Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Obesity (Silver Spring). Author manuscript; available in PMC 2010 October 1.
Published in final edited form as:
PMCID: PMC2920341

Regional Adipose Tissue Associations With Calcified Atherosclerotic Plaque: African American–Diabetes Heart Study


Coronary artery calcified atherosclerotic plaque (CP) is strongly associated with nonsubcutaneous adipose tissue, particularly pericardial adipose tissue (PAT), in community-based studies. We tested for relationships between regional adipose tissue depots and CP in African Americans with longstanding type 2 diabetes. Infrarenal aorta, coronary, and carotid artery CP and pericardial, visceral, intermuscular, and subcutaneous organ-specific adipose tissue volumes were measured using single and multidetector computed tomography (CT) in 422 African Americans with type 2 diabetes. Generalized estimating equations using exchangeable correlation and the sandwich estimator of the variance were used to test for associations between CP and adipose tissue depots. Mean (s.d.) age was 56.5 (7.6) years, diabetes duration 10.3 (7.6) years, PAT 85.3 (36.1) cm3/45 mm and visceral adipose tissue (VAT) 174.9 (70.1) cm3/15 mm. Adjusting for age, gender, BMI, blood pressure, medications, proteinuria, smoking, lipids, and 25-hydroxyvitamin D, PAT was positively associated with the presence (P = 0.009) and quantity of coronary artery CP in African Americans (P = 0.004), as well as the quantity of infrarenal aorta CP (P = 0.004). As in European Americans, PAT is associated with CP in African Americans with type 2 diabetes. Ethnic differences in the relationships between organ-specific adipose tissue depots and atherosclerosis require further study.


Markedly lower levels of coronary artery calcified atherosclerotic plaque (CP) are present in African Americans with and without type 2 diabetes mellitus, relative to European Americans (15). These consistent ethnic differences occur despite exposure to similar or more severe conventional cardiovascular disease risk factors in African Americans, suggesting that risk factors may have differential impacts on atherosclerosis between races (6). However, both presence and amount of coronary artery CP are associated with increased risk of cardiovascular events in individuals from all ethnic groups (7).

Pericardial adipose tissue (PAT) volumes were associated with presence of coronary artery CP in 227 European American and 171 African American Multi-Ethnic Study of Atherosclerosis (MESA) participants, although only 84 MESA subjects had detectable CP (8). PAT was also associated with incident coronary disease in 998 MESA participants (220 were African American) without baseline cardiovascular disease (9). These findings support the theory that local effects of adipose tissue directly impact atherosclerotic risk in adjacent arteries. Pericardial and visceral adipose associations with CP have not been performed in large numbers of African Americans or in those with diabetes. Furthermore, few studies have explored the relationship of regional or total adipose tissue measures with CP in the carotid artery or abdominal aorta.

The African American–Diabetes Heart Study (DHS) is exploring the inherited and environmental causes of CP in African Americans with type 2 diabetes. PAT, visceral adipose tissue (VAT), intermuscular adipose tissue (IMAT) and subcutaneous adipose tissue (SAT) volumes were measured with computed tomography (CT) in African Americans with longstanding type 2 diabetes and relatively preserved kidney function. Relationships between organ-specific adipose tissue depots and coronary artery, carotid artery and aorta CP were computed.


Study populations

African Americans with type 2 diabetes recruited in the initial DHS and in the follow-up African American–DHS formed the study population (3). The original DHS recruited sibling pairs and used age-at-diabetes onset of 35 years, because most of the participants were European American. The African American–DHS was limited to unrelated African Americans and age-at-diabetes onset was 30 years, because type 2 diabetes is reported at younger ages in African Americans, relative to European Americans. Subjects who underwent prior coronary artery bypass surgery or carotid endarterectomy were excluded from these analyses as it was felt that the CP mass score in relevant arteries would be impacted by these procedures; those with prior myocardial infarction or stroke were included. The study was approved by the institutional review board at the Wake Forest University School of Medicine and all participants provided written informed consent.

Vascular imaging

Calcified atherosclerotic plaque was measured in the coronary, carotid, and infrarenal abdominal aorta arteries with single and multidetector CT systems using a standard electrocardiogram-gated CT scanning protocol based on those currently implemented in the National Heart Lung and Blood Institute’s MESA (10). The calcium mass score (SmartScore, General Electric (GE) Healthcare, Waukesha, WI), which accounts for the volume and density of CP on a pixel by pixel basis and is highly correlated with the standard Agatston score using a 90 Hounsefield Unit threshold, 0.5 mm2 minimum lesion size (two adjacent pixels), was employed. As noted by multiple authors, the Agatston scoring system has established limitations as relates to measurement error (11). Furthermore, the weighting factors used in the coronary arteries which are <4 mm in diameter, are not appropriate for the carotid arteries and abdominal aorta which are more than ten times greater in size. By using SI units of calcium mass we have a more stable measure which increases statistical validity, is superior for comparing between the vascular beds, and has the benefits of assessing the volume of CP while also taking into account the density or mass of calcification.

