STRENGTHS AND LIMITATIONS OF THIS STUDY

Sleep profile of patients with septo-optic-pituitary dysplasia (SOD) will be assessed considering different clinical and structural outcome’s contributors.

Importantly, sleep features of patients with SOD will be compared with subjects with agenesis of corpus callosum and with subjects with visual deficit.

A limitation of the study is the absence of a nocturnal polysomnography as objective sleep evaluation.

Introduction

Sleep disturbances occur in a great number of children with neurodevelopmental disorders.1 Sleep disturbances often tend to become chronic and multiple, and are unlikely to resolve without interventions. Different neurodevelopmental abnormalities may affect the timing and features of sleep, interfering with ascending arousal system signalling of neurotransmitters that control sleep-wake cycle. Sleep-wake is controlled by homeostatic and circadian processes, where the circadian cycle is influenced by many regulators such as cortisol, melatonin, sleep efficiency, temperature and clock genes.2 The development of this rhythm could be disrupted by structural abnormalities involving midline structures such as the suprachiasmatic nucleus or pineal gland, and the optic pathway.

Septo-optic-pituitary dysplasia (SOD), is a condition characterised by multiple structural brain abnormalities that may be associated with a broad phenotypic spectrum of neurodevelopmental disorders. SOD is traditionally defined as the combination of a classic clinical-radiological triad including midline brain defects, hypoplasia of the optic nerves and/or chiasm and hypothalamic-pituitary dysfunction.3 The typical midline malformation is represented by the absence or disruption of septum pellucidum, a thin transparent membrane located in the brain between the body and anterior horns of the lateral ventricles. Other recurrent midline brain abnormalities are thinning or agenesis of corpus callosum and structural abnormalities of the hypothalamic-pituitary axis. Optic nerve hypoplasia generally results in visual impairment, although a minority of patients may exhibit normal or near normal visual acuity values,4 and may be associated with a range of comorbidities including developmental delay/intellectual disability, seizures and autism spectrum disorder.5

Children with SOD may experience a variety of sleep difficulties.6 To date, only a limited number of studies have examined sleep characteristics in patients with SOD, primarily focusing on individual cases or small case series.7 8 Among the most reported findings are sleep fragmentation, disrupted circadian rhythms and reduced sleep efficiency. All of the structural and functional defects of SOD are potential mechanisms accounting for the lack of circadian rhythm. Both midline brain abnormalities and reduced visual input may contribute to alter melatonin secretion. However, a specific melatonin profile in patients with SOD has not been identified, and the single factors contributing to sleep disturbances in SOD have not been thoroughly investigated.

A range of sleep abnormalities has been reported in patients with isolated agenesis of corpus callosum (ACC) alone, including greater sleep onset delay, less sleep duration, greater bedtime resistance, sleep anxiety, night awakenings, parasomnias, sleep-disordered breathing and daytime sleepiness.6 Interestingly, polysomnographic studies of individuals with ACC have found increased slow-wave sleep but reduced spindles-slow waves coupling, decreased Rapid Eye Movement (REM) sleep, and decreased interhemispheric coherence.9–12

Since light plays a crucial role in circadian entrainment, individuals with reduced or absent light perception, such as those with severe visual impairment or blindness, are at an increased risk of experiencing sleep difficulties.6 The absence of vision is often associated with both increases in reported sleep disturbances and incidence of free-running circadian rhythms.13 Individuals who are completely blind experience greater sleep disturbances compared with those with residual vision.13 An alteration of the sleep architecture and its microstructure was sometimes reported in blind adults, as well as spindles development.14

This study aims to evaluate sleep quality, sleep characteristics and melatonin profiles in paediatric patients diagnosed with SOD and compare these findings with children who have isolated visual impairment (ie, related to disorders involving pre-geniculate structures) and those with agenesis of the corpus callosum. Considering the heterogeneity of SOD features and phenotype, investigating possible determinants and major structural contributors to sleep outcomes could shed light on sleep development and potentially on sleep control strategies.

