Correlations among Brain Gray Matter Volumes, Age, Gender, and Hemisphere in Healthy Individuals

Yasuyuki Taki,Ryuta Kawashima,Benjamin Thyreau,Shigeo Kinomura,Kazunori Sato,Hiroshi Fukuda,Ryoi Goto,Yong He

doi:10.1371/journal.pone.0022734

Abstract

To determine the relationship between age and gray matter structure and how interactions between gender and hemisphere impact this relationship, we examined correlations between global or regional gray matter volume and age, including interactions of gender and hemisphere, using a general linear model with voxel-based and region-of-interest analyses. Brain magnetic resonance images were collected from 1460 healthy individuals aged 20–69 years; the images were linearly normalized and segmented and restored to native space for analysis of global gray matter volume. Linearly normalized images were then non-linearly normalized and smoothed for analysis of regional gray matter volume. Analysis of global gray matter volume revealed a significant negative correlation between gray matter ratio (gray matter volume divided by intracranial volume) and age in both genders, and a significant interaction effect of age × gender on the gray matter ratio. In analyzing regional gray matter volume, the gray matter volume of all regions showed significant main effects of age, and most regions, with the exception of several including the inferior parietal lobule, showed a significant age × gender interaction. Additionally, the inferior temporal gyrus showed a significant age × gender × hemisphere interaction. No regional volumes showed significant age × hemisphere interactions. Our study may contribute to clarifying the mechanism(s) of normal brain aging in each brain region.

Highlights

IntroductionPrevious studies using magnetic resonance (MR) imaging have demonstrated a negative correlation between global or regional gray matter volume and age by applying cross sectional [1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16] and longitudinal designs [10,17,18,19,20,21,22,23]
Applying Akaike information criterion (AIC), we determined that the correlation between age and gray matter ratio was significant and best fit by a first-order polynomial function in men (R2 = 0.643; p,0.001) and a secondorder polynomial function in women (R2 = 0.624; p,0.001)
Because most regions showed significant age 6 gender interactions and the inferior frontal gyrus showed a significant age 6 gender 6 hemisphere interaction, we examined these interactions by dividing the subjects into decade-long age groups and analyzed the correlation between age group and regional gray matter volume for each brain region in men and women

Summary

Introduction

Previous studies using magnetic resonance (MR) imaging have demonstrated a negative correlation between global or regional gray matter volume and age by applying cross sectional [1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16] and longitudinal designs [10,17,18,19,20,21,22,23]. Several studies have focused on the effect of age 6 gender interactions on regional gray matter volume, the findings of these studies were restricted to specific areas, such as the cortical areas [31], thalamus [11], 13 manually traced cerebral regions of interest [32], or to limited age groups, such as elderly subjects [16]. Prior research has not clarified the effect of age 6gender interactions on the global and regional gray matter volumes of each gyral structure in the cerebrum, on deep gray matter structures, such as the thalamus and caudate nucleus, or on each lobe of the cerebellum. No reported study has revealed age 6 hemisphere or age 6 gender 6 hemisphere interaction effects on the regional gray matter volume of each gyral structure of the cerebrum, on such deep gray matter structures as the thalamus and caudate nucleus, and on each lobe of the cerebellum

Objectives

Methods

Results

Conclusion