Diffusion tensor imaging reveals brain abnormality patterns in a patient with semantic dementia☆○
Qiang Yuan1, Qin Chen1, Hehan Tang2, Xintong Wu1, Qiyong Gong○2, 3, Dong Zhou1, Ling Zou2

1Department of Neurology, West China Hospital of Sichuan University, Chengdu  610041, Sichuan Province, China
2Huaxi Magnetic Resonance Research Center, Department of Radiology, West China Hospital of Sichuan University, Chengdu  610041, Sichuan Province, China
3Division of Medical Imaging, Faculty of Medicine, University of Liverpool, Liverpool, L69 3GB, UK

Qiang Yuan☆, Doctor, Lecturer, Department of Neurology, West China Hospital of Sichuan University, Chengdu   610041, Sichuan Province, China  

Qiang Yuan and Qin Chen contributed equally to this work.

Corresponding author: Ling Zou, Doctor, Lecturer, Huaxi Magnetic Resonance Research Center, Department of Radiology, West China Hospital of Sichuan University, Chengdu   610041, Sichuan Province, China   zl_scu@yahoo.com.cn

Abstract
BACKGROUND: Previous studies regarding primary progressive aphasia (PPA) have focused on progressive, non-fluent aphasia. Little information is available with regard to the use of diffusion tensor imaging compared with conventional magnetic resonance imaging for the detection of subtle structural abnormalities.
OBJECTIVE: To investigate and localize brain abnormalities in a Chinese patient with semantic dementia.
DESIGN, TIME AND SETTING: A concurrent, non-randomized, case-controlled, neuroimaging, clinical trial was performed at the Department of Radiology, West China Hospital of Sichuan Uni-versity in March 2009.
PARTICIPANTS: One 75-year-old male patient, who was diagnosed with semantic dementia, and 21 age- and gender-matched healthy volunteers were recruited for the study.
METHODS: Diffusion tensor imaging was used to determine mean diffusion (MD) and fractional anisotropy (FA) in the brains of the patient and the 21 healthy subjects. Voxel-based analysis of MD and FA values was performed using statistical parametric mapping.
MAIN OUTCOME MEASURES: MD and FA value maps differences between patient and controls.
RESULTS: MD was significantly increased in both cerebra, but was predominant on the left side and expanded to outside of the language-related region. Reduced MD was not detected in any of the brains. FA was shown to be decreased in the corpus callosum, but was increased in the basal ganglia.
CONCLUSION: The present study provided clear in vivo magnetic imaging evidence of diffuse brain involvement in semantic dementia. Increases in MD were greater than in FA when brain diffusion alterations were detected, which suggested that MD could be a better marker of disease pro-gression.
Key Words: semantic dementia; primary progressive aphasia; diffusion tensor imaging; mean diffusion; fractional anisotropy; voxel-based analysis

INTRODUCTION
  
Primary progressive aphasia (PPA), an uncommon form of progressive dementia, is characterized by an isolated decline in language functions[1] and during initial stages of the disease, other cognitive domains are relatively spared. Clinical presentations of PPA are classified into distinct variants, such as logopenic aphasia, progressive non-fluent aphasia, or semantic dementia based on language phenotypes[2]. Recently, structural and functional PPA patterns have been revealed by neuroimaging techniques, such as magnetic resonance imaging (MRI), single-photon emission computed tomography (SPECT), and positron emission tomography (PET)[3]. However, the majority of these studies have focused on progressive non-fluent aphasia. Little is known about semantic dementia[4].
Diffusion tensor imaging, a relatively new MRI method, is used to detect subtle structural abnormalities that are not easily revealed by conventional MRI[5-6]. Voxel-based analysis is a useful diagnostic tool for conditions where regions and distributions of abnormalities remain unknown[5-8], as well as for the detection of individual brain characteristics in a single patient compared with a group of healthy controls[7, 9-10].
Therefore, diffusion tensor imaging and voxel-based analysis were utilized in the present study to determine in vivo brain abnormality patterns in a patient with semantic dementia.

