Exploring the emotional and behavioural profile in paediatric narcolepsy type 1: A case–control study
Giorgia Simoncini Malucelli and Anna Mercante have contributed equally to the manuscript.
Summary
Narcolepsy type 1 (NT1) is a central disorder of hypersomnolence often arising in childhood and adolescence. NT1 has a significant, but poorly defined, psychological impact. We aimed to investigate the psycho-social functioning of children and adolescents with NT1. We performed a cross-sectional, child and parent-reported questionnaire survey in 37 children and adolescents (6–17 years) with NT1, compared with age- and sex-matched controls. Questionnaires (SSHS, ESS-CHAD, CDI, MASC, CBCL, CRS-R, and SNAP-IV) evaluated various aspects of behavioural and emotional profiles, sleep habits, and daytime sleepiness. Subsequently, NT1 intra-group analysis was performed to investigate the effect of sex (males vs females) and pharmacological treatment (treated vs non-treated) on psychological features. The NT1 questionnaires total scores were then correlated with the clinical characteristics (age, body mass index [BMI], ESS-CHAD score, cerebrospinal hypocretin-1 [Hcrt-1] levels, and diagnostic delay). Patients with NT1 showed a higher tendency to depressive symptoms, anxiety, somatisation, inattention, hyperactivity, oppositional/defiant problems, and other maladaptive behaviours compared with controls. Among NT1 patients, females showed a higher propensity to anxiety, and non-treated patients displayed higher depressive symptoms. Psychological symptoms increased with age, BMI, and daytime sleepiness in patients with NT1, while a younger age was associated with more frequent somatisation symptoms. Lower cerebrospinal Hcrt-1 levels correlated with poorer social competencies, daily activities, and inattention. Diagnostic delay was associated with a higher impact of depressive symptoms and behavioural problems. NT1 in children and adolescents is associated with poorer functioning in multiple psychological domains calling for a multidisciplinary approach and monitoring to reduce disease burden and to prevent psychiatric consequences.
INTRODUCTION
Narcolepsy type 1 (NT1) is a rare chronic central disorder of hypersomnolence resulting from a hypothalamic hypocretin/orexin (Hcrt-1) deficiency of probable autoimmune origin (Partinen et al., 2014). In this condition, persistent excessive daytime sleepiness (EDS) is typically associated with fragmented nocturnal sleep and features of dissociated rapid eye movement (REM) sleep occurring at sleep onset/offset (sleep-related paralysis and hypnagogic hallucinations), during sleep (REM sleep behaviour disorder), or during wakefulness (cataplexy), the latter being the pathognomonic symptom (Kornum et al., 2017).
The prevalence of NT1 in the general population is estimated to be between 0.025% and 0.05% (Longstreth Jr. et al., 2007); it usually arises during childhood or adolescence, where it may present with a peculiar phenotype (Plazzi et al., 2018; Rocca et al., 2015; Wise, 1998), but it also occurs during adulthood (Krishnamurthy et al., 2014). It is a disease frequently underdiagnosed with an average diagnostic delay of 15 years (Thorpy & Krieger, 2014), which seems to be lesser (i.e., mean of 2 years) in recent children's case series (Plazzi et al., 2011; Ponziani et al., 2016).
Paediatric onset narcolepsy often impairs emotional development, social functioning, and life quality (Hovi et al., 2022; Rocca et al., 2016; Szakacs et al., 2019) and associates with hyperactivity, irritability, impulsivity, aggression, and emotional dysregulation (Plazzi et al., 2018; Rocca et al., 2016; Shelton & Malow, 2017), either being children-specific manifestations of EDS (Chervin et al., 2002) reflecting the attempt to resist sleep or expression of comorbidity such as attention deficit hyperactivity disorder (ADHD) (Carls et al., 2020; Hansen et al., 2020; Kim et al., 2020; Modestino & Winchester, 2013; Rocca et al., 2015). Hence, a wide range of neuropsychiatric disorders has been reported in children and adolescents with NT1 such as anxiety, depressive mood, and psychotic symptoms (Blackwell et al., 2017; Carls et al., 2020; Fortuyn et al., 2011; Hanin et al., 2021; Hovi et al., 2022; Kim et al., 2020; Modestino & Winchester, 2013; Parmar et al., 2019; Posar et al., 2014; Szakacs et al., 2015). Depressive symptoms often combine with difficulty in accomplishing good school and social results (Fortuyn et al., 2011; Morse & Sanjeev, 2018), with several performances being negatively affected by EDS. Patients and their parents also report memory/concentration impairment (Blackwell et al., 2017; Ingram et al., 2021). Furthermore, both endangerment of interpersonal relationships and social withdrawal (Morse & Sanjeev, 2018) have emerged, significantly contributing to the disease burden (Plazzi et al., 2018).
