Alexithymia, impulsiveness, emotion, and eating dyscontrol: similarities and differences between narcolepsy type 1 and type 2

Non-sleep symptoms, as depression, anxiety and overweight, are often encountered in narcoleptic patients. The purposes of this study are to evaluate mood, impulsiveness, emotion, alexithymia, and eating behavior in patients with narcolepsy type 1 (NT1) and narcolepsy type 2 (NT2) compared to healthy controls and to investigate possible correlations between clinical-demographic data, polysomnographic parameters, and subjective questionnaires. Consecutive patients affected by NT1 and NT2 underwent to Patient Health Questionnaire-9, Generalized Anxiety Disorder-7 Scale, Barratt Impulsivity Scale-11, Difficulties in Emotion Regulation Scale, Toronto Alexithymia Scale, and Eating Disorder Evaluation Questionnaire. Daytime sleepiness was assessed using Epworth sleepiness score. Data were compared with controls. Fourteen NT1, 10 NT2, and 24 healthy subjects were enrolled. Toronto Alexithymia Scale total score was significantly higher in NT1 than NT2. Compared to controls, NT1 patients exhibited significantly higher scores at Patient Health Questionnaire-9 and Difficulties in Emotion Regulation Scale. A positive correlation between hypnagogic hallucinations and Difficulties in emotion regulation was found. NT1 and NT2 share several psycho-emotional aspects, but whereas NT1 patients exhibit more depressive mood and emotion dysregulation compared to controls, alexithymic symptoms are more prominent in NT1 than NT2. Hypnagogic hallucinations, emotion dysregulation, and alexithymia appear to be correlated, supporting the hypothesis of mutual interaction of the above areas in narcolepsy.


Introduction
Narcolepsy is a chronic debilitating sleep disorder characterized by excessive daytime sleepiness (EDS), sleep paralysis, hypnagogic hallucinations, and fragmented nocturnal sleep. The latest edition of the International Classification of Sleep Disorders (ICSD-3) has classified narcolepsy into narcolepsy type 1 (NT1) and narcolepsy type 2 (NT2) on the basis of levels of hypocretin, also known as orexin, in cerebrospinal fluid (CSF) [1]. Cataplexy is a sudden loss of muscle tone during wakefulness that is evoked by strong, mainly positive, emotions [1]. Patients with NT1, previously reported as narcolepsy with cataplexy, have low levels of hypocretin-1 (HCRT-1) in the CSF, whereas patients with NT2 do not exhibit cataplexy and display normal levels of HCRT-1; however, few patients with narcolepsy and cataplexy with physiological levels of HCRT-1 on CSF have been reported [1]. In addition to sleep/wake-related symptoms, non-sleep disturbances, i.e., depression, anxiety and overweight, are often encountered; the reasons of such relationships possibly stem from the role of hypocretin in several cerebral networks but also from the impact of daytime sleepiness on quality of life. The hypocretins are neuropeptides produced by lateral hypothalamus neurons which widely project to the olfactory bulb, cerebral cortex, thalamus, hypothalamus, brainstem, and limbic system [2,3]. Therefore, hypocretins play a well-documented role in a variety of physiological processes, such as arousal and the maintenance of wakefulness [4,5], feeding behavior and energy 1 3 metabolism [6], neuroendocrine and autonomic functions [7,8], and emotion regulation and reward processing [9].
Narcolepsy is associated with a high comorbidity of both medical conditions and psychiatric disorders [10][11][12]. Several studies have reported increased depressive disorders and greater incidence of anxiety disorders in narcolepsy than controls, even in young patients [10,12,13], without differences between NT1 and NT2 [11,14]. A possible genetic link between narcolepsy and depression or others common causal factors still remains speculative [11].
Bayard and coll. observed that NT1 patients opt for choices with higher immediate reward regardless of higher future punishment showing a significant lack of perseverance at the urgency, premeditation, perseverance, sensation seeking (UPPS) Impulsive Behavior scale, compared to healthy subjects [15], whereas data on NT2 are still scarce. An abnormal amygdala function using functional brain magnetic resonance imaging (MRI) has been found in patients with NT1, which may result in abnormal emotional processing under both pleasant and unpleasant conditions [16].
Some authors reported an association of alexithymia in several sleep disorders, including narcolepsy [17]. It has also been assumed that this personality trait may play a role in dream alterations [18] and some authors consider alexithymia a risk factor for somatic symptoms and eating disorders [19]. However, the link between alexithymia and narcolepsy has not yet been well established.
Finally, several studies have shown an increase of body mass index (BMI) and eating concern both in NT1 and NT2 patients, compared with the general population [20], and a high prevalence of eating disorders have been reported in NT1 [21].
Although there is evidence of association between overweight and emotion dysregulation, including alexithymia, in general population [19], their mutual relationships in narcoleptic patients have not been studied so far. Therefore, we performed, for the first time, a comprehensive evaluation of depression, anxiety, impulsiveness, alexithymia, emotion, and eating dysregulation, in narcoleptic patients, comparing NT1 and NT2. Previous studies explored some of these features individually in narcolepsy, but they have not analyzed all these aspects together.
The purposes of this study are (i) to evaluate mood, impulsiveness, emotion, alexithymia, and eating behavior, through self-reported questionnaires, in NT1 and NT2 patients in comparison with healthy controls, and (ii) to investigate possible correlations between clinical and demographic data, polysomnographic findings, excessive daytime sleepiness (EDS), mood, impulsiveness, emotion, alexithymia, and eating behavior in narcoleptic patients.

