Trotti and her colleagues were concerned enough about participants dozing off during imaging to monitor them via EEG, revealing that falling asleep in the scanner was common—even among controls. It suggests that imaging may be capturing a mixture of states when people are either asleep or trying strenuously to stay awake, rather than placidly daydreaming. In future studies, EEG during imaging to control for actual sleep could become standard practice.

  THE WRIST IS NOT THE BRAIN

  We’ve been discussing what neurobiology IHers may have in common. One factor that separates them may be the sheer amount of time they spend sleeping, since the long sleep form of IH may be distinct in its pathophysiology. Over the last decade, American, European, and Japanese sleep neurologists have begun to agree that a measure of habitual sleep time or overall sleep need should be used to diagnose IH, rather than relying on the MSLT. Several research groups have found that how quickly someone enters sleep on the MSLT does not match up well with increased sleep need or decreased alertness.

  It’s difficult to argue that someone is getting insufficient sleep when they report sleeping eleven hours a day or more. But then the question becomes how a doctor or other clinician establishes whether someone really does sleep that much. In practice, listening to the patient is important, but simply asking them what they remember is not that reliable. People often make errors in estimating actual shut-eye time, both in recall and in sleep diaries.

  A currently recommended clinical practice is to have a patient wear a wrist actigraphy device for a week or two, particularly before an MSLT. The devices are similar or the same as those that tell people how many steps they’ve taken and are convenient to take home. The devices don’t detect sleep but can tell whether someone is motionless for enough time that they are probably asleep.

  Plenty of studies compare actigraphy to the gold standard of polysomnography in the sleep lab—and some findings have been discouraging. In a study from Trotti and colleagues, actigraphy devices did not appear to perform consistently enough to help make decisions for individuals.34 In one study, 102 patients, most of whom were undergoing evaluation for sleep apnea, were asked to wear actigraphy devices while they were undergoing an overnight sleep exam. This allowed investigators to compare their measurement of sleep time against standard EEG.

  The Emory study evaluated the performance of two devices: the Philips Actiwatch, which is FDA cleared for clinical and research use, and the Jawbone UP3, a consumer product. On average, the Actiwatch overestimated sleep time by more than twenty-five minutes, and the Jawbone did so by almost one hour. On the individual level, the authors concluded that “a clinician attempting to apply the Actiwatch result would not know whether it was overestimating sleep time by more than 2 hours or underestimating it by almost 1.5 hours.”

  Plante’s group has performed similar studies on consumer sleep devices’ shortcomings.35 A Finnish study was more optimistic, finding close agreement between actigraphy and polysomnography when performed at home in a group of 281 patients being evaluated for sleep disorder diagnosis.36 The authors concluded: “Actigraphy is far from being perfect, but there is no better method to assess sleep time across several nights.”

  It’s an academic concern for now, but in the future, precise diagnosis may depend on accurately measuring how many hours someone sleeps per week at home. In the United States, insurance reimbursement for actigraphy with sleep disorder diagnoses has been inconsistent, and many insurers’ policies say that the practice is not reliable enough to be reimbursed.37

  Better technology could help. People with IH and related disorders could benefit from knowing whether medication changes or low-key interventions make a verifiable difference in their sleep quality. An overnight sleep test is expensive, costing thousands of dollars—and it only takes a one-night snapshot of someone’s sleep. The tension between accuracy in the sleep lab versus convenience and lower cost at home exists in other aspects of sleep medicine. In the last decade, sleep apnea diagnosis has been shifting to the home, in response to pressure from insurance.38 The end result is more convenient, but both patients and physicians get less information out of the encounter.

  While actigraphy devices that only track motion may not be accurate enough to measure sleep for diagnostic purposes, multisensor trackers, monitoring heart rate and sometimes temperature and skin conductance, may approach the level of accuracy needed.39 Portable EEG headsets, earbuds, or patches that adhere to the body may be able to record sleep phases and measure sleep duration at home with sufficient accuracy. Such devices represent the future of the field, whether clinicians want to deal with the flood of consumer technology or not. Apple and other consumer electronics companies clearly see the potential for a broad market and are likely to continue to produce more convenient gadgets in the future.

