Racial differences in demographics, acute complications, and outcomes in patients with subarachnoid hemorrhage: a large patient series

Transcription

1 J Neurosurg 103:18 24, 2005 Racial differences in demographics, acute complications, and outcomes in patients with subarachnoid hemorrhage: a large patient series DAVID ROSEN, M.D., ROBERTA NOVAKOVIC, M.D., FERNANDO D. GOLDENBERG, M.D., DEZHENG HUO, M.S., M.D., MARIA E. BALDWIN, M.D., JEFFREY I. FRANK, M.D., AXEL J. ROSENGART, M.D., PH.D., AND R. LOCH MACDONALD, M.D., PH.D. Section of Neurosurgery (Department of Surgery), Neurocritical Care and Acute Stroke Program, and Department of Health Studies, Pritzker School of Medicine, University of Chicago, Illinois Object. Few studies have focused on the impact of racial differences in demographics, clinical characteristics, acute complications, and outcomes of patients with aneurysmal subarachnoid hemorrhage (SAH). The purpose of this study was to examine this issue. Methods. The authors evaluated prospectively collected data on 1711 adult patients with aneurysmal SAH who were entered into two randomized, double-blind, placebo-controlled trials conducted at neurosurgical centers in North America between 1991 and Admission characteristics, treatment modalities, in-hospital complications, and 3-month outcomes assessed by application of the Glasgow Outcome Scale were compared using the chi-square test, a t-test, the Wilcoxon rank-sum test, and multiple logistic regressions based on a significance level of 0.05 in 241 African-American, 1342 Caucasian, and 128 other racial minority patients. Caucasian patients were significantly older than patients of other races (p ). African-American patients more frequently had a history of hypertension (p ) and an elevated blood pressure at the time of admission (p ). African-Americans and other racial minorities were more likely to have internal carotid artery aneurysms and Caucasians were more likely to have posterior circulation aneurysms (p = ). Rates of in-hospital complications were not significantly different except that pulmonary edema occurred more commonly in Caucasians (p = 0.036). After an adjustment was made for significant admission characteristics, the 3-month outcome was not significantly different among the races. Conclusions. Race was not found to be a prognostic factor for outcome after aneurysmal SAH. The higher SAH mortality rate previously observed in African-American patients is likely a result of a higher incidence of SAH in this group. These findings highlight the importance of primary prevention programs aimed at modifying risk factors for SAH. KEY WORDS race cerebral infarction cerebral ischemia subarachnoid hemorrhage vasospasm T HERE have been numerous epidemiological studies of SAH, 12,15,19,22,26,29 but few have focused on ethnic differences in SAH incidence and outcome. 1,3,16,21,27,28 In Cincinnati, Ohio, the age- and sex-adjusted incidence of SAH in African-Americans is 12 per 100,000 compared with six per 100,000 among Caucasian residents. 2 Nationally, the SAH mortality rate is approximately 50% higher in African-Americans. 1,19 The link between these findings remains unclear: the higher SAH mortality rates observed in African-Americans could be caused by a higher incidence of SAH, worse outcomes, or a combination of these factors. 21 Moreover, in many of these studies death was not used as an end point to gauge the impact of SAH because most patients with SAH do not die. Perhaps the studies may not have provided a correction for baseline characteristics Abbreviations used in this paper: CI = confidence interval; CT = computerized tomography; GOS = Glasgow Outcome Scale; ICA = internal carotid artery; ICH = intracerebral hemorrhage; IVH = intraventricular hemorrhage; MABP = mean arterial blood pressure; OR = odds ratio; SAH = subarachnoid hemorrhage; WFNS = World Federation of Neurosurgical Societies. 18 among races. In addition, almost nothing is known about whether there are differences in the demographics, clinical characteristics, and in-hospital courses of patients of different races. 8 In this study, we addressed some of these questions by evaluating prospectively collected data on 1711 adult patients with aneurysmal SAH who were entered into two randomized, double-blind, placebo-controlled drug trials conducted at neurosurgical centers in North America between 1991 and These trials included almost an order of magnitude more African-American patients than those included in most prior studies. Our purpose was to test the hypothesis that Caucasian, African-American, and other racial minority patients differ in admission characteristics, treatment modalities, in-hospital complications, and 3-month outcomes. Clinical Material and Methods Patient Population The database contained information on 1720 adult patients with aneurysmal SAH who had been recruited into

