Relation of Subcutaneous and Visceral Adipose Tissue to Coronary and Abdominal Aortic Calcium (from the Framingham Heart Study)

Article Outline

• Abstract
• Methods
• Results
• Discussion
• References
• Copyright

Body fat distribution might be differentially associated with subclinical cardiovascular disease. We examined whether the body mass index, waist circumference, and subcutaneous and visceral adipose tissue are associated with the prevalence of either coronary or abdominal aortic calcium in the Framingham Heart Study. Participants (n = 3,130, mean age 52 years, 49% women) free of clinical cardiovascular disease from the Framingham Heart Study underwent multidetector computed tomographic assessment to quantify the subcutaneous and visceral fat volume and coronary and abdominal aortic calcification. Coronary artery calcification and abdominal aortic calcium were examined in relation to the body mass index, waist circumference, subcutaneous adipose tissue, and visceral adipose tissue in age-, gender-, and multivariate-adjusted models. All measures of adiposity were associated with coronary aortic calcium in the age- and gender-adjusted models (all p <0.008). All relations were attenuated in the multivariate models (all p >0.14). The body mass index, waist circumference, and visceral adipose tissue (but not the subcutaneous adipose tissue) were associated with abdominal aortic calcification in the age- and gender-adjusted models (all p <0.012). However, all relations were attenuated in the multivariate models (all p >0.23). Similar findings were observed in the quartile-based analyses. In conclusion, the general measures of obesity and measures of central abdominal fat are related to the coronary aortic calcium and abdominal aortic calcium levels. However, these cross-sectional associations are attenuated by cardiovascular disease risk factors, possibly because they mediate the association between adiposity measures and subclinical cardiovascular disease.

Obesity and the associated metabolic risk factors are associated with cardiovascular disease.¹ Obesity is also associated with coronary artery calcium,², ³, ⁴, ⁵ a marker of atherosclerotic burden, that is positively associated with coronary heart disease and cardiovascular disease events.⁶, ⁷, ⁸ Because abdominal adiposity might be more atherogenic than generalized adiposity,⁹, ¹⁰ measures of central obesity, particularly visceral abdominal tissue, might be more strongly associated with coronary aortic calcium (CAC) or abdominal aortic calcium (AAC). Previous studies that have examined this question were conducted in small, selected samples, limiting the generalizability of their findings.⁴, ¹¹ The purpose of the present analysis was to comprehensively evaluate the association between body fat depots and the prevalence of CAC and AAC in a community-based sample.

Back to Article Outline

Methods 

The study participants were from the Framingham Heart Study Multi-Detector Computed Tomography study, a substudy derived from the Framingham Heart Study Offspring and Third Generation cohorts. The study design has been previously described.¹², ¹³, ¹⁴ From June 2002 to April 2005, 3,529 participants (2111 Third Generation and 1418 Offspring participants) underwent multidetector computed tomography for assessment of the aortic and coronary calcium. Study inclusion was weighted toward participants who resided in the Greater New England area and those from larger Framingham Heart Study families; additional details regarding the study exclusion criteria have been previously described.⁹ Of the 3,529 participants who underwent imaging, 3,495 had interpretable coronary and aortic calcium measures, 3,338 also had interpretable subcutaneous adipose tissue (SAT) and visceral adipose tissue (VAT) measures, 3,143 were free of clinical cardiovascular disease, and 3,143 attended a contemporaneous examination, 3,130 of whom had a complete covariate profile, resulting in a final sample size of 3,130 participants.

The institutional review boards of the Boston University Medical Center and Massachusetts General Hospital approved the study. All subjects provided written informed consent.

