Kleine-Levin syndrome: the fish that got away

Quinn Eastman
22 min readApr 27, 2022

For hundreds of people around the world, Kleine Levin syndrome (KLS) is a mysterious demon that periodically blankets their lives in sleep. KLS involves episodic hypersomnia, sleeping most of the day for periods sometimes lasting weeks at a time, as well as an altered mental state, sometimes including disinhibition or an altered appetite for food or sex.

During an episode, people with KLS report feeling like they are in a bubble, or that they don’t know whether they are dead or alive. Then the blanket lifts, leaving the person anxious about if or when it will come back.

As far as my book on sleep disorders, KLS was the fish that got away. I began research on KLS in connection with idiopathic hypersomnia, the focus of my book. I later decided to cut a planned KLS chapter, in order to limit word count. My book mainly compares IH to narcolepsy, which has an extensive history and better-understood biology. It makes the point (I hope) that by banding together, IHers and their supporters have begun to advance research and achieve a degree of recognition.

KLS made things more complicated, because it points toward other disorders, such as bipolar disorder, migraine headache and autoimmune encephalitis. I still think that comparing IH with KLS yields some insights, about both the disorders and the communities that have grown around them. Hence, this post.

KLS shows how a worldwide community of concerned families can coalesce and accomplish something significant: the first large-scale genetic study, finally published in 2021. (The study has received no attention from news organizations — editors, take note!)

At the same time, the definition of KLS has been changing, away from the lurid behaviors described in early case reports, while some experts have argued that a fraction of KLS diagnoses might actually be something else. Uncertainty around the proper definition of the disorder may have diluted or weakened the impact of the genetic study. This parallels the discussion around whether narcolepsy and IH diagnoses should be reclassified.

The literature on KLS goes back to the 1920s, before IH was recognized by the medical community. However, KLS is much rarer than IH. Only about a thousand cases have been diagnosed anywhere. Although recent advances may change the picture, there is not a definitive genetic marker for KLS. That’s why pharmaceutical industry interest in genetically well-defined rare diseases has left the KLS community behind.

The symptoms of IH are extensions of experiences most people have had, perhaps after staying up all night. Even sleep drunkenness — a distinctive aspect of IH — can be simulated by being woken abruptly while in deep slumber. In contrast, KLS is otherworldly and more difficult to grasp. KLS is classified as a sleep disorder, but it straddles the eroding boundary between neurology and psychiatry.

Unlike with IH, leading experts think stimulants or other wake-promoting medications are generally not the answer for people in a KLS episode. Stimulants might curtail actual sleep, but they would not bring someone with KLS out of their altered mental state, and (more likely) would make them irritated. A KLS episode was considered something that can be accommodated and perhaps limited, but not dispelled completely.

While IH has a new FDA-approved medication, KLS remains an off-label frontier. The two drugs that experts have settled on as possibly bringing some benefit are lithium, known for its ability to moderate mood swings in bipolar disorder, and corticosteroids, used to control inflammation in multiple sclerosis. Could KLS’s altered mental state and periodic episodes be viewed as an extension of bipolar disorder? Or does KLS resemble autoimmune disorders such as lupus and multiple sclerosis, in which patients have flares with detectable markers of inflammation?

It’s OK not to be OK

Delanie Weyer’s KLS episodes started in her last year of high school, in a small town in Minnesota. The first time one occurred, she was at a school event, alarming teachers and students. Delanie would develop a “blank zombie stare” and all she wanted to do was sleep, which she did for 15 to 20 hours a day. She would get up only to eat, drink or use the bathroom. She would sleep through holidays and family gatherings. Then the unnamed thing would release its grip.

“I was barely coherent. It was like living in a dream,” she said. “At that time, I didn’t know how to explain my KLS to anyone and even myself.”

Outside of those episodes, Delanie appeared healthy. She played volleyball in high school, and then made her way through nearby St. Cloud State. She worked in Alaska at a charter fishing company for three summers. But she dreaded having that overwhelming feeling come back, which it did a handful of times.

Her worst episode, lasting five weeks, began when she was preparing to join her family for a summer camping trip. Before it started, she felt anxious, but she brushed it off as pre-travel jitters. The symptoms quickly intensified.

