Narcolepsy–cataplexy and schizophrenia in adolescents

Sleep Medicine xxx (2013) xxx–xxx

Contents lists available at ScienceDirect

Sleep Medicine journal homepage: www.elsevier.com/locate/sleep

Original Article

Narcolepsy–cataplexy and schizophrenia in adolescents Yu-Shu Huang a,b,c, Christian Guilleminault d,⇑, Chia-Hsiang Chen c,e,f, Ping-Chin Lai g, Fan-Ming Hwang h a

Sleep Center, Chang Gung Memorial Hospital and University, Linkou, Taiwan Child Psychiatry Department, Chang Gung Memorial Hospital and University, Linkou, Taiwan c Psychiatry Department, Chang Gung Memorial Hospital and University, Linkou, Taiwan d Stanford University Sleep Medicine Division, Stanford, CA, USA e Institute of Medical Sciences, Tzu-Chi University, Hualien, Taiwan f Division of Mental Health and Addiction Medicine, Institute of Population Health Sciences, National Health Research Institutes, Zhunan, Taiwan g Kidney Research Center, Department of Nephrology, Chang Gung Memorial Hospital and University, Linkou, Taiwan h Department of Education, National Chia-Yi University, Chiayi, Taiwan b

a r t i c l e

i n f o

Article history: Received 1 January 2013 Received in revised form 21 September 2013 Accepted 24 September 2013 Available online xxxx Keywords: Narcolepsy Cataplexy Adolescent Schizophrenia HLA Chinese and HLA DQ B1-03:01

a b s t r a c t Background: Despite advances in the understanding of narcolepsy, little information the on association between narcolepsy and psychosis is available, except for amphetamine-related psychotic reactions. Our case-control study aimed to compare clinical differences and analyze risk factors in children who developed narcolepsy with cataplexy (N–C), schizophrenia, and N–C followed by schizophrenia. Methods: Three age- and gender-matched groups of children with N–C schizophrenia (study group), N–C (control group 1), and schizophrenia only (control group 2) were investigated. Subjects filled out sleep questionnaires, sleep diaries, and quality of life scales, followed by polysomnography (PSG), multiple sleep latency tests (MSLT), routine blood tests, HLA typing, genetic analysis of genes of interest, and psychiatric evaluation. The risk factors for schizophrenia also were analyzed. Results: The study group was significantly overweight when measuring body mass index (BMI) (P = .016), at narcolepsy onset compared to control group 1, and the study group developed schizophrenia after a mean of 2.55 ± 1.8 years. Compared to control group 2, psychotic symptoms were significantly more severe in the study group, with a higher frequency of depressive symptoms and acute ward hospitalization in 8 out of 10 of the subjects. They also had poorer long-term response to treatment, despite multiple treatment trials targeting their florid psychotic symptoms. All subjects with narcolepsy were HLA DQ B1⁄0602 positive. The study group had a significantly higher frequency of DQ B1⁄-03:01/06:02 (70%) than the two other groups, without any significant difference in HLA-DR typing, tumor necrosis factor a (TNFa) levels, hypocretin (orexin) receptor 1 gene, HCRTR1, and the hypocretin (orexin) receptor 2 gene, HCRTR2, or blood infectious titers. Conclusion: BMI and weight at onset of narcolepsy as well as a higher frequency of DQ B1⁄-03:01/06:02 antigens were the only significant differences in the N–C children with secondary schizophrenia; such an association is a therapeutic challenge with long-term persistence of severe psychotic symptoms. Ó 2013 Elsevier B.V. All rights reserved.

1. Introduction Narcolepsy with cataplexy (N–C) [1] is associated with daytime somnolence, disrupted nocturnal sleep, and episodes of abrupt complete or partial loss of muscle tone; the loss of muscle tone is mostly triggered by laughter or abrupt emotional involvement, with the disappearance of deep tendon reflexes during cataplexy. Hypnopompic and hypnagogic hallucinations and sleep paralysis are observed at various frequencies [2]. Nocturnal polysomnography ⇑ Corresponding author. Address: Stanford University Division of Sleep Medicine, 450 Broadway St, MC 5704, Redwood City, CA 94063, USA. Tel.: +1 650 723 6601; fax: +1 650 725 8910. E-mail address: [email protected] (C. Guilleminault).

(PSG) may uncover sleep-onset rapid eye movement periods (SOREMPs), though the multiple sleep latency test (MSLT) on the following day shows two or more SOREMPs during the five 20-min naps and less than 8 min mean sleep latency. N–C is associated with the presence of the HLA DQB1⁄06:02 allele, independent of ethnicity in at least 92% of the cases [3]; in addition, the cerebrospinal fluid analysis often reveals absence of or pathologically low levels of hypocretin. Presence of the HLA DQB1⁄06:02 allele has a sensitivity of 89.3% and a specificity of 76%; the presence of two or more SOREMPs at MSLT has a sensitivity of 87.9% and a specificity of 96.9%; and the complete absence of or low levels of hypocretin has a sensitivity of 83.3% and a specificity of 100% in patients with narcolepsy [4].

1389-9457/$ - see front matter Ó 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.sleep.2013.09.018

Please cite this article in press as: Huang Y-S et al. Narcolepsy–cataplexy and schizophrenia in adolescents. Sleep Med (2013), http://dx.doi.org/10.1016/ j.sleep.2013.09.018

2

Y.-S. Huang et al. / Sleep Medicine xxx (2013) xxx–xxx

Autopsy material shows destruction of the hypocretin (orexin) neurons that produce peptide hypocretin (orexin) in the lateral hypothalamus [5]. The notion that N–C may be an autoimmune disease has received more and more support due to further genetic analyses [6]. N–C is now frequently recognized in prepubertal and young teenagers due to systematic investigation and the emphasis on recent reports switches to the understanding of the health problems associated with hypocretin deficiency, not limited to sleepiness and cataplexy. Despite previous work indicating an association of narcolepsy in adult schizophrenics and the absence of recognition of narcolepsy syndrome [7–10], we still have little knowledge on the relationship between N–C and schizophrenia to date. The prevalence of N–C combined with schizophrenia has been previously estimated to be in 1–18 cases in a population of 2 million based on independent prevalence rates [11]. Stimulants and amphetamine-like drugs in particular have been used in the treatment of narcolepsy and the development of psychotic disorders in association with drug intake in narcoleptics also has been implicated in the literature [12]. Many studies have emphasized the strong association between narcolepsy and HLA, particularly the HLA-DQ and the HLA-DR protein [3]. In addition, a genetic association between specific HLA-DR genes and schizophrenia also has been shown [6,9,11,13], but the association between HLA alleles and schizophrenia is a field with continuous research efforts [14,15]. The question of a potential interaction between narcolepsy and schizophrenia, especially in children diagnosed with N–C, has mostly been unexplored particularly when looking at the chronologic development of the two syndromes and also questioning if the clinical presentation of the first syndrome gave any clue on occurrence of the second morbidity. We present the results of a retrospective investigation of our prospectively collected narcoleptic cases. We compared our findings obtained from N–C children who also developed schizophrenia to findings of age-matched children with N–C and age-matched children with schizophrenia only.

