ArticlesPrevalence of autism spectrum disorder phenomenology in genetic disorders: a systematic review and meta-analysis
Introduction
Autism spectrum disorder (ASD) is a broad term for a group of behaviourally defined neurodevelopmental disorders that historically includes autistic disorder, childhood autism, pervasive developmental disorder—not otherwise specified (PDD-NOS), and Asperger's syndrome.1, 2 ASD is defined by the presence of abnormalities or impairments in social interaction and communication, with accompanying restricted or repetitive behaviours, activities, or interests. Estimates for the prevalence of ASD in the general population range from one in 100 people3, 4 to one in 68.5 However, despite the high prevalence and robust research documenting heritability,6 a genetic aetiological cause has yet to be identified. This absence might be partly because of the methodological challenges caused by the behavioural heterogeneity within the spectrum.7
ASD phenomenology seems more prevalent in individuals with specific genetic and metabolic syndromes than in those without these sydromes.8, 9 Study of the prevalence and phenomenology of ASD within and across these syndromes could disentangle the genetic and biological pathways that underlie idiopathic ASD.7, 10, 11 Recent findings show that particular rare de-novo or transmitted copy number variations substantially increase risk of ASD.12 Through studying the downstream disruption caused by these variations, researchers are identifying candidate ASD-associated genes. From these biological markers, it might be possible to identify specific cognitive deficits that underpin characteristic idiopathic ASD behaviours.
Evidence is emerging that individuals with certain genetic and metabolic syndromes might have an atypical profile of ASD phenomenology, which would support a distinction between syndromic variants of ASD and idiopathic ASD.13, 14, 15 A pragmatic strategy to evaluate the presence of syndromic variants of ASD would be to conduct detailed analysis of ASD phenomenology in syndromes in which prevalence estimates for ASD are consistently high. However, despite many systematic reviews16, 17, 18, 19 no meta-analytic studies have documented the consistency of prevalence data within syndromes, calculated variation of prevalence estimates between syndromes, or compared prevalence estimates to those for the general population.
Several methodological challenges complicate synthesis of prevalence data for ASD phenomenology across syndromes. First, the diagnosis of ASD in clinical practice requires rigorous multicomponent assessment, with information collected from many sources and across contexts.19 This depth and breadth of diagnostic assessment is rarely replicated in research, and any prevalence statistics are thus more accurately described as estimates of the presence of ASD phenomenology, rather than estimates of the presence of diagnostically defined ASD. Additionally, many studies rely solely on screening measures, which take less time and resources. However, screening measures often have low specificity and sensitivity14, 20 and thus the prevalence data have wide confidence intervals. Diagnostic measures have greater sensitivity and specificity, however, prevalence data might be biased because accuracy is lowest for marginal or unusual cases, such as those with intellectual disability.20
The association between ASD phenomenology and greater severity of intellectual disability21 is hypothesised to contribute to the behavioural phenotypes associated with genetic syndromes, such that associated degree of disability, rather than the presence of the syndrome itself, more fully accounts for the presence of ASD in these groups.22 Although this pattern of association is often evident in individual syndrome groups, sometimes the association between ASD and intellectual disability is less robust. For example, in Cri du Chat syndrome, intellectual disability is usually severe and prevalence of ASD is relatively low compared with syndromes with similar levels of intellectual disability, even after controlling for intellectual functioning.23 Conversely, individuals with fragile X syndrome seem at increased risk of ASD phenomenology despite a wide range of intellectual functioning in the group—although within this group, ASD phenomenology is negatively associated with IQ.14 Individuals with Cornelia de Lange syndrome show higher amounts of ASD phenomenology than do individuals with intellectual disability of heterogeneous aetiology with comparable adaptive functioning.24 Available models of this association are based on individual empirical studies that are limited by small samples and cohort effects. The delineation of robust rates of ASD phenomenology for each syndrome through meta-analytic methods would advance attempts to evaluate comprehensively the association between ASD phenomenology and intellectual disability in syndromes.
In this systematic review and meta-analysis, we aimed to describe and evaluate the scientific literature for ASD phenomenology in genetic and metabolic syndromes and to generate pooled estimates for prevalence of ASD phenomenology within each syndrome, weighted by the quantity and quality of the available evidence; to make preliminary comparisons of the pooled prevalence estimates across syndromes; and to compare pooled prevalence estimates in the syndromes with estimates of ASD in the general population.
Section snippets
Search strategy and selection criteria
With use of a 2009 review, we generated a list of 21 syndromes most likely to be associated with ASD.15 The review highlighted genetic syndromes in which at least one empirical study had been done and previous systematic reviews had described the syndrome to be associated with ASD. We did literature searches in Ovid PsycINFO, Ovid MEDLINE, Ovid Embase, and PubMed Central for English-language papers published from database creation up to early 2014 (appendix). Searches were done by combining
Results
We identified 32 230 papers and selected 168 papers, across 16 syndromes, as suitable for qualitative review (figure 1, table 1). Across syndromes, only nine (5·4%) papers met criteria for the highest quality rating for sample identification, whereas 89 (53·0%) obtained the highest quality rating for syndrome confirmation, and 43 (25·6%) for ASD assessment. Only one (0·6%) paper met the highest quality rating for all three quality criteria. Nine (5·4%) papers were excluded from the pooled
Discussion
In this meta-analysis, we present pooled data and cross-syndrome comparisons for the prevalence of ASD phenomenology in rare genetic and metabolic syndromes. To our knowledge, this was the first meta-analysis of the prevalence of ASD phenomenology across many genetic syndromes and thus extended findings from previous systematic reviews.14, 15, 16, 17 Our wide search criteria and screening of both abstracts and titles during the initial search stages allowed the identification and inclusion of a
References (30)
- et al.
Prevalence of disorders of the autism spectrum in a population cohort of children in South Thames: the Special Needs and Autism Project (SNAP)
Lancet
(2006) - et al.
Genetic heterogeneity between the three components of the autism spectrum: a twin study
J Am Acad Child Adolesc Psychiatry
(2006) - et al.
Genetic advances in autism: heterogeneity and convergence on shared pathways
Curr Opin Genet Dev
(2009) - et al.
Searching for ways out of the autism maze: genetic, epigenetic and environmental clues
Trends Neurosci
(2006) - et al.
Convergence of genes and cellular pathways dysregulated in autism spectrum disorders
Am J Hum Genet
(2014) - et al.
Autism in fragile X syndrome: a category mistake?
J Am Acad Child Adolesc Psychiatry
(2010) - et al.
Childhood autism and associated comorbidities
Brain Dev
(2007) - et al.
Intellectual disability and its relationship to autism spectrum disorders
Res Dev Disabil
(2009) Rethinking the nature of genetic vulnerability to autistic spectrum disorders
Trends Genet
(2007)- et al.
Development quality criteria to evaluate nontherapeutic studies of incidence, prevalence, or risk factors of chronic diseases: pilot study of new checklists
J Clin Epidemiol
(2011)