FROM FUNCTIONAL TO IDIOPATHIC

  In his early work in Prague, Roth divided his patients into the categories of narcolepsy and hypersomnia in a more intuitive or subjective way than specialists are used to today. According to his system, in narcolepsy, sleep occurs quickly and forcefully, almost irresistibly, but does not last as long as with hypersomnia. Roth did not insist that cataplexy was necessary for an “essential narcolepsy” diagnosis, although it was present in about two-thirds of cases.14

  With people with hypersomnia, Roth noted that they usually had the ability to sit or lie down before sleep overwhelmed them, but their naps lasted longer than in narcolepsy, and their nightly sleep time was longer and steadier as well.

  From the point of view of the twenty-first century, an important element is missing from Roth’s classification scheme: REM sleep. It is unclear when Roth first learned about REM sleep. Early on, Roth did cite one paper by Dement on what was later called “paradoxical sleep” in cats,15 and he began recording eye movements in his own sleep lab in 1965.

  Half a century ago, the separation between neurological and psychiatric disorders was not as strict as it is today. In Roth’s 1957 book Narcolepsy and Hypersomnia from the Aspect of Physiology of Sleep, he divided hypersomnias into the categories of “organic” and “functional.” “Organic” meant that the origin was clear or structurally discernible, coming from encephalitis, traumatic injury, or a tumor. “Functional” held everything outside that category, including what would later be called IH.

  In neurology, the word functional sometimes has a connotation of “merely psychological.” For Roth, functional hypersomnia was a big tent. He gave some functional cases the label “vegetative dystonia,” because of symptoms like headaches, dizziness, or digestive problems. Others were neurotic, with “hysterical personality” as one contributing factor. Roth was willing to consider physiological causes, such as genetics, endocrine disturbances, or chronic sleep deprivation, for the neurotic type of functional hypersomnia. Over time, Roth reevaluated apparent neurotic cases and decided that most of them belonged in the IH category, because of heritability in some cases, along with the stable character of the symptoms.

  Hypersomnia with sleep drunkenness, or simply sleep drunkenness, began as a third category of functional hypersomnia. In the 1950s version of his book, sleep drunkenness gets a separate chapter. At that point, we can see a contrast between how Roth thought of sleep drunkenness and other types of functional hypersomnia. Generally, he recommended psychotherapy or behavioral therapy for most cases of functional hypersomnia, sometimes in combination with the mild stimulant ephedrine. He advised avoiding amphetamines, because of side effects: “Psychoton [amphetamine] is only prescribed by exception in functional hypersomnia, in cases in which all therapeutic methods have not succeeded, and even then only temporarily. If possible, we do not give psychoton in those cases where a larger neurotic component has been detected.”16

  In contrast to other types of functional hypersomnia, with sleep drunkenness, Roth was not shy about recommending strong stimulants, such as amphetamine—the same treatments he favored for narcolepsy. He was also enthusiastic about methylphenidate (Ritalin) and phenmetrazine (Preludin), both of which first appeared in the 1950s. For sleep drunkenness, he specifically recommended a low dose of amphetamine before bed to make waking up easier in the morning.

  CAVEAT AND CONTEXT

  Would Roth’s patients receive the same diagnoses today? Roth started forming his ideas about hypersomnia before it was recognized how common sleep apnea was. He began studying sleep apnea in the 1960s, and he thought it was responsible for a small number of cases in his cohort. In hindsight, Nevšímalová said she thought some cases Roth diagnosed as idiopathic hypersomnia may have actually been sleep apnea. In some cases of functional hypersomnia, Roth did notice obesity and headaches, which are often associated with sleep apnea. However, he reported that half had low blood pressure and half normal; this is less characteristic of sleep apnea.

  Putting sleep apnea aside, to appreciate Roth’s insights, we need to ask why somebody hadn’t identified IH before. At the time, the dominant view in English-language medical literature was that long-lasting cases of sleepiness, which were sometimes given the label hypersomnia, were psychiatric disorders. Hypersomnia was considered a symptom, not a disease of its own. Other diseases that enhanced sleepiness were considered to have an endocrine or psychiatric origin. Epidemic encephalitis (discussed more in chapter 11) attracted attention in the 1920s and 1930s but then faded away. Neurologists had enough difficulty wrestling narcolepsy away from psychiatrists and, in effect, decided that hypersomnia was too vague or borderline to bring with it.

