Mild cognitive impairment: Prognosis and treatment

INTRODUCTION — Mild cognitive impairment (MCI) is an intermediate clinical state between normal cognition and dementia. While specific subtle changes in cognition are frequently observed in normal aging, there is increasing evidence that some forms of cognitive impairment are recognizable as an early manifestation of a neurodegenerative condition that will ultimately lead to dementia [1].

The utility of this paradigm centers around the recognition that there is a continuum of cognition between normal aging and dementia, and dementia represents a rather advanced state of cognitive impairment.

This topic will review the prognosis and treatment of MCI. The epidemiology, pathology, and clinical assessment of MCI and topics related to dementia, including diagnosis, treatment, risk factors, and prevention of dementia, are discussed separately. (See "Mild cognitive impairment: Epidemiology, pathology, and clinical assessment" and "Treatment of Alzheimer disease" and "Risk factors for cognitive decline and dementia" and "Prevention of dementia" and "Clinical features and diagnosis of Alzheimer disease".)

PROGRESSION TO DEMENTIA — Older adults with MCI are at increased risk for progressing to dementia. Across multiple studies, those with MCI are approximately three times more likely to develop dementia over the next two to five years compared with age-matched controls [2].

On the other hand, a certain percentage of patients with MCI improve, even to normal, over a one- to three-year follow-up time, especially if a treatable cause of the MCI is identified [3-8]. The proportion who do improve ranges from 14 to 56 percent across various studies [2]. This finding emphasizes the expected clinical heterogeneity of MCI. However, such patients remain at significantly increased risk for a future diagnosis of dementia compared with those who have never received a diagnosis of MCI [8-10]. In spite of short-term clinical instability of the diagnosis, the vast majority of persons once labeled with MCI will progress to dementia [9,10].

Part of the variability between studies is due to the different populations studied, for example, patients referred to memory clinics versus epidemiologic cohorts [11].

Alzheimer disease (AD) and other forms of dementia are associated with increased mortality. In one community health center-based cohort study, MCI was associated with a small increased mortality hazard (hazard ratio [HR] 1.2) [12]. Other studies have also found that the presence and severity of MCI are associated with reduced survival [13].

Rates of progression — When counseling patients age 65 years and older who are diagnosed with MCI that is suspected to be of neurodegenerative origin, we estimate the risk of progression to dementia to be 10 percent per year and 15 percent over two years.

The overall incidence rate of dementia in the general older adult population is estimated at 1 to 3 percent per year [5,14,15]. By contrast, reported annual rates of progression from MCI to dementia have ranged from 5 to 16 percent, with lower rates of progression observed in population-based studies and higher rates in clinical centers and some treatment trials [4-6,14-25]. Rates of conversion also vary according to the duration of follow-up; one analysis compared longer-duration (>5 years) with shorter-duration studies and found lower rates with longer durations of follow-up, suggesting that the risk of conversion decreases over time [26]. In this study, the cumulative incidence was 33 percent. Typically, patients convert within a period of two to three years; however, longer intervals of up to eight years have been reported [27].

Predictors — Given the heterogeneity of outcomes among individuals with MCI, many studies have investigated factors that might identify those with MCI who will develop dementia. Combinations of risk factors may be more predictive than individual risk factors [28]. A formula using multiple variables for predicting MCI in the general population has been described but has not been independently validated [29].

Sex and education level have not been consistently shown to be predictive of progression to dementia [25,30]. Individual risk factors that are more consistently associated with this risk are discussed in the following sections. Risk factors for dementia in the general population are discussed separately. (See "Risk factors for cognitive decline and dementia".)

Age — Aging is a primary predictor of progression of MCI to AD [4,22,31-35]. With every year of age increase, MCI is slightly more likely to convert to AD.

In persons younger than 50 years, MCI is relatively unlikely to represent a predementia condition [22], unless there is a family history of early-onset dementia associated with an autosomal dominant mutation.

Neuropsychological testing — A variety of observational studies suggest that neuropsychological testing is helpful in defining individuals with MCI who are at risk for dementia. Patients who meet MCI criteria according to neuropsychological test performance have a higher rate of conversion to dementia than those who have a clinical dementia rating (CDR) scale summary score of 0.5 (table 1), which is based on an office assessment of cognition and daily functioning [36].