2. Methods 2.1. Subjects Our pediatric sleep center is the only pediatric sleep center for Taiwan (23 million inhabitants) and is the referring center for all children with abnormal sleepiness. The center is responsible for treatment and follow-up of all diagnosed children. During the 4 year prospective study, 151 children fulfilled all the International Classification of Sleep Disorders, second edition, criteria for diagnosis of Narcolepsy [16], but only 102 children presented with N–C. The prospectively collected 102 N–C children represented our total clinical initial group. Out of the 102 N–C children, 10 (9.8%) developed schizophrenia. Most of the children were diagnosed with hypersomnia and N–C first and subsequently developed Schneider first-rank symptoms within 3 years’ duration (mean time of schizophrenia onset, 2.55 ± 1.8 y). As shown below, the diagnosis of schizophrenia was based on thorough clinical evaluation by a child psychiatrist and positive testing on different scales. The children with N–C and schizophrenia (n = 10) comprised our study group. We also formed two control groups: control group 1 comprised age- and gender-matched N–C children without schizophrenia (n = 37/92); and control group 2 comprised a group of age- and gender-matched schizophrenic children without N–C (n = 13). In reviewing past teenage schizophrenia cases seen during the last 10 years in our large university pediatric psychiatric clinic, no schizophrenic child developed narcolepsy thereafter. Because the pediatric sleep laboratory and clinic are both part of the pediatric psychiatric division, all of these

previous patients received a similar evaluation as those in the study group. Children in the control and study groups completed similar testing procedures. Study approval was obtained from the Institutional Review Boards of Chang Gung Hospital, Taiwan. Written informed consent was obtained from subjects and their legal representatives following a detailed explanation of the study. 2.2. Diagnostic evaluation 2.2.1. N–C evaluation All cases underwent a standardized evaluation based on the international recommendations (International Classification of Sleep Disorders, second edition) for the diagnosis of N–C and hypersomnia [16]. Children underwent a general pediatric clinical evaluation. Body weight and height of subjects were assessed in a standardized fashion to calculate body mass index (BMI). Complete neurologic evaluation including electroencephalogram (EEG) (while awake and while asleep) and brain magnetic resonance imaging were systematically performed. Routine blood tests (complete blood cell count with differential count and biochemical blood tests [i.e., blood sugar, thyroid-stimulating hormone, thyroxine, liver function, and renal function]) were obtained. Systematic HLA typing was performed on all children. Blood also was drawn for other potential genetic studies. Parents and children filled out the following sleep questionnaires (validated in Mandarin): the Pediatric Daytime Sleepiness Scale (PDSS) [17], the Epworth Sleepiness Scale (ESS) if the child was old enough to drive [18], sleep diaries for 14 days, and the Stanford Narcolepsy Questionnaire [19]. They also were asked to fill out sleep diaries for a minimum of 14 successive days, with notation of indicators, triggers, and duration of each cataplexy attack; and four daily visual analog scales (VAS) (scored 1–100), which observed excessive daytime sleepiness (EDS), presence of hypnagogic/hypnopompic hallucination, and sleep paralysis. Each patient underwent 14 days of actigraphy tests, which observed the amount of sleep inactivity during the night and during the daytime. Nocturnal PSG also was observed for a minimum of 7 h, with monitoring of the following variables: EEG (C3/A2, C4/A1, Fz/A1–A2, and O1/A2); right and left electrooculogram; chin and legs electromyography; electrocardiography with a modified V2 lead; nasal cannula pressure transducer; mouth thermistor; chest and abdomen inductive plethysmography bands; neck microphone; and finger oximetry, from which oximetry curve and finger plethysmography were extracted and recorded. The following morning, an MSLT was administered to each patient at 2-h intervals consisting of five 20-min naps; mean sleep-latency and presence of SOREMPs were calculated. After the PSG, patients were submitted to video-monitored challenges reported by family members to trigger cataplectic attacks; such video monitoring allowed the observers to replay the attacks to determining the presence of complete and partial attacks, investigate the segments of the body involved during an ‘‘attack,’’ and to affirm the presence of cataplexy. Interaction with family members usually was successful in inducing cataplectic attacks. Attacks provoked in the sleep medicine facilities were witnessed by a physician expert in narcolepsy, and deep tendon reflexes were checked during the attack and just after recovery when the attack was of sufficiently long duration [20]. 2.2.2. Schizophrenia evaluation A structured psychiatric interview was conducted using the Schedule for Affective Disorder and Schizophrenia for school-aged children, adolescent version (K-SADS-E) [21] by experienced child psychiatrists. Schizophrenia was diagnosed based on Diagnostic and Statistical Manual of Mental Disorders, fourth edition, Text

Please cite this article in press as: Huang Y-S et al. Narcolepsy–cataplexy and schizophrenia in adolescents. Sleep Med (2013), http://dx.doi.org/10.1016/ j.sleep.2013.09.018

Y.-S. Huang et al. / Sleep Medicine xxx (2013) xxx–xxx

Revision (DSM-IV-TR) diagnostic criteria [22]. Inventories used to assess psychiatric health included the Positive and Negative Symptom Scale (PANSS), the Beck Depression Inventory, the Beck Anxiety Inventory, and the Clinical Global Impressions severity scale (CGI-S) [23–25]. Children also had a clinical evaluation performed by a sleep specialist and demonstrated absence of features suggestive of narcolepsy at PSG testing. All children were regularly followed over a minimum of 6 years, with adjustment of respective treatment as needed. 2.2.3. Medications All patients with N–C except for one received medication to control EDS. The Taiwan health recommended guidelines for treatment of EDS in narcolepsy are as follows: children are initially treated with methylphenidate (MPH) for 1–2 months at a dose of 0.3–0.7-mg/kg daily. Thereafter, the recommendation is to switch narcoleptic patients to 200 mg of modafinil in the morning. In addition, subjects with N–C also may receive anticataplectic treatment. Patients with schizophrenia were treated with antipsychotic medications as needed. 2.2.4. Family history evaluation Positive family history of narcolepsy was affirmed based on demonstration of EDS and cataplexy in siblings, parents, or second-degree relatives. We asked for medical documentation of the reported problems. Similarly, positive family history of psychiatric disorders was based on the presence of a DSM-IV-TR-documented psychiatric diagnosis in subsequent family members. 2.3. Statistical analysis 2.3.1. Data scoring Questionnaires were scored according to the recommended scales [17–19,21,23–25]. PSG and MSLT data were analyzed following the recommendations of the American Academy of Sleep Medicine [26]. Sleep–wake data were obtained by identifying slow-wave sleep (nonrapid eye movement [NREM] sleep stage 3) and dissociation of stage 3 and 4 NREM sleep following the international criteria from the Rechtschaffen and Kales atlas [27]. Short EEG arousals were scored following the American Sleep Disorders Association guidelines [28], and mean sleep latency and presence of SOREMPs were extracted from MSLTs. 2.3.2. Comorbid associations Comorbidities were determined following DSM-IV-TR criteria. The statistical analyses were presented as mean, standard deviation, and percentages. Comparisons of quantitative variables were performed using the nonparametric Mann–Whitney U test and the Student t test for independent samples. The statistics from the v2 test were used to compare percentages. We first performed an analysis of variance across all three groups and a post hoc test with Bonferroni adjustment for comparison. If a variable was not normally distributed, a Kruskal–Wallis analysis of variance on ranks and a Dunn test were performed. Pearson product moment correlation coefficients were obtained to determine the relationship between BMI and different sleepiness scales. Risk factors for schizophrenia were analyzed using multiple logistic regression analyses. The adjusted odds ratios (OR) for each significant factor are reported. 3. Results Demographics and clinical symptoms of the three groups of children are presented in Table 1. All narcoleptic patients revealed no abnormalities on neurologic evaluation (i.e., EEG, brain magnetic resonance imaging) on entry into the study. All blood tests

