Doi:10.1016/s0140-6736(00)02502-2

Differences in risk factors, atherosclerosis, and cardiovasculardisease between ethnic groups in Canada: the Study of HealthAssessment and Risk in Ethnic groups (SHARE) Sonia S Anand, Salim Yusuf, Vladmir Vuksan, Sudarshan Devanesen, Koon K Teo, Patricia A Montague, Linda Kelemen, Cheelong Yi, Eva Lonn, Hertzel Gerstein, Robert A Hegele, Matthew McQueen, for the SHARE Investigators IntroductionEthnicity-based research can identify new clues to Background Cardiovascular disease rates vary greatly between pathogenesis of a disease, since the populations under study ethnic groups in Canada. To establish whether this variation are heterogeneous in genetic and lifestyle characteristics.
can be explained by differences in disease risk factors and There is increasing evidence that rates of cardiovascular subclinical atherosclerosis, we undertook a population-based disease vary between ethnic groups. Previous studies study of three ethnic groups in Canada: South Asians, Chinese, indicate that people of South Asian origin have higher rates of this disease than people of European origin, a finding that Methods 985 participants were recruited from three cities cannot be explained by differences in conventional (Hamilton, Toronto, and Edmonton) by stratified random cardiovascular risk factors, such as smoking, raised blood sampling. Clinical cardiovascular disease was defined by history pressure, diabetes, or high cholesterol.1,2 Conversely, people or electrocardiographic findings. Carotid atherosclerosis was of Chinese origin have lower rates of cardiovascular disease than people of European origin, and with the exception ofglucose intolerance, have a more favourable risk factor (smoking, hypertension, diabetes, raised cholesterol) and novel risk factors (markers of a prothrombotic state) were measured.
These findings were confirmed by a recent analysis of Findings Within each ethnic group and overall, the degree of Canadian mortality data, in which South Asian Canadians carotid atherosclerosis was associated with a higher prevalence had the highest rates of cardiovascular disease, Canadians of of cardiovascular disease. South Asians had the highest European origin had intermediate rates, and Chinese prevalence of this condition compared with Europeans and Canadians had the lowest rates.5 To examine this variation Chinese (11%, 5%, and 2%, respectively, p=0·0004). Despite we measured the prevalence of cardiovascular disease in this finding, Europeans had more atherosclerosis (mean of the Canadians of South Asian, Chinese, and European origin maximum intimal medial thickness 0·75 [0·16] mm) than South and then identified whether these differences could be Asians (0·72 [0·15] mm), and Chinese (0·69 [0·16] mm). South explained by: conventional risk factors; novel risk factors Asians had an increased prevalence of glucose intolerance, (fibrinogen, plasminogen activator inhibitor-1 [PAI-1], higher total and LDL cholesterol, higher triglycerides, and lower lipoprotein (a), homocysteine); or atherosclerosis as HDL cholesterol, and much greater abnormalities in novel risk measured by B-mode carotid ultrasonography.
homocysteine, lipoprotein (a), and plasminogen activator The Study of Health Assessment and Risk in Ethnic groups Interpretation Although there are differences in conventional (SHARE) is a prospective investigation of atherosclerosis, and novel risk factors between ethnic groups, this variation and cardiovascular disease, and its determinants in three ethnic the degree of atherosclerosis only partly explains the higher groups in Canada: South Asians, Chinese, and Europeans.6 rates of cardiovascular disease among South Asians compared Canadians were classified as South Asian if their ancestors with Europeans and Chinese. The increased risk of originated from India, Pakistan, Sri Lanka, or Bangladesh; cardiovascular events could be due to factors affecting plaque Chinese if their ancestors originated from China, Taiwan, or rupture, the interaction between prothrombotic factors and Hong Kong; and European if their ancestors originatedfrom Europe. Participants were between 35 and 75 years, atherosclerosis, or as yet undiscovered risk factors.
