We studied the effects of gemfibrozil on glucose and fatty acid metabolism in subjects with mild endogenous hypertriglyceridemia. Twenty subjects (serum triglycerides, 3.2 +/- 1.4 mmol/L; age, 52 +/- 7 years; body mass index, 27.8 +/- 1.8 kg/m2) were randomly allocated to receive either placebo or gemfibrozil 1,200 mg daily for 12 weeks in a double-blind study. Gemfibrozil decreased serum total and very-low-density lipoprotein (VLDL) triglycerides by 53% and 57%, respectively, and serum apolipoprotein (apo) B concentration by 21%. Gemfibrozil had no effect on the diurnal concentration of free fatty acids (FFA). Neither did gemfibrozil change diurnal blood glucose or serum insulin concentrations. The endogenous glucose production rate remained unchanged in both groups during the treatment period, and was similarly suppressed by hyperinsulinemia. The rate of insulin-induced whole-body glucose disposal increased similarly both before (basal 10.8 +/- 1.8, low-dose insulin 10.5 +/- 2.1, and high-dose insulin 20.9 +/- 11.9 mumol.kg-1.min-1) and after (11.1 +/- 1.7, 10.7 +/- 1.2, and 18.6 +/- 7.9, respectively) gemfibrozil treatment. Rates of oxidative and nonoxidative glucose metabolism remained unchanged during gemfibrozil treatment. Basal pretreatment and posttreatment FFA turnover rates were similar in both study groups, as were the rates of substrate oxidation. In summary, gemfibrozil proved to be an effective serum triglyceride-lowering agent in patients with mild hypertriglyceridemia, but had no effect on the insulin sensitivity of glucose metabolism or of antilipolysis. These data support the idea that triglycerides per se do not cause insulin resistance, and that the triglyceride-lowering effect of gemfibrozil is not mediated via antilipolytic action.