Based on this evidence, we aimed to define the psychological aspects of NT1 in childhood and adolescence compared with healthy sex- and age-matched controls, using a comprehensive protocol of standardised questionnaires and scales to investigate the related emotional and behavioural profile. We also aimed to evaluate the correlations between clinical features and psychological aspects in NT1.
METHODS
Study design
This study has a cross-sectional, case–control design. It was conducted between January and September 2019, during routine outpatient visits performed at the Center for Narcolepsy, IRCCS Istituto delle Scienze Neurologiche, Bologna (Italy). Participation in the study was free; informed consent to participate in the study was provided by parents, with assent provided by patients. The local medical ethical committee approved the study (protocol number: 17009).
Subjects
We prospectively included consecutive patients diagnosed with NT1 attending the Center for Narcolepsy for follow-up or first evaluations. The diagnosis was established according to the International Classification of Sleep Disorders-Third Edition (ICSD-3) criteria (American Academy of Sleep Medicine, 2014).
Inclusion criteria were the following: (1) diagnosis of NT1; (2) age < 18 years; (3) absence of known comorbid psychiatric-neurological disorders at clinical examination.
Demographical and clinical data collected included sex, age, height, and weight to calculate body mass index (BMI) and BMI z-score, Epworth Sleepiness Scale for Children and Adolescents (ESS-CHAD) (Wang et al., 2022) score, diagnostic delay (DD), and cerebrospinal Hcrt-1 (CSF Hcrt-1) level when available. Subjects with a BMI > 95th percentile were considered obese (Ponziani et al., 2016).
Patients with NT1 were matched for age and sex to healthy controls (HCs), each evaluated by a psychologist trained in sleep medicine to exclude concomitant sleep and/or psychiatric disorders, as well as reported learning difficulties. The children and adolescent controls were recruited among public school students. The matching criteria were the following: age (±1 year) and no significant difference in the distribution of males and females among the two samples (NT1 group: 21/37 males, HCs group: 18/37 males; χ2 = 0.488; p = 0.485).
Psychological assessment
The psychological assessment consisted of different questionnaires and scales administered to the subjects and their parents inquiring about symptoms of anxiety, depression, ADHD, and other affective and social-behavioural issues. Also, sleep habits and the presence of EDS were recorded. Younger children were assisted in the compilation by their caregivers.
- School Sleep Habits Survey (SSHS): a questionnaire including 49 items about sleep–wake habits during school days and weekends, information about sleep schedule regularity, school performance, daytime sleepiness, behaviour problems, and depressive mood. We utilised the modified Italian version of the SSHS developed by Wolfson and Carskadon (Wolfson & Carskadon, 1998). The following variables of the SSHS were analysed in the current work: sleep latency during weekdays and weekends, and total sleep time during weekdays and weekend.
- Epworth Sleepiness Scale for Children and Adolescents (ESS-CHAD): an 8-item self-administered questionnaire adapted to the paediatric age to measure daytime sleepiness and the usual chances of dozing off or falling asleep while engaged in different activities (Wang et al., 2022). An ESS score of 10 or more was considered to indicate EDS.
- Children Depression Inventory (CDI): a 27-item instrument aimed to estimate the cognitive, affective, and behavioural signs of depression, quantifying their severity in children and adolescents between the ages of 7 and 17 (Kovacs, 1985), its abnormal cut-off score is 16 or higher.
- Multidimensional Anxiety Scale for Children (MASC): a self-report questionnaire that indexes the range and severity of anxiety symptoms in children and adolescents (aged 7–17 years). It consists of 39 items distributed across four major factors: Physical Symptoms, Harm Avoidance, and Social Anxiety subdivided into two subfactors each (Tense/Restless – Somatic/Autonomic, Perfectionism – Anxious Coping, Humiliation/Rejection – Performance Fears) and Separation Anxiety. It also provides a Total Scale score and an Anxiety Disorder Index to identify youths prone to developing an anxiety disorder and to monitor treatment effects (March et al., 1997). The MASC defines t-scores >60 as being in the clinical range, and between 55 and 59 as being in the borderline range 55–59.
