at baseline, sinus rhythm was present on one follow-up electrocardiographic examination in 10.6% of patients who received irbesartan as compared with 9.7% of patients who received placebo (relative risk, 1.10; 95% CI, 0.94 to 1.28; P = 0.26). In patients with sinus rhythm at baseline, atrial fibrillation was present at one follow-up electrocardiographic examination in 36.8% of patients who received irbesartan 933
T h e n e w e ngl a nd j o u r na l o f m e dic i n e A Stroke, Myocardial Infarction, or Death from Vascular Causes 0.4 Cumulative Hazard Rate No. at Risk 0.3 0.2 0.1 Hazard ratio, 0.99 (95% CI, 0.91 1.08) P=0.85 0.0 0 1 2 3 4 4.5 Years since Randomization B Stroke, Myocardial Infarction, Death from Vascular Causes, or Hospitalization for Heart Failure 0.4 Cumulative Hazard Rate No. at Risk 4498 4518 0.3 0.2 0.1 4195 4220 3912 3926 3647 3669 2737 2781 2160 2170 Hazard ratio, 0.94 (95% CI, 0.87 1.02) P=0.12 0.0 0 1 2 3 4 4.5 Years since Randomization 4498 4518 4035 4084 3690 3741 3402 3466 2523 2598 Figure 1. Kaplan Meier Curves for the Coprimary Study Outcomes. 1979 2019 as compared with 38.1% of patients who received placebo (relative risk, 0.97; 95% CI, 0.85 to 1.10; P = 0.61). did not significantly reduce the risk of hospitalization for atrial fibrillation (hazard ratio, 0.95; 95% CI, 0.85 to 1.07; P = 0.41) (Fig. 2A) or the risk of cardioversion (hazard ratio, 0.97; 95% CI, 0.86 to 1.10; P = 0.67). Transtelephonic Monitoring Substudy A total of 185 patients who were in sinus rhythm at randomization participated in the substudy of recurrence of atrial fibrillation (86 patients who received irbesartan and 99 patients who received placebo). Of 1202 episodes of atrial fibrillation captured during transtelephonic monitoring, only 665 were symptomatic (55.3%). The most common symptoms were: palpitations (in 86.8% of the patients), fatigue (in 43.8%), breathlessness (in 31.9%), dizziness (in 18.4%), chest pain (in 13.9%), and sweating (in 13.1%). In all, 56.7% of symptomatic transmissions and 9.0% of asymptomatic transmissions showed atrial fibrillation. Atrial fibrillation recurred at least once in 68.6% of patients randomly assigned to irbesartan and in 62.6% of patients randomly assigned to placebo (hazard ratio, 1.14; 95% CI, 0.80 to 1.64; P = 0.46) (Fig. 2B). A recurrent-events analysis did not show an effect of irbesartan (hazard ratio, 1.21; 95% CI, 0.72 to 2.04; P = 0.46). Hospitalizations The number of patients with a first hospitalization for cardiovascular events was 1846 (40.9%) in the irbesartan group as compared with 1900 (42.2%) in the placebo group (relative risk, 0.97; 95% CI, 0.92 to 1.02). As shown in Table 3, there were fewer hospital admissions for cardiovascular events in the irbesartan group than in the placebo group (3817 vs. 4060; difference, 243; P = 0.003). There were also fewer hospitalizations for noncardiovascular events with irbesartan (difference, 60), so that the total number of hospitalizations was significantly reduced (6519, vs. 6822 with placebo; difference, 303; P = 0.004). The mean duration of a hospital stay for cardiovascular events was 9.55 days in the irbesartan group as compared with 9.84 in the placebo group (P = 0.28). Thus, there was a significant reduction in the total hospital days for cardiovascular events among patients who received irbesartan (36,447 vs. 39,960; difference, 3513; P<0.001). Data on the duration of hospitalizations for noncardiovascular events were not collected. Prespecified Subgroup Analyses There was no significant interaction for the effects of irbesartan on either coprimary outcome according to whether patients were enrolled in ACTIVE A or ACTIVE W or according to baseline use or nonuse of ACE inhibitors and baseline blood pressure (Fig. 3). The reductions in hospitalizations for heart failure in the irbesartan group were consistent whether patients had a history of heart failure at baseline (hazard ratio, 0.90; 95% CI, 0.75 to 1.08) or did not have a history of heart failure (hazard ratio, 0.81; 95% CI, 0.69 to 0.96; P = 0.40 for interaction). A measure of left ventricular dysfunction was available in 4810 patients, among whom 1296 (26.9%) had a normal ejection 934