2 Racial differences in patients with subarachnoid hemorrhage two prospective, randomized, double-blind, placebo-controlled clinical trials of the drug tirilazad mesylate. 13,23 The trials were conducted at 65 neurosurgical centers in North America between 1991 and Tirilazad mesylate is a steroid derivative that is devoid of glucocorticoid effects but has a potent antioxidant activity; it has been tested as an antivasospastic and neuroprotective drug. Results in patients treated with tirilazad or placebo were analyzed together because tirilazad had no statistically significant relationship to any parameter reported in this manuscript. Unfortunately, any other additional information from these important, interesting, and worldwide studies is unavailable at present due to changes in the corporate governance of the pharmaceutical company originally responsible for the studies. Race data were missing for nine cases in the database. We therefore took as our study group the 1711 patients for whom this information was available. All these patients harbored an angiographically documented saccular aneurysm with SAH confirmed on CT scanning or lumbar puncture. Patients with all grades of SAH were included if they were at least 18 years of age and began drug or placebo treatment within 48 hours after onset of SAH. Patients with the following factors were excluded: 1) a nonsaccular aneurysm; 2) severe, concomitant medical, neurological, or psychiatric illness; 3) serious cardiovascular complications such as critically high, uncontrollable hypertension, any Q-wave myocardial infarction, serious cardiac arrhythmia, or severe congestive heart failure; 4) pregnancy or lactation; 5) use of steroid medications or calcium channel antagonists other than nimodipine before randomization; and 6) endovascular placement of a detachable coil in the ruptured aneurysm. Patients were randomized to receive placebo or tirilizad at various doses until Day 10 after SAH. The treating physician determined the timing of aneurysm repair surgery, which occurred at a mean of 55 hours after SAH. Nimodipine treatment, hemodynamic therapy, and mechanical or pharmacological angioplasty were used at the discretion of the treating physician. All patients received nimodipine, 62% received some component of prophylactic hemodynamic therapy (hypervolemia, induced hypertension or hemodilution), and 23% were treated with hemodynamic therapy. Other experimental therapies such as steroids and calcium channel antagonists were not allowed. Clinical and Radiological Variables Age, sex, race, modified Glasgow Coma Scale 18 score, and the WFNS 9 SAH grade were recorded on admission. Race was recorded as Caucasian, African-American, or other. Histories of hypertension, angina or myocardial infarction, diabetes, hepatic disease, thyroid disease, migraine headaches, and previous SAH were recorded. Blood pressure and body temperature were documented on admission and on Day 8 post-sah. Radiographic data included the amount of subarachnoid blood on initial CT scans, which was graded as none, diffuse or localized thin, or diffuse or localized thick. Intracerebral hemorrhage and IVH as well as hydrocephalus were noted. The location and maximum diameter of the ruptured aneurysm demonstrated on angiography were recorded and stratified into groups ( 12, 13 24, or 25 mm). Presence of vasospasm on the initial angiogram was noted. Data collected on in-hospital complications included neurological worsening, symptomatic vasospasm, cerebral edema, cerebral infarction, hydrocephalus, and pulmonary edema. Neurological worsening was defined as a decrease of at least two points in the modified Glasgow Coma Scale score or an increase of at least two points in the motor score of the National Institutes of Health Stroke Scale lasting for at least 8 hours. 6 When neurological worsening occurred, the treating physician was required to submit standardized forms that defined the criteria for neurological deterioration, the primary and secondary causes of the deterioration, and the investigations conducted to determine whether the worsening was caused by vasospasm or some other cause such as an infection. Neurological worsening was attributed to symptomatic vasospasm based on the following criteria: 1) classic symptoms including onset between Days 5 and 12 after SAH, worsening headache, stiff neck and/or low-grade fever, insidious onset of confusion, disorientation and/or decline in level of consciousness, and focal deficit, which may fluctuate in severity; 2) a CT scan of the head that excluded other causes of neurological worsening, such as rebleeding or hydrocephalus; 3) no other identifiable cause of neurological worsening, such as electrolyte disturbance, hypoxia, or seizure; and 4) confirmation of vasospasm by findings of angiography or transcranial Doppler ultrasonography, which was recommended but not required. 