3

were within reference range, including tumor necrosis factor a (TNF-a) cytokines and IL-6. A significant difference in body weight and BMI at time of diagnosis of N–C was found. Most of the N–C children with schizophrenia were much more overweight on entry into the study, sometimes at the level of obesity, than the other two groups (BMI, P = .002; body weight, P = .026). A positive family history of narcolepsy was shown in four subjects in the narcolepsy with schizophrenia group, and a positive psychiatry family history was found to be higher in the narcolepsy with schizophrenia group (n = 4; sibling [n = 2], second-degree relatives [n = 2]) than in the pure narcolepsy control group (n = 2; sibling [n = 1], parent [n = 1]). 3.1. Timing of schizophrenia onset in subjects with narcolepsy Except for one subject who developed hypersomnia, cataplexy, and schizophrenia in that order but at nearly the same time, schizophrenic symptoms appeared much later than symptoms of N–C, and the diagnosis of schizophrenia was given an average of 2.55 ± 1.8 years following the onset of N–C. Before developing schizophrenia, narcoleptic subjects in the study group were no different than any of the other narcoleptic subjects in reports of hallucinations shown by the prospectively collected follow-up data, with systematic follow-up evaluation at least every 6 months. However, at the onset of psychotic symptoms, narcoleptic subjects with schizophrenia presented several more visual hallucinations of psychotic type than those with schizophrenia in the control group 2; delusions were more frequent and more severe in the schizophrenia control group 2, but there was no statistic significant difference compared to the study group (P = .67). Comorbidities observed in the three groups are presented in Table 2. The percentage of subjects with N–C and schizophrenia who presented with major depression was significantly higher than that noted in the other two control groups (P = .034). The results from various administered scales are presented in Table 3. As expected, schizophrenic control subjects showed no evidence of sleepiness (ESS, 6.08 ± 2.4; PDSS, 10.38 ± 3.12), though N–C controls had lower scores on the PANSS and CGI-S in psychotic symptoms. On the other hand, N–C subjects with schizophrenia and control subjects with isolated schizophrenia presented with similar scores on the PANSS-P (PANSS positive subscale), the ESS, and the PDSS at the onset of psychotic symptoms (Table 3). However, greater severity was noted in the study group: psychotic symptoms were severe from the onset of the schizophrenia in the narcoleptic study subjects (CGI-S, 4.90 ± 1.37; PANSS-P, 24.7 ± 4.06; PANSS-N [PANSS negative subscale], 25.7 ± 5.06). Eight out of 10 subjects had to be admitted to the acute psychiatry ward, with florid auditory hallucination and delusional behaviors. Subjects were followed for a mean of 3.45 ± 2.02 years since the onset of schizophrenia and were noted to have responded poorly to several antipsychotic medications and electroconvulsive therapy. To date, psychotic symptoms in these subjects have persisted and are still poorly controlled. 3.2. Results of specific investigations 3.2.1. Physical evaluation of BMI Significant correlations between BMI and CGI-S and PANSS scores were found in the study group but not in the two control groups (Table 4); higher BMI was associated with a greater severity on the PANSS-P (r = 0.632), PANSS-G (PANSS general psychopathologic subscale) (r = 0.644), and the CGI-S (r = 0.669) (psychotic symptoms). Risk factors for schizophrenia were analyzed by multiple logistic regression analyses and cross-tabulations of OR (see Table 5). The adjusted OR of each significant factor are reported, showing that BMI was the higher risk factor for schizophrenia (OR, 1.506; P = .002).

Please cite this article in press as: Huang Y-S et al. Narcolepsy–cataplexy and schizophrenia in adolescents. Sleep Med (2013), http://dx.doi.org/10.1016/ j.sleep.2013.09.018

4

Y.-S. Huang et al. / Sleep Medicine xxx (2013) xxx–xxx

Table 1 Demographics and clinical symptoms. Variable

Study group (n = 10) mean ± SD (count) (%)

Control group 1 (n = 37) mean ± SD (count) (%)

Control group 2 (n = 13) mean ± SD (count) (%)

P value

Current age (y) Boys Girls Age of narcolepsy onset (y) Age of schizophrenia onset (y) BMI (kg/m2) Family history of narcolepsy Family history of psychiatric disease

18.54 ± 2.99 5 (50.0%) 5 (50.0%) 11.25 ± 3.92 15.80 ± 1.36 27.39 ± 7.17a,c 4 (40.0%)c 4 (30.0%)a

18.82 ± 3.54 21 (56.8%) 16 (43.2%) 12.59 ± 3.41 – 23.62 ± 4.86a 5 (13.5%)b 2 (5.4%)a,b

18.38 ± 3.65 6 (46.2%) 7 (53.8%) – 16.38 ± 0.84 20.32 ± 3.13c 0 (0.0%)b,c 3 (23.1%)b

.268 .703

Symptoms and sign at time of diagnosis Hypersomnia Cataplexy Hypnogogic hallucination Sleep paralysis Parasomnia Delusion Auditory hallucination or visual hallucination

10 (100%)c 10 (100%)c 10 (100%) 10 (100%)a,c 8 (80.0%) 7 (70.0%)a 9 (90.0%)a

37 (100%)b 37 (100%)b 30 (81.1%) 25 (67.6%)a,b 22 (59.5%) 0 (0.0%)a,b 0 (0.0%)a,b

3 (23.1%)b,c 0 (0.0%)b,c 9 (69.2%) 2 (15.4%)b,c 6 (46.2%) 13 (100%)b 13 (100.0%)b

.0001 .0001 .166 .0001 .258 .0001 .0001

.269 .572 .002 .027 .016

Abbreviations: SD, standard deviation; y, years; BMI, body mass index. Study group: narcolepsy–cataplexy with schizophrenia.. Control group 1: narcolepsy–cataplexy without schizophrenia. Control group 2: only schizophrenia. Comparison of means: Kruskal–Wallis H test. Comparison of counts (%): v2 test. a Post hoc analysis showed significance of study group and control group 1. b Post hoc analysis showed significance of control group 1 and control group 2. c Post hoc analysis showed significance of control group 2 and study group. a,b BMI: Mann–Whitney U test (1 tailed).

