HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Figures Only Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Norfray, J. F. Articles by Provenzale, J. M. Search for Related Content PubMed PubMed Citation Articles by Norfray, J. F. Articles by Provenzale, J. M. Social Bookmarking                      What's this?

Alzheimer's Disease: Neuropathologic Findings and Recent Advances in Imaging

¹ Chicago Northside MRI Center, 2818 N Sheridan Rd., Chicago, IL 60657.
² Department of Radiology, Duke University Medical Center, Durham, NC 27710.

Received November 11, 2002; accepted after revision August 4, 2003.

 
Address correspondence to J. F. Norfray.

Introduction

Top Introduction Neuropathologic Findings Clinical Criteria for Diagnosis... Imaging Techniques for... Treatment of Alzheimer's Disease Monitoring Treatment of... Summary References

 
The neuropathologic features of Alzheimer's disease are characterized by progressive cortical loss in neural pathways that are important for memory. The clinical diagnosis of Alzheimer's disease requires documentation of progressive decline in memory by means of longitudinal clinical examinations and neuropsychologic tests. Using clinical criteria yields greater than 90% sensitivity in diagnosing dementia of any type but is less than 70% accurate in the specific diagnosis of Alzheimer's disease [1]. Cross-sectional imaging techniques such as CT and MRI provide a valuable noninvasive method for detection of the cortical atrophy that is typically seen in Alzheimer's disease. MR spectroscopy and metabolic imaging techniques, such as positron emission tomography (PET), are even more specific and provide evidence of unique metabolic changes in memory pathways, thus supplying a wealth of new information for investigators. Finally, newer imaging techniques, such as brain activation studies, may provide even greater diagnostic accuracy and at an early stage of the disease. By quantifying the progression of neuronal loss or dysfunction, these imaging techniques also allow investigators to monitor the success of drug therapies. This article will explain the fundamental pathophysiologic processes in Alzheimer's disease, discuss the development of imaging techniques used for assessment of patients with Alzheimer's disease, and explain how advanced imaging techniques can be used to monitor drug therapy.

Neuropathologic Findings

Top Introduction Neuropathologic Findings Clinical Criteria for Diagnosis... Imaging Techniques for... Treatment of Alzheimer's Disease Monitoring Treatment of... Summary References

 
Alois Alzheimer, a German neuropathologist, originally described Alzheimer's disease in 1906. In an autopsy of a 56-year-old woman with severe dementia, Alzheimer noted senile plaques outside neurons and neurofibrillary tangles within neurons [2]. One hundred years later, the presence of plaques and neurofibrillary tangles as determined by biopsy or autopsy remains the only definitive method for diagnosing Alzheimer's disease [3].

A few general comments are in order before discussing the specific neuropathology of senile plaques and neurofibrillary tangles. Both entities are caused by the deposition of abnormal proteins. In the case of senile plaques, the abnormal protein is amyloid; in the case of neurofibrillary tangles, the abnormal protein is designated as tau. Both tau and amyloid assemble into insoluble aggregates of neurofibrillary tangles and senile plaques that destroy neurons. For reasons that are not fully known, both proteins are deposited along the course of cortical memory pathways dominated by pyramidal cells and cause neuronal destruction along these pathways. Senile plaques, which are extraneuronal in location, initiate an inflammatory response that destroys neurons by lysis of cell membranes of adjacent neurons and their dendrites. These lysed neurons may or may not contain neurofibrillary tangles. The production of senile plaques and neurofibrillary tangles is progressive, and later stages of Alzheimer's disease show a greater number of neurofibrillary tangles involved with senile plaques. The intraneuronal neurofibrillary tangles fill the cytoplasm of axons and dendrites, preventing glucose transport and causing neuronal death. Depending on the specific genetic mutation (and therefore the specific disease process), senile plaques may precede neurofibrillary tangles, and vice versa. For instance, in late-onset Alzheimer's disease, neurofibrillary tangles are deposited before senile plaques, whereas in Down's syndrome amyloid is deposited before neurofibrillary tangles. The presence of neurofibrillary tangles in some (but not all) senile plaques as well as the variable timing of each type of deposit, indicates that neurofibrillary tangles and senile plaques develop independently.

Senile plaques are extraneuronal accumulations of dense amyloid. The blue–black staining of amyloid by iodine in the presence of sulfuric acid indicates the presence of carbohydrates, one of the major components of amyloid. A second major component is a protein fragment from a surface receptor of neurons [4–7] (Fig. 1). Neuronal protein and carbohydrates are assembled into dense beta sheets of amyloid in a process termed the "amyloid cascade" [8]. Free radicals initiate the amyloid cascade by oxidation of the neuronal protein fragments that create binding sites for carbohydrates [4, 9–11]. Free radicals are generated in neurons when the cytochrome oxidase activity of mitochondria fails to meet neuronal energy demands. Because Alzheimer's disease is characterized pathologically by both senile plaques and neurofibrillary tangles, and free radicals are solely implicated in the formation of senile plaques, accumulations of free radicals alone cannot be considered to cause Alzheimer's disease. The mitochondrial genotype can mutate during the aging process, thereby decreasing cytochrome oxidase activity and potentiating senile and presenile dementias [10]. The aggregation of proteins and carbohydrates into dense beta sheets of amyloid produces the senile plaques of Alzheimer's disease [4] (Fig. 2). In response to amyloid formation, microglia attack the dense core of insoluble amyloid and initiate the "inflammatory cascade" [12]. Thus, there are two cascades in Alzheimer's disease—one producing, and the other trying to remove, the senile plaques. More than 60 secretory products are released in an attempt to digest the abnormal beta sheet extracellular protein [12]. Unfortunately, the end product of one of these secretory products, complement, lyses adjacent normal ("bystander") neurons and their axons and dendrites. The glial scar may embed neurons that have not lysed if complement secretion at the periphery of the plaque is insufficient to lyse the cell. Microglia stimulate astrocytes to generate a glial scar that walls off the amyloid and embeds both the intact and the lysed axons and dendrites. Finally, the activated astrocytes release growth factors that promote clonal expansion of the microglia, thus further increasing the cellular inflammatory response.

View larger version (21K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 1. —Schematic drawing shows origin of protein fragments. Amyloid precursor protein (APP) is neuronal cell membrane surface receptor and is source of amyloid and nonamyloid protein fragments. Both fragments are released from surface receptor by different enzymatic secretases (, ß, ). Activated amyloid B protein (ABP) forms amyloid. Activated APP soluble (APPs) increases length and branching of axons and dendrites. Both fragments contain binding sites (BS) that activate fragments. Heparin sulfate (HS) bonds to both ABP and APPs; nerve growth factor (NGF) bonds to APPs.

View larger version (10K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 2. —Schematic drawing shows formation of amyloid. Activation of amyloid B protein (ABP) binding site by nidus bonds heparin sulfate of carbohydrate (GAGs), causing aggregation of individual APB fragments into beta sheet amyloid.