Materials and methods

This is an observational study. Participants will be enrolled from the Child Neuropsychiatric Unit of the IRCCS Mondino Foundation (Pavia, Italy). The multidisciplinary team has more than 10 years of experience in the clinical management of patients with neurodevelopmental disorders, cerebral malformations and bilateral visual impairment. The estimated duration of the study is 2 years.

Participants

Three groups of participants will be enrolled in the study, with participants screened consecutively according to the relevant inclusion and exclusion criteria as outlined below.

Patients with SOD (group A)

Inclusion criteria are: (1) clinical diagnosis of SOD with or without a defined genetic diagnosis, according to current diagnostic criteria, in the presence of optic nerve involvement; (2) absence of malformation of cortical development; (3) age 3–18 years; (4) best corrected grating or visual acuity ≥0.5logMAR (5) availability of at least two serial sleep electroencephalograms (EEGs) performed during clinical follow-up; (6) absence of epilepsy; (7) stable drug therapy, if present, during the previous 3 months.

Patients with visual impairment (group B)

Inclusion criteria are: (1) diagnosis of congenital or early acquired isolated bilateral Visual Impairment due to the involvement of the so-called ‘anterior’ visual pathway (ie, the pre-geniculate component of the retino-geniculate pathway) with or without a known genetic diagnosis (eg, isolated eye and/or optic nerve maldevelopment, inherited retinal dystrophies); (2) age 3–18 years; (3) best corrected grating or visual acuity ≥0.5logMAR without significant variations at the previous clinical follow-up; (4) availability of at least two serial sleep EEGs performed during clinical follow-up; (5) absence of epilepsy; (6) stable drug therapy, if present, during the previous 3 months.

Patients with agenesis of corpus callosum (group C)

Inclusion criteria are: (1) presence of isolated corpus callosum complete agenesis on brain MRI; (2) age 3–18 years; (3) best corrected grating or visual acuity ≥0.5logMAR; (4) availability of at least two serial sleep EEGs performed during clinical follow-up; (5) absence of epilepsy; (6) stable drug therapy, if present, during the previous 3 months.

In all three groups exclusion criteria are: (1) absence of informed consent; (2) presence of severe developmental delay, intellectual disability and/or severe motor impairment; (3) melatonin consumption; (4) presence of cortical visual impairment. In group B, patients are also excluded for the presence of a cerebral malformation/lesion.

Patient and public involvement

Patients and family associations will be acknowledged. We will share the results of this study in the community to highlight the importance of sleep outcome and influence on the overall disease.

Outcome measuresPrimary outcome

The main aim of this study is to describe sleep features (circadian rhythm, sleep quality and efficiency) of patients with SOD compared with patients with an isolated bilateral visual impairment and patients with corpus callosum agenesis.

Secondary outcomes

Secondary outcomes include: (1) investigating whether structural and clinical features of SOD syndrome (such as visual acuity, epileptic abnormalities on EEG, midline abnormalities, presence and localisation of other brain malformations) might influence sleep outcome; (2) description of melatonin profile in patients with SOD compared with those with visual impairment and those with ACC; (3) comparison of sleep EEG patterns between patients diagnosed with SOD and those in the other two groups.

A sleep EEG recording will be scheduled at the time of study inclusion and previously performed EEG will be reviewed. EEG assessment will include the analysis of background activity, sleep macro and microstructure with the evaluation of presence/absence/distribution of neurophysiological elements, and eventually interictal discharges.

Data collection and planned evaluations

Study data will be collected in case report forms (CRFs) designed ad hoc and entered into a dedicated database. Preliminary patient framing involves review of the medical history (clinical and possibly genetic diagnosis, brain MRI findings, remote and recent history, physiological history) according to available clinical records. At the time of study inclusion, the referring clinician will perform sleep screening, evaluation of comorbidities and neurological examination. Importantly, patients’ history of recurrent otitis media, snoring, tonsils enlargement and any other clue for conditions that might impair sleep will be ruled out.

