Sleep profoundly affects metabolic pathways. In healthy subjects, experimental sleep restriction caused insulin resistance (IR) and increased evening cortisol and sympathetic activation. Increased obesity in subjects reporting short sleep duration leads to speculation that, during recent decades, decreased sleeping time in the general population may have contributed to the increasing prevalence of obesity. Causal inference is difficult due to lack of control for confounders and inconsistent evidence of temporal sequence. In the general population, obstructive sleep apnoea (OSA) is associated with glucose intolerance. OSA severity is also associated with the degree of IR. However, OSA at baseline does not seem to significantly predict the development of diabetes. Prevalence of the metabolic syndrome is higher in patients with OSA than in obese subjects without OSA. Treatment with continuous positive airway pressure seems to improve glucose metabolism both in diabetic and nondiabetic OSA but mainly in nonobese subjects. The relative role of obesity and OSA in the pathogenesis of metabolic alterations is still unclear and is intensively studied in clinical and experimental models. In the intermittent hypoxia model in rodents, strong interactions are likely to occur between haemodynamic alterations, systemic inflammation and metabolic changes, modulated by genetic background. Molecular and cellular mechanisms are currently being investigated.