Sleep-related eating disorder after dose escalation of sodium oxybate in narcolepsy type 1

Abstract

Sodium oxybate (SO) is an effective therapy for narcolepsy type 1 (NT1) but has been infrequently linked to complex sleep-related behaviors. We describe a case of NT1 who developed sleep-related eating disorder (SRED) after progressive gradual titration of SO. A 15-year-old female was diagnosed with NT1. Symptoms remained disabling despite high doses of modafinil and venlafaxine, and thus SO was initiated and titrated over months. After increasing to a dose of 4 g twice per night, the family reported episodes of partial arousal with altered consciousness beginning 1–2 h after the first dose, where the patient left the bed, ate large quantities of calorie-dense foods, and appeared unresponsive, before returning to bed again with complete amnesia. These episodes contributed to significant distress for the patient and her family, and notable weight gain. Gradual dose reduction to 7 g/night combined with structured behavioral interventions (dietary modification, bedside snacks, and environmental safety measures) successfully resolved SRED while maintaining excellent cataplexy control, demonstrating that combined management strategies can avoid medication discontinuation.

Introduction

Narcolepsy is a hypersomnolence sleep disorder with a prevalence of 0.025–0.05% in the population.¹ The local prevalence in KSA is 0.04% according to Al Rajeh et al.² Sodium oxybate (SO) is the standard of care for treating cataplexy, excessive daytime sleepiness (EDS), and disrupted nocturnal sleep.³ Common side effects of SO include nausea, dizziness, vomiting, tremor, and enuresis.⁴ Moreover, in a few previous reports,5, 6, 7 SO has been linked to somnambulism and complex sleep behaviors, including sleep-related eating and even sleep driving.

We present a case of narcolepsy type 1 (NT1) where gradual SO dose escalation to 8 g nightly (4 g twice per night) precipitated sleep-related eating disorder (SRED), which improved after dose reduction and behavioral interventions, supporting a dose-dependent association. This case adds to the limited literature by showing that structured behavioral strategies can resolve SRED while maintaining control of narcolepsy symptoms. We also outline behavioral strategies and environmental safety measures that may help clinicians to manage this adverse effect as previous reports focused mainly on pharmacological adjustments.⁶

Case history

A 20-year-old female, who first presented at 15 years of age, was evaluated at the University Sleep Disorders Center at King Saud University Medical City, Riyadh, KSA. The patient was referred from neurology to rule out epilepsy. She had been in her usual state of health until 1 year prior when she developed EDS, which interfered with her school performance and daily activities, and her Epworth Sleepiness Scale score was 18/24. EDS was accompanied by severely fragmented nocturnal sleep, daytime fatigue, low energy, frequent sleep paralysis, and hypnopompic hallucinations (auditory and visual).

Approximately 4–5 months after the onset of EDS, she developed severe cataplexy characterized by recurrent episodes of sudden loss of muscle tone with preserved consciousness. The attacks initially involved the upper limbs and facial muscles (eyes and jaw) but progressed to involve the whole body. Episodes were triggered by laughter, stress, and intense anger. These cataplexy attacks were clinically significant, leading to complete collapses and falls. Polysomnography followed by multiple sleep latency tests supported NT1 according to ICSD-3 criteria (Table 1). Cerebrospinal fluid hypocretin-1 measurement and HLA typing were not done.

Table 1.

Summary of overnight polysomnography and multiple sleep latency test results.

Overnight polysomnography results
Measurement	Report
Light off	10:13 PM
Light on	7:00 AM
Time in bed	527.5 min
Total sleep time	218.5 min
Sleep latency	94 min
Sleep efficiency	41.4%
N1	32.7%
N2	67.3%
N3	0%
R	0%
REM cycle	1
AHI (NREM/REM)	0
AHI (supine/lateral)	0
Central apnea index	0
Obstructive apnea index	0/h
Mixed apnea index	0
Hypopnea index	0/h
Desaturation index	1.6/h
Time with O2 < 90%	0%
Time with O2 < 95%	0%
Lowest O2/average O2	95%/98%
Arousal index	44.2/h
Snoring	0%
Leg movement index	19.2/h
Summary of multiple sleep latency test
Mean sleep latency	00:00:15
Mean REM latency	00:06:50
SOREMP	3 Naps

Note. AHI: Apnea–hypopnea index; N1/N2/N3: non-rapid eye movement sleep stages 1–3; REM: rapid eye movement sleep; SOREMP: sleep-onset REM period; O₂: oxygen saturation.