As opposed to the dense plaques that are extraneuronal in location, neurofibrillary tangles are intraneuronal structures that were first identified by Alois Alzheimer [2]. Neurofibrillary tangles are caused by the hyperphosphorylation of single tau protein filaments into paired helical filaments [13] (Fig. 3). Tau is a normal protein that is specific to axons and that functions in intracellular transport and movement of synaptic vesicles. Neurofibrillary tangles impede axonal transport of vesicles and inevitably produce neuronal lysis within 3–4 years [13]. Neurofibrillary tangles can be identified in the fragmented axons and dendrites at the periphery of the neuritic plaques on histologic specimens by a characteristic silver staining. In essence, then, neuronal loss in Alzheimer's disease proceeds along two pathways. The unremitting development of dense plaques outside the neuron, with their activated microglia and astrocytes, causes neuronal loss [12]. Inside the neuron, the neurofibrillary tangles accelerate neuronal death by impeding intraneuronal transport [13].

View larger version (8K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 3. —Schematic drawing shows formation of neurofibrillary tangles (NFT). Tau, protein of microtubules, forms NFT when hyperphosphorylated by enzyme microtubule affinity regulating kinase (MARK). MARK is coupled to inositol signaling pathway (ISP).

Various mutations (or polymorphisms) of genes that cause Alzheimer's disease have recently been identified [14–16]. These mutations increase the rate of formation of senile plaques and neurofibrillary tangles and thereby influence the age of onset of dementia. Mutations on chromosomes 1 and 14, called "presenilins," generate abnormal enzymes that release protein fragments from the receptor of the cell membrane and form amyloid; these mutations cause aggressive early-onset familial Alzheimer's disease [14]. Chromosome 19 controls the gene for apolipoprotein E, a lipophilic plasma protein synthesized and secreted by astrocytes, which is internalized by neurons and affects the phosphorylation of tau. One apolipoprotein E mutation, the apolipoprotein E4 mutation, promotes the hyperphosphorylation of tau and has been shown to increase the risk of the more common late-onset familial Alzheimer's disease and sporadic Alzheimer's disease by forming neurofibrillary tangles and inducing aggregations of amyloid that result in plaque development [15, 16]. In addition, mutations on chromosome 21, in the amyloid precursor protein gene, alter the normal proteolysis of the receptor protein and cause the senile plaques seen in Down's syndrome as well as those seen in early onset, familial, autosomal dominant Alzheimer's disease [14]. Finally, spontaneous mitochondrial mutations occur that increase with advancing age and involve essentially all individuals older than 80 years. These devastating mutations allow free radicals to leak from mitochondria and precipitate soluble amyloid into senile plaques.

In patients with Alzheimer's disease, neurofibrillary tangles propagate along specific neuronal circuits involved in memory and cognition. The stages of propagation have been categorized into the transentorhinal, limbic, and neocortical stages of Alzheimer's disease; each histologic stage correlates with a clinical stage [17, 18]. In the transentorhinal stage, neurofibrillary tangles initially selectively develop in the parahippocampal gyrus; at this stage, patients have normal memory. During the limbic stage, neurofibrillary tangles dramatically increase in the parahippocampal gyrus and progress into the hippocampus; early memory impairment ("senior moments") is seen clinically. In the neocortical stage, the neurofibrillary tangles advance into the temporoparietal cortex and eventually involve the entire neocortex; at this point severe dementia is evident [17, 18].

Clinical Criteria for Diagnosis of Alzheimer's Disease

Top Introduction Neuropathologic Findings Clinical Criteria for Diagnosis... Imaging Techniques for... Treatment of Alzheimer's Disease Monitoring Treatment of... Summary References

 
Clinical criteria provide a greater than 90% sensitivity for diagnosing dementia of any type, including Alzheimer's disease, in a specialized clinical setting such as a memory disorders clinic, but they have a specificity of less than 70% for the actual diagnosis of Alzheimer's disease [1]. The sensitivity and specificity are worse in the community setting because uniform diagnostic criteria are not used. The clinical standards used to diagnose Alzheimer's disease were first defined in 1984. These standards require insidious onset; gradual progression of memory deficits; focusing on early deficits of recent memory; and later impairment of orientation, judgment, problem solving, community and home living, and personal care [3]. In addition to a diagnosis of clinically definite Alzheimer's disease, the diagnoses of probable Alzheimer's disease and possible Alzheimer's disease can be rendered; both these diagnoses allow variations in onset, presentation, and course. The term "probable Alzheimer's disease" refers to memory deficits seen on neuropsychologic testing (Mini-Mental State Examination [MMSE] score 23) and progressive worsening of memory and deficits in two or more cognitive functions, as documented by clinical and neuropsychologic testing. The clinical diagnosis of possible Alzheimer's disease includes the presence of a second disease that may produce dementia but that is not likely the cause of Alzheimer's disease [3].

Finally, a transitional stage between normal aging and Alzheimer's disease, termed "mild cognitive impairment," is diagnosed on the basis of early mild memory impairment, absence of deficits in cognitive domains other than memory, and progressive decline in cognitive functions leading to the development of dementia [19, 20]. Longitudinal clinical tests have shown that 20% of mild cognitive impairment patients with early memory deficits, 36% of mild cognitive impairment patients with two additional cognitive deficits, and 60% of mild cognitive impairment patients with three additional cognitive deficits will develop Alzheimer's disease [19].

The cognitive functions discussed are measured by a battery of clinical and psychometric tests, such as the MMSE and clinical dementia rating (which requires an independent informant such as a caretaker) [21, 22]. Normal cognitive performance scores for the MMSE are greater than or equal to 27.6 (maximum score, 30); and for the clinical dementia rating, less than 0.5 on a scale of 0–3 [21, 22].

A number of limitations of clinical criteria for diagnosing Alzheimer's disease are evident. First, clinical evaluation is difficult in patients with severe depression, aphasia, and apraxia. Second, longitudinal clinical testing is necessary to distinguish between the early memory loss in normal aging (which has a slow progression) and very early Alzheimer's disease (which has a more rapid progression). Third, other degenerating dementias (e.g., primary progressive aphasia, posterior cortical atrophy, corticobasal degeneration, and frontotemporal dementia) also show a similar decline in cognitive functions and can mimic Alzheimer's disease on the basis of clinical criteria alone [1, 23]. For these reasons, increased emphasis has been placed on the role of neuroimaging techniques to establish the diagnosis of Alzheimer's disease.

