angiotensin ii-antagonist in paroxysmal atrial fibrillation (antipaf

Angiotensin II-Antagonist in Paroxysmal Atrial Fibrillation(ANTIPAF) Trial

Andreas Goette, MD; Norbert Schon, MD; Paulus Kirchhof, MD; Gunter Breithardt, MD;Thomas Fetsch, MD; Karl Georg Hausler, MD; Helmut U. Klein, MD; Gerhard Steinbeck, MD;

Karl Wegscheider, PhD; Thomas Meinertz, MD

Background—Unlike antiarrhythmic drugs, the safety and beneficial effects of angiotensin II receptor blockade (ARB) inpatients with structural heart disease is well established. The clinical efficacy of ARBs to prevent atrial fibrillation (AF)so far only has been shown in patients with structural heart disease. Here, we report the primary outcome of theAngiotensin II-Antagonist in Paroxysmal Atrial Fibrillation (ANTIPAF) trial, which investigated the effect ofolmesartan medoxomil compared with placebo on AF burden in patients with paroxysmal AF without structuralheart disease.

Methods and Results—The ANTIPAF trial was a prospective, randomized, placebo-controlled, multicenter trial analyzingthe AF burden (percentage of days with documented episodes of paroxysmal AF) during a 12-month follow-up as theprimary study end point. Four hundred thirty patients with documented paroxysmal AF without structural heart diseasewere randomized to placebo or 40 mg olmesartan per day. Concomitant therapy with ARBs, angiotensin-convertingenzyme inhibitors, and antiarrhythmic drugs was prohibited. Patients were followed using daily transtelephonic ECG(tele-ECG) recordings independent of symptoms. The intention-to-treat population of the trial encompassed 425 patients(placebo group, n�211; olmesartan group, n�214). A total of 87 818 tele-ECGs were analyzed in these patients duringfollow-up (placebo group, 44 888 ECGs; olmesartan group, 42 930 ECGs). Thus, a mean of 207 tele-ECGs wererecorded per patient. The primary end point (AF burden) was not different between the 2 groups (P�0.770). Secondaryoutcome parameters, including quality of life, also were not different. In particular, time to first AF recurrence, time topersistent AF, and number of hospitalizations were not different between the 2 groups. The time to prescription ofrecovery medication (amiodarone) was the only parameter showing an intergroup difference, with earlier prescriptionof amiodarone in the placebo group (P�0.022).

Conclusions—One year of ARB therapy per se does not reduce the number of AF episodes in patients with documentedparoxysmal AF without structural heart disease.

Clinical Trial Registration—URL: http://www.clinicaltrials.gov. Unique identifier: NCT00098137.(Circ Arrhythm Electrophysiol. 2012;5:43-51.)

Key Words: angiotensin � arrhythmia � atrial fibrillation � remodeling � olmesartan � telemedicine

Atrial fibrillation (AF) is the most common sustainedarrhythmia and associates with relevant excess morbid-

ity and mortality.1–3 So far, we are unable to prevent many ofthe severe complications associated with AF, despite anti-thrombotic therapy and “aggressive” management of con-comitant heart disease.2–4 Specifically, the perceived benefit

of rhythm control therapy by antiarrhythmic drugs appears tobe offset by proarrhythmic side effects.2

Clinical Perspective on p 51

Pharmacological inhibition of the renin-angiotensin systemimproves survival in patients with structural heart disease and

Received April 23, 2011; accepted October 24, 2011.From the University Hospital Magdeburg (A.G.), Magdeburg, Germany; Department of Cardiology and Intensive Care Medicine, St Vincenz-Hospital

Paderborn (A.G.), Paderborn, Germany; Private Practice (N.S.), Muhldorf, Germany; Department of Cardiology and Angiology, University HospitalMunster (P.K.), Munster, Germany; University of Birmingham Centre for Cardiovascular Sciences (P.K.), Birmingham, UK; Institut fur Klinisch-Kardiovaskulare Forschung (T.F.), Munich, Germany; Klinik und Poliklinik fur Neurologie, Charite-Universitatsmedizin Berlin (K.G.H.), Berlin,Germany; University of Rochester Medical Center, Heart Research Follow-up Program (H.U.K.), Rochester, NY; Department of Internal Med I,University Hospital Munich (G.S.), Munich, Germany; Department of Medical Biometry and Epidemiology, University Medical Center Hamburg-Eppendorf, Hamburg (K.W.), Germany; and University Heart Center Hamburg (T.M.), Hamburg, Germany.

Trial results were presented at the HOT Line Session III at the European Society of Cardiology Congress 2010 in Stockholm, Sweden.The online-only Data Supplement is available with this article at http://circep.ahajournals.org/lookup/suppl/doi:10.1161/CIRCEP.111.

965178/-/DC1.Correspondence to Andreas Goette, MD, St Vincenz-Hospital Paderborn, Department of Cardiology and Intensive Care Medicine, Am Busdorf 2,

D-33098 Paderborn, Germany. E-mail [email protected]© 2011 American Heart Association, Inc.

Circ Arrhythm Electrophysiol is available at http://circep.ahajournals.org DOI: 10.1161/CIRCEP.111.965178

43

by guest on April 9, 2018

http://circep.ahajournals.org/D

ownloaded from

by guest on A

pril 9, 2018http://circep.ahajournals.org/

Dow

nloaded from



ownloaded from

by guest on A

pril 9, 2018http://circep.ahajournals.org/

Dow

nloaded from

http://www.clinicaltrials.gov

http://circep.ahajournals.org/lookup/suppl/doi:10.1161/CIRCEP.111.965178/-/DC1

http://circep.ahajournals.org/lookup/suppl/doi:10.1161/CIRCEP.111.965178/-/DC1

http://circep.ahajournals.org/




clearly is a safe intervention in most patients with AF.5,6

There is good experimental evidence that angiotensin IIreceptor blockade (ARB) therapy can prevent structuralremodeling6–9 and occurrence of AF in patients with struc-tural heart disease.5,10,11 Given the observation that AFinduces atrial fibrosis and contributes to electrophysiologicalchanges (2 main factors that can be attenuated by ARBtherapy early during the course of the arrhythmia), ARBtherapy appears to be a reasonable and safe additive antiar-rhythmic intervention. Furthermore, recent systematic meta-analyses5,10 suggested that ARB or angiotensin-convertingenzyme (ACE) inhibitor therapy may have direct antiarrhyth-mic effects. In contrast to antiarrhythmic drug trials, in whichtime to first AF recurrence has been an accepted primarystudy end point, ARB therapy may take several weeks andmonths to influence the arrhythmogenic atrial substrate.Therefore, the beneficial effects of ARB therapy may becomeapparent after long-term therapy only. However, the thera-peutic benefit of ARB inhibition has not yet been prospec-tively investigated in patients with paroxysmal AF in theabsence of concomitant ACE-inhibitor and antiarrhythmictherapy. The AFNET (German Network of Competence inAtrial Fibrillation), therefore, conducted the investigator-initiated Angiotensin II-Antagonist in Paroxysmal AtrialFibrillation (ANTIPAF) trial, a prospective, randomized,placebo-controlled trial analyzing the burden of AF over a12-month period as the primary study end point in theabsence of concomitant ACE inhibitor therapy and systematicuse of class I and III antiarrhythmic drugs.

MethodsStudy DesignThe ANTIPAF trial examined the hypothesis that blocking theangiotensin II type 1 receptor with olmesartan medoxomil (olmesar-tan) reduces the incidence of episodes of AF in patients withparoxysmal AF during 12 months by �25% compared with standardmedication without ARB therapy in a prospective, randomized,placebo-controlled, double-blind trial. The sponsor of the ANTIPAFtrial is AFNET, which is funded by the German Ministry of Researchand Education. Daiichi Sankyo Deutschland GmbH (Munich, Ger-many) provided an unrestricted grant to support the study.