and had lived in Canada for 5 years or more. South Asian and Chinese Canadians were identified by the previously validated method of unique surname classification.7,8 Briefly,a sampling frame was created for each ethnic group. For theSouth Asians and Chinese a database of unique surnameswas created manually and merged with a compact disk Departments of Medicine (S S Anand MD, S Yusuf FRCP, compilation of public telephone directories from across P A Montague BSc, L Kelemen RD, C Yi MSc, E Lonn MD, H Gerstein MD) Canada. The resulting lists were then limited to three cities, and Pathology and Molecular Medicine (M McQueen MBChB), Hamilton, Toronto, and Edmonton. The South Asian and McMaster University, Ontario, Canada; Department of Medicine, Chinese sampling frames were sorted by postal codes, which University of Toronto (V Vuksan PhD, S Devanesen MD); Department created implicit stratification by geographic region. From of Medicine, University of Alberta (K K Teo MBBch); and this list, a household was randomly selected and mailed an John P Robarts Research Institute, University of Western Ontario introductory letter, followed by up to 12 telephone calls inviting the individual with the earliest date of birth from the Correspondence to: Dr S Anand, Preventive Cardiology and household to participate in the study. After a South Asian or Therapeutics Research Program, McMaster University, HGH-McMaster Clinic, Hamilton, Ontario Canada L8L 2X2 Chinese respondent was confirmed for a clinic visit, their postal code was re-entered in the compact disk program, For personal use only. Not to be reproduced without permission of The Lancet.
and a list of individuals with the same postal code was maximum intimal medial thickness (IMT) for each generated. Individuals whose names were not South Asian segment. Our method was similar to the validated protocols or Chinese were presumed to be people of European origin.
used in the SECURE, ACAPS, PLAC-II, and PLAC-II, They were selected randomly and approached in the same and PREVENT studies.17,18,19,20 All ultrasound readings were done in the core laboratory in Hamilton by two trained andcertified readers. Videotaped examinations were reviewed qualitatively and a minimum of three frames from each After providing written informed consent, fasting blood prespecified arterial segment with the thickest IMT were samples were obtained from all participants, and 2-h post- digitised, and measurements made. The mean of the glucose load (75 g oral solution) blood samples were taken maximum IMT readings for the 12 segments was calculated from non-diabetic participants. All samples were frozen at for each participant. Additional measurements included the –70°C and transferred on dry ice to the core laboratory in Hamilton, Canada, for central analysis with standard coefficients for between and within ultrasonographer methodology. All blood analyses were done in core (r=0·91, r=0·90, respectively) and between and within laboratories without knowledge of an individual’s ethnicity, reader (0·88, 0·92, respectively) reliability indicated our data clinical history, or risk factor status. Total cholesterol, (taken from three centres by four sonographers and two triglycerides, and glucose were measured by enzymatic readers) yielded highly reproducible measures of carotid methods,9–11 and LDL cholesterol was calculated.12 HDL cholesterol was measured after precipitating the VLDL and Cardiovascular disease was defined as individuals with: LDL with phosphotungstic acid and magnesium chloride.13 coronary artery disease, defined as angina (Rose Lipoprotein (a) was quantitatively measured by automated questionnaire); a self-reported admission for a myocardial immunoprecipitation and turbidimetric detection on the infarction, silent myocardial infarction (major Q waves by Ciba-Corning 550 EXPRESS (Oberlin, OH, USA) with the Minnesota criteria21); percutaneous coronary angioplasty Incstar (Stillwater, MN, USA) test kit for lipoprotein (a).
(PTCA) or coronary artery bypass graft (CABG) surgery; or Homocysteine was measured by high-performance liquid cerebrovascular disease, defined by self-report or a previous chromatography with fluorescence detection.14 PAI-1 was measured with a two-stage indirect enzymatic assay by useof a Biopool commercial kit Spectrolyse (plasminogen) (Ventura, CA, USA).15 Fibrinogen was measured with a All analyses were done with the SAS system (version 6.12).
