Narcolepsy initially misdiagnosed as schizophrenia: A case report

Narcolepsy is a rare, primary and lifelong sleep disorder. Its main manifestations include excessive daytime sleepiness, cataplexy, hypnagogic hallucinations, sleep paralysis and nocturnal sleep disturbances. These symptoms can have a profound impact on patients' academic performance, work efficiency and overall quality of life.1 Schizophrenia is a prevalent and intricate severe mental disorder. Its characteristic symptoms encompass a diverse range of abnormalities in aspects such as thinking, emotion and behaviour.2 The symptomatic overlap between narcolepsy and schizophrenia may impede their differential diagnosis. This paper presents a case of narcolepsy that was initially misdiagnosed as schizophrenia; the patient's cardinal features were excessive daytime sleepiness and hypnagogic hallucinations. By highlighting this example, we seek to increase clinical awareness of narcolepsy and thereby reduce rates of misdiagnosis and missed diagnosis. A 23-year-old man presented to our sleep disorders unit with a 4-year history of hypersomnia. The onset dated to 2021 (final year of secondary education) and was not preceded by infection, head trauma or psychosocial stressors. Nocturnal sleep duration ranged from 23:00 to 07:00–08:00; despite reporting ‘satisfactory’ subjective sleep quality, he experienced persistent daytime somnolence that necessitated forced arousal to attend classes. Brief (<20 min) naps during recesses transiently restored alertness. Initially, cataplexy, hypnagogic hallucinations and sleep paralysis were consistently absent, and no medical consultation was sought at that time. During university studies, the patient adopted a flexible schedule that allowed daytime napping in his dormitory, but symptoms remained unaddressed. In January 2024, he began employment in logistics that required biweekly rotation between diurnal and nocturnal shifts. Initially tolerated, the schedule precipitated progressive fatigue and intensified daytime sleepiness; after 6 months, he resigned because of functional incapacity. Since July 2024, he has worked intermittently as a construction labourer in his hometown, able to maintain wakefulness only while physically active; cessation of activity is promptly followed by sleep, occasionally even while seated. In August 2024, the patient reported a vivid dream of his brother's death, after which he began to experience hypnagogic auditory hallucinations characterized by a female voice stating, ‘Someone has sent you money; return to your hometown and stop working’. These episodes occurred exclusively prior to sleep, lasted 5–10 min and were not accompanied by daytime hallucinations. Collateral history from family members documented motor and behavioural disturbances: stereotyped pacing, repetitive self-touching of the head, soliloquy and inappropriate laughter. Extensive investigation at an external institution—including cerebrospinal fluid (CSF) analysis for autoimmune and infectious encephalitides, brain MRI and routine blood screens—revealed no explanatory abnormalities. A diagnosis of schizophrenia was assigned, and a 1-week trial of an unspecified antipsychotic medication was discontinued owing to lack of perceived benefit. Three months later (November 2024), progressive hypersomnia and recurrent behavioural dyscontrol prompted presentation to our outpatient service. The patient received a provisional diagnosis of schizophrenia with associated behavioural disturbance and started taking amisulpride 400 mg/day and blonanserin 8 mg/day from November 2024 to April 2025 (approximately 5 months). Psychotic manifestations remitted; however, hypersomnia escalated to ≥18 h of sleep per day, precluding normal activities of daily living. Formal psychiatric assessment during admission suggested that the stereotyped behaviours reported by family were likely nonsomatic manifestations of extreme somnolence or emotional dysregulation rather than psychotic phenomena; no persistent delusional ideation or formal thought disorder was identified. In April 2025, the patient was re-evaluated at our centre and admitted for systematic assessment of refractory hypersomnia. The patient consistently denied persistent auditory hallucinations while awake, as well as delusional ideation involving persecution or referential intent. His sleep–wake history was as follows: he went to bed at 23:00, woke up spontaneously between 09:00 and 12:00 and napped from 13:00 to 16:00–17:00. Sleep was described as uninterrupted, with rapid sleep-onset latency even when sitting or standing. Daytime somnolence was prominent during periods of inactivity, but he could be roused easily and remained fully oriented thereafter. Nocturnal polysomnographic (PSG) symptoms (snoring, witnessed apnoea or parasomnia) were denied. Appetite, micturition