Current Concepts DCL

Current Concepts in Mild Cognitive ImpairmentRonald C. Petersen, PhD, MD; Rachelle Doody, MD, PhD; Alexander Kurz, MD; Richard C. Mohs, PhD;John C. Morris, MD; Peter V. Rabins, MD; Karen Ritchie, MPsych, PhD; Martin Rossor, MA, MDFRCP;Leon Thal, MD; Bengt Winblad, MD

T he field of aging and dementia is focusing on the characterization of the earliest stagesof cognitive impairment. Recent research has identified a transitional state betweenthe cognitive changes of normal aging and Alzheimer’s disease (AD), known as mildcognitive impairment (MCI). Mild cognitive impairment refers to the clinical condi-

tion between normal aging and AD in which persons experience memory loss to a greater extentthan one would expect for age, yet they do not meet currently accepted criteria for clinically prob-able AD. When these persons are observed longitudinally, they progress to clinically probable ADat a considerably accelerated rate compared with healthy age-matched individuals. Consequently,this condition has been recognized as suitable for possible therapeutic intervention, and severalmulticenter international treatment trials are under way. Because this is a topic of intense interest,a group of experts on aging and MCI from around the world in the fields of neurology, psychiatry,geriatrics, neuropsychology, neuroimaging, neuropathology, clinical trials, and ethics was con-vened to summarize the current state of the field of MCI. Participants reviewed the world scien-tific literature on aging and MCI and summarized the various topics with respect to available evi-dence on MCI. Diagnostic criteria and clinical outcomes of these subjects are available in the literature.Mild cognitive impairment is believed to be a high-risk condition for the development of clinicallyprobable AD. Heterogeneity in the use of the term was recognized, and subclassifications were sug-gested. While no treatments are recommended for MCI currently, clinical trials regarding poten-tial therapies are under way. Recommendations concerning ethical issues in the diagnosis and themanagement of subjects with MCI were made. Arch Neurol. 2001;58:1985-1992

The boundary between normal aging andearly or mild Alzheimer’s disease (AD) isan area of intense interest for theoreticaland practical reasons. Basic research, suchas the identification of secretase inhi-bitors and the development of an immu-nization model for the prevention ofamyloid deposition, underscores the im-portance of developing techniques forearly detection.1,2 Parallel with these

endeavors, clinical research aimed at iden-tifying the earliest signs of cognitive im-pairment has progressed.3-5

A slight impairment in cognitive func-tion, typically memory, with otherwisenormal performance has been designatedmildcognitive impairment (MCI)andhasbe-come a topic of considerable research.3,4,6

From June 24 through 26, 1999, investi-gators from around the world gathered inChicago, Ill, at theCurrentConcepts inMildCognitive Impairment conference to re-view the world’s literature regarding theconcept of MCI and to suggest directionsfor future research.

MILD COGNITIVE IMPAIRMENT

Mild cognitive impairment refers to a tran-sitional state between the cognition of nor-mal aging and mild dementia.7 As shownin Figure 1, the criteria for definitive AD,

From the Department of Neurology, Mayo Clinic Rochester, Rochester, Minn(Dr Petersen); Department of Neurology, Baylor College of Medicine, Houston, Tex(Dr Doody); Department of Psychiatry, Technische University, Munich, Germany(Dr Kurz); Department of Psychiatry, Mt Sinai School of Medicine, New York, NY(Dr Mohs); Department of Neurology, Washington University School of Medicine,St Louis, Mo (Dr Morris); Department of Psychiatry, The Johns Hopkins UniversitySchool of Medicine, Baltimore, Md (Dr Rabins); Department of Nervous SystemPathologies, French National Institute of Medical Research, Montpellier, France(Dr Ritchie); Dementia Research Group, National Hospital for Neurology andNeurosurgery, Queen Square, London, England (Dr Rossor); Departmentof Neurosciences, University of California, San Diego (Dr Thal); and Divisionof Geriatrics, Karolinska Institute, Stockholm, Sweden (Dr Winblad).

SPECIAL ARTICLE

(REPRINTED) ARCH NEUROL / VOL 58, DEC 2001 WWW.ARCHNEUROL.COM1985

©2001 American Medical Association. All rights reserved.

while still controversial, have been reasonably well es-tablished on a neuropathological basis.8 Similarly, in re-cent years, the criteria for clinically probable AD havealso been well characterized,9 and the correlation be-tween clinical and pathological findings in suspected casesof AD has been good.10 However, as Figure 1 depicts, most,if not all, patients with AD experience a subtle cognitivedecline before reaching the clinical threshold for the di-agnosis of clinically probable AD. This transitional pe-riod has become known as MCI and is now the subjectof several multicenter treatment trials involving more than4000 subjects. The current clinical trials use an adapta-tion of the MCI criteria shown in the Table.

Because other presentations of MCI exist, the typejust described has been termed amnestic MCI because itsdefinition emphasizes memory loss. Most of these sub-jects will progress to AD at a rate of 10% to 15% per year,compared with healthy control subjects who convert ata rate of 1% to 2% per year.3,4 Data from the Mayo Alz-heimer’s Disease Research Center, Rochester, Minn, whichhas been observing a group of these subjects for morethan 10 years, have demonstrated a conversion to AD ofup to 80% during approximately 6 years (Figure 2).Other definitions of MCI and the imaging and neuro-pathological features of these subjects are discussed insubsequent sections.

RATING SCALES

There are several useful rating scales available for char-acterizing subjects along a continuum from normal ag-ing through various stages of dementia.11,12 Although thesescales are useful for describing subjects at various levelsof involvement, they do not necessarily coincide with the

clinically relevant conditions of normal aging, MCI, andmild AD. For example, the Clinical Dementia Rating(CDR) is a popular scale used to classify subjects alonga continuum from normal aging through AD.11 This scaledescribes a continuum from normal (CDR 0) throughquestionable dementia (CDR 0.5) to mild (CDR 1), mod-erate (CDR 2), and severe (CDR 3) dementia. Althoughsome investigators believe that CDR 0.5 is equivalent toMCI, others contend that CDR 0.5 actually describes abroader population that includes subjects with MCI andmild AD.13 Another instrument, the Global Deteriora-tion Scale (GDS) stages subjects from GDS 1 (normal)to GDS 2 (normal with subjective memory impair-ment), GDS 3 (mild dementia), and GDS 4 through 7(more severe stages of dementia).12 Within this ratingscale, subjects with MCI could correspond to a GDS ofeither 2 or 3. These potential relationships are depictedin Figure 3. The amnestic form of MCI is an identifi-able entity that does not necessarily map to a specific stageon these rating scales, thereby justifying a separate ter-minology.

