Mexican Americans have exhibited increases in various coronary heart disease risk factors compared to European Americans but have also had reportedly lower coronary heart disease mortality from vital statistics studies. We hypothesized this apparent paradox might relate to lower levels of subclinical disease in Mexican Americans. A total of 105 adult Mexican Americans (42 men and 63 women, age 46 ± 14 years) and 100 European Americans (59 men and 41 women, age 50 ± 11 years) were studied using blood tests, transthoracic echocardiography, and computed tomography coronary artery calcium (CAC) scans. Despite a greater body mass index and triglycerides in Mexican Americans (p <0.001), the Mexican Americans demonstrated less subclinical disease than did the European Americans (14.4% vs 25.7% with CAC scores >0, p <0.05 and mean left ventricular mass [LV] of 146 vs 160 g, p <0.05). Also, the LV mass was significantly greater in Mexican Americans with than in those without CAC (mean 172 vs 140 g, p <0.05). On logistic regression analysis, age and diastolic blood pressure were associated with an increased likelihood of CAC (p <0.001 and p <0.01, respectively), and Mexican-American ethnicity was associated with a decreased likelihood of CAC (odds ratio 0.33, 95% confidence interval 0.12 to 0.87, p <0.05). On multiple regression analysis, male gender, body surface area, and systolic blood pressure were independently associated with an increased LV mass (all p <0.001). The body mass index was less strongly related to the LV mass than was the body surface area and was not related to CAC. In conclusion, Mexican-American ethnicity is associated with both a lower LV mass and a lower prevalence of CAC, although the differences in LV mass did not remain after adjustment for other factors. Although systolic blood pressure, body surface area, and male gender were most strongly associated with the LV mass, age and diastolic blood pressure, in addition to Mexican-American ethnicity, were the most important indicators of CAC.
Limited data are available on the natural history of coronary heart disease (CHD) and other cardiovascular diseases (CVDs) among United States Hispanics, the nation’s second largest minority group, of which Mexican Americans constitute a substantial segment. Other studies have demonstrated that despite increased rates of obesity, hyperlipidemia, and diabetes in Mexican Americans, the CHD mortality rates from vital statistic studies are lower, or at least not greater than, among the non-Hispanic United States populations. The Corpus Christi Heart Project reported a greater incidence of both hospitalized myocardial infarction and 28-day case-fatality rates among Mexican Americans than among European Americans. These investigators suggested that their finding of increased CHD risk factor levels and greater hospitalized myocardial infarction and case-fatality rates—in the face of reported lower CHD mortality rates in the vital statistics studies among Mexican Americans—was likely due to a misclassification of the cause of death and/or ethnicity on the death certificates. We hypothesized that this apparent paradox might relate to lower levels of subclinical disease phenotypes (e.g., left ventricular [LV] mass, as measured by echocardiography, and coronary artery calcium [CAC], as measured by computed tomographic [CT] scanning)—for a given risk factor burden in Mexican Americans. Two specific hypotheses were that significant differences are present in the prevalence or extent of CAC and LV mass (subclinical disease phenotype variables) between Mexican Americans and European Americans; and that differences in the levels of traditional CHD risk factors between Mexican Americans and European Americans are related to, but do not fully explain, the differences in these subclinical CVD phenotypes.
Methods
The present study was conducted among asymptomatic, community-based adult volunteers who were neither selected nor self-referred on the basis of any risk factor. These Mexican-American and European-American volunteers were recruited by printed advertisements and word-of-mouth—including through churches in the Mexican-American community. A total of 105 adult Mexican Americans (42 men and 63 women, age 46 ± 14 years) and 100 European Americans (59 men and 41 women, age 50 ± 11 years) were studied using blood tests, transthoracic echocardiography, and CT CAC scanning. The subjects were generally healthy; those with known CVD or taking cardiovascular or blood pressure medications were excluded. All subjects provided informed consent before enrollment in the study.
The present study followed the echocardiographic protocol designed by our laboratory for the National Heart, Lung, and Blood Institute–sponsored Cardiovascular Health Study (CHS) and the Coronary Artery Risk Development in Young Adults (CARDIA) study. In brief, 2-dimensionally–directed M-mode measurements of the LV and left atrium were made according to conventions established by the American Society of Echocardiography. The LV mass was derived from the formula described by Devereux et al : LV mass (grams) = 0.80 × 1.04 [(VSTd + LVIDd + PWTd) 3 – (LVIDd) 3 ] + 0.6, where VSTd is the ventricular septal thickness at end-diastole, LVIDd is the LV internal dimension at end-diastole, and PWTd is the LV posterior wall thickness at end-diastole. The LV mass was normalized for various body size measures, including height. Data previously published from our laboratory (in the CARDIA Study) on the technical components of variability for LV mass measurements included intrareader variability (coefficient of variation 8% and inter-reader variability 14%) (6).
