Diabetes and ischaemic stroke outcome.

AIM: To compare all-cause mortality, stroke recurrence and functional outcomes in people who have experienced stroke, with and without diabetes. METHODS: We captured data on population-based ischaemic strokes (2006-2012) in Nueces County, Texas. Data were collected from participant interviews and medical records. Differences in cumulative mortality and stroke recurrence risk by diabetes status were estimated at 30 days and 1 year using Cox models. Differences in 90-day functional outcomes (activities of daily living/instrumental activities of daily living score: range 1-4; higher scores worse) by diabetes status were assessed using Tobit regression. Effect modification by ethnicity was examined. RESULTS: There were 1301 ischaemic strokes, 46% with history of known diabetes. The median (interquartile range) age was 70 (58-81) years and 61% were Mexican American. People with diabetes were younger and more likely to be Mexican American compared with those without diabetes. After adjustment, diabetes predicted mortality (30-day hazard ratio 1.44, 95% CI 0.97-2.12; 1-year hazard ratio 1.47, 95% CI 1.09-1.97) but not stroke recurrence (1-year hazard ratio 1.27, 95% CI 0.78-2.07). People with diabetes had a worse functional outcome score that was explained by cardiovascular risk factors and pre-stroke factors. Diabetes was not associated with functional outcome in the fully adjusted model (final adjusted activities of daily living/instrumental activities of daily living score difference 0.11, 95% CI -0.07 to 0.30). Effect modification by ethnicity was not significant (P>0.3 for all models). CONCLUSIONS: Diabetes was associated with higher mortality and worse functional outcome but not stroke recurrence. Interventions are needed to decrease the adverse outcomes associated with diabetes, particularly in Mexican-American people.

Higher mortality in patients with right hemispheric intracerebral haemorrhage: INTERACT1 and 2.

BACKGROUND AND PURPOSE: Controversy exists over the prognostic significance of the affected hemisphere in stroke. We aimed to determine the relationship between laterality of acute intracerebral haemorrhage (ICH) and poor clinical outcomes. METHODS: A subsidiary analysis of the INTERACT Pilot and INTERACT2 studies--randomised controlled trials of patients with spontaneous acute ICH with elevated systolic blood pressure (BP), randomly assigned to intensive (target systolic BP <140 mm Hg) or guideline-based (<180 mm Hg) BP management. Outcomes were the combined and separate end points of death and major disability (modified Rankin scale (mRS) scores of 3-6, 6 and 3-5, respectively) at 90 days. RESULTS: A total of 2708 patients had supratentorial/hemispheric ICH and information on mRS at 90 days. Patients with right hemispheric ICH (1327, 49%) had a higher risk of death at 90 days compared to those with left hemispheric ICH after adjustment for potential confounding variables (OR, 1.77 (95% CI 1.33 to 2.37)). There were no differences between patients with right and left hemispheric ICH regarding the combined end point of death or major disability or major disability in the multivariable-adjusted models (1.07 (0.89 to 1.29) and 0.85 (0.72 to 1.01), respectively). CONCLUSIONS: Right hemispheric lesion was associated with increased risk of death in patients with acute ICH. The laterality of the ICH does not appear to affect the level of disability in survivors. TRIAL REGISTRATION NUMBER: URL: http://www.clinicaltrials.gov. Unique identifier: NCT00226096 and NCT00716079.

Setting priorities for stroke care and research.

Recent publications describing the sobering global increase in stroke mortality and global life years lost due to stroke despite improvements in developed countries have drawn focus on the severe impact of stroke in the developing world. At the same time, three recent interventional trials that failed to demonstrate an important role for catheter-based therapies in acute stroke have called into question this expensive use of technology. Coupling all of this new data leads to the natural conclusion that a focus on stroke prevention for the developing world, and for the poor in developed countries, should be where we set our priorities for the foreseeable future.

Enrollment of women and minorities in NINDS trials.