More severely affected patients (with greater deficits and/or a broader range of deficits on neuropsychological testing) have been shown to be at greater risk for dementia than those who are less affected, according to prospective cohort studies of patients [6,37-43]. Deficits that include a primary or prominent amnestic component as well as those that lead to slight impairment in daily functioning are also felt to identify patients at higher risk of dementia [44].

A number of specific test measures, such as delayed verbal recall, visual recognition memory, and other cued memory tasks, as well as measures of instrumental activities of daily living, have also been found to have high predictive value for dementia in some studies [28,33,38,45-53]. However, these have not been independently and prospectively validated in a manner that allows application to individual cases.

Follow-up neuropsychological testing may provide more helpful information regarding prognosis; a decrement in performance compared with prior testing is more sensitive than comparison with group norms [54]. However, the sensitivity and specificity of certain patterns of decline have not been rigorously evaluated. Educational level is an important modulator of test performance and rate of decline [39,41,54-56]. Practice effects, which can lead to test score improvement in normal older adults even after an interval of one to two years, may decrease the sensitivity of neuropsychological test decline as a predictor of AD [57]. Practice effects can be seen in normal subjects, but they tend to be short lived [58].

Psychological features — Neuropsychiatric symptoms are prevalent in individuals with MCI. (See "Mild cognitive impairment: Epidemiology, pathology, and clinical assessment", section on 'Neuropsychiatric symptoms'.)

Whether the presence of such features is predictive of conversion to dementia is uncertain. Studies with different designs, follow-up periods, and inclusion of nonamnestic types of MCI (naMCI) have found increased [59-62], decreased [63,64], or no change in the risk [65] of dementia when MCI is associated with depression. Other studies have suggested that apathy, anxiety, and dysphoria may be harbingers of incipient MCI [66-68].

APOE epsilon 4 — At this time, testing of apolipoprotein E (APOE) genetic status is not recommended in MCI patients.

APOE epsilon 4 (ε4) genotype has been associated with the risk of MCI and AD in the general population but has had mixed association with conversion to AD dementia among individuals with MCI. (See "Genetics of Alzheimer disease", section on 'Apolipoprotein E' and "Mild cognitive impairment: Epidemiology, pathology, and clinical assessment", section on 'Epidemiology'.)

Some studies have found that APOE ε4 is a strong risk factor for conversion from MCI to AD [19,34,38,69-72], but others have found only a marginal or even no association [30,31,33,73-75]. A meta-analysis of 35 prospective cohort studies found a moderate association of APOE ε4 and progression to AD (OR = 2.3), but low sensitivity (0.53) and positive predictive value (0.57) [76].

Cerebrovascular risk factors — Studies of the impact of vascular risk factors on the conversion of MCI to dementia have had variable results [70,77-79].

Evidence of cerebrovascular disease on magnetic resonance imaging (MRI) has also been associated with a higher risk of conversion to dementia. (See 'MRI' below.)

CSF biomarkers — Because the pathologic processes of AD and other degenerative dementias are likely well underway before clinical symptoms manifest, biomarkers would seem to have potential utility in the early diagnosis of dementia. However, these tests do not have an established role in the evaluation of patients with MCI in the clinical setting [2].

A number of smaller studies have examined the use of cerebrospinal fluid (CSF) markers for predicting conversion from MCI to dementia [28,47,80,81]. The CSF biomarkers most often found to be predictive include:

●Increased levels of tau or tau protein phosphorylated at Thr 181 [21,34,82-90]

●Lower levels of amyloid beta 42 (Aß42) peptide, a low ratio of Aß42 to Aß40 levels, and a low ratio of Aß42 to tau levels [21,85-87,89,91-103]

A 2014 meta-analysis of 14 observational studies evaluating CSF Aß42 as a tool for predicting conversion from MCI to AD found significant variation in assay thresholds, which precluded summary estimates of sensitivity and specificity [104]. Sensitivity in individual studies ranged from 36 to 100 percent and specificity ranged from 29 to 91 percent. The authors concluded that a low CSF Aß42 is of marginal clinical utility. The absence of a clinical treatment imperative for MCI makes this relatively invasive test less appealing at this point in time. New CSF analytical platforms are being developed to standardize the measurements of key analytes for predicting conversion from MCI due to AD to dementia due to AD.

