BACKGROUND: Youths with bipolar disorder are ideal for studying illness pathophysiology given their early presentation, lack of extended treatment, and high genetic loading. Adult bipolar disorder MRI studies have focused increasingly on limbic structures and the thalamus because of their role in mood and cognition. On the basis of adult studies, the authors hypothesized a priori that youths with bipolar disorder would have amygdalar, hippocampal, and thalamic volume abnormalities. METHOD: Forty-three youths 6-16 years of age with DSM-IV bipolar disorder (23 male, 20 female) and 20 healthy comparison subjects (12 male, eight female) similar in age and sex underwent structured and clinical interviews, neurological examination, and cognitive testing. Differences in limbic and thalamic brain volumes, on the logarithmic scale, were tested using a two-way (diagnosis and sex) univariate analysis of variance, with total cerebral volume and age controlled. RESULTS: The subjects with bipolar disorder had smaller hippocampal volumes. Further analysis revealed that this effect was driven predominantly by the female bipolar disorder subjects. In addition, both male and female youths with bipolar disorder had significantly smaller cerebral volumes. No significant hemispheric effects were seen. CONCLUSIONS: These findings support the hypothesis that the limbic system, in particular the hippocampus, may be involved in the pathophysiology of pediatric bipolar disorder. While this report may represent the largest MRI study of pediatric bipolar disorder to date, more work is needed to confirm these findings and to determine if they are unique to pediatric bipolar disorder.