Apo(a) isoforms do not predict risk for coronary heart disease in a Gulf Arab population

It is well recognized that blood lipoprotein A [Lp(a)] levels constitute an important risk factor for atherosclerotic vascular disease. In some populations, mainly Caucasian, Lp(a) levels and coronary heart disease (CHD) risk are determined by the pattern of apolipoprotein a [apo(a)] polymorphism. It is currently unclear if these observations apply to other populations and ethnic groups. The aim of the current study is to determine to what extent known apo(a) polymorphisms associate with development of CHD in a Kuwaiti Arab population. Serum Lp(a) levels were measured by enzyme-linked immunosorbent assay and apo(a) isoforms determined by a high-resolution sodium dodecyl sulphate/agarose gel electrophoresis with immunoblotting in two groups of Kuwaiti subjects: healthy controls (n = 140) and subjects with CHD (n = 140). Blood lipids and anthropometric parameters were also determined in these subjects by standard methods. Serum Lp(a) levels were greater in those with CHD than in those in the healthy group (P < 0.001). There was no consistent trend in the pattern of serum Lp(a) levels found with specific apo(a) isoforms in either group of subjects. There was, therefore, no simple relationship between the isoform pattern (and number of kringle-IV repeats) and serum Lp(a) concentration, unlike in certain other populations. Additionally, almost identical proportions of subjects in either group had single-banded (homozygous, approximately 70%), double-banded (heterozygous, approximately 23%) and no-band (null, approximately 7%) phenotypes. The distribution of the five identified isoforms (F, S1, S2, S3 and S4) also was almost identical for both groups of subjects, whether homozygous or heterozygous, and whether classified into fast-moving (F, S1 and S2) or slow-moving (S3 and S4) isoforms. We conclude that the frequency and pattern of distribution of apo(a) phenotypes did not differ significantly between healthy control Kuwaiti Arab subjects and those with CHD. It is thus unlikely that an individual's apo(a) phenotype can predict both serum Lp(a) level and risk for CHD, irrespective of race and/or ethnic grouping.