PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
 
J Geriatr Psychiatry Neurol. Author manuscript; available in PMC 2010 December 1.
Published in final edited form as:
PMCID: PMC2783244
NIHMSID: NIHMS133154

Cognitive Impairment in Older Adults Without Dementia: Clinical and Pathological Outcomes in a Community-Based Sample

Abstract

This study examines the clinical and neuropathologic characteristics of 37 participants in a community-based dementia series who had cognitive complaints at enrollment but did not meet dementia criteria. Participants had neuropsychological testing, were followed until death, and underwent autopsy. Twenty-four participants progressed to dementia, and their baseline characteristics were analyzed. Of the 24, 13 met criteria for neuropathologic Alzheimer disease (AD). The 13 participants who progressed to neuropathologic AD (mean intake age 78.5 ± 7.7, mean enrollment 6.4 ± 2.1 years) performed worse than 11 who progressed to neuropathologic non-AD dementias (mean intake age 79.0 ± 6.0, mean enrollment 6.0 ± 3.2 years) on baseline Wechsler Memory Scale (WMS) delayed logical memory (3.4 ± 2.9 vs 6.3 ± 3.9, P=.05) and delayed visual reproduction (1.4 ± 2.1 vs 3.1 ± 2.7, P=.02). These observations are consistent with the view that nondemented patients with underlying AD may be more likely to present with memory than nonmemory cognitive impairment.

Keywords: autopsy, neuropathology, dementia, mild cognitive impairment

Older people display a continuum of cognitive abilities that range from no impairment to the severe deficits that are seen in dementia. Within this continuum, some people have cognitive impairment but do not meet full dementia criteria. These patients are described using a variety of diagnostic terms including cognitive impairment not dementia (CIND)1 and mild cognitive impairment (MCI).2 They are of particular interest to clinicians because they have a higher likelihood of progressing to dementia and because they have higher rates of morbidity, and mortality than people with no cognitive impairment. For example, Bowen and colleagues found that study participants with isolated memory loss progressed to dementia at higher rates than cognitively normal controls.3 Moreover, others haven shown that people with CIND have higher rates of nursing home placement, medical comorbidity, and mortality.4, 5

Currently, MCI is the more widely used term to describe this population. The criteria for a diagnosis of MCI include a self or informant report of cognitive impairment, evidence of cognitive impairment on objective testing, and preserved basic activities of daily living.6 Petersen has proposed MCI subclassifications that include amnestic MCI, which describes individuals with memory-related cognitive impairment and a potential risk for Alzheimer's disease (AD), and non-amnestic MCI, which describes individuals with nonmemory cognitive impairment and a potential risk for non-AD dementias (e.g., dementia with Lewy bodies, frontotemporal dementia, and vascular dementia).6-8 At this time, only a handful of studies have explored the neuropathologic substrate of amnestic MCI, and even fewer studies have explored the neuropathologic basis of nonamnestic MCI.9-12

In this article, we describe neuropathologic outcomes of people who presented to a community-based dementia registry with cognitive complaints but did not initially meet criteria for dementia. We describe differences among participants who progressed to neuropathologic AD (NP-AD) and to neuropathologic non-AD (NP-nAD) dementias.

METHODS

Sample

Participant data was drawn from the University of Washington and Group Health Cooperative Alzheimer's Disease Patient Registry (ADPR). The ADPR served as a community-based registry of newly identified dementia cases from 1987 through 1996. Potential participants for the ADPR were identified through a dementia surveillance system in a local health maintenance organization, which included monitoring computed tomography logs, computerized hospital admission and discharge records, and clinic registration lists, as well as through referrals from primary care physicians.13 All the patients in the registry underwent an initial workup, which included a symptom history from the patient and an informant; a medical history; physical, neurological, and neuropsychological examinations; and laboratory testing. ADPR clinical investigators reviewed information from the initial workup and agreed on a consensus diagnosis for each patient according to the National Institute of Neurological and Communicative Disorders and Stroke–Alzheimer Disease and Related Disorders Association (NINCDS-ADRDA) and the Diagnostic and Statistical Manual of Mental Disorders, (Third Edition, Revised [DSM-III-R]) criteria.14, 15 Patients were followed annually with either an in-person examination or a telephone interview. Clinical and neuropsychological examinations were repeated if new signs or symptoms emerged, and diagnoses were changed only by consensus committee.

