Neuropsychological Predictors of Long-Term (10 Years) Mild Cognitive Impairment Stability

Abstract

Background:

Although the diagnosis of mild cognitive impairment (MCI) corresponds to a condition likely to progress to dementia, essentially Alzheimer’s disease, longitudinal studies have shown that some patients may not convert to dementia and maintain the diagnosis of MCI even after many years.

Objectives:

To determine whether patients that maintain the diagnosis of MCI in the long term (10 years) are really stable or just declining slowly, and to identify clinical and neuropsychological characteristics associated with long-term stability.

Methods:

The Cognitive Complaints Cohort (CCC) was searched for MCI cases who maintained that diagnosis for at least 10 years. For each long-term-stable MCI patient, two MCI patients that converted to dementia during follow-up, matched for age and education, were selected from the same database. The baseline and last neuropsychological evaluations for long-term-stable MCI and converter MCI were compared. Baseline neuropsychological predictors of long-term stability were searched for.

Results:

Long-term-stable MCI (n = 22) and converter MCI (n = 44) patients did not differ in terms of gender distribution, education, age at first assessment and time between symptom onset and first evaluation. Time of follow-up was on average 11 years for long-term-stable MCI and 3 years for converter MCI. The baseline and follow-up neuropsychological tests were not significantly different in long-term-stable MCI patients, whereas a general decline was observed in converter MCI patients. Higher scores on one memory test, the Word Delayed Total Recall, and on the non-verbal abstraction test, Raven’s Progressive Matrices, at the baseline predicted long-term (10 years) clinical stability.

Conclusions:

Some patients with MCI remain clinically and neuropsychologically stable for a decade. Better performances at baseline in memory and non-verbal abstraction tests predict long-term stability.

Keywords

Amnestic follow-up long-term mild cognitive impairment neuropsychological tests prediction stability

INTRODUCTION

Although the term “mild cognitive decline” had been previously used in the literature [1], this diagnosis became widely disseminated after a set of studies by a group of investigators at the Mayo Clinic in 1999. Petersen et al. defined mild cognitive impairment (MCI) as a condition characterized by subjective memory complaints, objective memory deficit, normal general cognitive performance, and maintained activities of daily living [2]. From the beginning, the concept of MCI assumed a continuum between normality and Alzheimer’s disease (AD) and corresponded to a condition likely to progress to dementia, essentially AD [2]. The rate of conversion of MCI to dementia in a clinical setting was about 10% to 15% per year [3]. Nevertheless, it soon became apparent that a few patients, despite fulfilling MCI criteria, might not convert to dementia and would maintain the diagnosis of MCI even after many years [4]. As research progressed, several MCI subtypes were recognized, confirming that amnestic MCI, involving predominant impairment of the memory domain, was indeed the subtype associated with a higher conversion to AD [5, 6].

Another important observation was that high rates of conversion to dementia were observed in clinical studies, enrolling participants from clinical sources, such as memory clinics, whereas in epidemiologic studies, recruiting participants in the community, the conversion rates were lower [7]. A meta-analysis of 41 MCI studies found an adjusted annual progression rate to dementia from MCI of 9.6% in clinical settings and of 4.9% in community settings [8]. Thus, in epidemiological studies, a substantial proportion of MCI patients does not convert to dementia and may even revert to normal [9], which is not the case in a clinical setting.

An important aspect is that most studies about conversion or stability of MCI had relatively short follow-up times, rarely exceeding 5 years [9]. On the other hand, the annual rate of conversion of amnestic MCI patients to AD or dementia is not fixed along time [10]. Visser and colleagues showed, in their 10-year follow-up study in a clinical setting, that the annual conversion rate was highest during the first years of follow-up and decreased at longer follow-up intervals. The annual conversion rate in subjects with amnestic MCI was on average 10.8% during the first 2 years of follow-up, 4.5% during the next 3 years, and 2.5% during the last 5 years [10]. It is thus possible that patients with MCI who appear stable will still convert to dementia if the study follow-up is long enough. In a similar vein, patients who sustain the diagnosis of MCI for long periods may show some cognitive decline, albeit not severe enough to induce major changes in activities of daily living and justify the diagnosis of dementia. Only studies with a long follow-up and detailed neuropsychological testing may reliably answer the question of whether there is a proportion of patients with MCI in a clinical setting who really remain stable in the long term, or whether they are deemed to decline cognitively at a faster or slower pace. This question has relevant implications for the prognosis of patients suffering from MCI, and theoretical consequences for the pathophysiological meaning of MCI.

