Age-related changes of lateral ventricular width and periventricular white matter in the human brain: a diffusion tensor imaging study.

Introduction

Aging is the accumulation of multidimensional deterioration of processing of biological, psychological, and social changes with expansion over time (Bowen and Atwood, 2004; Grady, 2012). Aging-related changes are typically accompanied by decline in cognitive function, urinary control, sensory-motor function, and gait ability (Bradley et al., 1991; Bowen and Atwood, 2004; Hedden and Gabrieli, 2004; Grady, 2012; Moran et al., 2012). In addition, a number of studies have suggested changes in brain structure with normal aging, such as decrease in cortical thickness or increase in ventricular width (Blatter et al., 1995; Tang et al., 1997; Uylings and de Brabander, 2002; Preul et al., 2006; Apostolova et al., 2012). In particular, ventricular enlargement has been suggested as a structural biomarker for normal aging and progression of some illnesses, such as Alzheimer's disease (Blatter et al., 1995; Tang et al., 1997; Uylings and de Brabander, 2002; Preul et al., 2006; Apostolova et al., 2012). However, the question of how this structural change in the brain in normal elderly affects change of white matters remains a topic of interest and concern.

Diffusion tensor imaging allows for evaluation of white matter due to its ability to capture and quantify water diffusion characteristics (Mori et al., 1999; Assaf and Pasternak, 2008; Neil, 2008). In normal white matter, water molecules move relatively freely in a direction parallel to nerve fiber tracts, however, their movements are restricted across tracts, which cause diffusion anisotropy in white matter. This selective restriction of water molecule movement allows for exploration of physical changes in white matter caused by normal aging using diffusion anisotropy. Some studies using DTI have reported on changes in periventricular white matter in patients with stroke or other brain injuries (Yeo et al., 2011; Hattori et al., 2012; Jang et al., 2013). In the current study, using DTI, we investigated ventricular enlargement with normal aging, and its effect on periventricular white matter.

Subjects and Methods

Subjects

Sixty subjects (31 males, 29 females; mean age, 49.1 ± 17.6 years; range, 20–78 years) with no history of neurologic or psychiatric disease and head trauma were recruited by advertisements in a regional daily newspaper for the study. All subjects provided written informed consent prior to the study. This study was approved by the Institutional Review Board of Yeungnam University Medical Center in South Korea.

Diffusion tensor image

Diffusion tensor imaging data were acquired using a 1.5T Philips Gyroscan Intera system (Hoffman-LaRoche, Mijdrecht, the Netherlands) equipped with a Synergy-L Sensitivity Encoding head coil using a single-shot spin echo-planar imaging sequence. For each of the 32 noncollinear and noncoplanar diffusion-sensitizing gradients, we acquired 60 contiguous slices parallel to the anterior commissure-posterior commissure line. The imaging parameters used were: matrix = 128 × 128; field of view = 221 × 221 mm2; echo time = 76 ms; repetition time = 10,726 ms; sensitivity encoding factor = 2; echo planar imaging factor = 67; b = 1,000 s/mm2; number of excitations = 1; and a slice thickness of 2.3 mm.

Eddy current-induced image distortions and motion artifacts were reduced using affine multi scale two-dimensional registration (Smith et al., 2004). Diffusion tensor imaging datasets were preprocessed using the Oxford Centre for Functional Magnetic Resonance Imaging of Brain Software Library. White matter was evaluated using DTI-Studio software (CMRM, Johns Hopkins Medical Institute, Baltimore, MD, USA).

Diffusion tensor imaging parameters were estimated in the following four regions of interest in periventricular white matter, which is concerned with cognitive function, urinary control, sensory-motor function, and gait ability (Newton et al., 2006; Jang, 2009; Han et al., 2010; Hong et al., 2010; Klein et al., 2010; Tadic et al., 2010; Yeo et al., 2012); the anterior corona radiata, the posterior corona radiata, genu of the corpus callosum, and splenium of the corpus callosum (). In the anterior corona radiata, we placed a region of interest in the area defined by the medial half of the line drawn from the most lateral point of the anterior horn of the lateral ventricle to the most lateral point of the white matter horizontally and the middle third of the line drawn from the most anterior point of the anterior horn of the lateral ventricle to the middle point of the lateral ventricle vertically. In the posterior corona radiata, we placed a region of interest in an area defined by the medial half of the line drawn from the most lateral point of the posterior horn of the lateral ventricle to the most lateral point of the white matter horizontally and the middle third of the line drawn from the middle point of the lateral ventricle to the most posterior point of the posterior horn of the lateral ventricle. In the corpus callosum, we placed a region of interest on the middle third of the genu of the corpus callosum, the splenium of the corpus callosum (Yeo et al., 2011; Jang et al., 2013). We measured fractional anisotropy and apparent diffusion coefficient in each region of interest.