Imaging Techniques for Assessment of Alzheimer's Disease

Top Introduction Neuropathologic Findings Clinical Criteria for Diagnosis... Imaging Techniques for... Treatment of Alzheimer's Disease Monitoring Treatment of... Summary References

 
In 1984, the Alzheimer's Disease and Related Disorders Association decided that the primary use of imaging techniques for evaluation of Alzheimer's disease should be to exclude other causes of dementia. However, some of the participants in that group predicted the potential value of imaging in the direct diagnosis of Alzheimer's disease [3]. With the development of therapeutic strategies for Alzheimer's disease, early diagnosis and monitoring of Alzheimer's disease have taken on increased importance. The following discussion highlights recent advances in imaging techniques as a tool not only for accurate diagnosis of Alzheimer's disease but also for identification of preclinical stages of Alzheimer's disease that are not discernible using solely clinical testing. The latter indication is targeted toward the identification of patients in whom the effects of neuronal loss seen on imaging studies precede decline in cognitive performance scores and in whom early intervention may slow disease progression. Although a number of imaging techniques have been advanced for assessment of Alzheimer's disease patients, few comparative trials have been performed. Furthermore, although in many studies differences between groups of healthy subjects and groups of Alzheimer's disease patients have been observed, application of findings to individual cases can be difficult and is best accomplished when a number of different findings are present.

Structural (Anatomic) Imaging
In 1986, 2 years after publication of the clinical criteria used to diagnose Alzheimer's disease, CT was first used to depict Alzheimer's disease–related atrophy of the medial temporal lobe cortex [24]. Linear measurements of the cerebrospinal fluid spaces surrounding the medial temporal lobes (reflecting temporal lobe atrophy) on CT were found to provide 88% accuracy in diagnosis of Alzheimer's disease [24]. The multiplanar capability of MRI has also been adapted to making the diagnosis of Alzheimer's disease; MRI provides a diagnostic accuracy of approximately 87%, which is very similar to that of CT [25]. Angled axial images parallel to, and angled coronal images perpendicular to, the long axis of the hippocampus can show enlargement of the parahippocampal fissures that reflects atrophy in both the hippocampus (cornu ammonis and dentate gyrus) and the parahippocampal gyrus (entorhinal cortex and subiculum) [25]. One study showed that the rate of enlargement of the parahippocampal fissures on MRI can, in fact, predict the development of Alzheimer's disease. However, Alzheimer's disease cannot be accurately distinguished from frontotemporal dementia on the basis of enlargement of these fissures. Another valuable MRI technique is MR surface display, which shows preferential atrophy in the temporoparietal regions and allows distinction from the frontal and temporal lobe volume loss seen in frontotemporal dementia [26, 27]. MR volumetric analysis of the hippocampus, parahippocampus, and amygdala has provided a means for identifying mild cognitive impairment by sensitive depiction of hippocampal volume loss [28] (Figs. 4A, 4B and 5A, 5B). In older patients with mild cognitive impairment, hippocampal atrophy determined by MR volumetric measurements is predictive of subsequent conversion to Alzheimer's disease, although the sensitivity and specificity of this technique have not yet been determined [28]. Even slight MR volumetric atrophy in Alzheimer's disease is functionally relevant because loss of verbal learning and recall are associated with left parahippocampal atrophy in patients in whom speech function is localized to the left hemisphere. Although the sensitivity and specificity of voxel compression mapping of serial MRIs are not yet determined, research studies using this technique have also indicated that minimal atrophy of the medial temporal lobes, posterior cingulate, and temporoparietal cortex precedes the appearance of clinical Alzheimer's disease [29].

View larger version (87K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 4A. —Measurement of hippocampal volumes using T1-weighted 3D spoiled gradient-recalled echo sequence in 75-year-old woman at time of diagnosis of mild cognitive impairment. This patient's cognitive status remained stable during 49-month follow-up period. (Reprinted with permission from [28]) Unenhanced coronal T1-weighted image shows relatively normal volume in body of hippocampus. Boundary used for hippocampal measurement is outlined in white.

View larger version (85K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 4B. —Measurement of hippocampal volumes using T1-weighted 3D spoiled gradient-recalled echo sequence in 75-year-old woman at time of diagnosis of mild cognitive impairment. This patient's cognitive status remained stable during 49-month follow-up period. (Reprinted with permission from [28]) Unenhanced coronal T1-weighted image at region more anterior than that shown in A also shows relatively normal volume in head of hippocampus. Boundary used for hippocampal measurement is outlined in white.

View larger version (84K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 5A. —Measurement of hippocampal volumes using T1-weighted 3D spoiled gradient-recalled echo sequence in 70-year-old woman at time of initial diagnosis of mild cognitive impairment. This patient progressed to dementia over 43-month follow-up period. Comparison of these images with those shown in Figure 4A, 4B shows possible use of volumetric imaging to predict subsequent course. (Reprinted with permission from [28]) Unenhanced coronal T1-weighted image shows relatively decreased volume in body of hippocampus. Boundary used for hippocampal measurement is outlined in white. Note that compared with patient shown in Figure 4A, hippocampal volume is decreased even though this patient is 5 years younger than patient shown in Figure 4A, 4B.

View larger version (80K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 5B. —Measurement of hippocampal volumes using T1-weighted 3D spoiled gradient-recalled echo sequence in 70-year-old woman at time of initial diagnosis of mild cognitive impairment. This patient progressed to dementia over 43-month follow-up period. Comparison of these images with those shown in Figure 4A, 4B shows possible use of volumetric imaging to predict subsequent course. (Reprinted with permission from [28]) Unenhanced coronal T1-weighted image at region more anterior than that shown in A also shows decreased volume in head of hippocampus. Boundary used for hippocampal measurement is outlined in white. Again, size of hippocampus in this patient is decreased compared with patient who did not progress to dementia, as shown in Figure 4B.

PET
PET allows assessment of glucose metabolism using FDG. In Alzheimer's disease, decreased glucose utilization is characteristically seen at rest in the temporal and parietal lobes; the decrease in glucose metabolism has been shown to clinically correlate with a decrease in MMSE score and to histologically correlate with the density of senile plaques and neurofibrillary tangles [30, 31]. In the brain, glucose metabolism provides about 95% of the energy required for proper function. Activated neurons have increased glucose consumption. Because neurons have a limited ability to store glucose, increased cerebral blood flow is needed to deliver the glucose to meet increased metabolic needs. Therefore, in the brain, glucose metabolism is coupled to blood flow. Astrocytes are the means of transporting glucose from capillaries to neuronal synapses. In Alzheimer's disease and other dementias, a decrease in synapses is seen before neuronal death. In dementing illnesses, decreased glucose metabolism is seen as a result of both decreased glucose transport (because of a decrease in the number of synapses) and a decrease in the number of neurons. Therefore, the reduction in cerebral metabolism is a true loss of neurons and synapses rather than simply the decreased glucose metabolism expected from a smaller volume of tissue seen in patients with atrophy [32].

The pattern of bilateral temporoparietal hypometabolism is a characteristic feature seen in Alzheimer's disease and allows Alzheimer's disease to be distinguished from other forms of dementia with which it can be confused on clinical grounds, such as frontotemporal lobe dementia [31, 33]. This distinction is possible with FDG PET because decreased metabolism is seen at the sites of neuronal loss that differ between the two forms of dementia (i.e., temporoparietal regions for Alzheimer's disease and frontal and temporal lobes for frontotemporal dementia). However, FDG PET cannot reliably distinguish Alzheimer's disease from another form of dementia, Parkinson's disease dementia, because the pattern of decreased glucose metabolism is similar in both diseases [31]. Vascular dementia also has a characteristic pattern of multiple focal deficits, but the diagnosis can be reached only in the context of a clinical history of stepwise decline in cognition and multiple cardiovascular risk factors and the appropriate CT or MRI findings.