In the present study, we identified, within the Cognitive Complaints Cohort, a prospective study to investigate the conversion to dementia in non-demented subjects with cognitive complaints, 22 patients that maintained the diagnosis of MCI for at least 10 years. These stable patients were compared to a group of MCI patients that converted earlier to dementia, matched for age and education, selected from the same cohort. The objectives of this study were to ascertain whether patients that maintain the diagnosis of MCI in the long term (10 years) are really stable or just declining slowly, and to identify clinical and neuropsychological characteristics associated with long-term stability in MCI patients.

MATERIALS AND METHODS

Participants

Participants were selected from the Cognitive Complaints Cohort (CCC), a prospective study that consecutively recruits non-demented patients with cognitive complaints, referred for a comprehensive neuropsychological assessment at the participating institutions (Faculty of Medicine of Lisbon; Memoclínica, a private memory clinic in Lisbon; and Dementia Clinic, Hospitais da Universidade de Coimbra) [11]. The study was conducted in accordance with the Declaration of Helsinki, and was approved by the local ethics committee.

Inclusion criteria

The inclusion criteria for the diagnosis of MCI were adapted from Petersen et al. [2], corresponding to amnestic MCI:

presence of memory complaints;

abnormal memory function, documented by impairment in the Logical Memory A test score. Logical Memory is a subtest of the Bateria de Lisboa para Avaliação das Demências (BLAD) [12, 13] (see below). For the memory function to be considered abnormal, we set the cutoff score of the Logical Memory A test at 1 SD below the age and education norms. Busse et al. [14], have observed, in the cohort of the Leipzig Longitudinal Study of the Aged, that the “MCI modified, 1.0 SD” criteria had the highest relative predictive power for the development of dementia. Interestingly, the DSM-5 considers a rather broad range to establish the cognitive deficit in mild neurocognitive disorder, namely 1–2 standard deviations below age- and education-adjusted norms [15].

normal general cognitive function, determined by the Mini-Mental State Examination (MMSE) [16] (see below) within normal values for the Portuguese population [13].

no or a minimal impairment in activities of daily living, determined by the Instrumental Activities of Daily Living Scale (IADL) [17] (see below)—that is to say, no more than one item from the IADL scale was altered.

Exclusion criteria

Presence of neurological (stroke, brain tumor, significant head trauma, epilepsy) or psychiatric disorders that may induce cognitive deficits; patients with major depression according to DSM-IV-TR [18] or serious depressive symptoms, indicated by a score >10 in Geriatric Depression Scale short version (GDS₁₅) [19 –21] (see below);

Presence of systemic illness with cerebral impact (hypertension, metabolic, endocrine, toxic, and infectious diseases);

History of alcohol abuse or recurrent substance abuse or dependence;

Medication use with possible cognitive side effects;

Seriously reduced vision or other sensory deficits likely to interfere with assessment;

Presence of dementia according to DSM-IV-TR [18]

The diagnosis of MCI was made by an experienced neurologist (AdeM), after multidisciplinary consensus using all available information.

Long-term-stable MCI

Patients with long-term-stable MCI must fulfil the criteria for MCI during at least 10 years. The database of the CCC was thoroughly searched for these patients.

Converter MCI

For each long-term-stable MCI patient, the first two MCI patients that converted to dementia during follow-up, matched for age and education, were selected in the CCC database, and considered converter MCI.

Neuropsychological assessment

The comprehensive neuropsychological assessment at each CCC visit is carried out by the same team of trained neuropsychologists, following a standard protocol and including several instruments:

MMSE [16]. The Portuguese version of the test adapted from Guerreiro et al. [13] was used. Normal values for the Portuguese population are >27 for more than 11 years of education and >22 for 11 or less years of education [13].

BLAD [12, 13]: The BLAD is a comprehensive neuropsychological battery, including tests from the Wechsler Memory Scale (WMS) [22], that evaluates multiple cognitive domains and has been validated for the Portuguese population. For the present study, the following tests were considered:

Memory and Learning: Digit Span Forward from WMS, Word Delayed Total Recall (delayed recall of 5 non-related words), Logical Memory (immediate and delayed free recall) from WMS, Verbal Paired Associate Learning from WMS (difficult verbal pairs);

Attention and Executive Functions: Cancellation task (cross out 16 letters “A” from a set of 100 letters), Digit Span Backward from WMS, Clock Draw (free drawing of a clock), Verbal Semantic Fluency (supermarket food items);

Abstract Thought: Raven’s Colored Progressive Matrices (Ab series), Interpretation of Proverbs (3 proverbs);

Orientation: Temporal Orientation (7 questions concerning temporal orientation);

Calculation: Basic Written Calculation (4 additions, 2 subtractions, 3 multiplications);

Visuo-Constructional Abilities: Cube Copy (drawing of a cube with perspective);

GDS [19 –21]: For this study a short-form (15 items) of the self-report instrument, in the Portuguese version, adapted from Barreto et al. [19], was used. The presence of depression was defined as clinical history of depression or GDS₁₅ >5 [23].

