Original Article

Effect of Intensity of Oral Anticoagulation on Stroke Severity and Mortality in Atrial Fibrillation

List of authors.
  • Elaine M. Hylek, M.D., M.P.H.,
  • Alan S. Go, M.D.,
  • Yuchiao Chang, Ph.D.,
  • Nancy G. Jensvold, M.P.H.,
  • Lori E. Henault, M.P.H.,
  • Joe V. Selby, M.D., M.P.H.,
  • and Daniel E. Singer, M.D.

Abstract

Background

The incidence of stroke in patients with atrial fibrillation is greatly reduced by oral anticoagulation, with the full effect seen at international normalized ratio (INR) values of 2.0 or greater. The effect of the intensity of oral anticoagulation on the severity of atrial fibrillation–related stroke is not known but is central to the choice of the target INR.

Methods

We studied incident ischemic strokes in a cohort of 13,559 patients with nonvalvular atrial fibrillation. Strokes were identified through hospitalization data bases and validated on the basis of medical records, which also provided information on the use of warfarin or aspirin, the INR at admission, and coexisting illnesses. The severity of stroke was graded according to a modified Rankin scale. Thirty-day mortality was ascertained from hospitalization and mortality files.

Results

Of 596 ischemic strokes, 32 percent occurred during warfarin therapy, 27 percent during aspirin therapy, and 42 percent during neither type of therapy. Among patients who were taking warfarin, an INR of less than 2.0 at admission, as compared with an INR of 2.0 or greater, independently increased the odds of a severe stroke in a proportional-odds logistic-regression model (odds ratio, 1.9; 95 percent confidence interval, 1.1 to 3.4) across three severity categories and the risk of death within 30 days (hazard ratio, 3.4; 95 percent confidence interval, 1.1 to 10.1). An INR of 1.5 to 1.9 at admission was associated with a mortality rate similar to that for an INR of less than 1.5 (18 percent and 15 percent, respectively). The 30-day mortality rate among patients who were taking aspirin at the time of the stroke was similar to that among patients who were taking warfarin and who had an INR of less than 2.0.

Conclusions

Among patients with nonvalvular atrial fibrillation, anticoagulation that results in an INR of 2.0 or greater reduces not only the frequency of ischemic stroke but also its severity and the risk of death from stroke. Our findings provide further evidence against the use of lower INR target levels in patients with atrial fibrillation.

Introduction

Nonvalvular atrial fibrillation increases the risk of ischemic stroke by a factor of five, presumably by an atrioembolic mechanism.1-4 Such cardioembolic strokes are more severe than other types of ischemic stroke.5-8 Consistent with these observations are data from numerous population-based studies showing that mortality rates are higher for strokes associated with atrial fibrillation.9-15 Randomized trials have shown that warfarin is highly effective in preventing stroke in patients with atrial fibrillation, most likely by minimizing the formation of atrial thrombi.4,16,17 The full efficacy of anticoagulation is seen only at an international normalized ratio (INR) of 2.0 and above.18,19 Although the effect of oral anticoagulants on the frequency of stroke is clear, their effect on the severity of stroke and stroke-related mortality among patients with atrial fibrillation has been less well studied.8,20 In particular, the effect of the INR on the severity of stroke in such patients is unknown. We assessed the effect of the intensity of anticoagulation on the severity of ischemic stroke and on the 30-day mortality rate after stroke in a large cohort of patients with nonvalvular atrial fibrillation. We also determined the rates of ischemic stroke and intracranial hemorrhage according to the intensity of anticoagulation to help physicians and patients make better-informed decisions regarding optimal INR values.

Methods

Study Population

The study cohort has been described previously.21 We identified all patients 18 years or older enrolled in Kaiser Permanente of Northern California, a large, integrated health care system, who had nonvalvular atrial fibrillation between July 1, 1996, and December 31, 1997. Nonvalvular atrial fibrillation was identified on the basis of physician-assigned diagnoses of atrial fibrillation during a routine visit (International Classification of Diseases, 9th revision, Clinical Modification [ICD-9-CM], code 427.31), electrocardiographic data bases, or both.22 Patients with diagnosed mitral stenosis or heart-valve repair or replacement, transient perioperative atrial fibrillation, or recent hyperthyroidism were excluded. The final cohort included 13,559 patients.