FDG PET allows the diagnosis of Alzheimer's disease in patients who are difficult to characterize by clinical criteria alone [31]. For instance, FDG PET can show glucose metabolism deficits in patients at risk for Alzheimer's disease (e.g., those with the apolipoprotein E4 allele and mild cognitive impairment) [34]. The combination of reduced metabolic rates and genetic risk factors provides a means for preclinical Alzheimer's disease detection. The sensitivity and specificity of the clinical diagnosis of probable Alzheimer's disease have been compared with those for FDG PET. In one study of 22 pathologically verified cases of dementia, the clinical diagnosis was reached using a combination of cognitive testing and conventional MRI, and FDG PET studies were interpreted blindly without conventional MRI [31]. Using this method, the sensitivity and specificity of diagnosis based on clinical criteria were 64% and 87%, respectively, compared with a sensitivity and specificity of FDG PET of 87% and 62%, respectively. The inclusion of a patient with dementia related to Parkinson's disease likely contributed to the lower specificity of FDG PET because both Alzheimer's and Parkinson's diseases exhibit bilateral hypometabolism in the temporal and parietal lobes on FDG PET. The specificity of both clinical testing and FDG PET were improved when patients with Parkinson's disease were excluded using conventional MRI. In another study, the sensitivity and specificity of FDG PET for the diagnosis of Alzheimer's disease were 94% and 73%, respectively [35]. Perhaps more important, the accuracy of FDG PET to diagnose Alzheimer's disease in the early stages of the disease (when patients have questionable or mild dementia) has been reported to be 89%, with a sensitivity of 95% and a specificity of 71%. Furthermore, among patients with a dementia (not solely Alzheimer's disease) that subsequently progressed after initial imaging, FDG PET had a prognostic sensitivity of 93% and a prognostic specificity of 76% [35]. Finally, PET can show additional areas of decreased glucose metabolism during performance of memory tasks in visual memory pathways and auditory memory pathways in patients with suspected Alzheimer's disease [30, 36].

Because glucose metabolism is coupled to cerebral blood flow, changes in cerebral blood flow can also be seen in Alzheimer's disease and measured on PET images using H₂¹⁵O. Reduced cerebral blood flow can be seen in the region between the prefrontal cortex and the hippocampus, which is thought to indicate an interruption in functional connectivity that is the cause of short-term memory impairment in Alzheimer's disease patients [37]. PET can also show areas of increased cerebral blood flow in patients with Alzheimer's disease that are not seen in the healthy elderly, which may indicate recruitment of neural circuits in maintaining memory [38]. Combining cerebral blood flow measurements on PET during the resting state with measurement of cerebral blood flow during performance of active memory tasks can also allow depiction of additional alternative networks formed to compensate for memory deficits [38].

A new PET tracer, 2-(1-{6-[(2[¹⁸F] fluoroethyl) (methyl) amino]-2napthyl}ethylidene) malononitrile (¹⁸F FDDNP), developed at the University of California Los Angeles, crosses the blood–brain barrier and binds to senile plaques and neurofibrillary tangles [39]. For the first time, the presence of plaques and tangles can be imaged in living patients with Alzheimer's disease. The molecular probe binds to plaques and tangles and thereby reflects their location and density. The greater degree of tracer accumulation correlates with memory performance scores and decreased glucose metabolism on FDG PET [39].

One of the important roles for PET in assessing patients with Alzheimer's disease is detection of loss of neurotransmitter receptors in pathways that underlie memory and cognition. Development of senile plaques and neurofibrillary tangles in the striatum and hippocampus causes dopamine and serotonic receptor loss that can be detected using PET. Likewise, the formation of senile plaques and tangles in the basal forebrain and entorhinal areas causes early loss of cholinergic neurons and nicotinic (cholinergic) receptors [40]. In the normal state, these nicotinic receptors are related to acquisition and retention of verbal and nonverbal information, which is the basis for the "cholinergic hypothesis" for the cognitive symptoms of Alzheimer's disease. The decrease in the number of these receptors in Alzheimer's disease provides the rationale for cholinesterase therapy in treating Alzheimer's disease–related memory deficits [40]. Many PET radiolabeled tracers are deployed in investigating neurotransmitter deficits [23]. Nicotinic receptor loss can be depicted on PET using radioligands such as ¹¹C-nicotine and has been correlated with a decrease in cognitive function. On this basis, PET has proven valuable in trials of memory-sparing agents by allowing direct visualization of drugs synthesized to improve both receptor binding and selective activation of receptors needed in memory retention [40]. Improved receptor binding and selective activation of receptors may be seen on PET, which is an investigational tool being used at this point to develop novel therapeutic agents rather than to assess individual patients.

Single-Photon Emission Computed Tomography (SPECT)
The principal agents used in brain SPECT of patients with Alzheimer's disease are ^99mTc hexamethylpropyleneamine oxime (^99mTc HMPAO) and ^99mTc ethylcysteinate dimer (^99mTc ECD) [41]. These agents accurately measure cerebral blood flow and therefore are valuable in identifying the reduced cerebral blood flow in the temporoparietal region seen in patients with Alzheimer's disease [42, 43]. The accuracy of SPECT in the diagnosis of Alzheimer's disease ( 92%) is greater than that afforded by clinical criteria alone ( 74%) [44]. Positive findings on SPECT improve the clinical diagnosis of both probable and possible Alzheimer's disease by showing reduced cerebral blood flow in the hippocampus and temporoparietal regions [45]. These findings correlate with the severity of cognitive impairment as assessed by the MMSE, having Pearson's correlation coefficients in the range of 0.39–0.54 [46, 47]. The progression of Alzheimer's disease has been tracked on serial SPECT examinations by showing a progressive reduction of cerebral blood flow in the left hippocampus, parahippocampus, and cerebral association cortex [48].

SPECT adds specificity by allowing the distinction of Alzheimer's disease from frontotemporal dementia and Jakob-Creutzfeldt disease [42–44]. However, SPECT cannot reliably distinguish between Alzheimer's disease and the dementia associated with Parkinson's disease because both entities show a temporoparietal distribution of hypoperfusion [44].

Advanced MRI Techniques
On perfusion MRI using dynamic susceptibility contrast-enhanced MRI, reduced relative cerebral blood volume can be seen in the temporal and parietal regions of patients with Alzheimer's disease, which correlates with the cerebral blood flow reduction seen on PET and SPECT [49–51]. The sensitivity of dynamic susceptibility MRI in diagnosing both moderate and mild Alzheimer's disease is reported to be approximately 90% [50, 51] (Figs. 6 and 7). These findings indicate that perfusion MRI may be a valuable adjunct to structural MRI for assessment of suspected Alzheimer's disease.