14 Cerebral infarction was classified based on the presence of new hypodensities not attributable to intracerebral hemorrhage or other nonischemic causes on CT scans obtained 3 months after SAH. Therapeutic interventions were recorded including surgical aneurysm application, anticonvulsant use at the time of admission, cerebral angioplasty, and hemodynamic therapy (hemodilution, induced hypertension and/or hypervolemia). Outcomes were assessed at 3 months post-sah by applying the GOS. 17 Statistical Analysis Patients were sorted into three groups based on race (African-American, Caucasian, or other). A univariate analysis was performed to detect significant differences in demographic, clinical, and radiographic variables among racial groups. Continuous variables were evaluated by performing two-sample t-tests or Wilcoxon rank-sum tests, and categorical variables by performing the chi-square test. The analyses were repeated for in-hospital complications and 3-month outcomes. Outcome was assessed on the five-point GOS and was also dichotomized into favorable and unfavorable outcome groups, with good recovery and moderate disability considered to be favorable and severe disability, vegetative state, and death considered unfavorable. Multivariate logistic regression was used to adjust for variables that differed among the races. Odds ratios and 95% CIs were calculated for in-hospital complications and outcomes. In this report all values are given as means standard deviations. A probability value lower than 0.05 was considered statistically significant. Results Of the 1720 patients, 1342 (78%) were Caucasian, 241 (14%) were African-American, and 137 (8%) were categorized as other (Table 1). One of the North American tiri- 19

3 D. Rosen, et al. TABLE 1 Demographic, clinical, and radiological characteristics of patients with SAH at admission stratified by race* African-American Caucasian Other Race Characteristic (241 patients) (1342 patients) (128 patients) p Value demographic mean age (yrs) male component (% of patients) clinical features admission WFNS grade (% of patients) I II III IV V admission blood pressure (mm Hg) systolic diastolic admission temperature 38 C (% of patients) preexisting condition (% of patients) hypertension angina/myocardial infarction diabetes hepatic disease migraine/headaches SAH thyroid disease radiological findings (% of patients) clot thickness 0.45 none focal thin focal thick diffuse thin diffuse thick IVH ICH location of ruptured aneurysm (%) ICA ACA MCA pst circ other size of ruptured aneurysm (%) 12 mm mm mm vasospasm at admission (%) * Values are expressed as the means standard deviations or percentages where appropriate. Abbreviations: ACA = anterior cerebral artery; MCA = middle cerebral artery; pst circ = posterior circulation. lazad trials included only women, which led to the inclusion of a lower than usual proportion of men in the present analysis. Female patients accounted for 83% of Caucasian patients, 90% of African-American patients, and 78% of patients categorized as other race. Significant racial differences were found for several demographic and clinical variables. For example, Caucasian patients were significantly older than patients of racial minorities, with mean age of years in Caucasians, years for African-Americans, and years for persons categorized as other race (p ). African-Americans were more likely to have a history of hypertension (61% compared with 37% of Caucasians and 29% of other racial minorities, p ). In addition, African-American patients had significantly higher mean systolic and diastolic blood pressures on admission: their MABP was 154/82 mm Hg compared with 141/75 mm Hg in Caucasian patients (p ) and 142/77 mm Hg in patients categorized as other race (p ). Several key characteristics were similar among the racial groups. There was no significant difference in the admission WFNS grade or in the histories of heart disease, angina, diabetes mellitus, hepatic disease, thyroid disease, migraine headaches, or previous SAH. The location of the ruptured aneurysm was the only radiographic variable that was significantly different among the races. Overall, a ruptured aneurysm located on the ICA was more likely to occur in non-caucasians (43% of African-Americans, 39% of other racial minorities, and 30% of Caucasians; p = ). Nevertheless, the study cohort had a high percentage of female patients, and women have been reported to have a higher incidence of ICA aneurysms, although specific racial differences among female patients have not been well described. 30 Therefore, aneurysm location was analyzed by race and sex. Significant racial differ- 20