The rationale and design of the ANTIPAF trial have beenpublished previously.12 In brief, patients were eligible to participateif they had documented paroxysmal AF (documentation of AF in atleast 1 ECG recorded during the 6 months before randomization),age �18 years, and written informed consent. Patients were random-ized to placebo or 40 mg olmesartan per day. No adjustment of studymedication was permitted. Patients with documented paroxysmal AFwere stratified by �-blocker use (Figure 1). In case of suspectedintolerance of the study medication, study medication was termi-nated. Study medication was provided by Daiichi SankyoDeutschland GmbH. Patients were excluded if they had an indicationfor therapy with an ARB or ACE inhibitor, were on such therapywithin the past month, or had received therapy with antiarrhythmicagents (sodium or potassium channel blockers within 4 half-lives;amiodarone within the past 3 months). Other exclusion criteria weredirect-current cardioversion within the past 3 months; symptomaticbradycardia; implanted pacemaker or implantable cardioverter-defibrillator with any antitachycardiac algorithm in use; cardiacsurgery or cardiac catheter ablation within the past 3 months; typicalangina pectoris symptoms at rest or during exercise; known coronaryartery disease with indication for intervention; significant valvulardisease; left ventricular ejection fraction �40%; diastolic bloodpressure �110 mm Hg at rest; symptomatic arterial hypotension;

known renal artery stenosis, serum creatinine level �1.8 mg/dL;relevant hepatic or pulmonary disorders; hyperthyreosis manifestedclinically or in the laboratory; known drug intolerance for ARB;women who were pregnant or breast-feeding; women of childbearingpotential who were not using a scientifically accepted method ofcontraception; participation in a clinical trial within the past 30 days;drug addiction or chronic alcohol abuse; and legal incapacity or othercircumstances that would prevent the patient from understanding theaim, nature, or extent of the clinical study. The Institute for ClinicalCardiovascular Research (Munich, Germany) was the responsiblecontract research organization for project and data management andwas the central core laboratory for the management of all transtele-phonic ECG (tele-ECG) devices and the standardized analysis of alltransmitted tele-ECGs.

A total of 430 patients were included in the study by 43 centersand were randomized to olmesartan (n�215) or placebo (n�215)and received the study medication (safety population). Of these, 425(placebo group, n�211; olmesartan group, n�214) had at least 1evaluable tele-ECG, thus allowing the calculation of the primary endpoint. By definition, these 425 patients comprised the intention-to-treat population presented in this analysis (Figure 1).

The following terms were used in the study to define outcomeparameters. Day with documented paroxysmal AF: the calendar daywith at least 1 readable ECG recording of �30 s duration showingAF. Documented AF: an episode of AF in any ECG recording lasting�30 s. Paroxysmal AF: ECG recordings within 7 consecutive daysafter initial detection of AF showing both documented AF and sinusrhythm. Suspected persistent AF: all ECG recordings within 7consecutive days after initial detection of AF showing documentedAF and initiating an extraordinary visit to perform a Holter monitorrecording. Persistent AF: a continuous AF in a Holter monitorrecording with a minimum of 18 hours readable after suspectedpersistent AF had been identified.

Primary End PointThe primary end point of the study was the percentage of days withdocumented episodes of paroxysmal or with suspected persistent or

Figure 1. Trial profile. Patients were stratified by �-blocker usebefore randomization. ITT indicates intention to treat; PAF, paroxys-mal atrial fibrillation; SR, sinus rhythm; tele-ECG, transtelephonic ECG.

44 Circ Arrhythm Electrophysiol February 2012



ownloaded from


permanent AF. The AF burden was calculated by the number of dayswith paroxysmal AF or with preceding documentation of suspectedpersistent AF (up to a maximum of 365 days) divided by the numberof measurement days, that is, days in follow-up with at least 1readable tele-ECG recording (up to a maximum 365 days).

Secondary Outcome ParametersSecondary end points of the study were as follows: (1) time to firstoccurrence of a documented relapse of AF, (2) quality of life, (3)time to persistent AF, (4) time to prescription of the recoverymedication, (5) percentage of days with documented episodes ofparoxysmal or with suspected persistent or permanent AF after 90days of therapy (number of days with paroxysmal AF or withpreceding documentation of suspected persistent or permanent AFduring follow-up beginning after a treatment wash-in phase of 90days), (6) number of hospitalizations for cardiovascular reasonsaccording to end point review, (7) number of intermediate medicalvisits without hospitalization for cardiovascular reasons according toend point review, (8) number of cerebrovascular events, and (9) timeto first occurrence of a symptomatic documented episode of AF.

Permitted Concomitant MedicationAntihypertensive therapy included diuretics, calcium channel block-ing agents, and antiadrenergic substances. The target arterial bloodpressure was �140/90 mm Hg for both treatment groups. Oralanticoagulation followed clinical necessity according to presentrecommendations.13 A temporary change to intravenous or subcuta-neous anticoagulation was permitted. �-blockers were used if ther-apy was started before study inclusion. The protocol defined amiod-arone as recovery medication in case of severe symptoms or heartfailure despite sufficient rate control with verapamil or digitalis.Furthermore, in the case of persistent rapid AV conduction despitethe use of high doses of AV nodal blocking drugs, amiodarone alsocould be used. Other antiarrhythmic drugs (ion channel blockingdrugs) were not permitted.

Follow-UpPatients were asked to record and transmit via telephone at least 11-minute ECG per day independent of symptoms. Each patientreceived his or her personal tele-ECG device for the entire follow-upperiod. A central core laboratory was used for management of alltele-ECG devices and for the standardized analysis of all transmittedtele-ECGs. Further parameters during follow-up were from a ques-tionnaire of subjective conditions; a quality-of-life questionnaire(SF-12 [Medical Outcomes Study Short Form-12]); serious adverseevents; physical examination; noninvasive systemic arterial bloodpressure as the mean of 2 measurements; 12-lead ECG at rest;laboratory measurements of sodium, potassium, creatinine, creatinekinase, transaminase, serum urea nitrogen, thyroid stimulating hor-mone, and international normalized ratio; 24-hour ECG: mean heartrate and rhythm; transthoracic echocardiography; actual medication;and treatment allocation. Questionnaires and follow-up visits werescheduled after 3, 6, 9, and 12 months.

Sample SizeThe preliminary sample size determination was proposed based onlimited knowledge about the distribution of the primary outcomemeasure. The distribution was expected to be skewed or even Jshaped. Therefore, a nonparametric comparison of the primaryoutcome variable was chosen for the analysis, using a 10% increasecompared with the sample size derived from the 2-sided t test. Theresults from the SOPAT (Suppression of Atrial Tachyarrhythmias)trial suggested that there is an effect size of one third; that is, themean difference between the treatments was one third of the SD. Atotal of 382 patients was required to detect this difference in a2-sided t test with a significance level of ��0.05 and a power of90% (Program N; IDV; Gauting, Germany). The 10% increase forthe nonparametric approach would then result in a total of 422evaluable patients (211 per group).