Definition and Identification of Events

From each patient's index date through August 31, 1999, we identified ischemic strokes by searching comprehensive automated hospitalization and billing-claims data bases for primary ICD-9-CM discharge diagnoses for ischemic stroke (codes 433.00 to 433.01, 433.10 to 433.11, 433.20 to 433.21, 433.30 to 433.31, 434.00 to 434.01, 434.10 to 434.11, 434.90 to 434.91, and 436.0).22 Medical records were reviewed by a clinical-outcomes committee of three physicians using a formal protocol. Cranial computed tomographic scans or magnetic resonance imaging studies (or both) were available for 98 percent of the patients. A validated ischemic stroke was defined as a neurologic deficit of sudden onset that persisted for more than 24 hours, corresponded to a vascular territory in the absence of primary hemorrhage, and was not explained by other causes (e.g., trauma, infection, or vasculitis). Patients with strokes that occurred during hospitalization or as a complication of a procedure were excluded. Patients who were evaluated in the emergency room and then discharged were included if there was evidence that the neurologic deficit had persisted for more than 24 hours. All events had to be independently verified by two committee members. Disagreements were resolved by a consensus of all three committee members. Any remaining disagreements were adjudicated by a consulting neurologist.

Characteristics of the Patients

Information on clinical variables, obtained from medical charts, included a history of ischemic stroke; a history of cerebrovascular disease, defined by known carotid or vertebrobasilar atherosclerosis or prior carotid endarterectomy; congestive heart failure; coronary artery disease; diabetes mellitus; hypertension; and acute in-hospital treatment with anticoagulants or thrombolytic therapy. Data on age, sex, and racial or ethnic group were obtained from administrative data bases.21

Antithrombotic Medication and INR at Admission

The use of warfarin or aspirin at the time of the outpatient stroke was determined from a review of emergency-room and hospital-admission notes. The INR value at the time of presentation was recorded. The use of warfarin or aspirin was explicitly noted in the medication list obtained at admission in the case of 95 percent of patients. Documentation was incomplete for 28 patients. We subsequently determined that five of these patients had used warfarin on the basis of serial outpatient prescriptions for warfarin recorded in the pharmacy data base of the health plan or recording of INR values up to the time of the stroke in the laboratory data base (or both). We categorized the 23 remaining patients as taking neither warfarin nor aspirin. The INR value at admission was missing for 10 of the 201 patients who were taking warfarin when they were admitted for an ischemic stroke. For two of these patients, we used an outpatient INR value determined within 72 hours before the event. The remaining eight patients were excluded. The antecedent level of anticoagulation control was assessed by calculation of the mean INR for each patient during the six months before the stroke. Ninety-three percent of the patients who had been taking warfarin had such prestroke INR values available.

Severity of Stroke and 30-Day Mortality

Because the focus of our study was the effect of prior oral anticoagulant therapy on the severity of stroke, we excluded 14 patients who had an intracerebral hemorrhage after in-hospital treatment with thrombolytic or heparin therapy for their stroke (3 were taking aspirin on admission, 5 were taking warfarin and had an INR of less than 2.0, and 6 were taking neither aspirin nor warfarin). No patient whose INR was 2.0 or greater on admission had a spontaneous intracerebral hemorrhage after the stroke.

We used the modified Rankin scale adapted by the Oxfordshire Community Stroke Project to classify the functional deficit at hospital discharge, on the basis of a physician's review of the medical record.23 Patients with a score of 1 or 2 were classified as having minor strokes with a residual neurologic deficit that did not interfere with independent living. Patients with a score of 3 or 4 were classified as having major strokes with residual neurologic impairment that prevented independent living. A score of 5 denoted a severe stroke that resulted in total dependence and was usually associated with depressed consciousness. Clinical information was insufficient to determine the extent of the neurologic deficit at discharge for 15 patients (3 patients had been taking aspirin at admission, 2 had been taking warfarin and had an INR of less than 2.0, and 10 had been taking neither aspirin nor warfarin). These 15 patients were excluded from the analyses of the severity of stroke but were included in the analyses of 30-day mortality. We did not formally assess the mechanism of stroke.3

Death from any cause during hospitalization was ascertained by a review of the medical charts, and deaths that occurred after discharge were identified by a review of the health plan's records and the California death registry.24 Thirty-day follow-up data on vital status were complete.