IADL [17]. The Portuguese version, done in the context of the LADIS project, was used [24].

Conversion to dementia

The diagnosis of dementia and AD was established according to the DSM-IV-TR [18] criteria, in a consensus meeting with the neurologist and the neuropsychologists.

Data analysis

Neuropsychological test raw scores of the baseline and last assessments of all long-term-stable MCI and matched converter MCI individuals were registered; z scores were also calculated according to the age and education norms for the Portuguese population with the equation [z = (x-mean)/SD]. In the case of converter MCI, the visit in which the diagnosis of dementia was made was considered the last assessment.

Statistical analysis was performed using IBM SPSS Statistics 23 for Windows (2015 SPSS Inc., an IBM Company, Chicago, IL) package. Significance was set at p < 0.05.

Baseline demographic data were compared between groups using Student’s t test for numerical variables and χ² Pearson test for categorical data.

The baseline and follow-up neuropsychological test scores for long-term-stable MCI and converter MCI were compared with One-Way ANOVA, followed by the post hoc Tukey’s test. Binomial logistic regression was used to identify baseline neuropsychological predictors of long-term stability.

RESULTS

Twenty-two patients in the CCC met criteria for long-term-stable MCI, representing 3.4% of the 655 MCI patients having follow-up in the CCC. Forty-four converter MCI patients matched in terms of age and education at baseline were selected. As expected from the matching procedure, the two groups did not differ in terms of age and education. Gender distribution was also similar. They did not differ in terms of the presence of depression either. Time between symptom onset and baseline assessment was similar for both groups. Mean time to conversion to dementia was 3.5±2.1 (median 3.0, range 1–9) years in converter MCI patients, whereas the mean time of follow-up in the long-term-stable MCI group was 11.1±2.1 (median 11.0, range 10–14) years (Table 1).

Table 1

Demographic and clinical characterization

	Long-term-stable MCI (n = 22)	Converter MCI (n = 44)	p-value
Gender, male/female, n (% female)	8/14 (63.6%)	16/28 (63.6%)	1.000^a
Education, y, mean (SD)	9.5 (5.0)	9.1 (4.8)	0.780^b
Age of first symptoms, y, mean (SD)	64.1 (8.3)	64.5 (7.5)	0.856^b
Time between symptoms onset and first assessment, y, mean (SD)	1.6 (1.2)	1.9 (1.2)	0.326^b
Age at first evaluation, y, mean (SD)	65.5 (8.3)	66.4 (7.7)	0.678^b
Presence of depression^c, n(%)	7(31.8%)	14 (31.8%)	1.000^a
Duration of follow-up, y, mean (SD)	11.1 (2.1)	3.5 (2.1)	<0.001^b

^aχ² Pearson test. ^bIndependent samples Student’s t test. ^cPresence of depression was defined as clinical history of depression or GDS >5.

Long-term-stable MCI patients maintained performances at the 10-year follow-up in all neuropsychological tests (there were no significant differences between baseline and follow-up scores) (Table 2).