View larger version (144K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 6. —Potential role of cerebral blood volume images for aiding in diagnosis of Alzheimer's disease. Cerebral blood volume map obtained using T2*-weighted perfusion imaging is shown in 83-year-old woman with probable Alzheimer's disease and Mini-Mental State Examination score of 11 (normal, 27.6). Cerebral blood volume in temporoparietal regions (outlined in white) measured approximately 65% of reference cerebral blood volume in cerebellum in this patient (compared with mean of 112% in healthy control subjects), indicating diminished cerebral blood volume in same regions that are generally seen in Alzheimer's disease patients studied with positron emission tomography and SPECT. (Reprinted with permission from [50])

View larger version (20K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 7. —Plot shows comparison of ratio of relative cerebral blood volume measurements in temporoparietal regions to relative cerebral blood volume measurements in cerebellar cortex (x-axis) with Mini-Mental State Examination (MMSE) scores (y-axis) in patients with possible or probable Alzheimer's disease. Comparison shows that, as level of cognitive functioning decreases, relative cerebral blood volume measurements in these brain regions also tend to decrease. Note that although distinct separation between relative cerebral blood volume in patients with normal scores and those with lowered scores is not seen, clear trend toward lower relative cerebral blood volume in subjects with lower scores is evident. (Courtesy of Renshaw P, Belmont, MA)

Diffusion-weighted MRI detects alterations in microscopic water motion within tissues, which is usually measured by the apparent diffusion coefficient value. Alterations in cell membranes and myelin sheaths can produce altered rates of water diffusion within tissues. An increased rate of water diffusion has been reported in the hippocampal gyri of Alzheimer's disease patients, likely caused by the disruption of membranes and myelin sheaths combined with the fragmentation of axons and dendrites or by expansion of extracellular fluid from the inflammatory response [12, 52]. In addition, increased apparent diffusion coefficient values have been found in the hippocampal gyri of patients with mild cognitive impairment (MMSE = 26.8 ± 2.9) and probable Alzheimer's disease (MMSE = 18.6 ± 5.5), suggesting that monitoring apparent diffusion coefficient values may help as a research tool in monitoring the progression of Alzheimer's disease, although at this point its usefulness in evaluating individual subjects has not been determined [52]. The exact role, if any, that diffusion imaging will play in assessing Alzheimer's disease is presently uncertain.

MR brain activation studies that assess the degree of brain activation during memory tasks using a blood oxygenation level–dependent technique offer another promising method for studying Alzheimer's disease. Studies of Alzheimer's disease patients have shown decreased activation of the left medial temporal lobe during auditory memory tasks and decreased activation of the right parietal region in response to visual memory tasks [53, 54]. Perhaps most impressive is the fact that MR brain activation studies appear to be able to show additional regions of activation in cognitively healthy subjects who have the additional risks of a family history of Alzheimer's disease and the presence of the apolipoprotein E4 allele [55, 56]. Such findings may represent the development of alternate or compensatory memory circuits that maintain the level of cognition in the presence of early neuronal damage to principal pathways (Fig. 8A, 8B). These findings lend credence to the notion that neuropathologic changes in Alzheimer's disease precede cognitive impairment. MR brain activation studies may eventually attain an established role in the early diagnosis of Alzheimer's disease.

View larger version (112K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 8A. —Differences in sites and sizes of regions of brain activation after memory task of patient with mild cognitive impairment and control subject. 67-year-old man shown in B was one of group of subjects with mild cognitive impairment who were found to have statistically less activation than control subjects in five of seven brain regions studied; similar trend that was not statistically significant was seen in other two brain regions. Brain activation studies in both groups were obtained in identical fashion and were analyzed using same statistical threshold, smoothing and clustering. (Courtesy of Petrella J, Durham, NC) Brain activation functional imaging performed during memory task in 71-year-old man with normal cognitive functioning, on no medication, shows robust activation (depicted in colored voxels) in right frontal lobe.

View larger version (119K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 8B. —Differences in sites and sizes of regions of brain activation after memory task of patient with mild cognitive impairment and control subject. 67-year-old man shown in B was one of group of subjects with mild cognitive impairment who were found to have statistically less activation than control subjects in five of seven brain regions studied; similar trend that was not statistically significant was seen in other two brain regions. Brain activation studies in both groups were obtained in identical fashion and were analyzed using same statistical threshold, smoothing and clustering. (Courtesy of Petrella J, Durham, NC) Brain activation functional imaging performed during memory task in 67-year-old man with mild cognitive impairment (receiving no medication) shows decreased activation (depicted in color voxel overlay) in right frontal lobe compared with individual shown in A. Note activation in left frontal lobe, which suggests possible contralateral hemisphere compensatory activation.

MR Spectroscopy
MR spectroscopy is a technique that can monitor the metabolic changes that accompany brain structural changes and therefore would be expected to play a valuable role in the imaging of Alzheimer's disease. The presence of senile plaques is associated with neuronal loss that is reflected in a decrease in the neuronal marker N-acetylaspartate [12, 57]. This N-acetylaspartate decrease can be readily detected on MR spectroscopy. An additional finding seen in patients with Alzheimer's disease is an increase in the metabolite myo-inositol, which is a metabolite of astrocytes that is produced by the astrocytes surrounding senile plaques [57, 58]. On the basis of these findings, MR spectroscopy has been used to improve the accuracy of diagnosing probable Alzheimer's disease [57] (Fig. 9A, 9B, 9C). Using the ratio of myo-inositol to N-acetylaspartate allows patients with Alzheimer's disease to be distinguished from healthy subjects with a sensitivity of 83% and a specificity of 98% [57]. Because sites of neuronal degeneration differ in Alzheimer's disease and frontotemporal dementia, SPECT and MR spectroscopy have allowed investigators to distinguish between the two entities using imaging techniques [44, 59]. Neuronal loss occurs in both Alzheimer's disease and frontotemporal dementia. However, in early Alzheimer's disease, neuronal loss is greatest in the medial temporal lobe, occurs to a lesser degree in the temporoparietal regions, and spares the frontal association areas. In early frontotemporal dementia, neuronal loss occurs in both the frontal lobes and medial temporal lobes. Discriminate analysis can be used to distinguish frontotemporal dementia patients from Alzheimer's disease patients and healthy subjects in 91% of cases [59]. Evidence also exists that MR spectroscopy can be used to monitor the progression of Alzheimer's disease by measuring progressive decreases in N-acetylaspartate, and that this finding is more sensitive than volume loss for assessing disease progression [60]. Furthermore, decreases in N-acetylaspartate and increases in myo-inositol have been shown to correlate with a decrease in cognitive function as measured by the MMSE [61, 62]. Finally, MR spectroscopy studies have allowed distinction of subjects with normal aging from those with mild cognitive impairment based on a decrease in N-acetylaspartate and an increase in myo-inositol [63, 64] (Fig. 10).

