To the Editor: We noted with interest the recent article by Eileen P. Ahearn, M.D., Ph.D., and colleagues (1). The authors found a high prevalence of white and gray matter abnormalities on cranial magnetic resonance imaging (MRI) studies in a family with a strong history of mood disorders. The ability of this pedigree to illuminate the genetics of bipolar disorder, however, is limited by several considerations.

1. The significance of the apparently high occurrence of radiological abnormalities in the pedigree is difficult to evaluate in the absence of a comparison group, particularly since over one-half of the abnormal scans showed only one or two MRI lesions of 3 mm or less. Six of the nine patients with bipolar disorder and MRI lesions were over the age of 50. Subtle MRI changes are common in older individuals, even in the absence of psychiatric symptoms (2).

2. The data presented in table 1 in the article by Dr. Ahearn and colleagues show a lack of association between MRI lesions and the diagnosis of mood disorder (χ²=1.22, df=1, n.s.), suggesting that radiological abnormalities and psychiatric phenomena may be unrelated. It is not clear, therefore, how the MRI changes could serve as a biological marker for bipolar disorder.

3. Bipolar disorder, like other psychiatric illnesses, represents a syndrome that can be produced by many different underlying processes. A wide variety of inherited metabolic disorders can appear as primary psychiatric illnesses, including adrenoleukodystrophy, Tay-Sachs disease, Huntington’s disease, Wilson’s disease, metachromatic leukodystrophy, and mitochondrial diseases (3–6). The ability of many different pathobiological pathways to give rise to similar psychiatric syndromes underscores the difficulty of trying to relate bipolar disorder to a single gene abnormality, such as the chromosome 19 mutation associated with cerebral autosomal-dominant arteriopathy with subcortical infarcts and leukoencephalopathy.

We agree with the authors that patients with psychiatric illnesses and abnormal neuroimaging findings should be investigated for underlying etiologies, including those listed previously. This line of investigation is likely to yield crucial information for the understanding of the pathophysiology of psychiatric disorders.