one-stage clotting assay by Clauss’s method.16 Rates of discrete variables for each ethnic group were ageand sex standardised with the direct method by use of all participants in the study as the standard population.22 Subclinical atherosclerosis was measured in all participants Continuous variables are presented as age-adjusted and sex- by carotid B-mode ultrasonography. In Hamilton and adjusted means for each ethnic group with their standard Toronto, a high resolution ACUSON/ACOUSTIC deviations, unless otherwise specified. ANCOVA was used to analyse continuous variables with adjustment for age and (Mountain View, CA, USA), equipped with a 7.5 MHz sex when appropriate. Post-hoc pairwise comparisons were broad bandwidth frequency carotid probe was used. In done with Tukey’s approach to adjust for multiple Edmonton, the ATL-UM9-HDI (Bothell, WA, USA) comparisons. For statistical comparisons of discrete imaging system equipped with a 10 MHz transducer was variables between ethnic groups, logistic regression was used used. All ultrasound examinations were recorded on S-VHS with age and sex as covariates. A multiple logistic regression tapes and subsequently digitised and analysed offline in a analysis was used to examine the multivariate-adjusted odds ratios for cardiovascular disease of conventional and novel ultrasonography consisted of a transverse scan followed by a risk factors, atherosclerosis, and ethnicity.
full circumferential scan aimed at obtaining images from six On the basis of published rates of cardiovascular disease well defined carotid artery segments for the right and left for South Asians, Chinese, and Europeans, we postulated carotid arteries (12 segments per patient), identifying the that South Asians would have the highest levels of South Asians vs
South Asians vs
Europeans vs
Family history of myocardial infarction† *Any cardiovascular disease=history of myocardial infarction, angina, PTCA, CABG, or stroke. †Age and sex standardised. ‡Age standardised.
Table 1: Key demographic characteristics of the SHARE cohort For personal use only. Not to be reproduced without permission of The Lancet.
South Asians vs
South Asians vs
Europeans vs
*PTCA=percutaneous coronary angioplasty, CABG=coronary artery bypass graft. †Any coronary heart disease=history of myocardial infarction, angina, silent myocardial infarction, PTCA,CABG, ‡Age and sex standardised. §Cardiovascular disease=history of myocardial infarction, angina, silent myocardial infarction, PTCA, CABG, or stroke.
Table 2: Prevalence of coronary heart disease and cardiovascular disease subclinical atherosclerosis, and that the Chinese would have shows key demographic data. Most South Asians (244 the least amount of atherosclerosis. Sample size calculations [71·3%]) were born in the Indian subcontinent; 150 (44%) suggested that 320 participants per ethnic group would reported their mother tongue as Punjabi, 51 (15%) as provide over 80% power to detect differences in the mean of Gujarati, 44 (13%) as Hindi, 23 (7%) as Tamil, 35 (10%) as the maximum IMT of at least 0·056 mm (about an 8% English, and the remaining (39 [11%]) participants spoke relative difference) between each ethnic group. This sample other languages. The most commonly practised religions size would also provide sufficiently high power to detect were Hinduism 118 (34%), Islam 91 (27%), Sikhism 86 important differences in cardiovascular disease and risk (25%), Christianity 31 (9%), and 15 (5%) other. Most factor prevalence between ethnic groups.
Chinese participants originated from mainland China (136[43%]), Hong Kong (64 [20%]), Malaysia (18 [6%]), Vietnam 18 (6%), Taiwan 11 (3%), and the remaining Chinese participants (70 [22%]) came from other countries.
Of the 7728 households selected in the initial random The most commonly practised religions were Christianity sample, 5769 (75%) were contacted, 3172 (55%) of these (131 [41%]), traditional Chinese (54 [17%]), Buddhism completed the screening telephone call, and 1566 (49%) (45 [14%]), and the remaining participants (87 [28%]) were eligible. Of this group, 985 (63%) individuals practised other religions. Of the European group, most completed all components of the clinic visit. The response reported their country of origin as Canada (237 [73%]), or rate among eligible individuals did not vary much between the UK (26 [8%]), with 63 (19%) originating from other ethnic groups: 342 of 579 (59%) South Asians responded, countries. The most common religion was Christianity (298 317 of 458 (69%) Chinese, and 326 of 529 (62%) [91%]). 120 (35%) South Asians, 107 (34%) Chinese, and Europeans. When data from non-responders (n=1527) were 113 (35%) Europeans reported their job category as compared with that from participants (985) no differences professional, although South Asians and Chinese were more in rates of smoking (183 [12%] vs 108 [11%]), heart disease likely to have received a university education than their (76 [5%] vs 55 [6%]), or cancer (41 [3%] vs 37 [4%]) were European counterparts (152, 44% vs 127, 40% vs 106, 33%).
identified. However, non-responders were less likely to be employed (931 [61%] vs 689 [70%], p=0·001), but more South Asian participants had a higher prevalence of likely to have had some post-secondary education (808 established cardiovascular disease than Europeans and [53%] vs (385 [39%], p=0·001) than were participants.
