Best Evidence Topics
  • Send this BET as an Email
  • Make a Comment on this BET

Should Fragile X be tested for in boys with a diagnosis of autistic spectrum disorder?

Three Part Question

In [boys with a diagnosis of autism] and [no family history of learning disability - negative test result] what is [the risk of fragile X syndrome]?

Clinical Scenario

You diagnose a 5-year-old boy with autistic spectrum disorder. His examination is unremarkable and there is no family history of learning difficulties. Should you perform a molecular genetic screen for fragile X?

Search Strategy

PubMed: the search was limited to 1991–2008, as molecular genetic testing was routinely used to test for this condition from 1991 onwards.Secondary sources
Clinical Evidence: no relevant items. Cochrane Library: no relevant items. EMBASE: no relevant items.
Searches were carried out using "Autism", "Autistic spectrum disorder" combined with "FRAX" and "fragile X".

Search Outcome

Thirty articles were retrieved and the abstracts were reviewed. Four relevant articles were included. No papers were specific to boys alone. Many of the articles had very small case numbers or used cytogenetic testing as opposed to molecular testing for diagnosis. Other studies reported a much higher incidence (7–8%) of fragile X syndrome, but as there was evidence of selection criteria bias we did not review these articles.

Relevant Paper(s)

Author, date and country Patient group Study type (level of evidence) Outcomes Key results Study Weaknesses
Fombonne et al,
Epidemiological study of 325 347 French children; 6100 were registered with the local authority as having special needsCohort study (level 2b)174 had an ICD-10 diagnosis of "classic" autism Prevalence: 5.35/10 000 for autism and 11/10 000 for PDD3 fragile X in children with autism and 5 in the PDD group

Prevalence of fragile X in autistic children was 1.7% and 1.4% in the PDD group
Not all children with autism were screened and those that were used cytogenetic testing

Likely to be an underestimate
Kielinen et al,
Population based study of 152 732 Finnish children <16 years old, data from hospital records and a database of the intellectually disabledCohort study (level 2b)187 had a DSM-IV diagnosis of autistic disorder (including classic autism and other PDD)4 fragile X, ie prevalence rate of fragile X in children with autistic disorder was 2.1%Universal screening for fragile X in this population Used a broader definition of autism


The incidence of autism in the general population is 6–10/10 000 in developing countries depending on the diagnostic criteria used (Baird). Diagnosable causes of autism currently account for 10% of all cases, with the other 90% being referred to as "idiopathic" autism. Autism is usually diagnosed between 2 and 5 years of age and the recurrence risk in affected families is in the region of 2–8% (Muhle).

Fragile X syndrome is inherited as an X-linked condition. The prevalence of fragile X in the general population is 1 in 2000 in males and 1 in 4000 in females (Song). It consists of a triad of features of unusual facial appearance, macro-orchidism in adulthood and cognitive impairment of variable severity. These features may not manifest until the teenage years. The genetic basis for the syndrome is an unstable mutation due to an expansion of a CGG triplet-stretch in the FMR1 (fragile X mental retardation protein) gene located on the X chromosome. About 30% of affected individuals have an autistic spectrum disorder (Hatton, Cohen).

One of the main diagnostic problems with fragile X syndrome is that the typical clinical gestalt may not become apparent until the child is older (Fisch, Feinstein). Therefore, genetic counselling, one of the main benefits of early syndrome diagnosis, may not be offered in time to parents before they conceive another child (FitzPatrick).

We reviewed two large population based studies. In France registration with the local authority is compulsory in order for a child to access special education and care in both state and private institutions and schools. The research team reviewed the information on children born between 1976 and 1985 contained in local authority records. They identified 174 children with autistic spectrum disorder. The French cohort1 had an incidence of 2.9% of all chromosomal abnormalities including fragile X. Three boys with fragile X were identified among the autistic boys (n = 174) and 10 boys with fragile X in the non-autistic group (n = 5926). This gives a prevalence rate of fragile X of 0.17% among non-autistic boys with special needs and 1.7% among children with autism.

The Finnish study (Kielinen) was a cohort study of 152 732 children living in northern Finland. The authors identified 187 children with a DSM-IV diagnosis of autistic disorder (they excluded Asperger syndrome because of uncertainty concerning DSM-IV criteria). The prevalence rate of fragile X in the population found to have autism was given as 2.1% (the overall population prevalence rate of fragile X in Finland is 0.3 per 1000). As screening for fragile X syndrome in Finland is conducted in all children with mental disabilities including autism, this figure is presumed to represent the true prevalence.

These studies lead to the conclusion that 50 children with autism would need to be tested to identify one with fragile X. This incidence is much higher than that in the general population. However, one would have to consider the other 49 tested children who will have negative results. The potential adverse effects include the long time lag between testing and results, false negatives and parental anxiety about the test results. Also, it may be difficult to obtain blood samples from children with a severe social communication disorder. The benefits to the family of early identification (in terms of genetic counselling) may be marked in a condition such as fragile X where the clinical signs may only become apparent when the child is older. Up to 80% of children with fragile X have mild to severe learning difficulties. However, parents and educators should be aware from an early age that many children with fragile X achieve above the level that would have been predicted from measured IQ (Feinstein ).

Editor Comment

PDD, pervasive developmental disorders.

Clinical Bottom Line

Fragile X syndrome is significantly more common in children with autism (2%) than in the general male population (0.05%). (Grade C) Approximately 50 children with autism would have to be tested to diagnose one child with fragile X syndrome. (Grade C) Earlier identification of fragile X syndrome enables earlier genetic counselling and more accurate prognostic information to be offered. (Grade D)


  1. Fombonne E, Du Mazaubrun C, Cans C, et al. Autism and associated medical disorders in a French epidemiological survey. J Am Acad Child Adolesc Psychiatry 1997;36:1561–9.
  2. Kielinen M, Rantala H, Timonen E, et al. Associated medical disorders and disabilities in children with autistic disorder: a population-based study. Autism 2004;8:49–60.
  3. Baird G, Simonoff E, Pickles A, 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;368:210–15.
  4. Muhle R, Trentacoste SV, Rapin I. The genetics of autism. Pediatrics 2004;113:e472–86.
  5. Song FJ, Barton P, Sleightholme V, et al. Screening for fragile X syndrome: a literature review and modelling study. Health Technol Assess 2003;7(16).
  6. Hatton D, Sideris J, Skinner M, et al. Autistic behaviour in children with fragile X syndrome: prevalence, stability, and the impact of FMRP. Am J Med Gen A 2006;140A:1804–13.
  7. Cohen D, Prichard N, Tordjman S, et al. Specific genetic disorders and autism: clinical contribution towards their identification. J Autism Dev Disord 2005;35:103–16.
  8. Fisch G. Is autism associated with the fragile X syndrome? Am J Med Gen 1992;43:47–55.
  9. Feinstein C, Reiss A. Autism: the point of view from fragile X studies. J Autism Dev Disord 1998;28:393–405.
  10. FitzPatrick DR, Pearson P, Halpin S, et al. A school based study of children with learning disability indicates poor levels of genetic investigation. J Med Genet 2002;39:e19.