Narcolepsy-like symptoms triggered by lemborexant in the context of hyperactive delirium in a patient with bipolar depression: a case report

Abstract

Lemborexant is a dual orexin antagonist and is considered a safe and effective hypnotic. Dual orexin antagonists induce physiological sleep by blocking orexin receptors. Although the blockade of orexin signaling has triggered narcolepsy-like symptoms in rodents, there is currently no evidence of lemborexant inducing narcolepsy-like symptoms in humans. We describe the case of a 79-year-old Japanese woman with bipolar depression who experienced lemborexant-induced cataplexy and sleep attack. Her previous results on the Multiple Sleep Latency Test excluded the diagnosis of narcolepsy. She experienced narcolepsy-like symptoms on 2 occasions after she was administered lemborexant, in the context of hyperactive delirium, but not in a relaxed state. Her case suggests that lemborexant could trigger narcolepsy-like symptoms in patients with hyperactive delirium, even those with no history of narcolepsy. This case also emphasizes that clinicians must be very careful when they prescribe lemborexant to patients who experience hyperactive delirium.

Citation:

Shibata S, Oda Y, Ohki N, et al. Narcolepsy-like symptoms triggered by lemborexant in the context of hyperactive delirium in a patient with bipolar depression: a case report. J Clin Sleep Med. 2022;18(5):1459–1462.

INTRODUCTION

Insomnia is a common and a representative sleep disorder. More than 30% of adults are considered to have some symptoms of insomnia, including arousal during sleep and early-morning awakening. As insomnia is often accompanied by multiple psychosomatic symptoms (eg, somnolence, lassitude, difficulty in concentrating, depression, and anxiety), insomnia is considered to pose significant risks for socioeconomic loss and damage due to poor work performance, as well as for increased industrial accidents.

In the past, benzodiazepines were extensively used to treat insomnia. Whereas benzodiazepines have gamma-aminobutyric acid receptor-mediated sedating effects, they are associated with the serious issues of addiction, tolerance, and cognitive impairment. In contrast to benzodiazepines, orexin receptor antagonists are thought to induce physiological sleep through a competitive inhibition of orexin, which is involved in the stabilization of awareness. ¹ Orexin receptor antagonists are thus speculated to have fewer negative side effects than benzodiazepines. Lemborexant, a dual orexin receptor antagonist (DORA), is a new psychopharmaceutical that was recently approved in the United States, Canada, and Japan to treat insomnia. The results of phase 3 randomized clinical trials of lemborexant have shown that this DORA can improve both sleep onset and sleep maintenance over the long term. ^{2, 3} These trials also revealed that lemborexant did not induce excessive daytime sleepiness or rebound insomnia. Lemborexant is thus expected to become a new treatment option for patients with insomnia in terms of both its efficacy and its safety.

The neuropeptides orexin-A and orexin-B have been speculated to be involved in the maintenance of arousal and the regulation of rapid eye movement sleep via the stimulation of the orexin-1 and -2 receptors. Orexin-2 receptors are considered to play a central role in the sleep-wakefulness cycle, and Beuckmann et al ⁴ demonstrated that lemborexant is the DORA with the strongest orexin-2 receptor blocking action. It is also well known that a severe loss of orexin neurons causes type 1 narcolepsy, which is characterized by the presence of daytime sleepiness, cataplexy, sleep attacks, sleep paralysis, and hypnagogic/hypnopompic hallucinations. It is thus likely that the blockade of orexin receptors, especially orexin-2 receptors, would induce narcolepsy-like symptoms.

Orexin systems have been implicated in the pathophysiology of other psychiatric disorders, including mood disorders and schizophrenia. For example, many clinical studies have demonstrated that depressed patients have decreased orexin-A levels in the cerebrospinal fluid or plasma. ⁵ Individuals with depression may thus be particularly susceptible to DORAs. However, there is currently no report that lemborexant induces narcolepsy-like symptoms in patients with bipolar depression.

Herein we present the case of an elderly patient with bipolar depression who experienced cataplexy and sleep attacks twice after she took lemborexant (5 mg) in the setting of hyperactive delirium; this effect of lemborexant was not produced when she was in a more relaxed state.

REPORT OF CASE

Our patient was a 79-year-old Japanese female with bipolar disorder and complete atrioventricular block. She had previously tested negative for narcolepsy under the following circumstances. At the age of 67 years, after the loss of her husband to acute myeloid leukemia, she had noticed increasing fatiguability and repeated attacks of unconsciousness lasting just a few seconds. Specialists in cardiovascular internal medicine, pulmonology, and neurology had referred her to a sleep study under the suspicion of narcolepsy. Nocturnal polysomnography showed an apnea-hypopnea index of 1.9 events/h, lowest SpO₂ at 86%, arousal index of 7.3. In addition, the Multiple Sleep Latency Test showed the mean sleep latency at 12 minutes and no sleep-onset rapid eye movement sleep periods. These findings were interpreted as negative for narcolepsy.

