| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 Clinica de Diagnóstico Por Imagem, Ave. das Américas 4666, sala
32522649-900, Rio de Janeiro, Brazil.
2 Multi-Imagem, Rio de Janeiro, Brazil.
3 Department of Radiology, Federal University of Rio de Janeiro, Rio de Janeiro,
Brazil.
Received May 22, 2007;
accepted after revision November 18, 2007.
Address correspondence to D. B. Parente
(dbraz-parente{at}uol.com.br).
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. Seventy-nine patients with memory impairment and 16 volunteer controls participated in the study. MRI was performed with a 1.5-T system. Conventional MR images and diffusion tensor images were obtained for all participants. The diffusion tensor imaging data were postprocessed, and low b-value, fractional anisotropy, and fractional anisotropy color-coded maps were calculated. With the three maps as an anatomic reference, fractional anisotropy was measured for hippocampal formations, superior longitudinal fascicles, posterior cingulate gyri, and the splenium of the corpus callosum. Kruskal-Wallis and Steel-type multiple-comparison nonparametric tests were performed for the statistical analysis.
RESULTS. The fractional anisotropy values for the splenium of the corpus callosum, bilateral posterior cingulate gyri, and bilateral superior longitudinal fascicles of patients with mild cognitive impairment and those with probable Alzheimer's disease were significantly lower than the values of controls. No differences were found in hippocampal formations in any group. No significant difference was found in fractional anisotropy values in comparisons of mild cognitive impairment versus possible Alzheimer's disease and probable Alzheimer's disease or comparisons of probable Alzheimer's disease and possible Alzheimer's disease.
CONCLUSION. Diffusion tensor imaging is a promising technique for the evaluation of patients with probable mild cognitive impairment. Early detection of the disease expands the treatment options, increasing the likelihood of a good clinical response and enhancing the quality of life of patients and their relatives. Further studies with larger populations are needed to confirm the role of diffusion tensor imaging in the evaluation of memory impairment.
Keywords: Alzheimer's disease • corpus callosum • diffusion tensor imaging • fractional anisotropy • mild cognitive impairment • MRI • posterior cingulate gyrus • splenium • superior longitudinal fascicles
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Alzheimer's disease (AD) is the main cause of dementia among the elderly, affecting more than 30 million persons worldwide [5–7]. It primarily affects the gray matter, first involving the entorhinal cortex and then the hippocampi and temporal lobes, posterior cingulate gyri, and entire brain cortex [8, 9]. Except for nonspecific areas of high signal intensity, white matter (WM) lesions in patients with AD are not evident on conventional MRI [6, 7]. In postmortem studies, however, WM lesions associated with AD have been documented, but the cause of the WM lesions in these patients remains unclear. Wallerian degeneration, axonal damage and gliosis, and myelin breakdown have been suggested as possible pathologic mechanisms of WM lesions in AD [10, 11].
Diffusion tensor imaging (DTI) is an MRI technique with increased sensitivity to in vivo modifications in the WM microstructure and is especially indicated for diseases causing axonal damage and demyelination. DTI is based on the nonbrownian movement of water molecules, the direction of which is determined by many factors, such as cell membranes, axonal membranes, and cytoskeletal structures. The anisotropic movement of water dominates in regions with high concentrations of axons. As a result, quantitative measurement of diffusion anisotropy can be an indicator of the integrity of cerebral WM [12–14].