  MARATHON SLEEP TESTS

  In contrast to home sleep studies, a few European sleep centers have devised diagnostic procedures in which patients are asked to sleep in the laboratory for as long as they can. The idea is to allow the patient to fully saturate their need for sleep and gauge just how excessive it is. An advantage here: the patient is hooked up to EEG leads, and the amount of time they spend sleeping can be verified directly.

  The most arduous of these is the protocol used in Montpellier, which calls for patients to sleep as long as they can over thirty-two hours (two nights and the day between).40 The conditions of enforced bed rest with dim light and no screens or reading material sound punitive; they are designed to be as boring as possible. The protocols in Paris and Bologna are less stringent and allow patients to move around or read during the day.

  As Dauvilliers and his colleagues note, many healthy adolescents or young adults will be able to sleep for eleven hours in the first twenty-four, because they habitually get less sleep than needed during the week and then catch up on the weekend. The second night is necessary to separate IHers from everyone else. The ideal cutoff for separating IH from controls was nineteen hours of sleep over the thirty-two-hour period. These standards may need to be adjusted for age, body mass index, or other factors once enough information is available.

  Dauvilliers has advocated use of a strict laboratory protocol, such as the one his group uses, to gauge how much sleep someone needs for IH diagnosis. And recently, that approach has yielded some rewards. His group has shown that long sleep IHers show an association between sleep inertia and slower reaction time, as measured by the venerable psychomotor vigilance test.41

  Here, we see a clash between the needs of clinical practice and of research. In clinical practice, marathon sleep tests would be inconvenient and costly, especially in the United States. For research studies, it is essential to have homogeneous and well-defined groups of patients. Otherwise long-sought biomarkers for IH, like the GABA-enhancing somnogen, will continue to slip away.

  CHAPTER 17

  THE FDA OPENS A DOOR

  Sometimes, our sciences create kinds of people that in a certain sense did not exist before.

  —Ian Hacking, 2006

  In March 2021, the Hypersomnia Foundation and Jazz Pharmaceuticals announced the launch of “I Have IH,” a campaign aimed at increasing awareness of IH among both health care professionals and the general public. On Facebook, the Hypersomnia Foundation posted photos showing that the campaign’s content was visible on a large video screen in New York City’s Times Square. Reacting to the post, Diana Kimmel expressed satisfaction. This was what she and other people with IH wanted. “Fellow IHers, we have been heard and seen and it feels good,” she wrote. The public display and the awareness campaign showed that IH’s time had come. They were the culmination of events that Anna Sumner’s interactions with Parker, Rye, and Jenkins had nudged forward several years before.

  Let’s acknowledge what Anna’s round of publicity did not accomplish. Despite the Emory researchers’ hopes, it has not brought about the identification of another endogenous sleep-promoting substance. The sleepy stuff still lurked in patients’ CSF samples, waiting for discovery. And despite Anna’s and others’ experiences, most sleep neurologists have not embraced flumazenil as an alternative to conventional stimulants. When IHers clamored for it, the drug did become available to those able to find a willing prescriber.

  Instead, the more far-reaching downstream effects of Anna’s speaking up have been to encourage IH patients’ efforts to organize themselves. Anna may have stood apart from the hypersomnia community, but as “a legend to everyone in the room,” she inspired others. Her example accelerated the community’s coalescence in the United States online and in person.

  Others around the world have played critical roles in IH’s emergence; this book tells just one part of the story. Many sleep neurologists had grown frustrated with the MSLT’s categories and had been dealing with difficult-to-treat IH patients for years. Starting with the reference point of Bedřich Roth and his colleagues in Prague, researchers in France have built up our current knowledge of what IH is, while investigators in North America have achieved advances with EEG markers and brain imaging. On the advocacy side, Rye and others close to him had direct roles forming the Hypersomnia Foundation, but other groups such as Hypersomnolence Australia started independently.