The participants underwent 8-slice multidetector computed tomography of the chest and abdomen, as previously described (LightSpeed Ultra, General Electric, Milwaukee, Wisconsin).¹⁵ The SAT and VAT volumes were assessed using an Aquarius 3D Workstation (TeraRecon, San Mateo, California) using fat pixel identification (image display window width of −195 to −45 Hounsfield units; window center −120 Hounsfield units). The interclass correlation for the inter-reader comparisons was 0.992 for VAT and 0.997 for SAT.¹⁵

To image the heart, on average, 48 2.5-mm contiguous slices were obtained using a prospectively electrocardiographic-triggered scanning protocol; image acquisition was initiated at 70% of the cardiac cycle (120 kVp, 400 mA, temporal resolution 330 ms) using a procedure that has been previously described.¹⁶ The multidetector computed tomographic scans were quantified for the presence and quantity of AAC and CAC by an experienced reader using a dedicated off-line workstation (Aquarius, Terarecon, San Mateo, California).¹⁷ CAC or AAC (present vs absent) was determined by the age- and gender-specific ninetieth percentile cutpoints derived from a healthy referent sample.¹⁸

The risk factors were obtained at Offspring examination cycle 7 or the first examination of the Third Generation. The body mass index (BMI) was defined as the weight (in kilograms) divided by the height in meters squared. The waist circumference was measured using a tape measure at the level of the umbilicus. Fasting samples were used to measure the plasma glucose, total and high-density lipoprotein cholesterol, and triglycerides. Diabetes was defined a fasting plasma glucose of ≥126 mg/dl at a Framingham Heart Study examination or current treatment with either a hypoglycemic agent or insulin. Participants who smoked ≥1 cigarette/day for the past year were defined as current smokers. Physician-administered questions were used to quantify alcohol use; drinks per week were dichotomized using the following cutpoints: >14 drinks/wk (men) or >7 drinks/wk (women). Hypertension was defined as a systolic blood pressure of ≥140 mm Hg, or a diastolic blood pressure of ≥90 mm Hg, or the use of antihypertension therapy.

All measures of adiposity were standardized to a mean of 0 and standard deviation of 1. We examined the relation between the quartiles of each adiposity variable and the presence of CAC or AAC, as well as continuous measures of adiposity and the presence of CAC or AAC using generalized estimating equations to account for familial relations in our study sample. Models were age, gender, and multivariate (age, gender, systolic blood pressure, hypertension treatment, total/high-density lipoprotein cholesterol, triglycerides, lipid treatment, diabetes) adjusted.

Back to Article Outline

Results 

The study sample characteristics are presented in Table 1. Overall, the mean age was 52 years, and nearly 1/2 of the participants were women; 15% (n = 457) had CAC and 19% had AAC.

Table 1. Study sample characteristics (n = 3,130)

Variable	Value
Age(years)	52(11)
Women	1,528(49%)
Systolic blood pressure(mm Hg)	122 ± 18
Hypertension treatment	528(17%)
Smoking	397(13%)
Total/HDL cholesterol	3.9 ± 1.4
Median triglycerides(25th, 75th)	102(71,154)
Total cholesterol(mg/dl)	197(35)
HDL cholesterol(mg/dl)	54(17)
LDL cholesterol(mg/dl)	118(31)
Lipid treatment	372(12%)
Diabetes	156(5%)
Body mass index(kg/m2)	27.7 ± 5.9
Waist circumference(inches)	38.1 ± 5.6
Subcutaneous adipose tissue(cm3)	2,827 ± 1,396
Visceral adipose tissue(cm3)	1,772 ± 1,004
% Coronary aortic calcium >90%	457(15%)
% Abdominal aortic calcium >90%	601(19%)

Data are presented as means(standard deviation) or numbers(percentages).

To convert from milligrams per deciliter to millimole per liter, divide by 38.67 for total cholesterol, HDL, and LDL and by 88.57 for triglycerides.

HDL = high-density lipoprotein; LDL = low-density lipoprotein.

In the age- and gender-adjusted models, the participants with a BMI, waist circumference, SAT, and VAT in the fourth quartile had a greater prevalence of CAC (all p <0.05; Table 2). For example, persons in the fourth quartile of VAT had a 1.56-fold increased odds of CAC (odds ratio 1.56, 95% confidence interval 1.15 to 2.10, p = 0.004). After multivariate adjustment, all relations were attenuated (all p >0.11).