CT studies were performed using either an Imatron C-100 electron beam CT scanner (n = 27, 14 Mexican Americans and 13 European Americans) in California or a 16-slice CT scanner (GE LightSpeed, GE Medical Systems, Milwaukee, Wisconsin; n = 178, 91 Mexican Americans and 87 European Americans) in Detroit, Michigan. The CAC scores were measured in Hounsfield units, as previously described.
Fasting venous blood samples were obtained for biochemical analysis of the serum electrolytes, creatinine, glucose, and fasting total, high-density lipoprotein, and low-density lipoprotein cholesterol and triglycerides. A spot urine sample was collected for albumin and creatinine determinations.
The data are presented as the mean ± SD or percentages. The CAC scores were analyzed as a dichotomized measure (present, CAC score >0; absent, CAC score = 0). Bivariate comparisons between the 2 ethnicities were performed using Student’s t test for continuous variables and the chi-square test for categorical variables. Multivariate analyses were used to determine which factors, including ethnicity, were independently associated with (1) the likelihood of CAC, using stepwise multiple logistic regression analysis, and (2) the extent of LV mass, using stepwise multiple linear regression analysis. Variables were selected for entry if p <0.15 on bivariate analysis.
Results
The study participant demographic, risk factor, and subclinical disease characteristics by ethnic group and gender are presented in Table 1 . The body mass index and triglycerides were significantly greater in the overall Mexican-American cohort than in the overall European-American cohort and in the Mexican-American men and Mexican-American women, considered separately, compared to the European-American men and European-American women. In addition, the glucose level was significantly greater in Mexican-American women than in European-American women. In contrast, age, height, low-density lipoprotein cholesterol, and LV mass were all significantly lower in the overall Mexican-American than in the overall European-American cohort. However, no significant difference was found between the Mexican Americans and European Americans (overall cohort) when the LV mass was indexed to body height or body surface area (data not shown). In addition, age and smoking prevalence were lower in the Mexican-American than in the European-American women, and low-density lipoprotein cholesterol was lower in Mexican-American men than in European-American men.
| Parameter | Total | Men | Women | |||
|---|---|---|---|---|---|---|
| MA (n = 104) | EA (n = 101) | MA (n = 39) | EA (n = 59) | MA (n = 65) | EA (n = 42) | |
| Age (years) | 47.1 ± 13.5 | 49.6 ± 10.8 ⁎ | 47.1 ± 13.5 | 47.7 ± 11.3 | 47.2 ± 13.6 | 52.3 ± 9.4 ⁎ |
| Height (inches) | 64.1 ± 3.8 | 67.6 ± 3.9 † | 67.2 ± 2.9 | 70.2 ± 2.2 † | 62.2 ± 3.0 | 63.9 ± 2.4 ‡ |
| Weight (pounds) | 178.8 ± 46.3 | 181.5 ± 36.9 | 198.1 ± 46.1 | 199.5 ± 27.7 | 167.2 ± 42.8 | 156.3 ± 33.4 |
| Body mass index (kg/m 2 ) | 30.6 ± 6.9 | 27.7 ± 4.2 † | 30.7 ± 5.8 | 28.4 ± 3.5 ⁎ | 30.5 ± 7.5 | 26.8 ± 4.9 ‡ |
| Triglycerides (mg/dl) | 182 ± 122 | 124 ± 67 † | 185 ± 94 | 136 ± 71 ‡ | 181 ± 137 | 108 ± 6 † |
| Low-density lipoprotein cholesterol (mg/dl) | 112 ± 29 | 123 ± 31 ‡ | 114 ± 28 | 128 ± 28 ⁎ | 110 ± 29 | 117 ± 34 |
| High-density lipoprotein cholesterol (mg/dl) | 54 ± 13 | 54 ± 13 | 46 ± 9 | 49 ± 11 | 58 ± 14 | 62 ± 13 |
| Total cholesterol (mg/dl) | 201 ± 35 | 203 ± 37 | 197 ± 33 | 206 ± 38 | 204 ± 36 | 200 ± 37 |
| Glucose (mg/dl) | 91 ± 14 | 89 ± 22 | 91 ± 13 | 91 ± 27 | 90 ± 14 | 85 ± 11 ⁎ |
| Systolic blood pressure (mm Hg) | 126 ± 17 | 125 ± 14 | 129 ± 13 | 127 ± 13 | 124 ± 19 | 122 ± 15 |
| Diastolic blood pressure (mm Hg) | 72 ± 11 | 75 ± 10 | 79 ± 10 | 77 ± 10 | 68 ± 10 | 71 ± 8 |
| Current/previous smokers | 47.1% | 64.4% | 56.1% | 62.7% | 41.5% | 66.7 ⁎ |
| Coronary artery calcium score >0 | 14.4% | 25.7% ⁎ | 20.5% | 30.5% | 10.8% | 19.1% |
| Left ventricular mass (g) | 145.5 ± 43.9 | 160.2 ± 49.9 ⁎ | 181.5 ± 41.2 | 185.3 ± 46.5 | 124.4 ± 29.5 | 124.9 ± 28.8 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