After several examinations, the patient was eventually diagnosed with complete atrioventricular block and subsequently underwent pacemaker implantation at our hospital. From that point onward, she never experienced attacks of unconsciousness. However, on leaving the hospital, she came to feel that she was too tired to live and described wanting to die to end her agony. She was thus referred to our department of psychiatry. In part because she had had episodes of hypomania in the past, we diagnosed her with bipolar II disorder (bipolar depression) and introduced valproate sodium. However, her symptoms gradually worsened. At the age of 69 years, she was admitted to our hospital.

Directly after her admission, she fell into a catatonic stupor. Pharmacotherapies had limited effectiveness against her psychotic symptoms. We thus decided to introduce modified electroconvulsive therapy (mECT). Considering that the patient wore a pacemaker, we initially performed mECT in the presence of a cardiac physician and a biomedical engineer after we contacted the pacemaker manufacturer about the safety and utility of mECT and after reviewing the current literature.

After 24 mECT treatments, the patient experienced symptomatic relief. After her discharge from our hospital, she achieved lasting remission with maintenance mECT for 5 years. In May 2018, we decided to terminate the maintenance mECT and follow the patient, without medicine, every few months.

In March 2021, following an operation in which the old pacemaker was replaced with a new one, the patient experienced a relapse of depression. She was thus admitted to our hospital and again underwent mECT. Since she developed nocturnal delirium, she was monitored at the nurses' station every night. On the 4th, 5th, and 39th days after admission, she took lemborexant (5 mg) for insomnia. When she took the lemborexant, it took her > 30 minutes on average to fall asleep. After 12 treatments with mECT, she experienced symptomatic relief and was discharged with a prescription for valproate sodium.

However, in July 2021, the patient became significantly more depressed and developed suicidal ideation. She had become emaciated and was unable to stand up on her own. She was thus involuntarily admitted to our hospital with her son’s consent. On the second day of admission, she developed nocturnal hyperactive delirium, which might have been due in part to advanced age, the change in her living arrangement, and depression-induced cognitive impairment. She was therefore monitored at the nurses' station every night. During this time, she stayed completely awake and alert in the daytime. Although we promptly treated her with lurasidone, little improvement of the patient’s depression or nocturnal hyperactive delirium was observed. We thus provided the repeated on-demand use of risperidone (0.5 mg).

On the 15th day of hospitalization, since the patient presented hyperactive delirium with excitement, we administered risperidone (0.5 mg) and an additional intake of lemborexant (5 mg) at 15 minutes after the risperidone intake. At 20 minutes after the lemborexant administration, following a chat with nurses in which she experienced hyperactive delirium, the patient experienced a sudden loss of generalized muscle tone and instantly fell sleep in front of the nurses. Although she was unresponsive to our attempts to wake her at the time, her oxygen saturation was measured at > 95% and her blood pressure was stable. We thus took no more action to awake her and monitored her overnight.

The next morning the patient recovered from her sleep attack and cataplexy without sequelae, and she did not remember the episode because of the nocturnal delirium. On the 16th day of hospitalization, we initiated bilateral, twice-weekly mECT. On the 17th day, after she took risperidone (0.5 mg) and lemborexant (5 mg) once again for delirium, the patient re-experienced narcolepsy-like symptoms in exactly the same way, again at 20 minutes after an administration of lemborexant in the setting of hyperactive delirium. We subsequently considered that lemborexant could have produced the patient’s narcolepsy-like symptoms. We thus stopped using the lemborexant and prescribed eszopiclone as a hypnotic. From that point forward, the patient experienced no more narcolepsy-like symptoms. After 12 treatments of mECT, she achieved a sustained resolution of depressive symptoms and delirium. At the time of discharge on the 63rd day of hospitalization, the Mini-Mental State Examination score was 29 (1 point lost in the task of delayed recall).

Written informed consent was obtained from the son of the patient for their anonymized information to be published in this article.

DISCUSSION

We have presented the case of a patient with lemborexant-induced cataplexy and sleep attack in the context of hyperactive delirium. In our patient’s case, she had incidentally undergone polysomnography and Multiple Sleep Latency Test in the past, at which point the diagnosis of organic narcolepsy had been excluded. Furthermore, there is little chance of post-ECT seizures, since the first attack occurred before mECT was recommenced. Although lemborexant had efficacy for insomnia when our patient was in a relaxed state, she experienced cataplexy and sleep attack just before falling asleep twice after an intake of lemborexant when she was experiencing hyperactive delirium. To our knowledge, this is the first published report describing lemborexant-induced cataplexy and sleep attack in a patient with no history of narcolepsy.

