Treating Mild Cognitive Impairment

12 • The Canadian Review of Alzheimer’s Disease and Other Dementias

Clinicians are becomingincreasingly aware ofthe number of elderly

individuals who fall into thatborderzone between normalityand dementia. While a numberof terms have been applied tothis group, by far the most com-mon term used is Mild CognitiveImpairment (MCI).1,2 In 2003,an international working groupagreed on general criteria forMCI: the subject was not judgedto be normal or demented, thecognitive decline was reportedby self and/or informant, therewas impairment on objectivecognitive tasks and there wasevidence of decline over time onsuch tasks. In addition, therewere preserved basic activitiesof daily living, or else minimalimpairment only in complexinstrumental functions.3,4

While this article does notaddress the larger question ofwhether MCI is a clinical entitythat deserves attention, (which hasbeen addressed in a number of arti-cles5-8), it does address current ther-apies for individuals with MCI,and assess their efficacy.

Reasons to Treat MCIThe first question to ask is: whyshould we treat MCI? One answer isthat the symptom of memory loss isupsetting to some patients. In otherwords, treatment may be sympto-matic because of the individual’sconcerns over memory loss in MCI.Notice that this level of concernvaries from individual to individual.Many patients recruited into a recentMCI trial at our centre stated thatthey did not feel sick, were not both-ered by their memory lapses, anddid not require medication. A sec-ond reason to treat MCI is to preventdevelopment of future dementiasince MCI is a “high-risk” state forfuture dementia (see Figure 1). Athird reason to treat MCI subjects isthat in fact many of the individuals

with MCI already have significantearly Alzheimer’s disease (AD)pathology9,10 and with time themajority of members of any MCIcohort will progress to AD. In thissense we are not attempting treat-ment to prevent AD, so much astreating to modify and slow the dis-ease at its earliest presentation.

While accepting the above, itmust be emphasized that MCI is het-erogeneous, and not all MCI individ-uals (at least in our experience andthat of others) progress to AD.11-13

Therefore in some cases, one wouldbe treating individuals to prevent pro-gression, who would not in fact haveprogressed. Thus, the “risk:benefit”ratio for treating MCI differs fromthat of AD. The bar for acceptablemedications must therefore be sethigher. The question is not whetherwe should treat, but what should con-stitute current recommended therapy.

Cognitive Training andStimulationLet us first look at non-pharmacolog-ic approaches. There are intriguinghints in the normal elderly that

Treating Mild Cognitive ImpairmentThe term Mild Cognitive Impairment (MCI) brings to light the recent efforts of physiciansto recognize the subtle classifications of normal and abnormal aging. While MCI clinicaltrials are currently underway, it will be years before their results are known. This articlereviews the current status of symptomatic and preventive therapies available for managingMCI, and discusses their efficacy.

by Howard Chertkow, MD, FRCPC

Dr. Chertkow is a Professor in theDepartment of Neurology andNeurosurgery, McGill University,and Co-Director, Jewish GeneralHospital Memory Clinic, McGillUniversity in Montreal, Quebec.

engagement in stimulating cognitiveactivities is associated with bettermemory and verbal abilities.14 Casecontrol and longitudinal studies haveshown that participation in intellectu-ally stimulating and social activitiesin midlife was associated withreduced risk of developing AD.15,16

What about randomized controlledtrials (RCTs) of cognitive stimula-tion? A number of studies haveshown that memory trainingimproved performance on targetedmemory tasks and that the effectsizes for the training effects were inthe moderate range and were sus-tained over a two-year follow-up.17,18

We do not know, however, if suchinterventions would in any way pre-vent or decrease dementia.

Several open and randomizedcontrolled trials have been reportedon the effect of cognitive trainingand intervention in MCI.19,20 Thesestudies provide encouraging findingsof benefit, but there are many ques-tions remaining.21,22 The effortrequired to implement cognitivetraining on a large scale is not trivial,and before widespread recommen-dation of this therapy can occur,more replication studies are requiredwith properly controlled RCTdesigns, larger sample sizes, andanalyses that control for Type 1 error.

Thus, the evidence at the pres-ent time is insufficient to concludethat organized cognitive interven-tion is beneficial in preventingprogression in MCI or warrantsprescription. On the other hand,

given that there is little or no“down-side” to cognitive activity,it is reasonable for physicians andtherapists to promote engagementin cognitive activity as part of anoverall “healthy lifestyle” formu-lation for elderly individuals withand without memory loss.

Physical TrainingThe situation with physical trainingand exercise is quite similar.Several longitudinal cohort studiescarried out in normal elderly indi-viduals indicate that physical exer-cise is associated with reduced cog-nitive decline and reduced risk ofdementia.23,24 However, there arealso studies that failed to find aprotective effect of physical exer-cise on cognitive decline and onincident dementia.16 Two recentmeta-analyses have been publishedregarding the impact of physicalexercise programs on the cognitivefunction of older adults.25,26 Both

meta-analyses reported moderateeffect sizes for the exercise trainingeffect on global cognitive scores andexecutive control. There are impor-tant implications of such research interms of potential public healthmeasures to prevent dementia andcognitive decline. More studies areneeded to assess the optimal exer-cise training modalities in olderadults, particularly in terms of inten-sity and duration. No studies havebeen carried out specifically withMCI persons to assess the effect ofphysical training on their cognitivecapacities and cognitive decline.

Keeping this in mind, the ThirdCanadian Consensus Conference onDiagnosis and Treatment ofDementia recommended that physi-cians and therapists may promotephysical activity at an intensity levelthat is adapted to the persons’overallphysical capacities, as part of a“healthy lifestyle” for older individu-als with and without memory loss.27

The Canadian Review of Alzheimer’s Disease and Other Dementias • 13

Figure 1

Classes of Therapy in MCI and AD

Secondary Prevention

AD

Primary Prevention

SymptomaticTreatment

Normal Mild CognitiveImpairment

Pre-symptomaticAD

Intervention

ClinicalState

BrainPathologicalState

Nodisease

Nosymptoms

Early brainchanges

No symptoms

AD brainchanges

Mild symptoms

Moderateto severechanges

Symptoms

Treatment of Exacerbatingand Co-morbid ConditionsThere are a series of other condi-tions that can exacerbate memoryloss in MCI, or even produce MCIin an otherwise cognitively normalelderly individual. Attention to

these factors is recommended, evenin the absence of formal RCTs.

