Background: In a previous study, we reported a 4-fold increase in serum leptin following total parenteral nutrition given after surgery. We hypothesised that the perioperative fasting and stress contributed to this, possibly mediated by increased serum insulin and cortisol. Objective: To test the hypothesis that fasting, in combination with glucocorticoids, sensitises the leptin response to subsequent energy intake. Design: Healthy volunteers were randomised into two groups, one group received dexamethasone (DEX), 0.1 mg twice daily for 3 days, while the other group served as a control. Each group was then subdivided into two feeding protocols. Protocol 1, where a standard meal was given at 0, 24, 36 and 48 h and protocol 2 where the same meal was given at 0 and 48 h. Blood samples were drawn before, as well as every other hour during the study period for determination of serum leptin, insulin, glucose and cortisol concentrations. Results: In all groups serum leptin increased significantly following each meal ðP , 0:01Þ: The rise in serum leptin in response to the standard meal was higher when the meal was taken in the evening ðP , 0:001Þ or following longer duration of fasting ðP , 0:02Þ: In those fasting for 48 h, leptin decreased by 60% and showed no diurnal variation. DEX intake increased leptin concentrations in those fasting for short periods ðP , 0:02Þ but not for 48 h. Conclusions: Long durations of fasting sensitise the response of leptin to subsequent energy intake and abolish the DEX-induced upregulation of leptin. Meal timing is an important factor determining the leptin diurnal rhythm, but other factors must contribute since the leptin response to a standard meal taken in the evening was greater than to the same meal taken in the morning.