Plasma biomarkers — Ongoing investigations are focusing on biomarkers in the serum or plasma that correlate with AD and other neurodegenerative dementias. An analysis of 120 signaling proteins in blood plasma found 18 that could be used to classify healthy controls from patients with AD [105-107]. These were then studied in a series of 47 patients with MCI; these proteins identified 20 of 22 patients who developed clinical AD and none of the eight patients who developed non-AD dementia. Seven of 17 patients with stable MCI at follow-up had proteins suggestive of AD and continue to be followed for possible future progression to AD. The replicability of plasma markers for predicting progression has been a challenge; however, studies using mass spectrometry are particularly encouraging [108,109].

Neuroimaging — Pathologic studies have shown that the earliest and most severe neurofibrillary manifestations of AD are found in the medial temporal lobe. Neuroimaging studies have focused attention on these and other areas to define structural abnormalities that may predict conversion from MCI to AD, as these appear to be more closely associated with MCI than other findings, such as white matter abnormalities [110].

MRI — Regional atrophy and cerebrovascular disease can be identified on MRI in cognitively normal patients who later develop dementia [111,112]. The more abnormalities present, the more likely a person is to progress.

●Temporal lobe atrophy may be an early specific marker for preclinical AD; similar patterns of temporal lobe atrophy are seen in patients with amnestic MCI (aMCI) and AD, but are not seen in individuals with multiple-domain naMCI, whose scans (at least in one study) are similar to healthy controls [113-115]. While most studies use specialized techniques for volumetric measurements, visual assessment using a standardized rating scale seems to perform nearly as well [116-118]. Efforts are underway to develop automated structural MRI techniques that allow patterns of atrophy to be quantitatively assessed [119].

Both the degree and progression of medial temporal lobe atrophy on MRI are associated with conversion to dementia in patients with MCI as well as in people with normal baseline cognition [28,30,33,53,75,86,87,90,116-118,120-138]. These changes can be observed one to two years prior to cognitive decline and, at least in one study, are better predictors of future AD than CSF biomarkers [87]. Similarly, higher baseline apparent diffusion coefficient (ADC) values in the hippocampus on diffusion-weighted MRI (DWI) may predict conversion to dementia in patients with aMCI [139,140]. Older age, APOE ε4, and poorer baseline cognition predicted progressive hippocampal atrophy in one study of patients with MCI [141]. Other longitudinal follow-up studies have found that thinning of the temporal and parietal cortex as determined by MRI morphometric analysis may be useful in predicting conversion from MCI to AD [135,142,143]. At least one study has suggested that a finding of temporal lobe atrophy is prognostically useful even when the underlying process is judged to be non-Alzheimer related [144].

However, there is overlap in the degree of atrophy for patients who do and do not progress to dementia, and the same is true for ADC values. While these findings are not clearly predictive of progression to dementia for individual patients, nonetheless, of the available markers, these MRI findings appear to be most proximate to conversion to dementia [87,145-147]. We note these findings as a marker of higher risk when we counsel patients.

Similarly, while some investigations suggest that these MRI changes may be useful in combination with other risk factors in predicting risk of AD, such multivariate models require independent validation [87,127].

●An increased extent of subcortical white matter hyperintensities and/or cortical infarction on MRI has been associated with MCI in some studies [112,114,148-155]. In one study of 170 individuals with MCI, vascular subcortical hyperintensities predicted the risk of vascular or mixed dementia over a median follow-up of four years [156]. More longitudinal studies are needed to assess whether an MRI finding of white matter hyperintensities or cerebral infarction represents a risk factor for further cognitive decline or progression to dementia [137].

The combination of these abnormalities may be more predictive of future dementia than their individual presence. In a longitudinal study of the very old, MRI markers of hippocampal atrophy and white matter hyperintensities were both independent predictors of conversion to AD over a two-year period [157]. However, the number of subjects who progressed to AD with a single imaging abnormality was small.