“Looking back, I have no idea how I got to the airport — I only remember little details here and there,” she said. “I nodded off many times as I drove and I remember cars honking at me. I found myself balled up in a corner at the airport, knowing this is it. It’s happening to me again.”

When Delanie met up with her mother and the rest of her family, they recognized that something was wrong; her eyes had a familiar look. At the campsite, she slept in a hammock most of the day, but had anxiety attacks riding in the car on bumpy mountain roads.

“Again, it was like I was out of touch with reality and I was living in a dream,” she said. “Nothing felt real but the panic.”

Her parents discussed bringing her to a nearby hospital, but that hadn’t been helpful before. Her mother Jean ended up deciding to drive her back to Salt Lake City airport and take her home. Along the way Delanie thought she saw people she knew, but who weren’t there.

“While I was driving, I was terrified that she was going to jump out of the car, because she was so distraught,” Jean Weyer said.

Delanie and her mother did venture to the emergency room at a hospital in Minnesota, where doctors assumed that she was recovering from alcohol or drugs. Delanie, who had been working at a residential treatment center for adolescents struggling with addiction, regarded the offer of outpatient treatment with disappointment. At home, she contemplated suicide.

“There were multiple times I had thought out how I could do it,” Delanie said. “But there was no way I could even attempt, as I was so out of touch with reality. I cried myself to sleep almost every night.”

Delanie was eventually diagnosed with KLS by Ranji Varghese, a sleep specialist in Minneapolis, several years after her episodes started. I met her and her mother at a 2018 conference held by the Kleine Levin Syndrome Foundation in California, where they met other people with KLS and their families for the first time.

At the conference, I was nervous about talking with the families of people affected by KLS, since some had been through harrowing or disturbing experiences. When they are in an episode, people with KLS can display child-like behaviors or inappropriate sexuality. KLS usually begins in adolescence and is often disruptive to someone’s education or career.

I had approached Delanie because she spoke about her condition for a local television news story the year before. She said her background working in the field of addiction and mental health motivated her; she wanted to reduce stigma. For her job as a school counselor starting that fall, her website told students: “It’s okay not to be okay. Let’s talk about it.”

As a preventive measure, Delanie was taking lithium. However, she was struggling with the side effects, such as dry mouth, sporadic nausea, tremors and blurred vision. Although she hadn’t had an episode since 2017 (follow-up email: she had another in 2020), she was unsure how to assess lithium’s benefits.

One of the challenges of conducting clinical studies in KLS is: what is the baseline frequency of episodes? Delanie’s episodes came once or twice per year — but others I interviewed reported experiencing episodes nearly every month. Its symptoms vary from person to person. This suggests a larger issue of heterogeneity. What if not everyone diagnosed with KLS has the same thing?

A proposal to banish Sleeping Beauty

KLS had caught my attention in 2015 when the Kleine Levin Syndrome Foundation held a conference in Atlanta, together with the Hypersomnia Foundation. Alanna Wong, a woman with KLS who has appeared on CNN (below), described being found sleeping on the bathroom floor of her school, and another time she spent three days during a wilderness trip huddled in her sleeping bag.

At the time, because of Wong’s example, it seemed like flumazenil and clarithromycin, the non-stimulant GABA antagonist drugs championed by Emory neurologist David Rye, might be helpful for people with KLS too. Somebody with KLS was in Rye’s 2012 disputed Science Translational Medicine paper. (Rye’s work has been central to IH’s emergence over the last decade and more on why his paper was controversial is in my book.)

When people living with KLS like Delanie Weyer and Alanna Wong have appeared on television, the tone is sometimes sensational. In these reports, KLS is often given the name “Sleeping Beauty Syndrome.” It’s an evocative phrase, but it’s sexist and it irritates members of the KLS community, who point out that it makes the condition sound like a fairy tale, rather than a nightmare.

In a few media stories, people with KLS have pushed back against the “Sleeping Beauty” label. And the episodic nature of KLS has encouraged a few people with KLS or their parents to tell their own stories, by chronicling episodes in online diaries and videos (example: klsproject.org).

