ORIGINAL ARTICLE
Cortical Metabolic and Nigrostriatal Abnormalities Associated With Clinical Stage-Specific Dementia With Lewy Bodies Shu-Hua Huang, MD,* Chiung-Chih Chang, MD, PhD,† Chun-Chung Lui, MD,‡ Nai-Ching Chen, MD,† Chen-Chang Lee, MSc,‡ Pei-Wen Wang, MD,* and Ching-Fen Jiang, PhD§ Purpose: The aims of this study were to investigate the hypometabolic regions of FDG PET compared with the nigrostriatal dopamine pathway abnormalities in TRODAT-1 scan in patients with dementia with Lewy bodies (DLBs) at mild and dementia stages as well as to validate the correlation among networks being constructed with clinical data. Materials and Methods: A total of 25 DLB patients were classified into 2 functional groups stratified by the Clinical Dementia Rating (CDR) Scale (CDR 0.5: n = 14, mild stage; CDR 1 or 2: n = 11, dementia stage) compared with 9 agematched controls. Neuroimaging survey was applied using information derived from FDG PET by performing voxel-based analysis and a semiquantitative 99m Tc-TRODAT-1 scan to correlate these results with the cognitive and Unified Parkinson's Disease Rating Scale. Results: Compared with normal database, the patients with mild stage showed hypometabolism in the temporal regions, anterior cingulate cortex, inferior orbital region, thalamus, and caudate nucleus. Although at the dementia stage, more extensive cortical hypometabolism involving occipital region were found. The dopamine transporter levels derived from TRODAT-1 scan had excellent discrimination in diagnosing DLB compared with age-matched normal controls (1.58 [0.2] and 1.84 [0.1], P < 0.01) but without significant differences between mild and dementia stages. The sophisticated cortical-brainstem networks by FDG PET and TRODAT-1 yielded good clinical correlation. Conclusions: The networks yielded from FDG PET and TRODAT-1 revealed good correlation with clinical data and that nigrostriatal pathway abnormalities are preceded by typical occipital hypometabolism in mild stage of DLB. Dopamine transporter levels may serve as early diagnostic tool and FDG PET as staging indicator for DLB pathology. Key Words: dementia with Lewy bodies, Clinical Dementia Rating, FDG, PET, TRODAT-1 (Clin Nucl Med 2015;40: 26–31)
D
ementia with Lewy bodies (DLBs) is recognized as the second most common cause of degenerative dementia associated with α-synuclein pathology.1 Deficits on tests of attention, executive function, and visuospatial abilities that interfere with normal social functions represent the central features for dementia. The brainstem-predominant, limbic or diffuse neocortical LBs (Lewy bodies) are associated with early characteristic core features of parkinsonism, fluctuation of consciousness, and visual hallucination.2 Early recognition of this disorder is essential because its severe neuroleptic hypersensitivity is highly linked with mortality. Compared with Alzheimer dementia (AD), DLB more commonly has early LB-related pathology in the midbrain, pontine, and
Received for publication April 2, 2014; revision accepted September 23, 2014. From the Departments of *Nuclear Medicine, †Neurology, and ‡Radiology, Kaohsiung Chang Gung Memorial Hospital and Chang Gung University College of Medicine, and §Department of Biomedical Engineering, I-Shou University, Kaohsiung, Taiwan. Conflicts of interest and sources of funding: Supported by the research grant CMRPG8A0621 from Chang Gung Memorial Hospital. Reprints: Shu-Hua Huang, MD, Department of Nuclear Medicine, Kaohsiung Chang Gung Memorial Hospital, 123, Ta-Pei Rd, Niaosung, Kaohsiung 833, Taiwan. E-mail: [email protected]. Copyright © 2014 by Lippincott Williams & Wilkins ISSN: 0363-9762/15/4001–0026
26
www.nuclearmed.com
basal forebrain nuclei that primarily affect the cholinergic or dopaminergic systems.3,4 Consequently, damage to the integrity of the nigrostriatal system was observed using dopamine transporter (DAT) SPECT as a means of supportive features for DLB.5 Although the areas affected by LB pathologies are more recognizable in the brainstempredominant DLB, the increasing evidence implies that the atrophy of the cingulate gyrus, insula, and anterior frontal and temporal lobes characterize the MRI patterns in DLB.6 Information derived from functional neuroimaging FDG PET provide consistent yet conflicting reports, from the viewpoint of disruption of diaschiatic neuronal networks, that occipital hypometabolism characterized this disease compared with AD.5–7 However, group studies mixing patients at different clinical stages may have biased their functional results into different topographic distribution features.5–10 Several investigators have not demonstrated the typical occipital hypometabolism or hypoperfusion pattern, particularly at the early stage of DLB.8–10 Whether the degenerative patterns in DLB, grounded by clinical severities, could be yielded by studying data from FDG PET remains to be elucidated. Because changes in FDG PET and TRODAT-1 are both considered as neuronal injury biomarkers in the revised DLB criteria,5 the primary purpose of this study was to identify the affected networks in DLB patients at the mild and dementia stages, stratifying the patients according to their Clinical Dementia Rating (CDR) Scale. The clinical parameters and neuropsychological tests11 were used to verify the clinical significance of the networks identified.
PATIENTS AND METHODS This study was approved by the institutional review board of Chang Gung Memorial Hospital and complied with the ethical standards established in the Declaration of Helsinki. The experiments were undertaken with the written informed consent of each subject and their caregivers (when appropriate). Twenty-five patients with clinical diagnosis of probable DLB1 were recruited (Table 1). All had undergone comprehensive neurologic, neuropsychological, and functional assessment based on a consensus rendered in a multidisciplinary conference.12 The exclusion criteria were (1) clinical history of stroke, (2) modified Hachinski ischemic score greater than 4 to avoid cerebrovascular pathology,13 and (3) confusional or fluctuated state with difficulty participating in neuropsychological tests and imaging studies. All of the enrolled subjects received a revisit of clinical criteria checkups for DLB1 upon screening. Once enrolled, the patients underwent cognitive testing and the neuroimaging studies (TRODAT1 and FDG PET). Nine age-matched and sex-matched healthy subjects were enrolled as controls for neuroimaging and clinical data comparison. The inclusion and exclusion criteria for the control subjects were based on previous studies.12,14
Cognitive Testing The Mini-Mental State Examination (MMSE)15 and Cognitive Ability Screening Instrument (CASI)11 were used to assess the general intellectual function. Nine subscores from CASI were analyzed. For the behavioral observations, the neuropsychiatric inventory was used,16 Clinical Nuclear Medicine • Volume 40, Number 1, January 2015
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Clinical Nuclear Medicine • Volume 40, Number 1, January 2015
Neuroimaging in DLB
TABLE 1. Clinical Data of Controls (n = 9) and Patients With DLB (n = 25) Groups by Cognitive Severity Characteristics
Control
All Patients
Group 1
Group 2
Case numbers Sex (Male/Female) Age, y Parkinsonism REM sleep behavior disorders Visual hallucinations Fluctuations CDR CDR sum of box Memory Orientation Judgment and problem solving Community affairs Home and hobbies Personal care CDR cognitive subdomain CDR functional subdomain MMSE CASI Total scores Mental manipulation Attention Orientation Long-term memory Short-term memory Abstract thinking Drawing Verbal fluency Language Neuropsychiatric inventory‡ Total scores Apathy Sleep disorder
9 5/4 75 (5.2)
0 0 0 0 0 0 0 0 0 0 27.4 (2.2)
25 15/10 77.9 (6.3) 17 24 22 20 0.8 (0.4)* 4.3 (3.3)* 1.02 (0.6* 0.82 (0.6)* 0.62 (0.7)* 0.70 (0.6)* 0.64 (0.6)* 0.52 (0.7)* 2.46 (1.8)* 1.86 (1.7)* 18.3 (6.5)*
14 9/5 76.9 (6.3) 8 13 13 11 0.5 2.04 (1.1)* 0.68 (0.2)* 0.39 (0.2) 0.18 (0.3) 0.29 (0.3) 0.29 (0.3) 0.21 (0.4) 1.25 (0.4)* 0.79 (0.8) 22.1 (4.5)*
11 6/5 79.3 (6.2) 9 11 9 9 ≥1 7.2 (2.9)*† 1.46 (0.5)*† 1.36 (0.6)*† 1.18 (0.6)*† 1.20 (0.5)*† 1.09 (0.5)*† 0.91 (0.7)*† 4.0 (1.6)*† 3.2 (1.5)*† 13.4 (5.4)*†
91.0 (5.7) 8.1 (1.3) 6.9 (1.1) 17.7 (0.7) 9.8 (0.7) 10.3 (1.3) 10.8 (1.39) 9.6 (0.7) 8.3 (1.9) 9.6 (0.9)
63.4 (23.8)* 4.0 (2.9)* 6.4 (1.2) 12.1 (5.8)* 7.3 (3.2)* 5.8 (3.9)* 8.1 (3.6)* 7.8 (3.0)* 4.9 (2.8)* 7.2 (2.7)*
80.8 (9.1) 5.7 (2.1)* 6.6 (1.2) 16.3 (1.9) 9.5 (0.9 8.4 (2.7) 10.5 (1.73) 9.0 (1.6) 6.4 (2.0) 8.4 (2.0)
44.5 (19.9)*† 2.1 (2.6)*† 6.1 (1.2) 7.5 (5.0)*† 4.7 (2.9)*† 2.9 (2.7)*† 5.5 (3.4)*† 6.6 (3.7)* 3.2 (2.8)*† 5.9 (2.8)*†
3.6 (3.9) 0 3.3 (3.9)
14.6 (13.5)* 1.5 (3.6) 5.6 (5.7)*
13.8 (7.3)* 0 8.1 (5.6)*
15.6 (15.4)* 3.4 (4.8)*† 2.6 (4.3)†
Data are presented as mean (SD). *P ≤ 0.01 compared with controls. †P ≤ 0.01 compared with the DLB group 1. ‡Only the total scores and significant subdomains in neuropsychiatric inventory are listed. REM, rapid eye movement.
