• rapid eye movement sleep;
  • REM sleep propensity;
  • short-term REM sleep homeostasis;
  • sleep cycle;
  • sleep deprivation;
  • ultradian rhythms

We measured cerebrospinal fluid (CSF) hypocretin-1 levels in 11 patients with narcolepsy–cataplexy, five with narcolepsy without cataplexy and 12 with idiopathic hypersomnia (IHS). All patients were Japanese. As reported in Caucasian patients, undetectable or very low hypocretin-1 levels were observed in most (9 out of 11) Japanese narcolepsy–cataplexy patients. Our hypocretin-deficient narcoleptics included three prepubertal cases within few months after the disease onset. All nine hypocretin-deficient patients were human leuckocyte antigen (HLA) DR2 positive, while two who had normal CSF hypocretin-1 levels were HLA DR2 negative. In contrast, none of the narcolepsy without cataplexy and IHS subjects had undetectable low levels. Low CSF hypocretin-1 is therefore very specific for HLA DR2 positive narcolepsy–cataplexy, and the deficiency is likely to be established at the early stage of the disease.