The patient was initially treated with 38 mg of methylphenidate (extended release) without a significant response, before switching to modafinil starting at 200 mg/day and gradually increasing to 400 mg/day in two divided doses, which she continued. She was first treated for cataplexy with venlafaxine at 150 mg/day, which was gradually escalated to a high dose of 300 mg/day.

Despite this escalation, cataplexy remained poorly controlled, prompting the initiation of SO, whereas venlafaxine was continued. SO was started at the recommended initial regimen of 2 g at bedtime and 2 g at 3 h later, with at least 2 h between the last meal and the first dose. She exhibited no noticeable improvement on this dose. Therefore, the dose was titrated gradually upward with close monitoring via clinical visits or virtual clinics every 2–4 weeks. The dose was escalated over a 1- to 2-month interval as follows: 2.25 g, 3 g, 3.5 g, and eventually 4 g twice nightly. Eventually, she exhibited significant improvements over 4–5 months. After approximately 4 months of treatment with 4 g SO twice nightly and 3 h apart, and despite satisfactory improvement, she reported episodes of partial arousal from sleep, during which she was not fully conscious, and would go to the kitchen and eat anything present in the refrigerator. She did not report eating raw food or preparing food, but she ate an apple on one occasion, which was prohibited due to her recent dental braces, and this episode resulted in damage to her braces.

Her family attempted to lock the kitchen to prevent possible self-harm. They also observed that she would talk or shout during these episodes, although she had no subsequent recollection of these behaviors. The episodes occurred nightly, sometimes twice per night (the first episode around midnight to 1 am within 2 h after the first dose of SO, and the second episode at the time of the second dose around 3–4 am) with no reported history of self-harm or injuries.

SO was then gradually reduced to 3.5 g twice nightly (7 g/night), administered 2.5–4 h apart, as well as behavioral interventions, including a high-fiber, complex-carbohydrate-rich dinner; accessible, healthy bedside snacks such as sliced fruits and vegetables, and wholegrain crackers; environmental safety modifications, including locking the kitchen at night and removing sharp objects from the bathroom and bedroom; and primary nighttime supervision by her mother from an adjacent room. The frequency and duration of episodes decreased over 3 months of follow-up. She continued to awaken hungry within 2 h of the first dose but was more oriented, ate bedside snacks, and returned to sleep without further episodes that night. Cataplexy and excessive sleepiness remained well controlled despite dose reduction (Figure 1).

Figure 1.

Clinical course of sleep-related eating disorder after sodium oxybate dose escalation in narcolepsy type 1.

Our patient had morbid obesity, with a body mass index (BMI) of ∼50 kg/m² at the time of diagnosis, which increased to 56 kg/m² after she developed SRED, so she was referred to the obesity clinic. The assessment and management plan included bariatric surgery because adherence to a healthy dietary program was challenging. She subsequently underwent successful bariatric surgery, resulting in significant weight loss and her BMI decreased to ∼30 kg/m².

Discussion

This case supports and extends previous observations that SO can precipitate complex parasomnias, including SRED, in patients with narcolepsy. Wallace et al.⁶ described a 42-year-old man with narcolepsy–cataplexy, obesity, obstructive sleep apnea, and depression who developed one episode of sleep driving and at least two episodes of sleep-related eating shortly after his SO dose was titrated to 8 g/night. Two weeks after reaching that dose, he had a cluster of six parasomnia events, where he drove to a convenience store while asleep and was awakened at the counter buying a candy bar, while he also awoke with chocolate smeared on his face on other mornings, and his bed partner reported other complex behaviors, all without subsequent recall. Most episodes occurred during the latter half of the night following the second SO dose.⁶ Our patient's presentation exhibited a similar temporal association with dose escalation, where SRED episodes occurred within 2 h of the first SO dose and again around the time of the second dose, typically involving highly palatable foods such as chocolate or sweets, with no reported complex behaviors or injuries. Her SRED improved after dose reduction combined with behavioral strategies, and structured counseling on safety measures. Over time, her nocturnal events evolved into brief, fully conscious awakenings with small, planned snacks while she remained in bed. In the case reported by Wallace et al.,⁶ the behaviors ceased when SO was discontinued, and after re-initiation at 7 g/night, they did not recur for 1 year, but a brief rechallenge at 8 g/night again provoked somnambulistic episodes, leading to permanent dose reduction. A pharmacovigilance summary of this case published in Reactions similarly emphasized the dose-dependent relationship and recurrence upon rechallenge.