OBJECTIVE: To determine policy-associated changes over time in 1) the enrollment of women and minorities in National Institute of Neurological Disorders and Stroke (NINDS)-funded clinical trials and 2) the trial publication reporting of race/ethnicity and gender. METHODS: All NINDS-funded phase III trials published between 1985 and 2008 were identified. Percent of African Americans, Hispanic Americans, and women enrolled in the trials was calculated for those trials with available data. Z tests were used to compare reporting and enrollment data from before (period 1) and after (period 2) 1995 when NIH enacted their policies regarding race, ethnicity, and gender. Percent of main trial publications reporting enrollment of African Americans, Hispanic Americans, and women was also calculated. RESULTS: Of the 56 trials identified, 100%, 48%, and 25% reported enrollment by gender, race, and ethnicity. Women constituted 42.1% of the trial population. Enrollment of women increased over time (36.9% period 1; 49.0% period 2, p < 0.001). African Americans constituted 19.8% of the enrollees in trials with available data and enrollment increased over time (11.6% period 1; 30.7% period 2, p < 0.001). Hispanic Americans constituted 5.8% of subjects in trials with available data and enrollment decreased over time (7.4% period 1; 5.0% period 2, p < 0.001). CONCLUSIONS: Improvements in reporting of race/ethnicity in publications and enrollment of Hispanics in NINDS trials are needed. While African American representation is above population levels, Hispanic Americans are underrepresented in NINDS trials and representation is declining despite Hispanics' increasing representation in the US population.

Do-not-resuscitate orders and predictive models after intracerebral hemorrhage.

OBJECTIVE: To quantify the accuracy of commonly used intracerebral hemorrhage (ICH) predictive models in ICH patients with and without early do-not-resuscitate orders (DNR). METHODS: Spontaneous ICH cases (n = 487) from the Brain Attack Surveillance in Corpus Christi study (2000-2003) and the University of California, San Francisco (June 2001-May 2004) were included. Three models (the ICH Score, the Cincinnati model, and the ICH grading scale [ICH-GS]) were compared to observed 30-day mortality with a chi(2) goodness-of-fit test first overall and then stratified by early DNR orders. RESULTS: Median age was 71 years, 49% were female, median Glasgow Coma Scale score was 12, median ICH volume was 13 cm(3), and 35% had early DNR orders. Overall observed 30-day mortality was 42.7% (95% confidence interval [CI] 38.3-47.1), with the average model-predicted 30-day mortality for the ICH Score, Cincinnati model, and ICH-GS at 39.9% (p = 0.005), 40.4% (p = 0.007), and 53.9% (p < 0.001). However, for patients with early DNR orders, the observed 30-day mortality was 83.5% (95% CI 78.0-89.1), with the models predicting mortality of 64.8% (p < 0.001), 57.2% (p < 0.001), and 77.8% (p = 0.02). For patients without early DNR orders, the observed 30-day mortality was 20.8% (95% CI 16.5-25.7), with the models predicting mortality of 26.6% (p = 0.05), 31.4% (p < 0.001), and 41.1% (p < 0.001). CONCLUSIONS: ICH prognostic model performance is substantially impacted when stratifying by early DNR status, possibly giving a false sense of model accuracy when DNR status is not considered. Clinicians should be cautious when applying these predictive models to individual patients.

Significance of perihematomal edema in acute intracerebral hemorrhage: the INTERACT trial.

BACKGROUND: Uncertainty surrounds the effects of cerebral edema on outcomes in intracerebral hemorrhage (ICH). METHODS: We used data from the INTERACT trial to determine the predictors and prognostic significance of "perihematomal" edema over 72 hours after ICH. INTERACT included 404 patients with CT-confirmed ICH and elevated systolic blood pressure (BP) (150-220 mm Hg) who had the capacity to commence BP lowering treatment within 6 hours of ICH. Baseline and repeat CT (24 and 72 hours) were performed using standardized techniques, with digital images analyzed centrally. Predictors of growth in edema were determined using generalized estimating equations, and its effects on clinical outcomes were estimated using a logistic regression model. RESULTS: Overall, 270 patients had 3 sequential CT scans available for analyses. At baseline, there was a highly significant correlation between hematoma and perihematomal edema volumes (r(2) = 0.45). Lower systolic BP and baseline hematoma volume were independently associated with absolute increase in perihematomal edema volume. History of hypertension, baseline hematoma volume, and earlier time from onset to CT were independently associated with relative increase in edema volume. Both absolute and relative increases in perihematomal edema growth were significantly associated with death or dependency at 90 days after adjustment for age, gender, and randomized treatment, but not when additionally adjusted for baseline hematoma volume. CONCLUSIONS: The degree of, and growth in, perihematomal edema are strongly related to the size of the underlying hematoma of acute intracerebral hemorrhage, and do not appear to have a major independent effect in determining the outcome from this condition.