Statistical AnalysisBasic descriptive statistics are presented for the intention-to-treat pop-ulation and according to treatment assignment. Nominal variables aregiven as frequencies and percentages. Continuous variables are given asan arithmetic mean and SD. Continuous variables were compared usingStudent t test, and categorical variables were compared using �2 tests.Analysis of the primary end point variable was performed by the 2-sidedWilcoxon test, where patients were stratified according to �-blockertreatment (identical to the van Elteren test). For visualization of thedistributions of the primary outcome in the 2 groups, a mirror histogramwas constructed (Figure 2). Because these distributions turned out to bebimodal, we performed as sensitivity analysis an alternative test ap-proach that goes without any distributional assumptions, a permutationtest of the means with 10 000 randomly selected stratawise permuta-tions, and 95% bootstrap CIs based on 10 000 repetitions. For time-to-event secondary outcome variables (parameters 1, 3, 4, and 9), Kaplan-Meier analyses, including log-rank tests, were performed, stratified by�-blocker treatment. Additionally, stratified Cox proportional hazardmodels were fitted to the data to calculate hazard ratios and cumulativeevent rates for quantification and visualization of treatment effects. Forthe other secondary outcome variables (parameters 5, 6, 7, and 8),2-sided Wilcoxon tests were performed, where patients were stratifiedaccording to �-blocker treatment analog to the prespecified analysis ofthe primary outcome. Quality-of-life data (SF-12 questionnaire, param-eter 2) were analyzed using ANCOVA with baseline adjustment. Theprobability of type I error was fixed to 5% 2 sided and not adjusted formultiplicity of secondary end points. Instead, the secondary end pointswere hierarchically ordered as previously given to allow for an exten-sion of a claim beyond the primary hypothesis (closed testingprocedure).14

Three post hoc analyses were performed for further exploration oftreatment effects. First, it was systematically tested whether therewere differential treatment effects between the �-blocker strata.Second, for time-to-event end points, we checked the proportionalhazards assumption by introducing a treatment�log(time) interac-tion as a time-dependent covariate to the Cox models and tested bylikelihood ratio tests whether the models were significantly improvedby assuming different time trends for the treatments. Third, becausethe use of recovery medication may have diluted other treatmenteffects, a supportive analysis of the primary end point only consid-ering patients without recovery medication was performed. Stataversion 11.0, SAS for Windows version 9.2, and R version 2.8software were used for the statistical analyses.

ResultsBaseline characteristics of the intention-to-treat population aregiven in Table 1. At the end of follow-up, systolic bloodpressures (placebo, 131.3�16.3 mm Hg; olmesartan,

Figure 2. Distribution of the primary study end point (AF burden)by study group. For visualization, a mirror histogram of the dis-tribution of the primary end point in the 2 comparison groups isshown (P�0.770). AF indicates atrial fibrillation.

Goette et al The ANTIPAF Trial 45



ownloaded from


131.4�19.1 mm Hg; P�0.863) and diastolic blood pressures(placebo, 80.4�8.9 mm Hg; olmesartan, 78.5�10.2 mm Hg;P�0.096 adjusted for baseline values) remained comparable inthe 2 groups. Medical therapy was balanced and did not changesignificantly in either group up to the end of follow-up.

Tele-ECG RecordingsA total of 97 159 tele-ECGs were recorded in the 425 patientsduring follow-up. A total of 87 818 tele-ECGs (placebogroup, n�44 888; olmesartan group, n�42,930) could be

used for analyses. The number of tele-ECGs not fulfillingquality criteria in the blinded analysis was 9,341 (9.6%) andcould not be used for further analyses. Thus, a mean of 207tele-ECGs were recorded per patient, with an average of 1.12tele-ECGs per patient and measurement day.

Primary End PointFigure 2 shows the distributions of the primary end point of AFburden by treatment. Both distributions are very similar andreveal an essential bimodality with a marked mode at small AF

Table 1. Baseline Characteristics

Baseline Characteristics Placebo (n�211) Olmesartan (n�214) P

Age, y 61.9�10.2 61.1�11.2 0.470

Women 95 (45.0) 81 (37.9) 0.133

Arterial hypertension 103 (48.8) 106 (49.5) 0.882

Diabetes mellitus 18 (8.5) 17 (7.9) 0.826

History of smoking 91 (43.1) 91 (42.5) 0.900

Coronary artery disease 17/201 (8.5) 11/205 (5.4) 0.218

NYHA class 0.596

0 196/211 (92.9) 197/212 (92.9)

Class I 8/211 (3.8) 6/212 (2.8)

Class II 6/211 (2.8) 9/212 (4.2)

Class III 1/211 (0.5) 0/212 (0.0)

Class IV 0/211 (0.0) 0/212 (0.0)

Left ventricular ejection fraction, % 63.3�7.7 63.2�7.8 0.866

Left atrial diameter �40 mm 68/196 (34.7) 75/205 (36.6) 0.693

Creatinine, mg/dL 0.88�0.21 0.92�0.20 0.049

Serum urea nitrogen, mg/dL 31.0�17.4 32.8�16.0 0.259

Medication

�-blocker (stratum) 149/211 (70.6) 152/214 (71.0) 0.926

Dihydropyridines 23/162 (14.2) 23/170 (13.5) 0.860

Diltiazem 1/162 (0.6) 1/170 (0.6) 0.973

Class IV antiarrhythmic agent 6/162 (3.7) 8/170 (4.7) 0.649

Verapamil 5/162 (3.1) 7/170 (4.1) 0.614

Ca antagonist 29/162 (17.9) 30/170 (17.7) 0.952

Diuretics 24/162 (14.8) 22/170 (12.9) 0.621

Nitrates 3/162 (1.9) 2/170 (1.2) 0.613

Statins 25/162 (15.4) 16/170 (9.4) 0.095

ASS 48/162 (29.6) 53/170 (30.6) 0.849

OAC 46/162 (28.4) 36/170 (30.6) 0.127

Heart rate, beats/min 64.3�11.8 64.0�12.6 0.749

Systolic RR, mm Hg 132.5�18.7 130.0�18.3 0.210

Diastolic RR, mm Hg 79.9�10.2 78.2�8.6 0.098

CHADS2 score* 0.5 (0–4); 0.7�0.9 0 (0–4); 0.7�0.9 0.9016

CHA2DS2-VASc score* 1 (0–7); 1.6�1.4 1 (0–6); 1.5�1.3 0.5605

Data are presented as mean�SD or n (%) �or n/total no. (%)�, unless otherwise indicated. The t test was used forcontinuous variables, and Pearson �2 test with continuity correction was used for categorical variables.

NYHA indicates New York Heart Association; ASS, aspirin; OAC, oral anticoagulants; CHADS2, congestive heartfailure, hypertension, age �75 years, diabetes mellitus, prior stroke or transient ischemic attack; CHA2DS2-VASc,congestive heart failure (or left ventricular systolic dysfunction), hypertension, age �75 years, diabetes mellitus, priorstroke or transient ischemic attack or thromboembolism, vascular disease, age 65–74 years, sex category; systolicRR, systolic blood pressure; diastolic RR, diastolic blood pressure.

*For CHADS2 and CHA2DS2-VASc scores, median (range) and mean�SD are given. Groups were compared usingWilcoxon-Mann-Whitney test.




ownloaded from


burdens up to 10%, but �0, and a smaller mode at AF burdens�90%, resulting in an unexpectedly high SD (pooled SD,0.262). The means were 14.7% in the placebo group and 15.1%in the olmesartan group and were not significantly different inthe prespecified analysis (P�0.770) or in the sensitivity analysis(P�0.865) (Table 2). Because of the bimodality of distributions,bootstrap confidence limits may be more reliable than t-test-based confidence limits. The 95% CIs of the difference betweenolmesartan and placebo based on bootstrapping were large,allowing for absolute differences of �4.5% and 5.4% or, withrespect to a mean AF burden of 14.9%, for relative differencesof �31% or 37%.

