Author + information
- Received July 2, 1999
- Revision received March 1, 2000
- Accepted April 13, 2000
- Published online September 1, 2000.
- Edward L.C Pritchett, MDa,* (, )
- Richard L Page, MD, FACC∗,
- Stuart J Connolly, MD, FACC†,
- Stephen R Marcello, MD, FACC‡,
- Daniel J Schnell, PhD‡,
- William E Wilkinson, PhD§,
- the Azimilide Supraventricular Arrhythmia Program 3 (SVA-3) Investigators
- ↵*Reprint requests and correspondence: Dr. Edward L.C. Pritchett, Room 04280, Duke South Hospital, DUMC-3477, Durham, North Carolina 27710
The purpose of this study was to assess the effectiveness of azimilide, a class III antiarrhythmic drug, in reducing the frequency of symptomatic arrhythmia recurrences in patients with atrial fibrillation, atrial flutter or both.
Atrial fibrillation is an increasingly common disorder of the heart rhythm, and most patients with this problem are identified because they have symptoms associated with their arrhythmia. New antiarrhythmic therapies are needed to treat patients with this problem.
A total of 384 patients with a history of atrial fibrillation, atrial flutter or both were randomly assigned to receive once daily doses of placebo or azimilide; recurrent symptomatic arrhythmias were documented using transtelephonic electrocardiogram (ECG) recording. Azimilide 50 mg, 100 mg or 125 mg was tested; the primary efficacy analysis compared the time to first symptomatic recurrence in the combined azimilide 100 mg and 125 mg dose groups with that in the placebo group using the log-rank test.
In the primary efficacy analysis, the time to first symptomatic arrhythmia recurrence was significantly prolonged in the combined azimilide 100 mg and 125 mg daily dose group compared with the placebo group (chi-square 7.96, p = 0.005); the hazard ratio (placebo:azimilide) for this comparison was 1.58 (95% confidence interval [CI] = 1.15, 2.16). In comparisons between individual doses and placebo, the hazard ratio for the 50 mg daily dose was 1.17 (95% CI = 0.83, 1.66; p = 0.37); for the 100 mg group, dose was 1.38 (95% CI = 0.96, 1.98; p = 0.08), and for the 125 mg group, dose was 1.83 (95% CI = 1.24, 2.70; p = 0.002).
Azimilide significantly lengthened the symptomatic arrhythmia-free interval in patients with a history of atrial fibrillation, atrial flutter or both.
Atrial fibrillation (AF) is a disorder of the heart rhythm in which normal sinus rhythm is punctuated by periods of abnormal rhythm that is recognized by patients as the sudden onset of symptoms including palpitations, chest pain and dyspnea. The Cardiovascular Health Study, a cohort study that is following a population >65 years old, showed that the incidence of atrial fibrillation is increasing and that most patients are identified because they have symptoms when their rhythm is abnormal (1). Current antiarrhythmic drugs are limited in their use by imperfect efficacy and uncertain safety. In addition, as the population ages, there will be an increased need for new, convenient antiarrhythmic drugs that reduce the frequency of symptomatic occurrences of atrial fibrillation. Azimilide is a novel class III antiarrhythmic drug that blocks both IKr and IKs channels and that has been developed for atrial fibrillation; the combined effect of IKr and IKs blockade may decrease the proarrhythmic potential of azimilide. Because clinical experience with azimilide is limited (2), in this study azimilide doses of 50, 100 and 125 mg daily (QD) were evaluated to test the hypothesis that azimilide lengthened the symptomatic arrhythmia-free period for patients with AF, atrial flutter or both.
To be eligible, male or female patients with age ≥18 years were required to have a history of symptomatic AF, atrial flutter or both and to be candidates for antiarrhythmic therapy based on the judgment of the investigator enrolling them. Investigators were required to provide an electrocardiogram (ECG) showing AF or atrial flutter that had been recorded within 24 months of the date of randomization, and patients were required to be in sinus rhythm at the time of randomization. Important cardiovascular exclusions were angina at rest; symptoms of heart failure at rest; thoracic surgery, cardiac surgery or myocardial infarction within two months; a history of torsade de pointes or any other polymorphic ventricular tachycardia; resting heart rate <50 beats/min or QTc on an ECG recording in sinus rhythm >440. Other exclusion criteria were blood urea nitrogen >50 mg/dl or serum creatinine >2.0 mg/dl.
Randomization and follow-up
A permuted block randomization scheme was used to assign equal numbers of patients to four treatment groups: placebo or azimilide 50 mg, 100 mg or 125 mg. The assigned treatment was given twice a day (BID) for three days (the loading period) and then the frequency was reduced to QD during the efficacy period for up to 180 days. If a patient had a symptomatic arrhythmia recurrence during the loading period and the arrhythmia was present at midnight of day 3, then the efficacy period was adjusted to begin at the first time the patient had documented resumption of sinus rhythm. If sinus rhythm was not restored (spontaneously or by pharmacologic or direct current cardioversion) by day 10, the patient was withdrawn from the trial.