and bowel habits were unchanged, and body weight had remained stable relative to the premorbid baseline. Past, personal and family histories were unremarkable; physical and neurological examinations were normal. On examination the patient was alert, fully oriented and goal-directed without formal thought disorder or psychotic symptoms; he endorsed intermittent dysphoria, anhedonia and fleeting worthlessness but denied sustained depression, suicidal ideation or functional impairment. Objectively, daily sleep time was markedly prolonged with sleep-onset latency <5 min when unstimulated, yet arousal was easily achieved and post-awakening cognition was grossly intact. Basic activities of daily living were preserved, insight was adequate and cognitive screening revealed no global intellectual deficits. Complete blood count, hepatic and renal panels, thyroid profile, serum immune indices, electrolytes, cardiac enzymes, lipid spectrum and nutritional biomarkers were all within reference limits. Positive and Negative Syndrome Scale (PANSS): total score = 82, positive sub-score = 9, negative sub-score = 32 and general psychopathology = 41. 9-item Patient Health Questionnaire (PHQ-9) = 17, indicating severe depressive symptoms. 7-item Generalized Anxiety Disorder scale (GAD-7) = 13, consistent with moderate anxiety. Epworth Sleepiness Scale (ESS): baseline 21 and post-treatment 14. Pittsburgh Sleep Quality Index (PSQI): baseline 5 and post-treatment 3. Genetic screening for narcolepsy-associated susceptibility loci revealed HLA-DQB1*05:02 and HLA-DQB1*06:02 alleles present in heterozygous form, TCRA rs1154155: CC homozygous, P2RY11 rs2305795: GA heterozygous. CSF analysis: Routine and biochemical parameters were unremarkable. CSF orexin-A (hypocretin-1) concentration, determined by radioimmunoassay, was 24.46 pg/mL. Electrocardiography, electroencephalography, cranial CT and thoracic CT were unremarkable. PSG and multiple sleep latency test (MSLT) findings are summarised in Table 1. Definitive diagnosis: Narcolepsy type 1 (NT1), as per International Classification of Sleep Disorders, 3rd ed., text revision (ICSD-3-TR). Diagnostic justification: (i) core phenotype: hypersomnia persisting ≥4 years with daily irrepressible sleep attacks. (ii) Objective confirmation: MSLT: mean sleep latency 0.6 min, ≥2 sleep-onset REM periods (SOREMPs) recorded (actual number: 5). CSF orexin-A: 24.46 pg/mL (well below the ICSD-3-TR threshold of ≤110 pg/mL). (iii) Exclusion: no evidence of chronic sleep insufficiency, circadian-phase misalignment, medication/substance effects, or another medical, neurological, or psychiatric disorder that more satisfactorily accounts for the phenotype. Differential diagnosis: (1) Schizophrenia: Although the patient reported prior auditory hallucinations, these were exclusively hypnagogic, brief and non-psychotic in quality. The apparent social withdrawal and psychomotor inertia are fully attributable to excessive daytime sleepiness rather than primary negative symptoms. Therefore, schizophrenia is currently excluded. (2) Sleep-related breathing disorder: Polysomnography disclosed only mild sleep apnoea–hypopnoea syndrome (AHI: 5.4 events/h). The magnitude of hypersomnia (≥18 h/day) is disproportionate to this degree of respiratory disturbance, making obstructive sleep apnoea an unlikely sole explanation. (3) Idiopathic hypersomnia: Prolonged nocturnal sleep and unrefreshing naps are present; however, the concurrent hypnagogic hallucinations and multiple SOREMPs (n ≥ 2) on MSLT are incompatible with ICSD-3 criteria for idiopathic hypersomnia. (4) Major depressive disorder: Transient dysphoria, anhedonia, and a PHQ-9 score of 17 were noted. Nevertheless, affective symptoms are neither pervasive nor of sufficient duration to fulfil DSM-5 criteria for a major depressive episode, and no neurovegetative syndrome was elicited on mental state examination. Depressive disorder cannot be entirely dismissed, but longitudinal observation is required to clarify possible comorbidity. Following confirmation of the diagnosis, amisulpride and blonanserin were discontinued. Pitolisant 9–36 mg/day was initiated to promote arousal. MSLT demonstrated short sleep-onset latencies with multiple SOREMPs; consequently, fluoxetine was titrated from 20 to 40 mg/day to enhance motivation and suppress REM sleep. A structured psychotherapeutic programme grounded in cognitive behavioural therapy for insomnia (CBT-I) was delivered, encompassing psycho-education, sleep hygiene counselling, and stimulus control techniques. Owing to an abbreviated admission (8 days), only two formal CBT-I sessions were completed. Adjunctive bilateral repetitive transcranial magnetic stimulation (rTMS) was administered over the left dorsolateral prefrontal cortex at 10 Hz, 120% resting motor threshold and 3000 pulses/session, on five