  The 1950s marked the height of Freudian influence in North American psychiatry. During this period, hypersomnia mainly showed up in English-language medical literature as “psychogenic,” and it was portrayed as an escape mechanism. A 1959 paper from the Mayo Clinic in the American Journal of Psychiatry summarized twelve cases with long, inappropriately timed and unrefreshing naps, combined with disturbed nighttime sleep.17 The authors wrote: “The pattern of unusual drowsiness and undue sleep was consistently a means for passive retreat from conflict.”

  In contrast, Freud and psychoanalysis were not respected in 1950s Czechoslovakia, where “psychiatric research was primarily biologically orientated, mental health promotion was marginalized, and those who suffered from severe psychiatric disorders were confined in psychiatric hospitals.”18 The ideological environment may have enabled Roth to move ahead with physiologically oriented research on narcolepsy and hypersomnia.

  More broadly, Roth’s contribution was recognizing that there was a group of people for whom hypersomnia was the root of their problems. Their excessive sleep did not resemble narcolepsy and was not obviously tied to encephalitis or traumatic brain injury. As a first step, seeing sleep drunkenness as a distinct neurological symptom, analogous to cataplexy, may have made this possible.

  WORLD TRAVELER

  The early 1950s, when Roth was forming some of his concepts, was a bleak period for Czechoslovakia, especially for those from a Jewish background. The communist government was consolidating its power with mass arrests, antisemitic purges, and show trials. Some of Roth’s cousins and family friends were imprisoned, his son said. Roth was unable to travel abroad, and communications were limited.

  Later in that decade, the political situation in Czechoslovakia relaxed enough for Roth to travel. In 1957, Roth was able to spend six weeks in France with Henri Gastaut, an authority on epilepsy. When Roth came to work with him, Gastaut was exploring the EEG signs of altered states of consciousness, such as syncope or fainting.19

  In Marseilles, Roth refined his knowledge of EEG. With Gastaut’s advice, he was able to analyze sleep recordings more rigorously. From their collaboration, a paper emerged on narcolepsy.20 Examining both Czechoslovak and French patients, Roth and Gastaut concluded that “narcolepsy and cataplexy are transient, non-critical phenomena that have nothing to do with epilepsy.… Essential narcolepsy must therefore depend on a functional disorder of the structure responsible for the vigilant state and sleep regulation.”

  Roth’s book on narcolepsy and hypersomnia, translated into German but not English, received respectful reviews. In 1961, Roth published a paper in the then-premier journal Electroencephalography and Clinical Neurophysiology on the EEG signs of “lowered vigilance,” a transitional state on the way to sleep.21 He also presented his work on narcolepsy at a conference in Rome. This exposure seems to have brought Roth to the attention of other sleep researchers.

  At Charles University, Roth’s career prospered. He became the go-to person for others studying external influences—light stimulation, radiation, decompression sickness, and industrial poisons—on brainwaves.22 In 1963, Roth published a paper with his wife, Nada, a pediatrician, who was studying epilepsy in children.23 Anniki Rothova recalled that when her mother had medical duties in the evening, “it was great fun” because instead of cooking dinner, her father would take her to a nearby cafeteria. Around this time, Roth and his family acquired a yellow Skoda, for which they’d had to save for several years. It would have been considered small in other countries, but for them, it was a luxury.

  CHICAGO AND SANTA MONICA

  In 1967, Roth came to the University of Chicago to work with Allan Rechtschaffen, a leading sleep researcher there. At that point, during a temporary political thaw in the mid-1960s, foreign travel was complicated but possible. This was Roth’s chance to study people with hypersomnia in the sleep lab overnight, which he had not been able to do conveniently in Prague.

