High-resolution single-photon emission computed tomography (SPECT) with brain-retained technetium-99m (99mTc)–labeled tracers may be used for 3-dimensional measurements of regional cerebral blood flow (rCBF). This article summarizes important findings in SPECT studies of Alzheimer's disease (AD). There are three distinct potential applications of SPECT in diagnosing AD: (a) as a diagnostic adjunct in patients with mild cognitive or behavioral symptoms, suggesting a possible dementia disorder; (b) as a diagnostic adjunct for demented patients in whom the history, physical examination, and laboratory studies are in agreement with a diagnosis of probable AD; and (c) for determining the relative contributions of degenerative and vascular pathology to the clinical picture in demented patients with mixed disease. A clinical diagnosis of probable AD is associated with focal perfusion deficits as measured by SPECT. Characteristically, hypoperfusion is observed in the frontal and temporoparietal association areas, whereas other brain regions are relatively spared. The changes are present in the early phases of AD. The topography of rCBF deficits displays a marked heterogeneity among patients and correlates with cognitive profiles. In patients with mild cognitive complaints, the presence of focal hypoperfusion on SPECT may increase the accuracy of the diagnosis by confirming the presence of brain disease. In patients with probable AD, the presence of temporoparietal rCBF deficits on SPECT increases the accuracy of the clinical diagnosis, in particular when associated with medial temporal lobe atrophy on cranial X-ray computed tomography (CT). The role of SPECT in diagnosing patients with mixed disease remains to be clarified.

In vivo functional brain imaging provides an opportunity to quantify and localize functional deficits associated with Alzheimer's disease (AD), in relation to dementia severity and heterogeneous cognitive profiles. Such imaging also provides a basis for distinguishing AD from other causes of dementia and for making an early diagnosis of disease. One imaging modality that can elucidate AD is positron emission tomography (PET), which is used to measure regional cerebral metabolic rates for glucose (rCMRglc) and regional cerebral blood flow (rCBF). Resting-state measurements with PET, when related to cognitive profiles in longitudinal studies, indicate that specific cognitive defects are preceded and predicted by reductions in rCMRglc in regions subserving the cognitive functions tested. Metabolic reductions and right/left metabolic asymmetries can be used to convert a “possible” to a “probable” diagnosis of AD by the National Institute of Neurological and Communicative Disorders and Stroke–Alzheimer's Disease and Related Disorders Association (NINCDS-ADRDA) criteria. Furthermore, discriminant analyses of PET metabolic patterns can identify patients at risk for AD with mild memory deficits as having probable AD. In the future, stimulation PET studies should augment the power of this discriminant analysis.

Electroencephalography is the only diagnostic instrument directly reflecting cortical neuronal functioning, and it remains an important clinical tool in the diagnosis and differential diagnosis of Alzheimer's disease (AD). Although a normal EEG is found in many patients with mild AD, a pathological EEG is an important finding, because such a result is not in line with differential diagnoses such as depression or pseudodementia within a dissociative disorder. The vast majority of patients with moderate to servere AD have a pathological EEG. A normal EEG in this patient group is more in line with subcortical dementia or frontal lobe degeneration than with AD. Compared to SPECT or routine structural brain imaging (cCT, MRT), EEG has a comparable diagnosis sensitivity and a higher specificity. For monitoring changes of brain function by serial recordings (e.g., during therapy with antidementia drugs), EEG is the best available method.

Roses: I reviewed Professor Smith's and Dr. Jobst's data with them at Oxford, and we are really not concluding anything very different. The tests of state used in their study, in this case the computed tomographic (CT) scan, work on essentially all patients with Alzheimer's disease (AD), no matter what their ApoE genotype. The positive predictor value of an E4 in the 60% or 75% of patients with AD who have a positive E4 is 97%, but for those 35% or so who are negative, the test has no value at all. So, in comparing its diagnostic usefulness, there will always be a proportion of patients for whom the test has no value.

Before I start reviewing this workshop, I would just like to draw your attention to Screening for Cognitive Impairment in General Practice—Towards a Consensus. This is a document that was prepared as a result of an Alzheimer's Disease International (ADI) workshop in Buenos Aires in 1995. ADI is now expanding its Medical and Research Board and would like to invite experts from the International Psychogeriatric Association to be on the advisory panel. The collective role of this ADI advisory panel will be to offer expert advice on all aspects of Alzheimer's disease (AD) management to family carers of individuals with this condition. I hope you will be able to participate in this important initiative.