When initiating randomized therapy, no titration was used and no dose adjustments were made for weight, gender, serum creatinine, associated diseases or concomitant medications. Treatment was initiated in both inpatients and outpatients. Symptomatic arrhythmia recurrence was documented using transtelephonic ECG monitoring. Electrocardiograms recorded during hospital stays or emergency department visits also were obtained; all ECGs from symptomatic arrhythmia recurrences were reviewed by an event committee that was blinded to the patient’s treatment assignment and arrhythmia history. Transtelephonic ECG monitoring also was used every two weeks during follow-up to record an ECG when patients were asymptomatic. Patients completed follow-up by having a symptomatic recurrence of their arrhythmia documented by ECG or by finishing 180 days with no arrhythmia recurrence.
The protocol explicitly specified primary, secondary and additional efficacy analyses. Statistical analyses were done using SAS/STAT (SAS Institute Inc., Cary, North Carolina). The primary outcome variable was the time to first symptomatic arrhythmia recurrence documented by an ECG consistent with AF, atrial flutter or paroxysmal supraventricular tachycardia (completion events). The Kaplan-Meier life-table method was used to display the outcome data and to estimate the median recurrence time for each group (3). Differences between groups were compared with the log-rank test. The Cox proportional hazards model was used to calculate a hazard ratio (placebo:azimilide) for each comparison (3).
The protocol-specified primary efficacy analysis compared the placebo group with the combined 100 mg QD and 125 mg QD doses during the efficacy period (excluding the three-day loading period); this analysis used a two-sided log-rank test and alpha = 0.05. Combining the 100 mg and 125 mg dose groups increased the statistical power of the primary data analysis compared with doing multiple analyses of individual doses. Additional protocol-specified efficacy analyses compared individual doses with placebo.
Because the primary efficacy analysis excluded symptomatic arrhythmia events that occurred during the loading period, all efficacy analyses were repeated using data from the loading period as well as the efficacy period (called day 1 analyses). These latter analyses, which were specified in the protocol, were planned to assess whether including the loading period, when blood azimilide concentrations should have been lower compared with the efficacy period, decreased the magnitude of the treatment effect.
A log-rank trend test was done across all doses to test for a dose response. A secondary efficacy analysis specified in the protocol compared heart rate during arrhythmia recurrences. Finally, asymptomatic arrhythmia occurrences were tabulated in the four treatment groups in an exploratory analysis.
Adverse events were counted for all randomized patients from day 1 and, therefore, included both the loading period and the efficacy period. Patients were required to withdraw for QTc >525. Adverse events were included in the safety analysis if they occurred within 30 days of patient withdrawal for any reason. Safety variables reported here are deaths, torsades de pointes, withdrawal for 12-lead ECG with QTc ≥525 and withdrawals for any other adverse event. The QTc was measured on day 4 from a 12-lead ECG and compared with the QTc measured on a baseline (pretreatment) ECG.
The first patient was enrolled on September 4, 1996, and the last patient finished follow-up on October 6, 1997. The protocol was approved by an institutional review board for clinical investigations at each study site. One protocol amendment was filed to change the statistical analysis plan; this amendment was filed before the study data base was locked or unblinded.
A total of 384 patients were recruited at 97 sites and randomized among the four treatments: 93 patients to placebo, 101 patients to azimilide 50 mg QD, 97 to 100 mg QD and 93 to 125 mg QD. Twenty-eight patients (5.3%) were in the hospital at the time they received their first dose of blinded therapy. Demographic and baseline cardiovascular characteristics were well balanced among the treatment groups (Table 1)as was the use of common concomitant medications (Table 2). Overall 76.8% of patients had some form of structural heart disease, which was defined as a diagnosis of coronary artery disease (with or without myocardial infarction), congestive heart failure, valvular heart disease, hypertension with cardiomyopathy, hypertension with left ventricular hypertrophy, left atrial enlargement by echocardiogram or conduction system disease. Thirty-four percent had a history of treatment with direct current cardioversion before randomization.
About one-third of the patients had a history of multiple arrhythmias recorded by ECG before randomization. Only 61% of patients had just a history of AF while 7% had just atrial flutter. Seventeen percent had prior ECGs showing both AF and atrial flutter and 13% had prior ECGs showing AF or atrial flutter and another regular rhythm without evident atrioventricular dissociation that was typical for paroxysmal supraventricular tachycardia.