DTI has been increasingly studied with the aim of early detection of AD. Several investigators [13–16] have shown that damage to WM in AD is not apparent on conventional imaging. In addition, it has been found [17] that the WM lesions in AD can be detected with DTI even before the gray matter injury becomes apparent. The exact regions of alterations in diffusibility and anisotropic diffusion diverge from one study to another, some authors reporting anterior differences and others posterior or temporal changes [13, 18, 19]. In patients with AD, decreased fractional anisotropy (FA) values have been reported in the WM of the temporal [13, 20] and parietal [16, 20] lobes, hippocampus [21], superior longitudinal fascicles [14, 16], cingulate gyrus [13, 15], and corpus callosum [14, 18]. These conflicting results may be due to differences in the selection of patient populations. The purpose of this study was to evaluate the FA values of the WM tracts in patients with memory impairment with the aim of differentiating the healthy population from patients with MCI and AD.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
MRI Acquisition
All MRI studies were performed with a conventional bird-cage coil on a
1.5-T unit (Magnetom Symphony Maestro, Siemens Medical Solutions). The imaging
protocol included sagittal T1-weighted images (TR/TE, 426/11; field of view,
230 x 230 mm; matrix size, 128 x 128; slice thickness, 5 mm),
axial FLAIR images (9,500/100; field of view, 220 x 220 mm; interpolated
matrix size, 256 x 162; slice thickness, 5 mm), and oblique coronal
T2-weighted images (4,290/120; field of view, 220 x 220 mm; matrix size,
320 x 216; slice thickness, 3 mm). In addition, echo-planar
diffusion-weighted images (b = 0; 1,000 s/mm2) were obtained, and
apparent diffusion coefficient maps were calculated in the axial plane.
To fully determine the diffusion tensor, echo-planar pulse sequences including bipolar diffusion gradients in six orthogonal directions were applied (3,200/95; matrix size, 128 x 128; field of view, 230 x 230 mm; slice thickness, 5 mm; interslice gap, 1.5 mm; bandwidth, 1,346 kHz; echo-planar imaging factor, 128; echo spacing, 0.83 millisecond; flip angle, 90°; number of signals averaged, 3). To optimize the measurement of diffusion in the brain, only two b values were used (b1 = 0; b2 = 1,000 s/mm2). The slices were positioned perpendicularly to the longitudinal axis of the hippocampal formation, covering most of the brain hemispheres.
MRI Analysis
All data were transferred to an off-line workstation (Leonardo, Siemens
Medical Solutions). Two experienced neuroradiologists (3 and 10 years of
experience) blinded to clinical data prospectively read all MR images in
consensus. Conventional images were used to detect structural anomalies that
would exclude a patient from the study.
The DTI data were postprocessed with DTI Task Card software (Massachusetts General Hospital), and low b value, FA, and FA color-coded maps were calculated. To perform the region of interest (ROI)-based analysis, both neuroradiologists simultaneously displayed the three maps to clearly identify the anatomic structures. As a result, they could visibly recognize the hippocampal formations, superior longitudinal fascicles, posterior cingulate gyri, and the splenium of the corpus callosum. To achieve standardized conditions for analysis and to avoid contamination of the data by adjacent structures, five-pixel circular ROIs were individually positioned on each of these sites for the memory impairment patients and controls. Figure 1A, 1B, 1C illustrates the location of all ROIs.
|
|
|
0.05), Steel-type
multiple compari sons for nonparametric data were used to test which pairs of
groups differed. |
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The FA values in the splenium of the corpus callosum, bilateral posterior cingulate gyri, and bilateral superior longitudinal fascicles of patients with MCI and those with probable AD were significantly lower than those of controls (Table 2, Fig. 2A, 2B, 2C, 2D, 2E, 2F, 2G). The analysis of the same regions comparing patients with possible AD and controls also showed reduced anisotropy in the first group, but the difference did not reach statistical significance (p = 0.08). Finally, no significant difference was found in FA values in the comparison of patients with and those with possible AD or probable AD or the comparison of patients with probable AD and those with possible AD. Table 3 shows the median FA values and 25th–75th percentiles in the ROIs.
|
|
|
|
|
|
|
|
|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The high prevalence of disease of the WM that connects to the associative cortex suggests that wallerian degeneration secondary to cortical neuronal loss is one of the main reasons for WM lesions in AD. Vascular risk factors and ischemic events, common among the elderly population, cause changes in microvasculature and, in addition to the microvasculature modifications caused by AD itself, increase the risk of the disease. Moreover, myelin breakdown, WM rarefaction, axonal damage, and gliosis may be found in these patients [10, 11, 25]. In our study, the reduction in FA among patients with cognitive impairment might have been related to the histopathologic features.