Participants
NT1 and NT2 patients, diagnosed according to the ICSD-3 criteria, were consecutively enrolled at the Sleep Center of Policlinico Tor Vergata from May 2018 until September 2019. Both treated and untreated narcoleptic patients were included. Age, sex, BMI, disease duration, and medical and psychiatric comorbidities of all participants were collected.
All patients underwent to 24-h Polysomnographic continuous recording (PSG), Multiple Sleep Latency Test (MSLT), and brain MRI. Patients affected by Obstructive Sleep Apnea Syndrome (AHI > 5/h) were excluded. Healthy subjects comparable for age and sex were recruited. Controls were screened by a neurologist with expertise in sleep medicine and subjects with Pittsburg Sleep Quality Index (PSQI) > 5 and/or Epworth Sleepiness Scale (ESS) > 10 were excluded [22]. All subjects gave their written informed consent to the procedures. The study protocol was approved by the Independent Ethical Committee of the University Hospital of Rome "Tor Vergata."

Questionnaires
All patients and controls were investigated by means of the following self-administered questionnaires: Patient Health Questionnaire-9 (PHQ-9), Generalized Anxiety Disorder-7 Scale (GAD-7), Barratt Impulsivity Scale-11 (BIS-11), Difficulties in Emotion Regulation Scale (DERS), and Toronto Alexithymia Scale (TAS-20). To assess the presence of obesity and eating disorders we used BMI measurement and Eating Disorder Evaluation Questionnaire 6th edition (EDE-Q). Subjective daytime sleepiness was assessed by ESS [22].
The Patient Health Questionnaire-9 (PHQ-9) is a brief self-administered nine-item depression-specific questionnaire whose diagnostic validity has been established [23]. The PHQ-9 score ranges from 0 to 27, since each of the 9 items can be scored from 0 (not at all) to 3 (nearly every day). PHQ-9 scores 5 to 9, 10 to 14, 15 to 19, and 20 to 27, correspond to mild, moderate, moderately severe and severe depression, respectively [24].
The Generalized Anxiety Disorder Scale (GAD-7) is a seven-item practical self-report anxiety questionnaire that proved valid in primary care and has good operating characteristics for detecting generalized anxiety. The optimal cut point is ≥ 10. Cut points of 5, 10, and 15 might be interpreted as representing mild, moderate, and severe levels of anxiety [25].
The Barratt Impulsivity Scale-11 (BIS-11) is a selfadministered thirty-item questionnaire for measuring impulsiveness trait [26]. The questionnaire comprises six first-order constructs (attention, motor impulsiveness, self-control, cognitive complexity, perseverance, cognitive instability) which in turn form the three second-order factors: Attentional Impulsiveness (IA) refers to the inability to focus on the task at hand, Motor Impulsiveness (IM) involves acting without thinking, and Non-Planning Impulsiveness (NPI) reflects a lack of future orientation. Higher scores indicate higher levels of impulsiveness, with a maximum total score of 120 [26].
The Difficulties in Emotion Regulation Scale (DERS) is a thirty-six-item self-reported questionnaire. The sum yields a global DERS score ranging from 36 to 180 with higher scores reflecting greater difficulties in regulating emotion across each domain [27]. The questionnaire comprises of six subscales assessing different dimensions of emotion dysregulation. These subscales include the following: (1) non-acceptance of emotion (e.g., "When I'm upset, I become angry with myself for feeling that way"); (2) Goals or difficulties engaging in goal-directed behaviors (e.g., "When I'm upset, I have difficulty concentrating"); (3) Impulse control difficulties (e.g., "I experience my emotions as overwhelming and out of control"); (4) Strategies, as limited access to effective regulatory strategies (e.g., "When I'm upset, I believe that there is nothing I can do to make myself feel better"); (5) Clarity as reduced emotional clarity (e.g., "I am confused about how I feel"); and (6) Awareness or lack of emotional awareness (e.g., "I pay attention to how I feel" that needs to be coded using a reverse score). A validated Italian version was used [28].
The Toronto Alexithymia Scale (TAS-20) is a selfreported questionnaire that comprised twenty items and three factors: (a) difficulty in identifying feelings and differences between feelings and bodily sensations; (b) difficulty in describing feelings; and (c) externally oriented thinking. According to empirically determined cut-off points, people who obtain a score of 61 or more in TAS-20 are identified as being patients with alexithymia [29]. A validated Italian version of TAS-20 was administered [30].
The Eating Disorder Evaluation Questionnaire 6th edition (EDE-Q) is a self-report version of the Eating Disorder Examination [31], consisting of thirty six items assessing features of eating disorders and generating four subscale scores: dietary restraint, eating concern, weight concern, and shape concern. Respondents rate the items on a seven-point scale, indicating the number of days out of previous twenty eight in which specific behaviors and attitudes occurred. Higher scores reflect greater degree of severity. The EDE-Q has received psychometric support, including adequate test-retest reliability and a good convergence with the EDE interview [31].

