HMG-CoA reductase inhibitors effectively reduce the concentration of low density lipoproteins (LDL) in plasma. Lipoprotein(a) [Lp(a)] may be as atherogenic as LDL. A few studies, only one of which was placebo controlled, suggest that the HMG CoA reductase inhibitors either do not affect Lp(a) or they increase Lp(a). The response of Lp(a) to HMG-CoA reductase inhibition has not been related to apolipoprotein(a) phenotypes in previous studies. We conducted a double-blind, placebo controlled study of pravastatin in 51 patients with familial hypercholesterolemia (FH) (n = 43) or probable FH (n = 8). All patients had LDL-cholesterol concentration above 4.1 mmol l-1 despite treatment with diet and bile acid sequestration. In patients assigned to pravastatin (n = 34), the mean concentrations of total cholesterol and LDL cholesterol fell significantly (P < 0.01) when compared to placebo. Lp(a) increased (P < 0.01) from a mean (+/- SD) of 33.6 +/- 40.8 mg dl-1 to 41.1 +/- 46.1 mg dl-1 on pravastatin but was unchanged during placebo treatment. The percentage increase in Lp(a) was the same in patients with different apo(a) phenotypes, and hence the absolute increase in Lp(a) was greatest in patients with the low molecular weight apo(a) phenotypes.