Of the 987 participants who enrolled in the ADPR, 243 did not initially meet NINCDS-ADRDA or DSM-III-R criteria for dementia (Figure 1). Although the ADPR protocol was primarily designed to identify and follow participants with AD, all who presented for enrollment were considered of interest to the registry and invited to continue, even if they did not have a dementia diagnosis. In all, 243 participants enrolled with a diagnosis of “no dementia,” of which 92 later withdrew from the study, and 151 participants completed follow-up. Of these 151 participants, we identified 46 who had autopsy data available and who we confirmed by chart review to have presented with cognitive complaints. For this analysis, we excluded 2 of these 46 participants because they had a primary diagnosis of psychosis. Seven participants were also excluded because their diagnosis was “no dementia” at death and they did not undergo an in-person examination within 18 months of their death. Although these 7 participants underwent an annual telephone interview, the absence of an in-persona evaluation close to the time of death meant that during the intervening time, the individuals may have progressed to undetected dementia. The remaining 37 participants are included in this analysis.

Figure 1
Flow chart depicting sample selection and the groups of interest (i.e., progression to NP-AD, progression to neuropathologic NP-nAD, died without dementia). Boxes with bold borders indicate participants who were included in this study.

Cognitive and Functional Assessments

Neuropsychological tests included the Mini-Mental State Examination (MMSE)16; trails A and B time and errors17; Wechsler Adult Intelligence Scale-Revised (WAIS-R) age-scaled information, comprehension, block design rotation, and similarities subscores;18 Wechsler Memory Scale (WMS);19 Mattis Dementia Rating Scale20; and Boston Naming Test21.

Activities of daily living were measured using the Blessed Dementia Scale, which assesses everyday activities, personal care, and personality changes.22

Neuropathologic Evaluation

Neuropathologic evaluations were performed by neuropathologists from the University of Washington Department of Pathology and the Alzheimer's Disease Research Center. The neuropathologic techniques performed in this study have been used in multiple additional autopsy samples.23-25 Examinations focused on the cingulate gyrus; superior and middle frontal gyri; medial orbital cortex; superior, middle, and inferior temporal gyri; inferior parietal lobule; medial occipital cortex; hippocampus; amygdala; entorhinal cortex; parahippocampal gyrus; hypothalamus; mamillary bodies; thalamus; midbrain; pons; medulla; and cerebellum. Tissues were stained with hematoxylin-eosin, thioflavin S, and the modified Bielschowsky silver method. Results from this sample have been described previously.23,26 For each case, we determined Consortium to Establish a Registry for Alzheimer Disease (CERAD) neuritic plaque scores and Braak staging for neurofibrillary tangles.27,28 Participants with Braak stage IV and CERAD score B and above were considered to have neuropathologic AD.29

Cerebrovascular damage was characterized for age and location using clinical records and neuropathologic assessments. Recent infarcts, occurring less than 1 year before death, were presumed to be terminal events that occurred after clinical assessments, and they were therefore considered unlikely to contribute substantially to any clinical evaluations performed during the study. These acute and subacute infarcts were therefore excluded as noncontributory vascular lesions.30

We systemically evaluated participants for Lewy-related pathology (Lewy body inclusions and Lewy body neurites) using alpha-synuclein immunostaining (antibody LB509, dilution, 1:400; generous gift, JQ Trojanowski). In each case, we assessed the medulla (including the dorsal motor nucleus of the vagus nerve, raphe nuclei, and lateral tegmentum), substantia nigra, amygdala, transentorhinal cortex, cingulate gyrus, and either the superior or middle frontal gyrus. The anatomic distribution and regional severity of Lewy-related pathology was used to classify each positive case (i.e., as brainstem-, amygdala-, limbic-, or cortical-predominant).23,31

In cases with a neuropathologic diagnosis of hippocampal sclerosis at autopsy, we performed additional immunostaining for TAR DNA-binding protein 43 (TDP-43) as has been described previously.24

Data Analysis

We compared categorical variables using the Fisher's exact test and continuous variables using the 2-sample student's t-test. For neuropsychological tests that participants did not attempt or did not complete, we assigned scores that indicated poor performance; these scores were then included in comparison analysis using the Wilcoxon signed rank test. Specifically, participants who did not complete a test were assigned a value of −88, and participants who did not attempt a test were assigned a value of −99. However, because Trails A and B time and errors have higher positive scores for poorer performance, we assigned a value of 888 to non-completers of this test, and we assigned a value of 999 to non-attempters of the test.

We define statistical significance as 0.05, and all P values reflect 2-sided statistical tests. We analyzed all data using Stata 9.0, College Station, Texas.

RESULTS

Sample Demographics and Clinical Characteristics

About half of the participants were male (19 of 37, 51%), and the majority of the participants achieved an education level beyond high school (22 of 37, 60%). One participant was an Asian or Pacific Islander, and the remaining 36 were Caucasian. Most of the participants who lived more than 2 years after enrollment progressed to dementia (24 of 29, 83%). Of the 24 participants who progressed to dementia, 18 progressed to dementia within 5 years of cognitive-symptom onset.