Secondary Outcome ParametersThe secondary outcome parameters are listed in Table 2. Withone exception, secondary outcomes did not differ betweentreatment groups. Time to prescription of recovery medica-tion (amiodarone) tended to be shorter in the placebo group(P�0.022). Recovery medication was initiated in 9 (4.2%)patients in the olmesartan group and 20 (9.5%) patients in theplacebo group. In any case, because the primary end pointwas not significant, no claim for an advantage of olmesartan

in secondary outcomes can be derived from the data becauseof the prespecified hierarchal order of hypotheses.

Post Hoc AnalysesIn the first post hoc analysis, only the time to prescription ofrecovery medication revealed significant differences in treat-ment effects between �-blocker strata. The tendency to fewerprescriptions of amiodarone was only present if patients wereon �-blocker therapy (no �-blocker hazard ratio, 1.09 [95%CI, 0.15–7.73]; �-blocker hazard ratio, 0.34 [95% CI, 0.14–0.83]; P�0.041). In the second post hoc analysis,treatment�log(time) interactions were found for time to firstoccurrence of a documented relapse of AF (P�0.004) (Figure3), time to prescription of the recovery medication (P�0.00),and time to first occurrence of a symptomatic documentedepisode of AF (P�0.004). In each of these end points,survival curves were crossing, with a significant trend towardbetter values at the end of the observation interval in theolmesartan group (Figure 3).

It could be argued that the difference in recovery medicationfavored the placebo group in the primary analysis of AF burden.Thus, we performed a third post hoc sensitivity analysis, exclud-

Table 2. Primary End Point and Secondary Outcome Parameters

End Point DescriptionPlacebo

�Mean (95% CI)�Olmesartan

�Mean (95% CI)�Olmesartan vs Placebo

Difference, % P

Primary end point Percentage of days of AF

Prespecified analysis* 14.7 (10.9 to 18.4) 15.1 (11.4 to 18.8) 0.4 (�4.6 to 5.5) 0.770†

Sensitivity analysis* 14.7 (11.0 to 18.3)‡ 15.1 (11.7 to 18.5)‡ 0.4 (�4.5 to 5.4)‡ 0.865§

Secondary end points Percentage of days of AF after atreatment wash-in phase of 90 d*

11.3 (8.0 to 14.5) 12.0 (8.6 to 15.4) 0.7 (�3.8 to 5.2) 0.563†

No. hospitalizations for cardiovascularreasons according to end point review†

0.17 (0.06 to 0.28) 0.19 (0.09 to 0.30) 0.02 (�0.12 to 0.17) 0.958†

No. intermediate medical visits withouthospitalization for cardiovascularreasons according to end point review*

1.87 (1.31 to 2.43) 1.84 (1.28 to 2.40) �0.03 (�0.79 to 0.73) 0.127†

No. cerebrovascular events* 0.007 (�0.004 to 0.019) 0.003 (�0.009 to 0.014) �0.005 (�0.021 to 0.011) 0.265†

Quality of life: SF-12 physical sumscores*

45.6 (44.0 to 47.1) 45.1 (43.6 to 46.7) �0.5 (�2.6 to 1.7) 0.678

Quality of life: SF-12 mental sumscores*

50.8 (49.1 to 52.6) 52.5 (50.7 to 54.2) 1.6 (�0.7 to 4.0) 0.173

Cumulative Event Rate at12-mo Placebo, %

Cumulative Event Rate at12-mo Olmesartan, % Hazard Ratio (95% CI) P

Time to first occurrence of adocumented relapse of AF�

78.5 76.8 1.031 (0.828 to 1.283) 0.786¶

Time to persistent AF� 9.0 9.2 1.080 (0.571 to 2.042) 0.812¶

Time to prescription of the recoverymedication�

8.8 3.4 0.410 (0.186 to 0.904) 0.022¶

Time to first occurrence of asymptomatic documented episodeof AF�

70.9 63.9 0.868 (0.685 to 1.100) 0.240¶

AF indicates atrial fibrillation; SF-12, Medical Outcomes Study Short Form-12.*Strata adjusted mean values are shown.†Van Elteren test.‡Calculations were performed per bootstrapping, 10 000 repetitions.§Permutation test with 10 000 realizations.�Adjusted estimates from stratified Cox model.¶Stratified log-rank test.




ownloaded from


ing from the primary analysis the 29 patients who received arecovery medication during follow-up. We found that patientswith recovery medication on average had more AF burden(which may have been the reason for considering recoverymedication), and correspondingly, in a repetition of the primaryanalysis in the remaining 396 patients, arithmetic means werelower (placebo group, 13.0%; olmesartan group, 14.7%; differ-ence, 0.017). However, the difference between groups remainednonsignificant (P�0.246), indicating that the differential use ofrecovery medication did not mask an olmesartan effect withrespect to AF burden. In summary, use of recovery medicationwas rare and did not explain the lack of an effect of olmesartanon the primary outcome parameter.

Serious Adverse EventsSixty-five serious adverse events were reported by investigatorsand assessed by the blinded critical event committee. Forty-oneof the events were classified as nonserious, whereas 24, whichoccurred in 21 patients, were classified as serious adverse eventsaccording to ICH (International Conference on Harmonisationof Technical Requirements for Registration of Pharmaceuticalsfor Human Use) good clinical practice guidelines (Table 3). Nosignificant differences were found between groups. There were3 deaths in the trial. One 62-year-old patient in the placebo groupdied of carcinoma of the pancreas. Two deaths occurred in theolmesartan group. One 69-year-old patient died of an assumedacute myocardial infarction in the course of heavy physicalexercise after he had experienced significant dyspnea and fatiguefor some days; his left ventricular ejection fraction at baselinewas 56% without history of coronary artery disease. The otherpatient was a 74-year-old woman who died suddenly withoutpreceding clinical symptoms. She had diabetes mellitus type IIand a moderately reduced left ventricular ejection fraction atbaseline without a history of coronary artery disease. Otherserious adverse events in the placebo group were 1 acutepancreatitis, 2 supraventricular tachycardias, and 1 acute myo-cardial infarction. Other serious adverse events in the olmesartan

group were 1 episode of a nonsustained wide-QRS complextachycardia (12 beats), 1 supraventricular tachycardia, 1 septicshock with acute respiratory distress syndrome, and 1 carcinomaof the pancreas. In all adverse events adjudicated as serious, thecommittee saw no relationship to the study medication. Insummary, the safety profile of the olmesartan group showed nodifference compared with the placebo group.

DiscussionThe ANTIPAF trial showed that the use of 40 mg olmesartanin patients with paroxysmal AF is safe but does not reduce theAF burden compared with placebo during 1-year follow-up.Furthermore, 8 of 9 defined secondary outcome parameters,such as quality of life, time to first symptomatic and asymp-tomatic recurrence of AF, time to persistent AF, and numberof hospitalizations for cardiovascular reasons, were not dif-ferent between groups. Although in the presence of �-blockerprescriptions, time to prescription of the recovery medication(amiodarone) tended to be shorter in the placebo group, theprimary end point (AF burden) was still comparable in asubanalysis restricted to patients not receiving the recoverymedication. In any case, because the primary end point wasnot significant, no claim for an advantage of olmesartan insecondary outcomes can be derived from the data due to theprespecified hierarchal order of hypotheses.