with scores ranging from 0 to 144. The parkinsonian features were assessed using the Unified Parkinson's Disease Rating Scale (UPDRS).
Patient Grouping According to the Cognitive Functional Severity Cognitive severity was assessed using the CDR.11 Fourteen patients with a CDR score of 0.5 were defined as group 1, representing mild DLB stage, and 11 patients with CDR score of 1 or 2 were defined as group 2, as the dementia stage.
Imaging Acquisition and Analysis TRODAT-1 Acquisition All the patients were injected intravenously with a single bolus dose of 740 MBq (20 mCi) 99mTc-TRODAT-1. The brain SPECT/CT (Symbia T; Siemens, Erlangen, Germany) images were obtained 4 hours © 2014 Lippincott Williams & Wilkins
later. The SPECT/CT scanner was equipped with low-energy, highresolution collimators and a dual-slice spiral CT. The acquisition parameters for the SPECT were a 128 128 matrix with 60 frames (40 s/frame). The scan parameters for the CT were 130 kV, 17 mA, 5-mm slices, and image reconstruction with a medium-smooth kernel. The SPECT images were attenuation-corrected based on the CT images and scatter-corrected with Flash 3-dimensional (3D) algorithm (ordered subsets expectation and 3D maximization with resolution correction) with 8 subsets and 8 iterations.
TRODAT-1 Quantitative Analysis Regions of interests were drawn on the caudate and putamen of each hemisphere in the corresponding anatomical regions derived from brain MRI scans. The fusion technique and regions selected for analysis followed the published protocol.17 The occipital cortex served as a background area. The ratio of the specific to nonspecific striatal www.nuclearmed.com
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
27
Clinical Nuclear Medicine • Volume 40, Number 1, January 2015
Huang et al
(caudate and putamen) TRODAT-1 binding in each region was calculated by mean region of interest (ROI) counts divided by mean occipital cortex counts.
an automated anatomic labeling template.19 The example of the FDG PET brain analysis is shown in Figure 1. Image analysis was performed by an experienced nuclear medicine physician who was blinded to the results of neurological testing.
FDG PET Acquisition The patients fasted for at least 6 hours and were then intravenously injected with FDG in doses of 370 Bq (10 mCi). After injection, the patients were conducted to a quiet dimly lit environment with minimal background noise, where they stayed for 40 minutes. A 3D mode PET/CT was then taken for 10 minutes. A polycarbonate head holder was used to reduce head movements during the scan. The CT images were used for attenuation correction and fusion. Helical CT images were acquired first using the following parameters: 140 kV, 170 mA (maximum), and 3.75-mm-thick section. Images were reconstructed using an ordered subsets expectation maximization algorithm (OSEM) algorithm with 2 iterations and 16 subsets, and displayed as a 128 128 matrix.
FDG PET Qualitative Analysis Qualitative analysis used the software package Scenium (Siemens Medical Solutions USA, Inc molecular imaging) accompanied by a database prepared from age-matched normal controls. Individual preprocessed images from each subject were compared with the database for statistical analysis, whereas the extent of glucose uptake ratio was expressed as a set of Z values. Each Z value was equal to the division of difference between the mean of normal controls and the result subject by the SD of the normal controls, shown as a projection chart in ROIs. The qualitative interpretation of the hypometabolism was defined by Z less than −2 in this study.18 The ROIs were selected from
Statistical Analysis All values were expressed as mean (SD). The χ2 test was used to compare the significance between categorical variables. The MannWhitney U test or Kruskal-Wallis test with Bonferroni correction was used, as appropriate, to compare continuous variables between groups. Spearman rank-order correlation was used to explore the correlation between the intensity of images in ROI (both in PET and TRODAT-1) and clinical scores adjusted for age and sex. Receiver operating characteristic (ROC) curves were generated to determine the cutoff of abnormal level for TRODAT-1 scan. Sensitivity and specificity were assessed as equally significant. All statistical analyses were performed using the Statistical Package for Social Science (SPSS version 11.0 for Windows; Chicago, Ill). Statistical significance was set at P < 0.05.
RESULT Cognitive Data Analysis The demographic data and cognitive performances are listed in Table 1. Regarding parkinsonian features, the UPDRS scores in the DLB patients were 45.5 (18.7), with 37.2 (17.2) for group 1 and 54.6 (16.3) for group 2. The UPDRS scores in group 2 were significantly higher than those of group 1 (P < 0.01).
FIGURE 1. Example of a normal control on which FDG PET brain analysis was performed using the Scenium software. A, The FDG PET imaging was rendered into a normalized volume providing prototype ROI map and is shown as a Z score map (B), which revealed no qualitative hypometabolism in the bilateral cerebral cortex. 28
www.nuclearmed.com
© 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Clinical Nuclear Medicine • Volume 40, Number 1, January 2015
TABLE 2. Analysis of TRODAT-1 Values in the Striatum
TRODAT-1 (controls) All patients (n = 25) Group 1 (n = 14) Group 2 (n = 11)
Striatum
Caudate Nucleus
Putamen
1.84 (0.11) 1.58 (0.20)* 1.57 (0.21)* 1.59 (0.19)*
1.90 (0.14) 1.62 (0.23)* 1.62 (0.23)* 1.62 (0.23)*
1.75 (0.12) 1.56 (0.18)* 1.53 (0.17)* 1.58 (0.18)*
Data presented as mean (SD). Note: group 1, patients with CDR of 0.5; group 2, patients with CDR 1 (n = 9) or 2 (n = 2). *P < 0.01 compared with normal controls.