Figure 1

Measurement of ventricular width and regions of interest in the four regions of periventricular white matter in normal subjects.

(A) Measurement of ventricular width and calculations of relative width ratios for the ventricular body, anterior horn, and posterior horn. (B) Regions of interest in the four regions of periventricular white matter. Anterior corona radiata (CR), posterior CR, genu of the corpus callosum (CC), and splenium of the CC. Yellow blocks mean the loca-tion of region of interest for the anterior and posterior corona radiata, and blue lines mean the outline of region of interest for the definition of anterior and posterior corona radiata.

Measurement of ventricular width

We used the ventricular body index, anterior horn index, and posterior horn index, which could estimate the effects of hydrocephalus on the lateral ventricle (Jang et al., 2013) (). The relative width ratios for the ventricular body index, anterior horn index, and posterior horn index were defined as follows: the ratio between the maximum distance of the ventricular walls at the body of the ventricle level and the maximum width of the brain on the axial image, the ratio between the maximum widths of tips on both anterior horns and maximum width of the brain at the same level on the coronal image, and the ratio between the maximum widths of tips on both posterior horns and the maximum width of the brain at the same level on the coronal image were defined as the ventricular body index, anterior horn index, and posterior horn index, respectively.

Statistical analysis

Spearman's correlation test was used for determination of correlation between age, diffusion tensor imaging parameters of each region of interest, and each index of ventricular width. Bonferroni correction was used to correct for multiple comparisons of correlation test. Software (version 15.0; SPSS, Chicago, IL, USA) was used in performance of statistical analyses, and statistical significance was set at P < 0.001.

Results

Correlation between ventricular parameters and age

Regarding the ventricular parameters, we found significant correlation with age in all indexes for ventricular width. In particular, ventricular body index (P < 0.001) and anterior horn index (P < 0.001) showed strong positive correlation with age. In addition, the posterior horn index showed moderate positive correlation with age (P < 0.001; and ).

Table 1

Correlation between each ventricular index and age

Figure 2

Scatter diagram for correlation test between ventricular index and age.

Spearman's correlation test was used for determination of correlation between ventricular index and age. *P < 0.001. CR: Corona radiate.

Correlation between diffusion tensor imaging parameters and age

In the anterior corona radiata, we observed weak to moderate positive correlation between fractional anisotropy and age (P < 0.001), and apparent diffusion coefficient and age (P < 0.001). In addition, posterior corona radiata also showed moderate positive correlation between fractional anisotropy and age (P < 0.001). In contrast, apparent diffusion coefficient value of posterior corona radiata and diffusion tensor imaging parameters of the genu and splenium of the corpus callosum did not show significant correlation with age (P > 0.001) (; ).

Table 2

Correlation between diffusion tensor imaging parameters in each region of white matter and age

Figure 3

Scatter diagram for correlation test between diffusion tensor imaging parameters and age.

Spearman's correlation test was used for determination of correlation between diffusion tensor imaging parameters of each region of interest and age. *P < 0.001. VBI: Ventricular body index; AHI: anterior horn index; PHI: posterior horn index.

Correlation between diffusion tensor imaging parameters in white matter and ventricular parameters

In terms of correlation between ventricular parameters and regions of white matter, we found mild to moderate positive correlation between ventricular parameters and fractional anisotropy value of anterior corona radiata (P < 0.001), and fractional anisotropy value of posterior corona radiata (P < 0.001). In addition, apparent diffusion coefficient value of anterior corona radiata also showed mild to moderate positive correlation with each ventricular parameter (P < 0.001). On the other hand, apparent diffusion coefficient value of posterior corona radiata, and fractional anisotropy and apparent diffusion coefficient value of the genu and splenium of the corpus callosum did not show significant correlation with ventricular parameters (P > 0.001; ().