Most patients with AF have concomitant cardiovasculardiseases like hypertension, heart failure, or valvular heartdisease. Such diseases usually generate a clear indication forACE inhibitor or ARB therapy, which have been shown to beeffective and safe. In addition to ventricular changes, thesecardiovascular diseases have been found to affect substan-tially the structure of atrial tissue and, thereby, the occurrenceof AF.1–3,5,10 At the molecular level, angiotensin II, oxidativestress, and proinflammatory mediators are of particular im-portance to induce proarrhythmic structural remodeling andatrial ectopy in the area of the pulmonary veins.7–9,11,15,16

Therefore, the use of ACE inhibitors or ARBs appears as anattractive approach to treat patients with AF. Experimentaldata and several clinical trials support this concept; however,previous studies suggest that the concept to prevent structuralatrial remodeling is more efficient for primary prevention of

Figure 3. Cumulative incidence rates of atrial fibrillation recur-rence by study group showed no difference between treatmentgroups (P for hazard ratio�0.786). In the second post hoc anal-ysis, treatment�log(time) interactions were found for time to firstoccurrence of a documented relapse of atrial fibrillation (P fortrend differences�0.004).

Table 3. Serious Adverse Events by Study Group in theSafety Set

Type of Adjudicated SAEPlacebo(n�215)

Olmesartan(n�215) P

Acute heart failure 1 (0.47) 1 (0.47) 1.000

Bleeding 1 (0.47)* 1 (0.47) 1.000

Angina pectoris 0 (0) 1 (0.47) 0.317

Death 1 (0.47) 2 (0.93) 0.562

Cerebrovascular complications 2 (0.93) 1 (0.47) 0.562

Syncope 2 (0.93) 0 (0) 0.156

Other 4 (1.86) 4 (1.86)† 1.000

Any SAE 11 (5.12) 10 (4.65) 0.823

Data are presented as n (%).SAE indicates serious adverse event.*One patient had 3 events.†One patient had 2 events.




ownloaded from


AF, especially in patients with concomitant cardiovasculardiseases.5,10,17–20

For secondary prevention, the results are conflicting.5,10,21–26

First, prospective studies examined the effect of ARBs and ACEinhibitors in association with amiodarone to prevent AF.21,22,26

Of note, Madrid et al27 found a dose-dependent effect ofirbesartan on AF recurrence in 120 patients after cardioversion.However, these trials were relatively small, and systematic ECGmonitoring to detect silent AF was not used.21–23 In contrast, theGISSI-AF [Giuppo Haliano per Lo Studio della Soprarrivenzanell’Infanto Mioccondiso-Atrial Fibrillation] trial showed noeffect of ARB therapy on the recurrence of AF in �1400patients with underlying cardiovascular diseases, diabetes mel-litus, or left atrial enlargement. However, the study did notexclude patients on ACE inhibitor therapy.24 Approximately60% of all patients received concomitant ACE inhibitor therapy.Negative results were also found in a recent retrospectivesubanalysis of ALLHAT (Antihypertensive and Lipid-LoweringTreatment to Prevent Heart Attack Trial), which examinedbaseline prevalence and in-trial incidence of new-onset AF oratrial flutter and their influence on clinical outcomes in 42 418men and women with hypertension aged �55 years with at least1 additional cardiovascular risk factor. AF/atrial flutter occurredin 641 (2.0%) participants and, excluding doxazosin, did notdiffer by antihypertensive treatment group (lisinopril, amlodip-ine, and thiazide-like diuretics).28 Of note, in the high-riskhypertensive population, preexisting and new-onset AF/atrialflutter were associated with increased mortality. Similar to theANTIPAF results, ALLHAT raised substantial questions aboutwhether previous studies suggesting lower new-onset AF/atrialflutter rates with ACE inhibitors or ARBs are generalizable. Inaccordance, the ACTIVE (Atrial Fibrillation Clopidogrel TrialWith Irbesartan for Prevention of Vascular Events) program andONTARGET (ONgoing Telmisartan Alone and in combinationwith Ramipril Global Endpoint Trial) also could not showpositive effects of ARBs on the occurrence of AF in patientswith cardiovascular risk factors.25,29

Nevertheless, meta-analyses still favor the use of ACEinhibitors and ARBs to reduce the occurrence of AF.5,10 Arecent meta-analysis included 23 randomized controlled trialsof 87 048 patients.10 In primary prevention, 6 trials inhypertension, 2 trials in myocardial infarction, and 3 trials inheart failure were included (some being post hoc analyses ofrandomized controlled trials). In secondary prevention, 8trials after cardioversion and 4 trials assessing the medicalprevention of recurrence were included. Overall, renin-angio-tensin system inhibition reduced the odds ratio for AF by33%, but there was substantial heterogeneity among trials. Inprimary prevention, renin-angiotensin system inhibition waseffective in patients with heart failure and in those withhypertension and left ventricular hypertrophy but not inpatients postmyocardial infarction overall. In secondary pre-vention, renin-angiotensin system inhibition often was ad-ministered in addition to antiarrhythmic drugs, includingamiodarone, further reducing the odds for AF recurrence aftercardioversion by 45% and in patients on medical therapy by63%. Thus, it appears that ARBs or ACE inhibitors may beeffective in the setting of structural heart disease through their

inherent action on the renin-angiotensin system and, inparticular, if combined with amiodarone.

On the basis of existing data, it remained to be determinedwhether ARBs or ACE inhibitors per se have an antiarrhythmiceffect, which should be tested in patients with AF in the absenceof structural heart disease, where AF often may depend onectopic activity mostly originating from the pulmonary veins. Toour knowledge, the ANTIPAF trial, is the first study to evaluatethe role of ARBs for secondary prevention in patients withparoxysmal AF without significant cardiovascular diseases inthe absence of ACE inhibitors and ion channel blocking antiar-rhythmic drug therapy. A major effort was undertaken todocument AF recurrences during the follow-up period by dailytele-ECG monitoring. Because it was assumed that it may takeseveral weeks to months for ARBs to modulate atrial structureand electrophysiology, the primary outcome parameter was notthe first recurrence of AF but of AF burden, which is clinicallymore important for the patient. Thus, the duration of follow-upwas extended far beyond the first recurrence of AF. In addition,not only the occurrence of symptomatic AF was identified(secondary outcome parameter), but also any type of recurrenceindependent of symptoms was counted in the primary end point.During 1-year follow-up, the ANTIPAF trial failed to show anantiarrhythmic effect of olmesartan. Even after exclusion of thefirst 90 days of therapy, which appears as a sufficient time periodfor ARBs to affect profibrillatory structural and molecular atrialchanges, ARB therapy had no influence on the recurrence of AF.In addition, the time to persistent AF was not altered. So far, itwas concluded from many studies that ARBs might reduce orabolish atrial remodeling processes and, thereby, attenuate theprogression from paroxysmal to persistent AF. However, and inclear contrast to experimental and first clinical findings, thepresent study showed that the progression of AF from paroxys-mal to persistent AF was not altered by ARB therapy. Thiscannot be explained by an inadequate dosage of olmesartanbecause the ANTIPAF trial used the highest permitted dose ofthe drug. Thus, it appears highly unlikely that the overall resultsof the trial are caused by insufficient drug effects or bydifferences of concomitant factors like systemic blood pressure.Of note, the present data set (including �80 000 tele-ECGs)shows very high variability of AF recurrences in patients withparoxysmal AF. Thus, it appears very unlikely that nonsystem-atic and intermittent ECGs or Holter monitor recordings canassess the true burden of AF. This finding should be consideredin future attempts to assess the efficacy of pharmacological andnonpharmacological antiarrhythmic approaches. The only pa-rameter that was found to be different in the 2 study groups wasthe time to prescription of the recovery medication (amiod-arone), which was given earlier and more frequently in theplacebo group, and is consistent with the observation thatpatients with recovery medication, on average, had more AFburden. However, AF burden in the 2 treatment groups was stillcomparable when the 29 patients who received recovery medi-cation were excluded from analysis. The recovery medicationwas allowed in the present study because it was believed to beunethical to compare ARB therapy with placebo in patients withsymptomatic AF over an extended period without a possibilityfor an accepted and effective antiarrhythmic therapy. Therefore,the recovery medication (amiodarone) was included in the




ownloaded from


protocol. The initiation of amiodarone therapy was prespecifiedas a secondary outcome parameter in the trial. As assumed,patients with recovery medication had more AF episodes com-pared with others. Nevertheless, the absolute number of patientsreceiving the recovery medication was low, and the primaryoutcome was unchanged.