Between participants who died without dementia (n=13) and participants who progressed to dementia (n=24), gender, education status, age at symptom onset, age at death, and enrollment MMSE scores were not significantly different. The time between enrollment and death was shorter for participants who died without dementia than for participants who progressed to dementia (3.0 ± 3.7 years versus 6.3 ± 2.6 years, P < 0.01). This difference in time between enrollment and death was a critical confounding factor in our study sample, because an earlier death limited the clinical expression of dementia. In addition, the sample was small and statistical adjustment for this issue was not possible. Therefore, we focused further statistical analysis only on participants who progressed to dementia.

The participants who progressed to dementia were divided into NP-AD (Braak stage IV and CERAD neuritic plaque score B or higher) and NP-nAD groups. Table 1 provides descriptive demographic statistics for all 37 participants divided into 3 groups: died without dementia, progressed to NP-AD and progressed to NP-nAD groups. Statistical analyses were performed only on the progressed to NP-AD and NP-nAD groups. Neuropathologic AD participants were more likely to be female (P=.01), but there were no other statistically significant differences between the NP-AD and the NP-nAD groups including age at symptom onset, years to dementia progression, and enrollment MMSE scores.

Table 1
Demographic and Clinical Data for All Participants

Enrollment neuropsychological testing scores for the NP-AD and NP-nAD groups are summarized in Table 2. The NP-AD group performed worse than the NP-nAD group on WMS delayed logical memory (3.4 ± 2.9 versus 6.3 ± 3.9, P = 0.05) and WMS immediate logical memory (6.5 ± 2.7 versus 8.8 ± 3.1, P = 0.06) though the latter did not quite reach statistical significance. The NP-AD group also performed worse than the NP-nAD group on WMS delayed visual reproduction (1.4 ± 2.1 versus 3.1 ± 2.7, P = 0.02) and WAIS-R information (10.2 ± 2.2 versus 12.3 ± 2.5, P = 0.05). In contrast, the NP-AD group performed better than the NP-nAD group on the enrollment Mattis Dementia Rating Scale initiation and perseveration subscore (34.2 ± 3.4 versus 31.3 ± 3.1, P = 0.04).

Table 2
Enrollment Neuropsychological Testing for Participants Who Progressed to Dementia

Neuropathologic Characteristics

Braak stage frequency for the 24 participants who progressed to dementia (including NP-AD and NP-nAD) was bimodal in distribution, with II (n=7) and V (n=8; Figure 2) as the most frequent Braak stages. Of the 13 participants who progressed to NP-AD 9 had one or more concomitant pathologies, the most common being Lewy related pathology (6 of 13, 46%; Table 3). Of the 11 participants who progressed to NP-nAD, 5 had limbic or neocortical Lewy related pathology (2 with a clinical diagnosis of Parkinson's disease and 1 with concomitant vascular disease); 3 had isolated vascular disease; 1 had progressive supranuclear palsy; 1 had a subdural hematoma, limbic and neocortical cell loss, and associated gliosis that did not fulfill criteria for a specific neuropathologic diagnosis (TDP-43 negative); and 1 had no neuropathologic lesions to explain clinical dementia (Table 3).

Figure 2
Distribution of Braak staging for all participants. The category “progressed to dementia” includes participants who progressed to Alzheimer disease and participants who progressed to non–Alzheimer's disease dementias.
Table 3
Clinical and Neuropathological Findings for All Participants

The most frequent Braak stages for the 13 participants who died without dementia were II (n=5) and III (n=6; Figure 2). Most participants who died without dementia had some gross or microscopic evidence of vascular disease (9 of 13, 69%; Table 3). Four of the participants who died without dementia also had Lewy related pathology (2 with a clinical diagnosis of Parkinson's disease), and 3 had hippocampal sclerosis with TDP-43 immunopositive pathology (Table 3).

DISCUSSION

In our study, we describe the clinical and neuropathologic characteristics of participants from a community-based dementia registry who had cognitive complaints but did not meet dementia criteria at enrollment. A unique feature of our sample is the inclusion of participants with cognitive impairments not limited solely to amnestic or nonamnestic domains.

Approximately two thirds of our study participants progressed to clinical dementia, and 13 of these participants had NP-AD whereas 11 had NP-nAD. Participants in our sample who progressed to NP-AD generally had worse memory scores on enrollment neuropsychological testing than participants who progressed to other dementia types. This observation is consistent with the view that nondemented patients with underlying NP-AD processes are more likely to present with memory impairment (e.g., amnestic MCI) than nonmemory cognitive impairment (e.g., nonamnestic MCI). Several autopsy studies have suggested a relationship between memory complaints or memory impairment in nondemented older adults and AD pathologic changes. A study using linear regression models, for example, found that measures of subjective memory complaints in people with and without AD were associated with AD pathologic changes measured by amyloid beta and tau immunohistochemistry.32 Likewise, a study of Japanese-American men found that memory complaints of participants without dementia were predictors for development of NP-AD.31 Another study examined the brains of 23 normal controls, 10 participants with amnestic MCI, and 10 patients with early AD; these researchers found that early-AD changes were more severe in amnestic MCI cases than in controls and that neurofibrillary tangles in the hippocampus and entorhinal cortex correlated with memory function.10 However, it is important to note that the question of cognitive complaints being a predicator of dementia is complex, and there is significant debate in the literature outside of autopsy studies.34