For instance, it is increasinglyclear that stress, via cortisol levels,acts as a direct toxin on the hip-pocampus, capable of amplifyingdisease-related hippocampal dys-function.28-30 Recent studies onemotional factors such as pronenessto distress show this to be a signifi-cant risk factor for AD.31 Attempts toreduce stress levels in MCI seem areasonable goal. Although there havenot been direct clinical studies,untreated depression will exacerbateand amplify memory loss.35-39

An important study in theKungsholmen district of Stock-holm demonstrated that a poor orlimited social network increased therisk of dementia by 60% and a sig-nificant gradient was found forincreasing degrees of social connec-tions It appears that an extensivesocial network seems to protectagainst dementia.40 Clearly thisrequires a lifelong commitment to

building social interactions, butthis may be a modifiable risk, andsocial interaction can be encour-aged in MCI individuals.

Also, patients with sleep disor-ders often present with memoryloss, and this seems a reasonable

factor to modify and control.32-34

Assessment and treatment of sleepapnea in MCI patients is recom-mended if there is a sleep complaint.

Symptomatic Therapy There is no current treatment forMCI sufficiently substantiated tohave obtained government approvalfrom FDA or Canadian governmentregulators. This contrasts with thesituation in Europe, where medica-tions such as Ginkgo biloba andhydergine have been approvedbroadly for “memory impairment.”While some reviews have tended tobe fairly positive about treatmenteffects,41 symptomatic therapy isgenerally found to be disappointing,although the occasional patientappears to have a significantimprovement on each of the medica-tions listed in Table 1.

Cholinesterase inhibitors. Thethree available cholinesteraseinhibitors (CIs) in NorthAmerica—donepezil, rivastigmine

and galantamine—are approved fortreatment of mild to moderate AD,not MCI. These all produce modestimprovement and stabilization in themajority of patients.42,43

Symptomatic treatment of thememory complaints in MCI withCIs is generally disappointing.Clinicians have anecdotally report-ed that certain MCI patients benefitfrom treatment with CIs in terms ofmemory and global function.Salloway et al studied 270 patientsacross 20 centres meeting criteriafor amnestic MCI.45 Half weretreated with donepezil 10 mg forsix months and a series of cogni-tive and global tests were adminis-tered. Two thirds of the donepezil-treated cohort completed the study,and change in a paragraph recalltest as well as the Clinical GlobalImpression of Change-MCI instru-ment were used as the primary out-come measures. Neither of thesemeasures showed significant benefi-cial effects of therapy at the end ofsix months. However, a major sec-ondary measure, the ADAS-Cog, didshow a symptomatic benefit.Subjectively, patients treated withdonepezil reported greaterimprovement in memory functionthan those given placebo. Theyreported feeling sharper mentally,more organized, and more confi-dent of their memory. All of thissuggests that at least some MCIindividuals will have a significantclinical benefit from CIs, but over-all the effects are mild. The recent


There are a series of other conditions that canexacerbate memory loss in MCI, or even produce MCI inan otherwise cognitively normal elderly individual.Attention to these factors is recommended, even in theabsence of formal RCTs.

Canadian Consensus Conference ondementia therapy did not recommendCIs as therapy in MCI.27

Ginkgo biloba. At present,Ginkgo biloba is commonly pre-scribed in Europe for all memory-impaired patients, with the idea thatit improves blood and oxygen flowto the brain and supports memoryfunction, mental sharpness and cir-culation. There are few if any well-designed clinical trials that supportthis conclusion.46 There was oneplacebo-controlled study of Ginkgobiloba in AD, with a high dropoutrate. This showed a significantsymptomatic benefit in AD, albeitapproximately a quarter of the effica-cy of CIs.47 It is notable that in somecountries such as Germany, Ginkgobiloba is routinely prescribed for ADbecause of its greater accessibilityand lower cost to patients.

There is an ongoing long-termstudy testing the hypothesis thatGinkgo biloba, as an anti-oxidant,might prevent onset or slow pro-gression of AD. Data are not yetavailable. The Canadian ConsensusConference concluded that there iscurrently fair evidence to recom-mend against the use of Ginkgobiloba therapy in MCI.

Nootropics. There are a numberof over-the-counter “dietary supple-ments” which have been suggestedto strengthen and protect neurons ofthe brain involved in memory, serv-ing as “memory nutrients.” These“nootropics” have non-specificmechanisms of action, with putative

effects on energy metabolism,cholinergic mechanisms, excitatoryamino acid receptor-mediated func-tions, as well as hormonal mecha-nisms.49 In this class, one would listphosphatidyl-serine (PS), acetyl-l-carnitine and piracetam. These areavailable through health-food storesas diet supplements, not medica-tions. Evidence of their efficacy isslim, but they have few if any sideeffects. Presumably, these nootropicswould have symptomatic rather thanpreventive effects in MCI.

PS is obtained from cows andmore recently a form derived fromsoy lecithin is being sold. A PSstudy from 1991 had subjects withmild memory loss, similar to MCI,taking PS 300 mg for three months.The subjects showed some modestimprovement in their memory. Theeffects tended not to occur in every-one and there was no benefit in ADpatients.50 There have been noserious studies of PS in the pastten years. Piracetam was tested inseveral studies in individuals who

The Canadian Review of Alzheimer’s Disease and Other Dementias• 15

Table 1

Symptomatic and Preventive Therapies for MCI

Symptomatic therapies

Cholinesterase inhibitors

Ginkgo biloba

Nootropic medications- phosphatidyl-serine (PS)- acetyl-L-carnitine,- choline (phosphatidylcholine, citicoline)- piracetam

Memory stimulants- ampakines - NMDA receptor modulation)- CREB modulators

Preventive therapies

Anti-oxidants- vitamin E- other anti-oxidants (selegiline, vitamin C)

Homocysteine

Omega fatty acids

Cholinesterase inhibitors

Anti-inflammatory agents

Estrogen

Statins

Anti-amyloid therapies- beta- and gamma-secretase inhibitors- GAG-mimetics

Therapy for vascular risk factors


may have had MCI.51 There was evi-dence of mild improvement in mem-ory and attention.

