Globally, cerebrovascular disease (CVD), which includes coronary heart disease (CHD), strokes, and peripheral arterial diseases, is the leading cause of death and is a major disease burden. Although age adjusted CVD death rates have declined in several Western countries in the last decades, CVD rates have markedly increased in low- and middle-income countries and now approximately 80% of the burden occurs in these countries. Several factors contribute to the trend toward increasing rates of CVD in developing countries. First, decreasing mortality rates from acute infectious disease have resulted in increased life expectancy, which in turn leads to a higher proportion of individuals reaching middle and old age, ages at which they are at risk of atherothrombotic events. Second, lifestyle and socioeconomic changes associated with urbanization may lead to higher levels of risk factors for CVD. Third, some populations or individuals (e.g., because of lifestyle or genes) may be particularly susceptible to the adverse effects of a particular environmental factor; for example, South Asians appear to develop diabetes more often than individuals of European descent, which may be the result of an inherent susceptibility or perhaps of unique diets, which include highly refined carbohydrates.
Current knowledge about CHD and CVD prevention is primarily derived from studies involving Western and European populations. It is unclear to what extent the findings from these studies apply globally. Some studies have suggested that risk factors vary in their relationship to CHD in different populations, but these findings may not be reliable, given the small numbers of events in each study and their variable methodologies. Furthermore, even if the relationship (e.g., odds ratio [OR]) of a risk factor to CHD is similar across populations, the prevalence of risk factors vary, resulting in differing population to attributable risks (PARs). Therefore, a large study using standardized methods, aimed at discovering the relationship between risk factors and CHD in a number of countries, and representing different regions and ethnic groups, was found to be necessary.
Methodology
Participants in the Study
The INTERHEART study is a large-scale, standardized, case-control study involving about 28,000 cases (with first myocardial infarction) and matched controls from 52 countries from every inhabited continent. Study participants were recruited from 262 centers from 52 countries in Asia, Europe, the Middle East, Africa, Australia, North America, and South America. The choice of countries for INTERHEART was based on a balance between a desire to represent each major region of the world and feasibility.
All patients presenting to the coronary care unit or its equivalent with first myocardial infarction (MI) within 24 hours of symptom onset were eligible for enrollment. Patients were eligible if they had characteristic symptoms plus electrocardiographic changes indicative of a new MI. Patients with cardiogenic shock or a significant chronic medical illness were excluded. Controls were age-matched (up to 5 years older or younger) and gender-matched to each case and recruited from the hospital or surrounding community. Persons with significant chronic medical illness or a previous diagnosis of heart disease or history of chest pain on exertion were excluded.
Data Collection
A questionnaire was administered in the hospital by trained staff, who collected information about demographic factors, socioeconomic status (education, income), lifestyle (smoking, leisure time, physical activity, dietary patterns), psychosocial factors (depression, locus of control, perceived stress, life events), personal and family history of cardiovascular disease, and risk factors (hypertension, diabetes mellitus). Height, weight, waist and hip circumferences, and heart rate were determined by a standardized protocol. Although blood pressure at the time of examination was recorded in both cases and controls, these levels would be systematically affected by MI and any blood pressure–lowering treatment that had been administered. Therefore, only self-reported history of hypertension was used in the analysis.
Nonfasting blood samples (20 mL) were drawn from every individual and centrifuged within 2 hours of admission, separated into six equal volumes, and frozen immediately at −20° C or −70° C after processing. They could be obtained within 24 hours of symptom onset in two thirds of cases.
Data on smoking were missing in 1.1% of participants, hypertension in 0.6%, diabetes in 0.7%, psychosocial variables in 11%, physical activity in 1.1%, diet in 2.1%, and waist and hip measurements in 3.5%. Blood samples were available in 21,508 of 27,098 cases (79%) and controls.
Definitions
We defined current smokers as individuals who had smoked any tobacco in the previous 12 months and included those who had quit within the past year. Former smokers were defined as those who had quit more than 1 year earlier. Tertiles for waist-to-hip ratio were calculated using cut-off values of 0.90 and 0.95 in men and 0.83 and 0.90 in women. Deciles and Quintiles of the ratio of apolipoprotein B (apo B)–to–apolipoprotein A1 (apo A1) were calculated using cut-off values derived from all controls (men and women). Individuals were judged to be physically active if they were regularly involved in moderate (walking, cycling, or gardening) or strenuous exercise (jogging, football, vigorous swimming) for 4 hours or more per week. Regular alcohol use was defined as consumption three or more times per week.
Statistical Methods and Analyses
The sample size calculation for INTERHEART took into consideration the sample size requirements for each participating country or region—smaller countries that are similar were clustered together. The main analyses of this study used models fitted with unconditional logistic regression, adjusted for the matching criteria. The methods used here for analyses were checked against other methods for general agreement of key results.
