Ventral striatal volume is associated with cognitive decline in older people: a population based MR-study

Abstract

Striatal degeneration may contribute to cognitive impairment in older people. Here, we examine the relation of degeneration of the striatum and substructures to cognitive decline and dementia in subjects with a wide range of cognitive function. Data are from the prospective community-based Honolulu Asia Aging Study of Japanese American men born 1900-1919. Brain magnetic resonance imaging (MRI) (1.5 T) was acquired on a stratified subsample (n = 477) that included four groups defined by cognitive status relative to the scan date: subjects without dementia (n = 347), subjects identified as demented 2-3 years before brain scanning (n = 30), at the time of scanning (n = 58), and 3-5 years after scanning (n = 42). Volumes of the striatum, including the accumbens, putamen, and caudate nucleus were automatically estimated from T1 MR images. Global cognitive function was measured with the cognitive ability screening instrument (CASI), at four examinations spanning an 8-year interval. Trajectories of cognitive decline were estimated for each quartile of striatal volume using mixed models, controlling for demographic variables, measures of cerebro-vascular damage, global brain atrophy, and hippocampal volume. Diagnosis of dementia before, during, and after brain scanning was associated with smaller volumes of n. accumbens and putamen, but not with caudate nucleus volume. Subjects in the lowest quartile of n. accumbens volume, both in the total sample and in the subjects not diagnosed with dementia during the study, had a significantly (p < 0.0001) steeper decline in cognitive performance compared with those in the highest quartile. In conclusion, volumes of the n. accumbens and putamen are closely associated with the occurrence of dementia and n. accumbens volume predicts cognitive decline in older people. These associations were found independent of the magnitude of other pivotal markers of cognitive decline, i.e. cerebro-vascular damage and hippocampal volume. The present study suggests a role for the ventral striatum in the development of clinical dementia.

Figure 1. Anatomical and functional division of striatum

Presented is a 3D model of the striatum, based on the segmentation of those voxels that were labeled striatum in > 20% of the segmentation masks. Model was created with GAMEs (Ferrarini et al., 2007) and displayed here for explanational purposes.

Figure 2. Predicted CASI-score over time per quartile of striatal volume

X-axis: Age = Age of observation Y-axis: CASI Score = Predicted mean CASI (Cognitive Ability Screening Instrument) score over 8-year interval, adjusted for age, age at baseline, educational level, ICV, CES-D score, lacunes grade, presence of cerebral infarcts, WMH grade, bi-frontal distance/innertable distance, ApoE 4 allele, hippocampal volume quartiles. Spaghetti plots of a random sample of 10 subjects per striatal volume quartile are displayed.

Figure 3. Predicted CASI-score over time per quartile of striatal volume (non-demented people only)

X-axis: Age = Age of observation Y-axis: CASI Score = Predicted mean CASI (Cognitive Ability Screening Instrument) score over 8-year interval, adjusted for age, age at baseline, educational level, ICV, CES-D score, lacunes grade, presence of cerebral infarcts, WMH grade, bi-frontal distance/innertable distance, ApoE 4 allele, hippocampal volume quartiles. Spaghetti plots of a random sample of 10 subjects per striatal volume quartile are displayed.