Overall, despite lofty claims fordramatic effects of nootropics,(generally from those involved inmarketing of these agents), theproven benefit of each of theseagents is modest. One broadreview stated in conclusion “All inall, we believe that the current data

do not allow strong scientificallybased recommendations for any ofthese memory nutrients (includingPS and ginkgo). However, the dataalso do not allow us to concludethat these nutrients are ineffectivein boosting memory.”46

Memory Stimulants andFuture Smart DrugsThe media have been reporting the“imminent” arrival of medicationsthat will impact the neurochemicalprocesses of memory itself. Suchmedications have the potential tocompensate for the neurochemicaldeterioration thought to be part ofMCI and even AD, without chang-ing or retarding the underlyingpathological processes.52 The can-didate drugs directed at improv-ing memory fall into one of two

categories: those that target the ini-tial induction of long-term potentia-tion, and those that target the laterstages of memory consolidation.Drug candidates include ampakineswhich are already beginning toenter Phase II clinical trials for MCItreatment.53-56 The efficacy and side-effect profiles of these cognitiveenhancers are unknown. In the sec-ond category we find drugs aimed at

increasing CREB (cyclic-AMPresponse element binding protein),the element-binding protein which inturn activates genes to produce pro-teins that strengthen the synapse inresponse to experience—the basis oflong-term memory. A number ofmechanisms are being explored thatcan impact on the CREB level.57

Human trials are still years away.

Prevention of AD byIntervention at the MCI StageThe most critical interest from physi-cians, patients, and the pharmaceuticalindustry is in pharmacologic interven-tions that can be instituted at the MCIstage to prevent progression to demen-tia, specifically AD. Most readers willbe aware that no such effectivemedication currently exists, hasbeen substantiated by multiple or

convincing randomized placebo-con-trolled clinical trials, or is available inpharmacies. There are nine kinds ofpotential prevention therapies thatcannot currently be recommended(see Table 2). These each couldtheoretically delay or prevent pro-gression to AD from the MCI state,but have failed to reach sufficientstrength of evidence to be recom-mended. The theoretical argumentsare based on understanding of ADpathophysiology or evidence frompopulation studies. Studies testingmost of these mechanisms are cur-rently underway. It is important topoint out, however, that retrospectiveobservational studies are not the sameas carefully controlled randomizedintervention studies. This has beenbrought harshly to the forefront by thefailure of a number of RCTs to con-firm efficacy of interventions derivedfrom epidemiologic studies. In thesecases, either the population evidencewas simply wrong, or the “criticalperiod” for the pharmacologic inter-vention was missed, and MCI wassimply too late a time to treat. We willmention only one of these, vitamin E,along with positive therapy recom-mendations for treating vascular riskfactors, and perhaps suggestingdietary manipulations.

Should We Be Using Vitamin Eand Anti-oxidants?Vitamin E, an anti-oxidant, has had aroller-coaster profile as a medicationto prevent dementia, and this high-lights the challenges and difficulties

At present, Ginkgo biloba is commonly prescribed inEurope for all memory-impaired patients, with theidea that it improves blood and oxygen flow to thebrain and supports memory function, mentalsharpness and circulation. There are few if any well-designed clinical trials that support this conclusion.46

in deriving preventive therapies for achronic disease like AD. The case foranti-oxidant therapy to prevent onsetof AD (as well as other neurodegen-erative diseases and aging in general)is relatively strong, and new evidencecontinues to accumulate, althoughnone reach the level of recommenda-tions for therapy. Oxidative damagecan be found in a number of neu-rodegenerative conditions includingAD.58,59 In a widely cited study ofvitamin E in AD,60 patients takinghigh-dose vitamin E for up to 24months reached the functional mile-stone of institutionalization more

slowly than individuals on place-bo. The treatment was regarded assafe; vitamin E was not associatedwith increased risk of death.Indeed, an identical number ofsubjects taking vitamin E diedduring the course of the trial com-pared to patients taking placebo.Based on this single study, and thetheoretical benefits of anti-oxi-dants in preventing AD and cogni-tive decline as well as aging ingeneral, a large number of AD andMCI subjects are currently pre-scribed vitamin E, or obtain itthemselves from pharmacies.

The evidence for a benefit fromvitamin E in preventing or delayingAD derives from a set of epidemio-logic studies.61-63 In contrast tothis is new evidence from thelarge “Memory ImpairmentStudy,” in which individuals withMCI were recruited and random-ized into a vitamin E therapy arm, adonepezil therapy arm, or a placeboarm. The crucial primary endpointwas the number of subjects classi-fied as progressing to dementia atthe end of three years. This studyassessed the effects of five daily cap-sules of vitamin E (2000 IU) and


Table 2

Non-recommended Therapies Aimed at Preventing Progression from MCI to AD

Class Name of medications Mechanism of action Evidence for use Negative trials

Vitamin E Vitamin E Reduces oxidative stress Population studies Meta-analyses [61-63], RCT [60] [64,65]

Alternative Vitamins A, C, Reduce oxidative stress As above, also [81] RCT [64]anti-oxidants Gingko biloba

Vitamins B Vitamins B6, B12, Reduce homocysteine Framingham Study No RCTs as yetcomplex folic acid levels [77]

Omega fatty Docosahexaenoic Role in neuronal Observational No RCTs as yetacids acid (DHA) communication studies

