Intermuscular Adipose Tissue and Subclinical Coronary Artery Calcification in Midlife: The CARDIA Study (Coronary Artery Risk Development in Young Adults)

Abstract

Objective: Excess deposition of fat within and around vital organs and nonadipose tissues is hypothesized to contribute to cardiovascular disease (CVD) risk. We evaluated the association of abdominal intermuscular adipose tissue (IMAT) volume with coronary artery calcification in the CARDIA study (Coronary Artery Risk Development in Young Adults) participants.

Approach and results: We measured IMAT in the abdominal muscles, visceral adipose tissue and pericardial adipose tissue, and coronary artery calcification using computed tomography in 3051 CARDIA participants (56% women) at the CARDIA year 25 examination (2010-2011). Mean IMAT volume and mean IMAT/total muscle volume (IMAT normalized for muscle size) were calculated in a 10-mm block of slices centered at L3-L4. Multivariable analyses included potential confounders and traditional cardiovascular disease risk factors. Compared with the lowest quartile, the upper quartile of abdominal IMAT volume was associated with higher coronary artery calcification prevalence (odds ratio [95% confidence interval], 1.6 [1.2-2.1]) after adjusting for cardiovascular disease risk factors. Results were similar for highest versus lowest quartile of IMAT normalized to total muscle volume (odds ratio [95% confidence interval], 1.5 [1.1-2.0]). Significant associations of higher IMAT and normalized IMAT with coronary artery calcification prevalence persisted when body mass index, visceral adipose tissue, or pericardial adipose tissue were added to the models.

Conclusions: In a large, community-based, cross-sectional study, we found that higher abdominal skeletal muscle adipose tissue volume was associated with subclinical atherosclerosis independent of traditional cardiovascular disease risk factors and other adipose depots.

Keywords: adipose tissue; body mass index; muscle; prevalence; risk factors; vascular calcification.

Figure 1

Single CT slice through the abdomen at the L3–L4 lumbar level. Right and left muscle groups are indicated. Left and right sides were averaged for analyses.

Figure 2

Unadjusted Agatston CAC scores by quartiles of overall IMAT volume (chi square analysis p<.0001).

Figure 3

a–d. Probability of positive CAC scores of 1–19.9, 20–99.9, and ≥100 Agatston Units across quartiles of IMAT volume derived using multivariable multinomial logistic regression. Probabilities are adjusted for age, sex, race, sex*race, center, height (except when BMI is included), education, physical activity, alcohol consumed, smoking history, diabetes, systolic BP, BP med use, HDL cholesterol, triglycerides, cholesterol med use, and CRP (panel 3a) with or without BMI (panel 3b) or VAT (panel 3c) or PAT (panel 3d). P-value is based on multinomial logistic regression model.

Figure 4

a–d. Probability of positive CAC scores of 1–19.9, 20–99.9, and ≥100 Agatston Units across quartiles of normalized IMAT derived using multivariable multinomial logistic regression. Probabilities are adjusted for age, sex, race, sex*race, center, height (except when BMI is included), education, physical activity, alcohol consumed, smoking history, diabetes, systolic BP, BP med use, HDL cholesterol, triglycerides, cholesterol med use, and CRP (panel 4a) with or without BMI (panel 4b) or VAT (panel 4c) or PAT (panel 4d). P-value is based on multinomial logistic regression model.