Amyloid PET — Amyloid positron emission tomography (PET) tracers (florbetapir F-18; flutemetamol F-18; florbetaben F-18; Pittsburgh Compound B C11, PiB) that measure amyloid lesion burden in the brain have been developed as tools to aid in the diagnosis of AD in vivo, aid in early diagnosis and prognosis, speed development of anti-amyloid medications, and differentiate AD from other causes of dementia. (See "Clinical features and diagnosis of Alzheimer disease", section on 'Neuroimaging'.)

While many F-18 compounds have been approved by the US Food and Drug Administration (FDA) for qualitative assessment of beta amyloid plaque density, they are not intended for use as a diagnostic agent in patients with dementia and are not universally available or reimbursed by all insurers. While a negative amyloid scan makes it extremely unlikely that a symptomatic person with dementia has AD, there is ongoing uncertainty about the predictive value of a positive scan for individuals with both normal cognition and MCI.

The prevalence of cerebral amyloid pathology on PET in adults with normal cognition as well as MCI varies by age and APOE status. In a meta-analysis of patient-level data from 55 studies of amyloid PET in 8694 adults, the prevalence of a positive amyloid PET scan rose progressively by decade among patients with normal cognition: 10 percent at 50 years of age, 16 percent at 60 years, 23 percent at 70 years, 33 percent at 80 years, and 44 percent at 90 years [158]. Among individuals with MCI, the prevalence of a positive scan was also higher at each decade, ranging from 27 percent at 50 years of age to 71 percent at 90 years. APOE ε4 carriers were two to three times more likely to have a positive scan than noncarriers.

Longitudinal follow-up studies of patients with MCI have shown that those with amyloid tracer retention have a higher rate of progression to dementia than those who do not [138,159-165], with an estimated sensitivity for conversion to AD of 83 to 100 percent but lower specificity (46 to 88 percent) [166]. Higher C11-PiB retention appeared in one study to correlate with a more rapid rate of progression [160]. In one cohort study, the combination of a positive flutemetamol F-18 labeled PET, low hippocampal volumes on MRI, and cognitive status consistent with aMCI identified a high probability of progression to dementia in 36 months [167].

More studies with longer follow-up times, more patients, and pathologic correlation are needed to define the prognostic value of amyloid PET imaging in patients with MCI. In all likelihood, amyloid PET will prove to be most useful in combination with other biomarkers. In a study that evaluated the relative value of the various imaging modalities in predicting progression from MCI to dementia due to AD, the presence of amyloid coupled with evidence of neurodegeneration (hippocampal atrophy on MRI or FDG-PET hypometabolism) showed the highest likelihood of progression [168].

FDG-PET — Identification of regional patterns of cortical hypometabolism using 18-F fluorodeoxyglucose PET (FDG-PET) may be useful for predicting conversion from MCI to AD, especially in the presence of the APOE ε4 allele [169-175]. In one cohort, abnormal results on PET combined with impaired episodic memory identified a group 11.7 times more likely to convert to AD compared with individuals who tested normally on both measures [100]. However, PET studies alone only marginally predicted cognitive decline in this patient group. A systematic review of 14 studies examining the value of FDG-PET in predicting the conversion of MCI to AD found that specificity was highly variable across studies, and there was no defined threshold for determination of test positivity [176]. This emphasizes the need to consider any diagnostic testing and prognostication in MCI in the context of the clinical encounter [177]. The limited availability of PET in most medical centers is a current obstacle to widespread use.

Tau PET imaging — Tau PET imaging has become available and is being investigated, and the tracer, flortaucipir F-18, was approved by the FDA in 2020. Tau PET may offer an important diagnostic tool for predicting cognitive decline. However, more data are needed to document the clinical utility of tau PET imaging in predicting progression from MCI to dementia [178-180].

Other risk factors

●Slow gait – Combined with cognitive complaints, gait slowing may identify individuals at higher risk of MCI and future dementia [181-183]. Some have proposed calling this state the motoric cognitive risk (MCR) syndrome, defined as the presence of cognitive complaints and slow gait in older individuals without dementia or mobility disability [181].