Cholinesterase Donepezil, Increase synaptic Memory Impairment RCTs [64],[66]inhibitors rivastigmine, acetylcholine availability Study [64] (for ApoE4

galantamine carriers)

Anti-inflammatory Ibuprofen, Reduce inflammatory Observation study RCT [83]agents indomethacin, response [82]

prednisone

Estrogen Estradiol, Multiple mechanisms of Observation studies RCTs: WHIMS raloxifene action [84,85], RCT [86] [87,88] Nurses

Health Study [89]

Statins Atorvastatin, Block liver enzyme Observational studies No RCTs in MCI five others essential for cholesterol [90, 91], RCT [92] as yet

production

Anti-amyloid GAG-mimetics, Prevents production Promising RCTs No RCTs in MCI therapy immunotherapy (immunotherapy) or [93,94], but as yet

aggregation (GAG-mimetics) immunotherapy of A-beta amyloid fragments complications (encephalitis)

found no overall benefit in the vita-min E group in terms of preventionof progression to AD at the end ofthree years.64

Furthermore, a recent meta-analysis raised questions about thesafety of vitamin E when given atsuch a high dose. Miller et al exam-ined the number of deaths in 19 clin-ical trials of vitamin E, including atotal of 136,000 subjects.65 None ofthe individual studies showed anincrease in risk of death for subjectstaking vitamin E alone. However,when the studies were arranged bydose of vitamin E (above or belowthe 400 IU/day median dose), itappeared that individuals taking lowto moderate doses of vitamin E had avery slight protection against death

while those taking high-dose vitaminE were at a very slightly higher riskof death. There are, however, numer-ous methodological weaknesses inthis study.

Despite the meta-analysis, therisk of vitamin E also appearsminimal. Might a single 400 IUvitamin E tablet be beneficiallyand safely prescribed for an MCIindividual? They receive this inmy own clinic, albeit with somehesitation. The recent CanadianConsensus Conference concluded

that there is currently fair evi-dence to recommend against theuse of vitamin E therapy in MCI.

Can Cholinesterase InhibitorsSlow and Prevent AD?The “Memory Impairment Study”described earlier was an importantthree-year trial co-sponsored by theNational Institute of Aging (NIA),the Alzheimer Disease CooperativeStudy group (ADCS), and Pfizer. Inthis study, individuals with MCIwere recruited and randomized intoa vitamin E arm (2000 IU daily), adonepezil arm (10 mg daily), or aplacebo arm. The crucial primaryendpoint was the number of subjectsclassified as progressing to dementiaat the end of three years. The result

was negative—no significant differ-ences between the three groups werefound at three years.64 This disap-pointing result seems to put to restthe possibility of CIs as effectivetherapies to prevent AD, but there aresub-analyses that seem still to offerpromise. For instance, it is clear thatthe group of individuals takingdonepezil performed better thanthe others over the first 18 months,in terms of neuropsychologicalmeasures and global outcomes. Itis also clear that the majority of

individuals progressing to AD hadan Apo-E4 allele, and if the analysisis restricted only to those individu-als, there were indeed less “conver-sions to dementia” with donepezilat the end of three years. Currently,debate rages on whether this “nega-tive study” might be reinterpreted asa positive result for a particular sub-group of patients. There are also theusual methodological concerns(heterogeneity of patients, weakoutcome measure) that make it hardto achieve significant results evenwith large numbers of subjects.

The other main CIs are alsobeing assessed for their potentialto slow progression to AD. Thegalantamine trial was a two-yearstudy focusing on amnestic MCIpatients with memory below acutoff on paragraph recall. Therewas no difference in the primaryanalysis of conversion fromamnestic MCI to AD. There didappear to be a reduced rate ofwhole-brain atrophy in the patientstreated with galantamine.66

However, therapy with this med-ication was associated with asmall but statistically significantincreased risk of dying. Thesponsoring company discontin-ued the trial and has not recom-mended therapy for MCI withgalantamine.67

The bottom line is that fromthe current evidence, if there is abenefit of CIs in slowing progres-sion to AD, it appears to be tran-sient as well as limited.68


There are nine kinds of potential prevention therapiesthat cannot currently be recommended. These eachcould theoretically delay or prevent progression to ADfrom the MCI state, but have failed to reach sufficientstrength of evidence to be recommended.

Therapy for Vascular RiskFactorsIt is clear that vascular damageimpacts on the occurrence of ADand mixed dementia. Risks forvascular disease (diabetes, hyper-tension, smoking, obesity, hyper-lipidemia) are being proven to be riskfactors for the development ofdementia. In the Rotterdam Study inthe Netherlands, individuals with dia-betes had nearly double the risk ofdementia.69 Presumably, the micro-vascular damage from diabetes is theculprit, although it is possible thathigher-than-normal levels of glucosein the blood might be toxic. TheFramingham Heart Study demon-strated an impact of hypertension oncognition six years later.70 TheCardiovascular Health Study showedthat cognitive decline occurred evenwithout frank stroke in individualswith vascular risk factors.71

These data give us additionalapproaches to therapy of MCI, name-ly aggressively treating vascular riskfactors. Several randomized con-trolled studies of antihypertensivesafter stroke have shown a clear effectin reducing the subsequent incidenceof dementia.72-74 One clinical trialevaluated the role of hypertensiontreatment in individuals with mildcognitive deficits broadly defined asa Mini-Mental Status Examinationscore between 20 and 28.75 Patientswith the best response to treatment, interms of reduction of their diastolicblood pressure, significantly impro-ved on two cognitive tests.

Note that this therapy isunproven in the sense that no onehas yet mounted a long-termstudy proving that intervention atthe MCI stage will be effective inreducing or preventing dementiaor AD. However, treating theserisks makes sense in its ownright—a patient with uncontrolledhypertension should be treatedanytime. Thus, in recommenda-tions to family physicians, thoseworking to prevent AD are nowgiving strong advice to “do whatyou already do”—namely, aggres-sively treat any risk factors for

vascular disease, in a patient withMCI. The risk of dementia thusrepresents an additional reason totreat the patient.