Sleep quality will be assessed through the completion of standardised sleep questionnaires by patients and their caregivers, as well as participation in an interview to examine sleep habits and hypnic profiles. Sleep questionnaires employed will include: (1) Pittsburgh Sleep Quality Index (PSQI) (a self-completed questionnaire assessing sleep quality over a 1-month time interval); (2) Epworth Sleepiness Scale (ESS) (a self-completed questionnaire assessing daytime sleepiness); (3) children sleep habits questionnaire assessing any nature of sleep disturbance.

Sleep quality and sleep-wake cycle will be evaluated, based on child’s inability to go to bed, delay in falling asleep, sleep duration, overnight awakenings, anxiety related to sleep, parasomnia, respiratory disorders and daytime sleepiness.

Patients will be asked to wear an actigraph (Philips Respironics Actiwatch Spectrum) on the non-dominant hand for 7 to 14 days at T0 and for the same interval at T1 served as a quantitative measure of movement to test sleep efficiency (the ratio of total sleep time to sleep period), total nighttime sleep duration, rest activity and number of awakenings. During that period, patients/caregivers will be asked to compile a sleep diary.

Melatonin levels

For each subject, one 4 mL blood and one 2 mL saliva sample will be collected simultaneously during each assessment, always in the morning. Patients will be instructed to abstain from eating for at least 30 min before collecting saliva. Blood samples will undergo centrifugation within 12 hours (1500 g for 15 min), and the resulting serum will be stored at −80°C until measurement. Saliva will be collected via passive drooling, aspirated with a syringe, and transferred into 2 mL polypropylene tubes. Melatonin will be measured in both serum and saliva samples using a validated in-house liquid chromatography-mass spectrometry method.

Ethics and dissemination

This study has been approved by the local Ethics Committee (N°0049187/23). Written informed consent will be obtained from caregivers and participants above 14 years old. Parents will be exhaustively informed about the study during a counselling session.

The study findings will be shared through publication in an international peer-reviewed journal and presented at both national and international conferences.

Sample size and statistical methods

The three groups will be compared using three different parameters, namely the PSQI score, the ESS score and the mean activity score detected by the actigraph.

We estimated the sample size for conducting a MANOVA test with three outcome variables (PSQI score, ESS and mean activity score of the actigraph) across three groups. A sample of 45 participants (15 per group) would allow to detect an effect size of 0.16 (measured by Pillai’s trace) with a type I error rate of 0.05 and a power of 0.8.

The study will analyse three groups based on their PSQI score values, ESS questionnaire results and mean activity score values from actigraph data through a MANOVA test. If significance (threshold at 0.05) is detected, post-hoc comparisons will be conducted with appropriate corrections for significance level, such as Bonferroni adjustments.

Categorical variables will be presented as counts and percentages, compared across groups via χ2 or Fisher’s exact test. For continuous variables, we will report mean and SD or median and IQR if appropriate, with between-group comparisons made using analysis of variance for normally distributed variables or Kruskal-Wallis test. If significance is detected, post-hoc comparisons will be conducted through t test or Dunn test with Bonferroni adjustments.

Multivariate analysis, if necessary, will involve regression models. A p value of 0.05 or lower will denote statistical significance.

The statistical analysis will be carried out by the BioData Science Unit at the Mondino Foundation (Pavia, Italy), using the R Statistical Software (V.4.2.0; R Core Team 2022).

Discussion

The aims of this study are to describe the sleep features of SOD patients and investigate the potential role of the core clinical features of SOD spectrum in determining the sleep outcome. Indeed, sleep problems are frequent in this population and represent a substantial cause of distress to family. Moreover, these disturbances can exacerbate disease symptoms, cognitive and behavioural issues, and even seizures when they occur. Neurodevelopmental disorders often manifest with alterations in sleep that might result from both clinical and structural features. The mechanisms underlying neurodevelopmental impairments in children with SOD are complex and involve multiple factors, which may also extend to sleep patterns. In a recent review, Mann et al suggest that the prevalence of neurodevelopmental impairments in children within the SOD spectrum might be considerable.15 The high heterogeneity among the samples in relation to their neuroanatomical and clinical features prevented an assessment of their specific contributions to the phenotype. However, the authors suggest to consider standardised assessment of neurodevelopmental impairments alongside routine care, underlining the need for future research aimed at identifying the causal mechanisms.

