Impaired sensitivity to insulin is a frequent clinical situation and interest in measuring sensitivity to insulin has greatly increased since Reaven described the X syndrome in 1988. This syndrome is an association of increased serum triglycerides, high blood pressure, low levels of HDL and insulin resistance; affected patients have an increased risk of coronary artery disease. Several methods have been described to measure insulin resistance. They rely on the effect of insulin on glucose metabolism. The euglycaemic clamp interprets the interactive feedback control between insulin and glucose. Insulin is infused to reach a constant serum insulin level, then glucose is administered at predetermined levels. The blood glucose level thus corresponds to hepatic production and insulin sensitivity is expressed as the amount of glucose infused in mg/kglmin. Metabolic glucose clearance is the relationship between glucose consumption and blood glucose level during the test. This method is reliable, reproducible and eliminates the errors related to insulinindependent metabolism. Steady state plasma glucose relies on drugs which suppress endogenous insulin secretion. Glucose and insulin are then infused at constant rates and the resulting glucose and insulin levels express insulin sensitivity. This method is technically difficult to handle and is used less often. Bergman's minimal model and the modified insulin tolerance test offer other more practical and global methods with particular indications. Each method has its advantages and disadvantages in terms of precision, cost and duration. The choice of a method is basically made as a function of the clinical or research needs.