It has been reported that orexin-knockout mice were prone to experience somnolence and cataplexy manifested in narcolepsy. ⁶ It is possible that DORAs could also induce similar symptoms in patients. However, a large number of preclinical and clinical studies on DORAs have emphasized the low incidence of cataplexy using these medications. ^{1, 2} This point thus remains controversial. Mahoney et al ⁷ demonstrated that 2 DORAs, lemborexant and almorexant, induced cataplexy in mice dose dependently in the event of an intake of chocolate, which was used as a positive reinforcer, whereas cataplexy was quite rare in the absence of reward stimuli, even at high doses of lemborexant and almorexant. It was also reported that another DORA, suvorexant, induced cataplexy in mice under the same conditions. ⁸ We suspect that cataplexy develops during periods of arousal and is often triggered by feeling emotionally overwrought. Mahoney et al speculated that DORA-induced cataplexy in humans would be rare, since any such tendency would be masked by acute sedation and reduced activity in the brain’s reward pathways.

Given that there are few reports regarding DORA-induced narcolepsy-like symptoms, we also think that DORA-induced cataplexy would be rare. In fact, our patient did not experience narcolepsy-like symptoms under lemborexant while in a relaxed state. However, she experienced lemborexant-induced cataplexy and sleep attack at her fourth and fifth intakes, when she was experiencing hyperactive delirium after taking lemborexant. The narcolepsy-like symptoms occurred 20 minutes after the intake of lemborexant. Considering that lemborexant normally took >30 minutes to cause the patient to fall asleep, she would be not yet be substantially sedated by 20 minutes. We thus suspect that the confluence of hyperactive delirium and incomplete sedation may have induced her narcolepsy-like symptoms just before falling asleep. This case emphasizes the need for clinicians to be very careful when prescribing DORAs for patients with delirium who are prone to excitation.

Sleep medications are also associated with an increased risk of falls. Benzodiazepines and nonbenzodiazepine “Z”-drugs (ie, zolpidem and zopiclone) have a muscle relaxant effect via the α1–3 subunits of the gamma-aminobutyric acid receptors. DORAs do not have functions in gamma-aminobutyric acid receptors, but they act selectively on the orexin system. DORAs are therefore thought to not exert a muscle relaxant effect. ² In fact, several randomized control studies have shown that suvorexant is not associated with an increased risk of falls and fractures. ⁹ However, Ishibashi et al ¹⁰ recently demonstrated a risk of falls linked to suvorexant, but not ramelteon, which is a nonbenzodiazepine, a non-DORA that affects melatonin receptors. Considering that Ishibashi et al conducted the research with only elderly hospitalized patients, there were probably more than a few patients with delirium. It is thus possible that some of these falls were due to unrecognized cataplexy.

It was reported that in clinical trials, suvorexant induced hypnagogic hallucinations or sleep paralysis in ∼1% of the participants. ¹¹ As it is rare that patients succumb to cataplexy and sleep attack in the presence of medical professionals, a narcoleptic response in patients would be easily overlooked. In addition, since a large number of patients with hyperactive delirium do not remember these episodes, clinicians should be aware of this potential side effect of lemborexant.

Lemborexant is generally well-tolerated and provides long-term benefits for patients with insomnia. ³ Given that the clinical dose of lemborexant cannot strongly block orexin signaling as manifested by narcolepsy, lemborexant-induced narcolepsy-like symptoms are expected to occur fairly infrequently. However, if a patient takes lemborexant while in a state of high arousal and then experiences hyperactive delirium, this combination might trigger narcolepsy-like symptoms. As the affinity of lemborexant to orexin-2 receptors is the strongest among the DORAs, lemborexant might be more likely than other DORAs to induce narcolepsy-like symptoms. Given that patients with depression have decreased orexin-A levels in the cerebrospinal fluid or plasma, ⁵ this psychiatric comorbidity might make patients more prone to narcolepsy-like symptoms after an administration of lemborexant. Although there have been no reports regarding the relationship between psychiatric disorders and lemborexant-induced narcolepsy-like symptoms, clinicians should take extra precautions to counsel patients with depression on sleep hygiene when they prescribe lemborexant.

In conclusion, we administered lemborexant to a patient with bipolar depression under hyperactive delirium. Although lemborexant was effective for her insomnia in a relaxed state, she experienced lemborexant-induced cataplexy and sleep attack when she was experiencing hyperactive delirium. Clinicians should be very careful when prescribing DORAs for patients who are prone to experiencing hyperactive delirium.

ABBREVIATIONS

DORA

dual orexin receptor antagonist

mECT

modified electroconvulsive therapy

DISCLOSURE STATEMENT

All authors have seen and approved this manuscript. Work for this study was performed at Department of Psychiatry, Chiba University Hospital. Dr. Oda has received speakers’ honoraria from Otsuka, Sumitomo Dainippon, and Meiji Seika. Dr. Iyo has received consultant fees from Eli Lilly, Sumitomo Dainippon, Pfizer, and Abbott and reports honoraria from Janssen, Eli Lilly, Otsuka, Meiji Seika, Astellas, Sumitomo Dainippon, Ono, GlaxoSmithKline, Takeda, Mochida, Kyowa Hakko, MSD, Eisai, Daiichi-Sankyo, Novartis, Teijin, Shionogi, Hisamitsu, and Asahi Kasei. The other authors report no conflicts of interest.