Visual function in children with Joubert syndrome

To describe visual function in children with Joubert syndrome and to investigate its possible association with diagnostic and developmental aspects.

evident by the clinical severity of pathologies caused by its alteration. 4Like most ciliopathies, Joubert syndrome is mainly characterized by autosomal recessive inheritance.Exceptions are pathogenic variants in OFD1, with X-linked inheritance, 5 and heterozygous loss-of-function variants in SUFU, which are responsible for a neurodevelopmental syndrome at the milder end of the Joubert syndrome spectrum, with dominant inheritance and reduced penetrance. 6he most frequently mutated genes in Joubert syndrome are AHI1, CC2D2A, CEP290, CPLANE1, KIAA0586, MKS1, RPGRIP1L, and TMEM67. 7Typical clinical features in early life include hypotonia, abnormal respiratory pattern, abnormal eye movements (including ocular motor apraxia [OMA], an impairment of controlled, voluntary, and purposeful eye movement usually compensated by a thrusting movement of the head allowing change of fixation 8 ), ataxia, and developmental delay. 9The genetic and clinical spectrum of conditions presenting with the MTS has progressively widened over recent decades. 1 Variable associated features include fibrocystic kidneys, congenital liver fibrosis, and polydactyly. 9Ophthalmological involvement is frequent, variable, and depends to some extent on the involved gene. 10,11 cular manifestations may present either as maldevelopment (e.g.coloboma) or as degeneration (e.g.retinal dystrophy). 7The most frequently reported ophthalmological signs of Joubert syndrome are retinal dystrophy (20-30%), 10 retinal pigmentation abnormalities, and eye maldevelopment conditions such as colobomas, variably involving the optic disc and/ or choroid and occasionally coexisting with retinal degeneration. 7Other ophthalmological signs include nystagmus, strabismus, and ptosis. 7These characteristics may be associated with various degrees of visual impairment, which in turn may be related not only to ocular but also to visual system involvement.OMA is a key feature of Joubert syndrome, reported in up to 80% of patients, and the condition with the highest number of associated genes, AHI1 and SUFU being the most frequently reported. 6,7 unctional disorders may involve any of the three components of visual function (i.e. the abilities to see, to look, and to understand), which plays a pivotal role in a child's development and global functioning. 12Furthermore, along with multisystemic and ophthalmological involvement, a key finding in Joubert syndrome is developmental delay/intellectual disability, with neuropsychological profiles mainly characterized by deficits in processing speed and a relative sparing of verbal comprehension. 13Psychiatric and behavioural disorders are significantly more common in Joubert syndrome than in the general population as well. 13Neurodevelopmental disorders are often associated with or aggravated by visual impairment, as widely reported and discussed. 14,15  our knowledge, few studies have focused on the functional aspects of vision in children and adolescents with Joubert syndrome (see, for example, Brooks et al. 10 ), despite the incidence of both OMA and visual deficit related to retinal dystrophy in such condition.In this study, our aim was to describe the visual function in a large paediatric cohort of patients with Joubert syndrome and to investigate possible associations between visual function and diagnostic (neuroimaging and genetics) and developmental aspects.This would represent a first attempt to create a link between biological and global functioning in children with Joubert syndrome.

Participants
This was a retrospective cross-sectional work based on the review of the medical charts of a cohort of 59 children (33 males) consecutively referred to the developmental neuroophthalmology unit of a tertiary referral hospital for neurological conditions (IRCCS Mondino Foundation, Pavia, Italy) from 1st January 2002 to 31st December 2020 at their latest follow-up admission.We included all patients aged younger than 18 years who had received a diagnosis of Joubert syndrome based on neuroradiological confirmation of the MTS.Additional supportive clinical criteria were the presence of hypotonia, developmental delay, abnormal eye movements, and other cerebellar signs.Organ involvement was assessed by clinical examination and diagnostic tests according to published recommendations, 11 including fundus oculi evaluation, visual evoked potentials, electroretinogram (ERG), abdominal ultrasound, and routine blood tests to exclude involvement of kidney or liver conditions, brainstem auditory evoked potentials, and heart ultrasound.
Written informed consent was provided by the participants' legal guardian/next of kin according to the guidelines of the Declaration of Helsinki.