Table 2 Comorbidity. Variable

Experimental group (n = 10) mean ± SD (count) (%)

Major depression Obesity (BMI =25 kg/m2) ADHD history ODD Asperger syndrome PLMS OSA OCD Insomnia

5 9 3 1 2 0 4 0 3

(50.0%)a,c (75.0%)a,c (30.0%) (8.3%) (20.0%) (0.0%) (40.0%) (0.0%) (30.0%)

Control group 1 (n = 37) mean ± SD (count) (%)

Control group 2 (n = 13) mean ± SD (count) (%)

5 (21.6%)a 14 (37.8%)a,b 2 (5.9%) 4 (10.8%) 1 (2.7%) 4 (10.8%) 8 (21.6%) 0 (0.0%) 11 (29.7%)

2 2 2 0 0 1 2 1 5

(15.4%)c (15.4%)b,c (15.4%) (0.0%) (0.0%) (7.7%) (15.4%) (7.7%) (38.5%)

P value .034 .026 .357 .264 .289 .545 .355 .159 .838

Abbreviations: SD, standard deviation; BMI, body mass index; ADHD, attention-deficit/hyperactivity disorder; ODD, oppositional defiant disorder; PLMS, periodic limb movements during sleep (PLMI >5/h with daytime symptoms); OSA, obstructive sleep apnea (apnea–hypopnea index >5 events/h); OCD, obsessive–compulsive disorder. Comparison of means: Kruskal–Wallis H test. Comparison of counts (%): v2 test. Major depression was diagnosed using Diagnostic and Statistical Manual of Mental Disorders, 4th Edition, Text Revision (DSM-IV-TR) criteria. The diagnosis of insomnia was made according to criteria of DSM-IV-TR. a Post hoc analysis showed significance of study group and control group 1. b Post hoc analysis showed significance of control group 1 and control group 2. c Post hoc analysis showed significance of control group 2 and study group.

3.2.2. Sleep data In Table 3, analysis of variance showed a significant reduction of stage 2 NREM sleep in the N–C with schizophrenia group compared to the control group 2 (P = .006) on PSG. When compared to the group of schizophrenic control subjects, the two N–C subject groups showed significantly different mean sleep latency (1.74 ± 1.61; 2.97 ± 3.81 min) and number of SOREMPs (3.70 ± 1.34; 3.51 ± 1.38) compared to control group 2 schizophrenia in the MSLT, as expected. 3.2.3. HLA typing and blood tests All N–C patients were HLA DQB 1-0602 positive. There were three subjects (8.1%) with a homozygote for DQB1⁄06:02/⁄06:02 in the narcoleptic control group. There was no homozygote for

DQB1⁄06:02/⁄06:02 in the N–C subjects with schizophrenia and the schizophrenia-only control group 2, but there was a significant difference (P = .012) for DQB1⁄03:01/0 6:02 between the study group (70%), control group 1 (37.8%), and control group 2 (n = 0). Another interesting finding was the higher frequency of DQB1⁄06:01 in the schizophrenia control group 2. The results of the v2 test showed significant differences (P = .01) in the control group 2 schizophrenia subjects (46.1%), in the control group 1 narcolepsy subjects (8.1%), and the study group (10%). Analyses of the TNF-a cytokine and the hypocretin (orexin) receptor 1 gene, HCRTR1, and the hypocretin (orexin) receptor 2 gene, HCRTR2, showed no difference in any of the three investigated groups. We sequenced all exons of the dopamine D2 receptor gene, DRD2, investigating the possible mutations in the N–C subjects with schizophrenia, but no mutation in the

Please cite this article in press as: Huang Y-S et al. Narcolepsy–cataplexy and schizophrenia in adolescents. Sleep Med (2013), http://dx.doi.org/10.1016/ j.sleep.2013.09.018

5

Y.-S. Huang et al. / Sleep Medicine xxx (2013) xxx–xxx Table 3 Comparison of different administered scales and of sleep data. Variable

Experimental group (n = 10; a) mean ± SD (count) (%)

PANSS-P PANSS-N PANSS-G CGI-S ESS

24.70 ± 4.06) 25.70 ± 5.06 53.00 ± 7.23 4.90 ± 1.37 17.70 ± 2.00

PDSS

25.70 ± 4.60

VAS

Sleep variables MSLT variable Mean sleep latency (min) Number of sleep latency55 min Number of REM sleep (times) PSG variable AHI (times/h) AI (times/h) HI (times/h) Sleep efficiency % Awake % REM sleep % Stage 1% Stage 2% SWS % TST (min) Sleep latency (min)

Control group 1 (n = 37; b) mean ± SD (count) (%) – – – – 16.16 ± 3.00 22.19 ± 3.99 )

Control group 2 (n = 13; c) mean ± SD (count) n(%)

.018 .019 .004 .010 .0001

10.38 ± 3.12

.0001

23.08 ± 12.51

.0001

81.49 ± 12.18

1.74 ± 1.61

2.97 ± 3.81

10.24 ± 6.81

4.60 ± 0.97

4.03 ± 1.59

3.70 ± 1.34

3.51 ± 1.38

2.88 ± 3.97 0.38 ± 0.80 2.11 ± 3.13 83.81 ± 12.96 14.56 ± 12.08 18.18 ± 7.85 16.16 ± 9.27 40.13 ± 12.56 18.18 ± 11.11 378.35 ± 86.91 9.27 ± 15.47

P value

20.83 ± 3.16 19.77 ± 4.23 44.00 ± 4.24 3.69 ± 0.63 6.08 ± 2.40

92.50 ± 8.25

5.76 ± 7.60 1.43 ± 2.93 4.00 ± 5.57 76.32 ± 11.44 18.85 ± 10.87 20.44 ± 14.32 18.20 ± 4.84 28.25 ± 8.74 22.75 ± 10.17 324.10 ± 93.46 10.10 ± 15.80