Chinese (table 1). In addition, when this definition was From October, 1996, to October, 1998, 985 men and expanded to include silent myocardial infarction as detected women (342 South Asians, 317 Chinese, and 326 by electrocardiogram, South Asians had a greater burden of Europeans) completed the clinic visit. On average, people of disease than the Europeans and Chinese (table 2).
European ancestry had lived in Canada more than twice as Figure 1 shows mean of the maximum IMT values for all long as people of South Asian and Chinese origin. Table 1 participants with and without prevalent cardiovasculardisease. Overall, the values were significantly higher in people with such disease than those without (mean 0·81 [SD 0·16] vs 0·72 [0·15] mm, p=0·0001) after adjustment for age, sex, and recruiting centre. When this relation was examined within each ethnic group, a similar pattern was seen, confirming that the mean of the maximum IMT is a valid surrogate marker of cardiovascular disease in each ethnic group. The European group had the greatest mean maximum IMT (0·75 [0·16] mm), the Chinese had the lowest (0·69 [0·16]; p=0·0001), and the South Asians had an intermediate mean maximum IMT (0·72 [0·15]; p for South Asian vs European p=0·00098, South Asian vs Chinese=0·12), after adjustment for age, sex, and recruiting centre. This pattern was also noted when the single maximum IMT measurements were compared (European mean 1·27 mm [SD 0·54] vs Chinese 1·15 [0·53] and SouthAsian 1·23 [0·55], overall p=0·058). However, for any given Figure 1: Validation of carotid thickness (mean maximum IMT) degree of atherosclerosis, South Asians had an excess of as a correlate of cardiovascular disease*p<0·05; †p<0·01; ‡p<0·0001.
cardiovascular events compared with the other groups. For For personal use only. Not to be reproduced without permission of The Lancet.
South Asians vs
South Asians vs
Europeans vs
*Adjusted for age and gender. †LVH=Left ventricular hypertrophy. ‡Excluding known diabetes. §Natural log.
Table 3: Conventional and novel risk factors for cardiovascular disease* (mean [SD]) example, in those within the highest quartile of Chinese intermediate (5·19 [1·07] mmol/L), and the atherosclerosis, the rates of cardiovascular disease were 22 of Europeans the lowest (5·13 [1·08] mmol/L). A 2-h glucose 86 (26%) among South Asians, 13 of 104 (13%) among tolerance test was undertaken in all non-diabetic Europeans, and 3 of 45 (7%) among the Chinese group participants. By use of criteria of a fasting glucose of less than 7·0 mmol/L and a 2-h glucose of 7·8 mmol/L or moreand less than 11·1 mmol/L, 64 (19%) of South Asians had impaired glucose tolerance versus 49 (15%) of Chinese, and To examine why South Asians had more cardiovascular 37 (12%) of Europeans, (overall p=0·03). In addition, 34 disease than Europeans despite less carotid atherosclerosis, (10%) of South Asians, 21 (6%) of Europeans, and 15 (5%) we calculated the rates of conventional and novel risk factors of Chinese (overall p=0·03) were diagnosed as having between the ethnic groups. The Chinese group had the diabetes (fasting glucose of 7·0 mmol/L or more, or a 2-h highest rate of hypertension requiring medication, the South glucose of 11·1 mmol/L or more). Therefore, after exclusion Asians had the highest rates of diabetes, and the Europeans of people with diabetes at entry, the prevalence of newly were more likely to be current or former smokers (table 1).
diagnosed glucose intolerance (impaired glucose tolerance Although no difference in the number of individuals with or new diabetes) was 87 of 308 (28%) in South Asians vs 61 abnormal lipids receiving drug treatment was noted, South of 302 (20%) in Chinese, and 55 of 305 (18%) in Asians had the highest total cholesterol, LDL cholesterol, and triglycerides compared with the Chinese and No differences in mean systolic blood pressures were Europeans. In addition, the concentration of HDL-C was identified between groups, although South Asians had the significantly lower in the South Asians (table 3). Of those highest diastolic blood pressure, which persisted even after participants who did not have a diagnosis of diabetes at participants with hypertension requiring treatment were entry, South Asians had a significantly increased mean excluded. European men were the heaviest, South Asian fasting glucose concentration (5·47 [SD 1·10] mmol/L), the men intermediate, and the Chinese men were the lightest.