Table 2

Neuropsychological tests in long-term-stable and converter MCI patients

Cognitive domain Neuropsychological tests	Baseline		Follow-up
	Long-term stable MCI mean (SD) z scores (SD)	Converter MCI mean (SD) z scores (SD)	Long-term stable MCI mean (SD)	Converter MCI mean (SD)
Memory and Learning
Digit span forward	5.27(0.83)	4.77(0.61)	5.43(1.08)^#	4.24(1.22)
	0.29 (0.97)	–0.37 (0.75)
Word delayed total recall	9.80(1.51)	8.05(2.26) ^§	9.16(2.36)^#	5.86(2.85)^‡
	–0.70 (0.83)	–1.54 (1.36)
Logical memory (immediate free recall)	9.68(4.11)	6.07(3.97) ^§	9.73(4.36)^#	3.85(3.13)
	–0.73 (1.42)	–1.81 (1.32)
Logical memory (delayed free recall)	6.80(4.84)	4.12(4.29)	9.14(4.92)^#	1.48(1.96)
	–1.67 (1.48)	–2.49 (1.06)
Difficult Verbal Paired Associate Learning	3.67(2.78)	1.81(2.72)	4.35(3.20)^#	1.16(2.08)
	–1.21 (1.84)	–1.83 (1.51)
Attention and executive functions
Cancellation task-total	4.63(1.49)	3.86(1.49)	3.92(1.55)^#	2.74(1.24)^‡
	0.16 (1.15)	–0.25 (1.08)
Digit span backward	3.68(0.89)	3.50(0.73)	3.76(0.89)^#	2.34(1.29)^‡
	0.18 (1.15)	–0.03 (0.97)
Clock draw	2.70(0.47)	2.49(0.74)	2.65(0.59)^#	1.59(1.05)^‡
	0.295 (0.86)	–0.01 (1.56)
Verbal semantic fluency	16.23(3.82)	14.45(4.36)	16.82(4.25)^#	9.93(4.43)^‡
	0.00 (1.12)	–0.49 (1.38)
Abstract Thought
Raven’s Progressive Matrices	9.57(1.75)	7.31(2.33)^§	9.40(2.01)^#	5.33(2.48)^‡
	0.36 (1.03)	–0.49 (1.26)
Interpretation of Proverbs	7.00(1.54)	6.75(1.77)	6.90(1.73)^#	4.46(2.19)^‡
	0.79 (1.25)	0.52 (1.14)
Orientation
Temporal orientation	6.55(0.83)	6.02(1.12)	6.15(1.14)^#	3.00(2.49)^‡
	–0.26 (1.16)	–1.49 (2.01)
Calculation
Basic written calculation	13.41(1.00)	12.13(2.24)	11.74(2.85)^#	9.46(4.27)^‡
	0.38 (0.40)	–0.06 (1.24)
Visuo-constructional abilities
Cube copy	2.53(0.70)	1.97(1.06)	2.41(0.71)^#	1.39(1.23)
	0.45 (1.21)	–0.35 (1.79)

Raw test scores are shown. In the baseline neuropsychological results, z scores (SD) values are shown below raw scores. Long-term-stable MCI, MCI patients that have maintained this diagnosis for 10 years or more. Converter MCI, MCI patients that converted to dementia during follow-up. ^§Converter MCI patients had worse scores than long-term-stable MCI patients at baseline, One-Way ANOVA, Tukey post-hoc test. ^#Follow-up tests were not significantly different from baseline in long-term-stable MCI patients, One-Way ANOVA, Tukey post-hoc test. ^‡Follow-up tests worsened from baseline in converter MCI patients, One-Way ANOVA, Tukey post-hoc test.

On the contrary, converter MCI patients declined significantly between baseline and follow-up assessments in the domains of attention and executive functions, abstract thought, temporal orientation, calculation, and visuo-constructional abilities. They already scored low at the baseline and did not further decline significantly in the domain of memory and learning, except for the Word Delayed Total Recall test, in which a decay was noticed (Table 2).

Interestingly, at the baseline assessment, MCI patients that would remain stable for 10 years already performed consistently better than MCI patients deemed to convert to dementia in all the tests, the difference being statistically significant for Word Delayed Total Recall, Logical Memory – Immediate Free Recall and Raven’s Progressive Matrices (Table 2).

In order to identify neuropsychological predictors of long-term stability, the neuropsychological tests that were different at the baseline between long-term-stable MCI and converter MCI patients, namely Word Delayed Total Recall, Logical Memory – Immediate Free Recall and Raven’s Progressive Matrices, entered the binomial logistic regression model. Univariate logistic regression analysis revealed a significant association between high scores at baseline in all three tests and long-term MCI stability. In multivariate logistic regression analysis, only Word Delayed Total Recall and Raven’s Progressive Matrices retained significance in the prediction of long-term stability. For each additional score point in the Word Delayed Total Recall, the odds were 1.7, that is the probability of long-term stability increased by 70%. Each additional point in the total score of Raven’s Progressive Matrices increased the probability of long-term stability about 2 fold (Table 3).

Table 3

Neuropsychological predictors of long-term MCI stability

Neuropsychological tests	B	SE	Wald	Significance	Exp(B)
Word Delayed Total Recall	0.555	0.236	5.516	0.019	1.743
Logical memory (immediate free recall)	0.091	0.108	0.718	0.397	1.096
Raven’s Progressive Matrices	0.665	0.205	10.459	0.001	1.944

Multivariate binary logistic regression analysis.

DISCUSSION

The present work confirms that some patients with MCI maintain this diagnosis in the long term (10 years) and shows that they are able to maintain stable neuropsychological performance in all studied cognitive domains for an extended period.

Proportion of long-term-stable MCI patients

It should be noted that only a small proportion of patients with MCI (3.4% of MCI patients with follow-up in this cohort) maintain this diagnosis for a long period, presumably due to continual conversion to dementia [3] as well as persistent attrition of the cohort over the 10-year follow-up.

