Published online before print July 17, 2008, doi: 10.1161/STROKEAHA.107.508580
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
(Stroke. 2008;39:2736.)
© 2008 American Heart Association, Inc.
Original Contributions |
Racial/Ethnic Differences in Ischemic Stroke Rates and the Efficacy of Warfarin Among Patients With Atrial Fibrillation
From Department of Research and Evaluation (J.F.Y., W.C.), Kaiser Permanente Medical Center, Pasadena, Calif; Center for Interventional Therapeutics (CIVT) (S.S.B.), Columbia University Medical Center, New York, NY; Department of Cardiology (A.Y.-J.S., M.B.J.), Department of Electrophysiology (M.B.J., X.W.), Kaiser Permanente Medical Center, Los Angeles, Calif.
Correspondence to Albert Yuh-Jer Shen, Department of Cardiology, Kaiser Permanente Medical Center, 1526 Edgemont St, 2nd floor, Los Angeles, CA 90027; E-mail albert.y-j.shen{at}kp.org
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Background and Purpose— Warfarin reduces stroke risk in studies of predominantly white patients with atrial fibrillation (AF). Whether nonwhites also have lower rates of stroke while treated with warfarin is unclear.
Methods— A multiethnic stroke-free cohort hospitalized with nonrheumatic AF was identified in a large health maintenance organization. Stroke risk factors (advanced age, diabetes, hypertension, and heart failure), warfarin use, and anticoagulation intensity were assessed. Crude ischemic stroke rates were calculated by Poisson regression for each group while using and not using warfarin. Cox proportional hazard models were constructed to assess the independent effect of race/ethnicity on ischemic stroke.
Results— Between 1995 and 2000, we identified 18867 AF hospitalizations (78.5% white, 8% black, 9.5% Hispanic, and 3.9% Asian). Over the course of 63204 person-years follow-up (median, 3.3 years), 1226 ischemic strokes were identified. The percent-time on warfarin did not differ by race/ethnicity. The median percent-time on warfarin that international normalized ratio was 2 to 3 was 54.5% overall, but it was lower in blacks at 47.8%, whereas the other groups had a rate of 
54%. The rate ratios (95% CI) of ischemic stroke with warfarin compared to without warfarin for whites, blacks, Hispanics, and Asians were 0.79 (0.68 to 0.90), 0.92 (0.65 to 1.30), 0.71 (0.48 to 1.05), and 0.65 (0.34 to 1.23), respectively.
Conclusions— In this cohort, we did not observe a statistically significant lower rate of stroke with warfarin therapy among nonwhites (in particular blacks) with previous AF hospitalizations. The relatively small numbers of nonwhites renders our estimates less than precise and should be interpreted with caution.
Key Words: atrial fibrillation • racial differences • stroke • warfarin
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
With a prevalence of
1% among the adult population,1 atrial fibrillation (AF) is the most common arrhythmia that requires treatment. Extrapolations from population studies have estimated that 2.3 million Americans have AF.1,2 Outcome studies have shown that AF is a potent risk factor for stroke, increasing its risk by nearly 5- fold.3 In a pooled analysis of 5 randomized trials, adjusted dose warfarin reduced the annual rate of any stroke by 68%.4 However, 90% of patients in the pooled analysis were white, whereas stroke risk differs by race/ethnicity. There are sparse data on the benefits of anticoagulation among nonwhite patients. In population-based studies, the prevalence and incidence of all stroke types in blacks and Hispanics are higher than in whites.5–7 Whereas stroke is more prevalent among Asians predominantly because of greater rates of hemorrhagic stroke, ischemic stroke risk may be lower among Asians than whites.8,9 Stroke attributable to AF is more likely in whites than in nonwhites.5,10,11 In the Northern Manhattan Stroke Study of incident stroke in a multiethnic community, AF was prevalent in 29% of whites but in only 14% of Hispanics and in 18% of blacks.12 In a case control substudy, the population-attributable risk for AF, which estimates the proportion of strokes attributable to AF, was 0.20 in whites but only 0.02 in Hispanics and 0.03 in blacks.13 Another study reported that the rates of cardioembolic strokes among blacks and Hispanics were lower than the rates of nonembolic strokes when compared to whites.14 These data raise the possibility that the mechanism of stroke in nonwhite patients with AF may differ from that in whites. Nonwhites with AF may be more prone to nonembolic stroke, for which warfarin may not protect against.15
We recently reported on racial/ethnic differences in the risk of intracranial hemorrhage among patients with AF.16 The aim of the present study was to assess the magnitude of stroke risk reduction while using or not using warfarin in whites and nonwhites, and to assess whether ischemic stroke risk in a multiethnic cohort of hospitalized AF patients differs by race/ethnicity.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Kaiser Permanente Southern California (KPSC) is a health maintenance organization that currently serves 3.1 million members,
14% of the region’s population. Membership demographics, socioeconomic, and racial/ethnic composition are representative of California. All hospitalized patients are administratively assigned to 1 of the following racial/ethnic categories: white, black, Hispanic, Asian, or other. The study protocol was approved by our institutional review board.