Rates of Ischemic Stroke and Intracranial Hemorrhage According to the Intensity of Anticoagulation

We have previously reported on the effect of warfarin treatment on stroke rates in our cohort.25 Here, we report on the incidence of ischemic stroke and intracranial hemorrhage according to the INR among patients who were taking anticoagulant agents. We focus on intracranial hemorrhage because it is the category of hemorrhagic toxicity that approximates ischemic stroke in terms of clinical and functional effect. The duration of treatment within designated INR ranges was calculated with use of an adapted linear interpolation method.26 Warfarin status was assigned with use of a previously described algorithm incorporating pharmacy prescription information with laboratory INR data.21 To minimize error in interpolation, gaps in INR measurement exceeding 56 days were excluded. Events that occurred during these periods were not included in the INR-specific analyses of event rates.

Statistical Analysis

The severity of stroke was categorized according to the extent of the neurologic deficit at discharge: severe deficit (or in-hospital death), major deficit, or minor or no deficit. The correlation between categories of stroke severity and 30-day mortality was assessed with use of the Mantel–Haenszel chi-square trend test.27 We used proportional-odds ordinal logistic-regression models that allowed us to compare multiple outcome categories in order to assess the independent effect of antithrombotic treatment on the severity of stroke at hospital discharge.28 This approach fits a uniform log cumulative odds of progression across our three categories of severity as a function of antithrombotic therapy at admission and other covariates. The independent effect of antithrombotic therapy on 30-day mortality was assessed with the use of Cox proportional-hazards regression models.29 The proportional odds and the proportional-hazards assumptions were met for these regression models. Kaplan–Meier survival curves were constructed for the 30-day period after stroke for the four antithrombotic medication groups, and these groups were compared with use of the log-rank test.29

The study was approved by the institutional review boards at the Kaiser Permanente Medical Care Program of Northern California and at Massachusetts General Hospital. Because of the nature of the study, the requirement for informed consent was waived.

Results

Base-Line Clinical Features

Table 1. Table 1. Clinical Characteristics of 596 Patients with Nonvalvular Atrial Fibrillation Who Had an Ischemic Stroke, According to the Antithrombotic-Medication Status at Admission.

During the study period, 618 patients with atrial fibrillation and ischemic stroke were identified. Twenty-two were excluded from the analysis: 8 because of missing INR values at admission and 14 because they had an intracerebral hemorrhage after thrombolytic or heparin therapy for their stroke. Of the remaining 596 patients, 188 (32 percent) were taking warfarin at the time of their stroke, 160 (27 percent) were taking aspirin, and 248 (42 percent) were taking neither warfarin nor aspirin (Table 1). Overall, the mean age of the patients was 78 years, and 55 percent were women. Approximately one third of the patients had a history of congestive heart failure, coronary heart disease, or stroke. Nearly 70 percent had hypertension. The warfarin group had a slightly lower mean age than the other two groups, had fewer women, and had a higher prevalence of congestive heart failure, coronary heart disease, diabetes mellitus, and prior stroke. The median INR on admission was 1.7 in the warfarin group, and 62 percent of these values were less than 2.0. The antecedent level of anticoagulation control was nearly identical for patients with an INR of 2.0 or greater at the time of admission and those with an INR of less than 2.0, with a median INR of 2.2 in both groups.

Severity of Stroke and 30-Day Mortality

The severity of stroke was strongly correlated with 30-day mortality (P<0.001). Thirty-nine percent of patients with a severe stroke died within 30 days after discharge, as compared with 13 percent of those with a major deficit, 1 percent of those with a minor deficit, and none of those without neurologic sequelae.

Intensity of Anticoagulation and Severity of Stroke

Table 2. Table 2. Severity of the Neurologic Deficit at Discharge and 30-Day Mortality Rates, According to the Antithrombotic-Medication Status and International Normalized Ratio (INR) at Admission.

Among patients taking warfarin, 15 percent of those with an INR of less than 2.0 either died before discharge or were discharged after having a severe stroke, as compared with 5 percent of patients who had an INR of 2.0 or greater at presentation (Table 2). The proportion of patients who had a severe or fatal stroke did not differ significantly between those with an admission INR of 1.5 to 1.9 and those with an INR of less than 1.5. Twenty-two percent of patients who were not taking antithrombotic medication at the time of the stroke died before discharge or were discharged with a severe deficit, as compared with 13 percent of those taking aspirin.

