Autism is a behavioral syndrome usually presenting behavior abnormalities before the child is 30 months of age. These behavioral abnormalities include marked social deficits, specific language abnormalities and stereotyped, repetitive behaviors (Piven, 1990). Although the exact etiology of autism is not known it is now believed that it is a dysfunction of one or more unidentified brain systems and not the result of parental and environmental influences. Variations in symptomology and in prognosis among autistic persons depend on both the severity and the extent of the underlying brain dysfunction (Repin, 1991). Several studies have been conducted including neuroanatomic imaging, microscopic neuroanatomic observations and positron emission tomography in order to locate this dysfunction. Although some of the studies have reported abnormalities in various areas of the brain among the autistic patients, no common site or abnormality has yet been found.

The criteria for diagnosing autism are those listed in the revision of the third edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-III-R) of the American Psychiatric Association. According to this, the person must have been symptomatic since childhood and manifest a specified number of deficits that are abnormal for their developmental level in three aspects of behavior. These include qualitative impairment in reciprocal social interaction, qualitative impairment of verbal and non-verbal communication and imaginative activity, and a markedly restricted repertoire of activities and interests. These criteria were used to select the autistic patients for most of these studies.

The presenting symptomology of autistic children is the abnormal reciprocal social interaction. Infants resist cuddling and preschoolers walk past people as if they did not see them. Throughout their lives autistic children avoid eye contact and are unaware of other people’s feelings. They are unable to make friends and do not learn to play with other children. They often prefer social isolation.

Many autistic children have some sort of language disorder and this will often be another presenting complaint to the doctor from the parents. Some children are mute and understand little; others speak late and produce unintelligible sentences with no desire to communicate. These children often memorize television commercials and recite them with perfect imitation of the tone of voice and the rhythm of the speaker they are imitating. Although these children are unable to verbally communicate, they also rarely use gestures or imitate actions to communicate. They often will retrieve things and try to do things for themselves rather than try and ask for them.

Autistic children need routines and any deviation from routine severely upsets them. Repetitive movements such as rocking, humming and flapping hands is characteristic behavior. As these children become older, they may spend weeks to months learning about one specific topic.

The intelligence range of autistic children varies from profound mental deficiency to superior intelligence. "Idiots savants" is the name referred to autistic persons who have a superior skill in a narrow range of abilities such as calculations or rote verbal memory (Repin, 1991).

Although many studies have been done by various methods, one common site of brain abnormality has not been found consistently in all autistic persons. However, it does seem clear that most autistic persons have some type of brain abnormality.

Dr. Margaret I,. Bauman has reported on microscopic neuroanatomic observations of four brains of well-documented autistic patients. All of the brains were well-developed and showed no gross abnormalities. However each of the brains showed increased cell-packing density and reduced nerve cell size bilaterally in the hippocampal complex, subiculum, entorhinal cortex, portions of the amygdala, the mammillary body and the septal nucleus. Three of the four brains also showed this abnormality in the anterior cingulate gyrus. These abnormalities were found by comparison to identically processed age and sex matched control subjects. The specific regions of the limbic system thought to be involved in autism when lesioned are known to produce purposeless hyperactivity, impaired social interaction and stereotypic motor behavior. These behaviors are similar to the behaviors of autistic individuals. Because of these observations, it is now thought that the limbic system must be involved (Bauman, 1991)-

Dr. Bauman also reported a variable loss of Purkinje cells and to a lesser extent granule cells in the neocerebellar cortex. However, the expected retrograde degeneration of the principle inferior olivary climbing fibers was not found. Because of this, in comparison with primate development, it is now thought that the process that caused these abnormalities began or occurred before birth. How the abnormalities in the limbic system end in the cerebellum are related is not known. Also the relationships of cerebellar abnormalities to the clinical picture of autism is unclear.

Dr. Eric Courchesne from the neuroscience department at the University of California, San Diego, summarized the results of several neuroimaging studies. He found that there is no substantial evidence of abnormality in the cerebral hemispheres, corpus callosum, thalamus, basal ganglia, brain volume of the lateral ventricles and the third ventricle. Based on studies by Gaffney in which enlargement of the lateral ventricles was found, and on studies by Jacobson which found enlargement of the third ventricle, Dr. Courchesne found that about 15% of autistic patients have enlargement of some portion of the lateral ventricle, but the majority of neuroimaging studies have not found statistically significant enlargement in either the third or lateral ventricles (Courchesne, 1991).

