Determinants of lipoprotein(a) levels in a middle-aged working population

Abstract

Background. The association between blood lipids, apolipoproteins, fibrinogen, life-style-related factors and lipoprotein(a) was assessed in a cohort of middle-aged men. Methods. Male employees, working in local industry, were invited to participate in a health survey at their worksite. After exclusion of nine persons with prevalent diabetes and 14 subjects with a history of myocardial infarction or angina, data were available on 720 healthy Caucasian men. Results. Lipoprotein(a) concentration was measured using an enzyme-linked immunosorbent assay (ELISA), and distribution was found to be highly skewed with a median level of 9 mg.dl-1 (mean level 23.1 mg.dl-1). The percentage of subjects with lipoprotein(a) levels higher than 30 mg.dl-1 was 23.6%. Univariate analysis showed a significant association between lipoprotein(a) and age, total cholesterol, apolipoprotein B and fibrinogen. However, no relationship was found with body mass index, waist to hip ratio, smoking, blood pressure, alcohol consumption, diet, HDL cholesterol, apolipoprotein AI and apolipoprotein E concentration or apolipoprotein E polymorphism. In multivariate analysis, ln-transformed lipoprotein(a) correlated positively with apolipoprotein B (P < 0.0001) and fibrinogen (P = 0.004). Proportional changes in lipoprotein(a) concentration were predicted in relation to specified changes in biochemical and lifestyle variables. A 20 mg.dl-1 increase in apolipoprotein B and a 75 mg.dl-1 increase in fibrinogen levels were estimated to increase lipoprotein(a) concentration by 29.4% and 21.5% respectively. Conclusions. Our data confirm the existence of an independent association between lipoprotein(a) and fibrinogen and give evidence for correlation with apolipoprotein B.