Infantile autism and computerized tomography brain-scan findings: specific versus nonspecific abnormalities

Cluster analysis of autistic patients based on principal pathogenetic components

We have recently described four principal pathogenetic components in autism: (I) circadian and sensory dysfunction, (II) immune abnormalities, (III) neurodevelopmental delay, and (IV) stereotypic behaviors. Using hierarchical and k-means clustering, the same 245 patients assessed in our principal component analysis can be partitioned into four clusters: (a) 43 (17.6%) have prominent immune abnormalities accompanied by some circadian and sensory issues; (b) 44 (18.0%) display major circadian and sensory dysfunction, with little or no immune symptoms; (c) stereotypies predominate in 75 (31.0%); and (d) 83 (33.9%) show a mixture of all four components, with greater disruptive behaviors and mental retardation. The "immune" component provides the largest contributions to phenotypic variance (P = 2.7 x 10(-45)), followed by "stereotypic behaviors." These patient clusters may likely differ in genetic and immune underpinnings, developmental trajectories, and response to treatment.

Subtypes of autism by cluster analysis based on structural MRI data

The aim of our study was to subcategorize Autistic Spectrum Disorders (ASD) using a multidisciplinary approach. Sixty four autistic patients (mean age 9.4+/-5.6 years) were entered into a cluster analysis. The clustering analysis was based on MRI data. The clusters obtained did not differ significantly in the overall severity of autistic symptomatology as measured by the total score on the Childhood Autism Rating Scale (CARS). The clusters could be characterized as showing significant differences: Cluster 1: showed the largest sizes of the genu and splenium of the corpus callosum (CC), the lowest pregnancy order and the lowest frequency of facial dysmorphic features. Cluster 2: showed the largest sizes of the amygdala and hippocampus (HPC), the least abnormal visual response on the CARS, the lowest frequency of epilepsy and the least frequent abnormal psychomotor development during the first year of life. Cluster 3: showed the largest sizes of the caput of the nucleus caudatus (NC), the smallest sizes of the HPC and facial dysmorphic features were always present. Cluster 4: showed the smallest sizes of the genu and splenium of the CC, as well as the amygdala, and caput of the NC, the most abnormal visual response on the CARS, the highest frequency of epilepsy, the highest pregnancy order, abnormal psychomotor development during the first year of life was always present and facial dysmorphic features were always present. This multidisciplinary approach seems to be a promising method for subtyping autism.

Childhood autism: An appeal for an integrative and psychobiological approach

The difficulty that a person with autism has in establishing relationships, maintaining them (communicating and responding appropriately) is a common experience of those close to them., That impaired perceptual and cognitive processing can underlie this difficulty and the interactions of people with autism with the material environment has been established in the laboratory. The consequences at a psychological level of analysis may converge in the inadequacy of second-order representations of the world. An attenuation of such endogenous monitoring processes could also indirectly account for features of withdrawal and the stereotypies often observed. At another level of analysis there are delays in neurotransmission, in the CNS and a lack of flexibility of physiological response shown by evoked potential recordings. Tomographic studies of blood flow and metabolism illustrate a lack of correlation between information processing centres in the brain that may sometimes arise from diffuse gray matter atrophy. A "stop-go" form of modulation of central processing is mediated by anomalous ascending serotonergic and dopaminergic function (transmitters with inhibitory and switching functions). On these bases it is no wonder that representations are not formed and inappropriate and repetitive behaviors follow, although the link remains somewhat speculative. Both levels of analysis are useful for an explanation. As behavioral and pharmacotherapy, though helpful, are severely limited in their efficacy, more effort is required to synthesize the different levels of analysis into a psycho-biological approach, to remedial programs and new forms of therapy.

