To the Editor: Recently, J. Douglas Bremner, M.D., and colleagues (1) reported smaller hippocampal volumes in patients with major depression than in nondepressed comparison subjects. However, an even more striking finding was mentioned without discussion of its interesting implications: patients with major depression displayed amygdala volumes that were 25% larger than those of healthy comparison subjects. This finding was only marginally significant, which probably was because of the difficulty of measuring the amygdala, which results in a high variability of data. However, to our knowledge, this is the fourth report of amygdala hypertrophy associated with different depressive syndromes (2–4). With this observation being made by four different research groups in four different patient populations, the question arises as to what this finding of depression-related amygdala hypertrophy means. It might be a risk factor for depressive syndromes in general, or it might reflect only a state of chronically greater emotional information processing. The fact that it has been observed in patients with temporal lobe epilepsy and dysthymia supports the latter assumption. Dr. Bremner et al. could have helped answer this question by analyzing a possible relationship between amygdala volumes and the duration and severity of depressive symptoms.

Furthermore, the question arises as to how cerebral substructures might increase in volume. One might speculate that a specific dysfunctional mode of information processing results in greater focal perfusion and larger cellular size. Alternatively, dysfunctional emotional information processing might result in cell division in cerebral substructures. Recently, there have been reports in the literature demonstrating this mechanism (5).

Dr. Bremner and colleagues pointed out the possible association between hippocampal atrophy and higher glucocorticoid levels in major depression (1). The amygdala does have a direct efferent connection with the supraoptical and paraventricular nuclei. Both are the most important nuclei of corticotropin-releasing hormone secretion in the brain. One might speculate that the amygdala is involved in the control of the neuroendocrinological stress system. By analyzing a relationship between amygdala enlargement and hippocampal volume loss, Dr. Bremner et al. could possibly find some evidence for a distinctive role of these two limbic structures in the regulation of the neuroendocrinological stress system in affective disorder.