- Child Behaviour Checklist (CBCL) 6–18: is a two-section tool used to detect behavioural and emotional problems in the paediatric age (6–18 years). The first section provides ratings for competencies in various areas of the child's personal and social functioning as perceived by the parents (Activities, Social, and School performance), summed to provide a Total Competence score, its abnormal cut-off score is 40 or lower. The second contains 113 problem items that yield scores on Internalising (Anxious/Depressed, Withdrawn/Depressed, Somatic Complaints), Externalising (Rule-Breaking Behaviour, Aggressive Behaviour), and Total Problems (Internalising Problems, Externalising Problems, Social Problems, Thought Problems, Attention Problems), as well as scores on Diagnostic and Statistical Manual of mental disorders-IV (DSM-IV) related scales (Achenbach, 2001). Raw scores are converted to gender and age-standardised t-scores. The CBCL defines internalising, externalising, and total problems T-scores ≥65 as being in the clinical range, and between 60 and 65 as being in the borderline range. On the DSM-IV scales, T-scores >70 are considered clinical, and scores between 65 and 70 as borderline.
- Conners’ Rating Scales Revised (CRS-R): is an 80-item scale that evaluates behavioural symptoms of ADHD along with a broad spectrum of conditions that may occur in comorbidity in children and adolescents (age 3–17 years) (Conners, 1997). The parent rating scales include the following subscales: Oppositional, Cognitive Problems/Inattention, Hyperactivity, Anxiety–Shy, Perfectionism, Social Problems, Psychosomatic Concerns, ADHD index, Conner's Global Index, and DSM-IV Symptoms (Inattentive, Hyperactive–Impulsive). Raw scores are converted to gender and age-standardised T-scores. Clinical symptoms were defined with a T-score cut-off above 65, and borderline symptoms were defined with a T-score cut-off between 55 and 65.
- Swanson, Nolan, and Pelham Rating Scale (SNAP-IV, abbreviated version): it measures the core symptoms of ADHD, screening for inattention, hyperactivity/impulsivity, and oppositional defiant disorder (ODD) over a 26-item interview (Swanson et al., 2001). Parent SNAP-IV Rating Scale defines inattention, hyperactivity/impulsivity, and ODD with cut-off scores above 1.78, 1.44, and 1.88, respectively.
Statistical analyses
Data were explored with descriptive statistics for continuous (mean ± standard deviation) and categorical (N and percentage) values.
A Shapiro–Wilk test was performed to determine the data distribution for the continuous variables. Parametric data were analysed using unpaired t-tests, while non-parametric data underwent the Mann–Whitney U test. Categorical data were compared through the χ2 test.
First, we compared the clinical characteristics and the result of the psychological assessment in the NT1 population with the controls.
Subsequently, we performed an NT1 intra-group analysis sorting into sub-groups according to sex (males vs females), and current therapy (treated vs non-treated).
Finally, NT1 questionnaires’ total scores (CDI, MASC, CBCL, CRS-R, and SNAP-IV) were correlated with the clinical characteristics (Age, BMI, ESS-CHAD, CSF hcrt-1 levels, and DD) using Spearman's rank correlation coefficient (). We used Spearman's correlation coefficient because most of the data were not normally distributed. Furthermore, we did not use any “outlier” exclusion methods. For missing values, we used “exclude cases pairwise” (cases with missing values for one or both of a pair of variables for a correlation coefficient were excluded from the analysis).
For all the statistical tests, a significant difference was defined as a p value <0.05.
All data analyses were performed using the Statistical Package for Social Science (SPSS) version 25.0.
RESULTS
Demographic, sleep, and clinical features
A total of 37 patients with NT1 and also healthy controls were enrolled, respectively. The NT1 sample included 21 males (56.8%) and 16 females (43.2%), with a mean age of 12.2 years (range 6–17, SD ± 3.3). CSF Hcrt-1 measurements were available for 31/37 (83.8%; mean 22.4 pg/mL, SD ± 32.9). The HCs group included 18 males (48.6%) and 19 females (51.4%), with a mean age of 10.9 years (range 6–17, SD ± 2.8). No significant differences were observed between the NT1 and the HCs groups for age (p = 0.070) or sex (p = 0.485). Patients with NT1 showed a significantly higher ESS-CHAD score (NT1: 13.2 ± 4.1, HCs: 3.5 ± 2.8, p < 0.001), BMI (NT1: 23.9 ± 6.5, HCs: 20.6 ± 4.5, p = 0.032), BMI z-score (NT1: 1.24 ± 0.94, HCs: 0.59 ± 0.98, p = 0.020), and prevalence of obesity (NT1 group: 15/37, HCs group: 6/37, χ2 = 4.26, p = 0.039). Subjective nocturnal sleep latency (SL), evaluated with the SSHS, was lower in the NT1 group compared with HCs on weekdays (NT1: 6.4 ± 6.5 min, HCs: 14.3 ± 9.7 min, p < 0.001), detailed results of sleep questionnaires are reported in Table S1. Significant results of the comparison of clinical features between NT1 and HCs groups with p-values are shown in Figure 1. Fourteen NT1 subjects were on medication at the time of evaluation (12 with sodium oxybate, and two with modafinil).