In summary, results of recent clinical trials do not supportthe use ARBs or ACE inhibitors in patients with AF withoutconcomitant structural heart disease. These findings haveinfluenced the 2010 European Society of Cardiology AFguidelines,30 which do not recommend upstream therapy withACE inhibitors and ARBs for prevention of AF in patientswithout cardiovascular disease. Thus, the present results ofthe ANTIPAF trial clearly support this recommendation.

LimitationsA limitation of the study is that there was an essentialbimodality in both groups: AF burden was rather low in moststudy patients and very high in a small subgroup (Figure 2).Thus, the power of the study was smaller than anticipated,and true improvements and deteriorations of the average AFburden of up to 30% cannot be excluded. Nevertheless, thestudy shows that the recurrence of AF is highly variable, andtherefore, the true burden of AF is difficult to predict.Continuous monitoring for extended periods (�1 year) withimplanted devices and analyses of very large patient popula-tions might be the only way to determine mild or modestdifferences among patients with paroxysmal AF. Neverthe-less, the use of daily tele-ECGs is in accordance with arecommended monitoring strategy in AF trials.31 It is con-ceivable, however, that long-term therapy with ARBs overseveral years may produce a long-term antiarrhythmic effectin patients with AF without structural heart disease. Thesurvival curves reveal trends that suggest possible differencesbeyond the time interval studied (Figure 3).

ConclusionsARB therapy per se does not reduce the number of AFepisodes in patients with documented paroxysmal AF withoutstructural heart disease during 1-year follow-up. Therefore,ARBs may not be recommended as first-line treatment in thisclinical setting if not indicated for other reasons.

AcknowledgmentsAll participating centers and participating physicians were located inGermany and are listed in the online-only Data Supplement.

Sources of FundingThis work was supported by German Ministry of Research andEducation (BMBF) through the German Network of Competence inAtrial Fibrillation (AFNET; grant 01GI0204; NCT00098137). DaiichiSankyo Deutschland GmbH supported the study with an unrestrictedresearch grant given to the German AFNET, which was administered bythe Central Office of the AFNET under the control of the University ofMunster (Web site: www.kompetenznetz-vorhofflimmern.de; contact,[email protected]).

DisclosuresDr Goette has received speaker fees from Daiichi SankyoDeutschland GmbH.

References1. Kannel WB, Abbott RD, Savage DD, McNamara PM. Epidemiological

features of atrial fibrillation: the Framingham Study. N Engl J Med.1982;306:1018–1022.

2. Corley SD, Epstein AE, DiMarco JP, Domanski MJ, Geller N, GreeneHL, Josephson RA, Kellen JC, Klein RC, Krahn AD, Mickel M, MitchellLB, Nelson JD, Rosenberg Y, Schron E, Shemanski L, Waldo AL, WyseDG. Relationships between sinus rhythm, treatment, and survival in theAtrial Fibrillation Follow-Up Investigation of Rhythm Management(AFFIRM) Study. Circulation. 2004;109:1509–1513.

3. Go AS, Hylek EM, Phillips KA, YC, Henault LE, Selby JV, Singer DE.Prevalence of diagnosed atrial fibrillation in adults. JAMA. 2001;285:2370–2375.

4. Nabauer M, Gerth A, Limbourg T, Schneider S, Oeff M, Kirchhof P, GoetteA, Lewalter T, Ravens U, Meinertz T, Breithardt G, Steinbeck G. TheRegistry of the German Competence NETwork on Atrial Fibrillation: patientcharacteristics and initial management. Europace. 2009;11:423–434.

5. Healey JS, Baranchuk A, Crystal E, Morillo CA, Garfinkle M, Yusuf S,Connolly SJ. Prevention of atrial fibrillation with angiotensin-convertingenzyme inhibitors and angiotensin receptor blockers: a meta-analysis.J Am Coll Cardiol. 2005;45:1832–1839.

6. Li D, Fareh S, Leung TK, Nattel S. Promotion of atrial fibrillation by heartfailure in dogs. Atrial remodeling of a different sort. Circulation. 1999;100:87–95.

7. Goette A, Staack T, Rocken C, Arndt M, Geller C, Huth C, Ansorge S,Klein HU, Lendeckel U. Increased expression of extracellular signal-regulated kinase and angiotensin-converting enzyme in human atriaduring atrial fibrillation. J Am Coll Cardiol. 2000;35:1669–1677.

8. Goette A, Arndt M, Rocken C, Staack T, Geller JC, Huth C, Ansorge S,Klein HU, Lendeckel U. Regulation of angiotensin II receptor subtypesduring atrial fibrillation in humans. Circulation. 2000;101:2678–2681.

9. Kumagai K, Nakashima H, Urata H, Gondo N, Arakawa K, Saku K. Effectsof angiotensin II type 1 receptor antagonist on electrical and structuralremodeling in atrial fibrillation. J Am Coll Cardiol. 2003;41:2197–2204.

10. Schneider MP, Hua TA, Bohm M, Wachtell K, Kjeldsen SE, SchmiederRE. Prevention of atrial fibrillation by renin-angiotensin system inhi-bition: a meta-analysis. J Am Coll Cardiol. 2010;55:2299–2307.

11. Nattel S. New ideas about atrial fibrillation 50 years on. Nature. 2002;415:219–226.

12. Goette A, Breithardt G, Fetsch T, Hanrath P, Klein HK, Lehmacher W,Steinbeck G, Meinertz T. Angiotensin II-Antagonist in Paroxysmal AtrialFibrillation (ANTIPAF)-Trial: rationale and study design. Clin Drug Invest.2007;27:697–705.

13. Fuster V, Ryden LE, Cannom DS, Crijns HJ, Curtis AB, Ellenbogen KA,Halperin JL, Le Heuzey JY, Kay GN, Lowe JE, Olsson SB, PrystowskyEN, Tamargo JL, Wann S, Priori SG, Blanc JJ, Budaj A, Camm AJ, DeanV, Deckers JW, Despres C, Dickstein K, Lekakis J, McGregor K, MetraM, Morais J, Osterspey A, Zamorano JL, Smith SC Jr, Jacobs AK, AdamsCD, Anderson JL, Antman EM, Hunt SA, Nishimura R, Ornato JP, PageRL, Riegel B. ACC/AHA/ESC 2006 guidelines for the management ofpatients with atrial fibrillation-executive summary: a report of theAmerican College of Cardiology/American Heart Association Task Forceon practice guidelines and the European Society of Cardiology Com-mittee for Practice Guidelines (Writing Committee to Revise the 2001Guidelines for the Management of Patients with Atrial Fibrillation)Developed in collaboration with the European Heart Rhythm Associationand the Heart Rhythm Society. Eur Heart J. 2006;27:1979–2030.

14. Maurer W, Hothorn LA, Lehmacher W. Multiple comparisons in drugclinical trials and preclinical assays: a priori ordered hypotheses. In:Vollmar J, ed. Biometrie in der Chemisch-Pharmazeutischen Industrie,Band 6. Stuttgart, Germany: Fischer; 1995:13–18.

15. Goette A, Bukowska A, Lendeckel U, Erxleben M, Hammwohner M,Strugala D, Pfeiffenberger J, Rohl FW, Huth C, Ebert MP, Klein HU,Rocken C. Angiotensin II receptor blockade reduces tachycardia-inducedatrial adhesion molecule expression. Circulation. 2008;117:732–742.