Evaluation protocol
The evaluation protocol for all the children included ophthalmological tests (anterior segment, fundus oculi, cycloplegic refraction, visual evoked potentials, ERG), basic visual functions (visual acuity, contrast sensitivity, fixation, smooth pursuit, saccades, extrinsic ocular movements, and visual axis alignment), and developmental/cognitive and behavioural assessment.A multidisciplinary team of trained professionals with experience in the diagnosis and rehabilitation of visual disorders (child neuropsychiatrists, therapists, psychologists, ophthalmologists, and orthoptists) performed and reviewed the evaluations, which were recorded.Basic visual functions were assessed and graded according to the Visual Function Score, a standardized tool derived from our centre's experience as well as from clinical practice guidelines (see, for example, Osborne et al. 16 and Vinuela-Navarro 17 ), and previously presented by Signorini et al. 12 See Figure S1 for a graphic representation of the multidisciplinary assessment.Best-corrected visual acuity was measured, when possible, according to the age and cooperation of the children, with a Snellen optotype or LEA Vision Test, 18 both in monocular and binocular vision and both for near (40 cm) and for far (3 m) distance.When possible, according to the age and clinical picture, a crowded test (multiple symbols) was used (in 31 children); otherwise, a single symbol test was adopted (in three children).When best-corrected visual acuity was impossible to assess because of the individual's age, collaboration, and complexity of clinical picture, bestcorrected grating acuity was evaluated. 19Contrast sensitivity is the ability to detect an image's photometric contrast and spatial frequency; it was evaluated with a Lea Low Contrast Symbols Test 20 or Hiding Heidi 21 on the basis of the child's age and level of cooperation.Fixation is the ability to maintain fixation on a stationary target.Smooth pursuit is the ability to track a slow-moving object both on a horizontal and on a vertical arc, maintaining it on the fovea.Smooth pursuit was considered altered if discontinuous or fragmented and based on a sequence of saccades.Saccades are rapid and voluntary re-fixation eye movements to bring the target into the foveal region.They may be impaired in terms of accuracy (i.e.dysmetria), velocity, or latency.The child was instructed to shift gaze between two visual targets both horizontally and vertically; for younger children, the target was quickly introduced into their visual field.Saccade initiation failure associated with compensatory signs (jerky head movements or thrusts in reaching the target) was considered a sign of OMA.Extrinsic ocular motility was assessed monocularly (ductions) and binocularly (versions) with the head still; the child was asked to follow a target in eight gaze positions.For visual axis alignment, corneal light reflex and cover tests were used to detect strabismus.
Developmental and cognitive assessment relied on standardized tests, according to the age and cooperation of the children.Standardized developmental or cognitive evaluation was possible for 45 children (76.27%).Developmental evaluation was based on the Reynell-Zinkin Scales for visually impaired children 22 or Griffiths Mental Development Scale. 23Cognitive tests included Raven Progressive Matrices, 24 Leiter-R, 25 and the Wechsler Intelligence Scales, 26 chosen on the basis of the age and verbal abilities of the child.The frequency of test application in the assessment is summarized in Table S1.Clinical observation and parents' interviews focused on the possible presence of behavioural problems, including autistic-like behaviours such as stereotypical movements.

Data analysis and statistics
Data were analysed by using the free software R version 4.1.2(Free Software Foundation, Boston, MA, USA).Variables are described in terms of raw counts and percentages.To evaluate whether clinical features, including visual function parameters, were associated with the development of neuropsychological skills under cognitive, visuocognitive, and learning abilities perspectives, we conducted χ 2 tests for independence for each neuropsychological variable.
We applied a permuted Bonferroni correction for nonparametric data for post hoc tests performed in case of significant effects to adjust the p-value of multiple comparisons (significant value: p < 0.05).