4 Racial differences in patients with subarachnoid hemorrhage TABLE 2 Therapy use by race African-American Caucasian Other Race Treatment (241 patients) (1342 patients) (128 patients) p Value prophylactic hemodynamic therapy (% of patients) anticonvulsant use (% of patients) surgery to clip aneurysm (% of patients) hrs from SAH to surgery* 24 (22) 27 (26) 29 (25) 0.10 dose of tirilazad (% of patients) placebo mg/kg/day mg/kg/day mg/kg/day therapeutic hemodynamic therapy (% of patients) cerebral angioplasty (% of patients) * Values represent medians (interquartile ranges). ences were only found among female patients. Among female patients, a higher rate of ruptured ICA aneurysms was found in non-caucasians (45% of African-Americans, 41% of other racial minorities, and 30% of Caucasians) and a higher rate of posterior circulation aneurysms in Caucasians (20% of Caucasians, 10% of African-Americans, and 11% for other ethnic minorities; p = ). Although not statistically significant, a similar trend in the location of ruptured aneurysms was observed in male patients. Identical but statistically significant differences in the distribution of ruptured aneurysms were noted when racial groups were stratified based on a history of hypertension (p 0.05 for patients with and those without hypertension). No significant racial differences were observed in aneurysm size, subarachnoid blood clot thickness, ICH, IVH, or evidence of vasospasm on admission. Most treatment modalities did not significantly differ among ethnic groups including the use of tirilizad, timing of aneurysm surgery, and treatment for vasospasm including hemodynamic therapy and angioplasty (Table 2). Use of anticonvulsant medications was significantly different (p = ), with 92% of African-American, 83% of Caucasian, and 80% of other racial minority patients receiving anticonvulsant therapy. Analysis of in-hospital complications showed significant racial differences in the incidence of pulmonary edema after SAH (p = 0.036; Table 3). Sixteen percent of Caucasians, 11% of African-Americans, and 10% of patients categorized as other racial minority experienced pulmonary edema. No significant racial difference was observed in the likelihood of neurological worsening, cerebral vasospasm, infarction, cerebral edema, hydrocephalus, or elevated temperature on Day 8 post-sah. Outcome at 3 months posthemorrhage did not differ significantly among the racial groups. Approximately 55% of patients had a good recovery (GOS Score 5) with a mild or no residual neurological deficit and 15% of patients died (GOS Score 1). Because there were differences in baseline characteristics among the racial groups, multiple logistic regression modeling was used to delineate further the relationship between race and various end points including in-hospital complications and outcome. Factors were adjusted for known prognostic indicators of outcome after aneurysmal SAH as well as factors associated with outcome in the tirilazad database. 20,32 Table 4 contains the results of the regression model adjusted for age, neurological grade, MABP at admission, history of hypertension, location of ruptured aneurysm, anticonvulsant treatment, and country where treatment was given. Even after an adjustment was made for demographic variables including those that differed significantly among the races, no difference in 3-month outcome was observed. Patients categorized as other race had a significant increased risk of cerebral infarction (OR 1.61, 95% CI , p = 0.045) and hydrocephalus (OR 1.74, 95% CI , p = 0.016). None of the other in-hospital complications significantly differed among the racial groups. Discussion This analysis of 1720 patients with aneurysmal SAH from the US, Canada, and Mexico is the largest comparison of the demographics, clinical features, and in-hospital courses of patients who were Caucasian, African-American, and other races. We show that African-American patients tend to be younger and have a greater incidence of preexisting hypertension and higher blood pressure on admission. Non-Caucasian