The identification and differential diagnosis of dementia is especially challenging in its early stages, partly because of the difficulty involved in distinguishing it from the mild memory decline that can occur with normal aging and from mild cognitive manifestations of other neuropsychiatric conditions such as depression. This situation, which has long been clinically frustrating, has become more pressing now that several prescription medications for the treatment of mild to moderate Alzheimer’s disease (AD) are available.

As reported by the Agency for Health Care Policy and Research in 1996, “early recognition of the condition has important benefits,” and yet, “early-stage dementia is often unrecognized or misdiagnosed.”¹ This is especially troubling for dementias like Alzheimer’s that are due to neurodegenerative disease. These patients potentially have the most to gain from effective therapies that intervene as early as possible in the course of progressive, irreversible damage to brain tissue.

Conventional methods for evaluating AD patients are inadequate for making reliable diagnostic and prognostic assessments in the early stages of the disease. It has become increasingly evident during the past several years, however, that measurement of regional cerebral metabolism with PET imaging can sensitively detect such disease at the time of its earliest symptomatic expression, and even preclinically. Accordingly, about 40% of patients who undergo evaluation for cognitive decline at the University of California, Los Angeles Memory Clinic are assessed with brain PET scans. The following clinical case compares a conventional evaluation process with the dementia evaluation process as it is carried out in the Memory Clinic.

Ms. GC, who is 52 years old, came to the UCLA Memory Clinic earlier this year voicing the complaint, “It is hard to put thoughts together.” She reported a progressive decline in her memory abilities over the last several years. In the two and a half years prior to her evaluation at UCLA, the patient had had numerous visits with her internist, two visits with a neurologist to whom the internist referred her, two MR scans, ongoing therapy from a psychiatrist, and formal testing by a licensed neuropsychologist. Diagnostic conclusions were as follows:

• The last note from her internist simply stated, “Memory lapses persist.” His final assessment was “hypothyroidism, depression, fibromyalgia.”
• Ms. GC’s neurologist stated after her first visit, “Most of the patient’s symptoms are likely psychiatrically based. She is currently in therapy and being treated.” After her second visit, his note stated simply that he would need to observe her over time. His entire documented plan consisted of the statement “We will plan to see the patient in follow-up.”
• Subsequent to the neurologist’s second evaluation, a neuropsychologist evaluated the patient and stated in her summary, “The results suggest neurological difficulties. It can be said that posttraumatic stress disorder factors into the results… However, neurological problems should be ruled out ASAP.” Beyond this, the only concrete suggestion was for the patient to seek individual counseling, as well as marriage counseling, because of the patient’s “concern for her husband’s feelings about her difficulties.”

By the end of this period, the patient had been variously diagnosed with depression, hypothyroidism, anxiety, migraines, fibromyalgia, attention deficit disorder, and posttraumatic stress disorder stemming from childhood abuse. Prescribed medications included only the serotonin-specific antidepressant citalopram, which she had been taking for 18 months, levothyroxine, which she had been taking for several years for hypothyroidism secondary to Hashimoto’s thyroiditis, and estrogen replacement therapy.

With the support of her husband, a retired psychologist, Ms. GC was evaluated for the first time by a geripsychiatrist at the UCLA Memory Clinic. She was reported to be “alert, mildly anxious, euthymic, friendly, cooperative, with bright affect.” Her performance on the mini-Mental Status Examination (MMSE), however, was clearly abnormal, as she scored only 18 of a possible 30 points and remembered none of three short-term recall items.

PET and quantitative EEG were ordered, which were both completed within three weeks. The electroencephalogram report found “normal waking and drowsy EEG,” and the quantitative portion was interpreted as “normal brain maps, consistent with the patient’s conventional EEG.” In contrast, the PET study, which imaged regional brain metabolism using fluorine-18 fluorodeoxyglucose, was markedly abnormal. The scan revealed areas of moderately severe cortical hypometabolism, especially affecting bilateral midparietal cortex and left inferior frontal and temporal cortex, but sparing bilateral sensorimotor and visual cortices (Figure 1). This pattern is characteristic of neurodegenerative dementia, most commonly AD.