In a study that pooled data from four prospective cohort studies totalling 4812 individuals without baseline dementia, MCR was associated with a twofold increased risk of both incident cognitive impairment and dementia over a mean follow-up time ranging from five to nine years [182]. Slow gait was defined as a walking speed ≥1 standard deviation below age- and sex-specific means on a timed-walk test over a fixed distance. Forty percent of individuals with MCR also met criteria for MCI, and both MCR and MCI predicted incident cognitive impairment when examined in the same multivariate model (adjusted HRs 1.63 and 1.36, respectively). While MCR was associated with the AD dementia subtype in two of the four cohorts in this study [182], it has been associated with vascular dementia in other studies [181]. Across 22 cohorts worldwide, the pooled prevalence of MCR among older adults is 10 percent [182].

●Olfactory dysfunction – Olfactory dysfunction has also been identified as a predictor or prodrome of subsequent AD dementia, both in older individuals with normal cognition and in those with MCI [184-189]. One study used the Brief Smell Identification Test (B-SIT), a scratch-and-sniff card containing six food- and six nonfood-related smells, to assess olfactory function in a population-based cohort of older adults aged 70 to 89 years with normal baseline cognition (n = 1430) or MCI (n = 221) [184]. Over a median follow-up time of three years, patients with amnestic MCI (aMCI) in the lowest quartile of olfactory function (B-SIT scores 0 to 5) had fivefold higher risk of progression to AD dementia compared with those in the highest quartile (B-SIT scores 10 to 12). Among individuals with normal cognition at baseline, B-SIT scores in the lowest quartile conferred a twofold higher risk of aMCI and no additional risk for naMCI. The associations remained significant after adjusting for baseline cognitive scores and other risk factors, including APOE status.

TREATMENT — This section covers studies that specifically evaluated treatments for MCI. Treatment of dementia is discussed separately. (See "Treatment of Alzheimer disease" and "Management of the patient with dementia".)

For all patients with MCI, we treat reversible causes of cognitive impairment and modify vascular risk factors; we also inform patients of the potential benefits of exercise and cognitive interventions.

Treat reversible causes — A substantial minority of patients with MCI improve or remain stable over time. In addition, some patients have reversible causes of cognitive impairment that may be responsible for some or all of their deficits. Therefore, all patients with MCI should be evaluated and treated for reversible causes of cognitive impairment, including [2]:

●Medication side effects (see "Risk factors for cognitive decline and dementia", section on 'Medications')

●Sleep disturbances (see "Risk factors for cognitive decline and dementia", section on 'Obstructive sleep apnea' and "Risk factors for cognitive decline and dementia", section on 'Sleep disturbances')

●Depression (see "Diagnosis and management of late-life unipolar depression")

●Vitamin B12 deficiency or hypothyroidism

Vascular risk factor modification — Patients with MCI should be screened and treated for vascular risk factors, especially hypertension. Modification of vascular risk factors may ameliorate the risk of cognitive decline in patients with MCI [190]. Cerebrovascular disease is felt to play a role in Alzheimer disease (AD) as well as vascular dementia. (See "Epidemiology, pathology, and pathogenesis of Alzheimer disease", section on 'Acquired risk factors'.)

There have been no randomized controlled clinical trials to support this; however, treatment of hypertension has been shown to reduce the incidence of dementia in the general population, particularly in midlife [191]. There is less compelling evidence for the efficacy of antiplatelet therapy, statin therapy, smoking cessation, and diabetes management in the treatment or prevention of dementia; nonetheless, these are recommended for other health benefits. Specific treatments are discussed in detail separately. (See "Treatment of vascular cognitive impairment and dementia".)

Aducanumab — Aducanumab is a recombinant monoclonal antibody directed against amyloid beta. The US Food and Drug Administration (FDA) approved aducanumab for the treatment of mild AD using the accelerated approval pathway [192]. Postapproval trials are required to verify a clinical benefit.