TRODAT-1 Binding Ratio Analysis In lateralization, there was no significant differentiation between the right and left caudate (P = 0.70) or putamen (P = 0.68) DAT binding ratio. Therefore, the average values were presented (Table 2). Compared with the age-matched controls, the patients with DLB had significantly lower DAT binding ratio in the striatum, caudate, and putamen. The ROC curve of the DAT binding ratio indicated that a cutoff value of 1.7 in striatal uptake ratio had 90% sensitivity and 81.5% specificity for diagnosing DLB, equal to Z = −1.5 of the age-matched controls. The ROC curves suggested that either the caudate or putamen had excellent discrimination in diagnosing DLB (area under the curve: caudate, 0.89;
Neuroimaging in DLB
putamen, 0.86). The comparison between group 1 and group 2, however, showed no statistical significance. There are significant inverse correlation between UPDRS total scores with striatal (r = −0.45, P = 0.02) and caudate (r = −0.49, P = 0.02) DAT binding ratio.
Qualitative Data Analysis of Abnormal Regions From FDG PET For regional PET analysis, there were no significant differences in hemispheric laterality. Using a Z of −2 as cutoff, the percentages of regional hypometabolism are shown in Figure 2. In group 1 analysis, the regions at which hypometabolism occurred were in the mesial temporal regions (n = 12, 85.7%), anterior cingulate cortex (ACC) (n = 11, 78.6%), thalamus (n = 11, 78.6%), inferior orbital region (n = 10, 71.4%), superior temporal and parahippocampal regions (n = 10, 71.4%), and caudate nucleus (n = 9, 64.2%). When the analysis was classified as different cortical lobes, the abnormality was found in the temporal lobe (n = 13, 92.8%), followed by frontal (n = 11, 78.6%), basal ganglia and thalami (n = 11, 78.6%), and parietal (n = 10, 71.4%) lobes. Only 1 patient (7.1%) presented with hypometabolism involving the occipital lobe. Regarding group 2, the hypometabolic areas were diffusely found in the frontotemporal (n = 11, 100%), parietal (n = 10, 90.9%), basal ganglia (n = 10, 90.9%), thalami (n = 10, 90.9%), and occipital (n = 9, 81.8%) lobes (Table 3). Compared with the patients in group 1, those in group 2 exhibited significantly increased hypometabolism in the occipital regions, cingulate, superior orbital, putamen, and inferior temporal region (Fig. 1).
FIGURE 2. Differences in hypometabolic regions in 2 groups of DLB. Hypometabolism in FDG PET was classified using a Z score of −2. *P < 0.05; group 1, patients with CDR of 0.5; group 2, patients with CDR 1 (n = 9) or 2 (n = 2). © 2014 Lippincott Williams & Wilkins
www.nuclearmed.com
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
29
Clinical Nuclear Medicine • Volume 40, Number 1, January 2015
Huang et al
TABLE 3. The Average Z Scores (SD) and Percentage of Hypometabolism in Various Cortical Lobes in DLB Patients and Groups Frontal
Parietal
Temporal
Occipital
Basal Ganglia
Thalamus
Patient
Mean (SD)
%*
Mean (SD)
%
Mean (SD)
%
Mean (SD)
%
Mean (SD)
%
Mean (SD)
%
Total Group 1 Group 2
−1.1 (1.7) −0.4 (1.5) −1.8 (1.7)†
88.0 78.6 100
−0.7 (1.8) −0.3 (1.3) −1.3 (2.0)†
80.0 71.4 90.9
−1.5 (2.0) −1.0 (1.1) −2.4 (2.3)†
96.0 92.8 100
−0.3 (1.9) 0.5 (1.2) −1.1 (2.2)†
40.0 7.1 81.8
−1.5 (1.9) −1.0 (1.8) −2.0 (1.9)†
88.0 78.6 90.9
−2.5 (1.4) −1.9 (1.3) −3.1 (1.4)†
84.0 78.6 90.9
Group 1, DLB patients with CDR 0.5 rating; group 2, DLB patients with CDR 1 or 2 rating. *Percentage of hypometabolism according to regional cortical areas of mean Z score below −2.0. †P < 0.01 compared with DLB group 1.
Clinical Correlation Analysis The region-based correlations of the Z scores derived from PET and clinical data adjusted for age and sex are shown in Table 4. The results suggested that a wide network derived from PET correlated with the MMSE and CASI scores. The hypometabolism regions inversely correlated with the UPDRS scores except for the basal ganglia.
DISCUSSION In a cross-sectional manner, this study explored the affected neuronal network patterns by using complementary data from PET and TRODAT-1 in patients with DLB grounded by clinical severity. Our analysis showed that hypometabolism in the temporal regions, ACC, inferior orbital region, thalamus, and caudate nucleus feature the mild stage of DLB. Patients at the dementia stage had more extensive regional networks involving the occipital region that were good correlations with the cognitive and UPDRS domains. According to the nigrostriatal system in evaluating DAT by TRODAT-1, the abnormality in the nigrostriatal dopamine pathway appeared at mild stage of DLB but without significant progression at more advanced stages. The presence of early nigrostriatal pathway impairment and regional metabolic differences at various degrees of dementia suggested that the variable burden of a pathological protein load in the brain might cause a different temporal sequence of neuropathological changes and subsequent clinical symptoms in patients with DLB.20 Albin et al21 were the first to report FDG PET hypometabolism in the occipital association cortex and primary visual cortex regions in DLB compared with AD, followed by Imamura et al,6 who used a larger sample size and linked occipital hypometabolism with visual hallucination. Current consensus criteria for DLB suggest that a generalized low uptake on PET metabolic scan with reduced occipital activity is a supportive feature.1 Our PET study results indicated that few patients with mild DLB and most patients at more advanced stage exhibited hypometabolism in the occipital area. In most of the autopsyconfirmed DLB reports, FDG PET was visually rated,22–24 and most of the cases may have died at the end-stage of the disease. Under such conditions, the samples may have included limited cases at the early clinical stages. Base on the results of this study, occipital hypometabolism reported in relevant literature may not be the pathogenic signature of mild-stage DLB but may be considered a supportive feature of dementia-stage DLB.25 This statement is based on the observation that several studies have failed to establish occipital hypofunction in earlystage DLB patients,8–10 the lack of occipital atrophy in MRI measurements,1 and that LB seldom appear in occipital regions at early stage.26 Occipital hypometabolism helps discriminate between mild-stage and dementia-stage DLB in this study and may be served as an indicator for staging. According to the mild-stage DLB patient analysis, the hypometabolic regions were in the temporal regions, ACC, inferior orbital region, thalamus, and caudate nucleus. Interestingly, a part of the hypometabolic network in mild-stage DLB was identical to 30
www.nuclearmed.com
the anatomical structures with gray matter atrophy characterized on MRI patterns in DLB.6 These included the ACC, anterior frontal, and temporal lobes, implying that cortical metabolism might be affected earlier than cortical atrophy is. In dementia-stage patients, the hypometabolism state starts to increase diffusely, particularly in the occipital regions. Favorable correlations between the identified networks from the PET with the cognitive features determined for DLB patients were observed. This is consistent with a previous PET study in that the global cortical FDG has been found to be the best indicator of global cognitive functioning in DLB patients.27 In contrast to the findings for Parkinson disease,28 the TRODAT1 analysis revealed a reduction of signals in the caudate and putamen in DLB patients with a sensitivity and specificity of 78% and 94%.29 In this study, the decline of striatum DAT uptake develops even before the development of dementia and correlates well with clinical motor disability. The loss of ventrolateral nigra dopaminergic neuron represents a consistent feature early in DLB.30 A thorough longitudinal pathology study identified that the severity of dopamine cell loss correlated with disease duration.31 However, analysis of the TRODAT-1 binding ratio showed no significant differences between 2 functional groups. This study result supported a higher hierarchy in diagnostic ranking rather than staging by using DAT imaging. This study limitations include the fact that the diagnoses were not pathologically confirmed. Considerable clinical overlaps occurred between DLB with AD and with Parkinson dementia. Substantial effort has been expended to limit the study cohort by using clinical probable DLB.1 The second limitation is the small number of cases, and largescale trials are required to validate the results.