Adipose tissue imaging

PAT and VAT were measured from volumetric CT acquisitions to reduce variability related to slice location using the Volume Analysis software (Advantage Windows Workstation, GE Healthcare, Waukesha, WI) and a threshold of −190 to −30 as the definitions of adipose tissue containing tissue. PAT is the combined adipose tissue superficial (paracardial) and deep (epicardial) to the pericardium (12); however, the pericardium extends superiorly to encase the great vessels and inferiorly borders the diaphragm. Our methods for measuring PAT segments a volume for measurement that covers 45 mm in length along the z axis (cephalo-caudad) of the individual based on origin of the left main coronary such that it extends 15 mm above and 30 mm below. This PAT volume includes the majority of the coronary arteries and myocardium and excludes PAT located superiorly around the aorta and pulmonary arteries and adjacent to the abdomen, as reported (13).

In the abdomen, VAT, SAT, and IMAT were measured on abdominal CT scans with technical factors: helical mode, 120 kVp, 250 mA, 4 × 2.5 mm collimation, standard reconstruction kernel, and a display field of view of 500 mm. The landmark for analysis was the first lumbar disk above the lumbar-sacrum junction, most commonly designated as L4–L5. A volume 15 mm in z axis length of the abdomen was segmented for the subcutaneous, abdominal wall and intra-abdominal compartments. VAT was defined as the adipose tissue containing pixels located within the abdominal cavity, SAT was defined as the adipose tissue containing pixels between the skin surface and lean tissue of the abdominal wall and IMAT was measured within the abdominal wall and paraspinal muscles. Studies in human cadavers revealed that the area measured by CT is an accurate estimate of VAT, SAT, and IMAT volume.

Statistical methods

A series of generalized estimating equations assuming exchangeable correlation for the sibling pairs and using the sandwich estimator to estimate the variance of parameter estimates while adjusting for familial correlation was computed to test for associations between each adipose tissue depot and CP (14). The exchangeability assumption is expected to hold here since the clusters are made of only first degree relatives. The modified Poisson approach (15) was used for analyses that treat the CP variables as binary (presence vs. absence) traits, whereas an identity link was used for quantitative traits following the appropriate transformations. The choice of the modified Poisson approach is justified by the fact that the fraction of participants with detectable CP was high, making the rare disease assumption untenable. The Box-Cox method was applied to identify the appropriate transformation of each outcome variable that would best approximate the distributional assumptions of conditional normality and homogeneity of variance of the residuals (16). The natural log of (CorCP + 1), (CarCP + 1), and (AorCP + 1) were analyzed. Prior to these transformations, observed values of aorta, carotid and coronary CP exceeding the 95th percentile were winsorized at their 95th percentile.

All analyses were run adjusting for (i) age and gender, (ii) age, gender, and BMI, (iii) age, gender, and 25-hydroxyvitamin D, and (iv) age, gender, C-reactive protein, BMI, high-density lipoprotein cholesterol, systolic blood pressure, triglycerides, blood pressure, smoking history, lipid-lowering medications, and 25-hydroxyvitamin D. Standard regression diagnostics for collinearity and influence were computed for each model reported. All analyses satisfied the normality assumption made about the distribution of the residuals.


The study population consisted of 422 African Americans with type 2 diabetes mellitus from 323 families. Table 1 contains demographic characteristics of the study group and Table 2 lists laboratory results. Participants had mean (s.d.) PAT volume 85.3 (36.1) cm3/45 mm, VAT 174.9 (70.1) cm3/15 mm, IMAT 11.1 (7.4) cm3/15 mm, and SAT 437.1 (184.7) cm3/15 mm. The correlation between the adipose tissue depots was highly significant (spearman correlation = 0.63, P value < 0.001). However, this correlation translates to only a 40% (0.632) reliability estimate between the variables; therefore, it was possible to detect statistically significant effect with one variable, but not another. We next computed the partial correlation between log (coronary CP + 1) with intermuscular and PATs, accounting for age. The correlations were (0.16, P value = 0.001) for log (coronary CP + 1) with PAT and (0.05, P value = 0.29) for log (coronary CP + 1) with IMAT. This suggests that the initial correlation between log (coronary CP + 1) with intermuscular adipose was due to the strong correlation that exists between these variables and age (age appears to act as a confounding variable). The fact that the partial correlation adjusted for age is not significant explains why, despite the strong correlation between intermuscular and PAT, any model that included age as a covariate shows no statistically significant association between log (coronary CP + 1) and IMAT.