Cohort Identification
Details of the study methods have been reported.16 By searching the first 3 discharge diagnosis codes, all hospitalizations for AF or atrial flutter between January 1, 1995 and December 31, 2000 were identified from among 2.2 million adult members. Patients with AF and atrial flutter were identified by International Classification of Disease, 9th Revision, Clinical Modification (ICD-9) codes 427.31 and 427.32, respectively. Review of the medical records including electrocardiograms of 100 randomly sampled patients showed a 96% positive predictive value of ICD-9 codes in diagnosing AF in this cohort. Both permanent and intermittent AF were included, because the risk of stroke does not differ.17 Atrial flutter was included because the risk for stroke even among "lone atrial flutter" parallels that of AF.18 The qualifying event was AF or atrial flutter in 6% of the cohort and AF only in the remainder. For this study, only whites, blacks, Hispanics, and Asians were included. The 2.8% of patients identified as belonging to the "other" group were excluded.
By referencing ICD-9 and Current Procedural Terminology codes, patients with a history of stroke or TIA, rheumatic heart disease, mitral balloon valvuloplasty, mitral valve replacement predating stroke/TIA, brain tumor, heart transplantation, and congenital heart disease were excluded. Because previous stroke is a potent predictor of recurrent stroke, for this study we searched only for patients without a history of stroke/TIA based on the admission and discharge diagnoses at the time of AF hospitalization. Patients with prosthetic mitral valves were excluded because of their particularly high risk of thromboembolic events and higher target international normalized ratio (INR) levels.19
Primary Outcome Measures
Follow-up of each patient began on the date of AF diagnosis and ended with an outcome event (fatal or nonfatal ischemic stroke) or a censoring event (intracranial hemorrhage, nonstroke death, mitral valve replacement, disenrollment from health plan, and the end of study on December 31, 2003). Patients who underwent mitral valve replacement were censored because they have a different stroke risk profile.20 Patients who underwent carotid endarterectomy within the 2 weeks before or after stroke were also censored to exclude patients who likely had carotid disease-related stroke.
Ischemic and hemorrhagic strokes were identified by ICD-9 codes 433 to 436 and 430 to 432, respectively. Data sources included KPSC hospital records, non-KPSC claims, and a computerized matching of Kaiser Membership data with California vital statistics death tapes. Only the primary diagnosis from hospital files and the underlying cause of death from mortality files were used. The medical records, including all neuroimaging reports and neurology consultation reports, of all stroke event hospitalizations were manually reviewed and qualifying events were adjudicated.
Covariate Assessment
Stroke risk factors were identified using ICD-9 codes, including heart failure, hypertension, and diabetes mellitus,21 in addition to age 75 years or older. Household income, 
12 years education,22 and statin use also affect stroke risk and were entered into the models.23 Patients who filled 
2 statin prescriptions in each of the follow-up years were considered statin users. Census block group level household income and education attainment were obtained through geocoding, a technique that links members’ address data to census geographic areas (block, block group, and tract). More than 95% of our members were mapped to census blocks, the smallest census geographic unit.