NORMAL AGING

The differentiation of MCI from normality is an impor-tant area of study. Whereas clear diagnostic and re-search criteria for dementia are now available in inter-

Function

Mild Cognitive Impairment

Probable AD

Definite AD

Age

Figure 1. Theoretical progression of a person developing Alzheimer’sdisease (AD). The inflection point in the curve indicates the onset of mildcognitive impairment before the development of clinically probable AD.Reprinted with permission from W.B. Saunders.

100

60

MildCognitiveImpairment

80

40

20

0 1 2 43 5 6 7Duration, y

Stab

le, %

Figure 2. Survival curve of persons characterized as having a mild cognitiveimpairment for 6 years. Approximately 80% have converted to dementiaduring this time. Reprinted with permission from John Wiley & Sons, Inc.

Mild Cognitive ImpairmentCDR and GDS

Normal MCI

CDR

GDS

AD

0

1 2 3 4

0.5 1

Figure 3. Comparison of the clinical diagnoses of normal aging, mildcognitive impairment (MCI), and Alzheimer’s disease (AD), compared withthe approximate stages on the rating scales, Clinical Dementia Rating (CDR)and Global Deterioration Scale (GDS). Adapted from the American MedicalAssociation.13

Criteria for Amnestic Mild Cognitive Impairment

Memory complaint, preferably corroborated by an informantImpaired memory function for age and educationPreserved general cognitive functionIntact activities of daily livingNot demented



nationaldiseaseclassificationsystems, thenotionofnormalaging is less well understood. As disabilities associatedwith aging become treatable, they may be consideredpathologic conditions rather than inevitable features ofthe physiological aging process. Because of evolving ideasof what constitutes normal aging, cross-sectional com-parisons of older and younger populations are likely tobe misleading because of these age-cohort effects.

Within population studies, the concept of success-ful aging is sometimes proposed as an alternative to nor-mality. Successful aging refers to a state of health in whichthere are measurable positive features across a spec-trum of health measures. It extends beyond cognitive andfunctional definitions by considering the value of self-related psychological well-being. The Canadian Study ofHealth and Aging concluded that self-rated psychologi-cal well-being is a potential outcome measure for longi-tudinal investigation of individuals with either no cog-nitive impairment or minimal cognitive impairment.14

Previous attempts at characterizing cognitive changesintrinsic to normal aging have produced several terms,such as benign senescent forgetfulness, age-associatedmemory impairment, and age-associated cognitive de-cline.15-17 These terms are generally meant to reflect theextremes of normal aging rather than to describe a pre-cursor of pathologic aging. While some investigations ofthese concepts demonstrate dementia conversion ratesthat are not different from those of healthy subjects,18 oth-ers have indicated an increased conversion rate.19,20 Sub-jects with MCI have a condition that is different fromnormal aging, and longitudinal outcome results indicatethat they are likely to progress to AD at an acceleratedrate.3,21

EPIDEMIOLOGY

There has been little work on the incidence or preva-lence of MCI as defined herein. This is partially becauseof the recent characterization of this condition. Popula-tion-based studies, especially those that ascertain casesbased on psychometric test results rather than symp-toms, may identify a population of individuals with mildcognitive dysfunction that differs from the amnestic MCIdescribed clinically. Prevalence rates for age-associatedmemory impairment in these studies range from 17% to34%,16,22-25 while age-associated cognitive decline has beenestimated as having a prevalence of 26%.17 Other inves-tigators have indicated that the prevalence rate for con-ditions such as age-associated memory impairment canrange up to 85%, depending on the specific memory testsused to fulfill the criteria.25 In the Canadian Study ofHealth and Aging, the classification of cognitive impair-ment with no dementia had a prevalence rate of 17%; thisis likely a more inclusive term than MCI alone.14

An important factor in considering the frequency ofMCI concerns the source of subjects for a given study. Itis likely that a small number of subjects with MCI willpresent to memory disorders or dementia clinics. Mostindividuals who present to a dementia clinic probably al-ready have dementia. However, if one surveys a commu-nity population, it is probable that more cases of MCI willbe found.

HETEROGENEITY

All individuals who present clinically with mild cogni-tive symptoms may not share the same fate ultimately.Some may go on to develop AD, while others may progressto another dementia. It is possible that some of the sub-jects will never progress to any significant extent. Thisbroad group of individuals with mild cognitive com-plaints could be considered as having MCI. Recogniz-ing that there are multiple sources of heterogeneity insuch a classification, it is desirable to further specify cri-teria for subsets of MCI.

Thus far, we have focused herein on the most com-mon subset of subjects with MCI, those with amnestic MCI.These are individuals who present with a subjective memorycomplaint, preferably corroborated by an informant, andhave an objective memory impairment compared with ageand education norms, but who are performing reasonablywell on indexes of general cognitive function and have gen-erally preserved activities of daily living. As discussed ear-lier, these subjects do not meet standard criteria for AD.While subjects with amnestic MCI usually progress to ADat a high rate,3,4,6 not all will progress (Figure 2). On theother hand, some subjects who present with amnestic MCImay have other pathologic processes involving the medialtemporal lobes, eg, hippocampal sclerosis.26 It is as yet un-certain whether clinical variations of AD or disorders withconcomitant AD, such as dementia with Lewy bodies, canpresent as amnestic MCI. There is evidence that subjectswith dementia associated with Lewy bodies have relativepreservation of the hippocampi, making this amnestic MCIpresentation less likely for subjects diagnosed as having de-mentia with Lewy bodies.27

It theoretically should be possible to select a differ-ent subset of subjects with MCI by altering the core defi-nition and criteria. Figure 4 depicts other hypotheticalpresentations of MCI in addition to amnestic MCI. Al-though none of these other prodromal conditions have beenvalidated, MCI could be defined as mild impairment in mul-tiple cognitive domains, without requiring memory defi-cit.28 These subjects may have multiple areas of cognitiveimpairment that fall outside of predicted norms, but noneare sufficiently severe to constitute dementia. Subjects withMCI defined in this fashion may also progress to AD, butthey could also progress to other disorders.