For women, Europeans and South Asians were the heaviestcompared with the Chinese. Similarly, South Asian and European men had the greatest amount of abdominal adiposity as measured by the ratio of waist-to-hip circumference compared with Chinese men, although South Asian women had significantly more abdominal adiposity than European and Chinese women (table 3).
South Asians had greater abnormalities in prothrombotic markers such as raised fibrinogen, PAI-1, lipoprotein (a), andhomocysteine concentrations than other groups (table 3).
Predictors of cardiovascular disease To identify predictors of cardiovascular events in all participants, we developed a multivariate logistic regression model with prevalent cardiovascular disease as the dependent variable, and the risk factors derived from the Framingham study23 (age, sex, systolic and diastolic blood Figure 2: Prevalence of cardiovascular disease for specific pressure, HDL-C, total cholesterol, smoking status [current, former, or never], diabetes and left ventricular hypertrophy For personal use only. Not to be reproduced without permission of The Lancet.
lipid and glucose abnormalities than the other ethnic groups, and a preponderance of abnormalities of novel risk factors, such as raised plasma concentrations of fibrinogen, plasminogen activator inhibitor 1, lipoprotein (a), and homocysteine. In the multivariate model we noted that South Asian ethnicity itself was a strong, independent determinant of cardiovascular events, in addition to conventional risk factors, subclinical atherosclerosis, and Identification of reasons for an independent effect of South Asian ethnicity on disease requires careful consideration, since conventional cardiovascular risk factors and atherosclerosis are believed to be the principal contributors to pathogenesis.24 Clinical cardiovascular events such as acute myocardial infarction, or stroke are believed to Table 4: Predictors of cardiovascular disease by multivariate result from unstable atherosclerotic plaques that rupture, leading to development of thrombosis and acute occlusion [on the 12-lead electrocardiogram]), atherosclerosis as of a critical artery.25 The processes that link the presence of measured by mean maximum IMT, novel risk factors atherosclerosis to the development of clinical disease are (lipoprotein a, PAI-1, homocysteine, and fibrinogen) and multiple, complex, and probably vary between different ethnicity, as the independent variables. The recruiting ethnic groups. Carotid atherosclerosis, as measured by mean centre and the number of years each participant lived in maximum IMT could be a measure of early and late Canada were entered as covariates. A Framingham risk atherosclerotic thickening. It might also be a marker for score was calculated, and participants grouped into risk small plaques that may not necessarily cause arterial lumen tertiles (low, moderate, and high). We compared the odds obstruction, but could rupture and cause coronary occlusion ratio (OR) for cardiovascular disease of the middle risk tertile and the lowest tertile by logistic regression (OR=2·49 mechanisms may also be important in pathogenesis.27 [95% CI: 0·67–9·20], p=0·18), and the highest and lowest Potential factors include differences in: composition and risk tertile (5·28 [1·45–19·16], p=0·007). Subclinical stability of atherosclerotic plaques; inflammation; triggers atherosclerosis (mean maximum IMT) was also a strong and rate of plaque rupture; and thrombogenic tendency.
independent risk predictor (8·49 [2·89–27·80], p=0·0002).