Assessment of Warfarin Use During Follow-Up
Approximately 92% of KPSC members obtain their drugs through a KPSC-owned pharmacy. Patients are typically dispensed a 100-day supply of medication, which may last longer by pill splitting. All members have laboratory benefits and typically have their prothrombin time (reported as INR) tested at a KPSC-owned laboratory with a target range of 2 to 3 for uncomplicated AF.
We searched our pharmacy and laboratory databases to identify patients’ warfarin prescriptions. Patients with a filled warfarin prescription were considered to be continuously using warfarin for the 100-day period thereafter. For between-prescription gaps of 100 days, patients were considered to be continuously using warfarin if they had an INR of 2 in the intervening days. This was to account for pill splitting and prescriptions filled at non-KPSC pharmacies for which we have no record of warfarin dispensation. A therapeutic INR value was considered to be valid for 60 days or until the warfarin supply was expected to have exhausted, whichever occurred first. Using this method, which was modified from the method of Go et al,24 an individual patient may contribute to the follow-up times of both warfarin-treated and nonwarfarin-treated groups. These results were used to calculate each patient’s percent-time on warfarin during follow-up. We validated this approach by manual review of the hospital admission records of the qualifying event of all patients (1400 records) who experienced stroke. Comparison of patients’ warfarin use status at the time of stroke defined by our computerized algorithm and that of our chart review showed a 91% consistency rate (
=0.79).
Assessment of Achieved Anticoagulation Intensity
Warfarin efficacy is highly dependent on achieved anticoagulation intensity. For all patients given at least 1 warfarin prescription, we recorded all INR values that were performed. To exclude patients who were prescribed but may not have used warfarin, we included for analysis only patients who had 3 INR values within the 100 days after a warfarin prescription. Because INR tests are frequently ordered irrespective of whether patients with a history of warfarin use was in fact using warfarin at the time the INR was performed, we limited our anticoagulation intensity analysis to INR tests that were performed when our pharmacy records indicate that the patients was using warfarin.
Statistical Analysis
Continuous variables are reported as mean±SD or median and interquartile range as appropriate. Continuous and categorical variables were compared among racial/ethnic groups with the Kruskal-Wallis or 
2 tests. Overall and risk factor-stratified crude event rates were calculated using Poisson log-linear regression with a generalized estimating equations approach and are reported as per 100 person-years of follow-up. Cox proportional hazard models were used to calculate the effect of race/ethnicity on the risk of stroke after adjusting for age, gender, hypertension, diabetes, heart failure, statin use, and percent-time on warfarin. The proportional hazards assumption was validated by fitting a smoothed line in a plot of the Schoenfeld residuals. Rate ratios and hazard ratios are reported with 95% CI. Adjusted survival curves were constructed from the Cox model by setting the covariates at their means. All analyses were conducted using SAS software (version 9.1 for Windows; SAS Institute).
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Cohort Characteristics
We identified 22196 subjects with first-time AF hospitalizations, of which 18867 qualified for analysis (21.5% nonwhite) (Table 1). The median age was 74 years, but each of the nonwhite groups was 5 to 6 years younger than whites. Half of the white patients were older than 75 years, whereas only approximately one-third of nonwhites were older than 75 years. Hypertension and heart failure were more prevalent among blacks. Whites were least likely to have diabetes. Blacks were most likely and Asians were least likely to have
3 stroke risk factors.
|
Percent-Time Using Warfarin and Anticoagulation Intensity
Median follow-up was 3.3 years. As shown in Table 2, 40% to 45% of all patients were not prescribed warfarin at any time, whereas 25% of patients used warfarin >80% of follow-up. There were no significant between-group differences in the percent-time on warfarin. Of the total cohort, 8992 patients were included in the anticoagulation intensity analysis. The racial/ethic distributions were not different between those included and not included in the anticoagulation intensity analysis. The median monthly number of INR values for whites, blacks, Hispanics, and Asians were 2.57, 2.52, 2.67, and 2.51, respectively (P=0.05). The median proportion of follow-up time that INR was 2 for the full cohort was 70%. However, blacks spent more time (36.8%) in the subtherapeutic range (INR <2) than the other groups (range, 30.3% to 31.8%). Additionally, when considering percent-time spent in the therapeutic range (INR 2 to 3), blacks were in the therapeutic range slightly less than half the time, whereas the other races were in the therapeutic range more than half the time. Hispanics and Asians were in the therapeutic range for a similar percentage of follow-up time as whites.
