Disturbances in normal glucose metabolism and homeostasis which manifest as hyperglycemia and glucose intolerance are often observed during clinical sepsis. Skeletal and myocardial muscle as well as whole body insulin resistance have been demonstrated in this laboratory and others during experimental and clinical sepsis. The existence of hepatic insulin resistance in sepsis has yet to be fully elucidated. This study was undertaken to assess hepatic insulin resistance during chronic hyperdynamic sepsis. Animals were randomly assigned to a septic (n = 7), sham (n = 7), or control (n = 7) group. Sepsis was induced in anesthetized dogs via a midline laparotomy whereby a fecal-soaked gauze sponge was placed amid the intestines. Sham animals underwent a laparotomy with mechanical manipulation of the intestines but no fecal implant. Control animals had no previous surgery. Sham and control dogs were pair-fed with the septic dogs. On postoperative day 7, after an overnight fast, animals were anesthetized, intubated, and ventilated. Via a left subcostal laparotomy, electromagnetic flow probes were placed to measure hepatic arterial and portal venous blood flows. Cannulae were placed in femoral, portal, and hepatic veins and femoral artery and used to calculate hepatic outputs of glucose, lactate, and oxygen during a basal period and hyperinsulinemic-euglycemic clamps which used intravenous insulin infusions which ranged from 0.4 to 4,000 mU/min. Mean arterial blood pressure decreased with increasing insulin concentrations in septic animals while no change was seen in control or sham animals. In control and sham animals, net hepatic glucose output (NHGO) decreased in response to increasing insulin levels. Septic animals showed no such inverse relationship and, moreover, showed no change in glucose output response to any insulin infusion, i.e., hepatic insulin unresponsiveness during sepsis. Net hepatic lactate output during basal pre-insulin period during sepsis was negative. This was in contrast to the positive outputs in control and sham animals. Glucose infusion rates (GIR) increased during insulin infusion but were not different between groups at any insulin infusion rate. These data demonstrated a hepatic insulin resistance (unresponsiveness) during chronic hyperdynamic, hypermetabolic sepsis.