Correspondence
Warfarin in the Prevention of Stroke Associated with Nonrheumatic Atrial Fibrillation
N Engl J Med 1993; 328:1041-1043April 8, 1993
- Article
To the Editor:
The article by Ezekowitz et al. on warfarin in the prevention of stroke associated with nonrheumatic atrial fibrillation (Nov. 12 issue)1 is another example of discrimination against women in health issues. Only men were enrolled in this study, yet women do have atrial fibrillation. Although most veterans are men, there are female veterans (I am one), and therefore the Department of Veterans Affairs should not be funding sex-discriminatory research. I am surprised that the female authors were not more conscious of these issues.
1 ReferencesJean E. Howard, M.D.
Brookhaven National Laboratory, Upton, NY 119731
Ezekowitz MD ,Bridgers SL ,James KE , et al. Warfarin in the prevention of stroke associated with nonrheumatic atrial fibrillation. N Engl J Med 1992;327:1406-1412
Full Text | Web of Science | Medline
To the Editor:
The failure to use the system recommended by the World Health Organization of standardizing the prothrombin-time ratio according to International Normalized Ratios (INRs) may have masked the full importance of the important study by Ezekowitz et al. The authors state that they used a therapeutic range for the prothrombin-time ratio of 1.2 to 1.5, “corresponding to an INR of 1.4 to 2.8.” The extent to which the warfarin dose may have been reduced in terms of traditional North American standards1 is only apparent, however, from the additional statement that the International Sensitivity Index (ISI) of the thromboplastin reagents used at the various centers ranged from 1.5 to 2.6. Thus, the INR corresponding to the target interval for the prothrombin-time ratio of 1.2 to 1.5 could have ranged from between 1.3 and 1.6 to between 1.6 and 2.85.
With these guidelines some centers may have been using a warfarin dose that was less intense than the dose recommended for the secondary prevention of established thrombotic disorders, particularly in the arterial circulation. Other centers using reagents with relatively high ISIs were probably using a low dose but one that was more in agreement with the established recommended range2. Reporting the relation between the incidence of stroke or hemorrhage and the intensity of anticoagulation according to the ISI of thromboplastin would have been more helpful than providing the relatively meaningless prothrombin-time ratios. The mean warfarin dose used in the study would also have been of great interest as a guide to the intensity of anticoagulation.
As can be seen in Table 1Table 1
Overall Intensity and Effect of Anticoagulation Provided in Four Studies of Patients with Atrial Fibrillation., the overall intensity of anticoagulation provided in the Veterans Affairs study was probably less than that in recent randomized studies3-5. At some centers that appear to have been successful in preventing further strokes, it may have been at an extremely low level, further encouraging the current trend toward the use of a minimal dose of warfarin to maximize the ratio of benefit to risk.5 ReferencesL. Poller, D.Sc., M.D., F.R.C.Path.
Withington Hospital, Manchester M20 8LR, United Kingdom1
Poller L ,Taberner DA . Dosage and control of oral anticoagulants: an international collaborative survey. Br J Haematol 1982;51:479-485
CrossRef | Web of Science | Medline2
Hirsh J ,Poller L ,Deykin D ,Levine M ,Dalen JE . Optimal therapeutic range for oral anticoagulants. Chest 1989;95:Suppl:5S-11S
Web of Science | Medline3
Petersen P ,Boysen G ,Godtfredsen J ,Andersen ED ,Andersen B . Placebo-controlled, randomised trial of warfarin and aspirin for prevention of thromboembolic complications in chronic atrial fibrillation: the Copenhagen AFASAK study. Lancet 1989;1:175-179
CrossRef | Web of Science | Medline4
Stroke Prevention in Atrial Fibrillation Study Group Investigators
Web of Science | Medline5
The Boston Area Anticoagulation Trial for Atrial Fibrillation Investigators
Full Text | Web of Science | Medline
To the Editor:
It would seem that the five recent randomized trials of the efficacy of warfarin in preventing strokes in patients with nonrheumatic atrial fibrillation should encourage physicians to prescribe this drug more often in such patients1-5. A reminder of how the patients were selected is worthy of mention. In the latest trial1 and in the largest of the five,2 patients with atrial fibrillation were screened, and approximately 93 percent were excluded from entry. Admittedly, up to 28 percent of these were patients in whom antithrombotic therapy was thought to be definitely indicated, but there was still a large proportion of subjects who were said to have had contraindications to this form of treatment. Other important reasons for exclusion included patients' refusal to enter the studies and refusal by the physicians to enroll patients in the studies.
Even among the patients who were randomly assigned to warfarin therapy, the withdrawal rate ranged from 19 to 38 percent in four of the trials1-3,5. In the Boston trial,4 there was only a 10 percent rate of permanent withdrawal in the warfarin group, but the proportion of subjects who were eventually recruited into the study was not stated.