The MCI DietGiven the factors identified above, itis interesting to note that dietaryinterventions are possible in thetreatment of MCI or prevention ofdementia. A healthy diet helps pre-vent hypertension (via reduced sat-urated fats and sodium), prediabetes(reduce sweets and caloric intakeand consume more fibre), andstroke (dietary change to reducecholesterol). Obesity is to be avoid-ed by dietary limitation and exer-cise. One study reported a higher

risk of AD in seniors who ate moresaturated and trans fat and lessunsaturated fat, but another studydid not find the same link.76

Analysis of the Framinghamstudy produced the somewhat sur-prising result that higher homo-cysteine levels were associatedwith increased risk of sporadicAD.77 It is known that increasedserum homocysteine is associatedwith histopathologic evidence ofvascular endothelial injury, vascularsmooth muscle proliferation, andprogressive arterial stenosis.78 Thefactors in homocysteine levels are

well known—vitamin supplements(folate, B6, B12) lower the levels,while caffeine, smoking, and lack ofexercise increase levels. Currentmanagement of elevated homocys-teine has been to increase folate inthe diet or treat with supplementswhen increased homocysteine wasgreater than 15 µmol/L. Simpletreatment with folate (3 mg daily),B6 (25 mg daily), and B12 (250 to500 µg daily) keeps the homocys-teine level low. Homocysteine levels(high or even normal) can in theorybe reduced by a good intake offolic acid, B6, and B12 found ingreen leafy vegetables. Severalmulti-vitamins a day will alsosupply these amounts.


Vitamin E, an anti-oxidant, has had a roller-coasterprofile as a medication to prevent dementia, and thishighlights the challenges and difficulties in deriving

preventive therapies for a chronic disease like AD.

Omega fatty acids, particularlyDHA (docosahexaenoic acid), canbe obtained by eating cold-waterfatty fish such as salmon, sar-dines, mackerel, and bluefish.There is evidence that individualswhose diets are high in omega-3fatty acids, especially DHA, havea 50% reduction in their risk ofdeveloping dementia.79,80 Some

doctors are now recommendingthat MCI patients eat such fish (ortake two 200 mg DHA capsules)three times weekly.79,80 This ther-apy also did not receive recom-mendation from the CanadianConsensus Conference on demen-tia, however. About 180 mg ofDHA daily intake is suggested andthis amount can be achieved by eat-ing the fish previously mentionedabout three times per week. Thus,there is a theoretical basis for con-siderable dietary manipulation inMCI, none yet supported by RCTs.

Current Treatment of MCIGiven the lack of clear prognosticmarkers, heterogeneity in the nat-ural history of MCI individuals,and lack of proven therapies toprevent decline, the managementof MCI patients remains largelynon-specific. The strongest evi-dence supports suggestions to

“maintain a healthy lifestyle” withadequate exercise, avoidance ofobesity, mental and physical stimu-lation, control of stress, treatmentof medical illnesses and depres-sion, and control of vascular riskfactors such as diabetes, hypeten-sion, and hypercholesterolemia.

Currently we lack proven phar-macologic approaches to prevent

cognitive decline or progressionfrom MCI to dementia. It makesgood sense to aggressively treatvascular risk factors in MCI indi-viduals using lifestyle interven-tions, diet and medications whennecessary. Pharma-cologic treat-ment of depr-ession is also indi-cated. Drugs with known anti-cholinergic activity, as well assleeping pills and sedatives,should be avoided.

Physicians should inform pa-tients that there is no current spe-cific treatment for MCI sufficient-ly substantiated to have obtainedgovernment approval. TreatingMCI individuals (or the healthyelderly) in order to prevent subse-quent AD using CIs, anti-inflam-matories, estrogen, statins, vari-ous antioxidants or even vitaminE, represents prescription beyondproven therapies. It is advisable torefer the eager MCI patient to a

research clinic where RCTs ofthese and other preventive med-ications are currently underway.

The Memory ImpairmentStudy raises significant therapeu-tic issues. Since there was a delayin progression to AD in MCIindividuals with a positiveApoE4 perhaps “a discussion oftherapy” with donepezil is war-ranted. But what discussion?Should MCI patients be offeredgenetic testing with ApoE priorto a therapy decision? Shoulddonepezil be offered with thehope that the symptomatic bene-fit will make up for our uncer-tainty regarding its long-termprevention role? Should its rolein MCI (and the role of CIs ingeneral) be downplayed as amajor part of our therapeuticarmamentarium, rather thanusing them up at the MCI stage?

In our own clinic, we haveexploited the dietary possibilitiesnoted above as a way to maxi-mize the potential beneficialeffects of anti-oxidants andomega fatty acids, and to controlhomocysteine levels. Ourpatients generally are encour-aged to take one or two multivit-amins daily, which deliver ade-quate doses of vitamin E (400IU), along with B6 and folatesupplementation. This is reallyall they are offered at the currenttime. The rest remains in therealm of current research andfuture possibilities.


Thus, in recommendations to family physicians, thoseworking to prevent AD are now giving strong advice to“do what you already do”—namely, aggressively treat anyrisk factors for vascular disease, in a patient with MCI.

References:1. Petersen RC, Mild cognitive impairment as

a diagnostic entity. J Intern Med 2004;256(3):183-94.

2. Chertkow H, Mild Cognitive Impairment.Curr Opin Neurol 2002; 15(4):401-7.

3. Winblad B, Palmer K, Kivipelto M, et al.Mild cognitive impairment—beyond con-troversies, towards a consensus: report ofthe International Working Group on MildCognitive Impairment. J Intern Med 2004;256(3):240-6.

4. Tombaugh TN, McIntyre NJ. The Mini-men-tal state examination: A comprehensivereview. J Am Geriatr Soc 1992; 40:922-35.

