BACKGROUND AND AIM: Fat mass generation requires an energy surplus and the activity of the peroxisome proliferator-activated receptor gamma (PPARgamma). We investigated if the PPARgamma ligand rosiglitazone influences substrate usage, energy expenditure (EE) and energy intake (EI) and, thereby, how PPARgamma activity contributes to susceptibility to obesity. METHODS: Twenty healthy males (20-29 years) were randomly assigned to receive a placebo (n = 10) or rosiglitazone (8 mg/d) (n = 10) for seven consecutive days, while staying in a respiration chamber. Food intake was ad libitum. Body composition was determined by underwater weighing (day 1) and deuterium dilution (day 1 and 8). RESULTS: Mean (+/-SE) EI was 15.9 +/- 0.9 MJ/d in the placebo group and 18.9 +/- 1.2 MJ/d in the rosiglitazone group. Mean EE was 11.3 +/- 0.3 MJ/d and 12.5 +/- 0.5 MJ/d for the placebo and rosiglitazone groups respectively. This resulted in a cumulative positive energy balance (EB) of 32.3 +/- 5.1 MJ for placebo and 44.7 +/- 6.9 MJ for rosiglitazone. There were no significant differences in EI, EE, and EB between treatments. Both groups did not adjust their fat oxidation to the increased fat intake, but fat oxidation decreased faster in the rosiglitazone group (significantly lower on days 6 and 7). During treatment with rosiglitazone, significantly more fat storage was seen in overweight subjects while this was not the case in the placebo group. CONCLUSIONS: Our results suggest a shift in substrate usage during PPARgamma stimulation leading to a preference for fat storage, especially in subjects with a higher BMI. Copyright (c) 2005 John Wiley & Sons, Ltd.