Among 398 participants, 219 (55%) had calcified coronary plaque. Individuals with calcified coronary plaque were older and were more likely to be male and white compared to those without calcified coronary plaque (). In addition, they also had greater height, higher systolic blood pressure and lower concentrations of HDL cholesterol, were more likely to be former smokers, anti-hypertensive medication users, and have diabetes. There were no statistically significant differences in alcohol drinking status, triglycerides concentrations, and lipid-lowering medication use between those with and without calcified coronary plaque. Waist circumference, the average volumes of pericardial fat, abdominal visceral fat and intermuscular fat, and the non-subcutaneous fat index were significantly greater in those with a calcified coronary plaque, but the mean abdominal subcutaneous fat volume was lower. Body mass index and liver attenuation did not differ significantly between the groups defined by presence of calcification.
Characteristics of 398 MESA participants according to the presence or absence of calcified coronary plaque, 2002–2005
Body mass index, waist circumference, and abdominal subcutaneous fat were all strongly correlated, whereas the non-subcutaneous fat index was highly correlated with pericardial fat and abdominal visceral fat (). Liver attenuation and intermuscular fat were moderately correlated with other fat measures.
Spearman correlation coefficients between fat measures in 398 MESA participants, 2002–2005
Logistic regression analysis was used to examine the association of fat measures with calcified coronary plaque (). After adjusting for demographics and height, an increment of one standard deviation in the non-subcutaneous fat index was associated with a 43% increase in the odds of calcified coronary plaque (p = 0.008) (Model 1). Body mass index, waist circumference, pericardial fat, and abdominal visceral fat were also significantly associated with calcified coronary plaque. These associations remained statistically significant after further adjustment of lifestyle factors (Model 2). After adjusting for demographics, lifestyle factors, cardiovascular risk factors and height, only the association between the non-subcutaneous fat index and calcified coronary plaque remained statistically significant, although a borderline significance of the association between pericardial fat and calcified coronary plaque was observed (Model 3). When both abdominal subcutaneous fat and the non-subcutaneous fat index were included in the same model adjusting for demographics, lifestyle factors and height, the non-subcutaneous fat index (OR = 1.40; 95% CI: 1.00, 1.94), but not abdominal subcutaneous fat (OR = 1.02; 95% CI: 0.74, 1.39), was associated with calcified coronary plaque. Excluding ever smokers or diabetic participants, the association between the non-subcutaneous fat index and calcified coronary plaque was found to be even stronger in the analysis adjusting for demographics, lifestyle factors and height (OR = 1.77; 95% CI: 1.10, 2.84). Physical activity and diet were assessed 2–4 years earlier than the other measures. With adjustment for demographics, lifestyle factors including smoking status, alcohol drinking status, total intentional exercises, total energy intake and fat intake, and height, the non-subcutaneous fat index (OR = 1.39; 95% CI: 1.05, 1.86) was still associated with calcified coronary plaque.
Odds ratios of calcified coronary plaque per one standard deviation increment in fat measures in 398 MESA participants, 2002–2005
To investigate log-linearity of the association between the non-subcutaneous fat index and calcified coronary plaque, the participants were categorized into quartiles according to this index, and odds ratios for calcified coronary plaque were calculated comparing each quartile with the lowest quartile after adjusting for demographics, lifestyle factors and height (). The second and the fourth quartiles, but not the third, were consistent with a log-linear relationship. No heterogeneity was observed for the association between the non-subcutaneous fat index and calcified coronary plaque according to either gender (p value for the interaction term = 0.74) or ethnicity (p value for the interaction term = 0.54).
Figure 1 Odds ratio (95% CI) of calcified coronary plaque according to quartiles of the non-subcutaneous fat index, adjusting for age, gender, ethnicity, smoking status, alcohol drinking status and height, in a logistic regression analysis in 398 MESA participants, (more ...)