Predicting ischaemic stroke subtype from presenting systolic blood pressure: the BASIC Project.

OBJECTIVE: We hypothesized that low presenting systolic blood pressure (SBP) predicted cardioembolic stroke aetiology. DESIGN: Active and passive surveillance were used to identify all ischaemic strokes as part of the Brain Attack Surveillance in Corpus Christi (BASIC) population-based study. Multinomial logistic regression was used to examine the association between stroke subtype and first documented SBP in the medical record. SETTING: Nueces County, TX, USA (313,645 residents in 2000). The community is urban with the majority of the population residing in the city of Corpus Christi. The area is served by seven adult acute care hospitals. PATIENTS: Three hundred and eight cases with completed ischaemic stroke and determined subtype aetiology between January 2000 and December 2002. RESULTS: Lower presenting SBP was associated with stroke subtype (P = 0.001). This association remained significant in the final model adjusted for age and history of coronary artery disease. The odds of cardioembolic versus small vessel occlusion increased by 20% (OR = 1.20, 95% CI: 1.07-1.35) for every 10 mmHg decrease in presenting SBP. Other covariates including race/ethnicity, gender, history of hypertension, and diabetes were neither significant predictors of stroke subtype, nor did they confound the association of SBP and stroke subtype. A 5 year increase in age increased the odds of cardioembolic subtype by 25% (OR = 1.25, 95% CI: 1.07-1.47). CONCLUSIONS: Lower initial SBP and older age at ischaemic stroke presentation were associated with cardioembolic stroke. Suspicion of cardioembolic stroke should be increased in those presenting with low SBP.

Gender and ethnic differences in subarachnoid hemorrhage.

BACKGROUND: Mexican Americans (MAs) comprise the largest component of the largest minority group within the United States. The purpose of this study was to examine ethnic and gender differences in the epidemiology, presentation, and outcomes after subarachnoid hemorrhage (SAH) in a representative United States community. Targeted public health interventions are dependent on accurate assessments of groups at highest disease risk. METHODS: All patients with nontraumatic SAH older than 44 years were prospectively identified from January 1, 2000, to December 31, 2006, as part of the Brain Attack Surveillance In Corpus Christi project, an urban population-based study in southeast Texas. Risk ratios for cumulative SAH incidence comparing MAs with non Hispanic whites (NHWs) and women with men were calculated. Descriptive statistics for other clinical and demographic variables were computed overall, by gender, and by ethnicity. RESULTS: A total of 107 patients had a SAH during the time period (7-year cumulative incidence: 11/10,000); of these, 43 were NHW (40% of cases vs 53% of the population) and 64 were MA (60% of cases vs 48% of the population). The overall age-adjusted risk ratio for SAH in MAs compared with NHWs was 1.67 (95% CI: 1.13, 2.47), and in women compared to men was 1.74 (95% CI 1.16, 2.62). Overall in-hospital mortality was 32.2%. No ethnic difference was observed for discharge disability or in-hospital mortality. CONCLUSIONS: Subarachnoid hemorrhage disproportionately affects Mexican Americans and women. Public health interventions should target these groups to reduce the impact of this severe disease.

Early care limitations independently predict mortality after intracerebral hemorrhage.

OBJECTIVE: Intracerebral hemorrhage (ICH) is associated with a high early mortality rate. We examined the impact of early do not resuscitate (DNR) orders and other limitations in aggressive care on mortality after ICH in a community-based study. METHODS: Cases of spontaneous ICH from 2000 to 2003 were identified from the Brain Attack Surveillance in Corpus Christi (BASIC) project, with deaths ascertained through 2005. Charts were reviewed for early (<24 hours from presentation) DNR orders, withdrawal of care, or deferral of other life sustaining interventions, analyzed together as combined DNR (C-DNR). Multivariable Cox-proportional hazards models were used to examine the association between short- and long-term all-cause mortality and early C-DNR, adjusted for demographics and established predictors of mortality after ICH. RESULTS: Of 18,393 subjects screened for cerebrovascular disease, 270 non-traumatic ICH cases were included. Cumulative mortality risk was 0.43 at 30 days and 0.55 over the study course. Early C-DNR was noted in 34% of cases and was associated with a doubling in the hazard of death both at 30 days (hazard ratio [HR] 2.17, 95% CI 1.38, 3.41) and at end of follow-up (HR 1.92, 95% CI 1.29, 2.87) despite adjustment for age, gender, ethnicity, Glasgow Coma Scale, ICH volume, intraventricular hemorrhage, and infratentorial hemorrhage. CONCLUSIONS: Early care limitations are independently associated with both short- and long-term all-cause mortality after intracerebral hemorrhage (ICH) despite adjustment for expected predictors of ICH mortality. Physicians should carefully consider the effect of early limitations in aggressive care to avoid limiting care for patients who may survive their acute illness.