Furthermore, another case was previously described of a 41-year-old woman with narcolepsy who developed SRED after long-term treatment with SO.⁷ Her compulsive eating behaviors during sleep, with complete amnesia the next day, resolved immediately after SO was temporarily discontinued and did not recur when the drug was slowly reintroduced.⁷ The rapid resolution upon discontinuation and non-recurrence upon gradual reintroduction in this case further demonstrates that SO can transiently enhance arousal disorders arising from non-rapid eye movement (NREM) sleep, possibly by increasing deep slow-wave sleep rebound.⁷

In addition to single-case reports, our observations are consistent with pharmacovigilance data indicating a strong signal linking SO to SRED. In a disproportionality analysis based on the WHO VigiBase database, Merino et al.⁵ identified 676 reports of drug-associated SRED, where SO accounted for 185 (27.4%) of these cases, with a markedly elevated reporting odds ratio of 204.2 (95% confidence interval: 172.4–241.8).

The pathophysiological mechanisms that underlie SO induced SRED remain incompletely understood but multiple factors are probably involved. SO enhances slow-wave sleep and the transition from deep sleep stages (N3) back to lighter sleep (N1) or wakefulness creates conditions that are conducive to NREM parasomnias.³ The hypocretin deficiency intrinsic to NT1 may increase susceptibility because the hypocretin system normally stabilizes sleep–wake transitions and modulates arousal thresholds. In addition, hypocretin neurons project to reward-associated brain regions and interact with the melanocortin system, which has been implicated in binge-eating behaviors.^8,9 Patients with NT1 exhibit higher rates of eating disorders even without SO treatment, and recent data from Italy indicate that 23.3% of patients met the criteria for clinical eating disorders (primarily eating disorder not otherwise specified and night eating syndrome) compared with 0% of the controls.⁹ This intrinsic vulnerability, combined with SO's effects on sleep architecture and arousal, may create a “double hit” phenomenon that predisposes to SRED.

The present case study suggests practical guidance for managing SO-associated SRED in NT1. Baseline screening for eating disorders and metabolic abnormalities may identify vulnerable patients because both conditions frequently predate treatment.^8,9 SRED emerges more commonly at doses of 7–8 g per night or higher, suggesting that slower titration or increased monitoring may be warranted near this threshold.^5,6 Interestingly, Kass et al.¹⁰ demonstrated that switching from twice nightly oxybate (median 8.8 g/night) to a once nightly extended-release dose (median 9 g/night) is a promising approach, where patients showed clear improvement in SRED and most also experienced a reduction in BMI, highlighting the importance of medication timing and total nightly dose. When SRED develops, combined approaches merit consideration. Reducing the dose by 1 g per night (from 8 g to 7 g in this case) eliminated SRED while preserving cataplexy control, indicating that discontinuation does not need to be the initial response. Behavioral modifications, including high-fiber evening meals, accessible bedside snacks, kitchen security, and family education, appear to be useful as adjunctive measures.¹¹

Conclusion

In conclusion, SO may trigger complex parasomnias such as SRED, particularly at higher doses in susceptible patients with NT1. The present case supports a dose-dependent association. A modest dose reduction combined with behavioral strategies effectively resolved SRED while maintaining narcolepsy control, indicating that discontinuing SO is not always necessary. Clinicians should screen for eating disorder symptoms at baseline and during drug titration with close follow-up and education. Further research is needed to clarify the mechanisms involved, define dose-related risks, and identify vulnerable patients.

Patient consent statement

Written informed consent was obtained from the patient for publication of this case report and any accompanying images or clinical information. The authors confirm that the patient's identity has been protected.

Ethical approval

This study was approved by the Institutional Review Board, King Saud University Medical City, King Saud University (Approval No. E-25-10516).

Author contributions

RAA: conceptualization of the case report, acquisition of clinical data, literature review, drafting the initial manuscript, and critical revision of the manuscript. RA: acquisition and interpretation of clinical data, literature review, drafting sections of the manuscript, and critical revision of the manuscript. HD: interpretation of clinical data, supervision and clinical oversight, and review of the draft manuscript. AB: senior supervision, verification of clinical information, critical revision of draft manuscript, and final approval of the version to be published. All authors contributed substantially to the work, approved the final manuscript, and agree to be accountable for all aspects of the work in ensuring accuracy and integrity. All authors have critically reviewed and approved the final draft and are responsible for the content and similarity index of the manuscript.

Declaration of generative AI and AI-assisted technologies in the manuscript preparation process

During the preparation of this work, the authors used ChatGPT (OpenAI), Grammarly, and QuillBot to assist with language editing, enhance readability, and refine the organization of the manuscript. After utilizing these tools, the authors reviewed and edited the content as necessary and take full responsibility for the accuracy and integrity of the published article.

Source of funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Conflict of interest

The authors have no conflicts of interest to declare.