OBJECTIVE It has been observed that the apo(a) isoforms of low molecular weight predict the family history of premature ECV. Our objective was to study the relationship between the levels of Lp(a) and the isoforms of apo(a) in parents and children. METHODS Twenty-four families, where at least one of the children had been diagnosed with familial or polygenic hypercholesterolemia and at least one of these children and one parent had Lp(a) > or = 30 mg/dl, were studied. They were classified according to whether Lp(a) was > or = < 30 mg/dl, with 44 and 38 subjects in each group, respectively. The total cholesterol, triglycerides, HDL and LDL, the Lp(a) and apo(a) isoforms were determined, as well as the relative concentration of the latter. RESULTS The molecular weight average of the major isoform was lower in the group with high Lp(a) levels (592 +/- 38 vs 656 +/- 65 kD, p < 0.001). An inverse correlation between Lp(a) levels and the major isoform size was found (r = -0.522, p < 0.001). The correlation of the Lp(a) levels in the child with that of his or her "best fit" (the best fit parent was that whose level was closer to that of the child) was very significant (r = 0.61, p < 0.01), but lower than the corresponding paternal apo(a) isoform levels (r = 0.801, p < 0.01). The study of the apo(a) isoforms in parents and their children suggests not only that the isoform is inherited, but also its concentration. CONCLUSIONS Subjects with small size apo(a) isoforms have higher Lp(a) levels. There is a better correlation between parents and children in the phenotypes of the apo(a) than in Lp(a) levels. The determination of these parameters allows the identification of a risk population.