Corpus Callosum Morphology in Attention Deficit-Hyperactivity Disorder: Morphometric Analysis of MRI George W. Hynd, Margaret Semrud-Clikeman, Alison R. Lorys, Edward S. Novey, Deborah Eliopulos, and Heikki Lyytinen Although behavioral evidence provides support for the notion that attention deficit- hyperactivity disorder (ADHD) is related to central nervous system dysfunction, there is little direct evidence to reveal which neurometabolic systems or brain struc- tures are involved. Recent magnetic resonance imaging (MRI) studies suggest that, compared to nondisabled controls, ADHD children may have a smaller right frontal region. Morphometric analysis of MRI scans was used in this exploratory study to determine whether correlated regional variation might exist in the corpus callosum of children with ADHD. While all MRI scans were judged to be clinically normal, morphometric analysis revealed that, compared to nondisabled controls, ADHD children had a smaller corpus callosum, particularly in the region of the genu and splenium, and in the area just anterior to the splenium. Interhemispheric fibers in these regions interconnect the left and right frontal, occipital, parietal, and posterior temporal regions. These results suggest that subtle differences may exist in the brains of children with ADHD and that deviations in normal corticogenesis may underlie the behavioral manifestations of this disorder. W hile it has always been presumed that the behavioral manifestation of attention deficit-hyperactivity disorder (ADHD) reflects a dysfunction of neural systems subserving the regulation of motor control, inhibition, and atten- tional processes, little direct evidence exists as to which neural systems or struc- tures are implicated (Hynd & Willis, 1988). As Barkley (1986) pointed out, controversy regarding the neurological etiology of ADHD has been one of the most dominant themes in childhood psy- chopathology. Zametkin and Rapoport (1986) noted nearly a dozen neuroanatomically based hypotheses, including Laufer and Den- hoffs (1957) early ideas about dysfunc- tional diencephalic structures (thalamus, hypothalamus), Wender's (1974) notions about decreased reticular activating system (RAS) activation, and Dykman, Acker- man, Clements, and Peters' (1971) theory regarding a deficient forebrain inhibitory system. Consistent with some of these theories, Lou, Henriksen, and Bruhn (1984) employed regional cerebral blood flow/computed tomography (rCBF/CT) in children with ADHD and found evi- dence of periventricular hypoperfusion implicating low metabolic activity in the central and frontal regions of the brain. A more recent report, by Lou, Henrik- sen, Bruhn, Borner, and Nielsen (1989), included additional subjects using the same rCBF/CT procedures. They found that hypoperfusion was most significant in the right striatum in children with only ADHD and bilaterally hypoperfused in children with ADHD with co-occurring neurological symptoms (mild MR, dys- phasia). It was concluded that low stria- tal activity, seemingly reversible with the administration of methylphenidate (Rita- lin), was characteristic of children with ADHD. Less invasive neuroimaging procedures may shed additional light on deviations in brain morphology that contribute to the behavioral symptoms associated with ADHD. However, an initial CT study found no significant differences between children with attention deficit disorder and controls in measurements of the anterior horns of the lateral ventricles, bifrontal width, and right and left hemi- spheres (Shaywitz, Shaywitz, Byrne, Cohen, & Rothman, 1983). Consequent- ly, it would seem that more traditional measures of brain variation, as employed in the Shaywitz et al. study, may not reveal subtle regional variation in the brains of children with ADHD —should such differences in fact exist. A more recent study, employing mag- netic resonance imaging (MRI) with care- fully diagnosed populations of children with developmental dyslexia and ADHD, found that the children with ADHD dif- fered significantly from nondisabled con- trols in the right frontal width measure (Hynd, Semrud-Clikeman, Lorys, Novey, & Eliopulos, 1990). These differences were not clinically observable and emerged only when reliable intracranial measures were made on the MRI scan. It is of some interest that both children with dyslexia and children with ADHD dif- fered from nondisabled controls in the pattern of brain morphology in the fron- tal region. The nondisabled controls evi- denced the expected pattern of L < R asymmetry of the frontal lobes, whereas the children with dyslexia and children with ADHD evidenced symmetry of the frontal region (L = R). While variation in regional morphology may indeed relate in some conceptually consistent fashion to the behavioral manifestations of dyslexia or ADHD, it is not unreasonable to suggest that sub- cortical structures might also reflect deviations in neurodevelopment in these children. One such subcortical structure is the corpus callosum, which can easily be seen on midsagittal MRI scans. The corpus callosum connects the homologous areas of the cortex (Inno- centi, 1981). Although few primary sen- sory or motor fibers transverse the cor- pus callosum (Kertesz, Polk, Howell, & Black, 1987), the corpus callosum seems vital to the transfer and facilitation of associative information between hemi- spheres (Lassonde, 1986). Deviations in morphology of the corpus callosum may be related to lateralized speech domi- nance (O'Kusky et al., 1988) or gender (de Lacoste-Utamsing & Holloway, 1982). The evidence in favor of these potential differences is controversial, however, as conflicting evidence exists (Kertesz et al., 1987; Oppenheim, Lee, Volume 24, Number 3, March 1991 141