In the Canadian trial,5 detailed documentation of the reasons for exclusion was available for 1430 patients screened over an eight-month period. Eighty-seven percent of these patients were excluded, and of the 13 percent who were eligible, only 54 percent gave consent and underwent randomization. A worrisome result of this trial was that the rate of fatal or major bleeding was 2.5 percent per year in the warfarin group, as compared with 0.5 percent in the placebo group.
Finally, it is unclear why Ezekowitz et al. “believe that the treatment of patients with intermittent atrial fibrillation should be similar to that of patients with chronic atrial fibrillation,” when they purposely excluded this group of patients from entering their study.
5 ReferencesDennis J. Barnes, M.R.C.P.
Guy's Hospital, London SEI 9RT, United Kingdom1
Ezekowitz MD ,Bridgers SL ,James KE , et al. Warfarin in the prevention of stroke associated with nonrheumatic atrial fibrillation. N Engl J Med 1992;327:1406-1412
Full Text | Web of Science | Medline2
Stroke Prevention in Atrial Fibrillation Study: final resultsCirculation 1991;84:527-539
Web of Science | Medline3
Petersen P ,Boysen G ,Godtfredsen J ,Andersen ED ,Andersen B . Placebo-controlled, randomised trial of warfarin and aspirin for prevention of thromboembolic complications in chronic atrial fibrillation: the Copenhagen AFASAK study. Lancet 1989;1:175-179
CrossRef | Web of Science | Medline4
The Boston Area Anticoagulation Trial for Atrial Fibrillation Investigators
Full Text | Web of Science | Medline5
Connolly SJ ,Laupacis A ,Gent M ,Roberts RS ,Cairns JA ,Joyner C . Canadian Atrial Fibrillation Anticoagulation (CAFA) Study. J Am Coll Cardiol 1991;18:349-355
CrossRef | Web of Science | Medline
Author/Editor Response
The authors reply:
To the Editor: Dr. Howard questions our omission of women from our study. Our intention was not to discriminate against women. When the study was designed, women made up 2.5 percent of the patients treated by the Department of Veterans Affairs. Thus, it would not have been possible to analyze the data on women separately from those on men. Women were included in the four other studies of atrial fibrillation1-4. No difference was found between men and women with respect to the incidence of stroke or the efficacy of treatment.
We agree with Dr. Poller's comments regarding INRs. We would have preferred to use INRs in our study, but were unable to do so, as we stated in the paper, because an international reference thromboplastin was not available for all the thromboplastin preparations used in the study.
Dr. Barnes' question regarding the similarities between patients in our study and those in the general population of patients with atrial fibrillation is important. In the paper we enumerated the reasons for excluding patients from our study. Twenty-two percent of the patients who were screened had intermittent atrial fibrillation and, according to the protocol, were excluded. The three studies that evaluated patients with intermittent atrial fibrillation found that the incidence of stroke and the effect of treatment were similar to those in patients with persistent atrial fibrillation2-4. In retrospect, it is unfortunate that we excluded these patients. Eleven percent of the patients were excluded because there was a definite indication for anticoagulant or antiplatelet therapy. Twenty percent of otherwise eligible patients were excluded for administrative reasons. The important figure is the 40 percent of patients who had contraindications to anticoagulation. Thus, on the basis of this study, we believe that approximately 50 to 60 percent of patients with atrial fibrillation can safely undergo anticoagulation.
We are now analyzing our data with respect to the anticoagulant dose. At present, we have no accurate computation of the warfarin dose given when the patients' values were in the therapeutic range.
4 ReferencesMichael D. Ezekowitz, M.D., Ph.D.
Yale University School of Medicine, New Haven, CT 06510Kenneth E. James, Ph.D.
Veterans Affairs Medical Center, Portland, OR 972011
Petersen P ,Boysen G ,Godtfredsen J ,Andersen ED ,Andersen B . Placebo-controlled, randomised trial of warfarin and aspirin for prevention of thromboembolic complications in chronic atrial fibrillation: the Copenhagen AFASAK study. Lancet 1989;1:175-179
CrossRef | Web of Science | Medline2
The Boston Area Anticoagulation Trial for Atrial Fibrillation Investigators
Full Text | Web of Science | Medline3
Stroke Prevention in Atrial Fibrillation Study: final resultsCirculation 1991;84:527-539
Web of Science | Medline4
Connolly SJ ,Laupacis A ,Gent M ,Roberts RS ,Cairns JA ,Joyner C . Canadian Atrial Fibrillation Anticoagulation (CAFA) Study. J Am Coll Cardiol 1991;18:349-355
CrossRef | Web of Science | Medline