The PET scan was followed by an MRI study with T1-weighted, T2-weighted, and FLAIR sequences. An experienced UCLA attending neuroradiologist found all scans within normal limits, with no evidence of infarction and no white matter lesions.

A week after the PET scanning, a UCLA neuropsychologist administered a battery of tests to Ms. GC and concluded that although the other diagnoses she received in the past could be a factor in her performance, they would not explain the severity or scope of her current deficits. Based on the initial assessment, combined with results of neuroimaging and neuropsychologic testing, the geripsychiatrist diagnosed probable AD. Treatment was initiated with cholinesterase inhibition, the pharmacotherapeutic strategy of choice for treating Alzheimer’s symptoms. The patient was also referred to the Alzheimer’s Disease Association for education and support, and informed that she should refrain from driving.

At Ms. GC’s follow-up appointment one and a half months later, she and her husband reported improvement. Her husband described her as more alert and more engaged with others. Her MMSE score was 21, and she named two of three items correctly in a test of short-term recall. The psychiatrist affirmed the diagnosis as early-onset dementia, Alzheimer’s type. At their most recent visit (less than a week before this writing), the patient and her husband announced that they plan to move to northern California to be closer to their children. They will return to Los Angeles periodically to be followed in the Memory Clinic.

Neuronuclear Imaging

When neuroimaging is obtained in the evaluation of dementia, patients are usually referred for a structural imaging examination-MRI or CT of the brain. Conventional MRI or CT of patients with symptoms of dementia may be useful for identifying unsuspected clinically significant lesions, present in approximately 5% of patients.² In patients with AD, however (which is much more common), such scans are typically read as normal, or as demonstrating the nonspecific finding of cortical atrophy, or, in the worst case, as revealing ischemic changes that are misinterpreted as indicating cerebrovascular disease as the primary or sole process responsible for the cognitive decline. This misinterpretation results in failure to institute appropriate pharmacotherapy, such as donepezil, rivastigmine, or galantamine, all of which are FDA-approved only for “mild to moderate dementia of the Alzheimer’s type.”

It is unfortunately not rare for this type of misinterpretation to occur, even among expert clinicians, as is evidenced by the findings of a multicenter study that involved seven university-affiliated AD diagnostic and treatment centers. Among patients diagnosed with “vascular dementia” after clinical and structural neuroimaging evaluations, and in whom other dementia diagnoses were specifically thought to be absent, less than 30% actually had isolated cerebrovascular disease, and 55% had AD upon pathological diagnosis.³

Clinicians and researchers have accumulated substantial experience over the last two decades in using PET for the identification and differential diagnosis of dementia. Hundreds of patients with clinically diagnosed and, in some cases, histopathologically confirmed AD have been studied using PET measures of cerebral blood flow, glucose metabolism, or oxygen utilization. The principal findings, which have been the subject of review by several authors,^4-8 are as follows:

• a consistent pattern of focally decreased cerebral metabolism that involves neocortical association areas (especially parietotemporal cortex), but largely spares basal ganglia, thalamus, cerebellum, and cortex mediating primary sensory and motor functions; and
• a correlation of the extent of hypometabolism with the severity of cognitive impairment⁵ and, often, right/left hemispheric asymmetry.⁹

Blinded clinical evaluations of PET scans can differentiate patients with AD from those with other dementias and from cognitively intact individuals.^5,10,11

Single-photon emission computed tomography (SPECT) has historically been the most widely available functional brain imaging modality and thus has become the most commonly used method for the clinical evaluation of dementia. Most of the clinical and research SPECT studies for this application are perfusion-based. Although the specific radiopharmaceuticals and instrumentation differ from those used in PET, the principles of interpretation, as well as the neurobiological processes underlying its use, are similar. The primary differences are the typically poorer spatial resolution of SPECT images relative to PET images and occasional loss of the generally

parallel relationship between cortical metabolism (usually measured with PET) and perfusion in the presence of certain cerebrovascular disorders.

