Beginning in the early 1950s, Prof. Roth systematically extended his clinical studies of patients with daytime somnolence, and in 1957, he clinically analyzed a cohort of 248 cases [3]. These were divided into two groups: 155 cases of narcolepsy and 93 cases of different types of hypersomnia. The latter cases were classified as: (1) functional type (50 cases), in which pathological sleep was not induced by any known disease, (2) cases of organic origin (29 patients) determined by some known underlying disease, and (3) sleepiness with post-dormital drunkenness (14 cases), specified later as the idiopathic form of hypersomnia with sleep drunkenness.

The discovery of rapid eye movement (REM) sleep [4–6] gave impetus to polysomnographical (PSG) studies of patients with daytime somnolence, previously regarded as narcolepsy. Dement et al. [7] were the first to suggest that patients affected by excessive diurnal somnolence, but not accompanied by signs of REM sleep, and symptoms of cataplexy, hypnagogic hallucinations, or sleep paralysis, should be considered to be suffering from hypersomnolence other than narcolepsy.

At that time, Bedrich Roth had no opportunity to study nocturnal sleep recordings in Prague. That was why he decided to accept an invitation from Allan Rechtschaffen to visit his sleep laboratory in Chicago and examine patients with sleep drunkenness in the USA. The birth of a new clinical entity supported by PSG findings seemed rather amusing there. When Bedrich Roth arrived, everybody in the USA believed that this disease existed only in Prague. However, Roth arranged a short interview in the local television explaining the clinical symptoms of the disease (long nocturnal sleep with difficulty awakening and long-lasting daytime naps) and asked TV viewers for cooperation. Everybody in the Chicago team was really surprised to see, exactly then and there, the TV show lineup of people waiting to be examined by Bedrich Roth. After a clinical interview, he chose ten patients and the first PSG findings of this ailment were published [8] in patients with idiopathic hypersomnia who underwent PSG recording for two nonconsecutive nights. The organization of sleep was completely normal except for its long duration (12 h or more). The percentage of REM and nonrapid eye movement (NREM) sleep was normal, as was the periodicity of the sleep cycles, the number of which was simply increased. These findings were later published [8].

Three years later, a complete clinical description of hypersomnia with sleep drunkenness (58 cases), enriched by long-term nocturnal monitoring (9 cases), appeared in the literature [9] giving a clear picture of this clinical entity. Sleep drunkenness was characterized by difficulty awakening accompanied by confusion, disorientation, poor motor coordination, slowness, and repeated dosing off. Patients reported that these symptoms occurred almost every morning, and nearly all reported abnormally prolonged sleep. Of 58 cases of hypersomnia with sleep drunkenness, 52 were apparently idiopathic and 6 were possibly symptomatic of organic brain disturbance. A familial history of the disorder was found in 36 % of the idiopathic cases. No specific EEG or PSG abnormalities were noted except for relatively increased heart and respiratory rates and extended sleep.

In the late 1960s, the Prague school focused on the pathophysiology of narcolepsy and different types of hypersomnia [10]. Narcolepsy seemed to be associated with REM sleep disturbances and in most instances also with disturbances in NREM sleep, whereas hypersomnia was regarded as involving exclusively the NREM system. The authors assumed that most of the independent narcolepsy cases (without cataplexy) had a mechanism similar to that in hypersomnia patients. This hypothesis was supported also by study of dreams [11]. According to clinical data analyzing 451 patients, 200 were diagnosed with idiopathic narcolepsy, 78 with symptomatic narcolepsy , 47 with hypersomnia with organic basis, and 114 with hypersomnia without organic basis (31 of whom with sleep drunkenness), 2 with independent cataplexy, and 10 with independent sleep paralysis. Hypnagogic hallucinations and vivid, terrifying dreams were frequent in narcolepsy, especially in those suffering also from cataplexy and/or sleep paralysis. In hypersomniac patients, these symptoms were rare. Polygraphic examination of 75 daytime recordings with 215 awakenings showed that 97.4 % of patients awakened during paradoxical sleep reporting dreams; 80 % of them had experienced vivid dreams with a strong affective component and visual and acoustic perceptions. During synchronous sleep, dreams were reported in 34 % of awakenings, usually with vague content. Vivid dreams occurred in only 10 % of awakening during synchronous sleep, and these came mostly from within 10 min before or after paradoxical sleep.

Although the first description of a familial occurrence of hypersomnia was reported in Roth’s monograph [3], it was only rarely mentioned in later publications. In 1968, Bonkalo [12] described two siblings with a pure form of hypersomnia. A larger material was published in the early 1970s again by the Czech authors [13–14]. They wrote a genealogical study of the families of 30 patients with hypersomnia and 100 patients with narcolepsy . Idiopathic hypersomnia was found to run in the families of more than one third of the cases. The existence of transition from hypersomnia to isolated narcolepsy in patients with heredofamilial occurrence showed a pathogenetic relationship of these disturbances. According to the authors, transfer of the hereditary predisposition towards hypersomnia and isolated narcolepsy is most probably of an autosomal dominant type, while in narcolepsy with cataplexy and other symptoms of sleep dissociation, a multifactorial type of heredity was supposed.