Table 3. Table 3. Independent Predictors of the Severity of Ischemic Stroke in Patients with Nonvalvular Atrial Fibrillation.

After adjustment for potential confounders in the proportional-odds model, the medication group remained an independent risk factor for the severity of stroke when patients who had an INR of 2.0 or greater were compared with those who had an INR of less than 2.0 or those who were taking neither aspirin nor warfarin (Table 3).

Intensity of Anticoagulation and 30-Day Mortality

Figure 1. Figure 1. Kaplan–Meier Estimates of Survival in the 30 Days after an Ischemic Stroke among Patients with Nonvalvular Atrial Fibrillation, According to the Antithrombotic-Medication Status at Admission.

Patients who were taking warfarin are separated into two groups: those with an international normalized ratio (INR) of less than 2.0, and those with an INR of 2.0 or greater. The P value refers to the overall difference among the groups.

The 30-day mortality rate among patients who were taking neither warfarin nor aspirin when they had an ischemic stroke was 24 percent (Table 2 and Figure 1). Among patients who were taking warfarin at the time of the stroke and who had an INR of 2.0 or greater, 6 percent died within 30 days, as compared with 16 percent of those with an INR of less than 2.0. The 30-day mortality rate was essentially the same among patients who were admitted with an INR between 1.5 and 1.9 and those with an INR of less than 1.5 (18 percent and 15 percent, respectively). The 30-day mortality rate among patients who were taking aspirin when they had a stroke was 15 percent, which was nearly identical to the rate among patients who were taking warfarin and whose INR was less than 2.0.

Table 4. Table 4. Independent Risk Factors for 30-Day Mortality after Ischemic Stroke in Patients with Nonvalvular Atrial Fibrillation.

As compared with an INR of 2.0 or greater, an INR of less than 2.0 at the time of stroke conferred an adjusted relative hazard of 3.4 for death within 30 days (95 percent confidence interval, 1.1 to 10.1) (Table 4). Patients who were taking neither warfarin nor aspirin at the time of the stroke had the highest hazard ratio for 30-day mortality (4.9), as compared with those who were taking warfarin and who had an INR of 2.0 or greater. Among patients who were taking aspirin, the hazard ratio was 2.5, but the confidence interval included a hazard ratio of 1.0. Other independent risk factors for 30-day mortality included older age, congestive heart failure, and diabetes mellitus (Table 4).

Intensity of Anticoagulation and Rates of Stroke and Intracranial Hemorrhage

Table 5. Table 5. Incidence Rates of Ischemic Stroke and Intracranial Hemorrhage among Patients with Nonvalvular Atrial Fibrillation Who Were Taking Warfarin, According to the International Normalized Ratio (INR) at the Time of the Stroke.

To provide a more comprehensive accounting of the clinical sequelae of various INR values, we calculated the incidence rates of ischemic stroke and intracranial hemorrhage according to the INR among the patients who were taking warfarin (Table 5). The rate of ischemic stroke was highest at INR values of less than 2.0, especially values of less than 1.5. By contrast, there was no marked absolute increase in the rate of intracranial hemorrhage at INR values of less than 4.0.

Discussion

Prior studies have demonstrated that the incidence of ischemic stroke among patients with atrial fibrillation is greatly reduced by warfarin therapy that results in an INR of at least 2.0, a relation confirmed by our results.18,19 We also found that an INR of 2.0 or greater markedly reduces the severity of stroke and the short-term mortality rate. Patients with atrial fibrillation and an INR of less than 2.0 who had an ischemic stroke faced a risk of death within 30 days that was more than three times the risk among patients with an INR of 2.0 or greater. The outcomes were essentially equally poor among warfarin users with an INR of 1.5 to 1.9 at admission and those with an admission INR of less than 1.5. The outcome was even worse for those who were receiving no antithrombotic medication. The outcome among patients who were taking aspirin when they had a stroke was similar to that among patients who were taking warfarin and whose admission INR was less than 2.0. The lessened severity of stroke associated with more intense anticoagulation may reflect a reduced proportion of strokes from a cardiac source, a smaller thrombus, or both.3,5,8