Primary efficacy analysis
Among the 384 randomized patients, 367 entered the efficacy period in sinus rhythm after the end of the loading period. The primary efficacy analysis compared placebo patients (n = 87) who entered the efficacy period with a combined group of patients who received azimilide 100 mg QD and patients who received azimilide 125 mg QD (n = 181). The log-rank test p value for this comparison was 0.005 (chi-square 7.96), and the hazard ratio was 1.58 (95% confidence interval [CI] = 1.15, 2.16). The median time to completion event in the placebo group was 17 days compared with 60 days in the combined azimilide group (Fig. 1). The result of the primary efficacy analysis was consistent across subgroups (Fig. 2); that is, efficacy in none of the subgroups was significantly different from efficacy in the overall primary analysis.
When the loading period (day 1 to day 3) and the efficacy period were both included in the comparison of placebo (n = 93) with the combined azimilide group (n = 190), evidence of efficacy was comparable: chi-square 9.4, p = 0.002; hazard ratio 1.61 (95% CI = 1.19, 2.19). There appeared to be a very small, but favorable, effect of azimilide during the loading period; 19.3% of placebo patients had events during the loading period compared with 15.8% of azimilide (100 and 125 mg) patients.
Individual dose efficacy
Higher doses of azimilide were associated with better efficacy (Fig. 3) in the comparison of individual dose groups with the placebo group measured during the efficacy period. A log-rank trend test across all doses was significant (chi-square = 9.8, p = 0.002); that is, as dose increased from placebo to 125 mg QD, efficacy improved. The hazard ratios uniformly improved across doses: 50 mg 1.17 (95% CI = 0.83, 1.66; p = 0.37), 100 mg 1.38 (95% CI = 0.96, 1.98; p = 0.08) and 125 mg 1.83 (95% CI = 1.24, 2.70; p = 0.002). Also, the median time to first symptomatic arrhythmia recurrence was 17 days in the placebo group compared to 22 days in the azimilide 50 mg QD group, 41 days in the 100 mg group and 130 days in the 125 mg group.
Effect of azimilide on arrhythmia and heart rate recorded during completion event and on asymptomatic arrhythmias
The Event Committee interpreted ECGs recorded at the time of completion events as AF in 92% of placebo patients, 86% of the azimilide 50 mg QD group, 95% of the 100 mg QD group and 98% of 125 mg QD group. The few remaining ECGs were consistent with atrial flutter or were typical of paroxysmal supraventricular tachycardia.
The mean heart rates shown on the ECGs recorded at the time of completion events were 123.1 beats per min in the placebo group compared with 113.1 beats per min in the azimilide 50 mg QD group (p = 0.04), 112.9 in the 100 mg QD group (p = 0.06) and 113.6 in the 125 mg QD group (p = 0.08). For the comparison of groups used in the primary efficacy analysis (combined azimilide 100 mg and 125 mg dose group with placebo, as specified in the protocol), the heart rate on azimilide was significantly slower (123.1 beats/min vs. 113.2 beats/min, p = 0.04).
Asymptomatic atrial fibrillation and atrial flutter were occasionally recorded by transtelephonic ECG monitoring at the time of the biweekly routine contacts. These asymptomatic rhythms were recorded from 9.2% of patients in the placebo group compared with 8.1% of the azimilide 50 mg QD group, 4.3% of the 100 mg dose group and 4.4% of the 125 mg dose group (p > 0.20 for all comparisons vs. placebo).
The variables analyzed to assess safety in all randomized patients included deaths, episodes of torsade de pointes and withdrawals due to adverse events. There were no deaths in the placebo group or the azimilide 100 mg dose group; there were two deaths in the azimilide 50 mg dose group and one in the 125 mg group. One of the deaths in patients receiving azimilide 50 mg QD was a 71-year-old woman who died of a stroke 29 days after beginning randomized therapy and 21 days after discontinuing randomized therapy. The other death in a patient receiving azimilide 50 mg QD was unobserved, presumed sudden death 35 days after randomization. This patient was a 61-year-old man with diabetes and a history of myocardial infarction and exercise-induced angina. The death in a patient receiving azimilide 125 mg QD was a 70-year-old man with a history of coronary artery disease, bypass grafting, heart failure, stroke, transient ischemic attacks and carotid endarterectomy two weeks before randomization. This patient had unwitnessed and presumed sudden death eight days after randomization.
One patient in the 100 mg dose group, an 86-year-old woman, had torsade de pointes on day 4 of blinded therapy while she was in atrial flutter with 2.72 s pauses. She recovered and discontinued protocol therapy.