Finally, MCI could conceivably present as impair-ment in a single cognitive domain other than memory. Forexample, a pronounced language disturbance mightprogress to primary progressive aphasia, or an alteration

Mild CognitiveImpairment

(Single Nonmemory Domain)

Frontotemporal DementiaLewy Body DementiaVascular DementiaPrimary Progressive AphasiaParkinson's DiseaseAlzheimer's Disease

Mild CognitiveImpairment

(Multiple Domains Slightly Impaired)

Alzheimer's DiseaseVascular Dementia? Normal Aging

Mild CognitiveImpairment(Amnestic)

Alzheimer's Disease

Figure 4. Heterogeneity of the term mild cognitive impairment.



in attentional abilities or comportment and a dysexecu-tive syndrome might progress to frontotemporal demen-tia. Although no such MCI cases have been reported, itis reasonable to posit a prodromal period for frontotem-poral dementia, analogous to amnestic MCI as a precur-sor of AD.

It has become apparent that heterogeneity of MCIis derived from differences in causes, clinical symp-toms, and research methods. For clarity, it is recom-mended that the term MCI be qualified with an appro-priate modifier, such as amnestic MCI, to inform thereader as to the specific criteria being used to character-ize the condition and its most likely outcome.

NEUROPSYCHOLOGICAL ASSESSMENT

Screening instruments are routinely used to quantify thedegree of cognitive impairment in patients with dementiaand are likely to be particularly helpful early in a dement-ing illness, when functional and behavioral disturbancesare absent.29,30 The value of neuropsychological measuresin helping to identify early cases of dementia has been docu-mented by cross-sectional and longitudinal studies. Cross-sectional comparisons of patients clinically diagnosed ashaving mild AD compared with older healthy persons haveshown that patients with mild AD consistently performworse than their comparison subjects.31-33 Patients with earlyAD are impaired on tests of memory, including tests of de-layed recall, and often on tests of new learning.34,35 Defi-ciencies in language, executive function, and attention arealso present in many patients with mild AD, and deficitsin more than one cognitive domain are a better indicatorof mild AD than memory impairment alone.36 These cross-sectional data indicate that sensitive neuropsychologicalmeasures can help identify cases of AD at early stages.

Longitudinal studies5,36-38 of older healthy personsat high risk for developing dementia because of ad-vanced age have shown that persons who later developAD perform more poorly on cognitive tests at baselinecompared with those who remain free of dementia. Theselongitudinal data indicate that neuropsychological testsmay help to identify persons likely to convert to MCI orAD before they meet conventional diagnostic criteria.

Alzheimer’s disease and MCI cannot be diagnosedby neuropsychological tests alone, and clinical judg-ment is always required. First, performance on neuro-psychological tests is affected by many factors, includ-ing education, age, cultural background, and illnessesother than AD.39 Second, neuropsychological measurescannot fully distinguish among different types of de-mentia, because there is substantial overlap in neuropsy-chological profiles.40,41 The usefulness of any battery foridentifying cases of MCI will depend on its composition,size, and supporting data. A brief battery, including mea-sures of new learning, delayed recall, attention, and ex-ecutive function, could provide valuable information forscreening and diagnosis if interpreted properly.

NEUROIMAGING

Neuroimaging potentially is a powerful tool for the dif-ferential diagnosis of cognitive impairment and for moni-

toring change. Cross-sectional and longitudinal stud-ies42-44 have used structural (computed tomography andmagnetic resonance imaging) and functional (single-photon emission computed tomography, positron emis-sion tomography, and magnetic resonance spectros-copy) modalities in the evaluation of MCI. There isagreement on the occurrence of hippocampal atrophy inamnestic MCI compared with cognitively intact con-trols, and on the finding that hippocampal atrophy canpredict the rate of conversion from MCI to AD.43 Hip-pocampal atrophy seen in MCI can be correlated withautopsy evidence of atrophy and neuronal loss.45

In addition to hippocampal atrophy, changes in theentorhinal cortex may enhance the sensitivity of an earlydiagnosis of MCI or AD. However, a recent study46 com-paring entorhinal measures with hippocampal volumesindicates that the entorhinal measurement may not al-ways be superior. Neuroimaging measures of change aremore informative than a measurement made at one pointin time. Rates of hippocampal atrophy for individuals whoconvert from control subjects to MCI or from MCI to ADare comparable, suggesting that the MCI group may betransitional to early AD.47

Functional neuroimaging may also aid in the de-tection of subjects with MCI.48,49 In group studies usingpositron emission tomography, early metabolic deficitsin the temporoparietal region have been demonstratedin those at risk for familial AD48 based on family historyand those with an apolipoprotein E4 genotype.50 Simi-larly, patients with a genetic risk for AD also show meta-bolic reduction in the posterior cingulate cortex.48 Mag-netic resonance spectroscopy may eventually providefurther evidence of functional changes in MCI, althoughinvestigations have been limited thus far.51

In summary, neuroimaging studies support the viewthat MCI, especially amnestic MCI, shares features withAD, such as hippocampal atrophy, so that the presenceor development of atrophy will therefore predict con-version to clinical AD. Hippocampal atrophy, however,may not be specific, and careful selection and follow-upof subjects are important. Future directions are likely toconcentrate on measuring rates of change on neuroim-aging studies and on finding the shortest intervals nec-essary to demonstrate significant changes.

NEUROPATHOLOGY

Little is known about the neuropathology of MCI and re-lated disorders, because there have been few longitudi-nal studies that include neuropathological confirma-tion. Subjects with mild impairment are unlikely to dieof MCI, so neuropathological information is obtained onlywhen someone who happens to have the clinical desig-nation of MCI dies of other causes. In a neuropathologi-cal series52 of 16 patients diagnosed clinically as havingMCI or AD and a CDR of 0.5, all demonstrated neuro-fibrillary tangles in the hippocampus and entorhinal cor-tex, while 7 also had many senile plaques in the neocor-tex. Other indexes of neuronal degeneration, such aschanges in synaptic markers and neuronal loss, were seenin the CA1 region of the hippocampus as assessed by un-biased stereology.