Statistical analyses
Descriptive statistics are reported as absolute numbers with mean ± standard deviation (SD). Mean item scores were compared between groups using the Student's t test for normally distributed variables. When the normality of the distribution was rejected, as assessed by Kolmogorov-Smirnov test, a non-parametric Kruskal-Wallis test was used for comparisons of median values among three groups, followed by post hoc testing using unpaired Mann-Whitney U tests with Bonferroni correction. Statistical significance was set at a p < 0.05; after Bonferroni's correction the statistical significance was set at p < 0.016. To compare treated and untreated narcoleptic patients, Mann-Whitney U test was applied and statistical significance was set at a p < 0.05. A non-parametric correlation analysis was conducted through Spearman's Rho correlation coefficient to assess the associations between the questionnaire scores and PSG and MSLT parameters in the whole sample of narcoleptic patients (NT1 and NT2). To reduce the risk of a Type 1 error, a Bonferroni's correction indicated a p value of 0.008 as acceptable (p value of 0.05 divided by the multiple comparisons of PHQ-9; GAD-7, BIS-11, DERS, TAS-20, EDE-Q). Statistical analyses were performed with SPSS Statistics version 20.0.

Demographic and clinical data
Twenty-five patients agreed to participate in the study. One patient was excluded for incomplete questionnaires.

Depression and anxiety
Patients affected by NT1 showed a significantly higher PHQ-9 global score than controls (p = 0.001). Ten out of 24 patients exhibited a PHQ-9 score > 10 (7 NT1 and 2 NT2) and one NT2 patient showed PHQ-9 score > 20 indicating moderate and severe degree of depression, respectively.
Mean GAD-7 global score was slightly higher in NT1 and NT2 patients than controls, without reaching statistical significance. GAD-7 score was > 10 (moderate anxiety) in one NT1 patient and > 15 (severe anxiety) in three out of 24 narcoleptic subjects (1 NT1 and 2 NT2).
No statistical differences were found in the comparison between NT1 and NT2 at PHQ-9 and GAD-7 questionnaires. Data are reported in Table 2.

Impulsivity
BIS-11 total score and the subscales scores did not differ among NT1, NT2, and healthy controls ( Table 2).