The QTc measured on day 4 was compared with the baseline QTc and showed a mean (± standard deviation) change of +1.0% (±6.6%), +5.3% (±8.2%), +6.2% (±8.3%), +8.5% (±9.1%) in the placebo, 50 mg, 100 mg and 125 mg azimilide groups, respectively. Seven patients were required to withdraw for QTc >525 on ECGs recorded 4 to 187 days after randomization. One of these patients was receiving azimilide 50 mg QD, three were receiving 100 mg QD, and three were receiving 125 mg QD. In addition, withdrawal from the study due to an adverse event occurred in 0, 4, 2 and 2 patients in the placebo, 50 mg, 100 mg and 125 mg dose groups, respectively. The four patients who withdrew from the azimilide 50 mg group had (one each) a skin rash, abnormal liver function tests, diarrhea and a flu-like illness with weakness, dizziness and paresthesias. Two patients withdrew from the 100 mg dose group for (one each) dizziness and syncope. The latter patient was also taking the antiarrhythmic drug disopyramide, a protocol violation; an explanation for the syncope was not found. The two patients who withdrew from the 125 mg dose group had nonsustained ventricular tachycardia during hypokalemia (one patient) and premature ventricular beats (one patient).
Efficacy of azimilide
This randomized clinical trial demonstrated that azimilide is an effective antiarrhythmic drug to reduce the frequency of symptomatic arrhythmia recurrences in patients with AF, atrial flutter or both, and it demonstrated the effective dose range for azimilide. In the primary efficacy analysis, the combined azimilide 100 mg and 125 mg dose group was significantly better than placebo (p = 0.005). The hazard ratio for this comparison was 1.58 (95% CI = 1.15, 2.16), and the median time to first symptomatic arrhythmia recurrence was lengthened from 17 days in the placebo group to 60 days in the combined azimilide group. The analysis of individual doses defined the effective dose range for azimilide. The 50 mg QD dose showed no relevant treatment effect while the 100 mg and 125 mg QD doses both showed clinically important effects.
An event committee reviewed all ECGs from symptomatic events and classified completion events as most consistent with either AF, atrial flutter or paroxysmal supraventricular tachycardia, and AF was the rhythm recorded during completion events in the vast majority of patients in all treatment groups. Importantly, higher doses of azimilide did not appear to increase the occurrence of atrial flutter compared with AF as has been reported with some other antiarrhythmic drugs (4). There was a modest decrease in heart rate during completion events when the combined azimilide 100 mg and 125 mg group was compared with the placebo group. This decrease in heart rate, however, was not accompanied by evidence of an increase in the occurrence of asymptomatic arrhythmias detected by routine biweekly monitoring (5). In fact, asymptomatic arrhythmias appeared to decrease slightly or higher doses of azimilide although the decrease was not statistically significant. It is not known, of course, whether asymptomatic AF is associated with any important adverse patient outcome, nor is it known whether treating asymptomatic AF with an antiarrhythmic drug is associated with beneficial effects. The effect of antiarrhythmic therapy on asymptomatic arrhythmias, however, warrants further study.
More than 90% of patients began randomized therapy with azimilide as an outpatient, and azimilide was well tolerated. Torsade de pointes occurred in only one patient in this study. Three deaths occurred during the study. All occurred in patients with coronary artery disease, and all were thought to have a cardiovascular cause. Two of the deaths were in patients receiving azimilide 50 mg QD, the lowest dose tested and a dose not associated with clinical efficacy. An efficacy study of this size with only three deaths among 384 randomized patients is insufficient to establish the presence or absence of an effect of azimilide on mortality.
Patient withdrawals due to adverse events occurred infrequently and did not appear to increase with the higher doses tested. A precise estimate, however, of the rate at which these events occur in patients taking azimilide will require studying more patients.
This randomized trial included some novel features. Simple inclusion and exclusion criteria were designed to enroll a study population that was representative of many patients with AF and atrial flutter who are prescribed antiarrhythmic drug therapy; patients with all types of AF were included rather than limiting the trial to a single classification such as “paroxysmal” or “chronic” as has been done in many previous trials (6–10). The patients’ arrhythmia histories demonstrated the frequent prior occurrence of multiple arrhythmias in these patients. Azimilide predominantly was begun in an outpatient setting, and no dose titration was used for patients receiving higher doses. Transtelephonic ECG monitoring provided objective documentation of the rhythm at the time of symptoms, and event committee review of all symptomatic ECGs assured a uniform standard for interpretation of these ECGs.