View larger version (63K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 9A. —MR spectroscopic findings in 78-year-old woman with probable Alzheimer's disease. Coronal T2-weighted fast spin-echo image shows severe cortical atrophy in both hippocampi.

View larger version (134K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 9B. —MR spectroscopic findings in 78-year-old woman with probable Alzheimer's disease. Axial T2-weighted spin-echo image (stimulated echo acquisition mode, TR/TE, 1,500/30; average of 128 scans; mixing time, 13.7 msec) shows bitemporal volume loss. Voxel for MR spectroscopy has been drawn on parietal lobe and posterior cingulate gyrus.

View larger version (14K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 9C. —MR spectroscopic findings in 78-year-old woman with probable Alzheimer's disease. Spectrum from voxel depicted in B shows decreased N-acetylaspartate (NAA) levels, with NAA-to-creatine (CRE) ratio of 1.13, indicating neuronal loss. Elevated myo-inositol (MI) levels, with MI-to-creatine ratio of 0.70, are consistent with gliosis. CHO = choline.

View larger version (11K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 10. —Spectra from posterior cingulate cortex (point resolved spectroscopy, TR/TE, 2,000/30; average of 128 scans) obtained from control subject (top row), patient with mild cognitive impairment (MCI, middle row), and patient with probable Alzheimer's disease (AD, bottom row). Myo-inositol (MI) to creatine (Cr) ratios show progressive increase in mild cognitive impairment (0.65) and Alzheimer's disease (0.81) compared with control subject (0.48). Note also significant reduction in N-acetylaspartate (NAA) to creatine (Cr) ratios in Alzheimer's disease subject (1.33) compared with patient with mild cognitive impairment (1.60) and control subject (1.71). CHO = choline. (Reprinted with permission from [64])

Studies comparing the sensitivity and specificity of clinical diagnosis of Alzheimer's disease and advanced MRI techniques (e.g., MR perfusion imaging) have not yet been performed. However, the potential for improved sensitivity and specificity of advanced MRI is supported by the ability to identify patients with mild cognitive impairment who will progress to Alzheimer's disease using structural imaging techniques, FDG PET, and brain activation studies [28, 29, 34, 56]. Some investigators maintain that specificity can be increased by positive findings on two advanced imaging techniques [63]. For instance, bilateral mesial temporal sclerosis and Alzheimer's disease will show similar imaging findings with atrophy of both hippocampal gyri, decrease glucose metabolism, and decrease N-acetylaspartate; however, short-TE MR spectroscopy differentiates between the two by showing a decrease myo-inositol in mesial temporal sclerosis versus an increase myo-inositol in Alzheimer's disease [57, 65]. Currently, one of the authors uses conventional MRI, volumetric imaging of the hippocampi, and MR spectroscopy to add diagnostic specificity.

Treatment of Alzheimer's Disease

Top Introduction Neuropathologic Findings Clinical Criteria for Diagnosis... Imaging Techniques for... Treatment of Alzheimer's Disease Monitoring Treatment of... Summary References

 
As mentioned earlier, cholinesterase inhibitors are a major therapy for treating memory loss symptoms in patients with Alzheimer's disease by prolonging the presence of acetylcholine at remaining cholinergic synapses, thereby aiding in signal transmission. However, this treatment is solely symptomatic because, although memory initially improves, it subsequently deteriorates because senile plaques and neurofibrillary tangles continue to develop and destroy neurons [66]. For this reason, increased emphasis has recently been placed on attempts to develop therapies that are designed to halt progression of structural changes (e.g., amyloid production, the inflammatory cascade, and dephosphorylation of neurofibrillary tangles).

Various means of decreasing the rate of amyloid production in Alzheimer's disease are under investigation, generally in animal models. Enzyme inhibitors have been developed that can prevent the release of abnormal protein from membrane receptors, which interrupts the amyloid cascade at various locations along the protein pathway [66]. Amyloid production can also be decreased by the use of antioxidants (e.g., vitamins A, C, and E) that remove free radicals and thereby inhibit carbohydrate binding to protein fragments, which is a fundamental process in the amyloid production cascade [67] (Fig. 1). Another method for decreasing the rate of amyloid formation is the application of peptides that can inhibit amyloid fibrillogenesis in vivo as well as dissemble preformed fibrils in vitro [68]. In a similar manner, a vaccine of amyloid B protein has been shown in a mouse model to prevent and reverse senile plaque formation [69]. During the phase IIA trial of another vaccine, AN1792, several individuals with mild or moderate Alzheimer's disease developed clinical signs of inflammation of the central nervous system. To overcome the effects of antibody binding to plaques that activate phagocytic activity in surrounding microglia, antibodies to soluble ß-amyloid have been developed that show an 80–90% reduction in ß-amyloid burden in the hippocampus and neocortex as well as reduced microglial activation [70]. Finally, treatment with ₂-macroglobulin, which complexes with amyloid, has been shown to clear amyloid from tissues [66].

The major method under investigation for altering the inflammatory cascade in Alzheimer's disease is the use of nonsteroidal antiinflammatory drugs. One such drug, indomethacin, arrests symptoms in patients with Alzheimer's disease by interrupting the astrocytic response by interleukin-1 antagonists [66, 71]. Long-term use of other nonsteroidal antiinflammatory drugs is reported to protect against Alzheimer's disease by inhibiting the enzymes cyclooxygenase-1 and cyclooxygenase-2 [72].

Another technique designed to retard progression of Alzheimer's disease is aimed at preventing hyperphosphorylation of tau. Divalent cations (e.g., manganese and magnesium) increase the tau phosphatases of the neuron and thereby promote release of phosphates from neurofibrillary tangles. The abnormal paired filaments dissociate when the phosphate bonds are broken and functioning tau is reformed [73].

Monitoring Treatment of Alzheimer's Disease

Top Introduction Neuropathologic Findings Clinical Criteria for Diagnosis... Imaging Techniques for... Treatment of Alzheimer's Disease Monitoring Treatment of... Summary References

 
The well-documented difficulties of using solely cognitive testing in Alzheimer's disease treatment trials (e.g., large number of subjects needed per treatment arm, difficulty in measuring clinical outcomes, and prolonged duration of trials) have been previously discussed. Neuroimaging techniques have been advocated as a means to obviate these difficulties. Researchers must decide the particular type of trials and patient populations in which to use these techniques. Generally speaking, research trials using imaging techniques are performed during a period of 12–36 months [28, 29, 34].

Patient selection depends on the objective of a particular trial: for example, stabilization of symptoms in patients with Alzheimer's disease (or probable Alzheimer's disease) as opposed to Alzheimer's disease prevention in patients with mild cognitive impairment. Another instance would be patients with probable Alzheimer's disease being selected for trials of therapies that slow or interrupt the progressive neuronal loss [74]. The choice of the specific patient population to be studied, in turn, can modify the size of the sample of subjects needed to achieve meaningful results—for instance, in a study assessing the efficacy of a therapy designed to prevent progression from mild cognitive impairment to fully established Alzheimer's disease, a smaller number of patients with mild cognitive impairment would be needed if subjects had one or more genetic risks for Alzheimer's disease, such as the presence of the apolipoprotein E4 allele [34].