Neuropsychological stability

This study is the first, to our knowledge, to ascertain real neuropsychological stability in all the studied cognitive domains in stable MCI individuals over a decade of follow-up, that is, no statistically significant worsening was observed in any cognitive domain. In contrast, the converter MCI patients worsened in all neuropsychological tests, although this decline was not statistically significant in tests where they already had a low score at baseline, namely Digit Span Forward, Logical Memory-Free Immediate and Free Delayed Recall, Difficult Verbal Paired Associate Learning and Cube Copy tests, presumably due to a floor effect.

Demographic factors as predictors

Several factors influencing conversion of MCI to dementia have been recognized; namely older age [10] and lower level of education increase the risk of conversion ([25], but see [26]). These factors were a priori controlled for by the design of the study, matching long-term-stable with converter MCI patients by educational level and age at baseline evaluation. Time between symptom onset and baseline evaluation as well as distribution of gender were also similar between groups.

Neuropsychological tests as predictors

The neuropsychological performance at baseline, in particular, tests measuring memory, were shown to predict conversion of MCI to dementia in previous clinical studies with shorter follow-up durations. Arnáiz et al. [4], in a clinical sample of 303 MCI patients, followed up for 3 years on average, found that tests assessing learning and retention, specifically Wechsler Memory Scale-Revised delayed recall, were the best predictors of conversion to AD. Sarazin et al. [27] observed, in a cohort of 251 patients with MCI, followed for up to 3 years, that the most sensitive and specific test for diagnosis of prodromal AD was the Free and Cued Selective Recall Reminding Test (FCSRT). Silva et al. [28] found that four commonly used verbal memory tests were able to predict conversion to dementia in 272 non-demented patients reporting subjective cognitive complaints followed up for 3 years on average, and that the California Verbal Learning Test had the highest predictive value, which was not improved by adding other memory tests. In a similar vein, Gómez-Tortosa et al. [29] followed up for 48 months on average a cohort of 210 cases with amnestic MCI. They were divided into two groups according to their initial recognition memory discrimination index (DI) on the Hopkins Verbal Learning Test, and conversion to dementia occurred significantly later in cases with higher DI. A multivariate regression model revealed DI and delayed recall as the strongest predictors of dementia. A recent systematic review [30], covering data for a total of 2,365 participants with MCI at entry, followed over an average of 31 months, found that Paragraph Delayed Recall, Word-list Free Delayed Recall with and without oriented encoding were tests with excellent overall accuracy for predicting progression to Alzheimer’s type dementia. Another recent literature review [31] found that the majority of studies on the prediction of conversion from MCI to AD dementia report delayed recall as the most sensitive neuropsychological measure.

It is also clear that other tests, namely assessing attentional and executive capabilities, may also contribute as predictors of MCI conversion to dementia. For instance, in a work by Tabert et al. [32], with 148 MCI patients followed up for almost 4 years, the percent savings from immediate to delayed recall on the Selective Reminding Test and the Wechsler Adult Intelligence Scale–Revised Digit Symbol Test score were the strongest predictors of time to conversion. Also, in Fleisher et al. study [33], 539 participants with amnestic MCI were followed during 3 years and it was found that progression from MCI to AD dementia was best determined by combining distinct cognitive measures, namely Delayed Paragraph Recall Test, Delayed 10-Word List Recall, Symbol Digit Modalities Test and the ADAS-cog total score. Along the same lines, Li et al. [34] studied 139 patients with amnestic MCI enrolled in the Alzheimer’s Disease Neuroimaging Initiative and observed that, not only the baseline Memory composite (scores on the Logical Memory and Rey Auditory Verbal Learning tests), but also the executive function composite (scores on the Trail Making, Digit Symbol Substitution, and spontaneous Clock drawing tests) could predict progression to AD after 3 years.

In the present work, with 10 years of follow-up, although one verbal memory delayed measure at baseline was associated with long-term stability, we found that performance on the Raven’s Progressive Matrices at the baseline was the stronger neuropsychological predictor of long-term clinical stability. Remarkably, each additional point in the total score of the Raven’s Progressive Matrices increased the probability of long-term stability about 2 fold. This finding may assume clinical relevance, but must be replicated in other MCI cohorts.

Raven’s Progressive Matrices [35] are considered a non-verbal reasoning measure of fluid-type intelligence [36]. So-called fluid intelligence tests are most predictive of a general ability to do well, calling for novel problem solving with simple visual or other kind of materials, reflecting current ability for abstract thought and reasoning. Interestingly, Elias et al. [37], in a prospective study in healthy community participants in the Framingham cohort, found poor abstract reasoning to be a strong predictor of conversion to dementia in the long run. Fluid intelligence has been considered a good proxy for cognitive reserve [38]. Cognitive reserve is associated with lower risks for incident dementia [39]. As mentioned above, in our study, education, a variable known to influence prevalence rates of dementia [40], was controlled for. However, other factors thought to contribute to cognitive reserve, such as occupation, premorbid IQ, and mental activities, were not specifically analyzed [39].