In a clinical pathological series53 of 62 patients (39without dementia, 15 with a CDR of 0.5, and 8 with es-tablished AD), all of those with CDR 0.5 ratings had theneuropathological features of AD. These investigators pos-tulated that senile plaques in the neocortex may repre-sent the initial event in AD and may escalate age-relatedformation of neurofibrillary tangles in the medial tem-poral lobe beyond the usual level. Neocortical senileplaques and entorhinal tangles were also seen in somesubjects with no detectable cognitive decline, suggest-ing that there may be a neuropathological picture of ADthat precedes an MCI state.

In another report54 of 10 autopsy cases with clinicaldiagnoses of amnestic MCI, 6 met neuropathological cri-teria for AD, while an additional 2 had argyrophilic graindisease. Two cases had neurofibrillary tangle–only dis-ease. Data from the Religious Orders Study have sup-ported clinical observations that the causes of MCI maybe heterogeneous, because they have shown that non-ADdisorders that involve the medial temporal lobe may ac-count for the memory impairment.28 When these rarenon-AD disorders are excluded, most pathologic condi-tions of MCI are likely to be early-stage AD.

In summary, most patients with amnestic MCI dem-onstrate neurofibrillary pathologic conditions in the me-dial temporal lobes. It is possible that these neurofibrillarychanges contribute to the memory impairment. Some evi-dence indicates that patients with CDR 0 ratings (nomemory impairment) also have neurofibrillary tangles inthe medial temporal lobe; hence, this finding does not al-ways correlate with a memory impairment. Most patientswith MCI appear to have the neuropathological changesof AD, and a few clinically similar subjects have non-ADpathologic conditions.

BIOMARKERS

Because MCI may be clinically and pathologically het-erogeneous, biomarkers may be particularly useful foridentifying subtypes. However, there are no definitive dataon the usefulness of biomarkers in classifying MCI.

Biomarkers in the cerebrospinal fluid (CSF) havebeen used as indicators of the degenerative process of AD.Several laboratories have shown that CSF tau concen-trations are significantly elevated in the CSF and thatA�-42 levels are significantly decreased in patients withAD, compared with age-matched cognitively healthy con-trol subjects.55 Tau levels are also significantly in-creased in mildly demented subjects with AD (Mini-Mental State Examination score of 25), yielding sensitivityand specificity values approaching 0.90. More recently,elevated CSF tau has been confirmed in autopsy-provenAD cases.56

Longitudinal studies57 have demonstrated that nearlyall subjects with MCI who convert to AD have high CSFtau values, whereas in nonprogressive MCI, tau levels re-main low. Furthermore, CSF tau levels do not seem toincrease during the course of AD.57 This suggests that themeasurement of CSF tau might be used effectively foridentifying incipient AD among patients diagnosed clini-cally as having MCI.58 In summary, biomarkers could beuseful in a multistep approach to the diagnosis of MCI

or dementia, and disease-specific biomarkers could beuseful in treatment intervention follow-up.

PROGRESSION

Several longitudinal studies3,4,6 of aging and dementiaprovide information regarding outcomes of patients withMCI. In addition, some studies36,37 of aging shed light oncognitive progression in individuals who are functioningnormally at one point in time, but who subsequently de-velop MCI or AD. Both types of studies provide informa-tion regarding the ultimate usefulness of the designationof MCI.

Using the criteria outlined in the table for amnesticMCI, investigators at the Mayo Alzheimer’s Disease Re-search Center longitudinally observed 155 individualswith a clinical diagnosis of MCI.3,21 These individualsevolved to dementia at a rate of approximately 12% peryear. Variables that tended to predict a more rapid de-cline included apolipoprotein E4 carrier status, poor per-formance on a cued recall test, and atrophic hippo-campi on MRI. As indicated in Figure 2, up to 80% ofthis group had evolved to dementia in approximately 6years.

In Toronto, Ontario, investigators observed a groupof subjects with memory complaints who had been re-ferred to primary care practitioners.4 These subjects werefollowed up longitudinally for 2 years, and approxi-mately 20% progressed to dementia. These investiga-tors believed that performance on a memory test and pos-sibly apolipoprotein E4 status were predictors of decline.

Investigators from the University of Washington Alz-heimer’s Disease Patient Registry, Seattle, also followedup a group of memory-impaired individuals who did notmeet criteria for dementia.6 These subjects were ob-served for up to 48 months, and approximately 50% ofthe group progressed to dementia during that time. Theseinvestigators did not find any specific psychometric teststhat predicted who was more likely to become im-paired.

In various studies3,4,6 using different MCI criteria,subjects with memory problems that fall short of demen-tia progress to dementia or AD at a rate of between 10%to 15% per year, making this condition an identifiablerisk factor for the development of AD. These studies, takentogether, indicate that different criteria in multiple stud-ies may be identifying the common phenomenology ofamnestic MCI.

TREATMENT

There is no evidence that MCI, once diagnosed, can besuccessfully treated. Inclusion criteria for MCI trials willlikely continue to focus on subjects with amnestic MCIwho do not meet diagnostic criteria for AD. The end pointsfor MCI trials will depend on the presumed benefits ofthe intervention. Some interventions may result in im-provement of symptoms. End points to demonstrate thistype of benefit will likely be limited to sensitive tests ofdelayed memory (verbal and visual) and clinicians’ globalratings. In other cases, slowing the rate of progressionmay be the goal, so delaying a clinical milestone (eg, de-



velopment of a second domain of difficulty) or conver-sion to clinically diagnosable AD will be the appropriateend points. Several clinical trials have already been ini-tiated to test the use of cholinesterase inhibitors, anti-oxidants, and anti-inflammatory drugs in patients withMCI. Most of these trials are based on the recently ini-tiated trial being conducted by the Alzheimer’s DiseaseCooperative Study.59

Safety and tolerability will be required for drugs usedin MCI trials, because many subjects will be treated atthe early stage of the disease when symptoms are mild.Determination of optimal drug dosage should be workedout in individuals with AD, when possible, because dos-age-finding studies expose individuals to a greater riskof adverse effects, and the risk-benefit ratio may be dif-ferent for subjects with MCI vs AD.