5. Petersen RC and JC Morris, Mild cognitiveimpairment as a clinical entity and treat-ment target. Arch Neurol 2005; 62(7):1160-3; discussion 1167.

6. Davis HS, Rockwood K. Conceptual-izationof mild cognitive impairment: a review. Int JGeriatr Psychiatry 2004; 19(4):313-9.

7. Luis CA, Lowenstein DA, Acevedo A, et al.Mild cognitive impairment: directions forfuture research. Neurology 2003; 61(4):438-44.

8. Gauthier S, Reisberg B, Zaudig M et al.Mild cognitive impairment. The Lancet2006; 367:1262-70.

9. Morris JC, Price AL. Pathologic correlates ofnondemented aging, mild cognitive impair-ment, and early-stage Alzheimer's disease. JMol Neurosci 2001; 17(2):101-18.

10. Morris JC, Storandt M, Miller JP, et al. Mildcognitive impairment represents early-stageAlzheimer disease. Arch Neurol 2001;58(3):397-405.

11. Ritchie K, Touchon J. Mild cognitive impair-ment: conceptual basis and current noso-logical status. Lancet 2000; 355:225-8.

12. Bocti C, Whitehead V, Fellow L, et al.Characteristics of patients with MildCognitive Impairment who do not progressto dementia. 57th Annual MeetingAmerican Academy of Neurology 2005;Miami Beach, Florida.

13. Fisk JD, Rockwood K. Outcomes of incidentmild cognitive impairment in relation tocase definition. J Neurol Neuro-surgPsychiatry 2005; 76(8):1175-7.

14. Kramer AF, Bherer L, Colcombe SJ, et al.Environmental influences on cognitive andbrain plasticity during aging. J Gerontol ABiol Sci Med Sci 2004; 59(9):M940-57.

15. Lindstrom HA, Pritsch T, Petot G, et al. Therelationships between television viewing inmidlife and the development of Alzheimer'sdisease in a case-control study. Brain Cogn2005; 58(2):157-65.

16. Wilson RS, Mendes De Leon CF, Barnes LL,et al. Participation in cognitively stimulatingactivities and risk of incident Alzheimer dis-ease. JAMA 2002; 287(6):742-8.

17. Verhaeghen P, Marcoen A, Goossens L.Improving memory performance in the agedthrough mnemonic training: a meta-analyticstudy. Psychol Aging 1992; 7(2):242-51.

18. Ball K, Berch DB, Helmers KF, et al. Effectsof cognitive training interventions with olderadults: a randomized controlled trial. JAMA2002; 288(18):2271-81.

19. Gunther VK, Schafer P, Holzner BJ, et al.

Long-term improvements in cognitive per-formance through computer-assisted cogni-tive training: a pilot study in a residentialhome for older people. Aging Ment Health2003; 7(3):200-6.

20. Olazaran J, Muniz R, Reisberb B, et al.Benefits of cognitive-motor intervention inMCI and mild to moderate Alzheimer dis-ease. Neurology 2004; 63(12):2348-53.

21. Belleville S, et al. Improvement of episodicmemory in persons with Mild CognitiveImpairment and healthy older adults:Evidence from a cognitive intervention pro-gram. (in press). Dement Geriatr CognDisord, 2006.

22. Rapp S, Brenes G, Marsh AP. Memoryenhancement training for older adults withmild cognitive impairment: a preliminarystudy. Aging Ment Health 2002; 6(1):5-11.

23. Lytle ME, Vander Bilt J, Pandav RS, et al.Exercise level and cognitive decline: theMoVIES project. Alzheimer Dis AssocDisord 2004; 18(2):57-64.

24. Laurin D, Verreault R, Lindsay J, et al.Physical activity and risk of cognitiveimpairment and dementia in elderly per-sons. Arch Neurol 2001; 58(3):498-504.

25. Colcombe S, Kramer AF. Fitness effects onthe cognitive function of older adults: ameta-analytic study. Psychol Sci 2003;14(2):125-30.

26. Heyn P, Abreu BC, Ottenbacher KJ. Theeffects of exercise training on elderly per-sons with cognitive impairment and demen-tia: a meta-analysis. Arch Phys Med Rehabil2004; 85(10):1694-704.

27. Chertkow H, Nasreddinne Z, Massoud F, etal. Mild Cognitive Impairment—Recommendations on Diagnosis andTherapy from the Third CanadianConsensus Conference on the Diagnosisand Treatment of Dementia. BMCGeriatrics, submitted 2006.

28. Lupien SJ, Nair NP, Briere S, et al. Increasedcortisol levels and impaired cognition inhuman aging: implication for depressionand dementia in later life. Rev Neurosci1999; 10(2):117-39.

29. Sapolsky R, Krey L, McEwen B. The neu-roendocrinology of stress and aging: theglucocorticoid cascadde hypothesis.Endocrinol Review 1986; 7:284-301.

30. Issa AM, Rowe W, Gauthier S, et al.Hypothalamic-pituitary-adrenal activity inaged, cognitively impaired and cognitivelyunimpaired rats. J Neurosci 1990;10(10):3247-54.

31. Wilson RS, Barnes LL, Bennett DA, et al.Proneness to psychological distress is asso-ciated with risk of Alzheimer's disease.Neurology 2003; 61(11):1479-85.

32. Bliwise DL. Is sleep apnea a cause ofreversible dementia in old age? JAGS 1996;44:1407-9.

33. Boeve BF, Silver MH, Ferman TJ, et al.Association of REM sleep behavior disorderand neurodegenerative disease may reflectan underlying synucleinopathy. Mov Disord2001; 16(4):622-30.

34. Riemann D, Hohagen F, Krieger S, et al.Cholinergic neurotransmission, REM sleepand depression. J Psychosom Res 1994;

38(Suppl 1):15-25.35. Adler G, Bramesfeld A, Jajcevic A, Mild

cognitive impairment in old-age depressionis associated with increased EEG slow-wavepower. Neuropsychobiology 1999; 40(4):218-22.