Danger of misdiagnosis

Despite the limited publicity KLS has gotten over the years, a danger for people with KLS is that their condition goes unrecognized or misinterpreted. One example: Ryan McKinney, a young man diagnosed with KLS, who died in 2006 after a stay in a South Carolina psychiatric hospital. At various points, doctors told his mother her son had mononucleosis, meningitis, brain cancer, bipolar disorder or schizophrenia. Mom Donna White-Bailey makes a point to call KLS “not life-threatening, but life-altering”.

In a local newspaper article, Ryan was remembered by friends as a playful, good-natured comedian. After high school, Ryan was planning to start training to become a master welder before his KLS episodes resurfaced. Ryan experienced his first episode when he was 13. During track practice at school, he would lie down the track and sleep, his mother recalled. During episodes, he had cravings for ice cream and cheeseburgers, which coincided with symptoms of extreme dizziness and headaches.

“It was like he was in a dream state when he was awake and I was visiting,” White-Bailey told the Lancaster News. “He would ask if this was real and how he got to the hospital.”

After his death, the official determination was cardiac arrhythmia. After reading a toxicology report she obtained afterwards, White-Bailey blamed the volume and variety of medications, such as antipsychotics, her son had been given in the psychiatric hospital. Talking with me, she called what he experienced abuse: “He had so many shots, his bottom couldn’t take it anymore.”

White-Bailey’s experiences led her to start an annual fundraiser that was a gathering point for the KLS community for several years.

“My goal is to be an effective advocate for KLS research and raise awareness that it does exist,” White-Bailey told the Lancaster News. “If I yell long enough and loud enough, then hopefully not another family will have to go through what we went through.”

Scratching the surface

Until about 20 years ago, research on KLS was anecdotal: reports describing one or a few cases, with speculation about possible mechanisms. That has been changing, because of the efforts of the KLS community. At the conference where I met Delanie, Emmanuel Mignot from Stanford presented the results of the genetic study, which was then unpublished. Mignot jokingly called himself an “international vampire,” having scraped together hundreds of KLS blood samples from around the world.

Mignot is a top expert on narcolepsy. His research has opened up understanding of risk factors for narcolepsy type 1 and the autoimmune mechanism. See this excellent profile in The Scientist by Anna Azvolinsky. His interest in KLS was originally kick-started by the KLS Foundation, embodied by Neal Farber, father of two children with the disorder, who also has a background in the biotech industry.

The KLS genetic study got off the ground when then-postdoc Isabelle Arnulf — now an authority on KLS and IH herself — came to Mignot’s lab from France, and Farber saw an opportunity. “Without the pressure of the KLS Foundation, nothing would have been done at Stanford,” Arnulf told me.

Isabelle Arnulf and her colleagues in Paris have amassed the largest amount of clinical experience with KLS.

The KLS Foundation’s funding allowed her to hire a research assistant, to send and collect questionnaires, and to obtain samples from healthy controls. Arnulf’s initial work with Mignot helped her build the KLS questionnaire that her research team uses now.

At the conference, the genetic study generated some consternation, because it showed a connection between KLS and both bipolar disorder and schizophrenia, at least for the strongest genetic “hit.” Farber, then research director for the KLS Foundation, said he was concerned about KLS families’ reaction to the findings, because they preferred to think of KLS as sleep-related or neurological rather than psychiatric.

But the findings raised more questions than answers. The top hit was a variant in a gene called TRANK1, associated with a 50 percent elevated risk for KLS. The same variant in TRANK1 had also been connected with bipolar disorder and schizophrenia, but the increased risk was less (20 percent and 8 percent, respectively).

Mignot pointed to recent research on schizophrenia, indicating that complications around the time of birth, such as preeclampsia, amplify the small cumulative effect of many genetic variations, including the TRANK1 variant. Those in turn could set the stage for brain disorders that appear later in life.

Schizophrenia and bipolar disorder have been tough, complex problems for geneticists to crack, even to a limited extent, and what the protein encoded by TRANK1 does in the brain is still unclear. The TRANK1 mutation may alter placental physiology or increase susceptibility to injury during a difficult birth, Mignot has suggested. Follow-up analysis has shown that the TRANK1 association with KLS is heightened by birth difficulties, and was stronger in individuals born before the 1980s, possibly because of improved perinatal care since that time.

