Abstract
We report six demented individuals with pathologically verified diffuse Lewy body disease (DLBD) studied with fluoro-deoxyglucose positron emission tomography (FDG-PET). Three subjects had pure DLBD and three subjects had combined DLBD and Alzheimer's disease (DLBD-AD) pathology. FDG-PET revealed evidence of diffuse cerebral hypometabolism in both pure DLBD and DLBD-AD with marked declines in association cortices with relative sparing of subcortical structures and primary somatomotor cortex, a pattern reported previously in AD. Unlike AD, however, these subjects also had hypometabolism in the occipital association cortex and primary visual cortex. These findings indicate the presence of diffuse cortical abnormalities in DLBD and suggest that FDG-PET may be useful in discriminating DLBD from AD antemortem.
NEUROLOGY 1996;47: 462-466
Diffuse Lewy body disease (DLBD) is a dementing process characterized pathologically by the occurrence of Lewy bodies in subcortical, allocortical, and neocortical structures. [1-7] This disorder is described in a pure form and in conjunction with Alzheimer's disease (AD) pathology. [8,9] DLBD is recognized increasingly as a common cause of dementia, with autopsy series of demented individuals reporting a frequency of DLBD ranging from 13 to 26%. [2,3,7] If these estimates reflect true prevalence, DLBD would be the second most common form of dementia in the developed world. The epidemiology, complete clinical spectrum, and pathogenesis of DLBD are unknown.
DLBD is presently a pathologic diagnosis. Clinical criteria have been described for the antemortem diagnosis of DLBD but these criteria have not been tested in a prospective study. [10,11] The development of methods to diagnose DLBD accurately prior to death would facilitate the study of DLBD. Noninvasive imaging with fluoro-deoxyglucose positron emission tomography (FDG-PET) may be a useful method for diagnosis of AD and other dementing disorders. We report results in demented individuals with pathologically verified DLBD who were prospectively studied with FDG-PET.
Results.
Clinical.
Clinical data are summarized in Table 1.
Table 1. Clinical characteristics of DLBD and DLBD-AD patients
PET studies.
In five patients, [sup 18 F]-2-fluoro-deoxy-D-glucose (FDG) PET images were obtained using a Siemens ECAT 931/08-12 imager (CTI, Knoxville, TN) as described previously. [12-14] Arterial sampling was done in three patients for measurement of the arterial input function. One patient was studied using a Cyclotron Corporation PC4600A imager with five sections obtained at 11.5 mm separation. The protocol allowed coverage of 20 cm with section separation of 5 mm using interleaved images. The in-plane resolution was 12 mm at full-width-at-half-maximum. PET image sets obtained with arterial input function measurements were converted to quantitative glucose metabolic rate with a standard single-image approach. [15]
Reconstructed image sets were transformed into the stereotaxic coordinate system and cerebral metabolic activity was extracted into the stereotaxic surface projection format as described previously. [12,13] Regional activity was averaged using region-of-interest analysis for parietal cortex, temporal cortex, frontal cortex, occipital association cortex, anterior cingulate cortex, posterior cingulate cortex, primary somatomotor cortex, parahippocampal gyrus, thalamus, striatum, cerebellum, and pons. The locations of these regions were defined by standard stereotaxic coordinates. [13,16] Since some imaging was performed in a qualitative manner, regional metabolic values were normalized to pontine metabolic activity prior to subsequent data analysis. Analysis of results from three patients with quantitative studies revealed normal metabolic activity in the pons, consistent with prior studies and supporting normalization of this data set. [14] Normalized regional metabolic data were compared with a normal database as described previously. [13,17] Regional percent reductions of glucose metabolism of DLBD and DLBD-AD cases were transformed to Z scores using the formula [(individual regional value) - (control regional mean)/(control regional standard deviation)] to assess significance of reductions in glucose metabolism. Z scores (ZS) greater than 3, indicating regional values greater than 3 standard deviations from the control mean, were considered significant.