Neighborhood environment and risk of ischemic stroke: the brain attack surveillance in Corpus Christi (BASIC) Project.

The authors explored whether neighborhood-level characteristics are associated with ischemic stroke and whether the association differs by ethnicity, age, and gender. Using data from the Brain Attack Surveillance in Corpus Christi Project (January 2000-June 2003), they identified cases of ischemic stroke (n = 1,247) from both hospital and out-of-hospital sources. Census tracts served as proxies for neighborhoods, and neighborhood socioeconomic status scores were constructed from census variables (higher scores represented less disadvantage). In Poisson regression analyses comparing the 90th percentile of neighborhood score with the 10th, the relative risk of stroke was 0.49 (95% confidence interval (CI): 0.41, 0.58). After adjustment for age, gender, and ethnicity, this association was attenuated (relative risk (RR) = 0.79, 95% CI: 0.63, 1.00). There was no ethnic difference in the association of score with stroke (p for interaction = 0.79). Significant effect modification was found for age (p for interaction < 0.001) and gender (p for interaction = 0.04), with increasing scores being protective against stroke in men and younger persons. Associations were attenuated after adjustment for education (men: RR = 0.77, 95% CI: 0.55, 1.07; persons aged <65 years: RR = 0.65, 95% CI: 0.41, 1.02). Neighborhood characteristics may influence stroke risk in certain gender and age groups. Mechanisms for these associations should be examined.

Projected costs of ischemic stroke in the United States.

BACKGROUND: There are barriers to acute stroke care in minority groups as well as a higher incidence of ischemic stroke when compared with non-Hispanic whites. OBJECTIVE: To estimate the future economic burden of stroke in non-Hispanic whites, Hispanics, and African Americans in the United States from 2005 to 2050. METHODS: We used U.S. Census estimates of the race-ethnic group populations age 45 years and older. We obtained stroke epidemiology and service utilization data from the Northern Manhattan Stroke Study and the Brain Attack Surveillance in Corpus Christi project and other published data. We estimated costs directly from Medicare reimbursement or from studies that used Medicare reimbursement. Direct and indirect costs considered included ambulance services, initial hospitalization, rehabilitation, nursing home costs, outpatient clinic visits, drugs, informal caregiving, and potential lost earnings. RESULTS: The total cost of stroke from 2005 to 2050, in 2005 dollars, is projected to be 1.52 trillion dollars for non-Hispanic whites, 313 billion dollars for Hispanics, and 379 billion dollars for African Americans. The per capita cost of stroke estimates are highest in African Americans (25,782 dollars), followed by Hispanics (17,201 dollars), and non-Hispanic whites (15,597 dollars). Loss of earnings is expected to be the highest cost contributor in each race-ethnic group. CONCLUSIONS: The economic burden of stroke in African Americans and Hispanics will be enormous over the next several decades. Further efforts to improve stroke prevention and treatment in these high stroke risk groups are necessary.

A scoring system for ascertainment of incident stroke; the Risk Index Score (RISc).

OBJECTIVES: The main objective of this study was to develop and validate a computer-based statistical algorithm that could be translated into a simple scoring system in order to ascertain incident stroke cases using hospital admission medical records data. METHODS: The Risk Index Score (RISc) algorithm was developed using data collected prospectively by the Brain Attack Surveillance in Corpus Christi (BASIC) project, 2000. The validity of RISc was evaluated by estimating the concordance of scoring system stroke ascertainment to stroke ascertainment by physician and/or abstractor review of hospital admission records. RESULTS: RISc was developed on 1718 randomly selected patients (training set) and then statistically validated on an independent sample of 858 patients (validation set). A multivariable logistic model was used to develop RISc and subsequently evaluated by goodness-of-fit and receiver operating characteristic (ROC) analyses. The higher the value of RISc, the higher the patient's risk of potential stroke. The study showed RISc was well calibrated and discriminated those who had potential stroke from those that did not on initial screening. CONCLUSION: In this study we developed and validated a rapid, easy, efficient, and accurate method to ascertain incident stroke cases from routine hospital admission records for epidemiologic investigations. Validation of this scoring system was achieved statistically; however, clinical validation in a community hospital setting is warranted.