CONCLUSIONS The neuroimaging studies are involved in comparing FDG PET and TRODAT-1 modalities at mild and dementia stages of DLB. The FDG PET and TRODAT-1 analyses suggest that a sophisticated cortical-brainstem network is correlated with MMSE, CASI, and UPDRS. The nigrostriatal pathway abnormalities are preceded TABLE 4. Spearman Rank-Order Correlation Between Regional Cortical Z Scores and Clinical Data Z Scores
MMSE
CASI
UPDRS
Frontal Temporal Parietal Occipital Basal ganglia Thalamus
0.594* 0.842* 0.570* 0.554* 0.652* 0.664*
0.490† 0.774* 0.451† 0.566* 0.654* 0.700*
−0.557* −0.567* −0.551* −0.670* −0.283 −0.475†
*P < 0.01. †P < 0.05.
© 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Clinical Nuclear Medicine • Volume 40, Number 1, January 2015
by typical occipital hypometabolism at the mild stage of DLB. Dopamine transporter levels may serve as early diagnostic tool and FDG PET as staging indicator for DLB pathology. ACKNOWLEDGMENTS The authors thank the patients and their caregivers for their time and commitment to this research. They also thank Miss Y.T. Lin from the Department of Neurology, Chang Gung Memorial Hospital-Kaohsiung Medical Center for scheduling all of the experiments. REFERENCES 1. McKeith IG, Dickson DW, Lowe J, et al. Diagnosis and management of dementia with Lewy bodies: third report of the DLB Consortium. Neurology. 2005;65: 1863–1872. 2. Hansen L, Salmon D, Galasko D, et al. The Lewy body variant of Alzheimer's disease: a clinical and pathologic entity. Neurology. 1990;40:1–8. 3. Braak H, Ghebremedhin E, Rub U, et al. Stages in the development of Parkinson's disease-related pathology. Cell Tissue Res. 2004;318:121–134. 4. Langlais PJ, Thal L, Hansen L, et al. Neurotransmitters in basal ganglia and cortex of Alzheimer's disease with and without Lewy bodies. Neurology. 1993;43: 1927–1934. 5. McKeith IG, Burn DJ, Ballard CG, et al. Dementia with Lewy bodies. Semin Clin Neuropsychiatry. 2003;8:46–57. 6. Imamura T, Ishii K, Hirono N, et al. Visual hallucinations and regional cerebral metabolism in dementia with Lewy bodies (DLB). Neuroreport. 1999;10: 1903–1907. 7. Minoshima S, Foster NL, Sima AA, et al. Alzheimer's disease versus dementia with Lewy bodies: cerebral metabolic distinction with autopsy confirmation. Ann Neurol. 2001;50:358–365. 8. Londos E, Passant U, Brun A, et al. Regional cerebral blood flow and EEG in clinically diagnosed dementia with Lewy bodies and Alzheimer's disease. Arch Gerontol Geriatr. 2003;36:231–245. 9. Cordery RJ, Tyrrell PJ, Lantos PL, et al. Dementia with Lewy bodies studied with positron emission tomography. Arch Neurol. 2001;58:505–508. 10. Varma AR, Talbot PR, Snowden JS, et al. A 99mTc-HMPAO single-photon emission computed tomography study of Lewy body disease. J Neurol. 1997;244: 349–359. 11. Huang CW, Chang WN, Huang SH, et al. Impact of homocysteine on cortical perfusion and cognitive decline in mild Alzheimer's dementia. Eur J Neurol. 2013; 20:1191–1197. 12. Chang CC, Liu JS, Chang YY, et al. (99m)Tc-ethyl cysteinate dimer brain SPECT findings in early stage of dementia with Lewy bodies and Parkinson's disease patients: a correlation with neuropsychological tests. Eur J Neurol. 2008;15:61–65. 13. Rosen WG, Terry RD, Fuld PA, et al. Pathological verification of ischemic score in differentiation of dementias. Ann Neurol. 1980;7:486–488.
© 2014 Lippincott Williams & Wilkins
Neuroimaging in DLB
14. Chang CC, Chang YY, Chang WN, et al. Cognitive deficits in multiple system atrophy correlate with frontal atrophy and disease duration. Eur J Neurol. 2009;16: 1144–1150. 15. Folstein MF, 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. 16. Cummings JL, Mega M, Gray K, et al. The Neuropsychiatric Inventory: comprehensive assessment of psychopathology in dementia. Neurology. 1994;44: 2308–2314. 17. Chang CC, Chang WN, Lui CC, et al. Clinical significance of the pallidoreticular pathway in patients with carbon monoxide intoxication. Brain. 2011;134: 3632–3646. 18. Mirzaei S, Knoll P, Koehn H, et al. Assessment of diffuse Lewy body disease by 2-[18F]fluoro-2-deoxy-D-glucose positron emission tomography (FDG PET). BMC Nucl Med. 2003;3:1. 19. Tzourio-Mazoyer N, Landeau B, Papathanassiou D, et al. Automated anatomical labeling of activations in SPM using a macroscopic anatomical parcellation of the MNI MRI single-subject brain. Neuroimage. 2002;15:273–289. 20. Snider BJ, Norton J, Coats MA, et al. Novel presenilin 1 mutation (S170F) causing Alzheimer disease with Lewy bodies in the third decade of life. Arch Neurol. 2005;62:1821–1830. 21. Albin RL, Minoshima S, D'Amato CJ, et al. Fluoro-deoxyglucose positron emission tomography in diffuse Lewy body disease. Neurology. 1996;47:462–466. 22. Jagust W, Reed B, Mungas D, et al. What does fluorodeoxyglucose PET imaging add to a clinical diagnosis of dementia? Neurology. 2007;69:871–877. 23. Silverman DH, Small GW, Chang CY, et al. Positron emission tomography in evaluation of dementia: regional brain metabolism and long-term outcome. JAMA. 2001;286:2120–2127. 24. Hoffman JM, Welsh-Bohmer KA, Hanson M, et al. FDG PET imaging in patients with pathologically verified dementia. J Nucl Med. 2000;41:1920–1928. 25. Middelkoop HA, van der Flier WM, Burton EJ, et al. Dementia with Lewy bodies and AD are not associated with occipital lobe atrophy on MRI. Neurology. 2001; 57:2117–2120. 26. Gomez-Tortosa E, Newell K, Irizarry MC, et al. Clinical and quantitative pathologic correlates of dementia with Lewy bodies. Neurology. 1999;53:1284–1291. 27. Klein JC, Eggers C, Kalbe E, et al. Neurotransmitter changes in dementia with Lewy bodies and Parkinson disease dementia in vivo. Neurology. 2010;74: 885–892. 28. Huang WS, Lee MS, Lin JC, et al. Usefulness of brain 99mTc-TRODAT-1 SPET for the evaluation of Parkinson's disease. Eur J Nucl Med Mol Imaging. 2004; 31:155–161. 29. McKeith I, O'Brien J, Walker Z, et al. Sensitivity and specificity of dopamine transporter imaging with 123I-FP-CIT SPECT in dementia with Lewy bodies: a phase III, multicentre study. Lancet Neurol. 2007;6:305–313. 30. Zaccai J, Brayne C, McKeith I, et al. Patterns and stages of alpha-synucleinopathy: relevance in a population-based cohort. Neurology. 2008;70:1042–1048. 31. Halliday GM, McRitchie DA, Cartwright H, et al. Midbrain neuropathology in idiopathic Parkinson's disease and diffuse Lewy body disease. J Clin Neurosci. 1996;3:52–60.