Many classes of agents can be considered for use insubjects with amnestic MCI. Cholinesterase inhibitors maybe considered for use for their symptomatic effects andcircumstantial laboratory and clinical evidence that theymay have a biologically modifying effect on the process-ing of amyloid or other mechanisms in the progressionof the disease.60 Hormones, particularly estrogens, havebeen epidemiologically demonstrated to be associated withdelayed onset of the appearance of AD.61,62

Histological and epidemiological studies63,64 sug-gest the presence of inflammation early in AD. Clinicaltrials of the effect of cyclooxygenase-2 (COX-2) inhibi-tors in patients with AD and MCI are under way. Anti-oxidants such as vitamin E have been demonstrated todelay the time to clinical end points in patients with mod-erate AD,65 and vitamin E is being studied in patients withMCI.59 Interruption or modification of protein metabo-lism is believed by many AD researchers to represent thebest approach to disease-modifying therapy.66 Agents thatinterfere with A� production and aggregation or that re-duce A� burden may confer protection against A�-mediated neurodegeneration in AD.66,67 Another molecu-lar approach focuses on the development of agents thatinterfere with tau phosphorylation or aggregation.68 Fi-nally, Gingko biloba may have antioxidant properties69

and will undergo large-scale testing in primary preven-tion trials. Treatments should not be prescribed basedon current speculations, but must await confirmation fromwell-designed clinical trials.

ETHICAL ISSUES

The concept of MCI raises several ethical issues. The pri-mary ethical questions concern the strength of the evi-dence that MCI is a valid pathologic condition, andwhether the benefits of establishing this new category out-weigh the risks to patients. Making an accurate diagno-sis is a key element in the practice of medicine and hasseveral benefits. It helps clinicians identify specific con-ditions and aids in the choice of appropriate treatment.In addition, the establishment of a diagnosis enhancescommunication among clinicians, helps provide infor-mation to patients, identifies a need for understandingand support, and stimulates research. On the other hand,labeling patients with a disease has psychological and psy-chosocial consequences. Therefore, investigators and cli-

nicians who are proposing MCI as a new syndrome mustdevelop the data needed to convince skeptics that it is avalid entity.70

Patients diagnosed as having amnestic MCI oftencomplain about memory loss. This suggests that indi-viduals with MCI have insight and should be told aboutthis diagnostic label. On the other hand, some individu-als with MCI, especially if ascertained by population-based testing, may be unaware of their memory impair-ment or may believe that their symptoms are related tonormal aging or lifelong problems. Patients who pre-sent clinically should probably be managed differentlyfrom patients discovered by applying research criteria tolarge populations.

Because amnestic MCI progresses to AD at a high rate,it is associated with significant morbidity, potential eco-nomic loss to individuals and society, and frustration anddistress in caregivers. Mild cognitive impairment itself istherefore an appropriate target for treatments to improvesymptoms and to reverse or prevent the condition.

RECENT LITERATURE

Since the Current Concepts in Mild Cognitive Impair-ment conference, several relevant studies have appearedin the literature. These will be mentioned without an evalu-ative comment because they were not reviewed by theconference attendees.

Investigators at Massachusetts General Hospital, Bos-ton, followed up a group of subjects with a diagnosis of ques-tionable dementia (CDR 0.5) for 3 years and found thatthis group progressed to AD at a rate of 6% per year.71 Thisstudy documented the importance of the clinical inter-view in the characterization of these subjects. An edito-rial comment13 on this article called attention to the dis-tinction between a clinical diagnosis of MCI and the useof rating scales. Another study72 on these subjects dem-onstrated the usefulness of magnetic resonance imaging–based volumetric measurements of the entorhinal cortexin predicting progression in individuals with question-able dementia. A study73 from France applied a modifiedversion of MCI criteria to a population of subjects beingobserved longitudinally and found that certain neuropsy-chological criteria for MCI, when retrospectively ap-plied, were unreliable in characterizing the progression ofsubjects over time. A Washington University research groupperformed an autopsy-based study of their subjects witha CDR 0.5 diagnosis and found that all had a dementingillness; 88% had neuropathological features of AD.74 Astudy75 from Sweden demonstrated the reliability of im-paired glucose metabolism and a cognitive measure of vi-suospatial performance in predicting progression from MCIto AD. Finally, a recent evidence-based medicine prac-tice review76 of early detection and MCI by the AmericanAcademy of Neurology recommended that subjects withMCI be identified and followed up because of their in-creased risk for developing AD.

SUMMARY

Mild cognitive impairment is becoming an increasinglyrecognized clinical entity. Most clinicians are aware of



patients who appear to have a mild deficit in memory orsome other aspect of cognition but who are not de-mented. There are important differences between clini-cal practice settings and research settings. For example,researchers use comprehensive neuropsychological test-ing to document that memory loss is clinically signifi-cant and to make certain there are no other domains ofsignificant cognitive impairment that might mimic cri-teria for mild AD. Clinicians should be wary about des-ignating a patient as having MCI in the absence of spe-cific criteria. Although therapies are being developed thatmight prevent progression of amnestic MCI to AD, thereis no evidence to support prescribing therapeutic agentsto patients with MCI at this time. Our conference dis-cussion highlighted the current thinking on MCI from arecent international conference. Mild cognitive impair-ment deserves recognition and further study because, aspreventive treatments for AD become available, it will be-come incumbent on clinicians to identify persons at riskfor AD and those with the earliest signs of clinical im-pairment. It may be that at least some of what is now des-ignated as MCI will be recognized in the future as theearliest symptomatic stage of AD. Such an evolution willbe stimulated by evidence that pharmacological therapyis beneficial at this stage and by the clinician’s ability torecognize earlier forms of AD.