As might be expected, SPECT studies of AD have yielded results similar to those of PET, but with generally less sensitivity and decreased overall accuracy. That expectation has been further borne out through side-by-side comparisons of the two modalities, including studies performed with high-resolution SPECT scanners^12,13 in AD patients with mild symptoms.¹⁴ The data suggest decreased accuracy of about 15% to 20% for SPECT relative to PET.

One study assessing the prognostic accuracy of SPECT in early dementia is notable for having followed patients longitudinally for a minimum of two years after they were imaged with SPECT with technetium-99m HMPAO.¹⁵ Two groups of patients, matched for mildness at time of presentation, were compared on the basis of their SPECT findings. One group (n = 18) progressed to probable AD, and the other (n = 27) remained stable for at least two years. The progressing group could be distinguished from the stable group with a sensitivity of 78% and a specificity of 71%. A similarly designed study with PET found that progressing (n = 46) and nonprogressing (n = 42) patients could be distinguished an average of three years in advance with a sensitivity of 93% and a specificity of 74%.¹⁶

The higher diagnostic and prognostic sensitivity of PET may be especially relevant to the goal of identifying disease in its earliest stages so as to target patients for therapy while neurodegeneration is minimal.

PET Sensitivity, Specificity

Studies comparing neuropathologic examination with PET imaging are the most informative in assessing diagnostic value (Table 1). In the largest such single-institution series, Hoffman and coworkers studied 22 patients with various types of dementia (including 64% with AD alone and 9% with AD plus additional neurologic disease, identified by pathologic diagnosis).¹⁷ Visual interpretation of scans by readers blinded to clinical information yielded estimates of sensitivity of 88% to 93% and specificity of 63% to 67% for identifying the presence of AD.

A multicenter study designed to compare dementia diagnosis using FDG-PET with neuropathologic diagnosis¹⁸ collected data from an international consortium of clinical facilities that had acquired both brain FDG-PET and histopathologic data for patients undergoing evaluation for dementia. Images and pathologic data were independently classified as positive or negative for presence of a progressive neurodegenerative disease in general and AD specifically. The PET results identified patients with AD and patients with any neurodegenerative disease with a sensitivity of 94% in both cases, specificities of 73% and 78%, and overall accuracies of 88% and 92%.

Clinical Evaluation

The quality standards subcommittee of the American Academy of Neurology (AAN)² is the source of the most comprehensive guidelines and standards for the clinical evaluation of dementia. It has issued a report identifying three class I studies^19-21 in which the diagnostic value of the recommended conventional clinical assessment could be meaningfully measured (Table 2). (Class I indicates “a well designed prospective study in a broad spectrum of persons with the suspected condition, using a ‘gold standard’ for case definition, and enabling the assessment of appropriate tests of diagnostic accuracy.”)

Only one of the three studies¹⁹ focused on evaluating dementia at a relatively early stage. To be included in that investigation, participants were required to have had onset of dementia symptoms within one year of entry. All of the 134 patients evaluated underwent a complete standardized diagnostic workup comprising a comprehensive medical history and physical, neurological examination, neuropsychologic testing, laboratory tests, structural neuroimaging, and an average of three additional years of clinical follow-up. Sensitivity of this assessment for AD was 83% to 85%, while specificity was 50% to 55%. It should be emphasized that this was not the diagnostic accuracy of initial clinical evaluation, but of an entire series of evaluations repeated over a period of years.

As suggested by the literature, diagnostic workups that do not include assessments of cerebral metabolism tend to be substantially less sensitive in the diagnosis of AD. Based on the recommendation of the AAN² that the criteria of the National Institute of Neurological and Communicative Disorders and Stroke and the Alzheimer’s Disease and Related Disorders Associations (NINCDS & ADRDA) for probable AD (rather than the more inclusive possible AD) should be routinely used, clinical sensitivity appears to range in the interval of 66 ±17%, while the sensitivity of PET ranges in the interval of 91 ±3% (Tables 1 and 2).