The normalized values indicated severe metabolic reductions in parietal (mean decrease 38%; ZS > 3), temporal (mean decrease 28%; ZS > 3), and frontal (mean decrease 28%; ZS > 3) association cortices, and in the posterior cingulate cortex (mean decrease 35%; ZS > 3). Regional metabolic activity in the primary sensorimotor cortex (mean decrease 20%; ZS < 3), thalamus (mean decrease 18%; ZS < 3), striatum (mean decrease 18%; ZS < 3), cerebellum (mean decrease 9%; ZS < 3), anterior cingulate cortex (mean decrease 22%; ZS < 3), and parahippocampal gyrus (mean decrease 13%; ZS < 3) was relatively preserved Figure 1. This profile was similar in both the DLBD and DLBD-AD groups and similar in magnitude and pattern to cerebral metabolic reductions noted previously in probable AD. [17-19] In addition, metabolic activities in the occipital association (mean decrease 32%; ZS > 3) and primary visual (mean decrease 32%; ZS > 3) cortices were markedly reduced (see Figure 1). The pattern and magnitude of regional metabolic reductions in primary visual and occipital association cortices were similar in both pure DLBD and DLBD-AD patients.
Figure 1. Representative DLBD and AD + DLBD FDG-PET images. CMRglc maps represent regional glucose uptake normalized to pontine glucose uptake. The higher the number, the greater the degree of FDG uptake. Note reduced FDG uptake in the frontal, parietal, and occipital cortices. Z-score maps represent Z scores relative to control data base. The higher the Z score, the greater the deviation from normal. In this case, high Z scores represent reductions in FDG uptake. In both DLBD and AD + DLBD images, frontal, parietal, and occipital cortices have markedly abnormal Z scores.
Neuropathology.
At autopsy, brains were hemisected and one hemisphere frozen and stored at -70 degrees C. The remaining hemisphere was fixed in neutral buffered formalin. The fixed hemisphere was sectioned coronally and inspected for gross changes. A total of 46 anatomic sites from cerebral cortex, subcortical nuclei, the hippocampal formation, and cerebellum were examined. Paraffin sections were stained with cresyl violet-luxol fast blue-eosin (CV-LFB-E), phosphotungstic acid-hematoxylin (PTAH), and Bielschowsky's silver impregnation. For identification of Lewy bodies, sections were immunostained with a polyclonal rabbit anti-human ubiquitin antiserum (Dako, Carpenteria, CA). [20]
There was no gross atrophy of the cortical mantle or white matter. The frontal, temporal, and occipital lobes appeared within normal limits. Cerebella and brainstems appeared normal in all cases with the exception of mild substantia nigra pallor.
With light microscopy, both pure DLBD and DLBD-AD cases exhibited mild to moderately severe neocortical degeneration consisting of neuronal loss and gliosis. In both groups, neocortical involvement was most severe in the entorhinal cortex, followed by anterior cingulate and parietal cortices. In the three pure DLBD cases, posterior cingulate and occipital cortices showed no pathologic changes but DLBD-AD showed moderate involvement of both these regions. In the hippocampus, both groups showed the most severe changes in CA4, followed by subiculum, CA3, CA2, and CA1. The piriform cortex, amygdala, and anterior hypothalamus showed moderately severe changes with no difference between pure DLBD and DLBD-AD. Moderate neuronal loss and gliosis of the basal nucleus of Meynert did not differ between the DLBD and DLBD-AD groups. In both groups, marked degenerative changes were present in the ventral tegmental area.
In DLBD-AD cases, neurofibrillary tangles and senile plaques, exceeding the number required to satisfy diagnostic criteria, [21] were found in the frontal, parietal, temporal, entorhinal, and posterior cingulate cortices and within the amygdala. Cortical Lewy bodies were demonstrated in the entorhinal, anterior cingulate, and insular cortices of both DLBD and DLBD-AD cases Figure 2. The overall distribution did not differ between DLBD and DLBD-AD cases although there was a trend toward a greater density of Lewy bodies in DLBD than in DLBD-AD cases.
Figure 2. Ubiquitin-immunoreactive Lewy bodies (arrows) in the anterior cingulate of a patient with pure DLBD. Magnification 240 times, before 75% reduction.
Discussion.
The clinicopathologic spectrum of Lewy body-associated disorders varies from idiopathic Parkinson's disease with Lewy body expression restricted to subcortical nuclei to so-called ``pure'' DLBD, in which both neocortical and subcortical Lewy bodies are abundant and the clinical picture is dominated by dementia. Neocortical Lewy body pathology is often associated with some pathologic features of AD, especially the occurrence of senile plaques. [2,3,7,9]
McKeith et al. [10,11] have proposed clinical criteria for DLBD (which they term senile dementia of Lewy body type) and have demonstrated good levels of sensitivity, specificity, and interrater reliability, but these criteria have not yet been verified prospectively. The development of imaging or other objective measures to supplement clinical criteria would be useful in studying DLBD.