www.nuclearmed.com
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
31
Cortical Metabolic and Nigrostriatal Abnormalities Associated With Clinical Stage-Specific Dementia With Lewy Bodies Shu-Hua Huang, MD,* Chiung-Chih Chang, MD, PhD,† Chun-Chung Lui, MD,‡ Nai-Ching Chen, MD,† Chen-Chang Lee, MSc,‡ Pei-Wen Wang, MD,* and Ching-Fen Jiang, PhD§ Purpose: The aims of this study were to investigate the hypometabolic regions of FDG PET compared with the nigrostriatal dopamine pathway abnormalities in TRODAT-1 scan in patients with dementia with Lewy bodies (DLBs) at mild and dementia stages as well as to validate the correlation among networks being constructed with clinical data. Materials and Methods: A total of 25 DLB patients were classified into 2 functional groups stratified by the Clinical Dementia Rating (CDR) Scale (CDR 0.5: n = 14, mild stage; CDR 1 or 2: n = 11, dementia stage) compared with 9 agematched controls. Neuroimaging survey was applied using information derived from FDG PET by performing voxel-based analysis and a semiquantitative 99m Tc-TRODAT-1 scan to correlate these results with the cognitive and Unified Parkinson's Disease Rating Scale. Results: Compared with normal database, the patients with mild stage showed hypometabolism in the temporal regions, anterior cingulate cortex, inferior orbital region, thalamus, and caudate nucleus. Although at the dementia stage, more extensive cortical hypometabolism involving occipital region were found. The dopamine transporter levels derived from TRODAT-1 scan had excellent discrimination in diagnosing DLB compared with age-matched normal controls (1.58 [0.2] and 1.84 [0.1], P < 0.01) but without significant differences between mild and dementia stages. The sophisticated cortical-brainstem networks by FDG PET and TRODAT-1 yielded good clinical correlation. Conclusions: The networks yielded from FDG PET and TRODAT-1 revealed good correlation with clinical data and that nigrostriatal pathway abnormalities are preceded by typical occipital hypometabolism in mild stage of DLB. Dopamine transporter levels may serve as early diagnostic tool and FDG PET as staging indicator for DLB pathology. Key Words: dementia with Lewy bodies, Clinical Dementia Rating, FDG, PET, TRODAT-1 (Clin Nucl Med 2015;40: 26–31)
D
ementia with Lewy bodies (DLBs) is recognized as the second most common cause of degenerative dementia associated with α-synuclein pathology.1 Deficits on tests of attention, executive function, and visuospatial abilities that interfere with normal social functions represent the central features for dementia. The brainstem-predominant, limbic or diffuse neocortical LBs (Lewy bodies) are associated with early characteristic core features of parkinsonism, fluctuation of consciousness, and visual hallucination.2 Early recognition of this disorder is essential because its severe neuroleptic hypersensitivity is highly linked with mortality. Compared with Alzheimer dementia (AD), DLB more commonly has early LB-related pathology in the midbrain, pontine, and
Received for publication April 2, 2014; revision accepted September 23, 2014. From the Departments of *Nuclear Medicine, †Neurology, and ‡Radiology, Kaohsiung Chang Gung Memorial Hospital and Chang Gung University College of Medicine, and §Department of Biomedical Engineering, I-Shou University, Kaohsiung, Taiwan. Conflicts of interest and sources of funding: Supported by the research grant CMRPG8A0621 from Chang Gung Memorial Hospital. Reprints: Shu-Hua Huang, MD, Department of Nuclear Medicine, Kaohsiung Chang Gung Memorial Hospital, 123, Ta-Pei Rd, Niaosung, Kaohsiung 833, Taiwan. E-mail: [email protected]. Copyright © 2014 by Lippincott Williams & Wilkins ISSN: 0363-9762/15/4001–0026
26
www.nuclearmed.com
basal forebrain nuclei that primarily affect the cholinergic or dopaminergic systems.3,4 Consequently, damage to the integrity of the nigrostriatal system was observed using dopamine transporter (DAT) SPECT as a means of supportive features for DLB.5 Although the areas affected by LB pathologies are more recognizable in the brainstempredominant DLB, the increasing evidence implies that the atrophy of the cingulate gyrus, insula, and anterior frontal and temporal lobes characterize the MRI patterns in DLB.6 Information derived from functional neuroimaging FDG PET provide consistent yet conflicting reports, from the viewpoint of disruption of diaschiatic neuronal networks, that occipital hypometabolism characterized this disease compared with AD.5–7 However, group studies mixing patients at different clinical stages may have biased their functional results into different topographic distribution features.5–10 Several investigators have not demonstrated the typical occipital hypometabolism or hypoperfusion pattern, particularly at the early stage of DLB.8–10 Whether the degenerative patterns in DLB, grounded by clinical severities, could be yielded by studying data from FDG PET remains to be elucidated. Because changes in FDG PET and TRODAT-1 are both considered as neuronal injury biomarkers in the revised DLB criteria,5 the primary purpose of this study was to identify the affected networks in DLB patients at the mild and dementia stages, stratifying the patients according to their Clinical Dementia Rating (CDR) Scale. The clinical parameters and neuropsychological tests11 were used to verify the clinical significance of the networks identified.
PATIENTS AND METHODS This study was approved by the institutional review board of Chang Gung Memorial Hospital and complied with the ethical standards established in the Declaration of Helsinki. The experiments were undertaken with the written informed consent of each subject and their caregivers (when appropriate). Twenty-five patients with clinical diagnosis of probable DLB1 were recruited (Table 1). All had undergone comprehensive neurologic, neuropsychological, and functional assessment based on a consensus rendered in a multidisciplinary conference.12 The exclusion criteria were (1) clinical history of stroke, (2) modified Hachinski ischemic score greater than 4 to avoid cerebrovascular pathology,13 and (3) confusional or fluctuated state with difficulty participating in neuropsychological tests and imaging studies. All of the enrolled subjects received a revisit of clinical criteria checkups for DLB1 upon screening. Once enrolled, the patients underwent cognitive testing and the neuroimaging studies (TRODAT1 and FDG PET). Nine age-matched and sex-matched healthy subjects were enrolled as controls for neuroimaging and clinical data comparison. The inclusion and exclusion criteria for the control subjects were based on previous studies.12,14
Cognitive Testing The Mini-Mental State Examination (MMSE)15 and Cognitive Ability Screening Instrument (CASI)11 were used to assess the general intellectual function. Nine subscores from CASI were analyzed. For the behavioral observations, the neuropsychiatric inventory was used,16 Clinical Nuclear Medicine • Volume 40, Number 1, January 2015
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Clinical Nuclear Medicine • Volume 40, Number 1, January 2015
Neuroimaging in DLB
TABLE 1. Clinical Data of Controls (n = 9) and Patients With DLB (n = 25) Groups by Cognitive Severity Characteristics
Control
All Patients
Group 1
Group 2
Case numbers Sex (Male/Female) Age, y Parkinsonism REM sleep behavior disorders Visual hallucinations Fluctuations CDR CDR sum of box Memory Orientation Judgment and problem solving Community affairs Home and hobbies Personal care CDR cognitive subdomain CDR functional subdomain MMSE CASI Total scores Mental manipulation Attention Orientation Long-term memory Short-term memory Abstract thinking Drawing Verbal fluency Language Neuropsychiatric inventory‡ Total scores Apathy Sleep disorder
9 5/4 75 (5.2)
0 0 0 0 0 0 0 0 0 0 27.4 (2.2)
25 15/10 77.9 (6.3) 17 24 22 20 0.8 (0.4)* 4.3 (3.3)* 1.02 (0.6* 0.82 (0.6)* 0.62 (0.7)* 0.70 (0.6)* 0.64 (0.6)* 0.52 (0.7)* 2.46 (1.8)* 1.86 (1.7)* 18.3 (6.5)*
14 9/5 76.9 (6.3) 8 13 13 11 0.5 2.04 (1.1)* 0.68 (0.2)* 0.39 (0.2) 0.18 (0.3) 0.29 (0.3) 0.29 (0.3) 0.21 (0.4) 1.25 (0.4)* 0.79 (0.8) 22.1 (4.5)*
11 6/5 79.3 (6.2) 9 11 9 9 ≥1 7.2 (2.9)*† 1.46 (0.5)*† 1.36 (0.6)*† 1.18 (0.6)*† 1.20 (0.5)*† 1.09 (0.5)*† 0.91 (0.7)*† 4.0 (1.6)*† 3.2 (1.5)*† 13.4 (5.4)*†
91.0 (5.7) 8.1 (1.3) 6.9 (1.1) 17.7 (0.7) 9.8 (0.7) 10.3 (1.3) 10.8 (1.39) 9.6 (0.7) 8.3 (1.9) 9.6 (0.9)
63.4 (23.8)* 4.0 (2.9)* 6.4 (1.2) 12.1 (5.8)* 7.3 (3.2)* 5.8 (3.9)* 8.1 (3.6)* 7.8 (3.0)* 4.9 (2.8)* 7.2 (2.7)*
80.8 (9.1) 5.7 (2.1)* 6.6 (1.2) 16.3 (1.9) 9.5 (0.9 8.4 (2.7) 10.5 (1.73) 9.0 (1.6) 6.4 (2.0) 8.4 (2.0)
44.5 (19.9)*† 2.1 (2.6)*† 6.1 (1.2) 7.5 (5.0)*† 4.7 (2.9)*† 2.9 (2.7)*† 5.5 (3.4)*† 6.6 (3.7)* 3.2 (2.8)*† 5.9 (2.8)*†
3.6 (3.9) 0 3.3 (3.9)
14.6 (13.5)* 1.5 (3.6) 5.6 (5.7)*
13.8 (7.3)* 0 8.1 (5.6)*
15.6 (15.4)* 3.4 (4.8)*† 2.6 (4.3)†
Data are presented as mean (SD). *P ≤ 0.01 compared with controls. †P ≤ 0.01 compared with the DLB group 1. ‡Only the total scores and significant subdomains in neuropsychiatric inventory are listed. REM, rapid eye movement.