The sensitivity of clinical evaluation can be increased to 90.5 ±5.5% by expanding the diagnosis of AD to include patients who meet NINCDS-ADRDA criteria for possible AD, but at the expense of specificity (55.5 ±5.5% in the class I studies). In contrast, at that level of sensitivity, the specificity using PET is 70 ±3%.

The AAN subcommittee reviewed nine additional studies that addressed the diagnostic accuracy of AD2 but classified them as having lower “quality of evidence” than those described in Table 1. The subcommittee found an average clinical specificity of 70%—similar to that of PET—across all these studies, while average sensitivity in that analysis was 81%, compared with the 91 ±3% reported for PET. In the two largest class II studies that uniformly employed NINCDS-ADRDA diagnostic criteria,^3,22 at a sensitivity of 90 ±1% achieved by including possible AD patients, specificity fell to 29 ±8%.

Clinical Role of FDG-PET

With a preponderance of evidence indicating improved accuracy when FDG-PET is incorporated into the diagnostic algorithm for evaluating early dementia, the question arises as to exactly when PET should be used. Beyond the specific advantages of more accurate diagnosis leading to more appropriate management, several other ramifications of the added information PET provides bear upon this issue. If accurate positive diagnoses are achieved early on, patients could be spared the repeated diagnostic tests performed over extended periods of time, and they and their physicians would less often experience the frustrations of ambiguous diagnostic conclusions. The information would also enhance families’ ability to plan for future patient care, especially in light of recent data indicating that the degree of hypometabolism identified by PET in certain affected brain regions predicts the rate of decline in memory that takes place in the years subsequent to PET examination.²³

These and other considerations support the notion that the best time to obtain PET scanning is early in the course of the clinical workup, as soon as it has been determined that the patient is an appropriate candidate for evaluation of cerebrocortical dysfunction. The guiding principle for that determination is: A patient who presents with an adverse change in cognition or behavior that has not been fully explained and fully reversed following standard diagnostic and treatment approaches should undergo PET imaging (Figure 2).

The algorithm depicted here is not intended to indicate all tests that should be done to fully evaluate all patients with cognitive impairment but, rather, represents the steps in the decision pathway that we recommend to determine whether a PET scan is obtained for patients with symptoms that could represent early manifestations of Alzheimer’s or related diseases.

Apropos to this, the recommendation of the AAN shifted between 1994 and 2001 with respect to whether CT/MRI studies should be ordered only for patients with a specific indication identified by history or examination,²⁴ or for essentially all patients undergoing initial evaluation for dementia.² This shift occurred largely because of intervening empiric documentation²⁵ showing that clinically important conditions were sometimes found on CT/MRI in patients who did not have any of the previously recommended criteria for undergoing neuroimaging. The algorithm represented in Figure 2 should not be interpreted as supporting the former position; it illustrates only those steps the evaluating physician must take before deciding whether to order FDG-PET.

The impact that such an approach can have is illustrated by the case of Ms. GC described earlier. Years after onset of cognitive symptoms and a conventional clinical workup for dementia, this patient still had received no specific diagnosis to explain her memory problems, and she was receiving no specific therapy for her dementia symptoms. Furthermore, she and her husband were offered neither a prognosis for her disorder nor meaningful support for her condition. One month after she presented to the UCLA Memory Clinic, a diagnostic PET scan was acquired, which demonstrated findings characteristic of neurodegenerative dementia, probably AD. This was corroborated by neuropsychologic testing a week later, and subsequent MRI ruled out other etiologies.

In less than three months, the patient had been diagnosed, was prescribed appropriate pharmacotherapy that resulted in some signs of improvement, was linked up with useful community resources, and was able to begin planning with her husband how they would conduct the next phase of their lives.

How much will following such an algorithm cost and, more to the point, how does that compare with the costs incurred when the additional information provided by PET is lacking? The cost of a dedicated brain PET amounts to less than the typical costs of a year of pharmacotherapy for unnecessary treatment of a patient misdiagnosed with AD, or one month’s lost productivity and independence of a patient for whom we fail to provide timely diagnosis and treatment.