Clinical features
The most relevant clinical characteristics of the cohort are detailed in Tables 1, S2, and S3.Ages ranged between 4 months and 23 years (median age 8 years, mean age 9 years 2 months, standard deviation 6 years 3 months).The cohort included eight couples of siblings.Excluding these children, 10 out of 59 patients (16.95%) had a family history of neuropsychiatric problems (epilepsy, intellectual disability, and neurodevelopmental disorder), and 8 out of 59 (13.56%) of ophthalmological conditions (congenital cataract, nystagmus, strabismus, inherited retinal dystrophy, and severe refractive errors).Global developmental delay was reported by caregivers and observed by professionals in 44 out of 59 (75%) patients.
OMA was the most frequent sign at disease onset, reported in 75% of children and variably associated with other signs of cerebellar dysfunction such as hypotonia, developmental delay, and ataxia.Other frequent neurological abnormalities were brisk deep tendon reflexes, electroencephalographic abnormalities, and altered respiratory patterns (short alternating episodes of apnoea and tachypnoea or episodic tachypnoea alone).Associated clinical features included ocular signs such as retinal dystrophy (defined according to current criteria 27,28 by poor vision, abnormal eye movements, poor pupil responses, fundus abnormalities involving macular, retinal vessels, and retinal pigmentation, abnormal ERG), renal involvement (congenital malformations, altered biochemical parameters, or overt nephronophthisis), liver abnormalities (elevated blood liver enzymes or liver fibrosis), endocrine signs (mainly short stature), cardiac malformations, altered brainstem auditory evoked potentials, and swallowing abnormalities.Facial dysmorphisms included prominent forehead, ptosis, prognathia, epicanthus, lower lip eversion with trapezoid-shaped mouth, tongue protrusion, and enophthalmos.Brain magnetic resonance imaging revealed posterior fossa abnormalities other than MTS in 26 out of 59 patients (44%), and more complex malformative pictures with supra-tentorial involvement in 11 out of 59 (18.6%): corpus callosum abnormalities (five children), polymicrogyria (two), signs related to perinatal distress (two), impaired myelination (one), and periventricular heterotopia (one).For genetic studies on genomic DNA, next-generation sequencing was performed either on a targeted panel of ciliary genes or on the whole exome, as previously reported. 2,29 rty-six patients were found to carry homozygous or compound heterozygous pathogenic variants in genes associated with Joubert syndrome.In 13 patients, a genetic diagnosis could not be reached.

Fundus oculi
More than half the children (59%) had a completely normal fundus oculi examination, while 23 (39%) presented an altered retinal pigmentation (i.e.bone spicule pigmentation or degeneration of retinal pigment epithelium), associated in most cases with other fundus oculi Note: Note that all 59 children (100%) presented with neurological signs while multisystemic involvement was variable.

Anterior segment
The anterior segment was normal in 53 children (89.8%).
Others presented with lens opacity or cataract, lens coloboma, or corneal leukoma.

Refractive errors
Refractive errors (i.e.myopia and hyperopia, defined as spherical, an equivalent of >0.50 diopters in absolute value) were reported in 45 children (76.3%), the most frequent being hyperopia.

Electrophysiological examination
Half of the children (30; 50.8%) presented altered visual evoked potentials.ERG was altered in 24 children (40.6%), with extinguished pattern in nine (15.3%).Some children with funduscopic anomalies relatable to retinal dystrophy were not testable with electrophysiological examination owing to lack of cooperation.

Visual function
Results are reported in Tables 2, S2, and S3.

Cognitive and behavioural profiles
On the basis of clinical observation or standardized testing, 69.5% of children presented with severe developmental delay or intellectual disability.Their severity prevented a standardized test being performed in 14, so evaluation relied on clinical observation and caregivers' interview on adaptive abilities.Stereotypical behaviours (motor or verbal) were observed in 25.4% of patients, and behavioural problems in 18.6%.