SUBJECTS AND METHODS

Design
A concurrent, non-randomized, 
case-controlled, neuroimaging, clinical trial.
Time and setting
This study was performed at the Department of Radiology, West China Hospital of Sichuan University in March 2009.
Subjects
A 75-year-old Chinese male was admitted to West China Hospital in March 2009 for deteriorating linguistic capacities, manifested primarily as difficulty in naming objects and mild problems with writing single words over the past 5 years. Comprehension and communication abilities were preserved. Three years prior to admission, comprehension and verbal-related memory were mildly impaired. One year prior to admission, the patient presented with disorientation and marked dyscalculia. According to the patient’s relatives, he did not exhibit difficulties in performing daily activities, including dressing, eating, and housekeeping. The family history was unremarkable, and physical examinations were normal. The patient underwent a neuropsychological test, and his linguistic skills were assessed using the Aphasia Battery of Chinese, Western Aphasia Battery[11] and Boston Diagnostic Aphasia Examination[11]. Conversational speech was fluent and accompanied sensory aphasia and difficulties in naming color and objects, but no problems with identifying objects following name presentation, which demonstrated an unaffected, figural, episodic memory. Although repetition ability was preserved, the patient exhibited difficulties with auditory comprehension, while sparing visual comprehension. The Mini-Mental State Examination[12] score was 16, and the Wechsler Memory Scale score was < 51. Executive functions were mildly impaired, as well as problem-solving and semantic knowledge. Short-term memory and parietal functions were intact, and conventional MRI revealed mild atrophy in the left frontal and temporal lobes.
Diffusion tensor imaging images from 21 normal controls were selected from the normal image database established prior to the study. All normal controls were healthy volunteers [10 males and 11 females, mean age (63.1 ± 9.27) years, and a range of 50–76 years] with normal conventional MRI results and no history of neurological disorders. Structural T1- and T2-weighted images were systematically reviewed to exclude potential abnormalities in control subjects. The study was approved by the local ethics committee. Informed written consent was obtained from the patient’s relatives and healthy volunteers.
Methods
MRI acquisition
MRI was performed in the patient and control groups. diffusion tensor imaging data were acquired using a 3.0-T magnetic resonance scanner (GE, Excite, Milwaukee, USA) and a spin-echo echo-planar imaging (SE-EPI) sequence. Scanning parameters are as follows: 70.8-ms echo-time, 10 000-ms repetition time, 3.0 mm slice thickness with no gap, 24-cm2 field of view, and 128 x 128 acquisition matrix. A single b-value of    1 000 s/mm2 was applied in 15 non-collinear directions with 42 slice positions. In addition, a reference image with no diffusion weighting was also obtained (b0 image).
Image processing and exploratory whole-brain analysis
Following image acquisition, data were transferred to an independent workstation (Huaxi Magnetic Resonance Research Center of the West China Hospital) for calculating diffusion tensor imaging indices. Maps of mean diffusivity (MD) and fractional anisotropy (FA) were calculated for each voxel using the diffusion-weighted images according to a previously described method[13]. To determine voxel-based analysis between patient and controls, diffusion maps from each subject were first normalized to a standard EPI template provided by SPM2 (http://www.fil.ion.ucl.ac.uk/spm/software/) and then smoothed to 6-mm full-width with a half-maximum, isotropic, Gaussian kernel. For statistical analysis, the patient was individually compared to the 21 control subjects using analysis of covariance, with age and gender as confounding variables[14-15]. Significant differences in diffusivity or anisotropy were detected at a threshold of P < 0.001 (corrected for multiple comparisons with P < 0.05).

RESULTS
 
Voxel-based whole-brain analysis revealed diffusion tensor-detected regions with significantly increased MD predominantly in the left cerebrum, as well as superior temporal lobes, insulae, anterior thalami, left anterior cingulum, parahippocampi and hippocampi, cerebellum, corpus callosum, and subcortical and paraventricular white matter (Figure 1). Reduced MD was not detected.

Significantly reduced FA was revealed in the corpus callosum, with increased FA in the basal ganglia (Figure 2).