patients, in particular female patients, have more ICA aneurysms and are less likely to experience pulmonary edema. Importantly, once admitted to the hospital, patient outcome at 3 months post-sah was not significantly different among the races. These data are not population based, and therefore one cannot draw conclusions about the overall incidence or prevalence of demographic and clinical characteristics or about overall outcome. One could argue that the data are relatively representative of the general populations of these countries based on the fact, for example, that the percentage of African- American patients in the study was similar to that in the US population. Race was categorized by the investigators at each site based on what the patient or relatives reported or on observations by the admitting medical staff. Formal criteria for categorization of race defined by the US government were published only after these data were collected. 19 There is controversy over the use of race as a research variable. Fullilove 10 has noted that race is an arbitrary system of visual classification that does not demarcate distinct subspecies of the human population. Williams and Jackson 33 have stated that race is a socially constructed categorization system. On 21

5 D. Rosen, et al. TABLE 3 In-hospital complications and outcome by race Percentage of Patients African-American Caucasian Other Race Factor (241 patients (1342 patients) (128 patients) p Value 5-point GOS good recovery moderate disability severe disability vegetative state death dichotomous GOS favorable unfavorable vasospasm neurological worsening cerebral infarction hydrocephalus pulmonary edema cerebral edema temperature 38 C on Day 8 post-sah the other hand, all three authors have acknowledged that social systems and some diseases vary according to current race classification systems, although whether these categories are based mainly on differences in socioeconomic status or on some other factors remains controversial. Among African-Americans and other racial minorities, the mean age of the patient at the onset of SAH was 3 years younger than that in Caucasians. In one previous report authors noted a similar age difference. 12 Most other data on race and SAH are based on analyses of death certificates and whole populations, and indicate that there is a higher in- TABLE 4 Comparisons of outcomes between races based on the multiple logistic regression* OR (95% CI) African-American Other Minorities Overall Outcome vs Caucasian vs Caucasian p Value 5-point GOS 1.22 ( ) 1.46 ( ) 0.10 dichotomous GOS 1.35 ( ) 1.50 ( ) 0.11 vasospasm 0.70 ( ) 0.78 ( ) any hemodynamic therapy 0.80 ( ) 0.96 ( ) 0.44 neurological 0.99 ( ) 0.86 ( ) 0.75 worsening cerebral infarction 1.24 ( ) 1.61 ( ) use of cerebral 0.66 ( ) 0.92 ( ) 0.34 angioplasty hydrocephalus 1.21 ( ) 1.74 ( ) pulmonary edema 0.77 ( ) 0.68 ( ) 0.30 cerebral edema 1.03 ( ) 1.26 ( ) 0.72 temperature 38 C on Day 8 post-sah 0.88 ( ) 0.96 ( ) 0.74 * Adjustments were made for age, neurological grade, MABP at admission, history of hypertension, location of ruptured aneurysm, anticonvulsant treatment, and country where treatment was given. Proportional odds model. cidence of SAH among African-Americans. 3,4,12,16 The reasons for differences among races in the incidence of SAH are not specifically addressed in this paper. There was a higher incidence of preexisting hypertension in African- American patients in this study, and one could speculate that this contributed to the findings because hypertension probably is a risk factor for SAH. 22,31 Authors of earlier studies have noted the higher incidence of hypertension among patients with African heritage in general and among those with SAH in the US and the United Kingdom specifically. 3,7,25 The present study is limited given the absence of data on other risk factors for SAH, including cigarette smoking and alcohol and cocaine use. 5,22,31 The only other significantly different clinical characteristic among the races was the location of the ruptured aneurysm. Among female patients, ruptured ICA aneurysms were found more frequently in non-caucasians and posterior circulation aneurysms more frequently in Caucasians. The reasons for sex and racial differences in aneurysm location are unknown. Most treatment modalities were used with similar frequency among the racial groups, with the only difference being the more frequent use of anticonvulsant medications in African-Americans. Racial discrepancies in medical care have been reported; regarding stroke, however, a comprehensive multicenter analysis of 1073 patients with ischemic stroke demonstrated that most treatment modalities were used with similar frequency in Caucasians and non-caucasians. 