Azimilide may be a promising new antiarrhythmic drug, but additional data on both efficacy and safety are required to establish its role in clinical practice. In this randomized clinical trial, azimilide was initiated in an outpatient setting, was administered once daily, did not require dose titration, showed promising efficacy and was well tolerated.(Appendix)
Investigators, Subinvestigators, Study Coordinators, Institution Name, Number of Patients Enrolled, City: Steven Turner, MD, Cynthia M. Turner, RN, BSN, Heart Clinic of the Ozarks, Inc. (24), Springfield, Missouri; Jay L. Dinerman, MD, Jacksonville Heart Center (20), Jacksonville Beach, Florida; Steven Promisloff, MD, Michael Toren, MD, Danielle Jones, Rebecca Baird, Hillsboro Cardiology (13), Hillsboro, Oregon; Laurence Yellen, MD, Thomas J. Karras, MD, Olga B. Anspe, CCRC, Nancy Junker, CCRC, Cardiology Association Medical Group of East San Diego, Inc. (13), San Diego, California; Daniel Gottlieb, MD, Barbara Westcott, MFA, South Seattle Consulting Physicians (11), Seattle, Washington; Imran K. Niazi, MD, Raed Sweidan, MD, Charles Lanzarotti, MD, Joann Kiemen, RN, Concetta Wagner, RN, Wisconsin Center for Clinical Research (11), Milwaukee, Wisconsin; Richard Wolf, MD, FACC, Abdel Ahmed, MD, Noah Chelliah, MD, Emad Dodin, MD, Dianne Vold, LPN, Kelly Hagen, RN, Denise Carter, RN, United Hospital (11), Grand Forks, North Dakota; David P. Margolis, MD, Oklahoma Heart (10), Tulsa, Oklahoma; C. Andrew DeAbate, MD (9), New Orleans, Louisiana; Rokay Kamyar, MD, Jack S. Madowitz, MD, David Jinich, MD, George P. Reynolds, RCPT, Institute of HealthCare Assessment, Inc. (9), San Diego, California; Nampalli Vijay, MD, Western Cardiology Associates (9), Denver, Colorado; C. Basil Williams, MD, G. Thomas Blanch, MD, Barbara Fowler, Ogden Research Foundation (9), Odgen, Utah; Bart G. Denys, MD, Sandeed A. Patel, MD, Kenneth E. Wong, MD, Stacy M. Henry, CCRC, Cardiovascular Institute of the South Thibodaux (8), Thibodaux, Louisiana; Jose J. Fernandez, MD, Donald Gleason, MD, Robert Hirsch, MD, Jerome Rothbaum, MD, Leonard Joffe, MD, Cynthia Cengic, RN, BSN, Cathy Revere Darby, Advanced Clinical Theapeutics, an affiliate of Affiliated Research Center (8), Tucson, Arizona; Thomas Parker, MD, San Gabriel Clinic (8), Georgetown, Texas; Leland Sprinkle, FACC, MD, M. Joseph McGreevy, MD, FACC, Ronald K. Goldberg, MD, FACC, Sylvia Gerry, BS, C-CVT, Mary Barrett, Cardiology Medical Group of San Diego (8), La Mesa, California; Stuart J. Connolly, MD, Cathrine LeFeuvre, Lorraine Morrison, RN, Hamilton General Hospital (7), Hamilton, Ontario, Canada; Michael Famularo, FACC, MD, Michael Wichman, MD, FACC, Spencer Kulick, MD, FACC, Robert Doria, MD, FACC, Steven Pontius, MD, FACC, Mark Sada, MD, Lawrence VonDollen, MD, FACC, Elizabeth Peck, RN, Coastal Cardiology (7), San Luis Obispo, California; Bruce Goldreyer, MD, Robert Grossman, MD, Norman Zinner, MD, Christina Dyer, Laura Blaszkiewicz, RN, Mandy Sauzzi-Carnes, RN, Western Clinical Research (7), Torrance, California; Michael Imburgia, MD, Teri Stickler, RN, BSN, Louisville Cardiology (7), Louisville, Kentucky; Ronald P. Karlsberg, MD, Joshua Penn, MD, Eli Gang, MD, Tracey S. Gerez, BS, MA, Susan E. Jackman, RN, MS, Cardiovascular Research Institute of Southern California (7), Beverly Hills, California; Steven West, MD, Southwest Florida Heart Group (7), Fort Myers, Florida; Steven Borzak, MD, Charles R. Webb, MD, Stephen Smith, MD, James Tisdale, MD, Theresa Cruz, RN, Henry Ford Hospital (6), Detroit, Michigan; Ira Mitchell Dauber, MD, J. Kern Buckner, MD, Steve Glow, FNP, Terry Wilcox, ANP, Christie Brown, FNP, Shari Bowden, RN, South Denver Cardiology