Study Objectives: Previous laboratory studies in narcolepsy patients showed altered core body and skin temperatures, which are hypothesised to be related to a disturbed sleep wake regulation. In this ambulatory study we assessed temperature profiles in normal daily life, and whether sleep attacks are heralded by changes in skin temperature. Furthermore, the effects of three months of treatment with sodium oxybate (SXB) were investigated. Methods: Twenty-five narcolepsy patients and 15 healthy controls were included. Core body, proximal and distal skin temperatures, and sleep-wake state were measured simultaneously for 24 hours in ambulatory patients. This procedure was repeated in 16 narcolepsy patients after at least 3 months of stable treatment with SXB. Results: Increases in distal skin temperature and distal-to-proximal temperature gradient (DPG) strongly predicted daytime sleep attacks (P < 0.001). As compared to controls, patients had a higher proximal and distal skin temperature in the morning, and a lower distal skin temperature during the night (all P < 0.05). Furthermore, they had a higher core body temperature during the first part of the night (P < 0.05), which SXB decreased (F = 4.99, df = 1, P = 0.03) to a level similar to controls. SXB did not affect skin temperature. Conclusions: This ambulatory study demonstrates that daytime sleep attacks were preceded by clear changes in distal skin temperature and DPG. Furthermore, changes in core body and skin temperature in narcolepsy, previously only studied in laboratory settings, were partially confirmed. Treatment with SXB resulted in a normalisation of the core body temperature profile. Future studies should explore whether predictive temperature changes can be used to signal or even prevent sleep attacks.