In 1976, Prof. Roth published a review of 642 personally observed cases including 368 cases of narcolepsy and 274 cases of hypersomnia [15]. The largest group of hypersomniac patients consisted of so-called functional hypersomnias (213 cases). These were divided into a group with short sleep cycle (191 cases) and another with long sleep cycle (22 cases). The author distinguished two main forms of short-sleep-cycle hypersomnia: (a) idiopathic monosymptomatic form (71 cases), marked solely by excessive daytime sleepiness with long naps, and (b) idiopathic polysymptomatic form (103 cases) , in which daytime sleepiness was accompanied by prolonged nocturnal sleep and usually also by awakening difficulties (sleep drunkenness or sleep inertia). The rest of functional short-cycle hypersomnias were patients with neurotic hypersomnia (5 cases) and hypersomnia with disorders of breathing during sleep (12 cases). However, nocturnal polygraphic recordings were made only in a minority of these cases, which is why the last mentioned group may have been underdiagnosed.

In the first diagnostic classification of sleep disorders [16], idiopathic hypersomnia was referred to as idiopathic central nervous system (CNS) hypersomnia as one of the disorders of excessive somnolence. The distinction between the two forms proposed by Roth was left out.

In a monograph Narcolepsy and hypersomnia, published one year later [17], Roth described a carefully selected group ( n = 167) of idiopathic hypersomnia patients. He characterized this disease as a short-cycle “functional” hypersomnia, not caused by known organic brain disease or by metabolic or toxic condition or of psychogenic origin. Its clinical picture included a short sleep onset and frequently prolonged nocturnal sleep with difficulty awakening in the morning, and accompanied by psychological and autonomic dysfunction including sexual disturbances. Two forms of short-cycle functional hypersomnia—monosymptomatic and polysymptomatic—have a chronic course and severe socioeconomic impact. He drew attention to its relationship to idiopathic narcolepsy, especially the monosymptomatic form, without cataplexy and other disassociated sleep dysfunction. For the treatment, he recommended central stimulants similar to those for narcolepsy. This excellent book served as the most important textbook for physicians and sleep researchers for a long time, as well as for the patients suffering from daytime sleepiness.

In 1981, Roth et al. [18] published a detailed study of neurological, psychological, and polygraphic findings in sleep drunkenness. Eight patients with idiopathic hypersomnia and eight controls were tested after normal sleep duration (patients 12 h, controls 8 h), and after sleep deprivation (patients after 8 and 6 h, controls after 4 and 0 h of nocturnal sleep). A state of sleep drunkenness, characterized by “microsleep” in polygraphic recording, was found in 19 of the patients, but only once in the controls. Clinically prominent features included cerebellar signs, hyporeflexia or areflexia, signs of vestibular involvement, and fine and gross motor dysfunction. The authors presumed that sleep drunkenness develops as a result of chronic relative sleep deprivation in those patients, whose sleep requirements are greater than in normal individuals.

Figure 27.2 illustrates a group of sleep researchers organizing a Symposium on Narcolepsy and Hypersomnia in Prague in honor of Prof. Roth .

A detailed description of nocturnal sleep as well as Multiple Sleep Latency Test (MSLT) results comparing different disorders of excessive daytime somnolence (EDS) came from the Stanford group in the early 1980s [19]. The largest group in a 100-patient cohort consisted of narcoleptic patients (41 with cataplexy, 5 without cataplexy). The rest of the EDS patients formed a rather heterogeneous group: idiopathic CNS hypersomnia (17), EDS associated with psychological and/or psychiatric problem (18), irregular sleep pattern (5), insufficient (disturbed) nocturnal sleep (4), abuse of stimulant drugs (3), neurological conditions (2). In five patients, EDS was associated with no objective abnormality. The authors found a clear intergroup difference in the nocturnal as well as daytime polygraphic examinations. Narcoleptics showed more severe EDS with a shorter MSLT latency and presence of sleep-onset rapid eye movements (SOREMs) (at least two, although their number varied even in the same patient) as compared with others. REM latency during the night was shorter; they had fewer REM segments and more awakenings and myoclonic jerks during sleep. In the MSLT, narcoleptics had a mean sleep latency of 3.3 min (standard deviation (SD) ± 3.3), patients with idiopathic CNS hypersomnia 6.5 min (SD ± 3.2), and patients with psychological disturbances and those with no objective abnormalities 10.6 min (SD ± 5.2) and 10.9 min (SD ± 3.9), respectively. Based on these data, the authors concluded that a mean sleep latency of 5.5 min and less indicates pathological sleepiness which was found in the majority of narcoleptic patients. A value between 6 and 10 min is a “gray area,” typical of idiopathic CNS hypersomnia, and mean sleep latency of 10 min and more indicates that pathological sleepiness is unlikely.