in Atrial Fibrillation fraction. In this subgroup, 122 patients were hospitalized for heart failure in the irbesartan group as compared with 150 in the placebo group (relative risk, 0.81; 95% CI, 0.63 to 1.02). Adverse Events More patients in the irbesartan group than in the placebo group discontinued the study medication because of symptomatic hypotension (127 vs. 64, P<0.001). did not significantly increase the number of patients with either a doubling of the serum creatinine level or an increase in the potassium level to 6.0 mmol per liter or more (37 patients who received irbesartan vs. 34 patients who received placebo). However, any renal dysfunction leading to drug discontinuation was more frequent in the irbesartan group than in the placebo group (43 patients vs. 24 patients, P = 0.02). Four patients in the irbesartan group required dialysis; three of whom were also receiving an ACE inhibitor; none of the patients in the placebo group required dialysis. Discussion A Hospitalization for Atrial Fibrillation 0.20 Cumulative Hazard Rate No. at Risk 0.15 0.10 0.05 Hazard ratio, 0.95 (95% CI, 0.84 1.07) P=0.41 0.00 0 1 2 3 4 4.5 Years since Randomization 4498 4518 4047 4062 3727 3756 3464 3492 B First Episode of Recurrent Atrial Fibrillation in the Substudy, as Assessed with the Use of Transtelephonic Monitoring 1.0 Conditional Probability of Recurrence No. at Risk 0.8 0.6 0.4 0.2 Hazard ratio, 1.14 (95% CI, 0.80 1.64) P=0.46 0.0 0 90 180 270 365 455 545 630 730 Days since Follow-up 99 86 58 50 46 42 45 38 44 37 Among patients with atrial fibrillation, most of whom had well-controlled hypertension and 60% of whom were receiving an ACE inhibitor, the addition of irbesartan did not reduce the risk of death from cardiovascular causes, stroke, or myocardial infarction or this composite outcome plus hospitalization for heart failure. Among patients in sinus rhythm at randomization, there was also no effect on recurrence of atrial fibrillation. However, there was a significant reduction in hospitalizations for heart failure (a secondary outcome), total hospitalizations, and the number of days of hospitalization for cardiovascular reasons. The results of the latter two outcomes were not prespecified and therefore should be cautiously interpreted. We hypothesized that the key mechanisms of benefit would be due to renin angiotensin blockade, as well as lowering of blood pressure. The lack of a significant reduction in the coprimary outcomes weakens these hypotheses but may be attributable to several other factors. Although irbesartan reduced systolic blood pressure by 6.8 mm Hg, more patients in the placebo group received multiple antihypertensive drugs, and their systolic blood pressure was also reduced by 3.9 mm Hg, so that the difference between the groups was modest (2.9 mm Hg). In fact, the observed 9% reduction in stroke among patients who received irbesartan is consistent with the modest reduction in blood pressure observed. An observational analysis from the Ongoing Telmisartan Alone and in Combination with Ramipril Global End-point Trial (ONTARGET; ClinicalTrials.gov number, NCT00153101) indicates a particularly strong relationship between systolic blood pressure and stroke at levels above 115 mm Hg, 15 and two recent trials 16,17 showed that lowering of blood pressure in high-risk populations with initial blood pressure similar to that in ACTIVE I reduced stroke. ACE inhibitors may reduce cardiovascular events despite modest reductions in blood pres- 41 34 2598 2680 39 33 2068 2088 Figure 2. Kaplan Meier Curves for Hospitalization Due to Atrial Fibrillation and for Recurrent Atrial Fibrillation. 37 28 935
