In Reply

SIR: Dr. Lauterbach has made some encouraging comments on our report that patients with a history of unipolar or bipolar depression have a smaller external globus pallidus (GPe) than control subjects. He referenced additional literature on the significance of pallidal pathology in depressive states. We would like to respond by addressing issues for further research in this field.

First, there is the question of whether diagnostic dichotomy is reflected by structural brain pathology. In our study we found no volume differences in the basal ganglia between subjects with bipolar and those with unipolar disorder. If this result is confirmed by studies of larger samples, then arguably the pathomorphology of depressive states is unrelated to diagnostic polarity. This finding could be of interest because the most recently reported data suggest that the brain structure in bipolar disorder differs regionally from that in unipolar depression.^1–3 Thus, although neuroanatomic circuits might differ in bipolar and unipolar depression, some elements might be consistently present in both disorders and show similar changes. For instance, limbic-associated basal ganglia might be common elements in the neural networks underlying depression in bipolar and in unipolar illness.

As emphasized by Dr. Lauterbach, the dorsal and ventral parts of the GPe mediate different effects on thalamic-frontocortical circuits and should be given more attention in future studies on the morphology of depression. Also, structural data are needed on the involvement in primary and secondary depressive disorders of the mediodorsal thalamus and specific frontocortical regions such as the dorsolateral, orbitofrontal, and medial frontal cortex, which are considered in neuroanatomic models of depressive behavior.^4,5 Such results would allow for better ascertainment of the frontal-subcortical circuits implicated in the pathomorphology of depression. Dr. Lauterbach's proposal, that structural changes in the GPe may result in depressive states irrespective of etiology, could be extended to wider parts of these circuits. If evidence of wider involvement emerges, then factors affecting these regions in the developing or the adult brain should be taken into account when planning future strategies for the treatment and prevention of depressive disorder.