page contents

A Review of Current Thoughts on Localized Structural Lesions in Autistic Disorder

Philip Omotosho

Penn State College of Medicine



Definition and Epidemiology

The DSM-IV diagnosis of autistic disorder requires qualitative impairments in social interaction and in communication, plus restricted repetitive and stereotyped patterns of behavior.

Autistic disorder (autism) is rare, occurring in 2-5 children per 10,000 live births. It is more common in males, with a male to female ratio of 3:1. Risk of recurrence in siblings is 3-5 percent, a risk about 75 times greater than that in the general population. Autism is among the group of disorders known as the Pervasive Developmental Disorders. The other disorders in the group include Rett’s disorder, childhood disintegrative disorder, and Asperger’s disorder.

Mental retardation is associated with 70 percent of cases and seizures with 33 percent. Autistic features are associated with other neurological diseases such as tuberous sclerosis, neurofibromatosis, fragile X syndrome, and phenylketonuria. These findings have led to the conclusion that there is a genetic predisposition to autism.


Structural Abnormalities in Autistic Disorder

Several abnormalities have been described in the last 20 years in the study of Autism, including:

  1. Delayed maturation of frontal lobe circuitry (Zilbovicius et al., 1993)
  2. Decreased functional connections within the cerebral cortex and between the cortex and subcortical regions (Horwitz et al., 1988)
  3. Abnormalities of the dentatothalamocortical pathways (Chugani et al., 1997)

These findings point not to a specific location in the brain but suggest that autism is due to connection abnormalities between neural networks that process information.

The following studies have found evidence for localized structural abnormalities, which could explain clinically observed autistic behavior.

The Hardan et al. Study (focuses on posterior fossa structures)

Specifics of the study:

  • Non-mentally retarded adolescents and adults (males 12 and older)
  • Case-control study with 16 autistic subjects and 19 controls
  • Controls were matched with autistic subjects with regard to age, gender, and IQ


Cerebellar hemispheres were significantly larger (13.8%) in the autism group compared with the control group, adjustments having been made for total brain volume (TBV). The volume abnormality did not involve the vermis, or the brainstem.

  1. This finding is consistent with the clinical observation in autism of increased TBV and head circumference.
  2. Alterations in volume consistent with neuropathological findings of abnormalities in this structure, such as decreased number of Purkinje cells. (Explain this).
  3. The cerebellum is thought to be involved in the modulation of thought, planning, learning and memory, and possibly language.

The cerebellum is affected in a variety of other neurological disorders, so it not a specific marker for autism, nor can it be exclusively responsible for clinically observed autistic behavior.

The Boddaert et al. Review

This paper reviews functional imaging done in autism in the last twenty years, but emphasizes two of the most recent studies (Ohnishi et al. and Zilbovicius et al.), which have independently arrived at similar conclusions. These two studies relied heavily on the use of the sophisticated imaging techniques, such as fMRI, PET, and SPECT. The studies basically fall into two different groups:

  • Rest measurements of regional glucose metabolism or cerebral blood flow
  • Studies performed during brain activation paradigms
  1. The Ohnishi study
    • SPECT study in 23 autistic children and 26 control children
    • Detected significant bitemporal hypoperfusion located bilaterally in the superior temporal gyrus (STG) and in the left frontal region.
  1. The Zilbovicius study
  • PET study in 21 autistic children and 10 control children
  • Detected significant bitemporal hypoperfusion in autistic children located bilaterally in the STG and in the right superior temporal sulcus.

Both of these studies used as controls children with idiopathic mental retardation; the results found in the autistic children may therefore not be dismissed as simply due to the mental retardation that so frequently is associated with autism.

Important points from these studies:

  • They provide the first substantial evidence for localized dysfunction of the cerebral cortex in school-aged children with primary autism
  • The temporal region dysfunction may be implicated in almost all of the clinical symptoms in autism (perceptive, emotional and cognitive deficits), since these associative regions are highly connected with the frontal and parietal lobes and the limbic and associated sensory systems
  • Autistic behavior has also been reported in clinical temporal lobe pathology, such as epilepsy and HSV encephalitis. There is also a strong association between temporal lobe tubers in tuberous sclerosis and autistic syndrome.

The studies performed during activation using auditory paradigms showed a decreased activation of the left posterior STG in autistic children compared to controls, suggesting an abnormal pattern of activation of the left temporal cortex.



  1. Boddaert N, Zilbovicius M (2002). Functional Neuroimaging and Childhood Autism. Pediatric Radiology 32:1-7
  2. Hardan AY, Minshew NJ, Harenski K, Keshavan MS (2001). Posterior Fossa Magnetic Resonance Imaging in Autism. J Am Acad Child Adolesc Psychiatry 40:666-672
Share This