|
Person-Years Follow-Up and Stroke Events
As shown in Table 3, there were 1226 qualifying ischemic strokes over a total of 63204 person-years of follow-up. The follow-up times for each group are as follows: white, 49324; black, 5004; Hispanics, 6206; and Asians, 2669 person-years. There were 1.66 and 2.11 strokes per 100 person-years using warfarin and not using warfarin, respectively. Blacks experienced higher rates of stroke than the other groups, whether or not they were using warfarin.
|
Stroke Reduction With Warfarin
Overall, the crude ischemic stroke rate ratio was reduced by 22% with warfarin treatment (Table 3). White patients had a 21% lower stroke rate with warfarin, but none of the nonwhite groups experienced a statistically significant reduction in stroke. There was no evidence of benefit among blacks (P=0.60); there was a trend toward benefit among Hispanics (P=0.089); there was a nonsignificant reduction among Asians (P=0.18).
Patients younger than 75 years had less than half the stroke rate of patients 75 years or older. Analysis of patients’ stroke risk factors showed that the number of comorbidities correlated with stroke risk. Only patients with 2 risk factors benefited from warfarin. Patients without hypertension or heart failure did not benefit from warfarin. Warfarin reduced the risk of stroke whether patients were using statins or had diabetes.
Adjusted Hazard Ratio for Ischemic Stroke
Cox proportional hazard models adjusting for age, gender, stroke risk factors, socioeconomic status, statin, and percent-time using warfarin showed that blacks and Hispanics had a 59% and 24% higher risk of stroke, respectively, compared to whites (Table 4). The risk of stroke for Asians was 23% higher but did not reach statistical significance. Women had a significant 33% higher risk of stroke than men. Patients with hypertension, heart failure, and diabetes had 93%, 31%, and 33%, greater risks of stroke respectively, compared to those without those comorbidities (P<0.001 for all). Statin use conferred a 32% lower risk of stroke. Household income and educational level did not influence stroke risk. The Figure shows the adjusted ischemic stroke event-free curves based on Cox models for the 4 groups.
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Racial Differences in Ischemic Stroke
The major adverse consequence of AF is stroke. Anticoagulation prevents
26 strokes per 1000 patients treated over the course of 1 year.4 However, not all strokes in a patient with AF are cardioembolic. In an analysis of ischemic stroke events in the Stroke Prevention in AF I, II, and III studies, 32% of classifiable strokes (24% of all strokes) were deemed to be noncardioembolic in nature.15 AF patients at high risk for embolic stroke have comorbidities that also predispose them to nonembolic strokes, including lacunar strokes, small vessel occlusion, and large-artery atherosclerosis. Whereas nonwhites are at greater risk for several subtypes of ischemic stroke, cardioembolic stroke attributable to AF is more frequently implicated in whites.5,10,25 Warfarin is ineffective and may be harmful for the prevention of nonembolic stroke.26 Because stroke in nonwhites are generally less likely to be embolic in whites, it is reasonable to ask, to what extent do nonwhites with AF benefit from warfarin? In a study of 16000 Medicare beneficiaries, Birman-Deych et al27 reported that blacks and Hispanics did not benefit from warfarin anticoagulationin, partly because of less adequate warfarin treatment and anticoagulation monitoring. However, the poorer efficacy of warfarin in nonwhites remained after matching blacks and whites on stroke risk factors and monitoring frequency.