While the connection with schizophrenia and bipolar disorder was intriguing, Mignot’s study was just scratching the surface. If the top genetic “hit” increases the risk for a very rare disease by just 50 percent, scientists have to do a lot more work to explain why it strikes a few people and not others.

Despite the complexity, the discussion resonated with Delanie and her mother, who talked about events surrounding her birth; she was three weeks overdue, with a difficult delivery, and then in neonatal intensive care with a high fever. Perhaps her father’s exposure to Agent Orange while serving in the military also could have contributed, they wondered.

“These things don’t surprise me anymore,” Jean Weyer said.

Migraine overlap?

The Stanford KLS study was designed as a GWAS (genome-wide association study). He and his colleagues were not looking for genes that cause KLS. They were looking for well-known variants — common misspellings or alternative spellings of genes — that were statistically associated with KLS. This approach had worked well for more common conditions, such as coronary artery disease, type 1 and type 2 diabetes, and inflammatory bowel diseases. For KLS, sleep researchers have already looked at variations in HLA genes, the strongest risk factors for narcolepsy type 1.

At the outset, pursuing the genetics of KLS made sense. In a handful of families, more than one family member was affected. Also, the prevalence seemed to be higher in the Ashkenazi Jewish population than in the general population.

More recently, some of Mignot’s peers have suggested that his genetic study was casting its net too widely. The results could be explained by a heterogeneous group of patients, they said. This parallels the situation with narcolepsy. Genetic studies looking at narcolepsy type 1 pointed strongly to genes related to the immune system, supporting an autoimmune origin. If narcolepsy type 2 cases were mixed in, they would dilute and possibly confuse the results. Maybe that’s what was happening with KLS.

At a researchers-only session at the conference, Guy Leschziner, a sleep specialist at a large hospital in London, said that almost half of the 40 KLS cases he had seen had migraine-like features. The altered perception symptoms people with KLS describe often resembled a migraine aura, particularly one affecting the brainstem, he said.

Leschziner said he had started thinking about possible KLS/migraine overlap after two cases, which he and a colleague reported in a 2016 paper. The first, a 49-year-old woman, had experienced periodic episodes of hypersomnia for several years, sleeping up to 16 hours a day for 4–5 days each month.

As a premonition before an episode, she would experience a bitter taste in her mouth, distorted vision, loss of speech and child-like laughter, followed by dizziness and deep sleep. A combination of ibuprofen and domperidone, an antiemetic also used for migraines, could terminate the attacks, Leschziner said. He had similar success with a 16-year-old boy with headaches and hypersomnia episodes. The boy responded well to sumatriptan, another anti-migraine drug. Both of Leschziner’s index patients had family histories of migraine.

Although headache was not part of the diagnostic criteria for KLS, compliations of KLS cases have recorded regular headaches and light sensitivity in a majority of them, Leschziner said.

“Are we particularly bad at diagnosing KLS?” he asked. “I’m not saying that KLS is a subset of migraine. Instead, what I am saying is that KLS and migraine may have a shared pathophysiological mechanism.”

Aside from the mechanism, if some cases of KLS could be treated with familiar medications, that would be a relief for bewildered families. But other researchers at the conference seemed hesitant to discuss Leschziner’s ideas about overlap with migraine.

Mignot invoked the ancient parable of blind men struggling to perceive the full form of an elephant, while only sensing parts of its body individually.

“Underneath, it’s still the same elephant,” he said.

For his part, David Rye argued that KLS is part of a continuum that includes IH and narcolepsy type 2, and that people with KLS tend to be abnormally sleepy even when they’re not in an episode. The person described in his 2012 paper had to drop out of school after eighth grade because he would sleep nearly all day for two or three weeks. In between episodes, he was lucid and could work for a family business, but he fell asleep quickly enough that he met the diagnostic criteria for narcolepsy or IH.

Rye presented data on several patients of his who met the criteria for IH, in terms of falling into daytime naps quickly when they weren’t in an episode. Do KLS and IH have common neurobiology? That question remains unresolved.

Clues from brain imaging or proteomics?