The choice of brain region to be imaged in trials of Alzheimer's disease progression is, to some extent, governed by the disease stage being studied. Hippocampal volume loss is one of the earliest structural changes to be seen in Alzheimer's disease. Therefore, studies of mild cognitive impairment often use a technique such as 3D volumetric studies that is sensitive to early changes in hippocampal size (i.e., a 3D TI-weighted spoiled gradient-recalled echo sequence with 124 contiguous partitions and 1.5- to 1.6-mm slice thickness) [28, 29]. In one study, the 3D data sets were oriented perpendicular to the axis of the left hippocampus [28]. The borders of 40–50 hippocampal sections were manually traced, interpolated into a 512 x 512 matrix, and viewed at a magnification of 2. Hippocampal volumes were normalized for interpatient variation and head size by dividing hippocampal volumes by total intracranial volumes. A normal database was created that allowed hippocampal volumes among individual patients to be corrected for sex, age, and head size. A significant relationship was seen between premorbid hippocampal volume and the development of Alzheimer's disease. Only 15% of subjects with normal hippocampal volume developed Alzheimer's disease compared with a 46% rate of development of Alzheimer's disease among subjects whose hippocampal volumes were in the lowest range [28]. In another study, follow-up 3D brain volumes were aligned to the initial volume by an automatic image registration technique that optimized voxel accuracy. Global brain atrophy was calculated with a brain boundary shift integral that allowed calculations in a few hours on a desktop workstation. Voxel-compression color maps were created to highlight location and degree of volume loss and were then overlaid on structural images. The subtraction technique showed cerebral volume loss in all individuals over time characterized by shrinkage of the cortex and expansion of the ventricles. Mean global rates of loss in individuals with mild or moderate Alzheimer's disease were significantly greater than those in healthy controls (2.29% per year vs 0.24% per year; p = 0.001), with striking changes in the medial temporal lobe [29]. These two changes indicate that substantial focal brain changes occur in Alzheimer's disease and that these changes can be sensitively measured. Trials studying agents designed to prevent development of Alzheimer's disease in patients with mild cognitive impairment focus on measuring volumes of the hippocampus as well as other selected brain regions, such as the medial temporal lobe and posterior cingulate cortex [29]. These regions are also the focus of studies using advanced MRI techniques, as described earlier. Patients with probable Alzheimer's disease have generally already experienced hippocampal volume loss by the time of diagnosis. Therefore, trials studying agents to prevent Alzheimer's disease progression generally focus on other brain regions, such as the cortex of the temporal and parietal lobes, using techniques such as 3D volumetric measurements [75].

Summary

Top Introduction Neuropathologic Findings Clinical Criteria for Diagnosis... Imaging Techniques for... Treatment of Alzheimer's Disease Monitoring Treatment of... Summary References

 
Whereas the histologic changes that underlie Alzheimer's disease have been known for approximately 100 years, the physiologic principles and genetic mechanisms governing formation of senile plaques and neurofibrillary tangles continue to be elucidated. During the past decade, neuroradiologists have moved beyond solely structural imaging techniques that assess volume loss in Alzheimer's disease. At present, physiologic imaging techniques, such as PET, MR hemodynamic imaging, and MR spectroscopy, are being coupled with ever more sensitive volumetric techniques for the assessment of Alzheimer's disease. The sensitivity and specificity profiles of these techniques will be better assessed in coming years, and methods to apply study findings to individual patients will need to be developed. These techniques will likely be increasingly used to improve the accuracy of diagnosis of Alzheimer's disease, identify patients at risk for developing Alzheimer's disease, and monitor the various therapeutic agents that are being developed to retard the progression of the disease.

References

Top Introduction Neuropathologic Findings Clinical Criteria for Diagnosis... Imaging Techniques for... Treatment of Alzheimer's Disease Monitoring Treatment of... Summary References

 