Analyses relating the severity of stroke in patients with atrial fibrillation to the intensity of anticoagulation require large numbers of subjects who have a stroke while taking anticoagulants. Given the efficacy of anticoagulants for atrial fibrillation, it is difficult to find adequate numbers of subjects. In our cohort, there were 596 analyzable ischemic strokes, with 188 occurring among patients who were taking warfarin. By contrast, among five early primary-prevention trials of patients with atrial fibrillation there were only 27 strokes among patients treated with warfarin, and in the three Stroke Prevention in Atrial Fibrillation trials combined, there were only 25 strokes among those treated with adjusted-dose warfarin.8,16

Our results have a direct bearing on the decisions regarding the use of anticoagulant therapy in patients with atrial fibrillation and the target INR level. Guidelines and formal decision analyses have focused exclusively on the efficacy of warfarin in preventing stroke in patients with atrial fibrillation.30-35 Our findings highlight an important incremental benefit of anticoagulation: strokes that occur among patients with adequate anticoagulation are far less likely to result in severe disability or death. In formal utility assessments, patients rate strokes very differently, depending on the severity of the resulting deficits. In particular, patients frequently rate severely disabling stroke as equivalent to or worse than death.36 Recent guidelines of the American College of Cardiology–American Heart Association–European Society for Cardiology suggest the use of a lower INR target for certain patients with atrial fibrillation who are older than 75 years of age.35 Our data indicate that an INR of less than 2.0 will substantially increase the likelihood of death and severe disability from atrial fibrillation–related stroke.18,37-42 We also observed little additional risk of intracranial hemorrhage with the use of warfarin until INR values exceeded 3.9. These findings weigh against the use of target INR values below 2.0.

Potential limitations of our study should be acknowledged. Ours was an observational assessment of the effect of treatment. Randomized trials of lower levels of anticoagulation would be difficult to implement, given prior negative results of such an approach, and the numbers of strokes would probably be too small for an assessment of the effect on the severity of stroke.18,19 We observed strong effects while simultaneously controlling for other important determinants of the outcome. Residual confounding by indication is particularly unlikely in our central comparison of two levels of anticoagulation, since both groups of patients were prescribed warfarin and had very similar INR values in the six months preceding their strokes.43 It remains possible that some patients who had a minor stroke either did not seek medical care or were treated as outpatients. The 30-day mortality rate in the group of patients in our study who were not receiving anticoagulant therapy (24 percent) was similar to 30-day mortality rates for atrial fibrillation–related stroke in such population-based studies as the Framingham Study (25 percent),15 the Oxfordshire Community Stroke Project (23 percent),9,23 and the Italian Acute Stroke Study (27 percent),14 arguing against significant bias in our ascertainment of stroke. Furthermore, it is highly unlikely that the rates of hospitalization for minor strokes would be systematically disproportionate on the basis of the use or intensity of anticoagulation.

In this study of ischemic strokes in a large cohort of patients with nonvalvular atrial fibrillation, therapeutic anticoagulation resulting in an INR of 2.0 or greater at the time of the stroke was associated with less severe neurologic deficit than the absence of antithrombolytic therapy or therapy that resulted in a lower INR. The 30-day mortality rate was 6 percent among patients who were taking warfarin and who had an INR of 2.0 or greater, as compared with 16 percent among those who were taking warfarin and who had an INR of less than 2.0, 15 percent among those who were taking aspirin, and 24 percent among those who were taking neither aspirin nor warfarin. There was no significant difference in mortality rates between patients taking warfarin who had an INR of 1.5 to 1.9 and those taking warfarin who had an INR of less than 1.5. The risk of intracranial hemorrhage in our cohort did not increase until INR values exceeded 3.9. Our results provide further support for the use of anticoagulation to achieve an INR of 2.0 or greater (e.g., 2.5) in patients with nonvalvular atrial fibrillation.

Funding and Disclosures

Supported by a grant from the National Institute on Aging (5 R01 AG15478).

Drs. Hylek and Singer report having received grant support from DuPont Pharma, now Bristol-Myers Squibb.

We are indebted to Dr. Philip A. Wolf of Boston University School of Medicine for providing expert consultation in neurology and epidemiology and to Kathleen A. Phillips and Angela M. Capra, M.A., of the Division of Research, Kaiser Permanente of Northern California, for providing technical assistance.

