RESUMO
This study included all patients referred to the out-patient department of our sleep disorders centre from 1993 to 1999 on account of excessive daytime sleepiness (EDS). As a first step, patients in whom a diagnosis was established following appropriate polysomnography were excluded: this included sleep apnea syndrome, increased upper airway resistance syndrome, narcolepsy, periodic movements during sleep or other parasomnia, and epilepsy. Patients regularly taking psychotropic substances or with psychiatric disorders were also excluded. Finally, 128 patients remained in whom no clear diagnosis had been established for EDS, 70 women and 58 men, their ages ranging from 16 to 77 years. They underwent a 48-h recording (night 1-MSLT-night 2-continuous day). The aim of the study was to establish, define and characterise different groups of undiagnosed EDS patients using clinical, electrophysiological and immunological data with the help of hierarchical cluster analysis. Eight groups were characterised: group 1: mild hypersomnia type 1 (n = 11); group 2: hypersomnia frequently associated with HLA type DR2-DQw1 (n = 11); group 3: mild hypersomnia type 2 (n = 28); group 4: morning recovery from disrupted sleep (n = 19); group 5: young "long sleepers with difficulty at waking up" (n = 17); group 6: idiopathic hypersomnia (n = 15); group 7: poor or short sleepers since childhood (n = 8); group 8: older poor sleepers with a late onset of symptoms (n = 19). Characteristic features of these different groups provided consistent and objective arguments leading to a more precise diagnosis for these patients, and helped the initiation of appropriate management and treatment.