with scores ranging from 0 to 144. The parkinsonian features were assessed using the Unified Parkinson's Disease Rating Scale (UPDRS).
Patient Grouping According to the Cognitive Functional Severity Cognitive severity was assessed using the CDR.11 Fourteen patients with a CDR score of 0.5 were defined as group 1, representing mild DLB stage, and 11 patients with CDR score of 1 or 2 were defined as group 2, as the dementia stage.
Imaging Acquisition and Analysis TRODAT-1 Acquisition All the patients were injected intravenously with a single bolus dose of 740 MBq (20 mCi) 99mTc-TRODAT-1. The brain SPECT/CT (Symbia T; Siemens, Erlangen, Germany) images were obtained 4 hours © 2014 Lippincott Williams & Wilkins
later. The SPECT/CT scanner was equipped with low-energy, highresolution collimators and a dual-slice spiral CT. The acquisition parameters for the SPECT were a 128 128 matrix with 60 frames (40 s/frame). The scan parameters for the CT were 130 kV, 17 mA, 5-mm slices, and image reconstruction with a medium-smooth kernel. The SPECT images were attenuation-corrected based on the CT images and scatter-corrected with Flash 3-dimensional (3D) algorithm (ordered subsets expectation and 3D maximization with resolution correction) with 8 subsets and 8 iterations.
TRODAT-1 Quantitative Analysis Regions of interests were drawn on the caudate and putamen of each hemisphere in the corresponding anatomical regions derived from brain MRI scans. The fusion technique and regions selected for analysis followed the published protocol.17 The occipital cortex served as a background area. The ratio of the specific to nonspecific striatal www.nuclearmed.com
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
27
Clinical Nuclear Medicine • Volume 40, Number 1, January 2015
Huang et al
(caudate and putamen) TRODAT-1 binding in each region was calculated by mean region of interest (ROI) counts divided by mean occipital cortex counts.
an automated anatomic labeling template.19 The example of the FDG PET brain analysis is shown in Figure 1. Image analysis was performed by an experienced nuclear medicine physician who was blinded to the results of neurological testing.
FDG PET Acquisition The patients fasted for at least 6 hours and were then intravenously injected with FDG in doses of 370 Bq (10 mCi). After injection, the patients were conducted to a quiet dimly lit environment with minimal background noise, where they stayed for 40 minutes. A 3D mode PET/CT was then taken for 10 minutes. A polycarbonate head holder was used to reduce head movements during the scan. The CT images were used for attenuation correction and fusion. Helical CT images were acquired first using the following parameters: 140 kV, 170 mA (maximum), and 3.75-mm-thick section. Images were reconstructed using an ordered subsets expectation maximization algorithm (OSEM) algorithm with 2 iterations and 16 subsets, and displayed as a 128 128 matrix.
FDG PET Qualitative Analysis Qualitative analysis used the software package Scenium (Siemens Medical Solutions USA, Inc molecular imaging) accompanied by a database prepared from age-matched normal controls. Individual preprocessed images from each subject were compared with the database for statistical analysis, whereas the extent of glucose uptake ratio was expressed as a set of Z values. Each Z value was equal to the division of difference between the mean of normal controls and the result subject by the SD of the normal controls, shown as a projection chart in ROIs. The qualitative interpretation of the hypometabolism was defined by Z less than −2 in this study.18 The ROIs were selected from
Statistical Analysis All values were expressed as mean (SD). The χ2 test was used to compare the significance between categorical variables. The MannWhitney U test or Kruskal-Wallis test with Bonferroni correction was used, as appropriate, to compare continuous variables between groups. Spearman rank-order correlation was used to explore the correlation between the intensity of images in ROI (both in PET and TRODAT-1) and clinical scores adjusted for age and sex. Receiver operating characteristic (ROC) curves were generated to determine the cutoff of abnormal level for TRODAT-1 scan. Sensitivity and specificity were assessed as equally significant. All statistical analyses were performed using the Statistical Package for Social Science (SPSS version 11.0 for Windows; Chicago, Ill). Statistical significance was set at P < 0.05.
RESULT Cognitive Data Analysis The demographic data and cognitive performances are listed in Table 1. Regarding parkinsonian features, the UPDRS scores in the DLB patients were 45.5 (18.7), with 37.2 (17.2) for group 1 and 54.6 (16.3) for group 2. The UPDRS scores in group 2 were significantly higher than those of group 1 (P < 0.01).
FIGURE 1. Example of a normal control on which FDG PET brain analysis was performed using the Scenium software. A, The FDG PET imaging was rendered into a normalized volume providing prototype ROI map and is shown as a Z score map (B), which revealed no qualitative hypometabolism in the bilateral cerebral cortex. 28
www.nuclearmed.com
© 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Clinical Nuclear Medicine • Volume 40, Number 1, January 2015
TABLE 2. Analysis of TRODAT-1 Values in the Striatum
TRODAT-1 (controls) All patients (n = 25) Group 1 (n = 14) Group 2 (n = 11)
Striatum
Caudate Nucleus
Putamen
1.84 (0.11) 1.58 (0.20)* 1.57 (0.21)* 1.59 (0.19)*
1.90 (0.14) 1.62 (0.23)* 1.62 (0.23)* 1.62 (0.23)*
1.75 (0.12) 1.56 (0.18)* 1.53 (0.17)* 1.58 (0.18)*
Data presented as mean (SD). Note: group 1, patients with CDR of 0.5; group 2, patients with CDR 1 (n = 9) or 2 (n = 2). *P < 0.01 compared with normal controls.