Comparative analysis of emotional and behavioural aspects between NT1 and HCs group
Emotional and behavioural aspects were assessed on CDI and MASC scales for patients and controls, and from the parents’ CBCL, CRS-R, and SNAP-IV scales.
Descriptive statistics and the results of statistical comparison of the psychometric evaluation are reported in Table 1; notably, the percentages shown in the table and the text include subjects who reported both a clinical and a borderline score.
| Cut-off | NT1 (mean ± SD) | HC (mean ± SD) | P (Mann–Whitney) | NT1, n° (%) | HC, n° (%) | p (χ2) | |||
|---|---|---|---|---|---|---|---|---|---|
| CDI | CDI total | C: >19 | 7.1 ± 5.7 | 4.2 ± 3.9 | 0.015 | 3 (8.1) | 0 (0) | 0.077 | |
| B: 17–19 | |||||||||
| MASC | Physical symptoms total | C: >60 | 43.7 ± 7.3 | 40.9 ± 5.7 | 0.068 | 4 (10.8) | 0 (0) | 0.040 | |
| B: 55–59 | |||||||||
| Harm avoidance total | C: >60 | 38.0 ± 8.1 | 32.4 ± 9.1 | 0.006 | 1 (2.7) | 1 (2.7) | 1.000 | ||
| B: 55–59 | |||||||||
| Social anxiety total | C: >60 | 48.8 ± 8.6 | 43.7 ± 8.0 | 0.010 | 6 (16.2) | 4 (10.8) | 0.496 | ||
| B: 55–59 | |||||||||
| Separation/panic | C: >60 | 52.2 ± 9.6 | 48.5 ± 12.9 | 0.172 | 14 (37.8) | 12 (32.4) | 0.626 | ||
| B: 55–59 | |||||||||
| MASC total | C: >60 | 43.8 ± 8.1 | 38.6 ± 10.2 | 0.017 | 3 (8.1) | 2 (5.4) | 0.643 | ||
| B: 55–59 | |||||||||
| Anxiety disorder index | C: >60 | 40.8 ± 7.2 | 38.4 ± 10.3 | 0.258 | 2 (5.4) | 4 (10.8) | 0.394 | ||
| B: 55–59 | |||||||||
| CBCL | Total competence | C: < 35 | 35.5 ± 8.9 | 40.3 ± 8.7 | 0.021 | 27 (73.0) | 17 (45.9) | 0.018 | |
| B: 35–40 | |||||||||
| Syndromic scales | Internalising problems total | C: >65 | 63.7 ± 6.3 | 48.0 ± 10.3 | <0.001 | 24 (64.9) | 6 (16.2) | <0.001 | |
| B: 60–65 | |||||||||
| Externalising problems total | C: >65 | 55.6 ± 7.8 | 44.8 ± 8.2 | <0.001 | 7 (18.9) | 2 (5.4) | 0.075 | ||
| B: 60–65 | |||||||||
| Total problems | C: >65 | 60.8 ± 7.1 | 45.1 ± 8.5 | <0.001 | 17 (50.0) | 2 (5.4) | <0.001 | ||
| B: 60–65 | |||||||||
| DSM-oriented scales | Affective problems | C: >70 | 70.4 ± 6.5 | 52.7 ± 3.7 | <0.001 | 32 (86.5) | 1 (2.7) | <0.001 | |
| B: 65–70 | |||||||||
| Anxiety problems | C: >70 | 59.5 ± 6.7 | 55.4 ± 6.4 | 0.009 | 11 (29.7) | 4 (10.8) | 0.043 | ||
| B: 65–70 | |||||||||
| Somatic problems | C: >70 | 58.5 ± 6.1 | 53.4 ± 4.2 | <0.001 | 8 (21.6) | 0 (0.0) | 0.003 | ||
| B: 65–70 | |||||||||
| Attention deficit/hyperactivity problems | C: >70 | 54.5 ± 5.0 | 51.7 ± 3.1 | 0.006 | 2 (5.4) | 0 (0.0) | 0.152 | ||
| B: 65–70 | |||||||||
| Oppositional defiant problems | C: >70 | 58.1 ± 6.9 | 52.5 ± 3.9 | <0.001 | 6 (16.2) | 1 (2.7) | 0.047 | ||
| B: 65–70 | |||||||||
| Conduct problems | C: >70 | 54.2 ± 6.7 | 52.0 ± 3.9 | 0.096 | 3 (8.1) | 1 (2.7) | 0.304 | ||
| B: 65–70 | |||||||||
| CRS-R | Conners’ ADHD index | C: >65 | 53.9 ± 14.2 | 43.1 ± 8.0 | <0.001 | 10 (32.3) | 4 (10.8) | 0.029 | |
| B: 55–65 | |||||||||
| Conners’ global index | C: >65 | 50.4 ± 12.1 | 41.6 ± 5.5 | <0.001 | 6 (19.4) | 1 (2.7) | 0.024 | ||
| B: 55–65 | |||||||||
| DSM-IV symptoms | C: >65 | 49.5 ± 11.5 | 42.0 ± 4.8 | 0.001 | 11 (35.5) | 2 (5.4) | 0.002 | ||
| B: 55–65 | |||||||||
| SNAP-IV | Inattention | C: >1.78 | 0.5 ± 0.5 | 0.3 ± 0.3 | 0.017 | 1 (2.7) | 0 (0.0) | 0.314 | |
| Hyperactivity/impulsivity | C: >1.44 | 0.2 ± 0.3 | 0.2 ± 0.2 | 0.341 | 0 (0.0) | 0 (0.0) | _ | ||
| Oppositional/defiant | C: >1.88 | 0.8 ± 0.7 | 0.2 ± 0.3 | <0.001 | 3 (8.1) | 0 (0.0) | 0.077 | ||
- Note: Bold values were used to highlight the statistically significant results.