Presumed etiology of long-term-stable MCI

The question of whether these patients with long-term-stable MCI have AD pathology, that is, suffer from prodromal AD from the start, or not, is rather intriguing. Nowadays, the AD biomarkers are commonly used to detect AD pathology and diagnose prodromal AD or MCI due to AD (e.g., [41 –44]). However, when these patients were recruited, at least 10 years ago, AD biomarkers were not routinely used in clinical practice.

It is known that there can be a very long interval, about 20 years, between first development of amyloid positivity and onset of dementia [45]. Clinical cohort studies suggest that there may be very subtle cognitive alterations that are detectable a decade or more before meeting criteria for MCI [46]. Thus, the AD pathophysiological process may course with a long preclinical stage [46]. Based on our findings, we could speculate that the disorder might also be quiescent for long periods at the MCI stage, at least in some patients. However, factors that might assume a neuroprotective role at this stage are still largely unknown [47].

Limitations and strengths

This cohort is constituted mainly by memory clinic patients, and the findings may not apply to clinical settings with different patient characteristics. The absence of data on Apolipoprotein E genotype is a limitation of the present study, since the ɛ4 allele is an important risk factor for AD [48].

The strengths of this study were that it was carried out in the context of a large cohort, in which the patients underwent comprehensive standardized neuropsychological assessments, and a long term 10-year follow-up was achieved.

Conclusions

We found that, in some MCI patients, real neuropsychological stability over a decade is possible and that long-term stability could be predicted on the basis of neuropsychological tests measuring memory and non-verbal abstract reasoning.

Footnotes

ACKNOWLEDGMENTS

We thank Memoclínica for the facilities provided.

Supported by a Grant from FCT – PTDC/EEI-SII/1937/2014.

Authors’ disclosures available online ().

References

Reisberg

, Ferris

, de Leon

, Crook

(1982) The Global Deterioration Scale for assessment of primary degenerative dementia. Am J Psychiatry 139, 1136–1139.

Petersen

, Smith

, Waring

, Ivnik

, Tangalos

, Kokmen

(1999) Mild cognitive impairment: Clinical characterization and outcome. Arch Neurol 56, 303–308.

Petersen

, Doody

, Kurz

, Mohs

, Morris

, Rabins

, Ritchie

, Rossor

, Thal

, Winblad

(2001) Current concepts in mild cognitive impairment. Arch Neurol 58, 1985–1992.

Arnáiz

, Almkvist

, Ivnik

, Tangalos

, Wahlund

, Winblad

, Petersen

(2004) Mild cognitive impairment: A cross-national comparison. J Neurol Neurosurg Psychiatry 75, 1275–1280.

Petersen

(2004) Mild cognitive impairment as a diagnostic entity. J Intern Med 256, 183–194.

Manly

, Tang

M-X

, Schupf

, Stern

, Vonsattel

J-PG

, Mayeux

(2008) Frequency and course of mild cognitive impairment in a multiethnic community. Ann Neurol 63, 494–506.

Petersen

, Roberts

, Knopman

, Boeve

, Geda

, Ivnik

, Smith

, Jack

(2009) Mild cognitive impairment: Ten years later. Arch Neurol 66, 1447–1455.

Mitchell

, Shiri-Feshki

(2009) Rate of progression of mild cognitive impairment to dementia–meta-analysis of 41 robust inception cohort studies. Acta Psychiatr Scand 119, 252–265.

Pandya

, Clem

, Silva

, Woon

(2016) Does mild cognitive impairment always lead to dementia? A review. J Neurol Sci 369, 57–62.

10.

Visser

, Kester

, Jolles

, Verhey

(2006) Ten-year risk of dementia in subjects with mild cognitive impairment. Neurology 67, 1201–1207.

11.

Maroco

, Silva

, Rodrigues

, Guerreiro

, Santana

, de Mendonça

(2011) Data mining methods in the prediction of Dementia: A real-data comparison of the accuracy, sensitivity and specificity of linear discriminant analysis, logistic regression, neural networks, support vector machines, classification trees and random forests. BMC Res Notes 4, 299.

12.

Garcia

(1984) Doença de Alzheimer, problemas do diagnóstico clínico. Tese de Doutoramento, Faculdade de Medicina de Lisboa.

13.