Comparison of Stroke Risk Profiles
Stroke incidence increases with age. In our cohort, whites were 6 to 7 years older than nonwhites. Similar to other studies, hypertension and diabetes were more prevalent among nonwhites.5,28 Although younger, the greater prevalence of comorbidities among nonwhites partially explains the increased stroke rates. Because diabetes and hypertension is associated with lacunar and atherosclerotic arterial stroke, the apparent lesser efficacy of stroke reduction with warfarin in nonwhites is consistent with the hypothesis that this group may have had more nonembolic strokes even in the setting of AF. Socioeconomic status, as measured by educational attainment and income, was generally higher among whites and Asians than among blacks and Hispanics. Statin use was most prevalent among Asians. These factors likely contributed to the observed difference in stroke rate.
Factors Affecting Stroke Risk
After multivariate adjustment, the stroke risk of blacks remained 59% higher than that for whites. The reason for this is unclear, but there may be other factors that lead to greater stroke risk that were not adjusted for in our models, such as poorer control of hypertension. In this cohort, women had a 33% higher risk of stroke than men, which is consistent with previous studies.29 Education and income were estimated using geocoding and were not found to be associated with stroke risk in this cohort. However, as members of a prepaid health plan, 92% of patients had medication coverage and all had access to laboratory and other healthcare resources (eg, anticoagulation, primary and specialist clinics). This may have partially negated the impact of any existing socioeconomic inequality on stroke incidence.
Efficacy of Warfarin and Anticoagulation Intensity
The efficacy of warfarin in preventing stroke is dependent on the adequacy of anticoagulation achieved. In a recent systematic review of 67 anticoagulation studies (69% of which were for AF) by van Walraven et al,30 patients were in the therapeutic range 63.6% of time overall. Among the 30 studies in a community setting, the percent time in range was 56.7%, similar to the present study (54.5%). All patients in our cohort have access to INR monitoring. In fact, the percent-time follow-up using warfarin and the number of INR tests per month were similar between groups. Whites, Hispanic, and Asians had similar achieved anticoagulation intensity (Table 2). Blacks spent more time in the subtherapeutic range (INR <2), which likely contributed to the observed absence of stroke reduction. It appears unlikely, however, that this relatively small difference (INR <2 was 36.8% in blacks compared to 31% in the other groups) in warfarin treatment adequacy fully explains the absence of stroke reduction with warfarin.
In our cohort, the incidence of stroke among nonblacks was 1.5 per 100 person-years using warfarin and 2.1 per 100 person-years while not using warfarin, which is slightly higher than that reported by Go et al.24 In that study, the percent-time on warfarin that INR was 2 to 3 was 62.5%, somewhat higher than the 54.5% found in our cohort. Furthermore, our cohort consisted of patients with a history of AF-related hospitalization and had higher prevalences of heart failure, hypertension, and diabetes. These differences likely contributed to the lower stroke reduction seen in our study (21%) compared to that of Go et al24 (50%). Birman-Deych et al27 reported a 39% reduction in stroke risk with warfarin that was limited to the white patients in their cohort. Blacks and Hispanics did not derive any benefit from warfarin. Both these studies included patients with previous stroke or TIA who are at high risk for recurrent events, however, whereas the present study excluded the majority of such patients.
Recent studies have shown that there are racial differences in the prevalence of certain CYP2C9 and VKORC1 polymorphisms that affect warfarin dose requirement.31–34 Such polymorphisms have not been shown to affect warfarin efficacy. However, whether the differences in warfarin pharmacokinetics and pharmacodynamics that results from these polymorphisms affect the stability of INR levels and, hence, the risk of stroke is unknown.
Clinical Implications
For AF patients with risk factors for stroke, anticoagulation with dose-adjusted warfarin is the standard of care. Nonwhites constituted 10% of those in the clinical trials on which such decisions are made,1 and may have a different predisposition to stroke as compared to whites. Anticoagulation is very effective for the prevention of cardioembolic stroke, but much less so for nonembolic stroke. Nonwhites in our cohort, particularly Hispanics and blacks, have more comorbidity and may have been at greater risk for nonembolic stroke. This may partially explain the lack of observed benefit in nonwhites. One must be cautious when interpreting our results, however. Blacks experienced the least reduction in crude stroke rate, but they also had the least adequate anticoagulation. Among Hispanics and Asians, the rate ratios were 0.71 and 0.65, respectively. Statistical significance was not achieved in either case, but the relatively small numbers of events inevitably lead to imprecise estimates. A benefit of warfarin in these groups has not been ruled out.