T h e n e w e ngl a nd j o u r na l o f m e dic i n e Table 3. Effect of on Hospital Admissions. Reason for Hospitalization (N = 4518) (N = 4498) Difference P Value no. of admissions Cardiovascular disease 3817 4060 243 0.003 Atrial fibrillation 941 976 35 0.37 Condition other than atrial fibrillation 2876 3084 208 0.004 Myocardial infarction 151 139 12 0.51 Stroke 383 426 43 0.12 Non central nervous system embolism 51 60 9 0.38 Heart failure 878 1067 189 <0.001 Transient ischemic attack 85 96 11 0.40 Noncardiovascular disease 2702 2762 60 0.33 Overall (cardiovascular plus noncardiovascular events) 6519 6822 303 0.004 sure in patients with previous vascular disease. In a meta-analysis of three of the largest trials of ACE inhibitors in patients with vascular disease, 3,5,18 these agents reduced major vascular events by 23%, 19 with the same modest lowering of blood pressure that was seen in ACTIVE I, suggesting that mechanisms that are independent of blood-pressure lowering may also play a role. However, in those studies, the majority of patients had previous vascular disease (as compared with only 40% in ACTIVE I), and few were receiving concomitant ARBs. The possibility that cerebrovascular and peripheral thromboembolic events may be reduced by irbesartan is suggested by parallel nonsignificant reductions in transient ischemic attacks and systemic embolic events, and the reduction in the composite of these outcomes plus stroke which was nominally significant. A subgroup analysis of the Perindopril Protection against Recurrent Stroke Study (PROGRESS) involving patients with atrial fibrillation who had a nonfatal stroke also suggests that lowering of blood pressure reduces the risk of stroke and transient ischemic attack among patients with atrial fibrillation. 20 The observation of reduced hemorrhagic stroke and hemorrhagic transformation of ischemic stroke among patients who received irbesartan is intriguing, since similar results were also observed among patients who received telmisartan in the Prevention Regimen for Effectively Avoiding Second Strokes (PRoFESS, NCT00153062) study, in which patients also received an antiplatelet agent. 21 Moreover, a subgroup analysis of the PROGRESS data indicates that lowering of blood pressure reduces the risk of intracerebral hemorrhage more rapidly and to a greater extent than it reduces the risk of ischemic stroke. 22 This finding suggests that aggressive lowering of blood pressure may prevent intracerebral bleeding in patients with atrial fibrillation receiving concomitant antithrombotic drugs; this hypothesis warrants prospective evaluation. Small trials and post hoc analyses indicate that blockers of the renin angiotensin aldosterone system may prevent recurrent atrial fibrillation in patients with heart failure or left ventricular hypertrophy. However, we observed no benefits in this regard, which is consistent with the Gruppo Italiano per lo Studio della Sopravvivenza nell Infarto Miocardico-Atrial Fibrillation (GISSI-AF) study of valsartan. 23 Collectively, these two trials conclusively indicate that an ARB has no effect in preventing atrial fibrillation in patients with intermittent atrial fibrillation. In ACTIVE I, we observed a reduction in hospitalizations for heart failure among patients who received irbesartan as compared with those who received placebo, even though 60% of all the patients were receiving an ACE inhibitor. These findings are consistent with the results of a trial of ARBs in patients with a low ejection fraction 24 and with the results of the Candesartan in Heart Failure: Assessment of Reduction in Mortality and Morbidity (CHARM) Preserved trial (NCT00634712), involving patients with a normal ejection fraction and heart failure, 25 but not with the results of the Telmisartan Randomized Assessment Study in ACE Intolerant Subjects with Cardiovascular Disease (TRANSCEND, NCT00153101) 26 936