Associations between visual function and other characteristics
We report the most relevant results emerging from the investigation of associations between ophthalmological characteristics/visual function and other diagnostic and clinical features.Data supporting the results are reported in Table S4.
Association between ophthalmological findings/visual function and neuroradiological aspects (involvement of central nervous system structures other than the cerebellum): retinal dystrophy and refractive errors seemed to be more frequent with isolated MTS, with a significance level just below the cut-off (p = 0.09 for retinal dystrophy and p = 0.059 for refractive errors).
Ophthalmological findings/visual function and genetic findings: we investigated the associations between the most represented genes in our cohort (i.e.AHI1, CC2D2A, CEP290, CPLANE1, RPGRIP1L) and ophthalmological findings/visual function.Retinal dystrophy was significantly more represented in populations with AHI1 and CEP290 (p < 0.001, post hoc p = 0.029).Consistently, roving movements and poor pupil responses seemed to be more frequently associated with CEP290 (p = 0.02 and p = 0.01 respectively; post hoc p = 0.039 and p = 0.03 respectively).AHI1 was more frequently associated than other genes with vessel thinning (post hoc p = 0.025 compared with CC2D2A, p = 0.016 compared with CPLANE1, p = 0.04 compared with RPGRIP1L).
Ophthalmological findings/visual function and cognitive/developmental aspects: a significant association between retinal dystrophy and developmental delay/intellectual disability was found (p = 0.047, post hoc p = 0.052).Furthermore, reduced best-corrected visual acuity both for near and for far distances was more frequent in the group of children with developmental delay/intellectual disability (p = 0.014 and p = 0.017 respectively); the absence of stereopsis was also significantly associated with developmental delay/intellectual disability.Finally, the presence of refractive errors and impaired fixation showed a borderline significance level of association with developmental delay/ intellectual disability (p = 0.076 and p = 0.09 respectively).