- Abbreviations: ADHD, attention deficit hyperactivity disorder; B, borderline cut-off; C, clinical cut-off; CBCL, child behaviour checklist; CDI, children's depression inventory; CRS-R, Conners’ rating scales revised; DSM, diagnostic and statistical manual of mental disorders; MASC, multidimensional anxiety scale for children; SNAP-IV, Swanson, Nolan, and Pelham Rating Scale.
Further details of these findings, including all tests’ subscales, can be found in Table S2.
NT1 patients and HCs questionnaires
At CDI, the patients NT1 with showed a higher score for depressive symptoms than the HCs. The MASC results for symptoms related to anxiety disorders were significantly higher in NT1 versus HCs as accounted for by the MASC Total, Harm Avoidance Total, and Social Anxiety Total subscales. In particular, the prevalence of subjects reporting values above the normal range was significantly higher in the NT1 group for the Physical Symptoms Total score. On the other hand, non-significant differences were observed in the Separation/Panic and Anxiety Disorder Index.
Parents questionnaires
According to the CBCL scale, patients with NT1 showed lower scores than HCs at the Total Competence subscale. The patients with NT1 also showed higher scores for Internalising, Externalising, and Total Problems. Abnormal scores were reached in a significantly higher number of patients with NT1 for the Internalising Problems Total and Total Problems. The DSM-Oriented Scales were also applied to the CBCL, and children with NT1 exhibited higher scores in all the following domains: Affective Problems, Anxiety Problems, Somatic Problems, Attention Deficit/Hyperactivity Problems, and Oppositional Defiant Problems. In detail, in the NT1 population, there was a significantly higher prevalence of subjects in the pathological range in the following subscales: Affective Problems, Anxiety Problems, Somatic Problems, and Oppositional Defiant Problems. A non-significant difference was observed for the subscale Conduct Problems.
At CRS-R, the parents of narcoleptic patients reported greater scores in the ADHD Index, Conners’ Global Index, and DSM-IV Symptoms subscales. Of these, a significantly higher prevalence of T-scores above the normal value was reached for the NT1 group compared with the HCs group in the following subscales: ADHD Index, Conners’ Global Index, and DSM-IV Symptoms.
The SNAP-IV displays significantly higher ratings for Inattention and Oppositional/Defiant in the NT1 group, while a non-significant difference was observed in the Hyperactivity/Impulsivity subscale.
NT1 intra-group analysis of clinical features and psychometric measures
NT1 sex and treatment
We did not observe significant differences between males and females for what concerns demographic and clinical characteristics. Female patients showed a higher MASC Total score compared with males (females = 47.1 ± 10.1; males = 41.4 ± 5.1, p = 0.041).
Patients under treatment showed a significantly lower value of BMI (treated = 21.2 ± 3.5; non-treated = 25.6 ± 7.4; p = 0.026); lower CSF Hcrt-1 levels (treated = 5.2 ± 7.6; non-treated = 31.9 ± 37.5; p = 0.011) and a lower total sleep time during the weekend (treated = 8.9 ± 0.9 h; non-treated = 10.1 ± 1.5 hours; p = 0.006). Regarding the psychological assessment, non-treated patients displayed a higher CDI Total score than the treated (treated = 4.3 ± 4.1; non-treated = 8.8 ± 6.0, p = 0.011). At CBCL, treated patients showed poorer scores in Total Competence than non-treated ones (mean Total Competence score treated = 31.3 ± 5.4; non-treated = 38.0 ± 9.8, p = 0.026). No other significant differences were observed in the subgroup analysis.