All statistical tests of hypotheses were two-sided. PARs and 99% (or 95%) confidence interval (CI) were calculated using the Interactive Risk Attributable Program (IRAP) software (from the National Cancer Institute, 2002). For variance estimates, the method by Engel and colleagues was used to assist in calculating CI after a logit transformation.
Results
Following exclusion of cases not meeting inclusion criteria or with insufficient data, 12,461 cases and 14,637 controls were included in the analysis; 9459 cases (76%) and 10,851 controls (74%) were male. The overall median age of cases with first acute MI was about 9 years lower in men than in women overall; the proportion of male cases was highest in countries with a younger age of presentation of acute myocardial infarction (AMI). The highest proportion of cases with first AMI at age 40 years or younger was in men from the Middle East (12.6%), Africa (10.9%), and South Asia (9.7%) and the lowest proportion was in women from China and Hong Kong (1.2%), South America (1.0%), and central and eastern Europe (0.9%).
Association of Separate and Combined Risk Factors with Myocardial Infarction
Eight risk factors (abnormal lipid levels, smoking, hypertension, diabetes, abdominal obesity, psychosocial factors, lack of consumption of fruits and vegetables and of regular physical activity) were significantly related to AMI ( P < .0001), except alcohol, which had a weaker association ( P = .03). After multivariate analyses, current smoking and an increased apo B-to-apo A1 ratio (top vs. lowest quintile) were the two strongest risk factors, followed by history of diabetes, hypertension, and psychosocial factors ( Table 2-1 ).
| Risk Factor | Prevalence | OR (99% CI) Adjusted for Age, Gender, and Smoking (OR 1) | PAR (99% CI) | OR (99% CI) Adjusted for All Other Risk Factors (OR 2) | PAR 2 (99% CI) | |
|---|---|---|---|---|---|---|
| % of Controls | % of Cases | |||||
| Current smoking * | 26.76 | 45.17 | 2.95 (2.72-3.20) | — | 2.87 (2.58-3.19) | — |
| Current + former smoking * | 48.12 | 65.19 | 2.27 (2.11-2.44) | 36.4 (33.9-39.0) | 2.04 (1.86-2.25) | 35.7 (32.5-39.1) |
| Diabetes | 7.52 | 18.45 | 3.08 (2.77-3.42) | 12.3 (11.2-13.5) | 2.37 (2.07-2.71) | 9.9 (8.5-11.5) |
| Hypertension | 21.91 | 39.02 | 2.48 (2.30-2.68) | 23.4 (21.7-25.1) | 1.91 (1.74-2.10) | 17.9 (15.7-20.4) |
| Abdominal obesity (2 v 1) † | 33.40 | 30.21 | 1.36 (1.24-1.48) | — | 1.12 (1.01-1.25) | — |
| Abdominal obesity (3 v 1) † | 33.32 | 46.31 | 2.22 (2.03-2.42) | 33.7 (30.2-37.4) | 1.62 (1.45-1.80) | 20.1 (15.3-26.0) |
| All psychosocial factors ‡ | — | — | 2.51 (2.15-2.93) | 28.8 (22.6-35.8) | 2.67 (2.21-3.22) | 32.5 (25.1-40.8) |
| Vegetables and fruits daily * | 42.36 | 35.79 | 0.70 (0.64-0.77) | 12.9 (10.0-16.6) | 0.70 (0.62-0.79) | 13.7 (9.9-18.6) |
| Exercise * | 19.28 | 14.27 | 0.72 (0.65-0.79) | 25.5 (20.1-31.8) | 0.86 (0.76-0.97) | 12.2 (5.5-25.1) |
| Alcohol intake * | 24.45 | 24.01 | 0.79 (0.73-0.86) | 13.9 (9.3-20.2) | 0.91 (0.82-1.02) | 6.7 (2.0-20.2) |
| Apo B-to-apo A1 ratio (2 vs. 1) § | 19.99 | 14.26 | 1.47 (1.28-1.68) | — | 1.42 (1.22-1.65) | — |
| Apo B-to-apo A1 ratio (3 vs. 1) § | 20.02 | 18.05 | 2.00 (1.74-2.29) | — | 1.84 (1.58-2.13) | — |
| Apo B-to-apo A1 ratio (4 vs. 1) § | 19.99 | 24.22 | 2.72 (2.38-3.10) | — | 2.41 (2.09-2.79) | — |
| Apo B-to-apo A1 ratio (5 vs. 1) § | 20.00 | 33.49 | 3.87 (3.39-4.42) | 54.1 (49.6-58.6) | 3.25 (2.81-3.76) | 49.2 (43.8-54.5) |
| All above risk factors combined ¶ | — | — | 129.20 * (90.24-184.99) | 90.4 (88.1-92.4) * | 129.20 * (90.24-184.99) | 90.4 (88.1-92.4) * |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