  While they disagreed with him about his research, Rye’s critics agreed with him that MSLT-based distinctions between narcolepsy type 2 and IH were flimsy. In terms of pathophysiology, narcolepsy type 1 was well defined, based on a deficiency in hypocretin and the loss of a specific group of neurons. In contrast, narcolepsy type 2 and what distinguished it from IH were not well defined. There was a growing consensus among sleep medicine specialists that IH deserved to be on an equal footing with narcolepsy type 2. In fact, a coming realignment may join those groups together.

  REDRAWING THE LINES

  Sleep specialists have been signaling that the section of the current International Classification of Sleep Disorders dealing with narcolepsy and idiopathic hypersomnia will be revised. This will affect how members of the hypersomnia and narcolepsy community deal with the world and how they see themselves.

  The revision was certainly needed. The evidence was clear that variability in MSLT results, along with the effects of antidepressants, can obscure the ability to distinguish between narcolepsy type 2 and IH.1 Rye and Trotti’s 2013 publication on repeated MSLTs giving discordant results helped open up the discussion; others have since confirmed their findings.2 Since current distinctions between narcolepsy type 2 and IH are not meaningful, it makes more sense to divide people up by different criteria.

  Accordingly, two recent papers from sleep neurologists have called for merging narcolepsy type 2 and IH without long sleep; IH with long sleep should be carved off into a separate category. Looking ahead to ICSD-4, both proposals would narrow how the term “idiopathic hypersomnia” is applied: to people with an excessive need for sleep, rather than excessive daytime sleepiness by itself.

  Where the proposals differed was in what the merged category would be called. The first proposal came from a transatlantic group including Trotti, Arnulf, and the Harvard sleep neurologist Kiran Maski. They published a paper that grew out of discussions at the 2018 International Symposium on Narcolepsy, concluding: “It is not justifiable to exclude patients having IH, especially those without a long sleep time, from pharmacological treatments currently approved or being tested for NT2.”3

  These authors proposed to name the new merged category “narcolepsy spectrum disorder.”4 The term, meant to acknowledge the heterogeneity in the category, had already been used to describe relatives of people with narcolepsy type 1. In this group, it also referred to the appearance of symptoms such as sleep paralysis and hypnogogic hallucinations at a rate intermediate between narcolepsy type 1 and IH.

  The second paper, from European experts, proposed to name the merged category “idiopathic excessive sleepiness.”5 It also contained a system of diagnostic criteria for narcolepsy and other disorders, labeling them “definite” or “probable,” along with possible subcategories. The European proposal was more rigorous about applying what has been learned about narcolepsy since 1999. It would shift the definition of narcolepsy toward hypocretin deficiency and away from a collection of symptoms connected to REM sleep. Are narcolepsy and hypersomnia two parts of a spectrum of excessive sleepiness? Or does the term “narcolepsy” only apply to hypocretin deficiency?

  FIGURE 17.1. Proposals for the fourth edition of the International Classification of Sleep Disorders.

  Source: Adapted from Rolf Fronczek et al., “To Split or to Lump? Classifying the Central Disorders of Hypersomnolence,” Sleep 43 (2020).

  The European proposal also raised questions about what would happen when some diagnoses were redefined as not narcolepsy, especially if hypocretin CSF measurements remained rare. After all, the FDA approved newer drugs such as pitolisant and solriamfetol for something called narcolepsy. Maski, together with Emmanuel Mignot and Tom Scammell, emphasized access in a commentary on the European proposal: “A new nomenclature may force clinicians to change diagnostic labels for established patients because suggested diagnostic tests are inaccessible, thus compromising care and drug reimbursement.”6 At the same time, recent approvals have left people with hypersomnolence resulting from other psychiatric or medical disorders (besides narcolepsy or IH) on the sidelines.