DISCUS SION
The main aim of this work was to describe the clinical profile of a cohort of children diagnosed with Joubert syndrome, with a focus on visual function.Our question arose from the diffusely reported visual involvement in Joubert syndrome (including both perceptual and oculomotor deficits) and from the demonstrated association between visual impairment and developmental frailties. 7,30 urthermore, we investigated possible associations between ophthalmological findings/visual function and genetic, neuroradiological, and cognitive aspects, to explore possible links between the involvement of genes/structures and functional impairment.Despite being exploratory and inconclusive, our results may plant the seed for a deeper reflection on the existing relations between genotype and phenotype, and between visual function and developmental aspects, in Joubert syndrome.In fact, although previous works have studied ophthalmological 10 and electrophysiological 31  as well as the association between oculomotor impairment (i.e.involving fixation, smooth pursuit, and saccades) and neurodevelopmental disorders. 15Gathering information on visual function, and not only on organ involvement, would be essential both for an integrated knowledge of a child's developmental profile and for planning more specific and individualized rehabilitation activities to promote autonomy, quality of life, and social participation.The considerable interplay between visual function (i.e.how the eye functions) and functional vision (i.e.how the person functions) requires a comprehensive and multidisciplinary approach in the evaluation and training of visual disorders. 32isual function profiles in our cohort were dominated by oculomotor impairment, as expected owing to cerebellar, and specifically vermian, involvement. 9,33 uch an impairment can range from a slight discontinuity in smooth pursuit to overt OMA.Unsurprisingly, OMA was the most reported presentation sign in our cohort, together with developmental delay and cerebellar signs. 9The first clinical sign of oculomotor impairment in children is typically a head tilt with preferential gaze position, a compensatory posture related to difficulty in saccade initiation. 34This feature was present in the quasi-totality of our cohort and can be considered an early and sensitive marker of Joubert syndrome.Fixation was impaired in half of our cohort, while all children presented relevant alterations of smooth pursuit, which in most cases was discontinuous and fragmented by abnormal eye movements such as nystagmus.Saccadic alterations and overt OMA were frequent.As expected, great interindividual variability (from slight alterations such as hypohypermetria to an extremely difficult elicitation) was observed and the horizontal plane was mainly involved.Almost one-third of children presented a remarkable limitation of ductions and/or strabismus, confirming that these represent common, although not typical, signs of Joubert syndrome. 7The described oculomotor dysfunctions are not specific for Joubert syndrome; however, in association with other signs (e.g.hypotonia and developmental delay), they may guide the diagnostic workup and address a child to magnetic resonance imaging with focus on the posterior fossa.Oculomotor function in Joubert syndrome is frequently affected by nystagmus, which may be a presentation sign in newborn infants 7,35 and have variable semeiology (horizontal, vertical, torsional, pendular, or see-saw components can be present and frequently combined).Most children in our cohort presented with nystagmus, more frequently horizontal and cyclorotatory.On the basis of our findings and clinical experience, we suggest an accurate observation of nystagmus semeiology and attention to associated signs: in the absence of funduscopic or ERG abnormalities, isolated nystagmus would probably be of cerebellar origin.On the contrary, when associated roving movements and poor pupil responses are present, a screening for retinal dystrophy would be highly recommended.Oculomotor function and nystagmus were evaluated for this work using clinical tools. 12Quantitative studies with eye movement recording systems would provide more reliable and reproducible data both for clinical and for research purposes.Unfortunately, these technologies are rarely available in the clinical setting and require a level of cooperation that may be difficult to obtain for children with neuromotor and cognitive disorders. 36Nonetheless, our qualitative findings confirm data from a video-oculography-based study revealing significant clinical heterogeneity in oculomotor impairment in children with Joubert syndrome, consistent with anatomical structures involved in this condition. 37or the perceptual aspects, most children in our cohort presented with reduced visual or grating acuity.Ocular manifestations in Joubert syndrome, such as eye maldevelopment or retinal dysfunction, have a variable incidence in different reports. 7Refractive errors were highly represented (76%) in our cohort, particularly in association with retinal dystrophy, as expected. 38A screening and regular follow-up for refractive errors would be recommended in children with Joubert syndrome with or without retinal dystrophy, because they could benefit from adequate correction.Furthermore, half of our cohort of children presented with some degree of retinal involvement, as revealed by funduscopic examination and/ or ERG.Retinal dysfunction may be present in Joubert syndrome without ophthalmoscopic signs of retinal dystrophy, 31 as confirmed by our results.A possible discrepancy between clinical (i.e.funduscopic) and ERG findings in children with Joubert syndrome has been reported 10 and could be due to very subtle signs of retinal degeneration or predict a lateronset retinal dystrophy.Besides, ERG abnormalities in isolation could be related to a primary cilium dysfunction that alters photoreceptor maturation and precedes photoreceptor degeneration. 39On the basis of current knowledge, ERG monitoring over time would be suggested in all children with Joubert syndrome, regardless of the genetic diagnosis.Exploring possible associations between visual and genetic aspects, we found that retinal dystrophy is significantly more represented in patients with mutations of AHI1 and CEP290, as reported in the literature, 3,7 when limiting analyses to the most represented genes (i.e.AHI1, CC2D2A, CEP290, CPLANE1, RPGRIP1L).The extension of analyses to all the genes represented in our cohort failed to reach statistical significance, probably because of the genetic heterogeneity and the low frequency of mutations involving the same genes, which limits the possibility of statistical inference.Genetic heterogeneity in our cohort was in line with that in the literature, 3 while gene frequencies were not consistent with other previous reports (e.g.we found a relatively high frequency of pathogenic mutations in the TMEM67 gene 40,41 ).These data may be due to a referral bias because they were collected from a centre for neuro-ophthalmological conditions and need to be further investigated through multicentre studies and considering larger cohorts.
No significant association was found between visual function and complex central nervous system malformations (involving other structures than the posterior fossa).This would indicate that cerebellar involvement per se may explain perceptual visual impairment in Joubert syndrome beyond ocular conditions.In fact, children nystagmus and OMA may struggle to perform an effective foveation, with possible repercussion on the performance during visual acuity tests, especially when a crowded target is provided.Retinal dystrophy was more represented when MTS was the only malformation (with borderline significance level).Again, the significant prevalence of perceptual and oculomotor disorders in our cohort may have reflected an inclusion bias.Nevertheless, our results are in line with previous reports on ocular manifestations of Joubert syndrome and prove that such manifestations can be the most represented in the very first months of life, because cerebellar signs such as ataxia and dysmetria and developmental delay may become more evident later in development (e.g. with the increase of motor initiative).
Cognitive functioning evaluated with standardized tests was considerably variable in our cohort, ranging from severe developmental delay/intellectual disability (69.5%) to an IQ within the normal range (30.5% of children).Such a result is in line with a previous study 42 and supports the hypothesis that cognitive functioning in patients with Joubert syndrome may have been traditionally underestimated owing to problems in motor coordination and language, as well as emotional and behavioural issues. 13,43 hereby, cognitive evaluations in children with Joubert syndrome should be performed by trained neuropsychologists considering all the possible functional impairments described in the syndrome, including visual (oculomotor and perceptual) disorders.Perceptual deficits (specifically the presence of retinal dystrophy, low vision, and altered stereopsis) were significantly more represented in children with developmental delay/intellectual disability in our cohort.Less significant associations were found between refractive errors (that were corrected when present) and impaired fixation and developmental delay/intellectual disability.The inability to start and maintain proper target fixation (e.g. because of saccadic dysfunction or nystagmus) can affect visual performance, particularly in visual crowding conditions, and have a negative impact on cognitive and academic performance (especially reading). 44,45 ur work mainly focused on the first two components of visual function (i.e. to see and to look) because it was not possible to gather comparable data on visual cognitive aspects (owing to the wide range of age and severity of clinical picture).Nevertheless, considering the role of cerebellum in the development of visual cognitive abilities (especially visuospatial and visuomotor), further studies investigating such aspects in children with Joubert syndrome considering bigger and more homogeneous cohorts would offer relevant insights for clinical practice.
In conclusion, this is one of the first studies to explore the visual functional profile of children with Joubert syndrome, beyond the ophthalmological aspects.Our findings confirm the recurring coexistence of perceptual and oculomotor impairment in children with Joubert syndrome and highlight the impact of such impairments on overall and cognitive development.Consequently, a thorough multidisciplinary assessment and follow-up of visual function is highly recommended when Joubert syndrome is suspected, both for diagnostic purposes (e.g. for the early detection of signs of retinal degeneration or OMA) and for planning individualized rehabilitation tailored on the child's functional profile. 27,32 tudies are needed to unravel the role of visual function impairment in cognitive and adaptive development of children with Joubert syndrome.Romaniello et al. previously reported challenges and barriers in daily life of adults with Joubert syndrome.In line with this, our results support a multidisciplinary rehabilitation approach considering not only neuromotor and cognitive aspects in children with Joubert syndrome, as previously reported, 46,47 but also sustaining visual function and functional vision, to limit the known detrimental effects of visual impairment on development and to improve outcomes such as academic and social inclusion and quality of life.