Significant results of the comparisons between the subgroups are shown in Figure 2. Other details of comparisons according to sex and therapy are reported in Table S3.
Correlations between clinical data and psychometric measures
The CDI total score showed a positive correlation with Age (ρ = 0.460, C.I. = 0.150 to 0.688, p = 0.004), BMI (ρ = 0.578, C.I. = 0.303 to 0.764, p < 0.001), and DD (ρ = 0.457, C.I. = 0.082 to 0.719, p = 0.017).
In CBCL, the Total Competence score directly correlated with CSF hcrt-1 levels (ρ = 0.676, C.I. = 0.414 to 0.835, p < 0.001), as well as Externalising and Total Problems ratings with BMI (ρ = 0.405, C.I. = 0.083 to 0.650, p = 0.013 and ρ = 0.411, C.I. = 0.068 to 0.667, p = 0.018, respectively). We also observed a positive correlation between the scores of (1) Attention Deficit/Hyperactivity Problems with Age (ρ = 0.431, C.I. = 0.115 to 0.668, p = 0.008), ESS-CHAD (ρ = 0.359, C.I. = 0.030 to 0.619, p = 0.029) and DD (ρ = 0.460, C.I. = 0.085 to 0.721, p = 0.016); (2) of Oppositional Defiant Problems with Age (ρ = 0.383, C.I. = 0.057 to 0.635, p = 0.019); and (3) of Affective Problems and Conducts Problems with BMI (ρ = 0.343, C.I. = 0.011 to 0.607, p = 0.038 and ρ = 0.386, C.I. = 0.061 to 0.637, p = 0.018, respectively). A negative correlation emerged between the Somatic Problems score and Age (ρ = −0.327, C.I. = −0.595 to −0.007, p = 0.049).
Considering the CRS-R, the ADHD Index was positively correlated with Age (ρ = 0.431, C.I. = 0.079 to 0.687, p = 0.016), BMI (ρ = 0.383, C.I. = 0.022 to 0.656, p = 0.034) and DD (ρ = 0.514, C.I. = 0.085 to 0.721, p = 0.006); the Conners’ Global Index with Age (ρ = 0.417, C.I. = 0.062 to 0.678, p = 0.020); the DSM-IV Symptoms score with Age (ρ = 0.583, C.I. = 0.278 to 0.781, p = 0.001), BMI (ρ = 0.373, C.I. = 0.010 to 0.649, p = 0.039), and DD (ρ = 0.523, C.I. = 0.167 to 0.758, p = 0.005).
Finally, the SNAP-IV Inattention score showed a negative correlation with CSF hcrt-1 levels (ρ = −0.454, C.I. = −0.702 to −0.108, p = 0.010).
The most relevant correlations are shown in Figure 3. All correlations with p-value are reported in Table S4.
DISCUSSION
Our results suggest more frequent symptoms related to internalising and externalising problems in children and adolescents with NT1, using a protocol of well-validated and reliable assessment measures, expanding the data already described in the literature (Dorris et al., 2008; Inocente et al., 2014; Nordstrand et al., 2019; Rocca et al., 2016; Shelton & Malow, 2017; Stores et al., 2006). A schematic representation of their emotional and behavioural profile and correlations is shown in Figure 4 and Figure 5.
Internalising problems
The CBCL and MASC questionnaires disclosed a greater occurrence of anxiety symptoms in the NT1 population with a tendency to avoidance and social withdrawal, possibly reflecting a dysfunctional strategy to prevent emotionally intense situations and, therefore, cataplectic episodes, as already suggested in previous studies in adults (Chen et al., 2020; Fortuyn et al., 2009).
Regarding mood disorder, a higher prevalence of affective problems emerged in our NT1 cohort, as suggested by the results of CBCL and CDI scales, reporting depressive symptoms and often aggressive conduct. The previous literature describing the comorbidity of narcolepsy and depression reported cumulative effects of adverse life events related to this disease (i.e., poor quality of life/social stigma), concurring to promote depression (Lee et al., 2017), and, in turn, isolation and difficulties of social integration. This hypothesis seems to be reinforced by the positive correlation with age, DD, and BMI, suggesting an eventual interaction between the disease and increased depressive symptoms over time. The positive impact that treatment itself appears to have on mood scales, may indicate that improving the control of NT1 symptoms contributes to reducing the occurrence of depressive manifestations.