36. Alexopoulos GS, Heterogeneity and comor-bidity in dementia-depression syndromes[editorial]. Int J Geriatr Psychiatry 1991;6(3): 125-7.

37. Coen RF, Kirby M, Swanwick GR, et al.Distinguishing between patients withdepression or very mild Alzheimer's diseaseusing the delayed-word-recall test. DementGeriatr Cogn Disord 1997; 8(4): 244-7.

38. Reischies FM, Neu P. Comorbidity of mildcognitive disorder and depression—a neu-ropsychological analysis. Eur ArchPsychiatry Clin Neurosci 2000; 250(4):186-93.

39. Alexopoulos G, Schultz SK, Lebowitz BD,et al. The course of geriatric depression with"reversible dementia": a controlled study.Am J Psychiatry 1993; 150(11):1693-9.

40. Fratiglioni L, et al. Influence of social net-work on occurrence of dementia: a com-munity-based longitudinal study. Lancet2000; 355(9212): 1315-9.

41. Ihl R. The impact of drugs against dementiaon cognition in aging and mild cognitiveimpairment. Pharmacopsychiatry 2003; 36Suppl 1:S38-43.

42. Lanctot KL, Herrman N, Yau KK, et al.Efficacy and safety of cholinesteraseinhibitors in Alzheimer's disease: a meta-analysis. CMAJ 2003; 169(6):557-64.

43. Patterson C, Gauthier S, Bergman H, et al.The recognition, assessment and manage-ment of dementing disorders: conclusionsfrom the Canadian Consensus Conferenceon Dementia. CMAJ1999; 160(suppl12):S1-S15.

44. Reisberg B, Doody R, Stoffler A, et al.,Memantine in moderate-to-severeAlzheimer's disease. N Engl J Med, 2003;348(14):1333-41.

45. Salloway S, Ferris S, Kluger A, et al.Efficacy of donepezil in mild cognitiveimpairment: a randomized placebo-con-trolled trial. Neurology 2004; 63(4): 651-7.

46. McDaniel MA, Maier SF, Einstein GO."Brain-specific" nutrients: a memorycure? Nutrition 2003; 19(11-12):957-75.

47. Le Bars PL, Katz MM, Berman N, et al. Aplacebo-controlled, double-blind, ran-domized trial of an extract of Ginkgobiloba for dementia. North AmericanEGb Study Group. JAMA 1997;278(16):1327-32.

48. Van Dongen M, Van Rossum E, KesselsA, et al. Ginkgo for elderly people withdementia and age-associated memoryimpairment: a randomized clinical trial. JClin Epidemiol 2003; 56(4):367-76.

49. Riedel WJ, Jolles J. Cognition enhancersin age-related cognitive decline. Drugand Aging 1996; 8(4):245-74.

50. Crook TH, Ferris SH, Alvarez XA, et al.Effects of phosphatidylserine in age-asso-ciated memory impairment. Neurology1991; 41(5):644-9.


51. Israel L, Melac M, Milinkevitch D, et al.Drug therapy and memory training pro-grams: a double-blind randomized trial ofgeneral practice patients with age-associ-ated memory impairment. IntPsychogeriatr 1994; 6(2):155-70.

52. Tully T, Bourtchouladze R, Scott R, et al.Targeting the CREB pathway for memoryenhancers. Nat Rev Drug Discov 2003;2(4):267-77.

53. Lynch G. Memory enhancement: thesearch for mechanism-based drugs. NatNeurosci, 2002. 5 Suppl: 1035-8.

54. Farah MJ, Illes J, Cook Deegan R, et al.Neurocognitive enhancement: what canwe do and what should we do? Nat RevNeurosci 2004; 5(5):421-5.

55. Pepeu G. Overview and perspective onthe therapy of Alzheimer’s disease from apreclinical viewpoint. Progress in Neuro-Psychopharmacology & BiologicalPsychiatry 2001; 25(1):193-209.

56. Danysz W. CX-516 Cortex pharmaceuti-cals. Curr Opin Investig Drugs 2002;3(7):1081-8.

57. Squire LR, Kandel ER, Memory: FromMind to Molecules. New York(1999):W.H. Freeman & Co.

58. Markesbery WR. Oxidative stress hypoth-esis in Alzheimer’s disease. Free RadicBiol Med 1997; 23(1):134-47.

59. Perry G, Nunomura A, Hirai K, et al. Isoxidative damage the fundamental patho-genic mechanism of Alzheimer’s andother neurodegenerative diseases? FreeRadic Biol Med 2002; 33(11):1475-9.

60. Sano M, Ernesto C, Thomas RG, et al. Acontrolled trial of selegiline, alpha-toco-pherol, or both as treatment forAlzheimer’s disease. The Alzheimer’sDisease Cooperative Study. N Engl J Med1997; 336(17):1216-22.

61. Morris MC, Evans DA, Brenias JL, et al.Vitamin E and cognitive decline in olderpersons. Arch Neurol 2002; 59(7):1125-32.

62. Engelhart MJ, Geerlings MI, Ruittenberg A,et al. Dietary intake of antioxidants andrisk of Alzheimer disease. JAMA 2002;287(24):3223-9.

63. Fotuhi M, Hayden KM, Zandi P, et al. Useof NSAIDs and antioxidant supplements incombination reduces the rate of cognitivedecline. the cache county study. Alzhei-mer’s & Dementia, the journal of theAlzheimer’s association 2005; 1:S97-8.

64. Petersen RC,Thomas RG, Grundman M,et al. Vitamin E and donepezil for thetreatment of mild cognitive impairment. NEngl J Med 2005; 352(23):2379-88.

65. Miller ER, Pastor-Barriuso R, Dalal D, etal. Meta-analysis: high-dosage vitamin Esupplementation may increase all-causemortality. Ann Intern Med2005;142(1):37-46.

66. Scheltens P, Fox NC, Barkhof F, et al.Effect of galantamine treatment on brain

atrophy as assessed by MRI in patientswith mild cognitive impairment. NeurobiolAging 2004; 25(suppl.2):S270-1.