consecutive weekdays; parameters conventionally employed for major depressive disorder were selected in view of the patient's prominent amotivation. Following initiation of the revised regimen, diurnal hypersomnia decreased markedly, hypnagogic auditory hallucinations remitted completely, and sub-syndromal depressive symptoms abated. After a 6-day interval, the patient re-performed the PSG and MSLT tests. The number of night awakenings decreased, and the amount of deep sleep increased. However, the total sleep time and sleep efficiency were lower than the baseline because the measurements were taken at the early stage of the post-treatment period. At the 2-month outpatient follow-up, the patient reported significant improvements in his sleep, including normalized sleep duration, sleep efficiency and the complete resolution of hypnagogic hallucinations. Social and executive functioning returned to nomal. The patient engaged appropriately with family members and independently performed household tasks. At the four-month post-discharge follow-up (telemedicine): the patient had achieved occupational re-integration and was employed as a server in a beverage outlet. Standardised self-report scores: PHQ-9 = 3 and Epworth Sleepiness Scale (ESS) = 6. Unfortunately, during these two follow-up visits, the patient refused to undergo another PSG and MSLT tests, and Positive and Negative Syndrome Scale could not be re-administered owing to time constraints. However, collateral history from both patient and family revealed no resurgence of psychotic or negative symptoms. The index case illustrates how extensively NT1 can masquerade as schizophrenia. From a diagnostic perspective, the most challenging aspect was to distinguish NT1 from schizophrenia. First, the patient's ‘hallucinations’ were confined to sleep/wake transitions and short durations and accompanied by preserved insight, which is characteristic of hypnagogic hallucinations rather than the persistent, daytime hallucinations typically seen in schizophrenia. Second, social withdrawal and apparent psychomotor inertia occurred predominantly during periods of severe daytime somnolence and improved significantly upon effective treatment of narcolepsy, whereas in schizophrenia, social withdrawal tends to be progressive and persistent. Third, although the patient reported transient attention and memory lapses during sleep attacks, these cognitive complaints were resolved with improved alertness, and no persistent global cognitive decline was observed—contrasting with the often enduring cognitive impairment in schizophrenia. Fourth, two sequential trials of antipsychotic medications (unspecified agent at an external hospital, followed by amisulpride and blonanserin at our institution) yielded little sustained benefit and even exacerbated hypersomnia, a pattern inconsistent with typical antipsychotic response in schizophrenia. Finally, objective measurements—notably the MSLT demonstrating a mean sleep latency of 0.6 min with five SOREMPs, which significantly decreased CSF hypocretin-1, and a high-risk immunogenetic profile—provided strong biomarker support for NT1, whereas during schizophrenia there was a lack of comparable diagnostic biomarkers. Collectively, these longitudinal observations argue against a primary psychotic disorder and firmly establish NT1 as the underlying condition. A growing literature highlights the overlap between narcolepsy and psychotic symptoms. Studies report that 33%–80% of NT1 patients experience hallucinations, which are typically visual or dissociative in nature.3-5 In atypical presentations, more vivid, complex hallucinations or dream–reality confusion may occur, sometimes leading to delusional elaboration. Moreover, a subset of patients develops comorbid schizophrenia spectrum disorders. One study found that 1.5% of patients diagnosed with NT1 received a subsequent diagnosis of schizophrenia within 1–2 years, particularly those among whom the onset was before 18 years of age.6 Another investigation of 102 NT1 children reported that 9.8% developed comorbid schizophrenia after a mean interval of 2.5 ± 1.8 years from NT1 onset.7 Compared with patients with schizophrenia, narcolepsy patients more frequently report visual hallucinations (simple images 15% and complex images 38%), non-verbal auditory hallucinations (e.g., footsteps, door slams and animal sounds: 50%) and tactile hallucinations (48%).3-7 Conversely, severe sleep–wake disturbances are highly prevalent in schizophrenia populations. Clinicians should therefore maintain vigilance for two possibilities: under-recognized narcolepsy presenting with psychiatric features and genuine comorbidity between narcolepsy and psychotic disorders. Genetic screening revealed that the patient is heterozygous for HLA-DQB1*06:02/05:02, conferring a high-risk haplotype for NT1. Furthermore, genetic