In a recent examination of the extent to which the costs of scanning would be offset by the costs saved through improved diagnostic accuracy, employing the formalized tools of decision analysis,³⁷ it was found that the added diagnostic accuracy obtained by adding FDG-PET to the routine clinical evaluation of patients with early symptoms of dementia could be achieved in an economically practical manner. In fact, the attendant improvement in accuracy allowed PET scans to essentially pay for themselves for all reimbursed costs of brain PET lower than about $2000. (The amount that is currently reimbursed for brain PET is typically several hundred dollars below that figure.)

Moreover, developments in instrumentation strategies, commercial PET radiopharmaceutical distribution, and reimbursement policies are rapidly making FDG-PET scanners feasible in routine clinical settings. The path has thus been paved for a significant contribution to the assessment of patients with cognitive symptoms through application of sensitive functional neuroimaging tools to directly visualize metabolic changes in the brains of those undergoing evaluation for dementia.

Dr. Silverman has indicated that his research is supported by funds from the U.S. Department of Energy, contract no. DE-FCO3-87ER60615, and the Los Angeles Alzheimer’s Association, Turken Family Foundation Award.

References

1. U.S. Department of Health and Human Services, Public Health Service, Agency for Health Care Policy and Research. Recognition and initial assessment of Alzheimer’s disease and related dementias, clinical practice guideline no. 19, Rockville, 1996.
2. Knopman DS, DeKosky ST, Cummings JL, et al. Practice parameter: diagnosis of dementia (an evidence-based review). Report of the Quality Standards Subcommittee of the American Academy of Neurology. Neurol 2001;56:1143-1153.
3. Victoroff J, Mack WJ, Lyness SA, et al. Multicenter clinicopathological correlation in dementia. Am J Psychiatr 1995;152:1476-1484. 1997;37:95-99.
4. Friedland RP, Jagust WJ. Positron and single photon emission tomography in the differential diagnosis of dementia. In: Duara R, ed. Positron emission tomography in dementia. NY: Wiley-Liss;1990:161-177.
5. Mazziotta JC, Frackowiak RSJ, Phelps ME. The use of positron emission tomography in the clinical assessment of dementia. Sem Nucl Med 1992;22:233-246.
6. Herholz K. FDG PET and differential diagnosis of dementia. Alzheim Dis and Assoc Disord 1995;9:6-16.
7. Silverman DHS, Small GW, Phelps ME. Clinical value of neuroimaging in the diagnosis of dementia: sensitivity and specificity of regional cerebral metabolic and other parameters for early identification of Alzheimer’s disease. Clin Positron Imaging 1999;2:119-130.
8. Pietrini P, Alexander GE, Furey ML, et al. The neurometabolic landscape of cognitive decline: in vivo studies with positron emission tomography in Alzheimer’s disease. Int J Psychophysiol 2000;37:87-98.
9. Duara R, Grady C, Haxby J, et al. Positron emission tomography in Alzheimer’s disease. Neurology 1986;36:879-887.
10. Kippenhan SJ, Barker WW, Pascal S, et al. Evaluation of a neural-network classifier for PET scans of normal and Alzheimer’s disease subjects. J Nucl Med 1992;33:1459-1467.
11. Powers WJ, Perlmutter JS, Videen TO, et al. Blinded clinical evaluation of positron emission tomography for diagnosis of probable Alzheimer’s disease. Neurology 1992;42:765-770.
12. Mielke R, Pietrzyk U, Jacobs A, et al. HMPAO SPET and FDG PET in Alzheimer’s disease and vascular dementia: comparison of perfusion and metabolic pattern. Eur J Nucl Med 1994;21:1052-1060.
13. Messa C, Perani D, Lucignani G, et al. High-resolution technetium-99m-HMPAO SPECT in patients with probable Alzheimer’s disease: comparison with fluorine-18-FDG PET. J Nucl Med 1994;35:210-216.
14. Mielke R, Heiss WD. Positron emission tomography for diagnosis of Alzheimer disease and vascular dementia. J Neural Transm 1998;53(suppl):237-250.
15. Johnson KA, Jones K, Holman BL, et al. Preclinical prediction of Alzheimer’s disease using SPECT. Neurol 1998;50:1563-1571.
16. Silverman DHS, Chang CY, Cummings JL, et al. Prognostic value of regional brain metabolism in evaluation of dementia: comparison with long-term clinical outcome. J Nucl Med 1999;40(no. 5, suppl):abstr.
17. 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.
18. Silverman DH, Small GW, Kung de Aburto MA, et al. Diagnostic accuracy of FDG-PET in evaluation of dementia: International multi-center pooled brain scan and autopsy data. J Nucl Med 2000;41:63P.
19. Lim A, Tsuang D, Kukull W, et al. Clinico-neuropathological correlation of Alzheimer’s disease in a community-based case series. J Am Geriatr Soc 1999;47:564-569.
20. Jobst KA, Barnetson LP, Shepstone BJ. Accurate prediction of histologically confirmed Alzheimer’s disease and the differential diagnosis of dementia: the use of NINCDS-ADRDA and DSM-III-R criteria, SPECT, x-ray, CT, and apo E4 in medial temporal lobe dementias. Oxford Project to Investigate Memory and Aging. Int Psychogeriatr 1998;10:271-302.
21. Holmes C, Cairns N, Lantos P, et al. Validity of current clinical criteria for Alzheimer’s disease, vascular dementia and dementia with Lewy bodies. Br J Psychiatr 1999;174:45-50.
22. Galasko D, Hansen LA, Katzman R, et al. Clinical-neuropathological correlations in Alzheimer’s disease and related dementias. Arch Neurol 1994;51:888-895.
23. Small GW, Ercoli LM, Silverman DH, et al. Cerebral metabolic and cognitive decline in persons at genetic risk for Alzheimer’s disease. Proc Nat Acad Sci 2000;97:6037-6042.
24. Practice parameter for diagnosis and evaluation of dementia (summary statement). Report of the Quality Standards Subcommittee of the American Academy of Neurology. Neurol 1994;44:2203-2206.
25. Chui H, Zhang Q. Evaluation of dementia: a systematic study of the usefulness of the American Academy of Neurology’s Practice Parameters. Neurol 1997;49:925-935.
26. Silverman DHS, Gambhir SS, Huang C, et al. Decision tree analysis for comparing the cost-to-benefit ratios of algorithms for evaluation of early dementia: conventional diagnostic work-up versus assessment incorporating use of brain FDG-PET. J Nucl Med: In press.