Perhaps the higher lipid and glucose abnormalities The addition of novel risk factors to the model identified reported in South Asians might not only be associated with PAI-1 (1·04 [1·01–1·08], p=0·006), and suggested degree of atherosclerosis, but could also lead to decreased lipoprotein (a) (1·29 [1·01–1·08], p=0·03) as independent stability of atherosclerotic plaques. This possibility predictors of cardiovascular events. Over and above these combined with a greater prothrombogenic environment factors, South Asian ethnicity was identified as a strong and because of higher concentrations of prothrombotic factors independent determinant of cardiovascular events, the OR could partly explain why South Asians have more of the disease for South Asians versus Europeans was 4·51 (1·46–13·89), p=0·02, after adjustment for conventional atherosclerosis compared with other ethnic groups.28,29 and novel risk factors, and atherosclerosis. However, despite There are two major implications of our study. First, we a more favourable risk factor profile and a lower prevalence have shown that ethnicity or some unmeasured factors linked of the condition, Chinese ethnicity was not associated with a to ethnicity might mediate the effect of atherosclerosis on the lower OR of cardiovascular disease compared with the risk of cardiovascular disease. Second, our study suggests that Europeans (1·05 [0·30–3·63], p=0·91) after adjusting for findings from European populations cannot be fully the above variables. No substantial interaction between extrapolated to individuals from other ethnic groups, ethnicity and conventional factors and atherosclerosis was emphasising the importance of collecting ethnic-specific data.
detected. Table 4 shows the adjusted OR for each predictor Our study has several strengths: first, it is prospective and used random sampling from the general Canadianpopulation in three cities, thereby increasing generalisability.
Second, we used standard methods across all ethnic groups.
We have confirmed the national mortality patterns that Third, the use of subclinical markers of disease makes it Canadians of South Asian origin have an increased unlikely that knowledge of disease status could alter an prevalence of cardiovascular disease compared with individual’s lifestyle or risk factors.
Canadians of European and Chinese origin.5 We reported There are some potential limitations to our study.
that carotid atherosclerosis was higher in participants with SHARE was a cross-sectional study, and although prevalent disease across all ethnic groups, supporting its participants were selected at random from the population, validity as a surrogate marker for disease. However, the there could have been systematic differences between those much lower amount of carotid atherosclerosis seen among who volunteered and non-responders. Thus we gathered key South Asians, even after adjustment for age, sex, and sociodemographic data from non-responders,6 but did not recruiting centre suggested a paradox: South Asians had less identify any substantial differences in disease prevalence or atherosclerosis than Europeans, yet they had higher smoking rates between the groups. Although non- cardiovascular disease rates. Indeed, for any given degree of responders were less likely to be employed, they had atherosclerosis, South Asians had a higher prevalence of received more post-secondary education than participants, cardiovascular disease than both Europeans and Chinese, and this difference was consistent across ethnic groups.
and Europeans tended to have a higher prevalence than the Further, the proportion of responders among those eligible Chinese. We also noted that South Asians had more plasma was similar across ethnic groups. We relied on individuals’ For personal use only. Not to be reproduced without permission of The Lancet.
recall of their personal medical history, which may be Anand S, Yusuf S, Vuksan V, et al. The Study of Health Assessment and limited by factors such as memory, social desirability, and Risk in Ethnic groups (SHARE): rationale and design. Can J Cardiol
1998; 14: 1349–57.
possible volunteer bias. However, concordant differences in Sheth T, Nargundkar M, Chagani K, Anand S, Nair C, Yusuf S.
Classifying ethnicity utilising the Canadian Mortality Data Base. Ethn revascularisation rates, or documented clinical events) also Health 1997; 2: 287–95.
differed between ethnic groups, indicating that recall bias is Harland JO, White M, Bhopal RS, et al. Identifying Chinese populationsin the UK for epidemiological research: experience of a name analysis of unlikely to be an explanation for the different cardiovascular the FSHA register. Public Health 1997; 111: 331–37.
disease rates between ethnic groups. Also, we measured Allain CC, Poon LS, Chan CSG, Richmond W, Fu PC. Enzymatic atherosclerosis by use of non-invasive methods, since determination of total serum cholesterol. Clin Chem 1974; 20: 470–75.
angiography is generally impractical in a population-based 10 McGowan MW, Artiss JD, Strandberg DR, Zak B. A peroxidase coupled method for the colorimetric determination of serum triglycerides. Clin study of healthy people. Our assumption that carotid Chem 1983; 29: 538–42.
atherosclerosis is a reasonable surrogate for coronary 11 Neeley WE. Simple automated determination of serum or plasma atherosclerosis is in accordance with several large studies.26,30,31,32 glucose by a hexokinase/glucose-6-phosphate dehydrogenase method.
