Methods of quantitative MRI brain mapping and possible significance in diagnosis and response to therapy of childhood autism and attention deficit hyperactivity disorder are reviewed by investigators at the Child Psychiatry Branch, National Institute of Mental Health, Bethesda, MD. Recent advances in MRI allow automated analysis in measurement of regional gray and white matter volumes across time and between subject groups. Standard curves of healthy and disease-specific regional maturation and volumetric changes during development may be helpful in diagnosis. Except for the cerebellum, volume measurements are more highly correlated for monozygotic than dizygotic twins, and heritability indices for brain regions may be calculated. Brain volume and region-specific brain tissue changes may identify patients at high risk for autistic disorder and ADHD.

In studies at the NIH, childhood-onset schizophrenia is characterized by region-specific progressive cortical gray matter loss in parietal, frontal, and temporal regions. A “wave” of back-to-front tissue loss, early parietal followed by frontal and temporal gray matter loss, occurs in adolescence. Loss of gray matter slows in early adulthood. In contrast, two large studies of 57 boys and 50 girls with ADHD showed smaller brain size, abnormal caudate, and decreased volume of the posterior inferior cerebellar vermis, when compared to 105 matched controls. Anatomical changes in patients with ADHD support a postulated dysfunction of cerebellar-striatal-prefrontal brain circuits. Although results of MRI studies at present are promising, they are not sufficiently homogeneous to indicate routine brain MRI in childhood neuropsychiatric disorders. [1]

COMMENT. These large prospective MRI studies of the brains of hyperactive, autistic, and healthy children and adolescents, undertaken at the NIH since 1990, have uncovered subtle structural changes in brain development that supplement postmortem findings and are potentially significant in diagnosis and management.

Impaired activation of prefrontal-temporal regions demonstrated by PET studies in autistic subjects at the Wellcome Department of Cognitive Neurology, London, UK [2]. While watching animated sequencies depicting two triangles moving interactively, a ‘mentalizing’ effect that is normally associated with specific brain activation is impaired in autistic compared to control subjects. It is theorized that autism is the result of impaired mentalizing, and is characterized by a lack of social insight and poor communicative skills (Baron-Cohen et al. 1985).