Table 1
Demographic characteristics of African American–Diabetes Heart Study participants
Table 2
Laboratory characteristics of African American–DHS study participants (mean (s.d.) median)

Table 3 contains results of analyses comparing presence or absence of CP in the coronary arteries, carotid arteries and aorta with regional adipose tissue. In the fully adjusted model (age, gender, BMI, C-reactive protein, BMI, high-density lipoprotein cholesterol, systolic blood pressure, triglycerides, blood pressure, smoking history, lipid-lowering medications, and 25-hydroxyvitamin D) a significant positive association between presence of coronary artery CP and PAT (parameter estimate 0.011; P = 0.009) was observed.

Table 3
Associations between presence/absence of CP and organ-specific adipose tissue volumes

Relationships between quantitative measures of CP and regional adipose tissue depots are presented in Table 4. PAT was strongly and positively associated with quantitative coronary CP (parameter estimate = 0.015, P = 0.004). PAT was also positively associated with aorta CP (parameter estimate = 0.014, P = 0.004). Significant relationships were not observed between other regional adipose tissue depots and CP using fully adjusted models.

Table 4
Associations between quantity of calcified plaque and organ-specific adipose tissue volumes


Adipose tissue has established autocrine, paracrine, and endocrine effects (17) and may directly influence the process of atherosclerosis through mechanisms related to release of adipocytokines and fatty acids (1820). This study extends previous reports by evaluating larger numbers of African Americans with type 2 diabetes, by including quantitative measures of CP from three vascular beds, and by including CT-derived adipose tissue volumes from four adipose tissue depots. A strong positive association was detected between PAT and both the presence and amount of coronary artery CP. In addition, PAT was significantly associated with the quantitative measure of infrarenal aorta CP.

This finding extends the initial reports of association between PAT and coronary artery CP (8,9,21). MESA participants lacked symptomatic coronary artery disease and generally had lower amounts of CP than the diabetes-affected subjects in this report. The positive relationship between PAT and coronary artery CP herein is similar to that observed in MESA subjects, where after adjustment for demographics, lifestyle factors and height, coronary artery CP was associated with a 1 s.d. increment in the PAT index (odds ratio 1.38, 95% confidence interval 1.04–1.84) (8). Furthermore, a preliminary MESA report limited to 398 participants with abdominal CT scans (171 African American; 227 European American) revealed VAT associations with coronary artery CP in European Americans (odds ratio 1.41; 95% CI 1.01–1.97), but not African Americans (odds ratio 1.17; 95% CI 0.75–1.83) suggesting that ethnic differences in development of CP based upon neighboring adipose tissue depots may exist (J. Ding, unpublished data). Interpretation of this preliminary finding will require analysis in the full MESA sample.

Despite higher mean BMI in African Americans compared to a random sample of 180 European Americans from the DHS, pericardial, visceral, intermuscular and SAT volumes were lower (data not shown) (22). It is tempting to speculate that deposition of adipose tissue in the pericardium, adipose tissue with a relatively high metabolic activity and cytokine secretion in close proximity to the coronary arteries, impacts the local development of CP. This report also reveals that PAT is associated with atherosclerosis in more distant vascular beds, such as the infrarenal aorta. VAT, nearer the aorta, was not associated with infrarenal aorta CP. Inflammatory mediators, leptin, adipo-nectin, and other as yet undetermined hormones/cytokines may prove to be associated with development of CP. The African American–DHS has a biorepository containing frozen serum, plasma and urine samples. Analyses of circulating factors and polymorphisms in the genes that regulate their synthesis will be performed once local recruitment is complete. These analyses may reveal the factor(s) underlying the strong association between PAT and systemic atherosclerosis.

As discussed, ethnic differences in CP have widely been observed, with African Americans at lower risk than European Americans despite presence of similar or more severe cardiovascular disease risk factors (15). It remains possible that VAT and PAT manifest ethnic differences in their effects on susceptibility to development of CP; this question will require additional analyses. An important limitation of this and the MESA reports are their cross-sectional nature. It will be important to quantify PAT and other regional adipose volumes using CT scans performed to measure CP in studies with longitudinal follow-up.

In conclusion, PAT and neighboring coronary artery CP were significantly and positively associated in African Americans with diabetes. In addition, significant positive associations were observed between PAT and infrarenal aorta CP. It will be important to determine whether ethnic differences in the relationships between adipose tissue depots and regional CP contribute to observed ethnic differences in CP mass, with African Americans having less CP relative to European Americans. The factors that mediate the relationship between PAT and atherosclerosis remain to be identified.


This study was supported in part by the General Clinical Research Center of the Wake Forest University School of Medicine grant M01 RR07122; NIDDK grant RO1 DK071891 (B.I.F.); NIAMS grant RO1 AR048797 (J.J.C.); and NHLBI grant R01 HL67348 (D.W.B.). The investigators acknowledge the cooperation of our participants, study recruiters, Carrie Smith and Cassandra Bethea, and CT analysts, Susan Pillsbury, Blair Gordy, and Melanie Wilder.



