[Comparative analysis of patients with narcolepsy-cataplexy, narcolepsy without cataplexy and idiopathic hypersomnia]

To evaluate the distribution of clinical, electrophysiological and biological variables, and their relationship with the CSF hypocretin-1 levels, in patients with central hypersomnias diagnosed as narcolepsy-cataplexy (NC), narcolepsy without cataplexy (NnC) and idiopathic hypersomnia (IH) based on the ICSD-2 criteria.

We performed in all patients a clinical interview, a nocturnal polysomnogram and a multiple sleep latency test (MSLT), HLA analysis and measurement of CSF Hcrt-1 levels (low < or = 110 pg/mL).

Out of 51 patients, 31 were classified as NC, 11 as NnC and 8 as IH. 34 patients (66.7%) had low CSF Hcrt-1 levels (29 NC, 3 NnC and 1 IH). In the NC group, 96.1% were HLA DQB1*0602 positive and 91% had low CSF Hcrt-1 levels. The most frequent variables found in NC patients and in those with a low CSF Hcrt-1 levels were cataplexy, fragmented nocturnal sleep, short refreshing naps, automatic behavior, HLA DQB1*0602, and, in the MSLT, a short mean sleep latency, a higher number of REM sleep episodes and a short mean latency of REM sleep episodes. A long nocturnal sleep time and morning sleep drunkenness, 2 variables used in the ICSD-2 for the diagnosis of IH, were not different among the three groups of hypersomnias.

Central hypersomnias have a superposition of several clinical, electrophysiological and biological variables that makes sometimes difficult the differential diagnosis. The measurement of CSF Hcrt-1 levels may help in the diagnosis of those patients with unclear clinical or electrophysiological forms.