  Some of Roth’s peers reportedly thought hypersomnia with sleep drunkenness was something peculiar to Czechoslovakia.24 To confirm that people with hypersomnia could be found elsewhere, Rechtschaffen had classified ads placed in the Chicago Tribune: “WANTED—For research studies, patients with clinical hypersomnia and postdormital hypersomnia and postdormital confusion but without narcolepsy. Call Sleep Lab, University of Chicago.”

  Roth was in Chicago for just a few weeks, staying in the university’s housing for international students. He and Rechtschaffen stayed up several nights examining the people they recruited. In these experiments, Roth was able to see that people with hypersomnia did not enter REM sleep quickly and had normal proportions of REM and non-REM sleep; in fact, the organization of sleep was “completely normal apart from its long duration.” Roth, Nevšímalová, and Rechtschaffen later wrote: “It is clear that the disorder is not specifically related to any single sleep stage. Rather, the disorder appears as an extension and ‘intensification’ of normal sleep.”25

  In 1967, Roth was part of a group of notable figures in sleep research: the Rechtschaffen and Kales committee.26 Roth was one of three from outside the United States, along with Michel Jouvet from France and Ian Oswald from the United Kingdom. They produced a standard handbook to evaluate polysomnography studies, which was used in the field for four decades. The meetings, held in Santa Monica, were contentious, since each lab had developed its own ways of reading the voluminous paper graphs produced by overnight studies. Rechtschaffen barred the door and told attendees they were not allowed to leave until they came to a consensus.27

  INVASION AND NORMALIZATION

  Roth was able to return to Prague and confirm that only narcolepsy with cataplexy involved quick entry into REM sleep, while “independent narcolepsy” (that is, narcolepsy without cataplexy) and hypersomnia did not. In the 1970s, he and Nevšímalová eventually performed overnight studies in Prague with their colleague Le Van Thanh, with similar results to what he found with Rechtschaffen: the REM/non-REM sleep architecture in hypersomnia was preserved.

  Roth’s work with Rechtschaffen did not appear in Archives of General Psychiatry until 1972. Geopolitics had intervened with the 1968 Soviet invasion of Czechoslovakia. Roth was depressed and nearly immobile for weeks after the invasion, according to his daughter. He hesitated over whether to accept a job offer in Western Europe but ultimately declined.

  Roth refused to sign a declaration of loyalty to the new government, something many in Czechoslovakia had to do to keep their jobs.28 The incoming chairman of the neurology department, Josef Vymazal, protected Roth from losing his post, but he was demoted and could not be listed as a faculty member who interacted with students.29

  In 1973, Roth’s daughter became sick of communist indoctrination and immigrated to the Netherlands, becoming an ophthalmologist. Her departure was a blow and contributed to another period of depression for Roth. He continued to write his daughter regularly.

  Political restrictions eventually eased enough for Roth to travel again, although he needed permission from a series of officials, including the minister of health, to leave the country. He participated in landmark conferences, such as the First International Symposium on Narcolepsy, held in southern France in July 1975. Roth’s talk at that meeting, titled “Functional Hypersomnia,” contains some of his clearest advocacy on behalf of people with IH: “It is necessary to stress, however, that the symptoms of idiopathic hypersomnia are, without any doubt, pathological manifestations. These patients suffer considerably and have many difficulties in their professional as well as their private lives. This is why they so often seek medical help.”

  INFLUENTIAL FRIENDS

  In the 1970s, Roth became friends with other sleep researchers, such as Michel Billiard in France and Roger Broughton in Canada. Broughton, who had also trained with Gastaut, was head of the sleep disorders clinic at Ottawa General Hospital. His research interests ranged from early work on sleepwalking and night terrors to circadian rhythms.

  Broughton and Roth forged a professional partnership. Broughton arranged for Roth to speak at international conferences and facilitated his travel. Roth’s patients became part of Broughton’s surveys on the social and economic burdens of narcolepsy. Roth presented data on his IH patients at the 1978 European Congress on Sleep Research, writing: “It is evident that patients with idiopathic hypersomnia experience very serious effects on a number of parameters including work, education, personality, memory and so forth, as has also been documented for narcoleptics.… These socio-economic consequences, in fact, often exceed those of narcoleptics. This is perhaps due to the greater amounts of daytime sleep and of the intensity and duration of diurnal drowsiness.”