As with IH (and unlike narcolepsy type 1), there is no biology-based laboratory test for KLS. Diagnosis is a matter of examining a patient’s history of periodic hypersomnia and excluding other possibilities. But brain imaging has been helpful in a few cases to confirm a tentative diagnosis. A neuroradiologist in Seattle established this for Alanna Wong.

Several brain imaging studies of people with KLS show that they have reduced blood flow (hypoperfusion) in parts of the brain such as the frontal cortex, even when they are not in an episode. More extensive hypoperfusion during an episode may correlate with symptoms of derealization/dissociation or impaired perception, such as patients not being able to recognize their own faces in the mirror. The brain imaging findings on KLS parallel those from IH, in which a Montreal study has found reduced blood flow in the prefrontal cortex during awake time. Still, it’s not clear where reduced blood flow stands, in terms of the hierarchy of cause and effect.

In 2021, Arnulf and colleagues published a compilation of brain imaging data on 138 patients, noting that longitudinal studies are needed. Future brain imaging or EEG studies may also benefit from recent work on states of dissociation.

Aditya Ambati, a postdoc then working with Mignot, presented data at the 2018 conference on protein markers in CSF (cerebrospinal fluid). He reported that when compared with controls, CSF samples from KLS patients contained higher levels of proteins related to innate immune function and macrophages/microglia. This suggests that a KLS episode may reflect a state of neuroinflammation or wound healing response, Ambati said.

For Ambati’s CSF study, only 22 people with KLS contributed samples; it can be difficult to get people with KLS to undergo a painful procedure like a spinal tap. But one potentially helpful aspect is that patients could serve as their own controls when out of episode. (Update: in July 2022, this research was published in SLEEP, with the number of KLS patient samples a bit higher.)

Changing criteria for KLS over time

Starting with the first publications on Kleine-Levin syndrome, researchers have debated what is a “true” or “genuine” case. In particular, they disagreed about whether only men fit the criteria, or both men and women could receive the diagnosis. Early on, women with a KLS diagnosis would have been considered “atypical” by some experts. But reports of families in which both men and women were diagnosed with KLS have worn down the idea that only men could have it.

MacDonald Critchley, 1900–1997

In 1942, MacDonald Critchley, a British neurologist with wide-ranging interests, named the disorder, following reports by Willy Kleine in Germany and American psychiatrist Max Levin. Critchley’s experience as a physician serving in the Royal Navy, at a time when only men were members, may have shaped his perspective toward whether women could have KLS.

In early case reports on KLS, patients displayed both hypersomnia and excessive appetite for food during episodes. The combination of sleepiness and hunger is what Levin and Critchley stressed, defining KLS as “The syndrome of periodic somnolence and morbid hunger.”

Even before Critchley, cases that may have fit Kleine-Levin criteria appeared in an early 19th century publication on “Chronische Idiopathische Schlafsucht” by Schindler and in an 1884 paper from American neurologist Charles Dana on “Morbid Drowsiness and Somnolence.”

In his 1962 paper, Critchley described one case, a young sailor, as revealing. In the middle of attacks lasting five to eight days, the sailor would sleep all night and all day, except for meals. If wakened, he was irritable and resentful and barely spoke.

“He ate considerable quantities of food, i.e. four portions of the main meal in addition to a vast number of sweets. Between helpings he would drop off to sleep again. It was seem as though he would go on eating automatically, in a wolfish way, but without any real evidence of hunger.”

Another young man was sent home from India after displaying unusual behavior:

“…most of that time he was asleep, but during his wakeful periods he was noisy, confused, truculent, garrulous and excessively hungry. He imagined that he was the head of a police force protecting important personages including Winston Churchill. Several times he thought he saw snakes and he talked to them and poured water over them. One or two afternoons he spent walking up and down the garden saluting every few steps. He made repeated attempts to seduce his nurse — who is said to have been old enough to be his grandmother. When asleep he lay curled up with his thumb in his mouth.”

Critchley spends several pages reviewing others’ cases and judging some of them atypical: “It is true that the literature contains a few references to somnolency and hunger in women. In such cases however, there are usually atypical features present, e.g. later age, fever or obvious hysterical traits, which combine to take such case records still further away from the clearly demarcated syndrome here described.”