Guerreiro

(1998) Contributo da Neuropsicologia para o Estudo das Demências. Dissertação de Doutoramento em Ciências Biomédicas, Faculdade de Medicina de Lisboa.

14.

Busse

, Hensel

, Gühne

, Angermeyer

, Riedel-Heller

(2006) Mild cognitive impairment long-term course of four clinical subtypes. Neurology 67, 2176–2185.

15.

American Psychiatric Association (2013) Diagnostic and statistical manual of mental disorders (DSM-5®), American Psychiatric Association Publishing, Arlington.

16.

Folstein

, Folstein

, McHugh

(1975) “Mini-mental state”: A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 12, 189–198.

17.

Lawton

, Brody

(1969) Assessment of older people: Self-maintaining and instrumental activities of daily living. Gerontologist 9, 179–186.

18.

American Psychiatric Association (2000) DSM-IV-TR.APA, 4th edition, text revision, Washington, DC.

19.

Barreto

, Leuschner

, Santos

, Sobral

(2008) Geriatric Depression Scale (GDS). In Escalas e Testes na Demência, Mendonça

, Guerreiro

, eds. Grupo de Estudos de Envelhecimento Cerebral e Demência, 3rd edition, Lisboa, pp. 71–72.

20.

Yesavage

, Brink

, Rose

, Lum

, Huang

, Adey

, Leirer

(1983) Development and validation of a geriatric depression screening scale: A preliminary report. J Psychiatr Res 17, 37–49.

21.

Sheikh

, Yesavage

(1986) Geriatric Depression Scale (GDS): Recent evidence and development of a shorter version. In Clinical Gerontology: A Guide to Assessment and Intervention, Brink

, ed. The Haworth Press Inc, NY, pp. 165–173.

22.

Wechsler

(1969) Manuel de l’échelle clinique de mémoire. Centre de Psychologie Appliquée, Paris.

23.

Almeida

, Almeida

(1999) Short versions of the geriatric depression scale: A study of their validity for the diagnosis of a major depressive episode according to ICD-10 and DSM-IV. Int J Geriatr Psychiatry 14, 858–865.

24.

Van Der Flier

, van Straaten

, Barkhof

, Verdelho

, Madureira

, Pantoni

, Inzitari

, Erkinjuntti

, Crisby

, Waldemar

, Schmidt

(2005) Small vessel disease and general cognitive function in nondisabled elderly. Stroke 36, 2116–2120.

25.

Artero

, Ancelin

, Portet

, Dupuy

, Berr

, Dartigues

, Tzourio

, Rouaud

, Poncet

, Pasquier

, Auriacombe

(2008) Risk profiles for mild cognitive impairment and progression to dementia are gender specific. J Neurol Neurosurg Psychiatry 79, 979–984.

26.

Cooper

, Sommerlad

, Lyketsos

, Livingston

(2015) Modifiable predictors of dementia in mild cognitive impairment: A systematic review and meta-analysis. Am J Psychiatry 172, 323–334.

27.

Sarazin

, Berr

, De Rotrou

, Fabrigoule

, Pasquier

, Legrain

, Michel

, Puel

, Volteau

, Touchon

, Verny

(2007) Amnestic syndrome of the medial temporal type identifies prodromal AD A longitudinal study. Neurology 69, 1859–1867.

28.

Silva

, Guerreiro

, Maroco

, Santana

, Rodrigues

, Marques

, de Mendonça

(2012) Comparison of four verbal memory tests for the diagnosis and predictive value of mild cognitive impairment. Dement Geriatr Cogn Disord Extra 2, 120–131.

29.

Gómez-Tortosa

, Mahillo-Fernández

, Guerrero

, Montoya

, Alonso

, Sainz

(2012) Outcome of mild cognitive impairment comparing early memory profiles. Am J Geriatr Psychiatry 20, 827–835.

30.

Belleville

, Fouquet

, Hudon

, Zomahoun

HTV

, Croteau

(2017) Neuropsychological measures that predict progression from mild cognitive impairment to Alzheimer’s type dementia in older adults: A systematic review and meta-analysis. Neuropsychol Rev 27, 328–353.

31.

Cerami

, Dubois

, Boccardi

, Monsch

, Demonet

, Cappa

, Geneva Task Force for the Roadmap of Alzheimer’s, Biomarkers (2017) Clinical validity of delayed recall tests as a gateway biomarker for Alzheimer’s disease in the context of a structured 5-phase development framework. Neurobiol Aging 52, 153–166.

32.

Tabert

, Manly

, Liu

, Pelton

, Rosenblum

, Jacobs

, Zamora

, Goodkind

, Bell

, Stern

, Devanand

(2006) Neuropsychological prediction of conversion to Alzheimer disease in patients with mild cognitive impairment. Arch Gen Psychiatry 63, 916–924.