Previous research showed that patients with SOD with abnormal sleep-wake rhythmicity presented a higher prevalence of developmental delay compared with those with normal rhythmicity (100% vs 15%, respectively). Additionally, those with sleep-wake rhythm disturbances showed a greater incidence of corpus callosum hypoplasia or absence (66%) compared with those with normal sleep-wake rhythmicity (30%). Similarly, Garcia-Filion et al 16 found that corpus callosum hypoplasia, but not absent septum pellucidum or pituitary gland malformations, was associated with increased risk of developmental delay.16 For this purpose, Alt et al and Signorini et al reported a high proportion of normal cognitive development in their participants diagnosed as SOD-plus (additional cortical malformations) but no data are available concerning sleep features in SOD-plus condition compared with SOD.5 17 Cognitive-developmental impairment has been the primary focus in research assessing children with SOD spectrum conditions, with the majority of studies assessing intellectual disability and developmental delay, rather than behavioural, emotional and sleep outcomes.17

Evidence on sleep profiles of patients with SOD is scarce. Rivkees8 described a young child with SOD who showed arrhythmicity in sleep patterns with random sleep distributed throughout the day and night. Webb and colleagues7 characterised sleep-wake cycles in a cohort of six children with SOD, analysing actigraphy and a 24 hours melatonin profile. All six children were found to have sleep fragmentation with poor sleep efficiency due to frequent and prolonged night awakenings. Moreover, melatonin profiles showed a substantial variation in the timing and amount of melatonin produced, although no consistent pattern among all children could be found.7

Given the known importance of light in circadian entrainment, visually deprived individuals may have an increased risk of sleep difficulties also due to abnormal melatonin secretion. Children with early-onset visual impairment may present altered sensorimotor processing, a developmental competence related to sleep spindles, possibly reflecting a disrupted sleep pattern.18 Moreover, a different pattern of maturation of cortical activity in visually impaired toddlers and children compared with typically developing peers has been reported.12 Such differences involved the development of sleep patterns as well and may be related to the atypical development of competencies such as sensorimotor process frequently.19 ,14 Aubin and colleagues examined 30-day actigraphy recordings in 11 blind individuals compared with sighted individuals. Although they did not find group differences, they did find greater variability in sleep efficiency and time of sleep onset in blind individuals. Further analyses demonstrated abnormal timing of melatonin consistent with an abnormal circadian rhythm, but preservation of cortisol secretion profile.12

Ingram and Churchill performed a survey of 66 children with ACC and found that overall 78% had clinically significant sleep problems. Sleep problems were significantly correlated with overall quality of life regardless of age.12 Considering the poor sleep quality of patients with ACC, corpus callosum involvement and role in sleep outcome should be considered when studying sleep features in patients with SOD.

Among the strengths of this study, it represents the first attempt to comprehensively investigate sleep outcomes in patients with SOD, and at identifying different structural and clinical factors that influence sleep features in the heterogeneous spectrum of SOD phenotype. Moreover, the study includes both objective and subjective sleep outcomes, combining the use of actigraphy, sleep EEG and standardised sleep questionnaires. The study findings may guide the clinician in describing the sleep profile detectable in SOD patients and could potentially facilitate targeted intervention that should improve the quality of life of patients and their caregivers.

The present study has several limitations. First, the estimated number of participants is small, however, in the context of rare diseases, modest sample sizes are common. Another limitation of the study is the absence of nocturnal polysomnography as an objective sleep evaluation, which will not be undertaken in this exploratory study. Moreover, the single detection of melatonin could be framed as a limitation of the study but a continuous serial monitoring would have necessarily implied hospitalisation of the patient, thus also possibly negatively influencing the sleep quality.

Ethics statementsPatient consent for publication

Consent obtained from parent(s)/guardian(s).

Acknowledgments

The authors thank Marco Fasce and Grazia Papalia and all the patients and their families who will participate in the study.