This pattern of association was also seen overall in our study Clin Chem 1972; 18: 509–15.
12 Friedewald WT, Levy RI, Fredickson DS. Estimation of the and within each ethnic group. Therefore, we believe carotid concentration of low density lipoprotein cholesterol in plasma, without atherosclerosis as measured by the mean maximum IMT is a use of the preparative ultracentrifuge. Clin Chem 1972; 18: 499–502.
good marker of generalised atherosclerosis.
13 Warnick GR, Nguyen T, Albers AA. Comparison of improved Given that South Asians and Chinese represent over half precipitation methods for quantification of high-density lipoprotein
cholesterol. Clin Chem 1985; 31: 217–22.
of the world’s population and have increasing rates of 14 Fortin L, Genest J. Measurement of homocysteine in the prediction of cardiovascular disease,33 identifying the unique risk factors of arteriosclerosis. Clin Biochem 1995; 28: 155–62.
these ethnic groups is important in development tailored 15 Eriksson E, Ranby M, Gyzand E. Determination of plasminogen strategies for the prevention of cardiovascular disease.
activator inhibitor in plasma using t-PA and chromogenic single point
poly-D-lysine stimulated assay. Thromb Res 1988; 50: 91–101.
16 Clauss A. Gerinnungsphysiologische schnellmethode zur bestimmung The principal investigators were S Yusuf and S Anand. P Montague was the des fibrinogen. Acta Haematol 1957; 17: 237–47.
overall study coordinator. L Kelemen was the nutrition coordinator.
17 Lonn EM, Yusuf S, Doris CI, et al for the SECURE Investigators. Study Statistics were done by T Guo and C Yi. J C Villar read ECGs and design and baseline clinical characteristics of the study to evaluate carotid S Rangarajan, S Chin, J Vayalumkal, A Doobay, and J Bosch managed the ultrasound changes in patients treated with ramipril and Vitamin E: data. Administration was done by R Mayhew and B Cracknell.
SECURE. Am J Cardiol 1996; 78: 914–19.
18 Furberg CD, Adams HP Jr, Applegate WB, et al. Effect of lovastatin on Hamilton General Hospital, Hamilton, Ontario, Canada. Investigators: early carotid atherosclerosis and cardiovascular events: asymptomatic S Anand, H Gerstein, E Lonn, M McQueen, S Yusuf. Study co-ordinator: Carotid Artery Progression Study (ACAPS) Research Group. Circulation P Montague. Ultrasound personnel: S Smith, A Moore-Cox.
1994; 90: 1679–87.
University of Alberta Hospitals, Edmonton, Alberta, Canada. Investigator: 19 Crouse JR, Byington RP, Bond MG, et al. Pravastatin, lipids, and K K Teo. Study co-ordinators: P Chan, L Hill. Ultrasound personnel: atherosclerosis in the carotid arteries (PLAC-II). Am J Cardiol 1995; 75:
St Michael’s Hospital, Toronto, Ontario, Canada. Investigators: 20 Byington RP, Miller ME, Herrington D, et al. Rationale, design and S Devanesen, V Vuksan. Study co-ordinators: S Hamad, U Zdravkovic.
baseline characteristics of the Prospective Randomized Evaluation of the Vascular Effects of Norvasc Trial (PREVENT). Am J Cardiol 1997; 80:
Hamilton General Hospital, Hamilton, Ontario, Canada. M McQueen 21 Blackburn H, Prineaqs RJ, Crow RS. The Minnesota Code: manual for (director), D L Nazir. Technicians: S Damjanovic, J Frisina, M Popp, electrocardiographic findings—standards and procedures for measurement and classification. 1982. Littleton, Mass, Wright.
Hemostasis LaboratoryHenderson General Hospital, Hamilton, Ontario, Canada. M Johnston 22 Armitage P, Berry G. Statistical methods in medical research, 3rd ed, Oxford: Blackwell Scientific publications, 1994.
