W e tested the hypothesis that glucose plus insulin determine the rate of fat oxidation in humans by controlling the rate of fatty acid entrance into the mitochondria. We gave constant infusions of [1-13C]oleate, a long-chain fatty acid, and [1-14C]octanoate, a medium-chain fatty acid, for 3 h in seven volunteers (basal). Immediately after the basal period, a hyperinsulinemic (insulin infusion = 120 mU x m(-2) min(-1)), hyperglycemic (plasma glucose = 140 mg/dl) clamp was started and continued for 5 h. During the last 3 h of the clamp, the infusions of [1-13C]oleate and [1-14C]octanoate were repeated. Intracellular acylcarnitine concentrations were measured in muscle biopsies obtained before and after the clamp. Plasma oleate enrichment and FFA concentration were kept constant by means of variable infusions of lipids and heparin. Oleate, but not octanoate, requires carnitine binding to gain access to the mitochondrial matrix; hence, if glucose and/or insulin limit long-chain fatty acid entrance into the mitochondria, then, during the clamp, long-chain acylcarnitine formation should be decreased, causing a decrease in oleate, but not octanoate, oxidation. Oleate oxidation decreased from the basal value of 0.7+/-0.1 to 0.4+/-0.1 micromol x kg(-1) x min(-1) (P < 0.05). In contrast, octanoate oxidation remained unchanged. Long-chain acylcarnitine concentration decreased from 855+/-271 in the basal state to 376+/-83 nmol/gram dry weight during the clamp (P < 0.05). We conclude that glucose and/or insulin determine fatty acid oxidation by controlling the rate of long-chain fatty acid entrance into the mitochondria.