U.S. Department of Defense


Date of this Version



Journal of Clinical Lipidology (2012) 6, 174–179; doi:10.1016/j.jacl.2011.11.005


BACKGROUND: Non-high-density lipoprotein (HDL) cholesterol is recommended as a secondary lipid goal treated initially with lifestyle modification. However, the relationship between non-HDL and subclinical atherosclerosis is unknown. We examined the independent relationships between coronary artery calcium (CAC), lipids including non-HDL, exercise, and diet among healthy male participants of the Prospective Army Coronary Calcium (PACC) Project.

METHODS: Male participants from the PACC Project (n = 1637, mean age 42.8 years; no history of coronary heart disease) were studied. We used validated surveys to measure dietary quality and habitual physical exercise. Fasting lipid concentrations and other cardiovascular risk variables were measured. Subclinical atherosclerosis was detected with the use of electron beam computed tomography for CAC. Factors independently associated with the presence of any detectable CAC (CAC score > 0), including standard CV risk variables, non-HDL, exercise, and diet, were evaluated with the use of logistic regression.

RESULTS: The mean Framingham risk score was 4.6 ± 2.6%; CAC was present in 22.4%. Fasting lipid concentrations showed mean LDL-C 128 ± 32 mg/dL, HDL-C 50 ± 13 mg/dL, TG-C 130 ± 86 mg/dL, and non-HDL-C 154 ± 37 mg/dL. Men with CAC had significantly greater levels of LDL-C (135 vs 127 mg/dL), TG (148 vs 124 mg/dL), and non-HDL-C (164 vs 151 mg/dL) and less habitual physical activity (P = 0.006). There were nonsignificant trends between prevalent CAC, greater amounts of dietary fat intake, and lower HDL-C. In successive multivariable logistic regression models for the dependent variable CAC, only non-HDL-C (odds ratio [OR] 1.012 per mg/dL; 95% CI 1.002– 1.023; P = .019) and age (OR 1.119 per year; 95% CI 1.063–1.178; P < .001) were independently associated with the presence of CAC, and exercise (OR 0.808; 95% CI 0.703–0.928; P = 0.003) was associated with the absence of CAC.

CONCLUSIONS: Non-HDL-C and exercise are independently predictive of the presence of subclinical CAC among healthy lower-risk middle-aged men.