Recent reports conflict on the effect that pentobarbital anesthesia has on basal glucose turnover in the rat. It is also unclear whether pentobarbital alters insulin suppressibility of hepatic glucose production (Ra). We examined these issues by performing basal and hyperinsulinemic euglycemic clamp studies in anesthetized and conscious animals. Ra and glucose utilization (Rd) were estimated using a steady-state infusion of 3-[3H]glucose. Pentobarbital anesthesia in normothermic rats transiently elevated plasma glucose but resulted in a sustained suppression of basal Ra (10.4 +/- 0.3 vs. conscious 13.2 +/- 0.9 mg.kg-1.min-1, P less than 0.05). In the insulin-stimulated state (110 mU/l), despite similar plasma glucose and insulin levels, clamp glucose infusion rate was significantly reduced in anesthetized animals (11.1 +/- 0.9 vs. conscious 23.6 +/- 1.3 mg.kg-1.min-1, P less than 0.001). This can be attributed to both a significantly lower insulin-stimulated Rd (15.4 +/- 1.3 vs. conscious 22.8 +/- 1.4 mg.kg-1.min-1, P less than 0.005) and reduced insulin suppression of Ra (4.3 +/- 0.8 vs. conscious -0.8 +/- 0.5 mg.kg-1.min-1, P less than 0.001; i.e., anesthetized 59% vs. conscious 100% reduction of basal Ra). Thus pentobarbital anesthesia significantly reduces basal Ra and induces hepatic insulin resistance (reduces Ra suppressibility). Pentobarbital effects are not dependent on induced hypothermia, but this exacerbates the metabolic perturbation. Caution should be used in extrapolating from the anesthetized to the conscious state.