in Atrial Fibrillation A Stroke, Myocardial Infarction, or Death from Vascular Causes Subgroup Total No. of Patients %/100 patient-yr of follow-up Hazard Ratio (95% CI) P Value for Interaction Overall ACE inhibitor use Yes No Age <65 yr 65 yr Baseline systolic blood pressure <126 mm Hg 126 140 mm Hg >140 mm Hg Previous coronary heart disease Yes No 9016 5444 3570 2840 6176 2278 3266 3467 3262 5754 5.40 5.86 4.71 3.35 6.45 5.06 5.34 5.70 7.23 4.43 5.40 5.92 4.73 3.06 6.64 5.81 4.84 5.76 7.10 4.54 0.91 0.24 0.15 0.49 0.7 0.9 1.1 1.3 Better Better B Hospitalization Due to Heart Failure, Stroke, Myocardial Infarction, or Death from Vascular Causes Subgroup Total No. of Patients Hazard Ratio (95% CI) %/100 patient-yr of follow-up 9016 7.30 7.70 Overall ACE inhibitor use Yes No Age <65 yr 65 yr Baseline systolic blood pressure <126 mm Hg 126 140 mm Hg >140 mm Hg Previous coronary heart disease Yes No 5444 3570 2840 6176 2278 3266 3467 3262 5754 8.17 5.94 4.92 8.46 7.28 7.00 7.50 9.60 6.04 8.81 6.12 4.71 9.25 8.45 6.99 7.90 10.12 6.43 0.7 0.9 1.1 1.3 P Value for Interaction 0.58 0.14 0.36 0.77 Better Better Figure 3. Hazard Ratios for Study Outcomes According to Subgroup. The size of squares is proportional to the number of patients in the subgroup. ACE denotes angiotensin-converting enzyme. or Nateglinide and Valsartan in Impaired Glucose Tolerance Outcomes Research (NAVIGATOR, NCT00097786). The reduction in hospitalizations for heart failure among patients who received irbesartan in ACTIVE I is supported by a reduction in recurrent events and episodes of heart failure that did not require hospitalization. However, ACTIVE I results differ from those of the in Heart Failure with Preserved Ejection Fraction (I-PRESERVE, NCT00095238) study, in which no benefit with respect to heart failure was observed among patients who received irbesartan. 27 ACTIVE I included twice as many patients as did I-PRESERVE, so it had a higher power to detect moderate reductions in hospitalizations for heart failure. An important observation in ACTIVE I was that hospitalizations for heart failure were more common than stroke in 937
in Atrial Fibrillation this population of patients with atrial fibrillation, and both increased the risk of death by a factor of 4. This finding suggests that preventing heart failure is as important as preventing strokes in this population. In summary, in patients with atrial fibrillation, irbesartan was associated with a modest reduction in blood pressure but did not significantly reduce cardiovascular events. It was associated with a reduction in heart failure and hospitalizations for cardiovascular causes. It is not known whether more aggressive lowering of blood pressure would be effective in patients with atrial fibrillation. Supported by Bristol-Myers Squibb and Sanofi-Aventis. Disclosure forms provided by the authors are available with the full text of this article at NEJM.org. appendix The affiliations of the writing group of the ACTIVE Investigators are as follows: Population Health Research Institute, McMaster University, Hamilton, ON, Canada (S.Y., J.S.H., J.P., S.C., S.J.C.); the Royal Brompton and Harefield NHS Foundation Trust, London (M.F.); Department of Neurology, University of Texas Health Science Center, San Antonio (R.G.H.); J.W. Goethe University, Frankfurt, Germany (S.H.H.); Division of Cardiology, Sunnybrook Health Sciences Centre, Toronto (C.D.J.); and Department of Medicine, Brigham and Women s Hospital, Boston (M.A.P.). References 1. Kannel WB, Abbott RD, Savage DD, McNamara PM. Epidemiologic features of chronic atrial fibrillation: the Framingham Study. N Engl J Med 1982;306:1018-22. 2. Kannel WB. Blood pressure as a cardiovascular risk factor: prevention and treatment. JAMA 1996;275:1571-6. 3. The Heart Outcomes Prevention Evaluation Study Investigators. Effects of an angiotensin-converting enzyme inhibitor, ramipril, on cardiovascular events in highrisk patients. N Engl J Med 2000;342:145-53. [Errata, N Engl J Med 2000;342:748, 1376.] 4. The SOLVD Investigators. Effect of enalapril on survival in patients with reduced left ventricular ejection fractions and congestive heart failure. N Engl J Med 1991;325:293-302. 5. Fox KM. Efficacy of perindopril in reduction of cardiovascular events among patients with stable coronary artery disease: randomized, double-blind, placebocontrolled, multicentre trial (the EUROPA study). Lancet 2003;362:782-8. 6. Neal B, MacMahon S, Chapman N, Blood Pressure Lowering Treatment Trialists Collaboration. Effects of ACE inhibitors, calcium antagonists, and other blood-pressure-lowering drugs: results of prospectively designed overviews of randomised trials. Lancet 2000;356:1955-64. 