Regional Specificity of Brain Atrophy in Huntington’s Disease G. M. Halliday, D.A. McRitchie, V. Macdonald, K. L. Double, R. J. Trent,* and E. McCusker† Prince of Wales Medical Research Institute, High Street, Randwick, 2031, *Department of Molecular Genetics, Royal Prince Alfred Hospital, Camperdown, 2050; and †Department of Neurology, Westmead Hospital, 2145, N.S.W., Australia Received June 12, 1998; accepted July 27, 1998 The present study analyzes the relationship between cortical and subcortical brain volumes in patients with Huntington’s disease. The brains of seven patients with a clinical diagnosis and positive family history of Huntington’s disease and 12 controls were collected at autopsy with consent from relatives. Detailed clinical assessments were available for all study subjects with genotype confirmation for patients with Huntington’s disease. Volume analysis of the brain on serial 3-mm coronal slices was performed as previously described. All patients with Huntington’s disease exhibited signifi- cant brain atrophy resulting from volume reductions in both cortical and subcortical grey matter. Atrophy of the cortex was relatively uniform, although the medial temporal lobe structures were spared. The caudate nucleus and putamen were strikingly reduced in all cases and this atrophy correlated with the sever- ity of cortical atrophy, suggesting an associated dis- ease process. The rate of cortical but not subcortical atrophy correlated with CAG repeat numbers. Loss of frontal white matter correlated with both cortical and striatal atrophy. Age of onset of chorea correlated with the amount of subcortical atrophy, while duration of chorea correlated negatively with atrophy of the white matter. These results suggest a more widespread and global disease process in patients with Huntington’s disease.  1998 Academic Press Key Words: chorea; cortex; basal ganglia; dementia; Huntington’s disease; volume. INTRODUCTION Huntington’s disease (HD) is an autosomal-dominant disorder with marked neuropathology in the striatum and globus pallidus and characteristic abnormalities of movement, cognition, and behavior (35, 42). In vivo studies of subcortical structures in HD have shown that the age of symptom onset and trinucleotide repeat length correlates with the amount of atrophy in the caudate nucleus (6, 19, 42, 43) and a corresponding reduction in blood flow (2, 23, 25, 36, 37, 44). Caudate atrophy correlates with a measured decline in Mini- Mental State Examination scores (7, 23), but not with the severity or duration of neurological symptoms (6). However, atrophy of the putamen appears to be greater than that in the caudate nucleus and correlates with neurological indices of disease severity (23), but not with age of onset or the length of the trinucleotide repeat (6). Further, both hypometabolism and atrophy of the caudate and putamen can be measured prior to the onset of symptoms in presymptomatic gene carriers (2, 5, 24) with the rate of disease progression in the striatum governed by the number of CAG repeats (faster rate with higher repeat numbers; 3, 43). These results highlight the regional specificity of deficits observed in patients with HD and the early detection of abnormalities in subcortical structures prior to signifi- cant symptomatology. The focus of research on subcortical structures is not surprising considering that the severity of basal gan- glia pathology forms the basis for current pathological diagnostic grades for HD (grade 0, normal striatum, to grade 4, severe striatal pathology; 52, 53). However, neuroimaging studies have demonstrated that the dura- tion and severity of HD symptoms correlates with decreases in glucose metabolism in specific cortical regions, in contrast to the relatively invariable striatal hypometabolism in HD (23, 38). In addition, significant cortical atrophy has been reported in patients with HD (4, 7, 33). These in vivo findings have been confirmed and extended in three neuropathological studies of standardized cerebral areas (14, 41, 48), where the severity of regional atrophy correlated with the neuro- pathological grade of the disease. Studies analyzing cortical neurons also report significant cell loss and/or neuritic formation in patients with HD (11–13, 18, 30, 49, 50), in addition to a widespread distribution of intranuclear inclusions (9, 15). These inclusions concen- trate in the cortex where their density correlates with the number of CAG repeats (9). The pathological find- ings are supported by recent in vivo imaging showing that frontal degeneration in HD correlates with cogni- tive impairment (4, 7) and the clinical severity of the disease (4, 22). It should be noted that degeneration of the frontal cortex is absent in asymptomatic gene carriers (22). These studies suggest that the degenera- EXPERIMENTAL NEUROLOGY 154, 663–672 (1998) ARTICLE NO. EN986919 663 0014-4886/98 $25.00 Copyright  1998 by Academic Press All rights of reproduction in any form reserved.