Table 3

Correlation between diffusion tensor imaging parameters in each region of white matter and ventricular index

Discussion

In the current study, we investigated the effects of ventricular enlargement in normal aging on periventricular white matter. We found that the fractional anisotropy and apparent diffusion coefficient values of the anterior corona radiata and posterior corona radiata showed positive correlation with age and its related ventricular enlargement. Fractional anisotropy value, which is the most widely used diffusion tensor imaging parameter, represents the degree of directionality of the microstructures (e.g., axons, myelin, microtubules), and the apparent diffusion coefficient value indicates the magnitude of water diffusion (Mori et al., 1999; Assaf and Pasternak, 2008; Neil, 2008). Therefore, mechanical pressure by ventricular enlargement appeared to cause higher packing of fibers and increased fiber density per unit area, resulting in increased fractional anisotropy and apparent diffusion coefficient values of anterior and posterior corona radiata. According to our result, ventricular enlargement with normal aging appears to have the greatest effect on anterior and posterior periventricular white matter. Consequently, this characteristic might be vulnerable to pressure by ventricular enlargement.

Decline of cognitive function, urinary control, sensory-motor function, and gait ability are well-known as aging related changes of normal elderly (Bradley et al., 1991; Bowen and Atwood, 2004; Hedden and Gabrieli, 2004; Grady, 2012; Moran et al., 2012). Many neural tracts from the frontal lobe, such as thalamocortical tracts between the thalamus and the prefrontal cortex, corticoreticulospinal tract, and the neural tract for urinary control are known to pass through the anterior corona radiata (Newton et al., 2006; Klein et al., 2010; Tadic et al., 2010; Yeo et al., 2012). By contrast, somatosensory and motor neural tracts, such as the corticospinal tract, spinothalamic tract, and medial lemniscus are known to ascend or descend through the posterior corona radiata (Jang, 2009; Han et al., 2010; Hong et al., 2010). In the current study, although the discrimination of neural tract on the anterior and posterior corona radiata was not available using region of interest based measure technique, we think that our results, which showed compression of the anterior corona radiata by ventricular enlargement, might be related to the above clinical manifestations of frontal lobe function. In addition, decline of fine motor function and sensory perception might be related to compression in the posterior corona radiata by ventricular enlargement. These results appear to be compatible with the results of the previous studies, which showed changing of white matter with cognitive impairments in aging (Vernooij et al., 2009; Ziegler et al., 2010; Bennett et al., 2011; Borghesani et al., 2013). By contrast, genu and splenium of the corpus callosum, which facilitate inter-hemispheric communication, showed correlation with some cognitive and sensory function, respectively, however, were not responsible for these functions (Bogousslavsky, 2005; Jea et al., 2008; Jang et al., 2012). In addition, the corpus callosum is a colossal commissure between cerebral hemispheres, consequently (Bogousslavsky, 2005; Jea et al., 2008), genu and splenium of the corpus callosum might be less affected by age-related ventricular enlargement.

Several previous studies have reported ventricular enlargement with Alzheimer's disease and hydrocephalus due to stroke or other brain injury (Uylings and de Brabander, 2002; Yeo et al., 2011; Hattori et al., 2012; Jang et al., 2013). On the other hand, even without exact estimation of change of periventricular white matter, only a few studies have demonstrated ventricular enlargement with normal aging (Preul et al., 2006; Apostolova et al., 2012). Using MRI and 3D multi slice modified driven equilibrium Fourier transform, Preul et al. (2006) evaluated cortical thickness and ventricular width throughout normal aging. They demonstrated that an increase in ventricular width occurs in normal aging, accompanied by a decrease in cortical thickness. Recently, Apostolova et al. (2012) reported an association of normal aging with hippocampal atrophy and ventricular enlargement. However, they suggested that progression of ventricular enlargement over time was relatively slow compared to patients with Alzheimer's disease.

In conclusion, we observed ventricular enlargement with normal aging and its related changes in periventricular white matter. We believe that our results are likely to be useful for diagnosis and management of normal aging. However, some limitations to this study should be considered in the interpretation of the results. First, current region of interest approach technique is operator-dependent and it is difficult to full reflect the underlying neural fiber. Therefore, region of interest approach may underestimate the fiber tracts. Second, we could not evaluate declination of cognitive functions and motor functions concerned with aging. Additional studies are needed to report clinical correlation between changes of periventricular white matter and normal aging.