Dr. Courchesne also concluded that the reduction of cerebellar size due to Purkinje neuron loss in the cerebellar hemispheres and sometimes the cerebellar vermis was conclusive by both magnetic resonance (MR) and autopsy studies. In ten autopsy cases, there was Purkinje cell loss in the cerebellar hemisphere and in five of ten cases there was Purkinje cell loss in the cerebellar vermis. The magnetic resonance studies have found that there is a reduction in the size of the neocerebellar vermis and the neocerebellar hemisphere which is conclusive with the Purkinje cell loss (Courchesne, 1991).

Due to the fact there is no evidence of atrophy or deterioration after full development and the results of MR images and autopsy studies, it is now believed that the neocerebellar loss occurs before birth. The exact time of onset is still questionable but the fact that the neocerebellum is embryologically and phylogenetically distant from other portions of the cerebellum may help pinpoint the time. This may help detect other regions of the brain that may be affected which were developing at this time too.

There are many correlation’s between the symptoms seen in cerebellar lesions and in symptoms seen in autism according to Dr. Courchesne, such as disturbances of smooth ocular pursuit, mutism and dysarthria, disturbed motivational behavior and reduced social interactions.

Dr. Nancy J. Minshew of the Western Psychiatric Institute and Clinic, University of Pittsburgh, Pennsylvania has used electroencephalographic (EEG) studies in the autistic population to try and locate the abnormalities on the brain structures. The abnormalities seen in EEGs and the frequency of seizures in autistic persons were among the earliest evidence of a biologic basis for autism. Abnormal EEG’s were found in 65% of 147 autistic children when repeated EEG’s were done according to Small in 1975 (Small, 1975). Depending on the state of alertness of the child and the number of recordings, the chance of an abnormal EEG was changed significantly. With a greater number of recordings, there was a greater frequency of abnormal EEGs. Abnormal EEG’s were also found more frequently in which different recordings included states of sleep, awake and drowsiness, instead of just one or two states. It was also noted that there was a higher incidence of abnormal EEG’s in mentally retarded autistic individuals but a significant amount of abnormal EEG’s were also found in the mildly and non-retarded autistic persons. The EEG abnormalities have involved all regions of the cortex without a predilection for any one region or hemisphere (Minshew, l991).

Dr. Minshew, through positron emission tomography (PET) and nuclear magnetic resonance spectroscopy studies, has concluded that the association cortex is involved in the pathophysiology of autism. The symptomology seen n autism is often associated with dysfunction of the cerebral cortex, such as deficits in abstract reasoning, abnormal use of pitch and rhythm in speaking, and lack of eye contact and facial expressions in communication. Therefore all clinical manifestations of autism can be accounted for by the dysfunction of the association cortex (Minshew, 1991).

Besides the studies cited, many more studies have been done without the exact discovery to locate the etiology of autism. Many of these studies have conflicting results. Some of the problems encountered that cause discrepancies in the results are the criteria used for selecting patients for the studies, and/or the improper use of a reference group.

Abnormalities have been found in the limbic system, the association cortex and the cerebellum according to the studies cited. Defects in all these structures correlate with some, but perhaps not all the symptomology found in autism; therefore it is difficult to tell what defects are causing which specific symptomology.

Many more studies will need to be done before the etiology of autism will be found. The studies will need to be done with standardized criteria for selecting patients and for selecting the reference group. Without this, the same problems that are occurring now will continue to occur.

BIBLIOGRAPHY:

American Psychiatric Association, Committee on Nomenclature and Statistics. Diagnostic and Statistical Manual of Mental Disorders. 3rd ed. revised. Washington, DC: American Psychiatric Association, 1987.

Bauman, M. L., "Microscopic Neuroanatomic Abnormalities in Autism." Pediatrics. Vol. 87., suppl. 5. ,May 1991.,pp. 791-796.

Courchesne, E. "Neuroanatomic Imaging in Autism", Pediatrics. Vol. 87, suppl. 5, May 1991, pp. 781-790.

Minshew, N. J. "Indices of Neural Function in Autism: Clinical and Biologic Implications." Pediatrics, vol. 87, suppl. 5. May 1991. pp. 774-780.

Piven, J. et. al. "Magnetic Resonance Imaging Evidence for a Defect of Cerebral Cortical Development in Autism." American Journal of Psychiatry, vol. 147, no. 6, June 1990, pp. 734-739.

Rapin, I. "Autistic Children: Diagnosis and Clinical Features." Pediatrics, vol. 87, suppl. 5, May 1991, pp. 751-759.

Small, J. O. "EEG and Neurophysiological Studies of Early Infantile Autism." Biological Psychiatry, vol. 10, 1975, pp. 355-397.