F

Post hoc analysis

a > b,a > c, b>c a > b,a > c, b>c a > b, a > c, b>c

11.04

.0001

a < c, b < c

1.50 ± 1.52

9.00

.0001

a > c, b > c

0.57 ± 0.79

15.68

.0001

a > c, b > c

12.35 ± 3.20 2.11 ± 5.30 12.11 ± 8.38 79.93 ± 10.12 14.87 ± 12.75 16.84 ± 7.90 23.11 ± 23.77 49.39 ± 15.52 10.58 ± 17.04 385.13 ± 57.73 24.44 ± 16.64

7.69 1,84 8.71 1.56 0.462 0.35 1.14 5.71 0.03 2.70 3.14

.001 .170 .001 .219 .633 .708 .329 .006 .972 .077 .052

a < c, b < c NS a < c, b < c NS NS NS NS a
Abbreviations: SD, standard deviation; PANSS, Positive and Negative Symptom Scale; P, positive subscale; N, negative subscale; G, general psychopathologic subscale; CGI-S, Clinical Global Impression severity scale; ESS, Epworth Sleepiness Scale; PDSS, Pediatric Daytime Sleepiness Scale; VAS, visual analog scale (0–100) for daytime sleepiness; AHI, apnea–hypopnea index; AI, apnea index; HI, hypopnea index; MSLT, multiple sleep latency test; min, minutes; REM, rapid eye movement; PSG, polysomnography; h, hour; NS, not significant; SWS, slow-wave sleep; TST, total sleep time. Comparison of means: Kruskal–Wallis H test. Analysis of variance was used for sleep data.

DRD2 gene was found in our subjects. Measurement of blood titers to examine recent infection at time of diagnosis of narcolepsy in our narcoleptic subjects did not identify any peak of specific infectious agents (i.e., antistreptolysin O titer was low [16.7% greater than 200 IU/ml]; C-reactive protein also was low [3.3% greater than 5 mg/L]) and the distribution was not significant between our study group and control narcoleptic groups. 3.2.4. Treatment of narcoleptic patients Forty-four out of 47 of our narcolepsy subjects followed the recommended narcolepsy treatment guidelines, with initial treatment with methylphenidate (MPH) for 1–2 months (0.3–0.7 mg/kg/day); thereafter, they switched to 200 mg of modafinil in the morning. These 44 narcolepsy subjects were on a chronic medication regimen of venlafaxine (75–150 mg/day) which also was administered for the treatment of cataplexy in 19 out of 47 narcoleptic subjects, including three patients who developed schizophrenia. Finally, three of the studied N–C subjects, including one subject who later became schizophrenic, were drug naïve. At the onset of schizophrenic symptoms, administration of modafinil and venlafaxine to 5 out of 10 patients was discontinued for a minimum of 3–6 months without any change in symptomatology. 4. Discussion To our knowledge, our study is the first study in which children with N–C were prospectively collected and schizophrenia was secondarily diagnosed at systematic follow-up. Excluding one subject who simultaneously developed N–C and schizophrenia, all other

subjects had narcolepsy syndrome first. Overall, our narcolepsy subjects had clinical symptoms at an earlier age than pediatric schizophrenia patients when comparing our narcoleptic database to our large child psychiatry division database. Our N–C subjects had an onset of symptoms at an earlier age than often mentioned in white children, but the cataplectic attacks observed in our clinical setting were typical and closely corresponded to what was reported by Guilleminault and Pelayo [29] and reemphasized by Serra et al. [30]. N–C has a similar incidence rate in Taiwan to that reported in populations from European and North American countries (i.e., 0.03% of the general population) [31]. Moreover, HLADQB1⁄06:02 also was found to be present at the same frequency as elsewhere, which has been shown in our previous studies [31]. The suggestion that the major histocompatibility complex (MHC), a complex genetic locus known to influence immune regulation, plays a role in schizophrenia is outdated [32], but the genome-wide association studies have given much more credence to its involvement in schizophrenia risk. Many of the HLA-linked autoimmune inflammatory disorders are found to co-occur significantly more in schizophrenia [33,34]. A number of various autoimmune diseases also have been found to be present with a higher or lower prevalence compared to what would normally be expected in the general population. In 1998 Grosskopf et al. [35] reported an increased frequency of the DQB1⁄06:02 allele associated with narcolepsy and multiple sclerosis. Studies on schizophrenia also have shown presence of a high number of autoantigens and antibodies to neurotransmitters, including dopamine, serotonin, acetylcholine, N-methyl-D-aspartate, and other receptors [36]. Two factors important for our association of schizophrenia

Please cite this article in press as: Huang Y-S et al. Narcolepsy–cataplexy and schizophrenia in adolescents. Sleep Med (2013), http://dx.doi.org/10.1016/ j.sleep.2013.09.018

6

Y.-S. Huang et al. / Sleep Medicine xxx (2013) xxx–xxx

with narcolepsy have been shown to increase the risk for schizophrenia and involve the MHC: autoimmune reactions and central nervous system (CNS) inflammation. It also has been shown that MHC molecules that have immune functions display greater variance in children than in adults, emphasizing the role of these genes in neuronal and synaptic plasticity and showing that it also can cause aberration in the establishment of neural circuit and repair if such role can be positive [37]. Because genome-wide association studies can easily consider ethnicity, recent results on schizophrenia have shown that some specific MHC variants were specifically noted in the Chinese and Japanese populations [37,38], a fact to keep in mind when considering our association between N–C and schizophrenia. Schizophrenia also has been related to neuroinflammation with activation of microglia [39]. However, our studies showed no difference in the TNF-a cytokine and the HCRTR1 and HCRTR2 genes between groups, and there was no evidence of abnormal blood levels of TNF-a cytokine in our children. There have been several proposed explanations for the association of schizophrenia, comorbid autoimmune diseases, and HLA, such as direct involvement of the HLA molecules in the development of both syndromes; closeness of the loci associated with the autoimmune disorder (narcolepsy) and schizophrenia in the HLA region; shared genetic diathesis of schizophrenia with the family of autoimmune diseases such as narcolepsy; and pleiotropic effects of HLA genes [15]. Our study does not support one hypothesis more than another at this time. However, there are some important findings in our pediatric subjects with N–C. In our study, 9.8% of our narcoleptic patients developed schizophrenia at an average age of 2.5 ± 1.8 years after narcolepsy onset. The narcolepsy disorder was the first entity to appear and was a syndrome that best supported the presence of an immune disorder involving the brain. Development of schizophrenia in 9.8% of our cases is a high number. The subjects in our study group were modestly younger than our narcolepsy control subjects (mean age of onset of narcolepsy, 11.25 ± 3.92 years vs 12.59 ± 3.41 years) and were still prepubertal or in the earlier stages of puberty at the age of onset. The occurrence of N–C as a brain autoimmune disease in subjects who are younger than usual explains that HLA genes will play a greater role on synaptic plasticity and may lead more to aberrations in the establishment of repairs [40,41]. Early puberty is of important ‘‘pruning’’ of synapses in all children [42]. Finally, 70% of our narcolepsy schizophrenic subjects were DQB1⁄03:01/DQB1⁄06:02 positive, which is the last haplotype that has been found to further increase susceptibility to N–C in Chinese individuals; it also has recently been found that this haplotype had a strong effect on earlier age of onset of the syndrome in Chinese individuals [43,44]. Earlier age of onset will place children at greater risk for developing more brain disorders at the time of synaptic pruning occurring with puberty, including schizophrenia [40]. Younger age of onset also meant that our narcolepsy schizophrenic subjects received methylphenidate at a slightly earlier age than the group with isolated N–C, and seven (70%) N–C subjects with schizophrenia were diagnosed with schizophrenia after using a CNS drug, MPH, (for 1–2 months), and modafinil thereafter. Moreover, 3 of the 7 subjects treated with alerting agents also received venlafaxine treatment. Auger et al. [12] reported the association between the development of psychotic symptoms and CNS stimulants intake (e.g., MPH, amphetamine), but the authors concluded that the psychotic symptoms were predominantly seen in patients with high drug dosage and that the symptoms abetted with CNS stimulants withdrawal. We do not favor a pharmacologic agent role in the development of schizophrenia in our young narcoleptics patients. In our subjects, dosages were low compared to those reported in Auger et al. [12]. In addition, the duration of treatment spanned 1 to 2 months only, with interruption of treatment at least one year before the development of psychotic