The patients described in this report exhibited clinical features described previously as useful in distinguishing DLBD from other types of dementia, including fluctuating cognitive impairment, visual hallucinations, and parkinsonism. [10,11] As with previous reports of DLBD, however, there is not a uniform clinical presentation. Two of our patients (cases 1 and 3) presented with parkinsonism rather than cognitive symptoms. The remainder presented with dementia and some, but not all, developed parkinsonism later in their course.
All patients had evidence of hypometabolism in numerous brain regions. Marked reductions were present in many cortices with relative sparing of primary sensorimotor and subcortical regions, a pattern associated previously with AD. In addition, there was evidence of marked hypometabolism in the occipital cortex, a finding not noted in most studies of AD. [13,18] Some prior FDG-PET studies have documented occipital hypometabolism in patients suffering from Parkinson's disease with dementia, [22] raising the possibility that the patients examined in these prior studies suffered from DLBD. The observation of occipital hypometabolism in DLBD has implications for the use of FDG-PET in studies of dementing illnesses. If DLBD is relatively common, then appropriate classification of demented patients as having DLBD or AD might be necessary for performance of clinical and therapeutic studies of demented patients. FDG-PET may be useful in this context or may allow refinement of clinical criteria to discriminate DLBD and AD. Larger numbers of subjects will need to be studied to confirm or refute these conjectures and more widely available techniques such as HMPAO-SPECT imaging may be useful in evaluating these questions.
Occipital hypometabolism may not be unique to DLBD. A subgroup of AD patients has marked pathology involving both primary visual cortex and occipital association cortex. [23-26] This AD subgroup presents clinically with Balint's syndrome and other complex disturbances of vision. FDG-PET studies of a small number of these AD patients indicate the presence of primary visual cortex and occipital association cortex hypometabolism. [27,28] Differentiation of this visual form of AD from DLBD or DLBD-AD may be possible on clinical grounds.
Widespread changes in cortical metabolism in patients with pure DLBD, despite the relatively restricted distribution of Lewy body pathology, is direct evidence of diffuse cortical dysfunction. Neocortical Lewy bodies tend to occur in the cingulate cortex, entorhinal cortex, insular cortex, frontal cortex, and amygdala, suggesting preferential involvement of limbic cortices. [1,4-8] Our FDG-PET data indicate widespread functional impairment of the cortical mantle in both pure DLBD and DLBD-AD without a definite correlation with the distribution of Lewy body pathology. The anterior cingulate cortex, a region that consistently shows a high density of Lewy bodies, had relatively preserved cerebral metabolism. Some cortical regions without evident major pathology, such as the posterior cingulate cortex and occipital cortex, demonstrated major reductions in metabolic activity. The pattern of cortical hypometabolism in DLBD may reflect diachisis due to disruption of intracortical connections.
Several lines of evidence suggest that DLBD is a distinct clinicopathologic entity. The unique composition and distribution of Lewy bodies, the characteristic pattern of neuronal loss, the possible existence of characteristic clinical features, and possibly distinctive neurochemical changes all serve to differentiate DLBD from AD and other dementias. [29-35] Our data suggest that DLBD is characterized also by a relatively distinctive pattern of cortical metabolic disturbances.
Acknowledgments
We thank Kris Wernette and Bruno Giordani for assistance. We thank the anonymous reviewers and Neurology editorial staff for constructive criticism.
- Copyright 1996 by Advanstar Communications Inc.
REFERENCES
Disputes & Debates: Rapid online correspondence
REQUIREMENTS
If you are uploading a letter concerning an article:
You must have updated your disclosures within six months: http://submit.neurology.org
Your co-authors must send a completed Publishing Agreement Form to Neurology Staff (not necessary for the lead/corresponding author as the form below will suffice) before you upload your comment.
If you are responding to a comment that was written about an article you originally authored:
You (and co-authors) do not need to fill out forms or check disclosures as author forms are still valid
and apply to letter.
Submission specifications:
- Submissions must be < 200 words with < 5 references. Reference 1 must be the article on which you are commenting.
- Submissions should not have more than 5 authors. (Exception: original author replies can include all original authors of the article)
- Submit only on articles published within 6 months of issue date.
- Do not be redundant. Read any comments already posted on the article prior to submission.
- Submitted comments are subject to editing and editor review prior to posting.