Limitations
This study has several limitations, including the inherent shortcomings attributable to the retrospective nature of our study. We do not have sufficient data to ascertain whether there are between-group differences in ischemic stroke subtype (embolic vs nonembolic). Because aspirin is over-the-counter, we do not have reliable records on its use. However, it is standard practice to avoid aspirin in all anticoagulated patients, except in those with a mechanical prosthetic valve. Similarly, we do not have reliable and accessible data on other risk factors such as smoking, adequacy of blood pressure control, alcohol use, left atrial size, and other echocardiographic measurements that may affect stroke risk.
When assembling the cohort, we excluded patients who had ICD-9 codes of previous stroke during the qualifying AF hospitalization. Review of 200 random records showed that 10% of patients did in fact have documentation of previous ischemic stroke. Because warfarin is effective in secondary stroke prevention as well, we do not believe this materially affects the interpretation of our findings. ICD-9 codes were used to identify AF and comorbidities in this study. Chart review showed that ICD-9 codes had a 96% positive predictive value for AF in our cohort, but its sensitivity is unknown. A recent study35 found ICD-9 coding for AF and other stroke risk factors to be >80% sensitive and specific. Our cohort consists of patients with previous AF hospitalization and may not be representative of all AF patients.
Warfarin prescription information may be incomplete for the 8% of KPSC members who did not have drug benefits. Importantly, despite analyzing all pertinent INRs, we cannot rule out the possibility that INR stability (day-to-day variation) differed between groups, which may in turn influence stroke risk.
Finally, racial/ethnic classification in our system is administratively assigned rather than self-reported. Our administrative database did not differentiate between East and South Asians, but regional demographics indicate that East Asians predominate.
Conclusion
In this retrospective study of patients hospitalized with AF, only whites were observed to have a statistically significant lower rate of stroke while using warfarin. There were lower rates of stroke in Hispanics and Asians that did not reach statistical significance. Because of the relatively small numbers of events among nonwhites, a protective benefit of warfarin in nonwhites, although not confirmed, cannot be ruled out by available data. Further studies are needed to explore this important question.
|
Acknowledgments |
|---|
The authors thank Saman Fakheri, M.D. (Kaiser Permanente Medical Center, Los Angeles), for his assistance in data collection.
Disclosures
Dr Fakheri was not compensated for his assistance and reports no conflicts of interest.
Received October 26, 2007; revision received February 1, 2008; accepted February 25, 2008.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
1. Feinberg WM, Blackshear JL, Laupacis A, Kronmal R, Hart RG. Prevalence, age distribution, and gender of patients with atrial fibrillation. Analysis and implications. Arch Intern Med. 1995; 155: 469–473.
2. Go AS, Hylek EM, Phillips KA, Chang Y, Henault LE, Selby JV, Singer DE. Prevalence of diagnosed atrial fibrillation in adults: national implications for rhythm management and stroke prevention: the AnTicoagulation and Risk Factors in Atrial Fibrillation (ATRIA) Study. JAMA. 2001; 285: 2370–2375.
3. Wolf PA, Abbott RD, Kannel WB. Atrial fibrillation: a major contributor to stroke in the elderly. The Framingham Study. Arch Intern Med. 1987; 147: 1561–1564.
4. Atrial Fibrillation Investigators. 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.
5. Stansbury JP, Jia H, Williams LS, Vogel WB, Duncan PW. Ethnic disparities in stroke: epidemiology, acute care, and postacute outcomes. Stroke. 2005; 36: 374–386.
6. Sacco RL, Boden-Albala B, Gan R, Chen X, Kargman DE, Shea S, Paik MC, Hauser WA. Stroke incidence among white, black, and Hispanic residents of an urban community: the Northern Manhattan Stroke Study. Am J Epidemiol. 1998; 147: 259–268.