Table 2. Association between different measures of adiposity and coronary artery calcium(CAC)

Variable	Quartile 2⁎	p Value	Quartile 3⁎	p Value	Quartile 4⁎	p Value
Body mass index
Age, gender adjusted	1.29(0.95–1.76)	0.11	1.29(0.95–1.75)	0.10	1.93(1.44–2.59)	<0.0001
Multivariate adjusted†	1.12(0.82–1.54)	0.48	0.99(0.72–1.37)	0.97	1.30(0.94–1.79)	0.11
Waist circumference
Age, gender adjusted	1.19(0.89–1.65)	0.24	1.32(0.97–1.79)	0.08	1.80(1.34–2.41)	<0.0001
Multivariate adjusted†	1.05(0.77–1.43)	0.77	1.01(0.73–1.39)	0.96	1.18(0.85–1.64)	0.32
Subcutaneous adipose tissue
Age, gender adjusted	0.91(0.68–1.23)	0.55	1.03(0.77–1.36)	0.85	1.33(1.00–1.78)	0.05
Multivariate adjusted†	0.80(0.59–1.08)	0.15	0.82(0.61–1.10)	0.19	0.97(0.72–1.31)	0.85
Visceral adipose tissue
Age, gender adjusted	0.93(0.69–1.26)	0.65	1.12(0.83–1.52)	0.46	1.56(1.15–2.10)	0.004
Multivariate adjusted†	0.75(0.55–1.02)	0.07	0.77(0.55–1.06)	0.10	0.87(0.61–1.23)	0.43

Data are presented as odds ratio(95% confidence intervals) for each quartile of CAC.

Quartile cutpoints for women—BMI: 22.8, 25.7, and 30.1 kg/m²; waist circumference: 31.8, 35.5, and 39.8 in.; SAT: 2,033, 2,857, and 4,084 cm³; VAT: 681, 1,167, and 1,804 cm³.

Quartile cutpoints for men—BMI: 25.3, 27.8, and 30.5 kg/m²; waist circumference: 35.3, 39.0, and 42 in.; SAT: 1,796, 2,398, and 3,196 cm³; and VAT: 1,457, 2,093, and 2,794 cm³.

When each measure of adiposity was modeled continuously (Table 3), BMI, waist circumference, SAT, and VAT were all associated with CAC in the age- and gender-adjusted models (all p <0.008). On multivariate adjustment, all relations were attenuated (p >0.14).

Table 3. Association between different measures of adiposity and coronary artery calcium(CAC) and abdominal aortic calcium(AAC)

Variable	Coronary Aortic Calcium		Abdominal Aortic Calcium
	OR per SD of Fat	p Value	OR per SD of Fat	p Value
Body mass index
Age, gender adjusted	1.21(1.05–1.39)	0.008	1.16(1.03–1.30)	0.012
Multivariate adjusted⁎	1.08(0.98–1.19)	0.14	1.01(0.91–1.11)	0.92
Waist circumference
Age, gender adjusted	1.26(1.14–1.39)	<0.0001	1.24(1.13–1.36)	<0.0001
Multivariate adjusted⁎	1.08(0.96–1.22)	0.18	1.03(0.92–1.16)	0.56
Subcutaneous adipose tissue
Age, gender adjusted	1.18(1.06–1.31)	0.002	1.09(0.99–1.19)	0.07
Multivariate adjusted⁎	1.06(0.95–1.19)	0.31	0.96(0.86–1.07)	0.42
Visceral adipose tissue
Age, gender adjusted	1.23(1.11–1.37)	<0.0001	1.34(1.21–1.48)	<0.0001
Multivariate adjusted⁎	1.01(0.99–1.14)	0.92	1.07(0.96–1.21)	0.23

Data are presented as odds ratio(95% confidence intervals) of CAC or AAC >90% relative to standard deviation change of fat measurements.

All adiposity measurements are standardized to a mean of 0 and standard deviation of 1.

OR = odds ratio.