Author Affiliations

From the General Medicine Division, Clinical Epidemiology Unit, Massachusetts General Hospital, Harvard Medical School, Boston (E.M.H., Y.C., L.E.H., D.E.S.); the Division of Research, Kaiser Permanente of Northern California, Oakland (A.S.G., N.G.J., J.V.S.); and the Departments of Epidemiology, Biostatistics, and Medicine, University of California, San Francisco, San Francisco (A.S.G.).

Address reprint requests to Dr. Hylek at the General Medicine Division, Clinical Epidemiology Unit, Massachusetts General Hospital, 50 Staniford St., 9th Fl., Boston, MA 02114.

References (43)

  1. 1. Wolf PA, Abbott RD, Kannel WB. Atrial fibrillation as an independent risk factor for stroke: the Framingham Study. Stroke 1991;22:983-988

  2. 2. Britton M, Gustafsson C. Non-rheumatic atrial fibrillation as a risk factor for stroke. Stroke 1985;16:182-188

  3. 3. Bogousslavsky J, Van Melle G, Regli F, Kappenberger L. Pathogenesis of anterior circulation stroke in patients with nonvalvular atrial fibrillation: the Lausanne Stroke Registry. Neurology 1990;40:1046-1050

  4. 4. Manning WJ, Silverman DI, Waksmonski CA, Oettgen P, Douglas PS. Prevalence of residual left atrial thrombi among patients with acute thromboembolism and newly recognized atrial fibrillation. Arch Intern Med 1995;155:2193-2198

  5. 5. Longstreth WT Jr, Bernick C, Fitzpatrick A, et al. Frequency and predictors of stroke death in 5,888 participants in the Cardiovascular Health Study. Neurology 2001;56:368-375

  6. 6. Sherman DG, Goldman L, Whiting RB, Jurgensen K, Kaste M, Easton JD. Thromboembolism in patients with atrial fibrillation. Arch Neurol 1984;41:708-710

  7. 7. Miller VT, Pearce LA, Feinberg WM, Rothrock JF, Anderson DC, Hart RG. Differential effect of aspirin versus warfarin on clinical stroke types in patients with atrial fibrillation. Neurology 1996;46:238-240

  8. 8. Hart RG, Pearce LA, Miller VT, et al. Cardioembolic vs. noncardioembolic strokes in atrial fibrillation: frequency and effect of antithrombotic agents in the Stroke Prevention in Atrial Fibrillation studies. Cerebrovasc Dis 2000;10:39-43

  9. 9. Sandercock P, Bamford J, Dennis M, et al. Atrial fibrillation and stroke: prevalence in different types of stroke and influence on early and long term prognosis (Oxfordshire community stroke project). BMJ 1992;305:1460-1465

  10. 10. Gustafsson C, Britton M. Pathogenetic mechanism of stroke in non-valvular atrial fibrillation: follow-up of stroke patients with and without atrial fibrillation. J Intern Med 1991;230:11-16

  11. 11. Kaarisalo MM, Immonen-Raiha P, Marttila RJ, et al. Atrial fibrillation and stroke: mortality and causes of death after the first acute ischemic stroke. Stroke 1997;28:311-315

  12. 12. Penado S, Cano M, Acha O, Hernandez JL, Riancho JA. Stroke severity in patients with atrial fibrillation. Am J Med 2002;112:572-574

  13. 13. Lamassa M, Di Carlo A, Pracucci G, et al. Characteristics, outcome, and care of stroke associated with atrial fibrillation in Europe: data from a multinational hospital-based registry (The European Community Stroke Project). Stroke 2001;32:392-398

  14. 14. Candelise L, Pinardi G, Morabito A. Mortality in acute stroke with atrial fibrillation. Stroke 1991;22:169-174

  15. 15. Lin HJ, Wolf PA, Kelly-Hayes M, et al. Stroke severity in atrial fibrillation: the Framingham Study. Stroke 1996;27:1760-1764

  16. 16. Risk factors for stroke and efficacy of antithrombotic therapy in atrial fibrillation: analysis of pooled data from five randomized controlled trials. Arch Intern Med 1994;154:1449-1457[Erratum, Arch Intern Med 1994;154:2254.]

