Neurodevelopmental Impairments in Children With Septo-Optic Dysplasia Spectrum Conditions: A Systematic Review

Major Points and Findings:

This paper does not concern stem cells or any other treatment. It is a systematic review from King’s College London covering every study that formally measured learning, behaviour or autism in children with septo-optic dysplasia (SOD), SOD-plus or optic nerve hypoplasia (ONH). We include it because developmental and behavioural difficulties are often what families most hope a treatment will change. The review shows how common these difficulties are, what appears to drive them, and how unreliable the measuring tools are in a child who cannot see well.

The main figures are high. About half of the children had intellectual disability or developmental delay, and about a third met a threshold for autism. The authors describe these as rough estimates from small, mixed studies and say that the autism figure in particular may be inflated.

Aim:

To describe the type and prevalence of neurodevelopmental impairments and autism spectrum disorder (ASD) symptoms in children with SOD spectrum conditions.

Methods:

PRISMA guidelines were followed and the protocol was registered in advance on the Open Science Framework. PubMed, EMBASE and PsycInfo were searched from inception to 16 July 2022, in English, with no other filters, and reference lists were hand-searched.

The review included original peer-reviewed observational studies of children up to 18 years with SOD, SOD-plus or ONH, provided they used a formal, validated assessment of cognitive, developmental, behavioural, social or emotional function, or of ASD (for example the Social Communication Questionnaire, the Child Behaviour Checklist, or DSM criteria). Grey literature, non-English papers and studies without standardised measures were excluded.

Each record was screened independently by two reviewers, with a further reviewer settling disagreements. Two people extracted the data independently. No quality appraisal tool was applied, because the study designs varied too much. Findings were combined narratively and no meta-analysis was done.

Results:

The search found 2,132 unique records. Of these, 215 full texts were assessed and 20 studies were included, published between 1984 and 2018 and covering 479 children. Twelve studies described their patients as having ONH only and eight included SOD or SOD-plus. The reviewers point out that in eight of the twelve “ONH” studies, many children also had hormone deficiencies or brain abnormalities and so would meet today’s definition of SOD.

Learning and development (14 studies): 175 of 336 children (52%) had intellectual disability (IQ below 70) or developmental delay. Individual studies ranged from 0% (two reports with only one and two children) to 71%.

• Three studies consistently found more delay or intellectual disability in children with ONH in both eyes than in those with one eye affected.
• Blindness by itself probably does not explain the delay. In one population-based study the link between bilateral ONH and intellectual disability was the same whether vision was very poor or only mildly reduced.
• Among brain and hormone factors, corpus callosum hypoplasia was linked with delay, while an absent septum pellucidum and pituitary malformations were not. Hypothyroidism was linked with delay in every developmental domain, independent of the corpus callosum. Every child with a disturbed sleep-wake rhythm was delayed, against 15% of those with a normal rhythm. Brain stem abnormalities were linked with worse delay.
• Unexpectedly, in one small series of SOD-plus, three of seven children had normal cognition, two were borderline and two subnormal.

Behaviour, emotion and social function (5 studies): 88 of 184 children (48%) showed impairment. In one sample, sensory processing difficulties were reported in 34 of 41 (83%). One study removed the usual confounders. Its 11 children with isolated ONH, normal intelligence and no more than moderate visual loss still scored worse than matched controls on anxiety and depression, withdrawal, thought problems, attention and aggression, and 4 of the 11 (36%) scored in the clinical range. Their behaviour scores tracked with reduced white matter integrity in the ventral cingulum on diffusion imaging.

Autism (5 studies): 65 of 187 children (35%) had an ASD diagnosis or scored above a clinical cut-off, with individual studies ranging from 17% to 55%. ASD was somewhat more frequent with profound than with severe visual impairment (36% against 27%) and with bilateral than unilateral ONH (24% against 10%), but neither difference was statistically significant. In a study that applied four assessment methods, adapted for blindness, to seven children, all four agreed in only three.

Conclusions:

The authors conclude that neurodevelopmental impairment across the SOD spectrum may be as common as 48 to 52%, and clinically relevant autism symptoms may affect up to 35%. They recommend that formal developmental and ASD assessment become part of routine care, using tools adapted for visual impairment alongside specialist judgement, because a diagnosis gives access to therapy and to educational support. They also recommend repeat assessment over time. A child whose behaviours stem from blindness tends to develop adaptive social skills, whereas a child with ASD continues to show delays.

Limitations they state:

• The samples differed so much in brain, hormone and eye involvement that the separate contribution of each could not be tested statistically.
• Study designs and assessment tools varied widely, and the same scale was sometimes used to measure different things.
• There is no validated autism measure for children with visual impairment. Behaviours such as echolalia can be a blind child’s way of learning language, so ASD may have been over-counted.
• ONH and SOD were used interchangeably in older papers, and ONH often “becomes” SOD later in childhood when hormone deficiencies appear.

Background Information:

The paper opens with a short case: a 13-year-old girl with SOD, on hydrocortisone, levothyroxine and desmopressin, with severe visual impairment, insufficient sleep and marked sensitivity to sound. Her behavioural difficulties were addressed only after her carers raised them. The authors use the case to show that behaviour and daily function are easily overlooked when the medical focus is on hormones and eyes.

On mechanisms, the authors suggest the hypothalamus may be a shared link and mention oxytocin and vasopressin as a research direction. No study has tested this in SOD.

Two points follow from this review for families considering treatments. Some contributors to delay, in particular hypothyroidism and disturbed sleep-wake rhythm, are treatable today by standard means and should be looked for and managed. Development in these children is also very variable, and the measuring tools are unreliable in the presence of visual loss. A change seen after any intervention in a single child is therefore very hard to interpret without a comparison group. No study in this review tested a treatment of any kind.