symptoms. Modafinil has not been reported to be associated with the onset of psychotic symptoms. Finally, both higher dosages of and withdrawal from all narcolepsy-related medications for a minimum of 3–6 months had no impact on the psychotic symptomatology, particularly the auditory hallucinatory behavior and delusional presentation, which remained severe. The difference between hypnagogic hallucinations of narcolepsy and those seen in psychotic patients were well-described by Ribstein [45] and were revisited by Fortyun et al. [46]; the difference between the narcoleptic and psychotic hallucination was clear in our subjects, despite a prominence of visual hallucinations at onset of the psychosis. The psychotic hallucination and other psychotic symptoms had poor responsiveness to the pharmacologic antipsychotic treatment, which consisted of typical and atypical antipsychotics and later electroconvulsive therapy, due to the severity of the symptoms. Douglass et al. [7,47] and Wilcox [8] reported on adult schizophrenia patients who also were found to have narcolepsy, and both authors emphasized the severity of the schizophrenic symptoms and their poor response to treatment including atypical antipsychotic agents. Douglass et al. [47] recommended the use of stimulants to ‘‘fully recover,’’ but our narcoleptic patients were placed back on their narcoleptic medical stimulant regimen including venlafaxine without improvement of their psychotic symptoms. To date, despite several years of psychiatric treatment, psychotic symptoms continue to be present in these children. The only consistent finding of our study was the presence of an abnormally elevated BMI at the onset of narcolepsy, and our analysis of risk factors for schizophrenia in this group using multiple logistic regressions showed that BMI was a significant risk factor. Increases in BMI often are seen in narcoleptic patients and have been related to the destruction of the hypocretin (orexin) neurons; the clinical impression is that overweight and obese narcoleptic patients are seen fairly quickly when the child is younger at the onset of symptoms. Our children who developed schizophrenia had a fast and abnormal weight increase. They were already overweight with an elevated BMI when first seen, but their weight increase was determined to be recent after observing their clinical history. The destruction of the hypocretin (orexin) pathways not only had an impact on sleep–wake cycles, but also on autonomic regulation and food intake [48]. This finding suggests that the initial symptoms of the narcolepsy syndrome were not obvious early on, which did not lead parents to consult with professionals, and that children may have had a fast dysfunction of food intake; or that the destruction of the hypocretin neurons may differ depending on the subject, with a more important demonstration of other symptoms than sleep–wake disturbances associated with hypocretin destruction when the subject is younger. The speed of destruction of the hypocretin neurons and amount of destruction at the onset of the syndrome is unknown. It is possible that the anatomic extension of the lesions in the lateral hypothalamus occured at different speeds across individuals. For example, would a larger area of destruction lead to increased uncontrolled activities in specific neurotransmitters, which may normally have been activated or inhibited during food intake, absorption, and other functions depending on the hypocretin (orexin) activity? The neuropeptide, dynorphin (peptides involved in the regulation of energy balance), and hypocretin in the hypothalamic neurons have been reported to be related to the regulation of energy balance [48]. However, it is impossible to provide any answers at this stage. We have mentioned a significantly higher frequency of the DQB1⁄03:01/06:02 allele in our study group (70%) compared to other narcolepsy patients, not only increasing susceptibility for N–C, but also most likely favoring an earlier age of onset in Chinese individuals; however, no homozygous differences [49] or higher frequencies of the DQB1⁄06:01 allele were found in the

Please cite this article in press as: Huang Y-S et al. Narcolepsy–cataplexy and schizophrenia in adolescents. Sleep Med (2013), http://dx.doi.org/10.1016/ j.sleep.2013.09.018

7

Y.-S. Huang et al. / Sleep Medicine xxx (2013) xxx–xxx Table 4 Correlation between the body mass index and different psychometric scales. BMI

ESS

Study group Control group 1 Control group 2

BMI

PDSS

.165 .128 .394

VAS

.006 .273 .321

.542 .159 .402

PANSS-P ⁄

.631 – .272

PANSS-N .491 – .284

PANSS-G

CGI_S

⁄

.669⁄ – .134

.644 – .235

Abbreviations: BMI, body mass index; ESS, Epworth Sleepiness Scale; PDSS, Pediatric Daytime Sleepiness Scale; VAS, visual analog scale; PANSS, Positive and Negative Symptom Scale; P, positive subscale; N, negative subscale; G, general psychopathological subscale; CGI-S, Clinical Global Impression severity scale. Pearson product moment correlation coefficients: ⁄correlation is significant at the P = .05 level (2 tailed).

Table 5 Risk factors for schizophrenia.