Currently used criteria for neuropathological diagnosis of Alzheimer's disease (AD) allow only the distinction of fully developed AD or ill-defined cases with less severe pathologic alterations. There is a need for a staging procedure that is able to supplement the currently used criteria, to allow for sufficient differentiation between the initial and intermediate stages of the illness, and to provide a characterization of the degree of involvement in the brain tissue of “nondemented, age-related controls.” The destructive process underlying AD is characterized by a typical distribution pattern of brain changes that is specific with respect to area, lamina, and even cell type. The most conspicuous change is the progressive deposition of abnormal proteins, both between and within the nerve cells. Conventional staining methods to identify these deposits lack sensitivity and specificity. Silver methods (specifically the Gallyas silver-iodide technique for neurofibrillary changes and the Campbell-Switzer silver-pyridin technique for brain amyloid) are by far the best suited to diagnostic work. They are inexpensive, simple to use, and far more reliable. This article describes the patterns of staining typically found with each of these techniques in the brains of patients with progressively severe AD and describes how the specific changes observed can be used as a staging system for diagnostic purposes.

The Consortium to Establish a Registry for Alzheimer's Disease (CERAD), a multicenter longitudinal study, has worked to establish standardized instruments for the evaluation of individuals clinically diagnosed as having Alzheimer's disease (AD). Since its formation in 1986, clinical neuropsychological, neuroimaging, and neuropathology assessment batteries have been developed. The neuropathology protocol not only establishes levels of certainty for the diagnosis of AD, but also facilitates correlations with clinical, genetic, and other data. We find a high rate of diagnostic accuracy: the clinical diagnosis of AD was confirmed in 176 of 201 (87.6%) CERAD dementia subjects. Coexistent vascular lesions, usually infarcts, were found in 32% of these confirmed AD cases, and 23% had coexisting Parkinson's disease (PD) changes (nigral degeneration and Lewy bodies at any site). In the remaining 25 cases (12.4%), neuropathologists attributed the primary cause of dementia to other conditions. In studies comparing neuropathology protocols for AD, several groups have found that the CERAD diagnosis most closely correlates with measures of dementia severity, such as the Mini-Mental State Examination (MMSE). Others have adopted the CERAD neuropathology protocol for multicenter brain banking, citing its relative simplicity and adaptability among laboratories, the experience of many AD centers with the protocol, and its application to other dementing conditions. The CERAD data set and batteries are increasingly used for a wide array of clinical, neuropathological, and genetic studies.

Reisberg: I would like to summarize very briefly. In the Consortium to Establish a Registry for Alzheimer's Disease (CERAD) study, one third of patients with Alzheimer's disease (AD) had concomitant cerebral vascular disease. This finding supports the original work of Tomlinson and Blessed, who indicated a similar percentage of patients with AD also had mixed cerebral vascular disease upon neuropathologic examination.

Within our workshop we discussed neuropathological assessments in different cultures and countries, evaluation of dementia type, and availability of autopsy. We agreed that evaluation of dementia cases should not be limited to the use of diagnostic criteria; but neuropathologic assessment should be utilized where appropriate. We agreed that it is vital to communicate this concept to general practitioners (GPs) and physicians working in nursing homes, where there is often a high proportion of dementia patients. When neuropathologic assessments are conducted correctly, they provide both quantitative and qualitative information about cell and tissue damage. The assessment should preferably be performed by a neuropathologist in order that a diagnosis plus an estimation of disease severity and extent can be made.

Positive predictive value (PPV) estimates how often a particular test predicts the eventual diagnosis, and it is the basis of any test that attempts to establish a diagnosis. Calculating the PPV depends directly on the entrance criteria for considering the diagnosis and the exit criteria of the generally accepted “gold standard” for establishing the ultimate diagnosis. For a relatively common disease such as AD, the practical usefulness of a diagnostic test also relates to the stage of the disease at which the test can be applied. An accurate and predictive test only has practical value when it can be used early in the course of the disease. Apolipoprotein E (Apo E) is the first genetic locus identified as a susceptibility polymorphism for a common complex disease using positional or candidate cloning strategies. There are three common alleles (ε2 to ε4) and six possible genotypes; about 30% of the population carries an ε4 allele, each additional copy of which is associated with an increased risk of AD and earlier age of onset distribution. In patients with cognitive impairment of the Alzheimer type, who have been diagnosed with possible or probable AD, the presence of an Apo E-ε4 allele is likely in more than 50% of patients for whom the PPV is greater than 97%. Apo E genotyping is the only diagnostic adjunct that has been subjected to rigorous measurements of specificity, sensitivity, PPV, and NPV. In the past, testing of dementia patients was directed at “ruling out” other treatable causes. Results from the initial studies of the PPV of Apo E genotyping for AD showed that the PPV of Apo E genotyping was greater than 97%. The paradigm for AD diagnosis is changing. Apo E genotyping can provide useful diagnostic information and that accuracy will improve as other susceptibility genetic risk factors are identified.

