Mood-state effects on amygdala volume in bipolar disorder
J Affect Disord. 2012 Aug;139(3):298-301
Authors: Foland-Ross LC, Brooks JO, Mintz J, Bartzokis G, Townsend J, Thompson PM, Altshuler LL
Abstract
BACKGROUND:
Prior structural neuroimaging studies of the amygdala in patients with bipolar disorder have reported higher or lower volumes, or no difference relative to healthy controls. These inconsistent findings may have resulted from combining subjects in different mood states. The prefrontal cortex has recently been reported to have a lower volume in depressed versus euthymic bipolar patients. Here we examined whether similar mood state-dependent volumetric differences are detectable in the amygdala.
METHODS:Forty subjects, including 28 with bipolar disorder type I (12 depressed and 16 euthymic), and 12 healthy comparison subjects were scanned on a 3T magnetic resonance image (MRI) scanner. Amygdala volumes were manually traced and compared across subject groups, adjusting for sex and total brain volume.
RESULTS:Statistical analyses found a significant effect of mood state and hemisphere on amygdala volume. Subsequent comparisons revealed that amygdala volumes were significantly lower in the depressed bipolar group compared to both the euthymic bipolar (p=0.005) and healthy control (p=0.043) groups.
LIMITATIONS:
Our study was cross-sectional and some patients were medicated.
CONCLUSIONS:
Our results suggest that mood state influences amygdala volume in subjects with bipolar disorder. Future studies that replicate these findings in unmedicated patient samples scanned longitudinally are needed.
PMID: 22521854
Depression often has an underlying pathophysiology which includes a pro-inflammatory state and impaired metabolism. Thus, it is not surprising that activity and cellular volume is reduced in such a state.
This study, however, has the limitation of using imaging studies to determine behavioral consequences of structure. Such studies have highly variable and often irreproducible results. It would have been interesting if they could a priori create two groups – each with half bipolar depressed and half controls – to see if these two groups can result in one having a significant difference. In some studies, doing this results in large differences between two groups with similar makeup. This shows how much in imaging studies is left up to chance occurrence or error. For example, there was one MRI study showing a significant difference in structure between men and women. But when the same study was re-done, this time with two groups with each containing half men and half women, the same magnitude of difference was found between the two groups. This lead the researcher to recant the initial conclusion that the MRI results were the result of male-female differences.