No abstract

Hippocampal volumes do not discriminate a typical clinical population of elderly depressed patients from age-similar normal control subjects. If hippocampal dysfunction contributes to a diagnosis of syndromal depression in the elderly, such dysfunction does not appear to be regularly reflected in structural abnormalities captured by volumetric measurement as conducted. On the other hand, relationships between hippocampal volumes and clinical phenomena in depressives, but not controls, suggest potentially meaningful interactions between hippocampal structure and the expression of major depression in the elderly.

Background and Purpose-Increased frequency and severity of signal hyperintensities have been regularly reported in elderly depressed patients compared with normal subjects, however, greater neuroanatomic localization of lesions has been limited. Methods-T2-weighted MRI scans in elderly depressed patients (nϭ35) and normal comparison subjects (nϭ31) were assessed for signal hyperintensities in lateralized discrete brain regions. Results-Logistic regression revealed that left frontal deep white matter (PϽ.005) and left putaminal (PϽ.04) hyperintensities significantly predicted depressive group assignment. Conclusions-Findings suggest that greater neuroanatomic localization of hyperintensities than heretofore appreciated may relate to late-life depression. (Stroke. 1998;29:613-617.)Key Words: depression Ⅲ elderly Ⅲ leukoaraiosis Ⅲ magnetic resonance imaging G reater severity and/or frequency of subcortical signal hyperintensities on T2-weighted MRI scans have repeatedly been reported in depressed geriatric patients compared with normal elderly control subjects.1 An increasing number of recent studies implicating these lesions in geriatric depression and some of its clinical features 2-5 support observations that hyperintensities are relevant as a susceptibility factor for and/or correlate of late-life depression. 1,6 In addition to increasing age, presence of cerebrovascular disease risk factors have been most consistently associated with deep white and subcortical gray matter MR signal hyperintensities. 7,8 Furthermore, evidence from imaging/pathological correlation studies in nondepressed subjects indicates that histopathological changes indicative of cerebrovascular disease likely underlie at least the more severe deep white and subcortical gray matter hyperintensities.9,10 In fact, some investigators consider such lesions "silent cerebral infarcts."11 Taken together, these data have contributed to the renaissance 12,13 of an important conceptual entity in geriatric depression that was alluded to 35 years ago 14 : cerebrovascular disease-mediated depression. A critical foundation block supporting this construct are the high rates of depression after stroke, wherein particular relationships with left-sided frontal and basal ganglia infarcts have been demonstrated in many studies. 15 However, in most studies of geriatric depressed patients, ratings of signal hyperintensities have hardly addressed either lesion lateralization or greater anatomic specificity (eg, cortical subdivisions, subcortical gray matter differentiation into thalamus and basal ganglia components), and mixed results with regard to hyperintensity location are common. 1,16 With this in mind, the present study was undertaken with the aim of examining whether more specific neuroanatomic localization of MR signal hyperintensities in elderly depressed patients suggests particular brain-behavior relationships and supports investigations of poststroke depression that implicate more anterior and left-sided cortical and subcortical abnormalities. Subjects...