TRODAT-1 Binding Ratio Analysis In lateralization, there was no significant differentiation between the right and left caudate (P = 0.70) or putamen (P = 0.68) DAT binding ratio. Therefore, the average values were presented (Table 2). Compared with the age-matched controls, the patients with DLB had significantly lower DAT binding ratio in the striatum, caudate, and putamen. The ROC curve of the DAT binding ratio indicated that a cutoff value of 1.7 in striatal uptake ratio had 90% sensitivity and 81.5% specificity for diagnosing DLB, equal to Z = −1.5 of the age-matched controls. The ROC curves suggested that either the caudate or putamen had excellent discrimination in diagnosing DLB (area under the curve: caudate, 0.89;
Neuroimaging in DLB
putamen, 0.86). The comparison between group 1 and group 2, however, showed no statistical significance. There are significant inverse correlation between UPDRS total scores with striatal (r = −0.45, P = 0.02) and caudate (r = −0.49, P = 0.02) DAT binding ratio.
Qualitative Data Analysis of Abnormal Regions From FDG PET For regional PET analysis, there were no significant differences in hemispheric laterality. Using a Z of −2 as cutoff, the percentages of regional hypometabolism are shown in Figure 2. In group 1 analysis, the regions at which hypometabolism occurred were in the mesial temporal regions (n = 12, 85.7%), anterior cingulate cortex (ACC) (n = 11, 78.6%), thalamus (n = 11, 78.6%), inferior orbital region (n = 10, 71.4%), superior temporal and parahippocampal regions (n = 10, 71.4%), and caudate nucleus (n = 9, 64.2%). When the analysis was classified as different cortical lobes, the abnormality was found in the temporal lobe (n = 13, 92.8%), followed by frontal (n = 11, 78.6%), basal ganglia and thalami (n = 11, 78.6%), and parietal (n = 10, 71.4%) lobes. Only 1 patient (7.1%) presented with hypometabolism involving the occipital lobe. Regarding group 2, the hypometabolic areas were diffusely found in the frontotemporal (n = 11, 100%), parietal (n = 10, 90.9%), basal ganglia (n = 10, 90.9%), thalami (n = 10, 90.9%), and occipital (n = 9, 81.8%) lobes (Table 3). Compared with the patients in group 1, those in group 2 exhibited significantly increased hypometabolism in the occipital regions, cingulate, superior orbital, putamen, and inferior temporal region (Fig. 1).
FIGURE 2. Differences in hypometabolic regions in 2 groups of DLB. Hypometabolism in FDG PET was classified using a Z score of −2. *P < 0.05; group 1, patients with CDR of 0.5; group 2, patients with CDR 1 (n = 9) or 2 (n = 2). © 2014 Lippincott Williams & Wilkins
www.nuclearmed.com
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
29
Clinical Nuclear Medicine • Volume 40, Number 1, January 2015
Huang et al
TABLE 3. The Average Z Scores (SD) and Percentage of Hypometabolism in Various Cortical Lobes in DLB Patients and Groups Frontal
Parietal
Temporal
Occipital
Basal Ganglia
Thalamus
Patient
Mean (SD)
%*
Mean (SD)
%
Mean (SD)
%
Mean (SD)
%
Mean (SD)
%
Mean (SD)
%
Total Group 1 Group 2
−1.1 (1.7) −0.4 (1.5) −1.8 (1.7)†
88.0 78.6 100
−0.7 (1.8) −0.3 (1.3) −1.3 (2.0)†
80.0 71.4 90.9
−1.5 (2.0) −1.0 (1.1) −2.4 (2.3)†
96.0 92.8 100
−0.3 (1.9) 0.5 (1.2) −1.1 (2.2)†
40.0 7.1 81.8
−1.5 (1.9) −1.0 (1.8) −2.0 (1.9)†
88.0 78.6 90.9
−2.5 (1.4) −1.9 (1.3) −3.1 (1.4)†
84.0 78.6 90.9
Group 1, DLB patients with CDR 0.5 rating; group 2, DLB patients with CDR 1 or 2 rating. *Percentage of hypometabolism according to regional cortical areas of mean Z score below −2.0. †P < 0.01 compared with DLB group 1.
Clinical Correlation Analysis The region-based correlations of the Z scores derived from PET and clinical data adjusted for age and sex are shown in Table 4. The results suggested that a wide network derived from PET correlated with the MMSE and CASI scores. The hypometabolism regions inversely correlated with the UPDRS scores except for the basal ganglia.
DISCUSSION In a cross-sectional manner, this study explored the affected neuronal network patterns by using complementary data from PET and TRODAT-1 in patients with DLB grounded by clinical severity. Our analysis showed that hypometabolism in the temporal regions, ACC, inferior orbital region, thalamus, and caudate nucleus feature the mild stage of DLB. Patients at the dementia stage had more extensive regional networks involving the occipital region that were good correlations with the cognitive and UPDRS domains. According to the nigrostriatal system in evaluating DAT by TRODAT-1, the abnormality in the nigrostriatal dopamine pathway appeared at mild stage of DLB but without significant progression at more advanced stages. The presence of early nigrostriatal pathway impairment and regional metabolic differences at various degrees of dementia suggested that the variable burden of a pathological protein load in the brain might cause a different temporal sequence of neuropathological changes and subsequent clinical symptoms in patients with DLB.20 Albin et al21 were the first to report FDG PET hypometabolism in the occipital association cortex and primary visual cortex regions in DLB compared with AD, followed by Imamura et al,6 who used a larger sample size and linked occipital hypometabolism with visual hallucination. Current consensus criteria for DLB suggest that a generalized low uptake on PET metabolic scan with reduced occipital activity is a supportive feature.1 Our PET study results indicated that few patients with mild DLB and most patients at more advanced stage exhibited hypometabolism in the occipital area. In most of the autopsyconfirmed DLB reports, FDG PET was visually rated,22–24 and most of the cases may have died at the end-stage of the disease. Under such conditions, the samples may have included limited cases at the early clinical stages. Base on the results of this study, occipital hypometabolism reported in relevant literature may not be the pathogenic signature of mild-stage DLB but may be considered a supportive feature of dementia-stage DLB.25 This statement is based on the observation that several studies have failed to establish occipital hypofunction in earlystage DLB patients,8–10 the lack of occipital atrophy in MRI measurements,1 and that LB seldom appear in occipital regions at early stage.26 Occipital hypometabolism helps discriminate between mild-stage and dementia-stage DLB in this study and may be served as an indicator for staging. According to the mild-stage DLB patient analysis, the hypometabolic regions were in the temporal regions, ACC, inferior orbital region, thalamus, and caudate nucleus. Interestingly, a part of the hypometabolic network in mild-stage DLB was identical to 30
www.nuclearmed.com
the anatomical structures with gray matter atrophy characterized on MRI patterns in DLB.6 These included the ACC, anterior frontal, and temporal lobes, implying that cortical metabolism might be affected earlier than cortical atrophy is. In dementia-stage patients, the hypometabolism state starts to increase diffusely, particularly in the occipital regions. Favorable correlations between the identified networks from the PET with the cognitive features determined for DLB patients were observed. This is consistent with a previous PET study in that the global cortical FDG has been found to be the best indicator of global cognitive functioning in DLB patients.27 In contrast to the findings for Parkinson disease,28 the TRODAT1 analysis revealed a reduction of signals in the caudate and putamen in DLB patients with a sensitivity and specificity of 78% and 94%.29 In this study, the decline of striatum DAT uptake develops even before the development of dementia and correlates well with clinical motor disability. The loss of ventrolateral nigra dopaminergic neuron represents a consistent feature early in DLB.30 A thorough longitudinal pathology study identified that the severity of dopamine cell loss correlated with disease duration.31 However, analysis of the TRODAT-1 binding ratio showed no significant differences between 2 functional groups. This study result supported a higher hierarchy in diagnostic ranking rather than staging by using DAT imaging. This study limitations include the fact that the diagnoses were not pathologically confirmed. Considerable clinical overlaps occurred between DLB with AD and with Parkinson dementia. Substantial effort has been expended to limit the study cohort by using clinical probable DLB.1 The second limitation is the small number of cases, and largescale trials are required to validate the results.