AC K NOW L E D G M E N T S
The members of the Developmental Neuro-ophthalmology Study Group are as follows: Chiara Bertone, Walter Misefari, Mauro Antonini, Daria Paini, Eleonora Perotto, Antonella Luparia, Lucrezia Olivier, Elisa Ercolino, Anna Pichiecchio.We thank the children and families for their participation in our clinical and research activity.We also thank all the professionals who share everyday life with us and the Mariani Foundation for its support.Open access funding provided by BIBLIOSAN.

F U N DI NG I N FOR M AT ION
This work was primarily supported by the Italian Ministry of Health with the grant Ricerca Corrente 2022-2024.The grant RF-2019-12369368 "Ricerca Finalizzata" granted by the Italian Ministry of Health to Enza Maria Valente supported genetic testing.

F
I G U R E 1 (a) Sagittal and (c) parasagittal T1-weighted and axial inversion recovery.(b) Images of a patient with the 'molar tooth' sign.Note thickened and horizontal superior cerebellar peduncles (white arrow, c) and hypodysplasia of the cerebellar vermis (a,b).A widened fourth ventricle and abnormally deep interpeduncular fossa are also evident (c).
Neurological and systemic characteristics of the cohort.
T A B L E 1 14aracteristics of Joubert syndrome, aspects of visual function have been poorly investigated.Nevertheless, the possible negative impact of perceptual deficits on development has been well documented,14Visual function characteristics of the cohort.
3,11Saccadic exploration (explorative pattern organization) is the ability to explore faces, objects, and the environment in different tasks with the eyes, clinically evaluated in multidisciplinary teams through video recordings.Stereopsis is evaluated with Lang Stereotest.Peripheral visual localization areas are a clinical observation of peripheral visual awareness.Please note that some participants were assessed for both best-corrected visual acuity and best-corrected grating acuity. Note: