The aim of this study was to determine the effect of gemfibrozil, compared with lovastatin, in patients with high levels of lipoprotein(a) and on plasma lipid profile. Twenty-seven nondiabetic patients with high levels of plasma lipids and lipoprotein(a), 19 male and eight female, aged 37-68 (mean +/- S.D. 54.2 +/- 7.5) years, were randomly assigned to 2 weeks of treatment with gemfibrozil 600 mg twice daily (14 pts.) or lovastatin 40-80 mg once daily (13 pts.). Patients had fasting plasma total cholesterol levels > or = 6.2 mmol/l, low-density lipoprotein > 4.14 mmol/l and lipoprotein(a) > 0.62 mmol/l. All patients but one had triglycerides > 2.82 mmol/l. There were no statistical differences between both groups in terms of age, sex, clinical diagnosis and previous medication. After 3 months of treatment, gemfibrozil reduced triglycerides (47.9% vs. 24.5%; P < 0.001), very low density lipoprotein (43.9% vs. 24.6%; P < 0.05), lipoprotein(a) (25.3% vs. 4.9%; P < 0.05) and increased high-density lipoprotein (34.4% vs. 11%; P < 0.01) more than lovastatin. Gemfibrozil and lovastatin reduced comparably total cholesterol (21.4% vs. 29.0%; P = NS) and low-density lipoprotein (26.5% vs. 37.3%; P = NS). The plasma levels of high-density lipoprotein and lipoprotein(a) were unchanged significantly by lovastatin. In conclusion, besides well-known efficacy in hyperlipidemia treatment, gemfibrozil also increased high-density lipoprotein and reduced lipoprotein(a), which may have important epidemiologic implications.