We detected reduced volumes of the amygdala bilaterally in adolescents and young adults with BD compared with HC participants, both at the time of initial scan and after a time interval of approximately 2 years. This finding is consistent with previous reports, from cross-sectional studies, of reduced amygdala volume in adolescents and adults with BD (12
). It extends our previous cross-sectional study (13
) by suggesting that morphological disturbances of the amygdala are relatively stable over a 2-year period in adolescents and young adults with BD.
We cannot infer the disturbances in specific cellular elements that might have contributed to the macroscopic abnormalities in volume detected in these data. Reduced volumes could derive from any number of cellular disturbances, including reductions in the numbers or sizes of neurons or their processes, alterations in packing density, or disturbances in the numbers or processes of glial cells (24
). Indeed, recent reports suggest the presence of reduced glial cell numbers, particularly reduced numbers of oligodendrocytes, in the amygdala of adults with mood disorders (24
Possible behavioral consequences of disturbances in the morphology of the amygdala in persons with BD may include deficits in adaptive responses to emotional and social stimuli (28
). Similar to individuals with amygdala lesions (33
), individuals with BD are impaired in their ability to recognize specific facial emotions (3
), and preliminary evidence suggests that these deficits are associated with abnormal functional activity in the amygdala of adults with BD (3
). While abnormal functioning of basal ganglia structures has been reported in adolescents with BD (35
), functional abnormalities specifically in the amygdala of adolescents with BD have not been reported. Therefore, whether these early structural abnormalities of the amygdala are associated with functional disturbances during adolescence in persons with BD is currently unknown.
The detection of structural abnormalities in the amygdala during adolescence suggests that these morphological characteristics may serve as potential early biological markers of this disorder. However, studies earlier in childhood are needed to determine precisely when in ontogeny these abnormalities arise and whether indeed they are present at the time of onset of the illness or whether instead they are consequences of either chronic or recurrent illness (38
) or of medication exposure. The presence of abnormalities in the amygdala during adolescence, a period of intense dynamic development of the orbitofrontal cortices and other regions connected to the amygdala that have been implicated in the pathophysiology of BD (39
), raises the possibility that targeting those abnormalities with treatments could help in preventing the progression of abnormalities in cortico-limbic circuits (37
). We did not detect medication effects on amygdala volumes, although this may be attributable to insufficient statistical power. Preliminary evidence does suggest that mood-stabilizing medications are associated with increased cortical gray matter in BD (45
) and that lithium and valproic acid treatment are associated with increased amygdala volume in pediatric BD (16
). This suggests that future neuroimaging studies, combined with systematic treatment of pediatric BD patients, are warranted.
The findings differ from those of a previous study that noted abnormal increases in amygdala volume with age in adolescents and young adults with BD (15
). However, differences between the two studies include the cross-sectional design and larger number of BD participants taking lithium or valproic acid in the previous study. The results of the present study, taken in the context of variable findings in volumes of the amygdala in adults with BD (9
), suggest that reduced amygdala volumes could represent an early-onset morphological subtype of BD. The ability to generalize from the pediatric BD sample studied may be limited by the inclusion criteria requiring that participants meet full DSM-IV criteria for acute BD episodes, the presence of rapid-cycling subtype in a majority of BD participants (a characteristic of only a subset of patients with BD and one observed more commonly in children than in adults) and the relatively low rate of other comorbidities, such as ADHD, that have been reported to occur frequently in adolescents with BD (47
). Heterogeneous features, such as a range in age from 10 to 21 during which time significant neurodevelopmental changes occur as well as a range of comorbid illnesses and medications in some of the subjects, could have diluted the power to detect effects that could be detected in a larger, more homogeneous sample. It is possible that one of these features, such as a particular medication contributed to the group differences observed.
In summary, this study provides preliminary evidence, in a limited subject sample, that volume reductions in the amygdala are stable over an interval of approximately 2 years in adolescents and young adults with BD. Larger longitudinal studies that include prepubescent subjects, as well as children at risk for developing the disorder, will be important for understanding the role of amygdala abnormalities in the development of BD.