67. Advisory, H.C.P., Health Canada EndorsedImportant Safety Information on Reminyl(Galantamine). January 2005. available at:http://www.hc-sc.gc.ca/hpfb-dgpsa/tpd-dpt/index_advisories_public_e.html.

68. Blacker D, Mild cognitive impairment-nobenefit from vitamin E, little fromdonepezil. N Engl J Med, 2005.352(23):2439-41.

69. Ott A, Van Harskamp F, Stolk RP, et al.Diabetes mellitus and the risk of demen-tia: The Rotterdam Study. Neurology1999; 53(9):1937-42.

70. Elias MF, Elias PK, Sullivan LM, et al.Lower cognitive function in the presenceof obesity and hypertension: theFramingham heart study. Int J Obes RelatMetab Disord 2003; 27(2):260-8.

71. Elkins JS, O’Meara ES, Longstreth WT, etal. Stroke risk factors and loss of high cog-nitive function. Neurology 2004; 63(5):793-9.

72. Forette F, Seux ML, Staessens JA, et al.Prevention of dementia in randomiseddouble-blind placebo-controlled SystolicHypertension in Europe (Syst-Eur) trial.Lancet 1998; 352(9137):1347-51.

73. Shepherd J, Blauw GJ, Murphy MB, et al.Pravastatin in elderly individuals at risk ofvascular disease (PROSPER): a ran-domised controlled trial. Lancet 2002;360(9346):1623-30.

74. Tzourio C, Dufouil C, Ducimetiere P, et al.Cognitive decline in individuals with highblood pressure: a longitudinal study in theelderly. EVA Study Group. Epidemiologyof Vascular Aging. Neurology 1999;53(9):1948-52.

75. Starr JM, Deary IJ. The effects of antihyper-tensive treatment on cognitive function:results from the HOPE study. J Am GeriatrSoc 1996; 44:411-5.

76. Morris MC, Evans DA, Brenias JL, et al.Dietary fats and the risk of incidentAlzheimer disease. Arch Neurol 2003;60(2):194-200.

77. Seshadri S, Beiser A, Selhub J, et al.Plasma homocysteine as a risk factor fordementia and Alzheimer’s disease. N EnglJ Med, 2002. 346(7):476-83.

78. Miller JW, Green R, Mungus DM, et al.Homocysteine, vitamin B6, and vasculardisease in AD patients. Neurology 2002;58(10): 1471-5.

79. Engelhart MJ, Geerlings MI, Ruitenberg A,et al. Diet and risk of dementia: Does fatmatter? The Rotterdam Study. Neurology2002; 59(12):1915-21.

80. Kalmijn S, Feskens EJ, Launer LJ, et al.Polyunsaturated fatty acids, antioxidants,and cognitive function in very old men.Am J Epidemiol, 1997. 145(1):33-41.

81. Perrig WJ, Perrig P, Stahelin HB. The rela-tion between antioxidants and memory

performance in the old and very old. J AmGeriatr Soc 1997; 45(6):718-24.

82. Etminan M, Gill S, Samii A. Effect of non-steroidal anti-inflammatory drugs on riskof Alzheimer’s disease: systematic reviewand meta-analysis of observational studies.BMJ 2003; 327(7407):128-31.

83. Aisen PS, Schafer KA, Grundman M, et al.Effects of rofecoxib or naproxen vs place-bo on Alzheimer disease progression: arandomized controlled trial. JAMA 2003;289(21): 2819-26.

84. Yaffe K, Sawaya G, Lieberburg I, et al.Estrogen therapy in postmenopausalwomen: effects on cognitive function anddementia. JAMA 1998; 279(9): 688-95.

85. Maki PM, Resnick SM. Effects of estrogenon patterns of brain activity at rest andduring cognitive activity: a review of neu-roimaging studies. Neuroimage 2001;14(4):789-801.

86. Yaffe K, Krueger K, Cummings SR, et al.Effect of raloxifene on prevention of dem-entia and cognitive impairment in olderwomen: the Multiple Outcomes of Ralo-xifene Evaluation (MORE) randomizedtrial. Am J Psychiatry 2005; 162(4): 683-90.

87. Shumaker SA, Espeland MA, Rapp SR, etal. Conjugated equine estrogens and inci-dence of probable dementia and mildcognitive impairment in postmenopausalwomen: Women’s Health Initiative Mem-ory Study. JAMA 2004; 291(24):2947-58.

88. Rapp SR, Espeland MA, Shumaker SA, etal. Effect of estrogen plus progestin onglobal cognitive function in postmeno-pausal women: the Women’s HealthInitiative Memory Study: a randomizedcontrolled trial. JAMA 2003;289(20):2663-72.

89. Kang JH, Weuve J, Grodstein F. Post-menopausal hormone therapy and risk ofcognitive decline in community-dwellingaging women. Neurology 2004;63(1):101-7.

90. Jick H, Zornberg GL, Jick SS, et al. Statinsand the risk of dementia. Lancet 2000;356(9242):1627-31.

91. Wolozin B, Kellman W, Rousseau P, et al.Decreased prevalence of Alzheimer dis-ease associated with 3-hydroxy-3-methyg-lutaryl coenzyme A reductase inhibitors.Arch Neurol 2000; 57(10):1439-43.

92. Sparks DL, Sabbagh MN, Connor PJ, et al.Atorvastatin for the treatment of mild tomoderate Alzheimer disease: preliminaryresults. Arch Neurol 2005; 62(5):753-7.

93. Schenk D. Amyloid-beta immunotherapyfor Alzheimer’s disease: the end of thebeginning. Nat Rev Neurosci 2002;3(10):824-8.

94. Schenk D, Barbour R, Dunn W, et al.Immunization with amyloid-beta attenu-ates Alzheimer-disease-like pathology inthe PDAPP mouse [see comments].Nature 1999; 400(6740):173-7.


Documents

Treating Mild Cognitive Impairment