Acknowledgments

Variables

v2

OR

95% CI

P value

ESS PDSS VAS BMI (kg/m2)

18.05 17.34 22.97 0.01

0.030 0.052 0.021 1.056

0.004–0.260 0.010–0.271 0.002–0.184 0.525–4.320

.295 .601 .069 .002

Abbreviations: OR, odds ratio; CI, confidence interval; ESS, Epworth Sleepiness Scale; PDSS, Pediatric Daytime Sleepiness Scale; VAS, visual analog scale; BMI, body mass index. Logistic regression analyses with schizophrenia as independent variable and crosstabulations of OR. ESS, cutoff point of 10. PDSS, cutoff point of 16. VAS (0–100) for daytime sleepiness, cutoff point of 55. BMI, cutoff point of 25 kg/m2.

schizophrenia control group 2 (46.1%) compared to the two other groups. Significant associations between HLA and schizophrenia were reported with the DQB1⁄06:02 allele in a Chinese sample by Nimgaonkar et al. [50] and with the DQB1⁄03 allele in a Saudi Arabian sample by Kadasah et al. [11]. The hypocretin (orexin) system has a wide network of interactions with its five clear pathways [48]; yet, we continue to ignore many of the interactions occurring outside of the impact on the sleep-wake system. There also are questions regarding the triggers of the autoimmune processes considered to be behind the hypocretin neuron destruction, even in narcolepsy. We could not identify a difference in the habits of our subjects with isolated N–C and those of our study group subjects. Moreover, extensive questioning did not reveal striking events that could be related to the onset of the narcoleptic syndrome or the schizophrenia symptoms. As mentioned above, we questioned if younger children’s brains are more sensitive to the autoimmune process strongly suggested to occur with the development of narcolepsy, or if there is an interaction between genes involved in the development of narcolepsy and some specific genes related to occurrence of schizophrenia [12,13,32–34]. We have no response to these hypotheses to date. The genetic studies that we performed did not bring any real insight, and a group of 10 narcolepsy schizophrenic subjects is not sufficient enough to perform a genome-wide analysis. These subjects not only raised many questions about the interaction between narcolepsy and schizophrenia, they also raised the question of the role of the HLA system in both disorders. Our narcolepsy schizophrenic teenagers are a therapeutic challenge, due to the severity of their clinical presentation and their poor response to various antipsychotic therapeutic trials. In addition, being obese early on in life causes important metabolic changes to develop quickly after the onset of narcolepsy and poses additional therapeutic challenges when dealing with schizophrenia. Conflict of interest The ICMJE Uniform Disclosure Form for Potential Conflicts of Interest associated with this article can be viewed by clicking on the following link: http://dx.doi.org/10.1016/j.sleep.2013.09.018.

This research was supported by the National Science Council: NSC 98-2314-B-182A-040-MY3 (Huang YS: PI). The authors would like to thank PhD Po Yu Huang and Professor Fan-Ming Hwang for their help with statistical analysis.

References [1] De la Gélineau J. Narcolepsie. Gaz Hop (Paris) 1880;53. 626–8, 635–737. [2] Guilleminault C, Dement WC, Passouant P. Narcolepsy. New York: Spectrum Publications; 1976. [3] Mignot E, Hayduk R, Black J, Grumet FC, Guilleminault C. HLA DQB1⁄0602 is associated with cataplexy in 509 narcoleptic patients. Sleep 1997;20:1012–20. [4] Mignot E. Narcolepsy: pathophysiology and genetic predisposition. In: Kryeger MH, Roth T, Dement WC, editors. Principles and practice of sleep medicine. Saint Louis (MO): Elsevier-Sauders; 2011. p. 939. [5] Blouin AM, Thannickal TC, Worley PF, Baraban JM, Reti IM, Siegel JM. Narp immunostaining of human hypocretin (orexin) neurons: loss in narcolepsy. Neurology 2005;65:1189–92 [published online ahead of print August 31, 2005]. [6] Hallmayer J, Faraco J, Lin L, Hesselson S, Winkelmann J, Kawashima M, et al. Narcolepsy is strongly associated with the T-cell receptor alpha locus. Nat Genet 2009;41:708–11 [published online ahead of print May 3, 2009]. [7] Douglass AB, Shipley JE, Haines RF, Scholten RC, Dudley E, Tapp A, et al. Schizophrenia, narcolepsy, and HLA-DR15, DQ6. Biol Psychiatry 1993;34:773–80. [8] Wilson J. Psychopathology and narcolepsy. Neuropsychobiology 1985;14:170–2. [9] Kishi Y, Konishi S, Koizumi S, Kudo Y, Kurosawa H, Kathol RG. Schizophrenia and narcolepsy: a review with a case report. Psychiatry Clin Neurosci 2004;58:117–24. [10] Talih FR. Narcolepsy presenting as schizophrenia: a literature review and two case reports. Innov Clin Neurosci 2011;8:30–4. [11] Kadasah S, Arfin M, Tariq M. HLA-DRB1 association with schizophrenia in Saudi Arabian patients. Int J Psychiatry Clin Pract 2011;15:112–7 [published online ahead of print November 30, 2010]. [12] Auger RR, Goodman SH, Silber MH, Krahn LE, Pankratz VS, Slocumb NL. Risks of high-dose stimulants in the treatment of disorders of excessive somnolence: a case-control study. Sleep 2005;28:667–72. [13] Krause D, Wagner J, Matz J, Weidinger E, Obermeier M, Riedel M, et al. Monocytic HLA DR antigens in schizophrenic patients. Neurosci Res 2012;72:87–93 [published online ahead of print September 22, 2011]. [14] Schwab SG, Hallmayer J, Freimann J, Lerer B, Albus M, Borrmann-Hassenbach M, et al. Investigation of linkage and association/linkage disequilibrium of HLA A-, DQA-, DQB1-, and DRB1-alleles in 69 sib-pairs and 89 trio-families with schizophrenia. Am J Med Genet 2002;114:315–20. [15] Debnath M, Cannon DM, Venkatasubramanian G. Variation in the major histocompatibility complex (MHC) gene family in schizophrenia: associations and functional implications. Prog Neuropsychopharmacol Biol Psychiatry 2013;42:49–62 [published online ahead of print July 17, 2012]. [16] American Academy of Sleep Medicine. The international classification of sleep disorders: diagnostic and coding manual. 2nd ed. Westchester (IL): American Academy of Sleep Medicine; 2005. [17] Yang CM, Huang YS, Song YC. Clinical utility of the Chinese version of the Pediatric Daytime Sleepiness Scale in children with obstructive sleep apnea syndrome and narcolepsy. Psychiatry Clin Neurosci 2010;64:134–40. [18] Chen NH, Johns MW, Li HY, Chu CC, Liang SC, Shu YH, et al. Validation of a Chinese version of the Epworth Sleepiness Scale. Qual Life Res 2002;11:817–21. [19] Anic-Labat S, Guilleminault C, Kraemer HC, Meehan J, Arrigoni J, Mignot E. Validation of a cataplexy questionnaire in 983 sleep-disorders patients. Sleep 1999;22:77–87. [20] Guilleminault C, Wilson R, Dement WC. A study on cataplexy. Arch Neurol 1974;31:255–61. [21] Orraschel H, Puig-Antich J. Schedule for affective disorders and schizophrenia for school-age children: epidemiologic. 4th ed. Ft. Lauderdale (FL): Nova University, Center for Psychological Study; 1987.