Presentation of intracerebral haemorrhage in a community.

BACKGROUND: Studies on intracerebral haemorrhage (ICH) from tertiary care centres may not be an accurate representation of the true spectrum of disease presentation. OBJECTIVE: To describe the clinical and imaging presentation of ICH in a community devoid of the referral bias of an academic medical centre; and to investigate factors associated with lower Glasgow coma scale (GCS) score at presentation, as GCS is crucial to early clinical decision making. METHODS: The study formed part of the BASIC project (Brain Attack Surveillance in Corpus Christi), a population based stroke surveillance study in a bi-ethnic Texas community. Cases of first non-traumatic ICH were identified from years 2000 to 2003, using active and passive surveillance. Clinical data were collected from medical records by trained abstractors, and all computed tomography (CT) scans were reviewed by a study physician. Multivariable linear regression was used to identify clinical and CT predictors of a lower GCS score. RESULTS: 260 cases of non-traumatic ICH were identified. Median ICH volume was 11 ml (interquartile range 3 to 36) with hydrocephalus noted in 45%. Median initial GCS score was 12.5 (7 to 15). Hydrocephalus score (p = 0.0014), ambient cistern effacement (p = 0.0002), ICH volume (p = 0.014), and female sex (p = 0.024) were independently associated with lower GCS score at presentation, adjusting for other variables. CONCLUSIONS: ICH has a wide range of severity at presentation. Hydrocephalus is a potentially reversible cause of a lower GCS score. Since early withdrawal of care decisions are often based on initial GCS, recognition of the important influence of hydrocephalus on GCS is warranted before withdrawal of care decisions are made.

Differences in intracerebral hemorrhage between Mexican Americans and non-Hispanic whites.

BACKGROUND: Mexican Americans (MAs) have higher incidence rates of intracerebral hemorrhage (ICH) than non-Hispanic whites (NHWs). The authors present clinical and imaging characteristics of ICH in MAs and NHWs in a population-based study. METHODS: This work is part of the Brain Attack Surveillance in Corpus Christi (BASIC) project. Cases of nontraumatic ICH were identified from 2000 to 2003. Multivariable logistic regression was used to assess the independent associations between ethnicity and ICH location (lobar vs nonlobar) and volume (> or = 30 vs < 30 mL), adjusting for demographics and baseline clinical characteristics. Logistic regression was also used to determine the association between ethnicity and in-hospital mortality, adjusting for confounders. RESULTS: A total of 149 MAs and 111 NHWs with ICH were identified. MAs were younger (70 vs 77, p < 0.001), more often male (55% vs 42%, p = 0.04), had a lower prevalence of atrial fibrillation (2.0% vs 13%, p < 0.001), and a higher prevalence of diabetes (39% vs 19%, p < 0.001). MA ethnicity was independently associated with nonlobar hemorrhage (OR 2.08, 95% CI: 1.15, 3.70). MAs had over two times the odds of having small (< 30 mL) hemorrhages compared with NHWs (OR = 2.41, 95% CI: 1.31, 4.46). NHWs had higher in-hospital mortality, though this association was no longer significant after adjustment for ICH volume, location, age, and sex. CONCLUSIONS: There are significant differences in the characteristics of ICH in MAs and NHWs, with MA patients more likely to have smaller, nonlobar hemorrhages. These differences may be used to examine the underlying pathophysiology of ICH.

Emergency department evaluation of ischemic stroke and TIA: the BASIC Project.

