Glucose production by the liver is controlled on a minute to minute basis by the plasma insulin and glucagon concentrations. Glucagon increases glycogen breakdown and stimulates the gluconeogenic pathway and insulin inhibits both glycogenolysis and gluconeogenesis [1]. Until ten years ago it was generally accepted that the insulin level within the hepatic sinusoids was responsible for the hormone's inhibitory effect on the liver. Recently, however, that concept has been challenged.

In 1987 this concept was first called into question [2]. It was noted that in obese non-diabetic humans suppression of glucose production could occur in response to insulin infusion even when the estimated portal insulin concentrations did not rise. Insulin was infused into a peripheral vein and euglycaemia was maintained with a glucose infusion. Endogenous insulin secretion, estimated from C-peptide concentrations, decreased by about 50% over the course of the experiment. Based on the fall in endogenous insulin release and the prevailing peripheral insulin concentration, the authors estimated that the portal vein insulin concentration probably remained unchanged. Nevertheless glucose production was suppressed by about 80% at the end of their study. It should be remembered, however, that about 20% of liver blood flow is derived from the hepatic artery, so a rise in arterial insulin would have increased the liver sinusoidal insulin concentration somewhat, even though the portal insulin level did not change. Likewise, although the authors assumed that their tracer data