For AAC, we found that in an age- and gender-adjusted model, only those participants with BMI, waist circumference, and VAT in the fourth quartile had a greater prevalence of AAC (all p <0.0002; Table 4). All relations were attenuated after multivariate adjustment (all p <0.58).

Table 4. Association between different measures of adiposity and abdominal aortic calcium(AAC)

Variable	Quartile 2⁎	p Value	Quartile 3⁎	p Value	Quartile 4⁎	p Value
Body mass index
Age, gender adjusted	1.15(0.88–1.50)	0.31	1.13(0.87–1.48)	0.36	1.70(1.30–2.21)	<0.0001
Multivariate adjusted†	0.99(0.74–1.31)	0.93	0.86(0.63–1.16)	0.31	1.09(0.80–1.49)	0.58
Waist circumference
Age, gender adjusted	1.09(0.84–1.43)	0.51	1.46(1.11–1.93)	0.007	1.69(1.29–2.23)	0.0002
Multivariate adjusted†	0.95(0.71–1.27)	0.73	1.10(0.81–1.48)	0.55	1.05(0.77–1.44)	0.77
Subcutaneous adipose tissue
Age, gender adjusted	0.93(0.71–1.22)	0.60	1.15(0.89–1.50)	0.29	1.09(0.83–1.42)	0.55
Multivariate adjusted†	0.82(0.62–1.09)	0.18	0.94(0.70–1.27)	0.70	0.75(0.55–1.01)	0.07
Visceral adipose tissue
Age, gender adjusted	1.02(0.77–1.35)	0.89	1.31(1.00–1.73)	0.43	2.00(1.52–2.64)	<0.0001
Multivariate adjusted†	0.81(0.60–1.09)	0.16	0.89(0.66–1.21)	0.22	1.08(0.78–1.50)	0.63

Data are presented as odds ratio(95% confidence intervals) per quartile of AAC.

For quartile cutpoints for BMI, waist circumference, SAT, and VAT, see Table 2 footnote.

When the BMI, waist circumference, SAT, and VAT were modeled continuously (Table 3), the BMI, waist circumference, and VAT were significantly associated with AAC in an age- and gender-adjusted model (all p <0.012). After multivariate adjustment, all associations were rendered statistically nonsignificant (all p >0.23).

Back to Article Outline

Discussion 

We found significant associations between all measures of adiposity and CAC in age- and gender-adjusted models. However, all relations were attenuated on multivariate adjustment. Similarly, for AAC, we observed associations between most measures of adiposity (excluding SAT) in age- and gender-adjusted models that were attenuated on multivariate adjustment. These findings suggest that shared risk factors between adipose tissue and CAC explain most of our observations.

In an early work, Mahoney et al³ found that a greater BMI in childhood, and greater BMI and waist/hip ratio in young adulthood was associated with the early presence of CAC among participants from the Muscatine Study. Of 2,951 black and white adults from the Coronary Artery Risk Development in Young Adults (CARDIA) study, Lee et al⁵ observed an association between waist girth and CAC in models adjusted for demographics. Similar to our findings, after adjustment for traditional cardiovascular disease risk factors, the associations were no longer significant. In contrast, of 443 cardiovascular disease-free persons without diabetes, Cassidy et al² found a positive association between waist circumference and CAC among low-risk, but not high-risk, persons after adjusting for age, gender, hypertension, and cholesterol. However, these studies did not use a direct measure of VAT and, instead, relied on waist circumference as a proxy.

Among the studies that used a direct measure of VAT, the findings were similar. Allison and Michael⁴ examined 3,028 self-referred persons and found that VAT was associated with CAC in men, but not in women. These findings are in our contrast to our findings and might have resulted from the selected nature of the sample (i.e., self-referred sample) and the use of self-reported covariates, which might have underestimated the true risk factor burden. In contrast, among 220 women and 190 men aged 55 to 80 years from the Rancho Bernardo Study, Kim et al¹¹ observed no association between VAT and CAC in multivariate models. However, we could not compare these findings directly with the results of the present study, because no age- or gender-adjusted model was presented in their report.