7. Morgenstern LB, Smith MA, Lisabeth LD, Risser JM, Uchino K, Garcia N, Longwell PJ, McFarling DA, Akuwumi O, Al-Wabil A, Al-Senani F, Brown DL, Moye LA. Excess stroke in Mexican Americans compared with non-Hispanic Whites: the Brain Attack Surveillance in Corpus Christi Project. Am J Epidemiol. 2004; 160: 376–383.
8. Rodriguez BL, D'Agostino R, Abbott RD, Kagan A, Burchfiel CM, Yano K, Ross GW, Silbershatz H, Higgins MW, Popper J, Wolf PA, Curb JD. Risk of hospitalized stroke in men enrolled in the Honolulu Heart Program and the Framingham Study: A comparison of incidence and risk factor effects. Stroke. 2002; 33: 230–236.
9. Ayala C, Greenlund KJ, Croft JB, Keenan NL, Donehoo RS, Giles WH, Kittner SJ, Marks JS. Racial/ethnic disparities in mortality by stroke subtype in the United States, 1995–1998. Am J Epidemiol. 2001; 154: 1057–1063.
10. Hajat C, Tilling K, Stewart JA, Lemic-Stojcevic N, Wolfe CD. Ethnic differences in risk factors for ischemic stroke: a European case-control study. Stroke. 2004; 35: 1562–1567.
11. Smith MA, Risser JM, Lisabeth LD, Moye LA, Morgenstern LB. Access to care, acculturation, and risk factors for stroke in Mexican Americans: the Brain Attack Surveillance in Corpus Christi (BASIC) project. Stroke. 2003; 34: 2671–2675.
12. White H, Boden-Albala B, Wang C, Elkind MS, Rundek T, Wright CB, Sacco RL. Ischemic stroke subtype incidence among whites, blacks, and Hispanics: the Northern Manhattan Study. Circulation. 2005; 111: 1327–1331.
13. Sacco RL, Boden-Albala B, Abel G, Lin I-F, Elkind M, Hauser WA, Paik MC, Shea S. Race-ethnic disparities in the impact of stroke risk factors: the Northern Manhattan Stroke Study. Stroke. 2001; 32: 1725–1731.
14. Tuhrim S, Godbold JH, Goldman ME, Horowitz DR, Weinberger J. The Minorities Risk Factors and Stroke Study (MRFASS). Design, methods and baseline characteristics. Neuroepidemiology. 1997; 16: 224–233.[CrossRef][Medline] [Order article via Infotrieve]
15. Hart RG, Pearce LA, Miller VT, Anderson DC, Rothrock JF, Albers GW, Nasco E. 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.[CrossRef][Medline] [Order article via Infotrieve]
16. Shen AY-J, Yao JF, Brar SS, Jorgensen MB, Chen W. Racial/ethnic differences in the risk of intracranial hemorrhage among patients with atrial fibrillation. J Am Coll Cardiol. 2007; 50: 309–315.
17. Hart RG, Pearce LA, Rothbart RM, McAnulty JH, Asinger RW, Halperin JL. Stroke with intermittent atrial fibrillation: incidence and predictors during aspirin therapy. Stroke Prevention in Atrial Fibrillation Investigators. J Am Coll Cardiol. 2000; 35: 183–187.
18. Halligan SC, Gersh BJ, Brown RD, Jr., Rosales AG, Munger TM, Shen W-K, Hammill SC, Friedman PA. The natural history of lone atrial flutter. Ann Intern Med. 2004; 140: 265–268.
19. Salem DN, Stein PD, Al-Ahmad A, Bussey HI, Horstkotte D, Miller N, Pauker SG. Antithrombotic therapy in valvular heart disease-native and prosthetic: the Seventh ACCP Conference on Antithrombotic and Thrombolytic Therapy. Chest. 2004; 126: 457S–482S.[CrossRef][Medline] [Order article via Infotrieve]
20. Cannegieter SC, Rosendaal FR, Wintzen AR, van der Meer FJ, Vandenbroucke JP, Briet E. Optimal oral anticoagulant therapy in patients with mechanical heart valves. N Engl J Med. 1995; 333: 11–17.