DISCUSSION 

During the initial stage of disease, the patient described in the present study exhibited an acquired, progressive inability to name or comprehend common concepts, with little or no distortion of phonological and syntactic language aspects, as well as relative sparing of other cognition characteristics, such as episodic memory, nonverbal problem-solving, and perceptual and visual-spatial skills. Semantic dementia diagnosis was made based on criteria proposed by Mesulam[16]. However, studies have shown that Alzheimer’s disease has a wide presentation and overlaps primary progressive aphasia[17]. Our patient was not thought to present with Alzheimer’s disease, because of the predominant language defect that persisted for 2 years prior to development of cognitive deficits. Recent memory decline, especially episodic memory decline, is an initial symptom for Alzheimer’s disease. Lewy body dementia is characterized by visual hallucinations, Parkinsonism, and fluctuating cognitive symptoms. In frontal temporal dementia, early changes include obvious personality changes and decreased social behavior. In progressive non-fluent aphasia, word-finding difficulty, relatively reserved language comprehension, and distortion of phonological and syntactic language aspects are the primary disease manifestations[18].
Recent PET and SPECT studies have shown decreased metabolism in the left temporal lobe in PPA patients. However, in semantic dementia patients, anterior brain regions exhibit decreased metabolism, especially in the temporal lobes (predominately the left side), with no hippocampus involvement[19].
In the present study, MD was increased in brain regions that were more diffuse than previously reported[20] and not limited to language-related areas. The diffuse, abnormal areas were predominant in the left cerebrum and most significant in the left superior temporal lobe. This was in accordance with previous studies, demonstrating that left anterior temporal hypometabolism is a predominant characteristic in semantic dementia. In addition, results revealed involvement of the right cerebrum (temporal lobe, insula, anterior thalamus, parahippocampus, and hippocampus), which has been previously reported in patients with progressive non-fluent aphasia[21-23]. This finding may provide important evidence for different stages of neurodegenerative processes in patients with semantic dementia. Regions with MD changes were extended to language-related areas, which suggested increased overlap with behavioral variant frontotemporal dementia[24-25].
Neuropathological studies in PPA patients have identified abnormalities in frontal, perisylvian, and temporal cortices[26]. Approximately 60% of patients present with nonspecific focal atrophy, also known as neuronal loss with gliosis, which lacks distinctive histopathological features[27]. These neurodegenerative alterations may result in increased extracellular space in specific brain regions, which could increase diffusivity. This could explain the MD increase in the present patient.
Results demonstrated reduced FA in the corpus callosum. This could be due to several processes, including a decreased number or density of axons, demyelination, or decreased orientational coherence of fiber tracts[28]. However, basal ganglia FA was significantly greater in the patient compared with the control subjects, and a previous study demonstrated atrophic basal ganglia PPA patients[27].
The basal ganglia are important components of cortical-subcortical circuits and are connected to each other through intrinsic connections that exhibit anisotropy. If white matter fibers that connect cortical areas to basal ganglia are selectively affected by Wallerian degeneration or neuronal dysfunction, the intact intrinsic connections could display increased coherence[29]. However, these underlying mechanisms remain unclear.
Furthermore, MD changes were more pronounced than FA, which suggested that this index is more sensitive for detecting brain alterations, in particular neurodegenerative processes. This could also indicate only mild fiber loss with no obvious decreased fiber tract orientation coherence, which could represent disease duration and severity.
The present study provided clear in vivo MRI evidence of diffuse brain involvement in a patient with semantic dementia. Increased MD was detected in the cerebra, which extended to areas outside of the language- related network, suggesting that MD could be a better marker of disease progression. This technique provides unique information that cannot be obtained from other imaging methods and allows objective assessment of lesion extent. Further follow-up studies and additional cases studies using diffusion tensor imaging compared to conventional MRI, as well as clinical correlations including neuropsychological functions, are needed to determine the dynamics of this phenomenon.

