Until recently, olfactory dysfunction was an unknown feature of narcolepsy. Orexin A, also called hypocretin-1, is abnormally decreased or undetectable in the cerebrospinal fluid of narcoleptic patients with cataplexies. As hypothalamic orexin-containing neurons project throughout the entire olfactory pathway, from the olfactory mucosa to the olfactory cortex, disturbed orexinergic transmission may crucially be involved in impaired olfactory performance of narcolepsy patients. In our study we analysed the olfactory performance (threshold, discrimination, identification and sum score of these measurements, the TDI score) of narcoleptic patients with cataplexies (n = 10) and of age-, gender-, BMI- and smoker/non-smoker-matched healthy controls (n = 10). We then in a double-blind, randomized, placebo-controlled cross-over design applied orexin A intranasally to seven of the patients and measured 2-phenyl-ethyl alcohol (PEA) single-staircase odour detection thresholds. Compared to the controls, patients showed significantly lower scores for olfactory threshold (patients: median 8.0, range 4.0-10.5; controls: median 9.4, range 7.5-13.3; P < 0.05), discrimination (patients: median 12.5, range 10-15; controls: median 15.0, range 12-16; P < 0.005), identification (patients: median 13.0, range 10-16; controls: median 14.0, range 13-16; P < 0.05) and TDI score (patients: median 33.4, range 30-36; controls: median 38.4, range 35-43; P < 0.0001). In all patients, the PEA olfactory threshold score increased after administration of orexin A (median 11.5, range 6.5-13.25) compared to placebo (median 7.75, range 6.25-11.25; P < 0.05). Our results support the hypothesis that mild olfactory dysfunction is an intrinsic symptom of narcolepsy with cataplexies. The observation that intranasal orexin A restores olfactory function is in favour of this hypothesis. Furthermore, our data support that the pathophysiological mechanism underlying olfactory dysfunction in narcolepsy is the lack of CNS orexin.