11,24 Race was not a significant factor in outcome after SAH in this study, a finding consistent with those of previous studies. 3,27 The occurrence of most in-hospital complications were similarly unaffected by race. Only pulmonary edema was more common in Caucasian patients. Results of the multivariate analysis showed a significant increase in cerebral infarction and hydrocephalus in patients categorized as other race. Studies based on a review of US death certificates and population-based assessments have demonstrated a higher mortality rate in African-Americans from SAH. 1,19 The present results help clarify the issue of whether 22

6 Racial differences in patients with subarachnoid hemorrhage the higher SAH-associated mortality rate in African-Americans is due to a higher incidence of SAH, a worse outcome after SAH, or a combination of these factors. These data also indicate that if a patient survives long enough to reach the hospital, race is no longer a factor in outcome. Studies of medical record discharge codes, although more prone to methodological problems, also have provided evidence indicating that the excessive SAH-associated mortality rate is due to the increased incidence of SAH and not to case fatality. 3,21 This evidence indicates but does not prove that the elevated incidence of SAH in African-Americans is likely to be the driving force creating racial disparities in SAH mortality rates. Perhaps there is a higher prehospitalization mortality rate from SAH in African-Americans. This study included some patients classified as members of other racial minorities; these could include Hispanic, Asian, and Native Americans. Higher SAH mortality rates have been observed in other ethnic minorities, but incidence data are lacking and comparative studies of in-hospital courses are scarce. 1 In the present analysis, we grouped all non African-American minorities in the other race category, obscuring any conclusions about individual racial groups. The strongest statement that can be made is that being a non African-American minority is not a risk factor for worse outcome after SAH. Limitations of the present analysis include the question of whether the racial groups are represented at a rate roughly similar to their occurrence in the general population. According to the March 2002 US census, 13% of the nation s population was African-American, 69% was Caucasian, and 18% was another race. These percentages indicate that the population of the present study overrepresented African-Americans and underrepresented other racial groups. A second limitation has to do with the fact that these patients were treated in a clinical trial. Whether the results would be different among patients of various races treated outside a clinical trial is not addressed here. Conclusions In summary, race was not a prognostic factor for outcome following aneurysmal SAH. Outcome was assessed at 3 months; ideally a longer-term follow-up review of this or a similar group of patients would be worthwhile to verify these findings. The higher SAH mortality rate previously observed in African-Americans is likely due to a higher incidence of SAH in this group. This finding highlights the importance of primary prevention programs aimed at modifying the risk factors for SAH. 5 References 1. Ayala C, Greenlund KJ, Croft JB, Keenan NL, Donehoo RS, Giles WH, et al: Racial/ethnic disparities in mortality by stroke subtype in the United States, Am J Epidemiol 154: , Broderick J, Brott T, Kothari R, Miller R, Khoury J, Pancioli A, et al: The Greater Cincinnati/Northern Kentucky Stroke Study: preliminary first-ever and total incidence rates of stroke among blacks. Stroke 29: , Broderick JP, Brott T, Tomsick T, Huster G, Miller R: The risk of subarachnoid and intracerebral hemorrhages in blacks as compared with whites. N Engl J Med 326: , Broderick JP, Brott T, Tomsick T, Miller R, Huster G: Intracerebral hemorrhage more than twice as common as subarachnoid hemorrhage. J Neurosurg 78: , Broderick JP, Viscoli CM, Brott T, Kernan WN, Brass LM, Feldmann E, et al: Major risk factors for aneurysmal subarachnoid hemorrhage in the young are modifiable. Stroke 34: , Brott T, Adams HP Jr, Olinger CP, Marler JR, Barsan WG, Biller J, et al: Measurements of acute cerebral infarction: a clinical examination scale. Stroke 20: , Burt VL, Whelton P, Roccella EJ, Brown C, Cutler JA, Higgins M, et al: Prevalence of hypertension in the US adult population. Results from the Third National Health and Nutrition Examination