A few papers on KLS have suggested that it involves a disruption of the function of the hypothalamus, a region of the brain that regulates appetite as well as sleep/wake. Others call it a “hibernation syndrome,” implying that it is an Altered States-like-throwback to the evolutionary past.

In a 1975 Neurology paper, Billiard, Guilleminault and Dement — all heavyweights in the history of sleep research — described a 13 year old girl who displayed periodic hypersomnia associated with her menstrual cycles.

Before episodes, the girl (called Janice in the paper) displayed aggressive or hostile behavior, followed by staying in bed for several days. She refused food and drink, and experienced occasional “visual delusions,” recognizing people around her who were not present.

“Whenever awake, the patient appeared very apathetic and uncommunicative, entirely inactive and somewhat bizarre, but without any confusion; she was able to respond appropriately when questioned,” the Neurologypaper says.

Billiard and colleagues seemed unsure as to whether Janice should be diagnosed with KLS under Critchley’s criteria, since she did not display a heightened appetite for food during episodes. They hypothesized that the hormone progesterone may be involved in triggering her episodes, especially since treatment with estrogen appeared to stop them.

For a time, KLS and “menstrual-related hypersomnia” were separate diagnoses in the International Classification of Sleep Disorders, until a 2014 revision united them into one category named KLS.

Myopathy crossover

In 2018, around the time of the KLS Foundation conference, a paper representing an alternative to Mignot’s GWAS approach was published. It was a significant discovery, based on study of a family from Saudi Arabia. A team of researchers based in Lausanne, led by Mignot’s rival Mehdi Tafti, worked with Saudi doctors to identify the gene responsible, called LMOD3.

Unlike Mignot’s GWAS, this study could make a strong case to have found a causative mutation. A grandfather, six of his children and one grandson all had KLS. Among 17 members of the family, a subtle mutation in LMOD3 was present in all affected individuals, and confoundingly, in one unaffected sibling.

This example shows how for a rare neuropsychiatric disorder, a “deep not wide” approach has yielded more powerful results than a broader GWAS approach — one of the points I aim to make in my book about IH. But one needs fantastic luck to find a family like this. (Popper et al described a Hawaiian family with several KLS-affected members in 1980, but I have not found any follow-up.)

The Saudi doctors first reported on the family in 2007. A brother and sister, in their late teens and early 20s, had come to their attention with acute symptoms. During episodes, the young man was described as having exposed himself and become “more aggressive, less polite.” Other family members were diagnosed retrospectively; the grandfather had died by the time the 2018 paper appeared.

Collectively, the family’s manifestations of KLS appear canonical enough, although males and females were both affected. Onset for all family members was in adolescence. Attacks, occurring a few times per year, lasted about two weeks. Precipitating factors were either flu-like symptoms in males, or menstruation in females. Hyperphagia was present in the majority of cases. Most of the affected family members also had ankylosing spondylitis, a chronic inflammatory form of arthritis affecting the spine, which has an autoimmune basis.

It is difficult to explain how a mutation in LMOD3 could produce the complex periodic symptoms of KLS. The protein encoded by LMOD3 plays a structural role in cells’ internal skeletons. It was thought to be important for the organization of muscle fibers, since mutations in the same gene had been linked to nemaline myopathy, an inherited neuromuscular disorder. However, the mutations that cause nemaline myopathy were different from those seen in the Saudi family.

Tafti’s lab was able to demonstrate that the LMOD3 gene was expressed in parts of the mouse brain that control sleep and wake, such as the hypothalamus. Scanning through KLS cases of European origin, they have found other mutations in LMOD3.

None of the Swiss or Saudi authors attended the 2018 meeting in California, and their paper wasn’t discussed much. Although Mignot’s team did not find associations with LMOD3 in their study, the findings from the Saudi family may have broader relevance. Since then, Tafti has reported that LMOD3 mutations are present in at least a quarter of KLS patients examined. Sleep researchers may end up with two subtypes of KLS, familial and sporadic — as with neurodegenerative diseases, such as Parkinson’s or amyotrophic lateral sclerosis.