7. Shi Y, Li D, Tardif JC, Nattel S. Enalapril effects on atrial remodeling and atrial fibrillation in experimental congestive heart failure. Cardiovasc Res 2002;54:456-61. 8. Moreno I, Caballero R, Gonzalez T, et al. Effects of irbesartan on cloned potassium channels involved in human cardiac repolarization. J Pharmacol Exp Ther 2003; 304:862-73. 9. Gerdts E, Wachtell K, Omvik P, et al. Left atrial size and risk of major cardiovascular events during antihypertensive treatment: Losartan Intervention For Endpoint Reduction in Hypertension trial. Hypertension 2007;49:311-6. 10. Healey JS, Baranchuk A, Crystal E, et al. Prevention of atrial fibrillation with angiotensin-converting enzyme inhibitors and angiotensin receptor blockers: a metaanalysis. J Am Coll Cardiol 2005;45:1832-9. 11. The ACTIVE Investigators. Effect of clopidogrel added to aspirin in patients with atrial fibrillation. N Engl J Med 2009; 360:2066-78. 12. The ACTIVE Writing Group. Clopidogrel plus aspirin versus oral anticoagulation for atrial fibrillation in the Atrial fibrillation Clopidogrel Trial with for prevention of Vascular Events (ACTIVE W): a randomised controlled trial. Lancet 2006;367:1903-12. 13. Lin D, Wei L, Yang I, Ying Z. Semiparametric regression for the mean and rate functions of recurrent events. J R Stat Soc B 2000;62:711-30. 14. McCullagh P, Nelder JA. Generalized linear models. 2nd ed. London: Chapman and Hall, 1989. 15. Sleight P, Redon J, Verdecchia P, et al. Prognostic value of blood pressure in patients with high vascular risk in the Ongoing Telmisartan Alone and in combination with Ramipril Global Endpoint Trial study. J Hypertens 2009;27:1360-9. 16. The NAVIGATOR Study Group. Effect of valsartan on the incidence of diabetes and cardiovascular events. N Engl J Med 2010;362:1477-90. [Erratum, N Engl J Med 2010;362:1748.] 17. The ACCORD Study Group. Effects of intensive blood-pressure control in type 2 diabetes mellitus. N Engl J Med 2010;362: 1575-85. 18. Braunwald E, Domanski MJ, Fowler SE, et al. Angiotensin-converting enzyme inhibition in stable coronary artery disease. N Engl J Med 2004;351:2058-68. 19. Dagenais GR, Pogue J, Fox K, Simoons ML, Yusuf S. Angiotensin-convertingenzyme inhibitors in stable vascular disease without left ventricular systolic dysfunction or heart failure: a combined analysis of three trials. Lancet 2006;368: 581-8. 20. Arima H, Hart RG, Colman S, et al. Perindopril-based blood pressure-lowering reduces major vascular events in patients with atrial fibrillation and prior stroke or transient ischemic attack. Stroke 2005;36:2164-9. 21. Yusuf S, Diener H-C, Sacco RL, et al. Telmisartan to prevent recurrent stroke and cardiovascular events. N Engl J Med 2008;359:1225-37. 22. Chapman N, Huxley R, Anderson C, et al. Effects of a perindopril-based blood pressure-lowering regimen on the risk of recurrent stroke according to stroke subtype and medical history: the PROGRESS trial. Stroke 2004;35:116-21. 23. The GISSI-AF investigators. Valsartan for prevention of recurrent atrial fibrillation. N Engl J Med 2009;360:1606-17. [Erratum, N Engl J Med 2009;360:2379.] 24. McMurray JJ, Ostergren J, Swedberg K, et al. Effects of candesartan in patients with chronic heart failure and reduced left-ventricular systolic function taking angiotensin-converting-enzyme inhibitors: the CHARM-Added trial. Lancet 2003;362: 767-71. 25. Yusuf S, Pfeffer MA, Swedberg K, et al. Effects of candesartan in patients with chronic heart failure and preserved leftventricular ejection fraction: the CHARM- Preserved Trial. Lancet 2003;362:777-81. 26. The TRANSCEND Investigators. Effects of the angiotensin-receptor blocker telmisartan on cardiovascular events in high-risk patients intolerant to angiotensin-converting enzyme inhibitors: a randomised controlled trial. Lancet 2008;372: 1174-83. 27. Massie BM, Carson PE, McMurray JJ, et al. in patients with heart failure and preserved ejection fraction. N Engl J Med 2008;359:2456-67. Copyright 2011 Massachusetts Medical Society. 938