REFERENCES

[1] Mesulam MM. Primary progressive aphasia--differentiation from Alzheimer's disease. Ann Neurol. 1987;22(4):533-534.
[2] Gorno-Tempini ML, Dronkers NF, Rankin KP, et al. Cognition and anatomy in three variants of primary progressive aphasia. Ann Neurol. 2004;55(3):335-346.
[3] Panegyres PK, McCarthy M, Campbell A, et al. Correlative studies of structural and functional imaging in primary progressive aphasia. Am J Alzheimers Dis Other Demen. 2008;23(2):184-191.
[4] Clark DG, Charuvastra A, Miller BL, et al. Fluent versus nonfluent primary progressive aphasia: a comparison of clinical and functional neuroimaging features. Brain Lang. 2005;94(1):54-60.
[5] Kanaan RA, Kim JS, Kaufmann WE, et al. Diffusion tensor imaging in schizophrenia. Biol Psychiatry. 2005;58(12):921-929.
[6] Chen ZQ, Ni P, Xiao H, et al. Changes in brain function and anatomical structure following treatment of hyperbaric oxygen for visual pathway abnormalities in 16 cases: evaluation of functional magnetic resonance imaging combined with diffusion tensor imaging. Neural Regen Res. 2008;3(2):117-123.
[7] Thivard L, Adam C, Hasboun D, et al. Interictal diffusion MRI in partial epilepsies explored with intracerebral electrodes. Brain. 2006;129(Pt 2):375-385.
[8] Abe O, Yamada H, Masutani Y, et al. Amyotrophic lateral sclerosis: diffusion tensor tractography and voxel-based analysis. NMR Biomed. 2004;17(6):411-416.
[9] Chen Q, Lui S, Wang JG, et al. Diffusion tensor imaging of two unrelated Chinese men with hereditary spastic paraplegia associated with thin corpus callosum. Neurosci Lett. 2008;441(1): 21-24.
[10] Chen Q, Lui S, Li CX, et al. MRI-negative refractory partial epilepsy: role for diffusion tensor imaging in high field MRI. Epilepsy Res. 2008;80(1):83-89.
[11] Lezak MD. Neuropsychological Assessment. 4th ed. New York: Oxford University Press. 2004.
[12] Folstein W,Folstein SE,McHugh PR. "Mini-Mental State": a practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res. 1975;12:189-198.
[13] Pierpaoli C, Jezzard P, Basser PJ, et al. Diffusion tensor MR imaging of the human brain. Radiology. 1996;201(3):637-648.
[14] Kiebel SJ, Poline JB, Friston KJ, et al. Robust smoothness estimation in statistical parametric maps using standardized residuals from the general linear model. Neuroimage. 1999;10(6):756-766.
[15] Crinion J, Ashburner J, Leff A, et al. Spatial normalization of lesioned brains: performance evaluation and impact on fMRI analyses. Neuroimage. 2007;37(3):866-875.
[16] Mesulam MM. Primary progressive aphasia. Ann Neurol. 2001; 49(4):425-432.
[17] Mesulam M. Primary progressive aphasia pathology. Ann Neurol. 2008;63(1):124-125.
[18] Mendez MF, Cummings JL. Dementia: A Clinical Approach. 3rd ed. Philadelphia: Butterworth-Heinemann. 2003.
[19] Rabinovici GD, Jagust WJ, Furst AJ, et al. Abeta amyloid and glucose metabolism in three variants of primary progressive aphasia. Ann Neurol. 2008;64(4):388-401.
[20] George A, Mathuranath PS. Primary progressive aphasia: a comparative study of progressive nonfluent aphasia and semantic dementia. Neurol India. 2005;53(2):162-165.
[21] Vandenbulcke M, Peeters R, Van Hecke P, et al. Anterior temporal laterality in primary progressive aphasia shifts to the right. Ann Neurol. 2005;58(3):362-370.
[22] Soriani-Lefevre MH, Hannequin D, Bakchine S, et al. Evidence of bilateral temporal lobe involvement in primary progressive aphasia: a SPECT study. J Nucl Med. 2003;44(7):1013-1022.
[23] Repetto C, Manenti R, Cotelli M, et al. Right hemisphere involvement in non-fluent primary progressive aphasia. Behav Neurol. 2007;18(4):239-243.
[24] Banks SJ, Weintraub S. Neuropsychiatric symptoms in behavioral variant frontotemporal dementia and primary progressive aphasia. J Geriatr Psychiatry Neurol. 2008;21(2):133-141.
[25] Rosen HJ, Allison SC, Ogar JM, et al. Behavioral features in semantic dementia vs other forms of progressive aphasias. Neurology. 2006;67(10):1752-1756.
[26] Ikeda K. Anterior type dementia from the viewpoint of neuropathology. Rinsho Shinkeigaku. 2004;44(11):882-885.
[27] Rosen HJ, Kramer JH, Gorno-Tempini ML, et al. Patterns of cerebral atrophy in primary progressive aphasia. Am J Geriatr psychiatry. 2002;10:89-97.
[28] Takao M, Tsuchiya K, Mimura M, et al. Corticobasal degeneration as cause of progressive non-fluent aphasia: clinical, radiological and pathological study of an autopsy case. Neuropathology. 2006;26(6):569-578.
[29] Ciccarelli O, Werring DJ, Wheeler-Kingshott CA, et al. Investigation of MS normal-appearing brain using diffusion tensor MRI with clinical correlations. Neurology. 2001;56(7):926-933.
 (Edited by Zhao WQ/Song LP/Wang L)