16. Goette A, Bukowska A, Dobrev D, Pfeiffenberger J, Morawietz H,Strugala D, Wiswedel I, Rohl FW, Wolke C, Bergmann S, Bramlage P,Ravens U, Lendeckel U. Acute atrial tachyarrhythmia induces angioten-sin II type 1 receptor-mediated oxidative stress and microvascular flowabnormalities in the ventricles. Eur Heart J. 2009;30:1411–1420.

17. Pedersen OD, Bagger H, KØber L, Torp-Pedersen C. Trandolapril reducesthe incidence of atrial fibrillation after acute myocardial infarction in patientswith left ventricular dysfunction. Circulation. 1999;100:376–380.




ownloaded from

www.kompetenznetz-vorhofflimmern.de


18. Vermes E, Tardif JC, Bourassa MG, Racine N, Levesque S, White M, GuerraPG, Ducharme A. Enalapril decreases the incidence of atrial fibrillation inpatients with left ventricular dysfunction. Circulation. 2003;107:2926–2931.

19. Schmieder RE, Kjeldsen SE, Julius S, McInnes GT, Zanchetti A, Hua TA.Reduced incidence of new-onset atrial fibrillation with angiotensin IIreceptor blockade: the VALUE trial. J Hypertens. 2008;26:403–411.

20. Wachtell K, Lehto M, Gerdts E, Olsen MH, Hornestam B, Dahlof B,Ibsen H, Julius S, Kjeldsen SE, Lindholm LH, Nieminen MS, DevereuxRB. Angiotensin II receptor blockade reduces new-onset atrial fibrillationand subsequent stroke compared to atenolol: the Losartan InterventionFor End Point Reduction in Hypertension (LIFE) study. J Am CollCardiol. 2005;45:712–719.

21. Ueng KC, Tsai TP, Yu WC, Tsai CF, Lin MC, Chan KC, Chen CY, WuDJ, Lin CS, Chen SA. Use of enalapril to facilitate sinus rhythm main-tenance after external cardioversion of long-standing persistent atrialfibrillation. Results of a prospective and controlled study. Eur Heart J.2003;24:2090–2098.

22. Yin Y, Dalal D, Liu Z, Wu J, Liu D, Lan X, Dai Y, Su L, Ling Z, She Q,Luo K, Woo K, Dong J. Prospective randomized study comparing ami-odarone vs. amiodarone plus losartan vs. amiodarone plus perindopril forthe prevention of atrial fibrillation recurrence in patients with lone par-oxysmal atrial fibrillation. Eur Heart J. 2006;27:1841–1846.

23. Belluzzi F, Sernesi L, Preti P, Salinaro F, Fonte ML, Perlini S. Preventionof recurrent lone atrial fibrillation by the angiotensin-II convertingenzyme inhibitor ramipril in normotensive patients. J Am Coll Cardiol.2009;53:24–29.

24. GISSI-AF Investigators; Disertori M, Latini R, Barlera S, Franzosi MG,Staszewsky L, Maggioni AP, Lucci D, Di Pasquale G, Tognoni G.Valsartan for prevention of recurrent atrial fibrillation. N Engl J Med.2009;360:1606–1617.

25. ACTIVE Steering Committee; ACTIVE Investigators; Connolly S, Yusuf S,Budaj A, Camm J, Chrolavicius S, Commerford PJ, Flather M, Fox KA, HartR, Hohnloser S, Joyner C, Pfeffer M, Anand I, Arthur H, Avezum A,Bethala-Sithya M, Blumenthal M, Ceremuzynski L, De Caterina R, Diaz R,Flaker G, Frangin G, Franzosi MG, Gaudin C, Golitsyn S, Goldhaber S,Granger C, Halon D, Hermosillo A, Hunt D, Jansky P, Karatzas N, KeltaiM, Lanas F, Lau CP, Le Heuzey JY, Lewis BS, Morais J, Morillo C, Oto A,Paolasso E, Peters RJ, Pfisterer M, Piegas L, Pipillis T, Proste C, Sitkei E,Swedberg K, Synhorst D, Talajic M, Tregou V, Valentin V, van Mieghem

W, Weintraub W, Varigos J. Rationale and design of ACTIVE: the AtrialFibrillation Clopidogrel Trial With Irbesartan for Prevention of VascularEvents. Am Heart J. 2006;151:1187–1193.

26. Madrid AH, Bueno MG, Rebollo JM, Marin I, Pena G, Bernal E, RodriguezA, Cano L, Cano JM, Cabeza P, Moro C. Use of irbesartan to maintain sinusrhythm in patients with long-lasting persistent atrial fibrillation: a prospectiveand randomized study. Circulation. 2002;106:331–336.

27. Madrid AH, Marin IM, Cervantes CE, Morell EB, Estevez JE, Moreno G,Parajon JR, Peng J, Limon L, Nannini S, Moro C. Prevention of recur-rences in patients with lone atrial fibrillation. The dose-dependent effectof angiotensin II receptor blockers. J Renin Angiotensin Aldosterone Syst.2004;5:114–120.

28. Haywood LJ, Ford CE, Crow RS, Davis BR, Massie BM, Einhorn PT,Williard A. Atrial fibrillation at baseline and during follow-up inALLHAT (Antihypertensive and Lipid-Lowering Treatment to PreventHeart Attack Trial). J Am Coll Cardiol. 2009;54:2023–2031.

29. The ONTARGET Investigators. Telmisartan, ramipril, or both in patientsat high risk for vascular events. N Engl J Med. 2008;358:1547–1559.

30. Camm AJ, Kirchhof P, Lip GY, Schotten U, Savelieva I, Ernst S, VanGelder IC, Al-Attar N, Hindricks G, Prendergast B, Heidbuchel H, Alfieri O,Angelini A, Atar D, Colonna P, De Caterina R, De Sutter J, Goette A,Gorenek B, Heldal M, Hohloser SH, Kolh P, Le Heuzey JY, Ponikowski P,Rutten FH; ESC Committee for Practice Guidelines, Vahanian A, AuricchioA, Bax J, Ceconi C, Dean V, Filippatos G, Funck-Brentano C, Hobbs R,Kearney P, McDonagh T, Popescu BA, Reiner Z, Sechtem U, Sirnes PA,Tendera M, Vardas PE, Widimsky P; Document Reviewers, Vardas PE,Agladze V, Aliot E, Balabanski T, Blomstrom-Lundqvist C, Capucci A,Crijns H, Dahlof B, Folliguet T, Glikson M, Goethals M, Gulba DC, HoSY, Klautz RJ, Kose S, McMurray J, Perrone Filardi P, Raatikainen P,Salvador MJ, Schalij MJ, Shpektor A, Sousa J, Stepinska J, Uuetoa H,Zamorano JL, Zupan I. Guidelines for the management of atrial fibrillation:the Task Force for the Management of Atrial Fibrillation of the EuropeanSociety of Cardiology (ESC). Europace. 2010;12:1360–1420.

31. Kirchhof P, Bax J, Blomstrom-Lundquist C, Calkins H, Camm AJ, CappatoR, Cosio F, Crijns H, Diener HC, Goette A, Israel CW, Kuck KH, Lip GY,Nattel S, Page RL, Ravens U, Schotten U, Steinbeck G, Vardas P, Waldo A,Wegscheider K, Willems S, Breithardt G. Early and comprehensive man-agement of atrial fibrillation: proceedings from the II AFNET/EHRA con-sensus conference on atrial fibrillation entitled ‘research perspectives in atrialfibrillation.’ Eur Heart J. 2009;30:2969c–2977c.