  Broughton helped Roth revise his book on narcolepsy and hypersomnia, which was finally translated into English.30 Since the 1950s, the science of sleep had made great advances, so they revamped most of the book, and Broughton cowrote five chapters. When discussing IH in the later edition of his book, Roth included a monosymptomatic form, characterized by excessive daytime sleepiness only, and a polysymptomatic form, which also included long sleep time and sleep drunkenness upon waking. Roth wrote: “I originally believed that those patients who, in addition to diurnal sleepiness also suffered from deep and protracted nocturnal sleep and from sleep drunkenness on arousal, had a separate nosological entity, which I called hypersomnia with sleep drunkenness. However, I later came to the conclusion that the two clinical pictures were variants of the same disease.”31

  It is possible to see Roth’s imprint on the 1979 DCSAD (Diagnostic Classification of Sleep and Arousal Disorders), whose hypersomnia sections were edited by Broughton. There IH is called “idiopathic CNS hypersomnolence,” although the descriptions do not exactly match Roth’s.32 We can also see Roth’s influence on publications from Dement’s and Guilleminault’s group at Stanford. Their 1975 paper on “hypersomnia with automatic behavior,” submitted just before the meeting in France, makes no reference to Roth.33 Jumping ahead to 1986, their comparison of narcolepsy and IH says in the introduction that Roth was “the first to clearly isolate the idiopathic hypersomnia syndrome.”34 At the same time, Guilleminault’s identification of “upper airway resistance syndrome” (UARS) as a borderline condition for sleep apnea chipped away at IH. Some patients who might have been diagnosed with IH were diagnosed with sleep apnea or UARS instead, because of clinicians’ improved ability to detect disturbances of breathing during sleep.

  In the 1980s, Roth and Nevšímalová adapted to new trends in sleep medicine. They and their patients contributed to early genetic studies of narcolepsy.35 They also criticized the MSLT, the emerging diagnostic standard. Their alternative, tested in a small number of people, took into account both sleep latency and the duration of sleep stages entered during daytime naps.36

  Roth died in November 1989 of Hodgkin’s lymphoma, just a few days before the Velvet Revolution. His friend Broughton has argued that he was “the true father of sleep medicine.”37

  CHAPTER 6

  THE ESSENCE OF SLEEPINESS

  In the case of narcolepsy, are there not cases where a metabolic problem leads to the more rapid and intense production of the hypnotoxin? Research into a toxic activity of this type in the CSF of such patients would be quite desirable.

  —Henri Pieron, 1927

  Let’s return to Emory, where as early as mid-2007, investigators thought that Anna’s cerebrospinal fluid contained something that was making her extremely sleepy. The identity of this substance was unclear. Because flumazenil worked well for Anna, Parker had invoked the term “endozepine,” which has a history of its own. In the late 1970s, researchers at the National Institute of Mental Health had been hunting for what they called “the brain’s Valium.” Their idea was that benzodiazepine drugs had natural counterparts: endogenous benzodiazepines, or endozepines, which might reveal something fundamental about how anxiety was regulated.1 What they found turned out to be more complicated, and by the twenty-first century, the concept of endozepines had fallen out of favor.

  Despite language that Rye would later use to explain what they had found, he and Jenkins thought the GABA-enhancing activity was something distinct from a benzodiazepine. A plan detailing how they proposed to identify it can be found in a 2009 grant application.2 This was one of the first places where Rye unfurled his somnogen flag: “We were compelled in the course of this work to consider an alternative hypothesis—namely, that hypersomnia might be caused by excess of a naturally occurring somnogenic substance. Indeed, we have discovered that an endogenous, positive allosteric modulator of GABA-A receptors causes hypersomnia and sleepiness in humans.”

  Rye and a scientist in his lab, Glenda Keating, had been investigating a molecule called Cocaine and Amphetamine-Regulated Transcript, or CART. They had thought CART might be like hypocretin: important for keeping people awake and missing in narcolepsy. However, measurements of CART in CSF samples from people with sleep disorders revealed that CART levels correlated more with the use of stimulants than with sleepiness. CART was not the next hypocretin.