Accepted for publication July 30, 2001.This study was supported by an educational grant

from Pfizer, Inc, New York, NY, and Eisai Inc, Teaneck, NJ.Drs Petersen and Doody contributed equally to the

preparation of the manuscript.Participants in the Current Concepts in Mild Cogni-

tive Impairment conference include: Ronald C. Petersen, PhD,MD (cochair), and Clifford Jack, Jr, MD, Mayo Clinic Roch-ester, Rochester, Minn; Rachelle Doody, MD, PhD (co-chair), Baylor College of Medicine, Houston, Tex; Hi-royuki Arai, MD, Tohoku University School of Medicine,Sendai, Japan; David Bennett, MD, Rush Alzheimer’s Dis-ease Center, Chicago, Ill; Christopher Clark, MD, and Ja-son Karlawish, MD, University of Pennsylvania, Philadel-phia; Jeffrey L. Cummings, MD, University of California,Los Angeles, School of Medicine; Steven T. DeKosky, MD,University of Pittsburgh Medical Center, Pittsburgh, Pa;Mony de Leon, MD, and Steven Ferris, PhD, New York Uni-versity Medical Center, New York, NY; Davangere P.Devanand, MD, and Mary Sano, PhD, Columbia Univer-sity, New York; Dennis W. Dickson, MD, Mayo Clinic Jack-sonville, Jacksonville, Fla; Howard Feldman, MD, Univer-sity of British Columbia, Vancouver; Giovanni B. Frisoni,MD, IRCCS San Giovanni di Dio–FBF, Brescia, Italy; Nor-man L. Foster, MD, University of Michigan, Ann Arbor;Douglas Galasko, MD, and Leon Thal, MD, Veterans Af-fairs Medical Center, University of California, San Diego;Michael Grundman, MD, University of California, San Diego;William Jagust, MD, University of California, Davis; Jef-frey Kaye, MD, Oregon Health Sciences University, Port-land; David Knopman, MD, University of Minnesota, Min-neapolis (now at Mayo Clinic Rochester); Amos D. Korczyn,MD, Tel Aviv University Medical School, Tel Aviv, Israel;Alexander Kurz, MD, Technische University, Munich, Ger-many; Paul Leber, MD, Neuro-Pharm Group, LLC, Po-

tomac, Md; Richard Mohs, PhD, Mount Sinai School of Medi-cine, New York; John C. Morris, MD, Joseph L. Price, PhD,and Martha Storandt, PhD, Washington University, St Louis,Mo; Peter V. Rabins, MD, and Juan C. Troncoso, MD, TheJohns Hopkins University School of Medicine, Baltimore, Md;Karen Ritchie, MPsych, PhD, French National Institute ofMedical Research, Montpellier, France; Kenneth Rock-wood, MD, Dalhousie University, Halifax, Nova Scotia; Mar-tin Rossor, MA, MD, FRCP, National Hospital for Neurol-ogy and Neurosurgery, London, England; Frederick A.Schmitt, PhD, University of Kentucky, Lexington; JacquesTouchon, MD, Hopital Gui de Chauliac, Montpellier, France;Peter Whitehouse, MD, PhD, Case Western Reserve Uni-versity, Cleveland, Ohio; and Bengt Winblad, MD, PhD, Ka-rolinska Institute, Stockholm, Sweden.

Corresponding author and reprints: Ronald C. Petersen,PhD, MD, Department of Neurology, Mayo Clinic Roches-ter, 200 First St SW, Rochester, MN 55905 (e-mail:[email protected]).

REFERENCES

1. Schenk D, Barbour R, Dunn W, et al. Immunization with amyloid-� attenuatesAlzheimer-disease–like pathology in the PDAPP mouse. Nature. 1999;400:173-177.

2. Vassar R, Bennett BD, Babu-Khan S, et al. �-Secretase cleavage of Alzheimer’samyloid precursor protein by the transmembrane aspartic protease BACE. Sci-ence. 1999;286:735-741.

3. Petersen RC, Smith GE, Waring SC, Ivnik RJ, Tangalos EG, Kokmen E. Mild cog-nitive impairment: clinical characterization and outcome. Arch Neurol. 1999;56:303-308.

4. Tierney MC, Szalai JP, Snow WG, et al. Prediction of probable Alzheimer’s dis-ease in memory-impaired patients: a prospective longitudinal study. Neurology.1996;46:661-665.

5. Rubin DH, Storandt M, Miller JP, et al. A prospective study of cognitive functionand onset of dementia in cognitively healthy elders. Arch Neurol. 1998;55:395-401.

6. Bowen J, Teri L, Kukull W, McCormick W, McCurry SM, Larson EB. Progressionto dementia in patients with isolated memory loss. Lancet. 1997;349:763-765.

7. American Psychiatric Association. Diagnostic and Statistical Manual of MentalDisorders, Fourth Edition. Washington, DC: American Psychiatric Association;1994.

8. The National Institute on Aging, and Reagan Institute Working Group on Diag-nostic Criteria for the Neuropathological Assessment of Alzheimer’s Disease. Con-sensus recommendations for the postmortem diagnosis of Alzheimer’s dis-ease. Neurobiol Aging. 1997;18(4 suppl):S1-S2.

9. McKhann G, Drachman D, Folstein M, Katzman R, Price D, Stadlan EM, for theNational Institute of Neurological and Communicative Disorders and Stroke–Alzheimer’s Disease and Related Disorders Association Work Group. Clinical di-agnosis of Alzheimer’s disease: report of the NINCDS-ADRDA Work Group un-der the auspices of Department of Health and Human Services Task Force onAlzheimer’s Disease. Neurology. 1984;34:939-944.

10. Galasko D, Hansen LA, Katzman R, et al. Clinical-neuropathological correlationsin Alzheimer’s disease and related dementias. Arch Neurol. 1994;51:888-895.

11. Morris JC. The Clinical Dementia Rating (CDR): current version and scoring rules.Neurology. 1993;43:2412-2414.

12. Reisberg B, Ferris SH, de Leon MJ, Crook T. The Global Deterioration Scale forassessment of primary degenerative dementia. Am J Psychiatr. 1982;139:1136-1139.

13. Petersen RC. Mild cognitive impairment or questionable dementia [editorial]?Arch Neurol. 2000;57:643-644.

14. Graham JE, Rockwood K, Beattie BL, et al. Prevalence and severity of cognitiveimpairment with and without dementia in an elderly population. Lancet. 1997;349:1793-1796.

15. Kral VA. Senescent forgetfulness: benign and malignant. CMAJ. 1962;86:257-260.

16. Crook T, Bartus RT, Ferris SH, Whitehouse P, Cohen GD, Gershon S, for theNational Institute of Mental Health Work Group. Age-associated memoryimpairment: proposed diagnostic criteria and measures of clinical change.Dev Neuropsychol. 1986;2:261-276.