Survey, Hypertension 25: , Dennis GC, Welch B, Cole AN, Mendoza R, Morgan J, Epps J, et al: Subarachnoid hemorrhage in the African-American population: a cooperative study. J Nat Med Assoc 89: , Drake CG: Report of World Federation of Neurological Surgeons Committee on a Universal Subarachnoid Hemorrhage Grading Scale. J Neurosurg 68: , Fullilove MT: Comment: abandoning race as a variable in public health research an idea whose time has come. Am J Public Health 88: , Goldstein LB, Matchar DB, Hoff-Lindquist J, Samsa GP, Horner RD: Veterans Administration Acute Stroke (VASt) Study: lack of race/ethnic-based differences in utilization of stroke-related procedures or services. Stroke 34: , Gross CR, Kase CS, Mohr JP, Cunningham SC, Baker WE: Stroke in south Alabama: incidence and diagnostic features a population based study. Stroke 15: , Haley EC Jr, Kassell NF, Apperson-Hansen C, Maile MH, Alves WM: A randomized, double-blind, vehicle-controlled trial of tirilazad mesylate in patients with aneurysmal subarachnoid hemorrhage: a cooperative study in North America. J Neurosurg 86: , Haley EC Jr, Kassell NF, Torner JC: A randomized controlled trial of high-dose intravenous nicardipine in aneurysmal subarachnoid hemorrhage. A report of the Cooperative Aneurysm Study. J Neurosurg 78: , Ingall TJ, Asplund K, Mähönen M, Bonita R: A multinational comparison of subarachnoid hemorrhage epidemiology in the WHO MONICA stroke study. Stroke 31: , Ingall TJ, Whisnant JP: Epidemiology of subarachnoid hemorrhage, in Yanagihara T, Piepgras DG, Atkinson JLD (eds): Subarachnoid Hemorrhage. Medical and Surgical Management. New York: Marcel Dekker, 1998, pp Jennett B, Bond M: Assessment of outcome after severe brain damage. A practical scale. Lancet 1: , Jennett B, Teasdale G, Braakman R, Minderhoud J, Heiden J, Kurze T: Prognosis of patients with severe head injury. Neurosurgery 4: , Johnston SC, Selvin S, Gress DR: The burden, trends, and demographics of mortality from subarachnoid hemorrhage. Neurology 50: , Kassell NF, Torner JC, Haley EC Jr, Jane JA, Adams HP, Kongable GL: The International Cooperative Study on the Timing of Aneurysm Surgery. Part 1: Overall management results. J Neurosurg 73:18 36, Kennedy BS, Kasl SV, Brass LM, Vaccarino V: Trends in hospitalized stroke for blacks and whites in the United States, Neuroepidemiology 21: , Kissela BM, Sauerbeck L, Woo D, Khoury J, Carrozzella J, Pancioli A, et al: Subarachnoid hemorrhage: a preventable disease with a heritable component. Stroke 33: , Lanzino G, Kassell NF: Double-blind, randomized, vehicle-controlled study of high-dose tirilazad mesylate in women with aneurysmal subarachnoid hemorrhage. Part II. A cooperative study in North America. J Neurosurg 90: , Oddone EZ, Horner RD, Monger ME, Matchar DB: Racial variations in the rates of carotid angiography and endarterectomy in pa- 23

7 D. Rosen, et al. tients with stroke and transient ischemic attack. Arch Intern Med 153: , Otten MW Jr, Teutsch SM, Williamson DF, Marks JS: The effect of known risk factors on the excess mortality of black adults in the United States. JAMA 263: , Phillips LH II, Whisnant JP, O Fallon WM, Sundt TM Jr: The unchanging pattern of subarachnoid hemorrhage in a community. Neurology 30: , Qureshi AI, Safdar K, Patel M, Janssen RS, Frankel MR: Stroke in young black patients. Risk factors, subtypes, and prognosis. Stroke 26: , Sacco RL, Hauser WA, Mohr JP: Hospitalized stroke in blacks and Hispanics in northern Manhattan. Stroke 22: , Sacco RL, Wolf PA, Bharucha NE, Meeks SL, Kannel WB, Charette LJ, et al: Subarachnoid and intracerebral hemorrhage: natural history, prognosis, and precursive factors in the Framingham Study. Neurology 34: , Sahs AL, Perret GE, Locksley HB, et al: Intracranial Aneurysms and Subarachnoid Hemorrhage: A Cooperative Study. Philadelphia: Lippincott, 1969, pp Teunissen LL, Rinkel GJ, Algra A, Van Gijn J: Risk factors for subarachnoid hemorrhage: a systematic review. Stroke 27: , Torner JC, Kassell NF, Wallace RB, Adams HP Jr: Preoperative prognostic factors for rebleeding and survival in aneurysm patients receiving antifibrinolytic therapy: report of the Cooperative Aneurysm Study. Neurosurgery 9: , Williams DR, Jackson JS: Race/ethnicity and the 2000 census: recommendations for African American and other black populations in the United States. Am J Public Health 90: , 2000 Manuscript received June 11, Accepted in final form March 2, Address reprint requests to: R. Loch Macdonald, M.D., Ph.D., Section of Neurosurgery, MC3026, University of Chicago Medical Center, 5841 South Maryland Avenue, Chicago, Illinois rlmacdon@uchicago.edu. 24