Sidebar

CME Educational Objectives

Upon completion of this activity, participants should be able to:

• Describe in qualitative terms the differences between the information that structural (CT, MRI) and functional (SPECT, PET) neuroimaging modalities can contribute to the evaluation of patients with symptoms of dementia.
• List indications useful in selectinpatients who are most likely to benefit from the addition of structural neuroimaging and functional neuroimaging to their diagnostic work-up.
• Discuss the biological basis for the usefulness of FDG-PET in the evaluation of patients in early stages of dementing illness.
• Quantitatively compare the literature-based accuracy statistics of clinical assessments for the detection of Alzheimer’s disease, done with and without PET, and understand to what extent differences in accuracy relate to differences in sensitivity versus specificity.

To complete post-test activity and evaluation and register for earned category 1 CME credit at no charge to you, go to DiagnosticImaging.com. If you would like to receive the evaluation form and post-test by mail or fax, contact Amy Tran at 415/947-6498.

This activity has been planned and implemented in accordance with the Essential Areas and policies of the Accreditation Council for Continuing Medical Education through the joint sponsorship of CME, Inc. and CMP Healthcare Group. CME, Inc. is accredited by the ACCME to provide continuing medical education for physicians.

CME, Inc. designates this educational activity for a maximum of 1.0 hours in category 1 credit toward the AMA Physician’s Recognition Award. Each physician should claim only those hours of credit that they actually spent on the educational activity.

Activities that have been approved for AMA category 1 credit and are relevant to the radiologic sciences will be awarded category A credit on a one for one basis by the American Society of Radiologic Technologies (ASRT).