The authors declared no conflict of interest.


1. Bild DE, Detrano R, Peterson D, et al. Ethnic differences in coronary calcification: the Multi-Ethnic Study of Atherosclerosis (MESA) Circulation. 2005;111:1313–1320. [PubMed]
2. Newman AB, Naydeck BL, Whittle J, et al. Racial differences in coronary artery calcification in older adults. Arterioscler Thromb Vasc Biol. 2002;22:424–430. [PubMed]
3. Freedman BI, Hsu FC, Langefeld CD, et al. The impact of ethnicity and sex on subclinical cardiovascular disease: the Diabetes Heart Study. Diabetologia. 2005;48:2511–2518. [PubMed]
4. Budoff MJ, Nasir K, Mao S, et al. Ethnic differences of the presence and severity of coronary atherosclerosis. Atherosclerosis. 2006;187:343–350. [PubMed]
5. Carnethon MR, Bertoni AG, Shea S, et al. Racial/Ethnic differences in subclinical atherosclerosis among adults with diabetes: the multiethnic study of atherosclerosis. Diabetes Care. 2005;28:2768–2770. [PubMed]
6. Gaillard T, Schuster D, Osei K. Metabolic syndrome in Black people of the African diaspora: the paradox of current classification, definition and criteria. Ethn Dis. 2009;19:S2–S1. [PubMed]
7. Detrano R, Guerci AD, Carr JJ, et al. Coronary calcium as a predictor of coronary events in four racial or ethnic groups. N Engl J Med. 2008;358:1336–1345. [PubMed]
8. Ding J, Kritchevsky SB, Hsu FC, et al. Association between non-subcutaneous adiposity and calcified coronary plaque: a substudy of the Multi-Ethnic Study of Atherosclerosis. Am J Clin Nutr. 2008;88:645–650. [PMC free article] [PubMed]
9. Ding J, Hsu FC, Harris TB, et al. The association of pericardial fat with incident coronary heart disease: the Multi-Ethnic Study of Atherosclerosis (MESA) Am J Clin Nutr. 2009;90:499–504. [PubMed]
10. Detrano RC, Anderson M, Nelson J, et al. Coronary calcium measurements: effect of CT scanner type and calcium measure on rescan reproducibility–MESA study. Radiology. 2005;236:477–484. [PubMed]
11. Callister TQ, Cooil B, Raya SP, et al. Coronary artery disease: improved reproducibility of calcium scoring with an electron-beam CT volumetric method. Radiology. 1998;208:807–814. [PubMed]
12. Wheeler GL, Shi R, Beck SR, et al. Pericardial and visceral adipose tissues measured volumetrically with computed tomography are highly associated in type 2 diabetic families. Invest Radiol. 2005;40:97–101. [PubMed]
13. Ding J, Kritchevsky SB, Harris TB, et al. The association of pericardial fat with calcified coronary plaque. Obesity (Silver Spring) 2008;16:1914–1919. [PMC free article] [PubMed]
14. Zeger SL, Liang KY. Longitudinal data analysis for discrete and continuous outcomes. Biometrics. 1986;42:121–130. [PubMed]
15. Zou G. A modified poisson regression approach to prospective studies with binary data. Am J Epidemiol. 2004;159:702–706. [PubMed]
16. Box GEP, Cox DR. An analysis of tranformations. J R Stat Soc Ser B. 1964;26:211–46.
17. Kershaw EE, Flier JS. Adipose tissue as an endocrine organ. J Clin Endocrinol Metab. 2004;89:2548–2556. [PubMed]
18. Frayn KN. Adipose tissue as a buffer for daily lipid flux. Diabetologia. 2002;45:1201–1210. [PubMed]
19. Marchington JM, Mattacks CA, Pond CM. Adipose tissue in the mammalian heart and pericardium: structure, foetal development and biochemical properties. Comp Biochem Physiol, B. 1989;94:225–232. [PubMed]
20. Avogaro A, de Kreutzenberg SV. Mechanisms of endothelial dysfunction in obesity. Clin Chim Acta. 2005;360:9–26. [PubMed]
21. Greif M, Becker A, von Ziegler F, et al. Pericardial adipose tissue determined by dual source CT is a risk factor for coronary atherosclerosis. Arterioscler Thromb Vasc Biol. 2009;29:781–786. [PubMed]
22. Bowden DW, Lehtinen AB, Ziegler JT, et al. Genetic epidemiology of subclinical cardiovascular disease in the diabetes heart study. Ann Hum Genet. 2008;72:598–610. [PMC free article] [PubMed]