The present review focuses on some aspects of diet and exercise behaviour as they affect insulin sensitivity in both diabetic and healthy subjects. Insulin resistance, a less than normal biological effect for a given insulin concentration, is present in a number of common conditions including non-insulin-dependent diabetes mellitus (NIDDM), essential hypertension, hypertriacylglycerolaemia and obesity. Evidence that insulin resistance may be important in the pathogenesis of these conditions, and the fact that they are frequently present in the same patient, has led to speculation that insulin resistance is a common aetiological factor (Reaven, 1988). There is also both epidemiological and experimental support for the view that raised circulating insulin concentrations, a consequence of impaired insulin sensitivity, are a risk factor for vascular disease (Pyorala, 1979; Ducimetiere et al. Stout, 1992). The ability of pharmacological agents used in the treatment of both MDDM and essential hypertension to alter insulin action is relevant to the choice of therapy in these diseases (Harper et al. 1996).Against this background of the potential importance of insulin resistance both to causation and treatment of common conditions, which carry a significant morbidity and mortality, effects of changes in diet and exercise upon insulin sensitivity are of considerable interest. In the present review only studies in human subjects are considered and the emphasis is on work where effects of insulin on glucose metabolism have been measured directly, usually involving the euglycaemic hyperinsulinaemic clamp technique.
EXERCISE AND INSULIN SENSITIVITY IN MANOver 25 years ago Bjorntorp et al. (1970) demonstrated lower insulin concentrations in response to oral glucose in obese patients after a programme of exercise which improved physical fitness. The effects were independent of changes in body weight. Since blood glucose concentrations did not change, the lower circulating insulin concentrations were interpreted as indicating increased sensitivity to insulin. Bjorntorp et al. (1972) also showed in cross-sectional studies that fit subjects had lower insulin concentrations than less-fit individuals. Subsequent workers have measured whole-body insulin action directly using the euglycaemic hyperinsulinaemic clamp technique and confirmed that those exercising regularly are more insulin sensitive than sedentary subjects (Rosenthal et al. 1983).Improved insulin sensitivity follows a single bout of exercise and this effect wears off after 2-5 d (Schneider et al. 1984;Mikines et al. 1988Mikines et al. , 1989. It has been considered that acute effects probably account for most of the improvements in insulin sensitivity seen during physical training programmes. Thus, whilst training in association with weight loss improves insulin action (Bogardus et al. 1984), if weight is maintained by increased energy intake and the effect of the last bout of exercise is controlled, insulin sensitivity remains unaltered despite an improvement in cardio...