Associates (6), Denver, Colorado; Jacques H.F. Lenis, MD, Peter Carmichael, RN, Nathalie Pilon, RN, INVASCOR Clinical Research (6), Longueuil, Quebec, Canada; Alan Mobley, MD, Bonnie Stevens (6), Conroe, Texas; Melvin Tonkon, FACC, FACP, MD, Cathy Fox, BS, Anaheim Heart and Research Institute (6), Anaheim, California; Jerome Blumenthal, MD, Jean Warren, RCVT, Cardiology & Internal Medicine Associates (5), Marietta, Georgia; Wynne Crawford, MD, C. McGavock Porter, MD, John L. Finklea, MD, H. Forrest Flemming, MD, David Jenkins, MD, David N. Goerge, MD, Paul B. Moore, MD, Laura Thompson, PA, R. Eric Crum, MD, Michael Salvia, MD, Robert Robichaux, MD, W. Ross Davis, MD, Eliyya G. Abbud, MD, Theresa Garrett, RN, R. Rena Grines, RN, Ernest Parker, RN, Mark Platt, RN, Montgomery Cardiovascular Associates, P.C. (5), Montgomery, Alabama; Harold Fleming, MD; Ronald Littlefield, MD; Deborah Weathers, RN; Susan Brown, CCRC; MedQuest, Inc. (5), Greer, South Carolina; Martin Frey, MD, Walter Hepp, MD, Mary Healy, ARNP, Kori Thomas, CNMT, EMT, Heart Center (5), Sarasota, Florida; William Grossman, MD, Charleston Cardiology (5), Charleston, South Carolina; David Kraus, FACC, MD, Massimo F. Giusti, MD, Lisa Burch, RN, Cindy Lee, RN, Cardiology Associates of Memphis (5), Memphis, Tennessee; Robin A. Kuritzky, MD, Royal Columbia Hospital (5), New Westminster, British Columbia, Canada; G. Calvin MacCallum, MD, Sheila Griffiths-Beresford, RN, Health Care Corporation of St. John’s General Hospital Site (5), St. John’s, Newfoundland, Canada; P.F. Adrian Magee, MD, Vasilies Papademetriou, MD, Puneet Narayan, MD, Mrinal Sharma, MD, Erica Marino, CRC, Mary Ellen Royalty, RN, Clinical Research Institute of Northern Virginia (5), Springfield, Virginia; Dennis Ruff, MD, Emanuel P. DeNoia, MD, Donna Griffin, LVN, Healthcare Discoveries (5), San Antonio, Texas; James W. Smith, MD, Karl P. Undesser, PhD, MD, Molly B. Kennedy, RT, MS, Boise Heart Clinic (5), Boise, Idaho; Stephen Winters, MD, Jay H. Curwin, MD, John S. Banes, Jr., MD, Jennifer Heric, RN, MSN, Morristown Memorial Hospital (5), Morristown, New Jersey; Robert Benson, FACC, MD, Candy Ledbetter, RN, BSN, Jacksonville Center for Clinical Research (4), Jacksonville, Florida; John Boulet, MD, ACRC/Arizona Clinical Research Center (4), Tucson, Arizona; Luis Campos, MD, Salah El Hafi, MD, Rick Keister, LVN, CRC, Med-Tech, Inc. (4), Houston, Texas; Harry Colfer, MD, Peter Levamovich, MD, Denise Antonishen, RN, Colleen Shaw, RN, Burns Clinic (4), Petoskey, Missouri; Stuart Damore, MD, Karen Ratner, LVN, Center for Clinical Research (4), Austin, Texas; Robert Davidson, MD, Tower Cardiology Research (4), Los Angeles, California; Mark Geller, MD, Anita R. Betschart, RN, BS (4), Pittsburgh, Pennsylvania; Richard Henthorn, MD, The Christ Hospital (4), Cincinnati, Ohio; William K. K. Hui, MBBS, FRCP (Edin), FRCP (C), FACC, MD, Neil Brass, BSc, MD, FRCP (C), Randall G. Williams, MD, FRCP (C), FACC, Kenneth O’Reilly, BSc, MD, FRCP, Linda Kvill, RN, BSC, N, Royal Alexandra Hospital (4), Edmonton, Alberta, Canada; John King, MD (4), Laguna Hills, California; Wai Hung Lee, MD, Joseph Covello, MD, Daniel Kreichbaum, PharmD, Helen Orvis, CCRC, Health Advance Institute (4), South Bend, Indiana; Louis Rosenfield, MD, Paul K. Albert, ARNP, MSN, Paul Eubanks, CMA, BSN, Cardiology Associates (4), Port Charlotte, Florida; Manfred Sandler, MD, E. Edward Proctor, MD, Searls Videlfsky, MD, Corrine F. Quinn, MD, James E. Barnhill, III, MD, Larisa Brust, RN, BSN, PrevisionMed, Inc. (4), Duluth, Georgia; Frederick Zugibe, MD; Anna R. Zugibe; Newark Cardiovascular Clinic (4), Newark, New York; Arthur Green, DO, James Roberts, MD, Benigno