Fewer reports have been published on the relation between adiposity, in particular, central obesity, and AAC. In 168 men, AAC severity, as measured on lateral x-rays, correlated with the truncal fat mass (an indirect measure of VAT) quantified by dual energy x-ray absorptiometry¹⁹; similar effects in postmenopausal women were not observed.²⁰ Among 148 patients with peripheral artery disease, Golledge et al²¹ found a sixfold increase in the odds of AAC in the upper compared with the lower tertile of the VAT compartment (measured as the transverse diameter of the abdominal cavity) in models adjusted for age, gender, hypertension, diabetes, smoking, and cholesterol. Although intriguing, these findings were limited by the small sample size, that all had peripheral arterial disease with a likely distribution of AAC far exceeding that of a population-based sample, the very wide confidence limits, and the use of an indirect measure of VAT.

The present findings highlight the strong association between measures of adiposity and subclinical disease, as well as the mediation of these associations by cardiovascular disease risk factors that are likely in the causal pathway from adiposity to subclinical disease.

The large sample size, community-based setting, volumetric assessment of visceral and subcutaneous fat with computed tomography were strengths of our investigation. The limitations included the cross-sectional study design, such that causality could not be inferred. Furthermore, the Framingham Offspring Study was composed of primarily white participants, limiting the generalizability of our findings to other ethnic groups. Finally, we did not assess the progression of arterial calcification, and it is possible that measures of adiposity could be associated with progression of CAC or AAC.

