A subanalysis was performed on data acquired from 67 subjects having serum cholesterol levels of 6.2 mmol/l or above (greater than or equal to 240 mg/dl) and triglyceride levels of 2.25 to 4.00 mmol/l (199-354 mg/dl) (excluding patients with familial hypercholesterolemia), who were participating in a multicenter study comparing lovastatin and gemfibrozil, to evaluate the role of these agents in the treatment of combined hyperlipidemia (type IIb phenotype). In stratum 1 (cholesterol measures of 62.-7.79 mmol/l [240-301 mg/dl]), patients received either lovastatin 20 mg nightly (n = 17) or gemfibrozil 600 mg twice daily (n = 8), and in stratum 2 (cholesterol levels greater than or equal to 7.8 mmol/l [greater than or equal to 302 mg/dl]), patients received either lovastatin 40 mg nightly (n = 23) or gemfibrozil 600 mg b.i.d. (n = 19) for 6 weeks. Low-density lipoprotein (LDL) cholesterol levels were reduced significantly more by lovastatin than by gemfibrozil (stratum 1, -23 versus +1%, stratum 2, -34 versus -12%, respectively). A treatment goal of 4.0 mmol/l (155 mg/dl) for LDL cholesterol was achieved by 59 and 35% of patients receiving lovastatin and by 0 and 11% of patients receiving gemfibrozil in strata 1 and 2, respectively. Gemfibrozil was more effective in reducing triglyceride levels and in increasing high-density lipoprotein (HDL) cholesterol in both strata, although increases in HDL/LDL cholesterol ratios were greater with lovastatin. We conclude that, although lovastatin was more useful in normalizing LDL cholesterol, neither agent was ideal for all patients with combined hyperlipidemia. Further development of treatment regimens is called for in this group of patients.