Emotion dysregulation
DERS global score was significantly higher in NT1 patients compared to controls (p = 0.009) ( Table 2).
Compared to healthy subjects, the scores at "Goals" and "Strategies" sub-items were higher in NT1 patients and "Clarity" sub-item was higher in NT2 patients, but without reaching the statistical significance after Bonferroni's correction (Table 2). Alexithymia TAS-20 global score was significantly higher in NT1 than in NT2 (p = 0.005) (Table2). Furthermore, the "difficulty in identifying feelings" (DIF) and the "difficulty in describing feeling" (DDF) subscale scores were higher in NT1 than in NT2 patients (p = 0.017 and p = 0.004), without reaching the statistical significance, after Bonferroni's correction as to TAS-DIF (Table 2).

Eating disorders
EDE-Q total score did not differ between NT1, NT2, and controls.
Although "dietary restrain" score was higher in NT2 than controls (p = 0.018) and "eating concern," "shape concern," and "weight concern" sub-items scores were higher in NT1 than in healthy subjects, these differences lost the statistical significance after Bonferroni's correction (Table2). No statistical differences emerged from the comparison between NT1 and NT2 ( Table 2).

Correlations between clinical, demographic, PSG data, and subjective questionnaires in narcoleptic patients
No correlation among age, age at narcolepsy onset, disease duration and PSG data, and Questionnaires scores were found in the whole sample of narcoleptic patients (NT1 and NT2). Correlations between PSG data and subjective questionnaires, as ESS and PHQ-9 correlation, lost statistical significance after Bonferroni correction.
All data are reported in Table 3.
Correlations between questionnaires total scores and their subscales were excluded.
All data are reported in Table 4.

Discussion
The peculiarity of our study is the comprehensive evaluation of the psycho-emotional profile investigating the mutual relationships between mood, emotion, eating behavior, alexithymic trait, impulsiveness, daytime somnolence, and PSG parameters in a limited but representative sample of NT1 and NT2 patients compared to healthy control group. An association between emotion dysregulation and sleep disorder has been described and difficulties in modulating emotional experiences, evaluated through DERS questionnaire, have been observed in patients suffering from insomnia [32]. For the first time the DERS questionnaire was administered in patients with narcolepsy.
We found a statistically higher DERS global score in NT1 patients than controls. The impairment in emotion processing and the possible occurrence of different emotional networks in NT1 are still debated [33][34][35]. The hypothalamus, including the hypocretin region in close anatomical connection with the amygdala, is crucial in human emotional processing [5]; therefore, it is expected that the dysfunction of hypocretin system, which is pathognomonic of NT1, may impair the above circuit. Considering the crucial role of emotions in cataplexy, several authors have hypothesized that NT1 could experience an emotional constriction to avoid cataplexy attacks [33,34]. Furthermore, in line with this hypothesis, in our sample NT1 patients exhibited higher difficulty in identifying and describing feelings at alexithymia scale than NT2. On the matter, findings of previous studies are controversial. It was reported that emotional judgment ability, evaluated by a classical emotional processing task (i.e., facial affect recognition) and explicit emotion regulation strategies detected through a self-reported questionnaire (Emotional Regulation Questionnaire), is comparable in NT1, NT2, and controls [35]. Moreover, since REM sleep and dreaming play a role in the cross-night regulation of negative emotions, it has been hypothesized that alexithymia may be related to dream alterations in NT1 [36]. In this regard, alexithymia is positively related with nightmare distress in other sleep disturbances [37,38].
We found a positive correlation between hypnagogic hallucinations and DERS total score, while the same correlation with TAS-DIF (p = 0.02) lost the statistical significance after Bonferroni's correction.
Our findings, for the first time, support the hypothesis that narcoleptic patients show an impairment in emotional processing, evaluated by means of DERS, and that emotion dysregulation is correlated with the presence of hypnagogic hallucinations.
In a previous study, an association between nightmare distress and the DIF component of alexithymia was described, both for patients with sleep disorders and for healthy subjects, and it has been assumed that both nightmare distress and DIF scores reflect a deficit in the processes of regulating emotions [17].
Emotion dysregulation may lead alexithymic individuals to exhibit excessive emotional outbursts when they are awake [17,37]. As above explained, dreaming and REM sleep play role in the cross-night regulation of negative emotions and the consolidation of emotional memories. In our study the positive correlation between hypnagogic hallucinations (associated with nightmares and REM sleep) and DERS supports the hypothesis of a mutual interaction between REM alteration and dysregulation of diurnal emotions, in narcoleptic patients.
Narcoleptic subjects, in particular NT1, exhibit higher score than controls in the depression-specific questionnaire (PHQ-9) without statistical differences between NT1 and NT2, in line with previous studies [11]. Nevertheless, no differences in anxiety-specific questionnaire (GAD-7) score between narcoleptic patients and controls were found in our population, in contrast with other studies describing Eating concern, SC Shape concern, WC Weight concern). r Spearman's rho correlation coefficient * p < 0.05, in light gray, denotes weak correlation ** p < 0.01, in dark gray, denotes strong correlation Bold values denote statistical significance after Bonferroni's correction (p value < 0.008)

