Cardiorespiratory fitness, which is determined mainly by the level of physical activity, is inversely related to mortality in the general population as well as in subjects with diabetes, the incidence of which is also increased by low exercise capacity. Exercise is capable of promoting glucose utilization in normal subjects as well as in insulin-deficient or insulin-resistant diabetic individuals. In diabetic subjects treated with insulin or insulin secretagogues, exercise may also result in complications, with too much insulin causing hypoglycaemia and not enough insulin leading to hyperglycaemia and possibly ketoacidosis; both complications may also occur several hours after exercise. Therefore, self-monitoring of blood glucose before, during (for exercise duration of more than 1 h) and after physical exercise is highly recommended, and also carbohydrate supplementation may be required. In the Italian Diabetes Exercise Study (IDES), measurement of blood glucose and systolic and diastolic blood pressure levels before and after supervised sessions of combined (aerobic + resistance) exercise in type 2 diabetic subjects with the metabolic syndrome showed significant reductions of these parameters, though no major hypoglycaemic or hypotensive episode was detected. The extent of reduction of blood glucose was related to baseline values but not to energy expenditure and was higher in subjects treated with insulin than in those on diet or oral hypoglycaemic agents (OHA). Thus, supervised exercise training associated with blood glucose monitoring is an effective and safe intervention to decrease blood glucose levels in type 2 diabetic subjects.