Gender-specific association of coronary artery calcium and lipoprotein parameters: the Heinz Nixdorf Recall Study

Background: Coronary atherosclerosis can be detected by computed tomography. The amount of coronary artery calcification (CAC) is related to cardiovascular risk factors, the strength of the gender specific relation between lipoprotein parameters and CAC has not extensively been studied. Especially, the role of routinely determined lipoproteins in contrast to less common and computed lipid parameters (e.g. ratios) remains to be clarified.

Methods and results: The study cohort (n = 3956, 52% women, age 45-75 years) was randomly selected from three cities of a German metropolitan area. Lipoproteins-low-and high density lipoprotein (LDL-C/HDL-C), total cholesterol, apolipoprotein A-1 and B (apoA-1/apoB) as well as lipoprotein (a) (Lp(a)) were measured, while non-HDL-C was calculated. All participants received an electron-beam computed tomography (EBCT) for quantification of CAC. Adjusted for age and cardiovascular risk factors, CAC increased by a factor of 1.97 (1.51-2.57, 95% CI) and 1.94 (1.53-2.45, 95% CI) comparing the fourth to the first quartile of LDL-C for men and women, respectively. This association with LDL-C was also found after dichotomization of CAC at thresholds >0, ≥ 100 and ≥ 400. The best association of CAC was, however, found to be apoB and the second best was non HDL-C, in both men and women. For apoB, the model including all risk factors reached an explained variance for CAC of 20.2% in men and of 21.6% in women. When using LDL-C as a given parameter according to the current practice and advice, HDL-C in men and apoB in women provided an additional but small benefit.

Conclusion: ApoB showed the best association with CAC compared to all other tested lipoproteins. Neither the ratio LDL-C/HDL-C nor apoB/apoA-1, or Lp(a) revealed a closer association with CAC. While lipoproteins are related to CAC more closely in women than in men, their association with CAC is, however, not particularly strong. Our results may influence primary and secondary prevention advices in order to improve detection of subclinical atherosclerosis, for which lipoprotein parameters can only play a minor role.