Please cite this article in press as: Huang Y-S et al. Narcolepsy–cataplexy and schizophrenia in adolescents. Sleep Med (2013), http://dx.doi.org/10.1016/ j.sleep.2013.09.018

8

Y.-S. Huang et al. / Sleep Medicine xxx (2013) xxx–xxx

[22] American Psychiatric Association. Diagnostic and statistical manual of mental disorders. 4th ed. Washington (DC): American Psychiatric Association; 2000. Text Revision. [23] Kay SR. Positive and negative syndromes in schizophrenia. New York (NY): Brunner-Mazel; 1991. [24] Beck AT, Ward CH, Mendelson M, Mock J, Erbaugh J. An inventory for measuring depression. Arch Gen Psychiatry 1961;4:561–71. [25] Beck AT, Epstein N, Brown G, Steer RA. An inventory for measuring clinical anxiety: psychometric properties. J Consult Clin Psychol 1988;56:893–7. [26] Iber C, Ancoli-Israel S, Chesson A, Quan SF. The American academy of sleep medicine manual for the scoring of sleep and associated events: rules, terminology and technical specifications. Westchester (IL): Academy of Sleep Medicine; 2007. [27] Hori T, Sugita Y, Koga E, Shirakawa S, Inoue K, Uchida S, et al. Sleep Computing Committee of the Japanese Society of Sleep Research Society. Proposed supplements and amendments to ‘A manual of standardized terminology, techniques and scoring system for sleep stages of human subjects’, the Rechtschaffen & Kales (1968) Standard. Psychiatry Clin Neurosci 2001;55:305–10. [28] EEG arousals: scoring rules and examples: a preliminary report from the Sleep Disorders Atlas Task Force of the American Sleep Disorders Association. Sleep 1992;15:173–84. [29] Guilleminault C, Pelayo R. Narcolepsy in pre-pubertal children. Ann Neurol 1998;43:135–42. [30] Serra L, Montagna P, Mignot E, Lugaresi E, Plazzi G. Cataplexy features in childhood narcolepsy. Mov Disord 2008;23:858–65. [31] Huang YS, Tafti M, Guilleminault C. Daytime sleepiness with and without cataplexy in Chinese–Taiwanese patients. Sleep Med 2006;7:454–7. [32] Cazuello CL, Simeraldi E, Penati G. The leucocyte antigenic system HLA-A as possible genetic marker of schizophrenia. Br J Psychiatry 1974;125:612–5. [33] Benros ME, Nielsen PR, Nordentoft M, Eaton WW, Dalton SO, Mortensen PB. Autoimmune diseases and severe infections as risk factors for schizophrenia: a 30-year population-based register study. Am J Psychiatry 2011;168:1303–10. [34] Chen SJ, Chao YL, Chen CY, Chang CM, Wu ECH, Wu CS, et al. Prevalence of autoimmune diseases in in-patients with schizophrenia: nation-wide population-based study. Br J Psychiatry 2012;200:374–80 [published online ahead of print March 22, 2012]. [35] Grosskopf A, Müller N, Malo A, Wank R. Potential role for the narcolepsy and multiple sclerosis associated HLA allele in schizophrenia subtypes. Scizophr Res 1998;30:187–9. [36] Carter CJ. Schizophrenia: a pathogenetic autoimmune disease caused by viruses and pathogens and dependent on genes. J Pathog 2011;2011:128318 [published online ahead of print May 26, 2011]. [37] Wu ZP, Washburn L, Bilousova TV, Boudzinskaia M, Escande-Beillard N, Querubin J, et al. Enhanced neuronal expression of major histocompatibility

[38]

[39]

[40] [41]

[42]

[43]

[44] [45]

[46]

[47]

[48]

[49]

[50]

complex class I leads to aberrations in neurodevelopment and neurorepair. J Neuroimmunol 2011;232:8–16 [published online ahead of print October 14, 2010]. Ikeda M, Aleksic B, Yamada K, Iwayama-Shiegeno Y, Matsuo K, Numata S, et al. Genome-wide association study of schizophrenia in a Japanese population. Biol Psychiatry 2011;65:472–8. Monji A, Kato T, Mizoguchi Y, Horikawa H, Seki Y, Kasai M, et al. Neuroinflammation in schizophrenia especially focused on the role of microglia. Prog Neuropsychopharmacol Biol Psychiatry 2013;42:115–21 [published online ahead of print December 13, 2011]. Feinberg I. Schizophrenia: caused by a fault in programmed synaptic elimination during adolescence? J Psychiatr Res 1982;17:319–34. Escande-Beillard N, Washburn L, Zekzer D, Wu ZP, Eitan S, Ivkovic S, et al. Neurons preferentially respond to self MHC class I allele products regardless of peptide presented. J Immunol 2010;184:816–23 [published online ahead of print December 16, 2009]. Feinberg I, Campbell IG. Longitudinal sleep EEG trajectories indicate complex patterns patterns of adolescent brain maturation. Am J Physiol Regul Integr Comp Physiol 2012;304:R296–303 [published online ahead of print November 28, 2012]. Mignot E, Lin L, Rogers W, Honda Y, Qiu X, Liu X, et al. Complex HLA-DR and DQ interactions confer risk of narcolepsy–cataplexy in three ethnic groups. Am J Hum Genet 2001;68:686–99. Han F, Lin L, Li J, Dong SX, An P, Zhao L, et al. HLA-DQ association and allele competition in Chinese narcolepsy. Tissue Antigen 2012;80:328–35. Ribstein M. Hypnagogic hallucination. In: Guilleminault C, Passouant P, Dement WC, editors. Narcolepsy. New York (NY): John Wiley and Sons; 1976. p. 145–60. Fortyun HA, Lappenschaar GA, Nienhuis FJ, Furer JW, Hodiamont PP, Rijnders CA, et al. Psychotic symptoms in narcolepsy: phenomenology and a comparison with schizophrenia. Gen Hosp Psychiatr 2009;31:146–54 [published online ahead of print January 24, 2009]. Douglass AB, Hays P, Pazderka F, Russell JM. Florid refractory schizophrenias that turn out to be treatable variants of HLA-associated narcolepsy. J Nerv Ment Dis 1991;179:12–7. Peyron C, Tighe DK, van den Pol AN, de Lecea L, Heller HC, Sutcliffe JG, et al. Neurons containing hypocretin (orexin) project to multiple neuronal systems. J Neurosci 1998;18:9996–10015. Pelin Z, Guilleminault C, Risch N, Grumet FC, Mignot E. HLA-DQB1⁄0602 homozygosity increases relative risk for narcolepsy but not disease severity in two ethnic groups. US Modafinil in Narcolepsy Multicenter Study Group. Tissue Antigen 1998;51:96–100. Nimgaonkar VL, Ruder WA, Zhang XR, Tsoi WF, Trucco M, Saha S. Further evidence for an association between schizophrenia and the HLA DQB1 gene locus. Schizophr Res 1995;18:43–9.

Please cite this article in press as: Huang Y-S et al. Narcolepsy–cataplexy and schizophrenia in adolescents. Sleep Med (2013), http://dx.doi.org/10.1016/ j.sleep.2013.09.018