Racial/Ethnic Differences in Subclinical Atherosclerosis Among Adults With Diabetes: The Multiethnic Study of Atherosclerosis

MESA is a multicenter study of 6,814 men and women aged 45–84 who were free from clinical CVD in 2000–2002 (8). From the cohort of 6,814 participants, we excluded 25 participants without glucose measurements and 5,761 without diabetes. Our final analysis sample included 204 whites, 387 blacks, 311 Hispanics, and 126 Chinese with diabetes. The prevalence of diabetes in the cohort by race/ethnicity is comparable to that reported in a general U.S. population survey (9).

Demographics, clinical characteristics, and health behavior data were collected in a standardized manner at all study sites by trained personnel; blood assays were processed at central laboratories (10,11). Diabetes was defined according to measured fasting glucose of at least 7.0 mmol/l (126 mg/dl), self-reported previous physician diagnosis, or use of diabetes medication. Computed tomography was performed using a cardiac-gated electron-beam computed tomography scanner or a prospectively electrocardiogram-triggered scan acquisition multidetector computed tomography system over phantoms of known physical calcium concentrations. High-resolution B-mode ultrasonography was used to capture images of the bilateral common carotid and internal carotid arteries using a Logiq 700 ultrasound machine (General Electric Medical Systems) at all centers (10). ABI, the ratio of ankle to arm blood pressure, was computed separately for each leg using a Doppler probe (12). Peripheral artery disease (PAD) was defined as ABI <0.90 in either ankle (13). All analyses were conducted using SAS version 9.0 (SAS Institute, Cary, NC).

The mean age and BMI of the sample were 67 years and 30.6 kg/m², respectively; 65% of participants were using diabetes control medications, ranging from a high of 68% in black and Hispanic participants to 55% in whites. Hypertension and treated hypercholesterolemia were present in 66 and 26% of the sample, respectively.

The highest prevalence of measurable CAC was in whites (78%), followed by Chinese (68%), Hispanics (58%), and blacks (54%) (overall 62%). CAC >400 (17% of the total) followed a similar pattern of prevalence with 24, 20, 15, and 14% of white, Chinese, Hispanic, and black participants, respectively, affected. Following adjustment for age, sex, high school education, health insurance, current alcohol intake, cigarette smoking, physical activity, hypertension, total and HDL cholesterol, cholesterol medication use, and BMI in a multivariable logistic regression model, the odds of having any CAC were lower in blacks (OR [95% CI] 0.32 [0.2–0.51]; P < 0.0001) and Hispanics (0.37 [0.23–0.60]; P < 0.0001) compared with whites with diabetes, but did not differ between Chinese and whites (0.66 [0.36–1.22]; P = 0.18). Similar patterns were present for CAC >400. Although prevalence of PAD varied by ethnicity (9.5, 8.5, 6.4, and 5% of black, white, Chinese, and Hispanic participants, respectively), the age- and sex-adjusted odds of having PAD comparing blacks (1.26 [0.67–2.37]; P = 0.68), Hispanics (0.59 [0.28–1.23]; P = 0.13), and Chinese (0.65 [0.26–1.59]; P = 0.29) to whites did not differ significantly from 1.0. Results were unchanged following adjustment for duration of diabetes control medications (a surrogate for diabetes duration).

Among participants with any measurable CAC, geometric mean scores calculated from linear regression models and compared using ANOVA did not differ significantly across ethnicities. Lower IMT scores were noted among Chinese compared with whites, and ABI values were lower among blacks compared with whites and other ethnicities (Table 1).

In this study of asymptomatic men and women without a prior diagnosis of atherosclerotic disease, we found excess atherosclerosis in whites with diabetes compared with blacks and Hispanics that was restricted to calcification in the coronary arteries. However, among those participants with measurable CAC, the levels did not differ significantly by ethnicity, which suggests that the burden among those with some calcification was similar. Our finding of smaller IMT among Chinese compared with other ethnicities is consistent with studies of clinical outcomes of persons with diabetes that demonstrate lower risks of myocardial infarction in Asians with diabetes versus whites (14). Conflicting results in Chinese for IMT and CAC indicate that further research is needed in this ethnic group.

The finding that blacks with diabetes have lower ABI than whites and Hispanics is consistent with findings in this population that were not restricted to persons with diabetes (7) and reports of PAD prevalence in other samples (6,7,11,15). One explanation for the absence of a statistically significant finding between ethnicity and PAD is that clinical PAD is often present with CVD (6). Although the mean age in this sample is similar to the age of onset of PAD, participants in this study were chosen specifically because they did not have clinical CVD. Consequently, the overall prevalence of PAD in this sample is low, and those with PAD did not have CVD comorbidities, which may contribute to ethnic disparities in PAD in other population samples.

This population-based study of adults with diabetes who are free from clinical CVD describes racial/ethnic variations in the presence of atherosclerosis across different vascular beds that are consistent with previous reports in this sample (1) and most others (2,3,4,5) that are not restricted to persons with diabetes. Further research to explore how these ethnic differences in subclinical atherosclerosis relate to clinical outcomes in persons with diabetes is warranted.

This research was supported by contracts N01-HC-95159 through N01-HC-95166 and N01-HC-95169 from the National Heart, Lung, and Blood Institute. M.R.C. was supported in part by a career development award from the National Heart Lung and Blood Institute (1K01 HL73249-01).