  17. 17. Hart RG, Benavente O, McBride R, Pearce LA. Antithrombotic therapy to prevent stroke in patients with atrial fibrillation: a meta-analysis. Ann Intern Med 1999;131:492-501

  18. 18. Adjusted-dose warfarin versus low-intensity, fixed-dose warfarin plus aspirin for high-risk patients with atrial fibrillation: Stroke Prevention in Atrial Fibrillation III randomised clinical trial. Lancet 1996;348:633-638

  19. 19. Hylek EM, Skates SJ, Sheehan MA, Singer DE. An analysis of the lowest effective intensity of prophylactic anticoagulation for patients with nonrheumatic atrial fibrillation. N Engl J Med 1996;335:540-546

  20. 20. van Walraven C, Hart RG, Singer DE, et al. Oral anticoagulants vs aspirin in nonvalvular atrial fibrillation: an individual patient meta-analysis. JAMA 2002;288:2441-2448

  21. 21. Go AS, Hylek EM, Borowsky LH, Phillips KA, Selby JV, Singer DE. Warfarin use among ambulatory patients with nonvalvular atrial fibrillation: the anticoagulation and risk factors in atrial fibrillation (ATRIA) study. Ann Intern Med 1999;131:927-934

  22. 22. International classification of diseases, 9th revision, clinical modification (ICD-9-CM). Vol. 1 and 2. Dover, Del.: American Medical Association, 1999.

  23. 23. Bamford J, Sandercock P, Dennis M, Burn J, Warlow C. A prospective study of acute cerebrovascular disease in the community: the Oxfordshire Community Stroke Project -- 1981-1986. 2. Incidence, case fatality rates and overall outcome at one year of cerebral infarction, primary intracerebral and subarachnoid haemorrhage. J Neurol Neurosurg Psychiatry 1990;53:16-22

  24. 24. Arellano MG, Petersen GR, Petitti DB, Smith RE. The California Automated Mortality Linkage System (CAMLIS). Am J Public Health 1984;74:1324-1330

  25. 25. Go AS, Hylek EM, Phillips KA, et al. Effectiveness and safety of warfarin to prevent thromboembolism among 13,559 patients with atrial fibrillation: how well do results of randomized trials apply to clinical practice? The ATRIA Study. Circulation 2002;106:Suppl II:II-515 abstract.

  26. 26. Rosendaal FR, Cannegieter SC, van der Meer FJ, Briet E. A method to determine the optimal intensity of oral anticoagulant therapy. Thromb Haemost 1993;69:236-239

  27. 27. Mantel N. Chi-square tests with one degree of freedom: extensions of the Mantel-Haenszel procedure. J Am Stat Assoc 1963;58:690-700

  28. 28. McCullagh P. Regression models for ordinal data. J Stat Soc [B] 1980;42:109-142

  29. 29. Kalbfleisch JD, Prentice RL. The statistical analysis of failure time data. New York: John Wiley, 1980.

  30. 30. Eckman MH, Falk RH, Pauker SG. Cost-effectiveness of therapies for patients with nonvalvular atrial fibrillation. Arch Intern Med 1998;158:1669-1677

  31. 31. Gage BF, Cardinalli AB, Albers GW, Owens DK. Cost-effectiveness of warfarin and aspirin for prophylaxis of stroke in patients with nonvalvular atrial fibrillation. JAMA 1995;274:1839-1845

  32. 32. Thomson R, Parkin D, Eccles M, Sudlow M, Robinson A. Decision analysis and guidelines for anticoagulant therapy to prevent stroke in patients with atrial fibrillation. Lancet 2000;355:956-962[Erratum, Lancet 2000;355:1466.]

  33. 33. Man-Son-Hing M, Nichol G, Lau A, Laupacis A. Choosing antithrombotic therapy for elderly patients with atrial fibrillation who are at risk for falls. Arch Intern Med 1999;159:677-685

  34. 34. Albers GW, Dalen JE, Laupacis A, Manning WJ, Petersen P, Singer DE. Antithrombotic therapy in atrial fibrillation. Chest 2001;119:Suppl:194S-206S

  35. 35. Fuster V, Ryden LE, Asinger RW, et al. ACC/AHA/ESC guidelines for the management of patients with atrial fibrillation: executive summary: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines and the European Society of Cardiology Committee for Practice Guidelines and Policy Conferences (Committee to Develop Guidelines for the Management of Patients with Atrial Fibrillation) developed in collaboration with the North American Society of Pacing and Electrophysiology. Circulation 2001;104:2118-2150

  36. 36. Gage BF, Cardinalli AB, Owens DK. The effect of stroke and stroke prophylaxis with aspirin or warfarin on quality of life. Arch Intern Med 1996;156:1829-1836

  37. 37. Hylek EM, Singer DE. Risk factors for intracranial hemorrhage in outpatients taking warfarin. Ann Intern Med 1994;120:897-902

  38. 38. The Stroke Prevention in Atrial Fibrillation Investigators. Bleeding during antithrombotic therapy in patients with atrial fibrillation. Arch Intern Med 1996;156:409-416

  39. 39. Landefeld CS, Goldman L. Major bleeding in outpatients treated with warfarin: incidence and prediction by factors known at the start of outpatient therapy. Am J Med 1989;87:144-152

  40. 40. van der Meer FJ, Rosendaal FR, Vandenbroucke JP, Briet E. Bleeding complications in oral anticoagulant therapy: an analysis of risk factors. Arch Intern Med 1993;153:1557-1562

  41. 41. Palareti G, Leali N, Coccheri S, et al. Bleeding complications of oral anticoagulant treatment: an inception-cohort, prospective collaborative study (ISCOAT). Lancet 1996;348:423-428

  42. 42. Evans A, Kalra L. Are the results of randomized controlled trials on anticoagulation in patients with atrial fibrillation generalizable to clinical practice? Arch Intern Med 2001;161:1443-1447

  43. 43. Miettinen OS. The need for randomization in the study of intended effects. Stat Med 1983;2:267-271

Citing Articles (836)

    Letters

    Figures/Media

    1. Table 1. Clinical Characteristics of 596 Patients with Nonvalvular Atrial Fibrillation Who Had an Ischemic Stroke, According to the Antithrombotic-Medication Status at Admission.
      Table 1. Clinical Characteristics of 596 Patients with Nonvalvular Atrial Fibrillation Who Had an Ischemic Stroke, According to the Antithrombotic-Medication Status at Admission.
    2. Table 2. Severity of the Neurologic Deficit at Discharge and 30-Day Mortality Rates, According to the Antithrombotic-Medication Status and International Normalized Ratio (INR) at Admission.
      Table 2. Severity of the Neurologic Deficit at Discharge and 30-Day Mortality Rates, According to the Antithrombotic-Medication Status and International Normalized Ratio (INR) at Admission.
    3. Table 3. Independent Predictors of the Severity of Ischemic Stroke in Patients with Nonvalvular Atrial Fibrillation.
      Table 3. Independent Predictors of the Severity of Ischemic Stroke in Patients with Nonvalvular Atrial Fibrillation.
    4. Figure 1. Kaplan–Meier Estimates of Survival in the 30 Days after an Ischemic Stroke among Patients with Nonvalvular Atrial Fibrillation, According to the Antithrombotic-Medication Status at Admission.
      Figure 1. Kaplan–Meier Estimates of Survival in the 30 Days after an Ischemic Stroke among Patients with Nonvalvular Atrial Fibrillation, According to the Antithrombotic-Medication Status at Admission.

      Patients who were taking warfarin are separated into two groups: those with an international normalized ratio (INR) of less than 2.0, and those with an INR of 2.0 or greater. The P value refers to the overall difference among the groups.

    5. Table 4. Independent Risk Factors for 30-Day Mortality after Ischemic Stroke in Patients with Nonvalvular Atrial Fibrillation.
      Table 4. Independent Risk Factors for 30-Day Mortality after Ischemic Stroke in Patients with Nonvalvular Atrial Fibrillation.
    6. Table 5. Incidence Rates of Ischemic Stroke and Intracranial Hemorrhage among Patients with Nonvalvular Atrial Fibrillation Who Were Taking Warfarin, According to the International Normalized Ratio (INR) at the Time of the Stroke.
      Table 5. Incidence Rates of Ischemic Stroke and Intracranial Hemorrhage among Patients with Nonvalvular Atrial Fibrillation Who Were Taking Warfarin, According to the International Normalized Ratio (INR) at the Time of the Stroke.