Attraction to autoimmune encephalitis

In addition to genetic studies, members of the KLS community have followed research on autoimmune encephalitis with interest, noting similarities between the disorders. Another reason to suspect autoimmunity is that flu-like illnesses often seem to precede KLS episodes.

Autoimmune encephalitis (AE) is caused by inflammation in the brain, sometimes in connection with a tumor or infection. in addition to sleep disturbances, symptoms of AE can include: seizures, bodily weakness or numbness, impaired memory or cognition, compulsive or repetitive behaviors, agitation, unusual or involuntary movements, and impairments of balance, speech or vision. In comparison with autoimmune encephalitis, KLS patients’ symptoms appear to be more restricted, in that movement disorders are not prominent.

Autoimmune encephalitis received a wave of publicity beginning in 2012, with Susannah Cahalan’s book (and later movie) Brain on Fire. It came at a time of increasing awareness of autoimmune encephalitis in the medical community as well.

For a patient experiencing acute symptoms who might have either autoimmune encephalitis or schizophrenia, their family would probably want them to have the former. Emory neurologist Jim Greene has called this preference the “Brain on Fire effect.”

Families’ preferences are understandable. Autoimmune encephalitis is considered neurological and possibly temporary, and treatable with anti-inflammatory or immunomodulatory drugs, while schizophrenia is presumed permanent and antipsychotic drugs have major adverse side effects.

It is not necessary to have seizures or go into catatonia, both experienced by Calahan, for someone to be diagnosed with autoimmune encephalitis. That’s because autoimmune encephalitis has what psychiatric disorders don’t: clear biomarkers, in the form of anti-NMDA receptor antibodies and other antibodies to signaling molecules in the brain.

Although autoimmune encephalitis is relatively rare, compared to schizophrenia, estimates of AE’s prevalence are increasing as more autoantibodies are identified, according to a 2018 Mayo Clinic study.

Josep Dalmau and colleagues identified the anti-NMDA markers in 2007. And the mechanism of the disorder makes sense. Autoantibodies’ effects on the brain resemble those of NMDA receptor antagonists such as phencyclidine and ketamine: dissociation and hallucinations.

In 2018, Mignot said that his lab had gone looking for anti-NMDA receptor antibodies and other markers of autoimmune encephalitis already: “It was one of those great ideas, which didn’t actually pan out.” Perhaps the critical autoantibody in KLS has not yet been identified? The search for the autoimmune mechanism of narcolepsy type 1 was also full of twists and turns.

Another similarly rare diagnosis that has been appearing more in the medical literature is neuromyelitis optica spectrum disorder, referring to inflammation of the optic nerve and spinal cord. NMOSD used to be considered a variant of multiple sclerosis, but now is its own diagnosis because of readily available markers, in the form of autoantibodies recognizing the protein aquaporin-4.

Hope in the form of burnout

For individuals with KLS and their families, hope (along with considerable uncertainty) can come with the pattern for episodes to decline in frequency after someone’s 20s or 30s. For a few people I have interviewed, KLS is something they can point to and say: That part of my life is over. This contrasts with other psychiatric illnesses such as schizophrenia, which do not fade out in the same way.

What more could neuroscientists learn from studying KLS, in which the altered mental state is of apparent endogenous origin, yet reversible? In the same way that the study of fragile X syndrome has provided insights into the neurobiology of autism, perhaps studying KLS could provide an entry point to a better understanding of schizophrenia?

The challenges for continued research on therapeutic options are that Kleine-Levin syndrome is very rare, does not present in a uniform fashion and may be heterogeneous. With a retrospective approach, Arnulf’s group has established evidence for the efficacy of lithium and corticosteroids, but clinicians elsewhere continue to try other medications, ranging from antiepileptic drugs to the venerable dye methylene blue. Additional insights may come from individuals who were diagnosed with KLS who are now conducting research on KLS themselves.

Thanks to advice from Michael Hamper, along with input from Mehdi Tafti, Guy Leschziner, Isabelle Arnulf, and in particular Delanie and Jean Weyer.



Quinn Eastman

Quinn was trained as a biochemist and then as a science journalist. His book on idiopathic hypersomnia will be published by Columbia University Press in 2023.