17. Levy R. Aging-associated cognitive decline. Int Psychogeriatr. 1994;6:63-68.18. Hanninen T, Hallikainen M, Koivisto K, et al. A follow-up study of age-associated

memory impairment: neuropsychological predictors of dementia. J Am GeriatrSoc. 1995;43:1007-1015.

19. Richards M, Touchon J, Ledesert B, Richie K. Cognitive decline in aging: are AAMIand AACD distinct entities? Int J Geriatr Psychiatr. 1999;14:534-540.



20. Lane F, Snowdon J. A longitudinal survey of suburban elderly. In: Lovibond P,Wilson P, eds. Clinical and Abnormal Psychology. Amsterdam, the Netherlands:Elsevier; 1989:365-376.

21. Petersen RC, Smith GE, Ivnik RJ, et al. Apolipoprotein E status as a predictor ofthe development of Alzheimer’s disease in memory-impaired individuals. JAMA.1995;273:1274-1278.

22. Barker A, Jones R, Jennison C. A prevalence study of age-associated memoryimpairment. Br J Psychiatr. 1995;167:642-648.

23. Coria F, Gomez de Caso JA, Minguez F, Rodriquez-Artalejo F, Claveria LE. Preva-lence of age-associated memory impairment and dementia in a rural community.J Neurol Neurosurg Psychiatr. 1993;56:973-976.

24. Larrabee GJ, Crook TH. Estimated prevalence of age-associated memoryimpairment derived from standardized tests of memory function. Int Psycho-geriatr. 1994;6:95-104.

25. Smith GE, Ivnik RJ, Petersen RC, Malec JF, Kokmen E, Tangalos E. Age-associated memory impairment diagnoses: problems of reliability and concernsfor terminology. Psychol Aging. 1991;6:551-558.

26. Dickson DW, Davies P, Bevona C, et al. Hippocampal sclerosis: a common patho-logical feature of dementia in very old (�80 years of age) humans. Acta Neuro-pathol (Berl). 1994;88:212-221.

27. Lippa CF, Johnson R, Smith TW. The medial temporal lobe in dementia with Lewybodies: a comparative study with Alzheimer’s disease. Ann Neurol. 1998;43:102-106.

28. Mufson EJ, Chen EY, Cochran EJ, Beckett LA, Bennett DA, Kordower JH. Ento-rhinal cortex �-amyloid load in individuals with mild cognitive impairment. ExpNeurol. 1999;158:469-490.

29. Folstein MF, Folstein SE, McHugh PR. “Mini-Mental State”: a practical methodfor grading the cognitive state of patients for the clinician. J Psychiatr Res. 1975;12:189-198.

30. Rosen WG, Mohs RC, Davis KL. A new rating scale for Alzheimer’s disease. AmJ Psychiatr. 1984;141:1356-1364.

31. Knopman DS, Ryberg S. A verbal memory test with high predictive accuracy fordementia of the Alzheimer type. Arch Neurol. 1989;46:141-145.

32. Storandt M, Hill RD. Very mild senile dementia of the Alzheimer type, II: psy-chometric test performance. Arch Neurol. 1989;46:383-386.

33. Welsh KA, Butters N, Hughes JP, Mohs RC, Heyman A. Detection and staging ofdementia in Alzheimer’s disease: use of the neuropsychological measures de-veloped for the Consortium to Establish a Registry for Alzheimer’s Disease. ArchNeurol. 1992;49:448-452.

34. Petersen RC, Smith GE, Ivnik RJ, Kokmen E, Tangalos EG. Memory function invery early Alzheimer’s disease. Neurology. 1994;44:867-872.

35. Welsh K, Butters N, Hughes J, Mohs R, Heyman A. Detection of abnormal memorydecline in mild cases of Alzheimer’s disease using CERAD neuropsychologicalmeasures. Arch Neurol. 1991;48:278-281.

36. Masur DM, Sliwinski M, Lipton RB, Blau AD, Crystal HA. Neuropsychological pre-diction of dementia and the absence of dementia in healthy elderly persons. Neu-rology. 1994;44:1427-1432.

37. Dartigues JF, Commenges D, Letenneur L, et al. Cognitive predictors of demen-tia in elderly community residents. Neuroepidemiology. 1997;16:29-39.

38. Howieson DB, Dame A, Camicioli R, Sexton G, Payami H, Kaye JA. Cognitive mark-ers preceding Alzheimer’s dementia in the healthy oldest old. J Am Geriatr Soc.1997;45:584-589.

39. Inouye SK, Albert MS, Mohs RC, Sun K, Berkman LF. Cognitive performance ina high-functioning community-dwelling elderly population. J Gerontol A Biol SciMed Sci. 1993;48:M146-M151.

40. Heyman A, Fillenbaum GG, Welsh-Bohmer KA, et al, for the Consortium toEstablish a Registry for Alzheimer’s Disease. Cerebral infarcts in patients withautopsy-proven Alzheimer’s disease. Neurology. 1998;51(pt 18):159-162.

41. Ferman TJ, Boeve BF, Smith GE, et al. REM sleep behavior disorder and demen-tia: cognitive differences when compared with AD. Neurology. 1999;52:951-957.

42. Jack CR Jr, Petersen RC, Xu YC, et al. Medial temporal atrophy on MRI in normalaging and very mild Alzheimer’s disease. Neurology. 1997;49:786-794.

43. Jack CR Jr, Petersen RC, Xu YC, et al. Prediction of AD with MRI-based hippo-campal volume in mild cognitive impairment. Neurology. 1999;52:1397-1403.

44. Fox NC, Warrington EK, Seiffer AL, Agnew SK, Rossor MN. Presymptomatic cog-nitive deficits in individuals at risk of familial Alzheimer’s disease: a longitudinalprospective study. Brain. 1998;121:1631-1639.

45. Bobinski M, de Leon MJ, Tarnawski M, et al. Neuronal and volume loss in CA1of the hippocampal formation uniquely predicts duration and severity of Alzhei-mer disease. Brain Res. 1998;805:267-269.

46. Xu YC, Jack CR Jr, O’Brien PC, et al. Usefulness of MRI measures of entorhinalcortex versus hippocampus in AD. Neurology. 2000;54:1760-1767.

47. Jack CR Jr, Petersen RC, Xu Y, et al. Rates of hippocampal atrophy correlatewith change in clinical status in aging and AD. Neurology. 2000;55:484-489.