Table 4
Spearman's rank correlation between subjective questionnaires in narcoleptic patients type 1 and type 2 augmented anxiety disorder in narcolepsy [10,14]. However, our findings are not fully comparable to previous data since different methods were utilized and because the distinction between NT1 and NT2 was not always performed in above-mentioned studies. We found a positive correlation between PHQ-9 total score and GAD-7, PHQ-9 and DERS total score, and PHQ-9 and TAS-DIF and also between GAD-7 and DERS total score. DERS total score and TAS-DIF are also positively associated with each other.
These findings support the hypothesis of mutual interaction between depression, anxiety, emotion dysregulation, and alexithymia in narcoleptic patients.
Regarding eating dyscontrol, in animal models the dysfunction in reward circuits may contribute to eating disorders (anorexia, bulimia, and binge eating) [39] and some authors observed in NT1 patients higher scores on the Eating Attitude Test [20] and various types of eating disorders, including Eating Disorder Not Otherwise Specified, with increased weight concern and binge eating [9,20]. In addition, impulsivity may be a characteristic of people who exhibit aspects of abnormal eating and also alexithymia has been associated to eating behavior [19,40]. While some authors have observed in NT1 higher eating, shape, and weight concerns at EDE-Q compared with BMI-matched controls [21], in our sample no differences between NT1 and NT2 subjects in eating, shape, weight concerns, and BMI were detected. We did not found a positive association between eating concern and impulsivity questionnaires, probably for the small sample size.
Finally, we did not found any differences in the comparison between treated and drug-naïve patients. We cannot fully explain these findings, but we assume that the similar ESS score in the two groups may partially justify this data. Moreover, it can be assumed that drugs used in narcolepsy have no effect on non-sleep symptoms.
Several limitations in our study need to be pointed out: Firstly, the small sample size. Secondly, CSF HCRT-1 measurements were available in a minority of cases; thus, possible correlations between scores at self-reported questionnaires and HCRT-1 levels were not assessed. Thirdly, treated and drug-naïve narcoleptic patients were included, but the recruitment of both groups may increase the generalizability of the results to narcoleptic patients in real-life conditions. Further studies, with large cohorts of patients and exploring possible effects of drug treatments in NT1 and NT2 patients, are needed.
In conclusion our findings suggest that (i) NT1 exhibit higher depressive disorders and emotional dysregulation compared to healthy subjects; (ii) NT1 and NT2 patients share several similarities mainly regarding depressive symptoms, anxiety, impulsiveness, and emotion dysregulation; (iii) compared to NT2, NT1 patients exhibit higher Eating concern, SC Shape concern, WC Weight concern); r Spearman's rho correlation coefficient * p < 0.05 in light gray, denotes weak correlation ** p < 0.01 in dark gray, denotes strong correlation Bold values denote statistical significance after Bonferroni's correction (p value < 0.008) 1 3 difficulties in identifying and describing feelings maybe related to cataplexy; iv) hypnagogic hallucinations are positively correlated with emotion dysregulation; and (v) depression, anxiety, emotion dysregulation, and alexithymia appear to be correlated, supporting the hypothesis of mutual interaction of the above areas in narcolepsy.
Funding Open access funding provided by Università degli Studi di Roma Tor Vergata within the CRUI-CARE Agreement.

Declarations
Conflict of interest On behalf of all authors, the corresponding author states that there is no conflict of interest.

Ethical approval
The study protocol was approved by the Independent Ethical Committee of the University Hospital of Rome "Tor Vergata." All subjects gave their written informed consent to the procedures.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.