In our series, the hippocampal FA values did not differ among all groups. Previously described [21, 26, 27] factors might have contributed to this finding. It is difficult to define the atrophic hippocampus on images with a low b value, used as an anatomic reference, which favor the partial volume effect with CSF. Hippocampal proximity to the temporal bone and mastoid cells results in an inhomogeneous magnetic field, which favors susceptibility artifacts on echo-planar images and results in a low signal-to-noise ratio. Finally, the gray matter and WM composition of the hippocampi is not easily differentiated during ROI positioning.
Our results corroborate those of previous studies in which investigators found reduced FA values in the WM tracts of patients with MCI and AD. The WM tracts most affected include the corpus callosum, temporal and parietal WM, and posterior portion of the cingulum [13, 15, 16, 18, 19]. Takahashi et al. [13] found significant reduction in FA values in the temporal subcortical WM, posterior part of the corpus callosum, and anterior and posterior cingulate bundles in patients with AD compared with the values among controls. Similarly, Naggara et al. [18] found a decrease in FA in the temporal lobe WM and splenium of the corpus callosum of patients with memory impairment. Likewise, Head et al. [19] found that AD is associated with additional vulnerability of the posterior fiber tracts of the parietal and temporal regions. In our series, patients with memory impairment had a significant decrease in FA values in the longitudinal superior fascicles, posterior cingulate gyri, and splenium of the corpus callosum, probably as a result of the involvement of selective regions connected with associated cortices.
Previous studies [15, 16] have shown differences in WM FA values in comparisons of MCI and AD patients with controls. However, those studies showed no differences in most WM tracts in comparisons of MCI patients with AD patients. Fellgiebel et al. [15] found significant differences in FA values of the posterior cingulate gyri in patients with MCI and controls and in comparisons of patients with AD and controls. Nevertheless, comparison of the FA values of patients with MCI and patients with AD versus controls showed no significant differences. Similarly, in a voxel-based analysis, Medina et al. [16] found significant reduction in anisotropy, mainly in the posterior WM regions of patients with MCI and those with AD, compared with controls. Those investigators also found no differences between patients with MCI and patients with AD. Likewise, in our study, patients with MCI and patients with AD had reduced anisotropy in several WM regions compared with controls. There was no significant difference, however, in the FA values of patients with MCI, patients with possible AD, and patients with probable AD.
Our study had several limitations. First, we did not follow the patients longitudinally to assess conversion to AD. In addition, although the ROIs were carefully positioned, we cannot exclude some degree of partial volume effect on the measurements. Furthermore, magnetic field heterogeneity and susceptibility artifacts might have impaired study of the hippocampi.
We found no differences in the FA values among patients with MCI, patients with possible AD, and patients with probable AD. However, these three groups had FA reduction in the regions first affected by AD, such as the splenium of the corpus callosum, the posterior cingulate gyri, and the superior longitudinal fascicles. These findings support the use of DTI for imaging of patients with MCI. As a result, early therapy can be implemented, increasing the chances of a positive clinical response and improving the quality of life of these patients and their relatives. Nevertheless, further studies with larger populations are needed to confirm the use of DTI in the evaluation of patients with memory impairment.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  Y. Zhang, N. Schuff, A.-T. Du, H. J. Rosen, J. H. Kramer, M. L. Gorno-Tempini, B. L. Miller, and M. W. Weiner White matter damage in frontotemporal dementia and Alzheimer's disease measured by diffusion MRI Brain, September 1, 2009; 132(9): 2579 - 2592. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. M. L. de Win, G. Jager, J. Booij, L. Reneman, T. Schilt, C. Lavini, S. D. Olabarriaga, G. J. den Heeten, and W. van den Brink Sustained effects of ecstasy on the human brain: a prospective neuroimaging study in novel users Brain, November 1, 2008; 131(11): 2936 - 2945. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