1. Kramer JH, Miller BL. Alzheimer's disease and its focal variants. Semin Neurol2000; 20:447 –454[Medline]
2. Thomas M, Isaac M. Alois Alzheimer: a memoir. Trends Neurosci 1987;10:306 –307
3. McKhann G, Drachman D, Folstein M, Katzman R, Price D, Stadlan EM. Clinical diagnosis of Alzheimer's disease: report of the NINCDS-ADRDA work group under the auspices of Department of Health and Human Services Task Force on Alzheimer's Disease. Neurology 1984;34 : 939–944[Abstract/Free Full Text]
4. Kisilevsky R, Fraser P. Proteoglycans and amyloid fibrillogenesis. In: Costa PP, ed. The nature and origin of amyloid fibrils. Chichester, England: Wiley, 1996:58 –72
5. Beyreuther K, Multhaup G, Masters CL. Alzheimer's disease: genesis of amyloid. In: Costa PP, ed. The nature and origin of amyloid fibrils. Chichester, England: Wiley, 1996:119 –131
6. Kang J, Lemaire H-G, Unterbeck A, et al. The precursor of Alzheimer's disease amyloid A4 protein resembles a cell-surface receptor. Nature 1987;325:733 –736[Medline]
7. Yankner BA. Mechanisms of neuronal degeneration in Alzheimer's disease. Neuron1996; 16:921 –932[Medline]
8. Hardy JA, Higgins GA. Alzheimer's disease: the amyloid cascade hypothesis. Science1992; 256:184 –185[Free Full Text]
9. Selkoe DJ, Yamazaki T, Citron M, et al. The role of APP processing and trafficking pathways in the formation of amyloid beta-protein.Ann N Y Acad Sci1996; 777:57 –64[Abstract]
10. Parker WD Jr, Davis RE. Primary mitochondrial DNA defects as a causative event in Alzheimer's disease. In: Beal MF, Howell N, Bodis-Wollner I, eds. Mitochondria and free radicals in neurodegenerative diseases. New York, NY: Wiley-Liss, 1997:319 –333
11. Smith MA, Sayre LM, Perry G. Morphological aspects of oxidative damage in Alzheimer's disease. In: Beal MF, Howell N, Bodis-Wollner I, eds. Mitochondria and free radicals in neurodegenerative diseases. New York, NY: Wiley-Liss, 1997:335 –342
12. McGeer PL, McGeer EG. The inflammatory response system of brain: implications for therapy of Alzheimer and other neurodegenerative diseases. Brain Res Rev1995; 21:195 –218[Medline]
13. Goedert M. Tau protein and the neurofibrillary pathology of Alzheimer's disease. Ann N Y Acad Sci1996; 777:121 –131[Abstract]
14. Selkoe DJ. Translating cell biology into therapeutic advances in Alzheimer's disease. Nature1999; 399[suppl]:A23 –A31[Medline]
15. Mahley RW, Nathan BP, Pitas RE. Apolipoprotein E: structure, function, and possible roles in Alzheimer's disease. Ann N Y Acad Sci 1996;777:139 –145[Abstract]
16. Blacker D, Tanzi RE. The genetics of Alzheimer's disease: current status and future prospects. Arch Neurol1998; 55:294 –296[Abstract/Free Full Text]
17. Braak H, Braak E. Frequency of stages of Alzheimer-related lesions in different age categories. Neurobiol Aging1997; 18:351 –357[Medline]
18. Hyman BT, Van Hoesen GW, Damasio AR. Memory-related neural systems in Alzheimer's disease: an anatomic study. Neurology1990; 40:1721 –1730[Abstract/Free Full Text]
19. Morris JC, Storandt M, Miller JM, et al. Mild cognitive impairment represents early-stage Alzheimer disease. Arch Neurol2001; 58:397 –405[Abstract/Free Full Text]
20. Bozoki A, Giordani B, Heidebrink JL, Berent S, Foster NL. Mild cognitive impairment predicts dementia in nondemented elderly patients with memory loss. Arch Neurol2001; 58:411 –416[Abstract/Free Full Text]
21. Folstein MF, Folstein SE, McHugh PR. "Minimental state": a practical method for grading the cognitive state of patients for the clinicians. J Psychiat Res1975; 12:189 –198[Medline]
22. Morris JC. The clinical dementia rating (CDR): current version and scoring rules. Neurology1993; 43:2412 –2413[Free Full Text]
23. Cummings JL. Cognitive and behavioral heterogeneity in Alzheimer's disease: seeking the neurobiological basis. Neurobiol Aging 2000;21:845 –861[Medline]
24. LeMay M, Stafford JL, Sandor T, et al. Statistical assessment of perceptual CT scan ratings in patients with Alzheimer type dementia. J Comput Assist Tomogr1986; 10:802 –809[Medline]
25. deLeon MJ, Golomb J, George AE, et al. The radiologic prediction of Alzheimer disease: the atrophic hippocampal formation. AJNR 1993;14:897 –906[Abstract]
26. Kitagaki H, Mori E, Yamaji S, et al. Frontotemporal dementia and Alzheimer disease: evaluation of cortical atrophy with automated hemispheric surface display generated with MR images. Radiology1998; 208:431 –439[Abstract/Free Full Text]
27. Kitagaki H, Hirono N, Ishii K, Mori E. Corticobasal degeneration: evaluation of cortical atrophy by means of hemispheric surface display generated with MR images. Radiology2000; 216:31 –38[Abstract/Free Full Text]
28. Jack CR, Petersen RC, Xu YC, et al. Prediction of AD with MRI based hippocampal volume in mild cognitive impairment. Neurology 1999;52:1397 –1403[Abstract/Free Full Text]
29. Fox NC, Crum WR, Scahill RI, Stevens JM, Janssen JC, Rossor MN. Imaging of onset and progression of Alzheimer's disease with voxel-compression mapping of serial magnetic resonance images. Lancet2001; 358:201 –205[Medline]
30. Desgranges B, Baron J-C, delaSayette V, et al. The neural substrates of memory systems impairment in Alzheimer's disease: a PET study of resting brain glucose metabolism. Brain1998; 121:611 –631[Abstract/Free Full Text]
31. 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[Abstract/Free Full Text]
32. Bokde AL, Pietrini P, Ibanez V, et al. The effect of brain atrophy on cerebral hypometabolism in the visual variant of Alzheimer disease. Arch Neurol2001; 58:480 –486[Abstract/Free Full Text]
33. Santens P, DeBleecker J, Goethals P, et al. Differential regional cerebral uptake of 18F-fluoro–2-deoxy-D-glucose in Alzheimer's disease and frontotemporal dementia initial diagnosis. Eur Neurol2001; 45:19 –27[Medline]
34. Small GW, Ercoli LM, Silverman DH, et al. Cerebral metabolic and cognitive decline in persons at genetic risk for Alzheimer's disease. Proc Natl Acad Sci U S A2000; 97:6037 –6042[Abstract/Free Full Text]
35. Silverman DH, Small GW, Chang CY, et al. Positron emission tomography in evaluation of dementia: regional brain metabolism and long-term outcome. JAMA2001; 286:2120 –2127[Abstract/Free Full Text]
36. Pietrini P, Furey ML, Alexander GE, et al. Association between brain functional failure and dementia severity in Alzheimer's disease: resting versus stimulation PET study. Am J Psychiatry1999; 156:470 –473[Abstract/Free Full Text]
37. Grady CL, Furey ML, Pietrini P, Horwitz B, Rapoport SI. Altered brain functional connectivity and impaired short-term memory in Alzheimer's disease. Brain2001; 124:739 –756[Abstract/Free Full Text]
38. Becker JT, Mintum MA, Aleva K, Wiseman MB, Nichols T, DeKosky ST. Compensatory reallocation of brain resources supporting verbal episodic memory in Alzheimer's disease. Neurology1996; 46:692 –700[Free Full Text]
39. Shoghi-Jadid K, Small GW, Agdeppa ED, et al. Localization of neurofibrillary tangles and betaamyloid plaques in brains of living patients with Alzheimer disease. Am J Geriatr Psychiatry2002; 10:24 –35[Abstract/Free Full Text]
40. Newhouse PA, Potter A, Kelton M, Corwn J. Nicotinic treatment of Alzheimer's disease. Biol Psychiatry2001; 49:268 –278[Medline]
41. vanDyck CH, Huie Lin C, Smith O, et al. Comparison of technetium-99m-HMPAO and technetium-99m-ECD cerebral SPECT images in Alzheimer's disease. J Nucl Med1996; 37:1749 –1755[Abstract/Free Full Text]
42. Vlasenko A, Petit-Taboué MC, Bouvard G, Morello R, Derlon J-M. Comparative quantitation of cerebral blood volume: SPECT versus PET. J Nucl Med1997; 38:919 –924[Abstract/Free Full Text]
43. Jagust WJ, Budinger TF, Reed BR. The diagnosis of dementia with single-photon emission computed tomography. Arch Neurol 1987;44:258 –262[Abstract]
44. Read SL, Miller BL, Mena I, Kim R, Itabashi H, Darby A. SPECT in dementia: clinical and pathological correlation. J Am Geriatr Soc 1995;43:1243 –1247[Medline]
45. Jagust W, Thisted R, Devous MD Sr, et al. SPECT perfusion imaging in the diagnosis of Alzheimer's disease: a clinical-pathologic study. Neurology 2001;56:950 –956[Abstract/Free Full Text]
46. Bartenstein P, Minoshima S, Hirsch C, et al. Quantitative assessment of cerebral blood flow in patients with Alzheimer's disease by SPECT. J Nucl Med1997; 38:1095 –1101[Abstract/Free Full Text]
47. Rodriguez G, Vitali P, Calvini P, et al. Hippocampal perfusion in mild Alzheimer's disease. Psychiatry Res2000; 100:65 –74