Alzheimer's disease (AD) is histopathologically characterized by the presence of numerous neurons with neurofibrillary tangles of paired helical filaments (PHF) in the neocortex, particularly the hippocampus. The major protein subunit of PHF, which also accumulate as neuropil threads and dystrophic neurites of the neuritic (senile) plaques, is the microtubule-associated protein τ. τ in AD brain is abnormally hyperphosphorylated. In addition to hyperphosphorylation, some of the AD τ is also ubiquitinated. The level of τ in AD brain is elevated severalfold, and this increase is in the form of the abnormally phospholylated τ. The levels of conjugated ubiquitin are also increased severalfold in AD brain. The cerbrospinal fluid (CSF) levels of both τ and ubiquitin are elevated in AD. However, some of the normal elderly and non-AD neurological control cases also show elevated levels of these two proteins in the CSF. At present, it is not clear whether the overlap in CSF τ or ubiquitin values between AD and the control groups is due to some unknown heterogeneity of AD, inclusion of presymptomatic AD cases in the control group, or lack of specificity of the CSF elevation of these markers to AD. Thus, until the results of prospective studies on many patients with AD and normal controls become available, the CSF τ and ubiquitin levels can be used only as additional tools in the psychometric and clinical diagnosis of AD.

This article describes the results of studies conducted to determine the usefulness of reflex changes as markers of disease severity in Alzheimer's disease (AD). Standardized and quantified muscle stretch reflexes, cutaneous, reflexes, and developmental (primitive) reflexes were studied in normal older adults, in individuals with mild memory impairment, and in patients with AD, in all clinical severity stages as assessed with the Global Deterioration Scale (GDS), the Mini-Mental State Examination (MMSE), and the Functional Assessment Staging (FAST) procedure. Changes in frequency and intensity of these individual reflex variables, as well as of variables consisting of combinations of these individual reflexes, appeared to be sensitive indicators of the progression of AD. These neurological reflex variables showed high Pearson correlations with the GDS (.72), the MMSE (.74), and the FAST (.80). Standardized quantified neurological reflex measures are useful as noncognitive, education-independent, and culture-independent markers of the course of AD.

To determine the association between cognitive dysfunction and motor behavior in older adults, 41 cognitively normal elderly (NL), 25 nondemented patients exhibiting mild cognitive impairment (MI) and at risk for future decline to dementia, and 25 patients with mild (early) Alzheimer's disease (AD) were examined using a wide array of motor/psychomotor and cognitive assessments. The three groups were recruited from an aging and dementia research center and were composed of well-characterized physically healthy volunteers, with similar ages and gender distributions. The outcome measures included 16 motor/psychomotor tests categorized a priori into gross, fine, and complex, as well as eight cognitive tests of memory and language. Relative to the NL group, MI individuals performed poorly on cognitive, fine, and complex motor measures but not on gross motor tests; AD patients performed worse on cognitive and all motor domains. Differences in complex motor function persisted after adjustment for performance on cognitive and on less complex motor tests. Classification analyses showed similar accuracies in discriminating NL from MI and NL from AD cases for both complex motor (79% and 92% accuracy, respectively) and cognitive tests (80% and 93% accuracy, respectively). Less complex motor tests produced poorer accuracies. Among nondemented subjects, education correlated with several cognitive scores but no motor scores. These results indicate that motor impairment is an important aspect of cognitive decline in older adults. Motor/psychomotor assessments were found to be comparably sensitive to traditional tests of cognitive function in identifying persons affected by the earliest stages of AD pathology and may improve identification of at-risk nondemented elderly, especially among diversely educated individuals.

Korczyn: Professor Roses, first, does ApoE genotype relate to cognitive decline or pathological change in the progression of Alzheimer's disease (AD)? They are certainly closely linked but they are not identical, because patients can develop severe pathological alterations without being demented. Second, because the positive predictive value is 100%, theoretically, a patient with ApoE4 cannot develop any other dementing disease.

Workshop 6 discussed peripheral markers from the perspective of the medical profession as a whole. We have considered each of the assessment methods and the limitations and problems associated with them, and graded them according to their usefulness, comprehensiveness, and cultural specificity.

Forette: I was just trying to highlight that it is dangerous to state that computed tomographic (CT) scans are not an essential tool for the specialist to rule out conditions other than Alzheimer's disease.

Burns: This meeting had a number of aims, one being to review and assimilate current knowledge about Alzheimer's disease (AD), with particular reference to primary care. The introduction of guidelines for GPs would perhaps be much better implemented at a national or even regional level.