Interestingly, the orexinergic/hypocretinergic system has been implied in anxiety and depressive disorders (Chen et al., 2015; Etkin & Wager, 2007; Phelps & LeDoux, 2005; Pizza et al., 2014) as orexin is involved in the regulation of emotional processing in the amygdala (Schwartz et al., 2008), possibly representing a physiological link between narcolepsy and anxiety/mood disorders (Chen et al., 2020). According to CBCL and CRS-R scales, somatic problems were also more frequent in narcoleptic patients. These complaints of physical symptoms (e.g., headaches, stomach aches) with no established medical basis, may represent secondary manifestations of anxiety or depression, as reported in the literature (Masi et al., 2000; Sackl-Pammer et al., 2018).
Externalising problems
The higher presence of ADHD traits has been confirmed by all the related tests performed on these areas (CBCL, CRS-R, SNAP-IV), though without reaching the pathological range. Previous studies suggested a significant association between NT1 and the peculiar aspects of ADHD (i.e., inattention, hyperactive–impulsive symptoms) (Lecendreux et al., 2015; Morse & Sanjeev, 2018; Shelton & Malow, 2017).
Similarly, a prevalent oppositional/defiant attitude was suggested by different tools (including CBCL, CRS-R, and SNAP-V-I) as already described for young subjects with NT1 by Rocca et al. (Rocca et al., 2016) and Szakacs et al. (Szakacs et al., 2015).
The CBCL, CRS-R, and SNAP-IV scales also displayed greater difficulty in attention processes, restlessness, irritability, and emotional unstableness. The hypersomnolence characterising NT1 negatively interferes with maintaining sustained attention. Likewise, restlessness and irritability may reflect the attempt of the child to cope with EDS through self-stimulatory behaviours. Importantly, ADHD traits in our sample correlated directly with age, BMI, and inversely with the hypocretin level, suggesting that older and obese patients may manifest worse ADHD symptoms while residual hypocretinergic transmission may have a protective role.
Other emotional and behavioural problems
Other areas of the child's personal and social functioning were impaired in our NT1 cohort. According to their caregiver's clinical scores in the CBCL-related scales, children and adolescents with NT1 showed low social and school performances, data also in line with previous evidences (Quaedackers et al., 2019; Rocca et al., 2016; Stores et al., 2006).
As described, intra-group analysis of patients with NT1 disclosed significant correlations between their psychological traits and clinical features. Female patients presented higher anxiety than males, also in accordance with the previous and general literature (Chen et al., 2020).
Pharmacological treatment was associated with minor depressive manifestations, suggesting that an effective anti-narcoleptic medication may mitigate the disease burden and eventually the related depressive symptoms (Mutti et al., 2022). On the other hand, treated patients showed worse adaptive functioning at the CBCL Total Competence subscales. The published literature on the subject is still inconclusive: Rocca et al. (2016) did not observe significant differences between treated and non-treated patients concerning the Competence subscale, while other authors described improved performances in treated patients, especially seen in academic spheres (Mansukhani & Kotagal, 2012). Conversely, our results seem to indicate that pharmacological treatment alone is insufficient to ameliorate the performances of narcoleptic patients, calling for further prospective approaches (Brunetti et al., 2023). Greater treatment efficacy occurs when behavioural and pharmacological treatments are combined (Janssens et al., 2020); this dual approach should therefore be considered in managing narcoleptic patients, particularly in cases of poor daily activity performance, and include behavioural training for psychological distress.
In this study, the burden of narcoleptic psychological symptoms appeared to be increased with age, BMI, and EDS. The latter result is in line with what has already been reported on EDS, corresponding to somatic, thought, attention problems, and ADHD (Lecendreux et al., 2015; Rocca et al., 2016). A high BMI has been associated with psychological problems in paediatric (Inocente et al., 2013) and adult narcoleptic patients (Barateau et al., 2020), as well as in the general population. Indeed, obesity may act in NT1 either as a discrete comorbidity as in the general population, or as a factor contributing to a more severe phenotype (Inocente et al., 2013). Our results confirm the association between obesity and NT1.
We also found that younger age is associated with more frequent reports of somatisation symptoms, as outlined by Masi et al. (2000), though a significant relationship between age and somatic symptom severity is only sometimes consistent (Sackl-Pammer et al., 2018).
Lower CSF Hcrt-1 levels correlated with poorer social competencies, daily activities, and inattention, pointing to a possible causative role of orexin deficiency as a biochemical factor also in attention problems. These data aligned with some of the preceding research on ADHD: patients presenting the inattentive subtype had lower Hcrt-1 levels in the CSF (Baykal et al., 2019). Unlike, Lecendreux et al. who did not observe any significant correlation between CSF Hcrt-1 levels and ADHD symptoms in patients with narcolepsy (Lecendreux et al., 2015). However, the CSF Hcrt-1 dosages in this study were available only for a minority of patients underpowering our statistical analysis.