analysis revealed the patient to carry the TCRA rs1154155 CC (homozygous) and P2RY11 rs2305795 GA (heterozygous) variants, indicating a high risk non-HLA susceptibility profile.8 In the present case, the coexistence of HLA-DQB1*06:02 with the high-risk TCRA rs1154155 and P2RY11 rs2305795 variants suggests a substantial immunogenetic susceptibility to NT1, in line with previous findings in Han Chinese cohorts. Rather than being an incidental finding, this multi-locus risk profile provides convergent support for an immune-mediated loss of hypocretin neurons as the core pathophysiological process. It is noteworthy that these narcolepsy-associated loci only minimally overlap with the major genetic risk reported for schizophrenia, further supporting that the patient's clinical picture reflects a primary sleep–wake disorder mimicking psychosis, rather than a psychotic disorder per se. The patient endorsed only intermittent dysphoria and anhedonia, nevertheless, his PHQ-9 score was 17 (severe depression) and his GAD-7 score was 13 (moderate anxiety). This discrepancy may be explained by the substantial overlap between symptoms of severe hypersomnia (e.g., fatigue, low energy and anergia) and neurovegetative features of depression. Thus, the elevated scores likely reflect the burden of chronic sleep–wake dysregulation rather than an independent mood disorder. This interpretation is consistent with the rapid amelioration of affective complaints following effective treatment of narcolepsy. Narcolepsy is a lifelong disorder that mandates prompt, individualised and multimodal management once the diagnosis is confirmed. Pharmacological therapy was therefore revised: All antipsychotic agents were discontinued and replaced by pitolisant (a selective histamine H3-receptor antagonist/inverse agonist) augmented with fluoxetine. Pitolisant augments endogenous histaminergic tone in the tuberomammillary nucleus, thereby promoting sustained wakefulness, and is endorsed as first-line treatment for narcolepsy with or without cataplexy in current international guidelines.9 The 2022 Chinese guidelines for the diagnosis and management of narcolepsy designate pitolisant as the first-line agent for excessive daytime sleepiness (grade I recommendation and level A evidence) and as the standard pharmacotherapy for cataplectic episodes (grade I recommendation and level A evidence).4 In a prospective open-label study of 30 narcolepsy patients, pitolisant administration was associated with a rapid and sustained reduction in subjective sleep propensity: ESS scores fell significantly within 2 weeks and remained stable thereafter (all post-baseline visits p < 0.01 vs. baseline). Objective diary data revealed a parallel suppression of cataplexy: Daily attack frequency decreased from 0.83 ± 0.22 at the baseline to 0.41 ± 0.18 after 1 week, corresponding to a 50.6% reduction (p < 0.001).10 Non-pharmacological interventions constitute an essential component of evidence-based narcolepsy management. Structured daytime napping, rigorous sleep hygiene education, multidisciplinary psychosocial support and targeted cognitive behavioural therapy constitute first-line non-pharmacological measures. rTMS is currently under investigation as a neuromodulatory adjunct; preliminary data suggest that focal stimulation of cortico-thalamo-limbic circuits may enhance sustained wakefulness and ameliorate comorbid affective symptoms. In the index patient, an individualised, multi-level treatment algorithm was therefore implemented that integrated these non-pharmacological strategies with pharmacotherapy. The primary objectives were (1) reduction of excessive daytime sleepiness and sleep fragmentation, (2) mitigation of mood dysfunction and (3) maximization of functional recovery and health-related quality of life. In conclusion, the presence of atypical clinical features necessitates heightened vigilance against diagnostic error. A systematic reassessment of sleep–wake history, supplemented by targeted PSG and neurobiological investigations, is essential to penetrate superficial symptomatic presentations and realign therapeutic priorities toward circadian restoration. We advocate the establishment of a formal, trans-disciplinary protocol bridging psychiatry and sleep medicine to minimise misclassification and optimise patient outcomes. Feng Daoyi: Writing—review and editing; writing—original draft; conceptualization. Liu Xin: Conceptualization; methodology; project project We to all members of the for Sleep and of the of for their and We the patient's family members for their and support during the of this case The no of This case report is on routine clinical and not Therefore, was not However, was from the patient for the of this The for the of the and the of the The data that support the findings of this study are from the corresponding upon