The clinical trials that supported the approval of aducanumab included patients with MCI attributed to AD as well as those with mild AD dementia (Mini-Mental State Examination [MMSE] >24, clinical dementia rating [CDR] score of 0.5, and positive amyloid positron emission tomography [PET] scan). Most patients in these trials had MCI rather than dementia. Thus, we will consider such patients eligible for treatment with aducanumab.

Specific details regarding patient selection for treatment with aducanumab, along with dosing and monitoring guidelines, are discussed separately. (See "Treatment of Alzheimer disease", section on 'Aducanumab'.)

Nonpharmacologic interventions

Exercise — Exercise programs ranging in duration from 6 to 12 months have shown modest benefit in some measures of cognitive function in small trials. A meta-analysis of 11 randomized trials in a total 1497 participants with MCI found that aerobic exercise improved global cognitive ability by 1 point on the MMSE (95% CI 0.5-1.45) and was associated with small improvements on tests of immediate and delayed recall but not other domains [193]. Trials have generally measured short-term outcomes, and it is not known whether exercise programs affect rates of conversion from MCI to dementia. Nonetheless, because regular exercise has other health benefits and generally limited risk, many experts recommend exercise programs for patients with MCI [2].

Cognitive interventions — Various modalities of cognitive rehabilitation including memory training, the use of external memory cues, and organizational aids have been shown to improve function in healthy older adults [194-196]. (See "Prevention of dementia", section on 'Lifestyle and activity'.)

It is not clear if individuals with MCI may benefit from similar programs [197,198]. Although multiple small studies have demonstrated short-term improvements in various cognitive domains after cognitive training programs compared with a control condition, the benefits tend to be small, and studies that have measured long-term outcomes generally demonstrate waning effects over time.

Not recommended or unproven — No medications or dietary supplements have been shown to have symptomatic or preventive benefits in patients with MCI [2,199-201].

Acetylcholinesterase inhibitors — Acetylcholinesterase inhibitors are not routinely recommended for patients with MCI [190,202]. Clinical trials of donepezil, galantamine, and rivastigmine in the treatment of MCI have not provided support for the use of acetylcholinesterase inhibitors in preventing progression of MCI to dementia [20,202-208]. Systematic reviews and meta-analyses of available studies have concluded that in patients with MCI, there is no evidence that treatment with cholinesterase inhibitors affects progression to dementia or improves cognitive test scores, and that there is significant evidence of an increased of treatment-associated adverse events, particularly gastrointestinal [199,209].

However, if memory difficulties are particularly troublesome to an individual patient, a trial of donepezil or other acetylcholinesterase inhibitor for symptomatic benefit may be warranted [20]. Patients and families should be informed of the potential adverse effects. The risks and benefits of acetylcholinesterase inhibitors in the treatment of dementia are discussed separately. (See "Cholinesterase inhibitors in the treatment of dementia".)

Anti-inflammatory agents — The rationale for using anti-inflammatory agents in patients with MCI is based on the evidence of an inflammatory component to the pathogenesis of AD, as well as some epidemiologic studies suggesting that nonsteroidal anti-inflammatory drug (NSAID) use in the general population is associated with a lower risk of AD. However, there is no evidence for the benefit of NSAIDs in the treatment of MCI. (See "Prevention of dementia", section on 'NSAID therapy'.)

In a large, randomized, controlled trial of 1457 patients with MCI, treatment with rofecoxib was not associated with a lower conversion rate to dementia at four years, nor did it affect secondary endpoint measures of cognition [210].

Herbs and nutritional supplements — There are no studies that suggest a benefit for herbal therapies or nutritional supplements in patients with MCI.

●Ginkgo biloba – A clinical trial evaluating the effects of ginkgo biloba in preventing dementia in cognitively normal older adults and those with MCI found no benefit for treatment in preventing cognitive decline or the development of dementia among individuals with MCI [211,212]. In a clinical trial of 2854 patients aged 70 years and older with memory complaints (not necessarily meeting criteria for MCI), treatment with gingko biloba (120 mg twice per day) was not associated with a reduced incidence of dementia after five years of treatment [213].