CLINICAL PERSPECTIVESeveral post hoc analyses from clinical trials suggest that upstream therapy with angiotensin II receptor blockade (ARB)is beneficial in patients with atrial fibrillation (AF). So far to our knowledge, a proof-of-principle study showing that ARBshave true antiarrhythmic potential has not been published. The ANTIPAF (Angiotensin II-Antagonist in Paroxysmal AtrialFibrillation) trial is the first prospective, randomized, placebo-controlled, multicenter trial analyzing the AF burden(percentage of days with documented episodes of paroxysmal AF) during a 12-month follow-up as the primary study endpoint. Four hundred twenty-five patients were followed using daily transtelephonic ECG recordings independent ofsymptoms. The extensive number of 87 818 transtelephonic ECGs were analyzed during follow-up. The primary end point(AF burden) as well as secondary end points were not different in the 2 treatment groups. Thus, 1 year of ARB therapyper se did not reduce the number of AF episodes in patients with documented paroxysmal AF without structural heartdisease. In contrast to GISSI-AF and the ACTIVE (Atrial Fibrillation Clopidogrel Trial With Irbesartan for Prevention ofVascular Events) program, no concomitant use of angiotensin-converting enzyme inhibitors and antiarrhythmic drugs wasallowed in ANTIPAF. Therefore, the ANTIPAF study proves that upstream therapy with ARB per se does not reduce therecurrence rate of AF. In the clinical setting, ARB should only be used in patients with AF if these substances are indicatedfor therapy of concomitant diseases like hypertension and heart failure.




ownloaded from


Georg Häusler, Helmut U. Klein, Gerhard Steinbeck, Karl Wegscheider and Thomas MeinertzAndreas Goette, Norbert Schön, Paulus Kirchhof, Günter Breithardt, Thomas Fetsch, Karl

Angiotensin II-Antagonist in Paroxysmal Atrial Fibrillation (ANTIPAF) Trial

Print ISSN: 1941-3149. Online ISSN: 1941-3084 Copyright © 2011 American Heart Association, Inc. All rights reserved.

Avenue, Dallas, TX 75231is published by the American Heart Association, 7272 GreenvilleCirculation: Arrhythmia and Electrophysiology

doi: 10.1161/CIRCEP.111.9651782012;5:43-51; originally published online December 7, 2011;Circ Arrhythm Electrophysiol.

http://circep.ahajournals.org/content/5/1/43World Wide Web at:

The online version of this article, along with updated information and services, is located on the

http://circep.ahajournals.org/content/suppl/2011/12/07/CIRCEP.111.965178.DC1Data Supplement (unedited) at:

http://circep.ahajournals.org//subscriptions/

is online at: Circulation: Arrhythmia and Electrophysiology Information about subscribing to Subscriptions:

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

document. Answer

Permissions and Rights Question andunder Services. Further information about this process is available in thepermission is being requested is located, click Request Permissions in the middle column of the Web pageClearance Center, not the Editorial Office. Once the online version of the published article for which

can be obtained via RightsLink, a service of the CopyrightCirculation: Arrhythmia and Electrophysiologyin Requests for permissions to reproduce figures, tables, or portions of articles originally publishedPermissions:



ownloaded from

http://circep.ahajournals.org/content/5/1/43

http://circep.ahajournals.org/content/suppl/2011/12/07/CIRCEP.111.965178.DC1

http://www.ahajournals.org/site/rights/

http://www.ahajournals.org/site/rights/

http://www.lww.com/reprints

http://circep.ahajournals.org//subscriptions/


Supplemental Material

Angiotensin II-Antagonist in

Paroxysmal Atrial Fibrillation (ANTIPAF)-Trial

Steering Committee (SC):

A. Goette and T. Meinertz (PI), G. Breithardt, T. Fetsch, H.U. Klein, G. Steinbeck

Data and Safety Monitoring Committee:

L. Seipel, Tübingen (Chairman); D. Andresen, Berlin; M. Gottwik, Nürnberg

Biometrical advisor: K. Wegscheider

Critical Event Committee (CEC):

M. Oeff, U. Tebbe, G. Einhäupl, K.G. Häusler

Participating and recruiting centres (number of included patients) Praxis Dr. Schön, Mühldorf (95) N Schön; Universitätsklinikum Magdeburg, Magdeburg (74) A Goette, A D’Allessandro, M Hammwöhner; Universitätsklinikum Münster, Münster (33) G Breithardt, P Kirchhof, S Müller, S Orwat, A Samol; Euregio Klinik Nordhorn, Nordhorn (26) M-L Leutermann-Oeinck; Uniklinik Bonn, Bonn (18) T Lewalter, S Remerie; Praxis Dr. Boscher, Biberach an der Riß (18) D Boscher; Praxis Dr. Sarnighausen, Lüneburg (12) H-E Sarnighausen; Praxis Dr. Taggeselle, Markkleeberg (11) J Taggeselle; Universitätsklinikum Hamburg-Eppendorf, Hamburg (10) T Meinertz, I Drewitz; Praxis Dr. AL-Zoebi, Wermsdorf (10) A Al-Zoebi; Uniklinik Mannheim, Mannheim (9) C Echternach, M Werle; Klinik Bassum, Bassum (9) B Hayen; Klinik Urban Berlin, Berlin (8) A Mielke; Uniklinik Aachen, Aachen (7) T Schimpf, E Saygili; Reha Zentrum Spreewald, Burg (7) W Kamke; Praxis Dr. Jahnke, Ulm (7) N Jahnke; Praxis Dr. Haßler, Plauen (7) N Haßler; Praxis Dr. Gabelmann, Kirchzarten (7) M Gabelmann; Klinikum Kassel, Kassel (6) J Neuzner; Städt. Klinik Brandenburg, Brandenburg (5) C Sprenger; Uniklinik Tübingen, Tübingen (5) N Rüb, A Henning; Robert Bosch Krankenhaus, Stuttgart (5) U Parade, C Holzer; Praxis Dr. Klingenheben, Bonn (4) T Klingenheben; Praxis Dr. Weimer, Reinfeld (3) J Weimer; Praxis Dr. Jeserich, Nürnberg (3) M Jeserich; Praxis Dr. Berg, Ühlingen (3) R Berg; Krankenhaus Hagenow, Hagenow (3) D Buchholz; Ahrtalklinik Bad Bodendorf, Bad Bodendorf (3) G Giesen; Herzzentrum Leipzig, Leipzig (2) G Hindricks; Praxis Dr. Schumacher, Wolmirstedt (2) M Schumacher; Praxis Dres. Kruck/ Bosch, Ludwigsburg (2) R Bosch; Praxis Dr. Killat, Haßloch (2) H Killat; Krankenhaus Wriezen, Wriezen (2) L Gruev; Katharinen Hospital Un-na, Unna (2) S Atri; Klinik Pirna, Pirna (1) B Axthelm; Klinik Garmisch-Partenkirchen, Garmisch-Partenkirchen (1) M Adelt; Evangelisches Krankenhaus Düsseldorf, Düsseldorf (1) E Vester; Clemenshospital Münster, Münster (1) B Knipper;Herzzentrum Lahr, Lahr (1) E von Hodenburg; Hellmig Krankenhaus Kamen, Kamen (1) HJ Dieckmann; Praxis Dr. Baar, Andernach (1) W Baar; Praxis Dr. Neubert, Rathenow (1) J Neubert; Praxis Dr. Schaefer, Aachen (1) P Schaefer; Praxis Dr. Ulrich, Kamen (1) C Ulrich.

angiotensin ii-antagonist in paroxysmal atrial fibrillation (antipaf

Documents