CONCLUSIONS The neuroimaging studies are involved in comparing FDG PET and TRODAT-1 modalities at mild and dementia stages of DLB. The FDG PET and TRODAT-1 analyses suggest that a sophisticated cortical-brainstem network is correlated with MMSE, CASI, and UPDRS. The nigrostriatal pathway abnormalities are preceded TABLE 4. Spearman Rank-Order Correlation Between Regional Cortical Z Scores and Clinical Data Z Scores
MMSE
CASI
UPDRS
Frontal Temporal Parietal Occipital Basal ganglia Thalamus
0.594* 0.842* 0.570* 0.554* 0.652* 0.664*
0.490† 0.774* 0.451† 0.566* 0.654* 0.700*
−0.557* −0.567* −0.551* −0.670* −0.283 −0.475†
*P < 0.01. †P < 0.05.
© 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Clinical Nuclear Medicine • Volume 40, Number 1, January 2015
by typical occipital hypometabolism at the mild stage of DLB. Dopamine transporter levels may serve as early diagnostic tool and FDG PET as staging indicator for DLB pathology. ACKNOWLEDGMENTS The authors thank the patients and their caregivers for their time and commitment to this research. They also thank Miss Y.T. Lin from the Department of Neurology, Chang Gung Memorial Hospital-Kaohsiung Medical Center for scheduling all of the experiments. REFERENCES 1. McKeith IG, Dickson DW, Lowe J, et al. Diagnosis and management of dementia with Lewy bodies: third report of the DLB Consortium. Neurology. 2005;65: 1863–1872. 2. Hansen L, Salmon D, Galasko D, et al. The Lewy body variant of Alzheimer's disease: a clinical and pathologic entity. Neurology. 1990;40:1–8. 3. Braak H, Ghebremedhin E, Rub U, et al. Stages in the development of Parkinson's disease-related pathology. Cell Tissue Res. 2004;318:121–134. 4. Langlais PJ, Thal L, Hansen L, et al. Neurotransmitters in basal ganglia and cortex of Alzheimer's disease with and without Lewy bodies. Neurology. 1993;43: 1927–1934. 5. McKeith IG, Burn DJ, Ballard CG, et al. Dementia with Lewy bodies. Semin Clin Neuropsychiatry. 2003;8:46–57. 6. Imamura T, Ishii K, Hirono N, et al. Visual hallucinations and regional cerebral metabolism in dementia with Lewy bodies (DLB). Neuroreport. 1999;10: 1903–1907. 7. Minoshima S, Foster NL, Sima AA, et al. Alzheimer's disease versus dementia with Lewy bodies: cerebral metabolic distinction with autopsy confirmation. Ann Neurol. 2001;50:358–365. 8. Londos E, Passant U, Brun A, et al. Regional cerebral blood flow and EEG in clinically diagnosed dementia with Lewy bodies and Alzheimer's disease. Arch Gerontol Geriatr. 2003;36:231–245. 9. Cordery RJ, Tyrrell PJ, Lantos PL, et al. Dementia with Lewy bodies studied with positron emission tomography. Arch Neurol. 2001;58:505–508. 10. Varma AR, Talbot PR, Snowden JS, et al. A 99mTc-HMPAO single-photon emission computed tomography study of Lewy body disease. J Neurol. 1997;244: 349–359. 11. Huang CW, Chang WN, Huang SH, et al. Impact of homocysteine on cortical perfusion and cognitive decline in mild Alzheimer's dementia. Eur J Neurol. 2013; 20:1191–1197. 12. Chang CC, Liu JS, Chang YY, et al. (99m)Tc-ethyl cysteinate dimer brain SPECT findings in early stage of dementia with Lewy bodies and Parkinson's disease patients: a correlation with neuropsychological tests. Eur J Neurol. 2008;15:61–65. 13. Rosen WG, Terry RD, Fuld PA, et al. Pathological verification of ischemic score in differentiation of dementias. Ann Neurol. 1980;7:486–488.
© 2014 Lippincott Williams & Wilkins
Neuroimaging in DLB
14. Chang CC, Chang YY, Chang WN, et al. Cognitive deficits in multiple system atrophy correlate with frontal atrophy and disease duration. Eur J Neurol. 2009;16: 1144–1150. 15. Folstein MF, 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. 16. Cummings JL, Mega M, Gray K, et al. The Neuropsychiatric Inventory: comprehensive assessment of psychopathology in dementia. Neurology. 1994;44: 2308–2314. 17. Chang CC, Chang WN, Lui CC, et al. Clinical significance of the pallidoreticular pathway in patients with carbon monoxide intoxication. Brain. 2011;134: 3632–3646. 18. Mirzaei S, Knoll P, Koehn H, et al. Assessment of diffuse Lewy body disease by 2-[18F]fluoro-2-deoxy-D-glucose positron emission tomography (FDG PET). BMC Nucl Med. 2003;3:1. 19. Tzourio-Mazoyer N, Landeau B, Papathanassiou D, et al. Automated anatomical labeling of activations in SPM using a macroscopic anatomical parcellation of the MNI MRI single-subject brain. Neuroimage. 2002;15:273–289. 20. Snider BJ, Norton J, Coats MA, et al. Novel presenilin 1 mutation (S170F) causing Alzheimer disease with Lewy bodies in the third decade of life. Arch Neurol. 2005;62:1821–1830. 21. Albin RL, Minoshima S, D'Amato CJ, et al. Fluoro-deoxyglucose positron emission tomography in diffuse Lewy body disease. Neurology. 1996;47:462–466. 22. Jagust W, Reed B, Mungas D, et al. What does fluorodeoxyglucose PET imaging add to a clinical diagnosis of dementia? Neurology. 2007;69:871–877. 23. Silverman DH, Small GW, Chang CY, et al. Positron emission tomography in evaluation of dementia: regional brain metabolism and long-term outcome. JAMA. 2001;286:2120–2127. 24. Hoffman JM, Welsh-Bohmer KA, Hanson M, et al. FDG PET imaging in patients with pathologically verified dementia. J Nucl Med. 2000;41:1920–1928. 25. Middelkoop HA, van der Flier WM, Burton EJ, et al. Dementia with Lewy bodies and AD are not associated with occipital lobe atrophy on MRI. Neurology. 2001; 57:2117–2120. 26. Gomez-Tortosa E, Newell K, Irizarry MC, et al. Clinical and quantitative pathologic correlates of dementia with Lewy bodies. Neurology. 1999;53:1284–1291. 27. Klein JC, Eggers C, Kalbe E, et al. Neurotransmitter changes in dementia with Lewy bodies and Parkinson disease dementia in vivo. Neurology. 2010;74: 885–892. 28. Huang WS, Lee MS, Lin JC, et al. Usefulness of brain 99mTc-TRODAT-1 SPET for the evaluation of Parkinson's disease. Eur J Nucl Med Mol Imaging. 2004; 31:155–161. 29. McKeith I, O'Brien J, Walker Z, et al. Sensitivity and specificity of dopamine transporter imaging with 123I-FP-CIT SPECT in dementia with Lewy bodies: a phase III, multicentre study. Lancet Neurol. 2007;6:305–313. 30. Zaccai J, Brayne C, McKeith I, et al. Patterns and stages of alpha-synucleinopathy: relevance in a population-based cohort. Neurology. 2008;70:1042–1048. 31. Halliday GM, McRitchie DA, Cartwright H, et al. Midbrain neuropathology in idiopathic Parkinson's disease and diffuse Lewy body disease. J Clin Neurosci. 1996;3:52–60.
www.nuclearmed.com
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
31