Despite our study cohort receiving the diagnosis in a relatively short time (mean 23 months), we found that the longer the DD, the higher seemed to be the occurrence of depressive symptoms and behavioural problems. A missed diagnosis of hypersomnolence prevents rapid and proper management, leading to undertreatment and possible exacerbation of related issues. Narcolepsy symptoms are often misinterpreted by parents, teachers, and doctors and attributed to other factors (e.g., laziness, lack of interest, poor cognitive skills), easily resulting in impaired learning processes and feelings like a failure or a lack of self-esteem. This element appears extremely important since it represents a possible psychopathologic risk factor for the paediatric narcoleptic population, which should be accurately assessed.
Limits and strengths
We acknowledge that our study presents several limitations. First, the questionnaires filled out by the children/adolescents and their parents are not diagnostic tools, despite being validated instruments to assess emotional and behavioural characteristics and to monitor the clinical course and the response to therapy (Achenbach, 2001; Conners, 1997; Kovacs, 1985; March et al., 1997; Murali & Kotagal, 2006; Swanson et al., 2001; Wolfson & Carskadon, 1998). Additionally, it should be noted that these tools investigate symptoms common to various psychopathological conditions, which sometimes overlap with those typical of narcolepsy (e.g., the items investigating the hours of sleep or perceptual alteration in CBCL). Therefore, the presence of certain conditions may be misinterpreted: for example, the significant difference observed in our study between NT1 and healthy controls in the subscale Thought Problems (CBCL) may be attributed not to defined psychopathologic alterations (i.e., psychotic aspects) but instead to the nature of narcoleptic symptoms (e.g., hypnagogic and hypnopompic hallucinations). However, we used a set of convergent tools that showed consistent findings in the different psychological domains explored, thus limiting the possibility of bias related to a specific questionnaire. More, though patients with defined psychiatric and cognitive comorbidities were excluded, we did not further investigate Specific Learning Disorders, one of the determinants of neurocognitive and school performance.
Despite these limits, our research applied a broader protocol of well-validated and reliable assessment measures than previous studies, identifying multiple possible associations between NT1 clinical features and emotional/behavioural traits. Validating the results of different converging tools with others addressing the same emotional and behavioural aspects provides a further homogeneous and reliable psychological picture. Lastly, the cross-sectional design of the project limits causation interpretation.
Further research including correlation between subjective assessments and objective metrics in a larger population size will help to clarify the results we observed.
CONCLUSION
In this study, we evaluated the psychosocial functioning of a paediatric NT1 population and investigated the relationship with specific clinical aspects of the disease, matching our findings to previous literature. We identified a peculiar emotional and behavioural predisposition, with a major tendency to develop depressive symptoms, anxiety, ADHD, oppositional/defiant behaviours, and somatisation symptoms. We also found that age, BMI, DD, and EDS, as well as CSF Hcrt-1, may concur to differently modulate emotional and behavioural features, competencies, and attention performances.
Taken together, these findings support the hypothesis that NT1 may have repercussions on global performance and quality of life, resulting in an increased risk of developing psychopathological distress.
Based on those observations, we suggest including a psychological evaluation with a standardised test battery in the early clinical assessment of NT1 to provide a comprehensive diagnostic-therapeutic approach and, possibly, further improve the quality of the patient's life using patient-tailored psychological support.
Further studies are needed to define the NT1 cognitive and emotional profile and its causative factors in children, clarifying whether NT1 may constitute a psychiatric risk factor or mere comorbidity.
AUTHOR CONTRIBUTIONS
Giorgia Simoncini Malucelli: Conceptualization; investigation; writing – original draft. Anna Mercante: Conceptualization; investigation; writing – original draft. Fabio Pizza: Conceptualization; supervision; data curation; writing – review and editing. Valerio Brunetti: Supervision; data curation; conceptualization; writing – review and editing. Francesco Biscarini: Data curation. Stefano Vandi: Data curation. Alice Mazzoni: Data curation. Christian Franceschini: Supervision. Giacomo Della Marca: Formal analysis; data curation. Catello Vollono: Formal analysis; data curation. Daniela Pia Rosaria Chieffo: Resources. Giuseppe Plazzi: Conceptualization; supervision; project administration; data curation; writing – review and editing.
FUNDING INFORMATION
None.
CONFLICT OF INTEREST STATEMENT
All co-authors declare no financial interest or conflicts of interest to disclose deemed to this study.
Open Research
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request.