23 Anderson K, Wilson P, Odell P, Kannel W. An updated coronary risk Sampling methodology: C Nair, M Nargundkar, Statistics: Canada.
profile: a statement for health professionals—AHA Medical/Scientific Genetics: R Hegele, Robarts Research Institute. Nutrition Consultants: Statement. Circulation 1991; 83: 356–62.
W Willett, J Hankin, M Jain Merck Frosst: T Montague, G Nakhele.
24 Ross R. Atherosclerosis: an inflammatory disease. N Engl J Med 1999; 340: 115–26.
AcknowledgmentsThis study was funded by Medical Research Council of Canada (MRC 25 Fuster V, Badimon L, Badimon J, Cheesebro J. The pathogenesis of Grant No. MT-12790, Merck Frosst Canada). S Anand was supported by a coronary artery disease and the acute coronary syndromes.
Heart and Stroke Foundation of Ontario Research Fellowship during the N Engl J Med 1992; 326: 242–50.
study and is a recipient of a Medical Research Council of Canada Clinician 26 Chambless L, Heiss G, Folsom A, et al. Association of coronary heart Scientist Award. S Yusuf is a recipient of a Medical Research Council of disease incidence and carotid arterial wall thickness and major risk Canada Senior Scientist Award and holds a Heart and Stroke Foundation of factors: The atherosclerosis risk in communities (ARIC) study Ontario Research Chair. K Teo has a Medical Scholarship from Alberta 1987–1993. Am J Epidemiol 1997; 146: 483–94.
Heritage Foundation for Medical Research. R A Hegele is a recipient of 27 Libby P. The interface of atherosclerosis and thrombosis: basic Career Investigator Award of the Heart and Stroke Foundation of Ontario.
mechanisms. Vascular Medicine 1998; 3: 225–29.
Grant support was from the Medical Research Council of Canada # MT- 28 Cappuccio F, Cook D, Atkinson R, et al. Prevalence, detection, and management of cardiovascular risk factors in different ethnic groups in
south London. Heart 1997; 78: 555–63.
29 Bhopal R, Unwin N, White M, et al. Heterogeneity of coronary heart McKeigue PM, Shah B, Marmot MG. Relation of central obesity and disease risk factors in Indian, Pakistani, Bangladeshi and European insulin resistance with high diabetes prevalence and cardiovascular risk in populations: cross-sectional study. BMJ 1999; 319.
South Asians. Lancet 1991; 337: 382–86.
30 O’Leary D, Polak J, Kronman R, Manolio T, Burke G, Wolfson S, for McKeigue PM, Ferrie JE, Pierpoint T, Marmot MG. Association of the Cardiovascular Health Study Collaborative Research Group.
early-onset coronary heart disease in South Asian men with glucose Carotid-artery intima and media thickness as a risk factor for intolerance and hyperinsulinemia. Circulation 1993; 152–61.
myocardial infarction and stroke in older adults. N Engl J Med 1999; Harland J, Unwin N, Bhopal R, et al. Low levels of cardiovascular risk 340: 14–22.
factors and coronary heart disease in a UK Chinese population.
31 Hodis H, Mack W, LaBree L, Selzer R, Liu C, L C, Azen S. The role of J Epidemiol Community Health 1997; 51: 636–42.
the carotid arterial intima-media thickness in predicting clinical coronary Woo KS, Chook P, Young RP, Sanderson JE. New risk factors for events. Ann Intern Med 1998; 128: 262–69.
coronary heart disease in Asia. Int J Cardiol 1997; 62 (suppl 1): 639–42.
32 Bots M, Hoes A, Koudstaal P, et al. Common carotid intima-media Sheth T, Nair C, Nargundkar M, Anand S, Yusuf S. Cardiovascular and thickness and risk of stroke and myocardial infarction. The Rotterdam cancer mortality among Canadians of European, South Asian, and Study. Circulation 1997; 96: 1432–37.
Chinese origin from 1979 to 1993: an analysis of 1.2 million deaths. Can 33 Reddy KS, Yusuf S. Emerging epidemic of cardiovascular in the Med Assoc J 1999; 161: 132–38.
developing countries. Circulation 1998; 97: 596–601.
For personal use only. Not to be reproduced without permission of The Lancet.

Source: http://idocc.ca/Guideline/References/10_Ethnic_Differences.pdf