Bobon, MD, Carolyn Goddard, RN, The Crucible Group (3), Tucker, Georgia; Steven Kutalek, MD; Christine Saari, MSN, CCRN, CCRC; Hahnemann University Hospital (3), Philadelphia, Pennsylvania; Donald Meacham, MD, Cardiology Consultants, P.A. (3), Little Rock, Arkansas; Robert Mittleman, MD, University of Massachusetts (3), Worcester, Massachusetts; Douglas Rothrock, MD, Scott B. Baron, MD, Daniel C. Fisher, MD, Teri Seim, RN, Pacific Coast Clinical Coordinators (3), Carmichael, California; U.R. Shettigar, MD, George H. Barbier, MD, M. Siddique, MD, M. Schmiedt, MD, M. Bialow, MD, Doreen Appunn, BSN, Rebecca Diemer, RN, VA Medical Center, Clinical Research (3), Bay Pines, Florida; Eric Spivack, MD, Clinical Cardiology Associates (3), Aventura, Florida; Jeffrey L. Anderson, MD, Sanjeev Trehan, MD, Joseph B. Muhlestein, MD, Kirti Salunkhe, MD, LDS Hospital (2), Salt Lake City, Utah; David Benditt, MD, University of Minnesota Medical School (2), Minneapolis, Minnesota; Jorge Bonet, MD, Kathryn G. Ilott, ARCT, Apex Clinical Research Ltd. (2), Victoria, British Columbia, Canada; Frank Cardello, MD, James V. Felicetta, MD, Nicolette Estrada, RN, MSFNP-C, Sharon Pebbus, RN, BSN, CCRC, Carl T. Hayden, VAMC (2), Phoenix, Arizona; Andrew Cohen, MD, Aurora Denver Cardiology Associates (2), Aurora, Colorado; Luis Constantin, MD, Steven Zelenkofske, DO, Mary Ann Gergits, RN, Heart Care Group (2), Allentown, Pennsylvania; Paul Dorian, MD, D. Newman, MS, Jan Mitchell, RN, St. Michael’s Hospital (2), Toronto, Ontario, Canada; Boris Kerzner, MD (2), Baltimore, Maryland; George Klein, MD, University Hospital (2), London, Ontario, Canada; Claude Nadeau, MD, Deborah G. O’Connor, LPN, Lewis-Gale Clinic, Inc. (2), Salem, Virginia; Kevin Rapeport, MD, George P. Reynolds, RCPT, Harold J. Guy, MD, Institute of HealthCare Assessment, Inc. (2), San Diego, California; Anita Robinson, MD, Future Scripts (2), Spokane, Washington; David Sackin, MD, Ira Klimberg, MD, J. Robert Mcghee, MD, Chan Das, MD, Sid Clevinger, MD, Cary Harbater, CCRC, Urology Center of Florida (2), Ocala, Florida; Joel Sklar, MD, Cardiology Associates of Marin & San Francisco Medical Group, Inc. (2), Larkspur, California; Karen Beckman, MD, Mario Gonzalez, MD, Tammy Denton, RN, University of Oklahoma (1), Oklahoma City, Oklahoma; Anil Bhandari, MD, Beverly Firth, RN, MN, Good Samaritan Hospital (1), Los Angeles, California; R.K. Bhargava, FRCPC, MD, Optimum Clinical Research (1), Oshawa, Ontario, Canada; Paul Colavita, MD, John Fedor, MD, Robert Svenson, MD, John Gallagher, MD, Samuel Zimmon, MD, Connie Dellinger, RN, BSN, Gale Schwarz, RN, The Sanger Clinic, P.A./Carolinas Medical Center (1), Charlotte, North Carolina; George Dennish, MD, Damluji-Bari Clinic (1), San Diego, California; David Fitzgerald, MD, Bowman Gray School of Medicine (1), Winston-Salem, North Carolina; James J. Heger, MD, Lisa Knepper, RN, BSN, Stucky Research Center (1), Fort Wayne, Indiana; Charles W. Karpen, MD, Yi W. Karpen, MS, Decatur Memorial Hospital (1), Decatur, Illinois; Shane Kimber, MD, University of Alberta (1), Edmonton, Alberta, Canada; Greg Koshkarian, FACC, FCCP, MD, Erica C. Forsyth, RN, Desert Cardiology of Tucson (1), Tucson, Arizona; Neal Lerner, MD, Janice Samimi, RN, Manitoba Clinic (1), Winnipeg, Manitoba, Canada; A. R. Zaki Masud, MD, Buffalo Heart Group (1), Buffalo, New York; David M. Mokotoff, MD, David W. Kohl, MD, John G. Finn, MD, Barry S. Weinstock; MD, Solomon Fishman, MD, Bruce Rice, PA-C, Christine Conner, RN, Bay Area Heart Center (1), St. Petersburg, Florida; Richard L. Page, MD, Jose A. Joglar, Patrick J. Welch, Lauren Nelson, BSN, RN, University of Texas Southwestern Medical Center (1), Dallas, Texas; Michael Ptasnik, MD, J.T. Okin, MD, R.K. Law, MD, J.T. Ferrell, MD, J.D. Rubinstein, MD, D.C. Thompson, MD, R.D. Spangler, MD, J.T. Svinarich, MD, F.G. Thurman, MD, C.K. Benedict, MD, Barbara J. McKinster, Cardiovascular Associates, PC (1), Denver, Colorado; Antone F. Salel, MD, A Medical Corporation (1), Encinitas, California; Robert Schnitzler, MD (1), San Antonio, Texas; Matthew Schwinger, MD, Eugene Dula, MD, Teri Matta, CCRC, Cardiovascular Consultants Medical Group (1), Van Nuys, California; Yoseph Shalev, MD, Terry Hornik, RN, Milwaukee Heart Institute (1), Milwaukee, Wisconsin; Sudeep Singh, MD, SARC Research Center (1), Fresno, California; J. Thomas Suh, MD, Kathy Vittum, RN, Regions Hospital (1), St. Paul, Minnesota; John Walker, MD, James E. Carley, MD, David L. Williams, MD, Dianne Tracy, RN, Cardiology Research Associates (1), Ormond Beach, Florida; John Wilson, MD, Cardiology Associates of Cincinnati, Inc. (1), Cincinnati, Ohio; Scientific Advisory Committee: Edward L.C. Pritchett, MD (Chairman), Duke University Medical Center, Durham, North Carolina; Stuart J. Connolly, MD, McMaster University, Hamilton, Ontario, Canada; Stephen R. Marcello, MD, Procter and Gamble Pharmaceuticals, Cincinnati, Ohio; Richard L. Page, MD, University of Texas Southwestern Medical Center, Dallas, Texas; Daniel Schnell, PhD, Procter and Gamble Pharmaceuticals, Cincinnati, Ohio; Elizabeth A. McCarthy, RN (nonvoting), Duke University Medical Center, Durham, North Carolina; Martin D. Phillips, MD (nonvoting), Procter and Gamble Pharmaceuticals, Cincinnati, Ohio; William E. Wilkinson, PhD (nonvoting), Duke University Medical Center, Durham, North Carolina. Data and Safety Monitoring Board: D. George Wyse, MD, PhD (Chairman), University of Calgary, Calgary, Alberta, Canada; Eugene Passamani, MD, Suburban Hospital, Bethesda, MD; William G. Stevenson, MD, Brigham and Women’s Hospital, Boston, Massachusetts; Nanette K. Wenger, MD, Emory University School of Medicine, Atlanta, Georgia; Janet T. Wittes, PhD, Statistics Collaborative, Inc., Washington, DC; William E. Wilkinson, PhD (nonvoting), Duke University Medical Center, Durham, North Carolina; Event Committee: Richard L. Page, MD (Chairman), University of Texas Southwestern Medical Center, Dallas, Texas; Mark D. Carlson, MD, University Hospitals of Cleveland, Case Western Reserve University, Cleveland, Ohio; Anne B. Curtis, MD, University of Florida, Gainesville, Florida; Mark E. Hamer, MD, Evanston Hospital, Northwestern University Medical School, Evanston, Illinois; Katherine T. Murray, MD, Vanderbilt University School of Medicine, Nashville, Tennessee; Elizabeth A. McCarthy, RN (nonvoting), Duke University Medical Center, Durham, North Carolina; Lesley R. Hudson, BS, MT, Quintiles Laboratory Limited, Smyrna, Georgia; Linda Brokaw, BS, Kathleen Walsh, BA, Karen Ferretti, Premier Research Worldwide, Philadelphia, Pennsylvania; Maureen Pieper, MSN, Mary Baugh, RPh, Peter E. Djuric, PharmD, Kendle International, Cincinnati, Ohio.
☆ This study was supported by grants from Procter and Gamble Pharmaceuticals, Cincinnati, Ohio.
- atrial fibrillation
- twice daily
- confidence interval
- Received July 2, 1999.
- Revision received March 1, 2000.
- Accepted April 13, 2000.
- American College of Cardiology
- Psaty B.M.,
- Manolio T.A.,
- Kuller L.H.,
- et al.
- Kalbfleisch J.D.,
- Prentice R.L.
- Page R.L.,
- Wilkinson W.E.,
- Clair W.K.,
- McCarthy E.A.,
- Pritchett E.L.C.
- Anderson J.L.,
- Gilbert E.M.,
- Alpert B.L.,
- et al.
- Boissel J.P.,
- Wolf E.,
- Gillet J.,
- et al.
- Juul-Moller S.,
- Edvardsson N.,
- Rehnqvist-Ahlberg N.
- Levy S.,
- Novella P.,
- Richard P.,
- Paganelli F.