Neuroimaging in infantile autism

Metabolic findings using [18F]fluorodeoxyglucose (FDG) with positron emission tomography (PET) and correlative anatomic findings with computed tomography (CT) or magnetic resonance imaging (MRI) were characterized in 13 children with infantile autism. Four of 13 patients had both an abnormal FDG-PET and an abnormal MRI, whereas seven of 13 patients had both a normal FDG-PET and a normal CT or MRI. Sixteen of a total of 195 brain areas qualitatively examined with FDG-PET had a hypometabolic abnormality on PET. Three of the five abnormal structural imaging studies revealed neuronal migrational anomalies (focal pachygyria). In two of the five patients with anatomic abnormalities, these were noted only after knowledge of the FDG-PET findings. Our experience reveals that anatomic and metabolic abnormalities can be found in children who exhibit autistic behavior. An FDG-PET study may provide evidence of metabolic dysfunction after an initially unremarkable MRI scan because subtle anatomic abnormalities (as those seen with neuronal migrational anomalies) may be found only after knowledge of a regional metabolic abnormality.

The neurobiology and genetics of infantile autism

Autism is a syndrome with multiple etiologies, as is made clear both by the evidence of neurobiological research and by the catalog of disorders that present with autistic behaviors. What remains unclear are the specific neuropathological mechanisms that produce autistic behaviors; for example, is there a common neuroanatomic pathology for all cases of autism, or can autistic behaviors emerge from different pathological sequences within the brain? Although it is premature to generalize, neuropathological studies appear to have identified common abnormalities in the cerebellum and limbic system of at least five autistic subjects. These subjects, with variable levels of mental retardation, demonstrated marked Purkinje cell loss in the cerebellar hemispheres, together with retained fetal neuronal circuitry in cerebellar nuclei and increased neuronal packing in specific regions of the limbic system, amygdala, and hippocampus. The architecture of the cerebral cortex was not affected. Although our knowledge of brain functioning is incomplete, alterations of the kind noted in the cerebellum and limbic system could reasonably produce autistic behaviors. For more detail, readers are directed to a review of cerebellar contributions to higher functions by Schmahmann (1991). Neuroimaging studies allow less resolution of brain structure than do neuroanatomic studies, and the reported findings from neuroimaging are somewhat contradictory. However, a number of investigators have reported structural abnormalities in ventricle size and cerebral hemispheric asymmetry using CT. MRI, which offers greater resolution, has uncovered some consistent findings, along with a variety of nonspecific abnormalities. Common abnormalities include reduced volume of cerebellar hemispheres and vermal lobules--findings not inconsistent with the above-mentioned neuropathological defects. It is also interesting to note that individuals with fragile X syndrome have similar cerebellar findings. PET and NMR studies of autism are at a preliminary stage, but these methodologies allow insight into the functioning of the brain, rather than simply brain anatomy. Recent PET studies indicating decreased association between paired regions of the brains of autistic subjects are of interest, particularly if they can be confirmed and refined by additional studies. Neurophysiological studies also offer insight into brain function, but are subject to numerous methodological criticisms. Nevertheless, recent reports of diminished P300 waves and absent NC components in autistic subjects seem to indicate fundamental defects in attention and secondary processing, which could help explain the self-stimulatory behaviors often seen in autism. The disturbances in brain development associated with autism can be produced in a number of ways, and at different times during development of the nervous system.(ABSTRACT TRUNCATED AT 400 WORDS)

Area and volume measurement of posterior fossa structures in MRI

This study was undertaken to determine the extent to which area measures of posterior fossa structures can be used confidently to represent structure volumes. MRI scans were obtained from three groups: fragile X males, males with other developmental disabilities, and males with normal IQ. The areas of the midbrain, pons, medulla, cerebellar vermis, and fourth ventricle were measured in midsagittal sections. Volumes of midbrain, pons, medulla, cerebellum, fourth ventricle, and third ventricle were obtained by measuring these structures in contiguous axial slices. In addition, the largest axial area for each structure was identified. Analysis revealed that midsagittal area measures for pons, medulla, and fourth ventricle were significantly correlated with structure volumes, and that all of the largest axial area measures were significantly correlated with structure volumes. However, only the midsagittal area measure of fourth ventricle and the largest axial area measures of fourth ventricle and cerebellum were correlated with volume measures with an r of .80 of greater. Results of this study suggest that area measures may not accurately represent three-dimensional structure size.