21. Gage BF, Waterman AD, Shannon W, Boechler M, Rich MW, Radford MJ. Validation of clinical classification schemes for predicting stroke: results from the National Registry of Atrial Fibrillation. JAMA. 2001; 285: 2864–2870.
22. Qureshi AI, Suri MF, Saad M, Hopkins LN. Educational attainment and risk of stroke and myocardial infarction. Med Sci Monit. 2003; 9: CR466–CR473.[Medline] [Order article via Infotrieve]
23. Collins R, Armitage J, Parish S, Sleight P, Peto R. Effects of cholesterol-lowering with simvastatin on stroke and other major vascular events in 20536 people with cerebrovascular disease or other high-risk conditions. Lancet. 2004; 363: 757–767.[CrossRef][Medline] [Order article via Infotrieve]
24. Go AS, Hylek EM, Chang Y, Phillips KA, Henault LE, Capra AM, Jensvold NG, Selby JV, Singer DE. Anticoagulation therapy for stroke prevention in atrial fibrillation: how well do randomized trials translate into clinical practice? JAMA. 2003; 290: 2685–2692.
25. Uchino K, Risser JM, Smith MA, Moye LA, Morgenstern LB. Ischemic stroke subtypes among Mexican Americans and non-Hispanic whites: the BASIC Project. Neurology. 2004; 63: 574–576.
26. Sandercock P, Mielke O, Liu M, Counsell C. Anticoagulants for preventing recurrence following presumed non-cardioembolic ischaemic stroke or transient ischaemic attack. Cochrane Database Syst Rev. 2003: CD000248.
27. Birman-Deych E, Radford MJ, Nilasena DS, Gage BF. Use and effectiveness of warfarin in Medicare beneficiaries with atrial fibrillation. Stroke. 2006; 37: 1070–1074.
28. Frey JL, Jahnke HK, Bulfinch EW. Differences in stroke between white, Hispanic, and Native Am patients: the Barrow Neurological Institute stroke database. Stroke. 1998; 29: 29–33.
29. Wolf PA, Mitchell JB, Baker CS, Kannel WB, D'Agostino RB. Impact of atrial fibrillation on mortality, stroke, and medical costs. Arch Intern Med. 1998; 158: 229–234.
30. van Walraven C, Jennings A, Oake N, Fergusson D, Forster AJ. Effect of study setting on anticoagulation control: a systematic review and metaregression. Chest. 2006; 129: 1155–1166.[CrossRef][Medline] [Order article via Infotrieve]
31. Takahashi H, Wilkinson GR, Nutescu EA, Morita T, Ritchie MD, Scordo MG, Pengo V, Barban M, Padrini R, Ieiri I, Otsubo K, Kashima T, Kimura S, Kijima S, Echizen H. Different contributions of polymorphisms in VKORC1 and CYP2C9 to intra- and inter-population differences in maintenance dose of warfarin in Japanese, Caucasians and African-Americans. Pharmacogenet Genomics. 2006; 16: 101–110.[Medline] [Order article via Infotrieve]
32. Sconce EA, Khan TI, Wynne HA, Avery P, Monkhouse L, King BP, Wood P, Kesteven P, Daly AK, Kamali F. The impact of CYP2C9 and VKORC1 genetic polymorphism and patient characteristics upon warfarin dose requirements: proposal for a new dosing regimen. Blood. 2005; 106: 2329–2333.
33. Rieder MJ, Reiner AP, Gage BF, Nickerson DA, Eby CS, McLeod HL, Blough DK, Thummel KE, Veenstra DL, Rettie AE. Effect of VKORC1 haplotypes on transcriptional regulation and warfarin dose. N Engl J Med. 2005; 352: 2285–2293.
34. Aithal GP, Day CP, Kesteven PJ, Daly AK. Association of polymorphisms in the cytochrome P450 CYP2C9 with warfarin dose requirement and risk of bleeding complications. Lancet. 1999; 353: 717–719.[CrossRef][Medline] [Order article via Infotrieve]
35. Kokotailo RA, Hill MD. Coding of stroke and stroke risk factors using international classification of diseases, revisions 9 and 10. Stroke. 2005; 36: 1776–1781.
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||