48. Kennedy AM, Frackowiak RS, Newman SK, et al. Deficits in cerebral glucose me-

tabolism demonstrated by positron emission tomography in individuals at riskof familial Alzheimer’s disease. Neurosci Lett. 1995;186:17-20.

49. Berent S, Giordani B, Foster N, et al. Neuropsychological function and cerebralglucose utilization in isolated memory impairment and Alzheimer’s disease.J Psychiatr Res. 1999;33:7-16.

50. Reiman EM, Caselli RJ, Lang S, et al. Preclinical evidence of Alzheimer’s diseasein persons homozygous for the E4 allele for apolipoprotein E. N Engl J Med. 1996;334:752-758.

51. Kantarci K, Jack CR Jr, Xu YC, et al. Regional metabolic patterns in mild cogni-tive impairment and Alzheimer’s disease: a 1H MRS study. Neurology. 2000;55:210-217.

52. Troncosco JC, Martin LJ, Dal Forno G, Kawas CH. Neuropathology in controlsand demented subjects from the Baltimore Longitudinal Study of Aging. Neuro-biol Aging. 1996;17:365-371.

53. Price JL, Morris JC. Tangles and plaques in nondemented aging and “preclini-cal” Alzheimer’s disease. Ann Neurol. 1999;45:358-368.

54. Petersen RC, Parisi JE, Dickson DW, et al. Neuropathologic substrate of mildcognitive impairment [abstract]. Neurobiol Aging. 2000;21(suppl 1S):S198.

55. Galasko D, Clark C, Chang L, et al. Assessment CSF levels of tau protein in mildlydemented patients with Alzheimer’s disease. Neurology. 1997;48:632-635.

56. Andreasen N, Vanmechelen E, Van de Voorde A, et al. Cerebrospinal fluid tauprotein as a biochemical marker for Alzheimer’s disease: a community based fol-low up study. J Neurol Neurosurg Psychiatr. 1998;64:298-305.

57. Sunderland T, Wolozin B, Galasko D, et al. Longitudinal stability of CSF tau lev-els in Alzheimer patients. Biol Psychiatry. 1999;46:750-755.

58. Galasko D. Cerebrospinal fluid opens a window on Alzheimer’s disease. Arch Neu-rol. 1999;56:655-656.

59. Grundman M. Vitamin E in Alzheimer disease: the basis for additional clinicaltrials. Am J Clin Nutr. 2000;71:630S-636S.

60. Lahiri DK, Farlow MR, Sambamurti K. The secretion of amyloid �-peptides isinhibited in the tacrine-treated human neuroblastoma cells. Brain Res Mol BrainRes. 1998;62:131-140.

61. Yaffe K, Sawaya G, Liebergurg I, Grady D. Estrogen therapy in postmenopausalwomen: effects on cognitive function and dementia. JAMA. 1998;279:688-695.

62. Waring SC, Rocca WA, Petersen RC, O’Brien PC, Tangalos EG, Kokmen E. Post-menopausal estrogen replacement therapy and risk of AD: a population-basedstudy. Neurology. 1999;52:965-970.

63. McGeer PL, Schulzer M, McGeer EG. Arthritis and anti-inflammatory agents aspossible protective factors for Alzheimer’s disease: a review of 17 epidemio-logic studies. Neurology. 1996;47:425-432.

64. Pasinetti GM. Cyclooxygenase and inflammation in Alzheimer’s disease: experi-mental approaches and clinical interventions. J Neurosci Res. 1998;54:1-6.

65. Sano M, Ernesto C, Thomas RG, et al, for The Alzheimer’s Disease CooperativeStudy. A controlled trial of selegiline, alpha-tocopherol, or both as treatment forAlzheimer’s disease. N Engl J Med. 1997;336:1216-1222.

66. Wolfe MS. Secretase targets for Alzheimer’s disease: identification and thera-peutic potential. J Med Chem. 2001;44:2039-2060.

67. Wolfe MS, Xia W, Ostaszewski BL, Diehl TS, Kimberly WT, Selkoe DJ. Two trans-membrane aspartates in presenilin-1 required for presenilin endoproteolysis and�-secretase activity. Nature. 1999;398:513-517.

68. Schneider A, Biernat J, von Bergen M, Mandelkow E, Mandelkow EM. Phos-phorylation that detaches tau protein from microtubules (Ser262, Ser214) alsoprotects it against aggregation into Alzheimer paired helical filaments. Biochem-istry. 1999;38:3549-3558.

69. Le Bars PL, Katz MM, Berman N, Itil TM, Freedman AM, Schatzberg AF, for the NorthAmerican EGb Study Group. A placebo-controlled, double-blind, randomized trialof an extract of Ginkgo biloba for dementia. JAMA. 1997;278:1327-1332.

70. Christensen H, Henderson AS, Jorm AF, Mackinnon AJ, Scott R, Korten AE. ICD-10mild cognitive disorder: epidemiological evidence on its validity. Psychol Med.1995;25:105-120.

71. Daly E, Zaitchck D, Copeland M, Schmahmann J, Gunther J, Albert M. Predict-ing conversion to Alzheimer’s disease using standardized clinical information.Arch Neurol. 2000;57:675-680.

72. Killiany RJ, Gomez-Isla T, Moss M, et al. Use of structural magnetic resonanceimaging to predict who will get Alzheimer’s disease. Ann Neurol. 2000;47:430-439.

73. Ritchie K, Artero S, Touchon J. Classification criteria for mild cognitive impair-ment: a population-based validation study. Neurology. 2001;56:37-42.

74. Morris JC, Storandt M, Miller JP, et al. Mild cognitive impairment represents early-stage Alzheimer disease. Arch Neurol. 2001;58:397-405.

75. Arnaiz A, Jelic V, Almkvist O, et al. Impaired cerebral glucose metabolism andcognitive functioning predict deterioration in mild cognitive impairment. Neu-roreport. 2001;12:851-855.

76. Petersen RC, Stevens J, Ganguli M, Tangalos EG, Cummings J, DeKosky ST. Prac-tice parameter: early detection of dementia: mild cognitive impairment. Neurol-ogy. 2001;56:1133-1142.



Current Concepts DCL

Documents

Transcript of Current Concepts DCL