●Phospholipids – Alterations in neuronal plasma membrane lipid content and structure have been observed in the aging brain and in AD dementia [214-216]. As a result, there has been interest in using precursors of the phospholipid membrane (eg, phosphatidylserine [PS] and phosphatidylcholine [PC]) as a way of repairing the plasma membrane and improving neuronal function. PC may also act as a precursor to acetylcholine metabolism.

Evidence of benefit in adults with cognitive impairment is currently lacking, however. There have been no trials of PS or PC in adults with strictly defined MCI. Several small trials have examined phospholipid supplementation in adults with AD, subjective memory problems, or age-related memory impairment. Results have been mixed, with some studies showing an apparent benefit [217-222] and others finding no effect [221].

●Multinutrient supplements – A randomized trial of a dietary supplement containing fish oil; uridine monophosphate; choline; vitamins B12, B6, C, E, and folic acid; phospholipids; and selenium showed no benefit on a composite cognitive performance outcome at 24 months in 311 patients with MCI due to AD [223]. The trial was limited by relatively high attrition and a smaller-than-expected change in the primary outcome over 24 months in both arms.

Others

●Transdermal nicotine was evaluated as a treatment for amnestic MCI (aMCI) in a placebo-controlled randomized study of 74 nonsmoking patients [224]. After six months of treatment, transdermal nicotine was associated with improvement in cognitive measures but not in ratings of clinician-rated global impression of change (CGIC).

●Intranasal insulin was compared with placebo in 104 adults with aMCI or mild to moderate dementia [225]. After four months, treatment for the combined group of aMCI and AD was associated with improved measures of memory, Alzheimer Disease Assessment Scale-Cognitive subscale (ADAS-cog) score, and functional abilities, suggesting that further study of this therapy is warranted [226].

●Growth hormone-releasing hormone was evaluated in a clinical trial in nondemented adults, including 66 with MCI; 20 weeks of therapy was associated with improved performance on cognitive tests compared with placebo [227]. Longer-term studies are needed to evaluate the safety and efficacy of this treatment approach.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Cognitive impairment and dementia".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Basics topic (see "Patient education: Mild cognitive impairment (The Basics)")

●Beyond the Basics topic (see "Patient education: Dementia (including Alzheimer disease) (Beyond the Basics)")

SUMMARY AND RECOMMENDATIONS

●Patients with mild cognitive impairment (MCI), particularly amnestic MCI (aMCI), are at risk for progression to dementia. A conservative estimate of the risk is 10 percent per year. Many, but not all, patients will progress to dementia due to Alzheimer disease (AD) or vascular cognitive impairment. (See 'Progression to dementia' above.)

●Neuropsychological test measures, cerebrospinal fluid (CSF) biomarkers, and neuroimaging studies are being evaluated as predictive tools for assessing patients' risk for conversion to dementia. These lack standardization, and there is no clinical imperative for general use. Apolipoprotein E epsilon 4 (APOE ε4) genotype testing is also not recommended as a way to predict likelihood of progression. (See 'Predictors' above.)

●All patients with MCI should be evaluated and treated for reversible causes of cognitive impairment, including medication side effects, sleep disturbances, depression, vitamin B12 deficiency, and hypothyroidism. (See 'Treat reversible causes' above.)

●Patients with MCI attributed to AD may be eligible for treatment with aducanumab. Patient selection is discussed separately. (See "Treatment of Alzheimer disease", section on 'Aducanumab'.)

●Patients with MCI and clinical or radiologic evidence of cerebrovascular pathology should be screened and treated for vascular risk factors. This approach has not been shown to alter conversion to dementia; however, there is convincing benefit for reduction of cerebrovascular and cardiovascular events. (See 'Vascular risk factor modification' above.)

●Regular exercise might have cognitive benefits in patients with MCI and is encouraged in older adults for general health benefits. (See 'Exercise' above.)

●Based on the available clinical trial data and lack of convincing benefit, we suggest not routinely treating MCI with cholinesterase inhibitors (Grade 2B). For an individual patient with troublesome memory difficulties, a trial of donepezil for symptomatic benefit may be warranted, although controlled study support for this recommendation is not available. Patients and families should be informed of the potential risks. (See 'Treatment' above.)

ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Eric M McDade, DO, who contributed to an earlier version of this topic review.