33.

Fleisher

, Sowell

, Taylor

, Gamst

, Petersen

, Thal

(2007) Clinical predictors of progression to Alzheimer disease in amnestic mild cognitive impairment. Neurology 68, 1588–1595.

34.

, Okonkwo

, Albert

, Wang

(2013) Variation in variables that predict progression from MCI to AD dementia over duration of follow-up. Am J Alzheimers Dis (Columbia) 2, 12–28.

35.

Raven

, Raven

, Court

(1956) Coloured progressive matrices: Sets A, AB, B. Oxford Psychologist Press.

36.

Staff

, Murray

, Deary

, Whalley

(2004) What provides cerebral reserve? Brain 127, 1191–1199.

37.

Elias

, Beiser

, Wolf

, Rhoda

, White

, D’Agostino

(2000) The preclinical phase of Alzheimer’s disease. A 22-year prospective study of the Framingham cohort. Arch Neurol 57, 808–813.

38.

Siedlecki

, Stern

, Reuben

, Sacco

, Elkind

, Wright

(2009) Construct validity of cognitive reserve in a multiethnic cohort: The Northern Manhattan Study. J Int Neuropsychol Soc 15, 558–569.

39.

Stern

(2012) Cognitive reserve in ageing and Alzheimer’s disease. Lancet Neurol 11, 1006–1012.

40.

The Canadian Study of Health and Aging (1994) Risk factors for Alzheimer’s disease in Canada. Neurology 44, 2073–2080.

41.

Albert

, DeKosky

, Dickson

, Dubois

, Feldman

, Fox

, Gamst

, Holtzman

, Jagust

, Petersen

, Snyder

(2011) The diagnosis of mild cognitive impairment due to Alzheimer’s disease: Recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement 7, 270–279.

42.

Dubois

, Feldman

, Jacova

, Hampel

, Molinuevo

, Blennow

, DeKosky

, Gauthier

, Selkoe

, Bateman

, Cappa

(2014) Advancing research diagnostic criteria for Alzheimer’s disease: The IWG-2 criteria. Lancet Neurol 13, 614–629.

43.

López

, Turrero

, Cuesta

, López-Sanz

, Bruña

, Marcos

, Gil

, Yus

, Barabash

, Cabranes

, Maestú

(2016) Searching for primary predictors of conversion from mild cognitive impairment to Alzheimer’s disease: A multivariate follow-up study. J Alzheimers Dis 52, 133–143.

44.

Dukart

, Sambataro

, Bertolino

(2016) Accurate prediction of conversion to Alzheimer’s disease using imaging, genetic, and neuropsychological biomarkers. J Alzheimers Dis 49, 1143–1159.

45.

Jansen

, Ossenkoppele

, Knol

, Tijms

, Scheltens

, Verhey

FRJ

, Visser

, the Amyloid Biomarker Study Group (2015) Prevalence of cerebral amyloid pathology in persons without dementia a meta-analysis. JAMA 313, 1924–1938.

46.

Sperling

, Aisen

, Beckett

, Bennett

, Craft

, Fagan

, Iwatsubo

, Jack

, Kaye

, Montine

, Park

(2011) Toward defining the preclinical stages of Alzheimer’s disease: Recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement 7, 280–292.

47.

Bos

, Vos

, Frölich

, Kornhuber

, Wiltfang

, Maier

, Peters

, Rüther

, Engelborghs

, Niemantsverdriet

, De Roeck

, Tsolaki

, Freund-Levi

, Johannsen

, Vandenberghe

, Lleó

, Alcolea

, Frisoni

, Galluzzi

, Nobili

, Morbelli

, Drzezga

, Didic

, van Berckel

, Salmon

, Bastin

, Dauby

, Santana

, Baldeiras

, de Mendonça

, Silva

, Wallin

, Nordlund

, Coloma

, Wientzek

, Alexander

, Novak

, Gordon

; Alzheimer’s Disease Neuroimaging Initiative, Wallin

ÅK

, Hampel

, Soininen

, Herukka

, Scheltens

, Verhey

, Visser

(2017) The frequency and influence of dementia risk factors in prodromal Alzheimer’s disease. Neurobiol Aging 56, 33–40.

48.

Strittmatter

, Saunders

, Schmechel

, Pericak-Vance

, Enghild

, Salvesen

, Roses

(1993) Apolipoprotein E: High-avidity binding to beta-amyloid and increased frequency of type 4 allele in late-onset familial Alzheimer disease. Proc Natl Acad Sci U S A 90, 1977–1981.