  The physicians who wrote the European proposal did not dismiss the issue of access to treatment. One of the lead authors, Gert Jan Lammers, has used the example of a patient who falls between gaps in current diagnostic categories.7 His hypothetical patient habitually sleeps 10.5 hours per day, has difficulty waking in the morning, experiences sleep drunkenness, and reports regularly feeling drowsy or falling asleep during the day. During an MSLT, he displays an average sleep latency of 9.3 minutes, which is above the current cutoff for narcolepsy type 2 or IH. But his overall sleep duration per twenty-four hours is under the ICSD-3 boundary of eleven hours. According to Lammers, this patient has a serious problem, but no diagnosis can be made when applying current criteria. At the same time, sleep deprivation as a possible cause for complaints can be overlooked, he said.

  Previous revisions of the ICSD touched on similar issues. Since the first ICSD, IH was split into two, then fused back together. The idea of merging narcolepsy type 2 and IH without long sleep had come up before.8 This time, the science has advanced far enough to bring about a shift whose implications go beyond drug reimbursement.

  If either proposal for reclassifying IH and narcolepsy type 2 is approved, thousands of people diagnosed with narcolepsy will be told that what they live with every day is now named something else. In the United States, people with narcolepsy type 2 diagnoses outnumber those with narcolepsy type 1. They currently have explanations for what they experience, and many have made career and life choices based on their understanding of their disorders. Some have found a community where they belong. Redrawing boundaries and renaming diagnoses might disrupt those supportive associations—although new ones can form.

  The proposed name change for narcolepsy type 2 carries some similarities to the redefinition of autism spectrum disorder in the DSM-5 (Diagnostic and Statistical Manual of Mental Disorders) a decade ago, along with current discussions around renaming schizophrenia.9 For narcolepsy and IH, issues may be sharper around medication access, since no medications are approved for the core symptoms of autism.10

  One prominent voice in the narcolepsy community, Julie Flygare, CEO of Project Sleep, called for more discussion before renaming and reclassifying. Flygare said there should be a broader process of consultation among patient groups and other stakeholders considering both the science and questions of identity and medication access. “We are not the only disease community to face an issue like this one, and I think sometimes we forget the wealth of information and advice we could get by looking out beyond the sleep field horizon to leverage expertise from the broader healthcare landscape,” Flygare said in an email. “My biggest opinion is that there are lots of opinions to be considered.”

  Hypersomnia Foundation board members Diane Powell and Betsy Ashcraft told me they were following the reclassification discussion and had not made a decision about it but were concerned about insurance coverage and future FDA approvals. “The question is really, what gets us to more research and more treatments approved?” Ashcraft said. “I’m not going to be fighting over what we call it.”

  When he was revising his book on narcolepsy and hypersomnia in the 1970s, Bedřich Roth seemed to anticipate future controversy over how to delineate the two conditions. He said a decision should be made “jointly by a group of specialists who have worked in this field and discussed the issue thoroughly.”11 In this context, it may be helpful to rename IH—possibly, a narrower redefinition—after Roth. The condition could be called Roth hypersomnia syndrome, honoring Roth as “the true father of sleep medicine” and removing the discouraging word idiopathic. It could also serve as a reminder of IH’s background in postwar Czechoslovakia, an environment where pharmaceutical companies had minimal influence.

  DEFINING THE GROUP OF PEOPLE WITH IH

  However the nomenclature gets updated, the self-organization of the hypersomnia community means that people with IH and related disorders will have a voice in matters they care about. In the past, people with IH appeared in research papers as controls: sleepy people who didn’t have narcolepsy or sleep apnea. Now, more recent research is focused on them, and non-IHers have someone to ask about the experience of having IH. For related sleep disorders, the Circadian Sleep Disorders Network and the Kleine Levin Syndrome Foundation have played a similar role. Together, their representatives can have seats at the table when dealing with pharmaceutical companies, insurers, the FDA, the National Institutes of Health, and the American Academy of Sleep Medicine.