Little is yet known about brain asymmetry patterns in normal development or how they are altered in neurodevelopmental disorders such as fetal alcohol syndrome (FAS). FAS, a permanent birth defect caused by maternal consumption of large quantities of alcohol during pregnancy, is a leading cause of preventable developmental disabilities. Even the earliest reports of the syndrome documented significant neurodevelopmental abnormalities of which microcephaly is the most commonly described.

Volumetric, voxel-based, and surface-based image analyses have been used to study structural abnormalities in vivo in the brains of individuals exposed to alcohol prenatally (ALC). Results from these studies show that brain maturation continues to be adversely affected long after prenatal alcohol exposure. For example, research has shown increased gray matter density and decreased white matter density bilaterally in posterior temporal and inferior parietal regions in ALC subjects, suggestive of abnormal myelination during adolescence. The left hemisphere seems to be more affected than the right, leading to speculation that the ALC subjects may have altered asymmetry in the perisylvian cortex.

Some caveat is warranted when interpreting gray matter asymmetry patterns given the prominent spatially dependent nonuniformity of voxel intensities frequently found in MRI data. We observed that the pattern of non-uniformity tends to run diagonally across the brain, at least on the magnet used to image all of the subjects studied here (i.e., Signa; General Electric, Milwaukee, WI). The right frontal lobes tend to have lower signal than the left, and the left occipital lobes tend to have a lower signal value than the right. This effect is not completely eliminated with the inhomogeneity intensity correction algorithm used here and also used by Watkins and colleagues in their study of gray matter asymmetry. Thus, it is possible that the gray matter asymmetry pattern observed by us, and by others before us, is not about morphological differences between the hemispheres but about differences in voxel intensity due to inhomogeneity artifact. However, the findings of white matter asymmetry in the posterior temporal lobes in post mortem data, not subject to scanner artifact, suggest that our observations are valid. Also, the ALC and control subjects (SDC and YC) were scanned in the same type of magnet, with presumably the same artifact, so the effect of inhomogeneity would be subtracted out in group difference analyses. Given this, we are confident about the altered gray matter asymmetry pattern reported here in the alcohol-exposed subjects, but issue more caution in interpreting the gray matter asymmetry maps within groups.

While we attribute the morphological abnormalities in the ALC subjects to the prenatal alcohol exposure, it is possible that other environmental factors (pre or postnatal) unique to the ALC group could also result in the aberrant asymmetry. For example, many of the mothers of ALC subjects studied here also smoked during pregnancy, but defining features characteristic of individuals with fetal alcohol syndrome are not found in children of mothers who smoked but did not drink heavily during pregnancy. Other drugs may have been used by the mothers of the ALC subjects as well, but they were characterized as “alcohol dependent” and not “drug dependent” by medical, legal, and family reports. Nutritional differences during pregnancy could also be a factor in the brain dysmorphology observed, but animal studies have shown brain growth restriction in animals with neonatal alcohol exposure when nutritional intake was well controlled. Nonetheless, environmental teratogens other than alcohol cannot be completely ruled out as contributing factors to brain dysmorphology in this human sample.