OBJECTIVE: To identify demographic and clinical variables of emergency department (ED) practices in a community-based acute stroke study. METHODS: By both active and passive surveillance, the authors identified cerebrovascular disease cases in Nueces County, TX, as part of the Brain Attack Surveillance in Corpus Christi (BASIC) Project, a population-based stroke surveillance study, between January 1, 2000, and December 31, 2002. With use of multivariable logistic regression, variables independently associated with three separate outcomes were sought: hospital admission, brain imaging in the ED, and neurologist consultation in the ED. Prespecified variables included age, sex, ethnicity, insurance status, NIH Stroke Scale score, type of stroke (ischemic stroke or TIA), vascular risk factors, and symptom presentation variables. Percentage use of recombinant tissue plasminogen activator (rt-PA) was calculated. RESULTS: A total of 941 Mexican Americans (MAs) and 855 non-Hispanic whites (NHWs) were seen for ischemic stroke (66%) or TIA (34%). Only 8% of patients received an in-person neurology consultation in the ED, and 12% did not receive any head imaging. TIA was negatively associated with neurology consultations compared with completed stroke (odds ratio [OR] 0.35 [95% CI 0.21 to 0.57]). TIA (OR 0.14 [0.10 to 0.19]) and sensory symptoms (OR 0.59 [0.44 to 0.81]) were also negatively associated with hospital admission. MAs (OR 0.58 [0.35 to 0.98]) were less likely to have neurology consultations in the ED than NHWs. Only 1.7% of patients were treated with rt-PA. CONCLUSIONS: Neurologists are seldom involved with acute cerebrovascular care in the emergency department (ED), especially in patients with TIA. Greater neurologist involvement may improve acute stroke diagnosis and treatment efforts in the ED.

Ischemic stroke subtypes among Mexican Americans and non-Hispanic whites: the BASIC Project.

Ischemic stroke subtype distribution was compared between Mexican Americans (MAs) and non-Hispanic whites (NHWs) in a community-based stroke surveillance study in Nueces County, TX. There was no difference in the distribution of stroke subtype by ethnicity (p = 0.19). There was a similar proportion of small-vessel and large-artery strokes between the two ethnic groups (p = 0.32). Differences in stroke rates among MAs and NHWs are not explained by the distribution of ischemic stroke subtypes.

A population-based study of acute stroke and TIA diagnosis.

BACKGROUND: Acute stroke therapy is heavily dependent on the diagnostic acumen of the physician in the emergency department (ED). OBJECTIVE: To determine this diagnostic accuracy in a population-based multiethnic stroke study. METHODS: The Brain Attack Surveillance in Corpus Christi (BASIC) Project prospectively ascertained all acute stroke or TIA cases in an urban Texas county of 313,645 residents without an academic medical center. Cases were validated by board-certified neurologists using source documentation. Case validation was used as the gold standard to compare the diagnosis given by the ED physician. RESULTS: From January 2000 to August 2002, a total of 13,015 patients were screened. Of these, 1,800 were validated as stroke/TIA. Overall sensitivity of the emergency physician for the BASIC-validated diagnosis was 92%, and positive predictive value was 89%. Of the cases that the emergency physician thought were stroke, 11% were validated as no stroke. In multivariable modeling, motor symptoms was an independent predictor of protection from false-negative ED diagnosis of stroke/TIA (odds ratio [OR] = 0.61; 95% CI 0.41 to 0.89). Protection from false-positive stroke/TIA diagnosis was predicted by sensory symptoms (OR = 0.43; 95% CI 0.28 to 0.66), motor symptoms (OR = 0.44; 95% CI 0.32 to 0.62), and severe neurologic deficit (OR = 0.33; 95% CI 0.14 to 0.78). History of stroke/TIA predicted false-positive stroke diagnosis (OR = 1.72; 95% CI 1.23 to 2.40). The majority of disagreements occurred in patients with generalized neurologic or acute medical, nonneurologic syndromes. CONCLUSIONS: Physicians practicing in the ED are sensitive for stroke/TIA diagnosis. The modest positive predictive value argues for a systems approach with neurology support so that proper decisions regarding acute stroke therapy can be made.

Use of Ephedra-containing products and risk for hemorrhagic stroke.

This case-control study examined the association between Ephedra use and risk for hemorrhagic stroke. For use of Ephedra at any dose during the 3 days before the stroke, the adjusted OR was 1.00 (95% CI 0.32 to 3.11). For daily doses of < or =32 mg/day, the OR was 0.13 (95% CI 0.01 to 1.54), and for >32 mg/day, the OR was 3.59 (95% CI 0.70 to 18.35). Ephedra is not associated with increased risk for hemorrhagic stroke, except possibly at higher doses.