Back to Article Outline

References 

1. Poirier P, Giles TD, Bray GA, Hong Y, Stern JS, Pi-Sunyer FX, et al. Obesity and cardiovascular disease: pathophysiology, evaluation, and effect of weight loss: an update of the 1997 American Heart Association scientific statement on obesity and heart disease from the Obesity Committee of the Council on Nutrition, physical activity, and metabolism. Circulation. 2006;113:898–918
   • View In Article
   • CrossRef
2. Cassidy AE, Bielak LF, Zhou Y, Sheedy PF, Turner ST, Breen JF, et al. Progression of subclinical coronary atherosclerosis: does obesity make a difference?. Circulation. 2005;111:1877–1882
   • View In Article
   • CrossRef
3. Mahoney LT, Burns TL, Stanford W, Thompson BH, Witt JD, Rost CA, et al. Coronary risk factors measured in childhood and young adult life are associated with coronary artery calcification in young adults: the Muscatine study. J Am Coll Cardiol. 1996;27:277–284
   • View In Article
   • Full-Text PDF (128 KB)
   • CrossRef
4. Allison MA, Michael WC. Body morphology differentially predicts coronary calcium. Int J Obes Relat Metab Disord. 2004;28:396–401
   • View In Article
   • MEDLINE
   • CrossRef
5. Lee CD, Jacobs DR, Schreiner PJ, Iribarren C, Hankinson A. Abdominal obesity and coronary artery calcification in young adults: the Coronary Artery Risk Development in Young Adults (CARDIA) Study. Am J Clin Nutr. 2007;86:48–54
   • View In Article
6. Wong ND, Hsu JC, Detrano RC, Diamond G, Eisenberg H, Gardin JM. Coronary artery calcium evaluation by electron beam computed tomography and its relation to new cardiovascular events. Am J Cardiol. 2000;86:495–498
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (302 KB)
   • CrossRef
7. Arad Y, Spadaro LA, Goodman K, Lledo-Perez A, Sherman S, Lerner G, et al. Predictive value of electron beam computed tomography of the coronary arteries (19-month follow-up of 1173 asymptomatic subjects). Circulation. 1996;93:1951–1953
   • View In Article
   • MEDLINE
8. Detrano R, Guerci AD, Carr JJ, Bild DE, Burke G, Folsom AR, et al. Coronary calcium as a predictor of coronary events in four racial or ethnic groups. N Engl J Med. 2008;358:1336–1345
   • View In Article
   • CrossRef
9. Fox CS, Massaro JM, Hoffmann U, Pou KM, Maurovich-Horvat P, Liu CY, et al. Abdominal visceral and subcutaneous adipose tissue compartments: association with metabolic risk factors in the Framingham Heart Study. Circulation. 2007;116:39–48
   • View In Article
   • CrossRef
10. Zhang C, Rexrode KM, van Dam RM, Li TY, Hu FB. Abdominal obesity and the risk of all-cause, cardiovascular, and cancer mortality: sixteen years of follow-up in US women. Circulation. 2008;117:1658–1667
    • View In Article
    • CrossRef
11. Kim DJ, Bergstrom J, Barrett-Connor E, Laughlin GA. Visceral adiposity and subclinical coronary artery disease in elderly adults: Rancho Bernardo Study. Obesity (Silver Spring). 2008;16:853–858
    • View In Article
    • CrossRef
12. Dawber TR, Kannel WB, Lyell LP. An approach to longitudinal studies in a community: the Framingham Heart Study. Ann NY Acad Sci. 1963;107:539–556
    • View In Article
13. Shurtleff D. Some characteristics related to the incidence of cardiovascular disease and death: Framingham Study, 18-year follow-up (Section 30). In:  Kannel WB,  Fordon T editor. The Framingham Study: An Epidemiological Investigation of Cardiovascular Disease (DHEW publication no. (NIH) 74-599). Washington, DC: Department of Health, Education, and Welfare; 1973;
    • View In Article
14. Splansky GL, Corey D, Yang Q, Atwood LD, Cupples LA, Benjamin EJ, et al. The Third Generation Cohort of the National Heart, Lung, and Blood Institute's Framingham Heart Study: design, recruitment, and initial examination. Am J Epidemiol. 2007;165:1328–1335
    • View In Article
    • MEDLINE
    • CrossRef
15. Maurovich-Horvat P, Massaro J, Fox CS, Moselewski F, O'Donnell CJ, Hoffmann U. Comparison of anthropometric, area- and volume-based assessment of abdominal subcutaneous and visceral adipose tissue volumes using multi-detector computed tomography. Int J Obes. 2007;31:500–506
    • View In Article
    • MEDLINE
    • CrossRef
16. Rosito GA, Massaro JM, Hoffmann U, Ruberg FL, Mahabadi AA, Vasan RS, et al. Pericardial fat, visceral abdominal fat, cardiovascular disease risk factors, and vascular calcification in a Community-Based Sample (The Framingham Heart Study). Circulation. 2008;117:605–613
    • View In Article
    • CrossRef
17. Hoffmann U, Siebert U, Bull-Stewart A, Achenbach S, Ferencik M, Moselewski F, et al. Evidence for lower variability of coronary artery calcium mineral mass measurements by multi-detector computed tomography in a community-based cohort—consequences for progression studies. Eur J Radiol. 2006;57:396–402
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (228 KB)
    • CrossRef
18. Hoffmann U, Massaro JM, Fox CS, Manders E, O'Donnell CJ. Defining normal distributions of coronary artery calcium in women and men (from the Framingham Heart Study). Am J Cardiol. 2008;102:1136–1141
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (81 KB)
    • CrossRef
19. Alexandersen P, Tanko LB, Bagger YZ, Jespersen J, Skouby SO, Christiansen C. Associations between aortic calcification and components of body composition in elderly men. Obesity (Silver Spring). 2006;14:1571–1578
    • View In Article
    • MEDLINE
    • CrossRef
20. Tanko LB, Bagger YZ, Alexandersen P, Larsen PJ, Christiansen C. Central and peripheral fat mass have contrasting effect on the progression of aortic calcification in postmenopausal women. Eur Heart J. 2003;24:1531–1537
    • View In Article
    • MEDLINE
    • CrossRef
21. Golledge J, Jayalath R, Oliver L, Parr A, Schurgers L, Clancy P. Relationship between CT anthropometric measurements, adipokines and abdominal aortic calcification. Atherosclerosis. 2008;197:428–434
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (291 KB)
    • CrossRef