To our knowledge, there is only one study that followed MCI patients longitudinally until they progressed to dementia and underwent autopsy.9 This study enrolled patients with amnestic MCI and found that the majority of participants progressed to AD. The researchers did not find a difference in enrollment cognitive measures between participants who progressed to AD and participants who progressed to non-AD dementias. However, their inclusion of only amnestic MCI patients, who by definition are predominantly memory impaired, would have made it difficult to detect neuropsychological differences.

Several studies investigating neuropsychological performance in patients with MCI have indicated an association between poorer performance on verbal memory tasks and progression to clinical AD. For example, Tabert and colleagues found that poorer performance on the Selective Reminding Test, which assesses immediate and delayed verbal recall, strongly predicted progression from amnestic MCI to AD within three years of the task.35 Sarazin and colleagues found that the Free and Cued Selective Recall Reminding Test, which measures verbal memory and semantic cueing, was a sensitive and specific test for progression to AD in amnestic MCI patients.36 In our study, the NP-AD group performed significantly poorer than the NP-nAD group on the WMS logical memory delayed recall task, which measures verbal memory. We also found poorer performance in delayed WMS visual reproduction and WAIS-R information tasks for the NP-AD group. In contrast, participants who progressed to NP-nAD performed poorer on the Mattis Dementia Rating Scale initiation and perseveration test at enrollment than the NP-AD group; as this test does not measure memory, the poorer performance suggests that more severe impairment in nonamnestic domains might be associated with underlying non-AD pathologic changes.

Because participants in our study were evaluated between 1987 and 1995 when MCI and CIND were not yet widely used diagnoses, the criteria for these diagnoses were not consistently applied. However, the neuropathologic results of our participants who had cognitive complaints and died without dementia, and who therefore approximate MCI and CIND samples clinically, are similar to the neuropathologic results in studies examining MCI patients. For example, Petersen and colleagues observed that the modal Braak stages for a sample of amnestic MCI subjects were II and III, which were found to be intermediate degrees of pathology when compared to normal and AD patients.11 Bennett and colleagues found that most of the MCI patients from the Religious Orders Study, whom they did not assign amnestic or non-amnestic subclassifications, also had intermediate levels of neurofibrillary tangle pathology (Braak Stages of III and IV).12 Our study found similar neuropathologic results; participants who died without dementia generally had less advanced pathologic changes than participants who progressed to dementia, including lower Braak stages (II and III) and a smaller proportion of participants with Lewy related pathologic changes.

A limitation of this study is that sample selection could have been influenced by bias. The primary goal of the ADPR was to follow patients with incident dementia and not patients with milder impairment, so it is likely that enrollment, retention, and autopsy-selection procedures favored participants who were not representative of the general medical community and who were perhaps more likely to progress to dementia. The percentage of study participants who progressed to dementia was high; of the 29 participants who lived longer than 2 years, 24 progressed to dementia and 18 were diagnosed with dementia within 5 years of symptom onset. However, this rate may not be substantially higher than other studies, where approximately half of CIND participants progressed to dementia in 5 years,5 and amnestic MCI patients progressed to AD at a rate of 10% to 12% per year.8 In addition, unfortunately, there were only 13 autopsied study participants who never developed dementia, and they had a significantly shorter time from enrollment to death in comparison to those who did progress to dementia. Thus, given the limitations of sample size and limited clinical follow-up time, we did not pursue additional analysis of this very interesting group of participants. Finally, the small sample size limits additional statistical analyses, such as adjustments for age, gender, or education level. This study is best interpreted as a description of a unique sample that lends clues to the relationship between clinical and neuropathological characteristics in the symptom course of AD and other dementias.

In summary, we examined participants in a community-based dementia sample who presented with cognitive complaints but did not meet dementia criteria at enrollment. Although there are limitations to this study, it nonetheless is a unique clinical comparison and description of neuropathologic outcomes for participants who progressed to AD and non-AD dementias.

Acknowledgements

This research was presented in part as a poster at the 10th International Conference on Alzheimer's Disease and Related Disorders, Madrid, Spain, July 2006. The study was supported by research grants from NIH (AG006781, NS048595) and the Department of Veterans Affairs Mental Illness Research, Education, and Clinical Centers Special Fellowship in Advanced Psychiatry. We gratefully acknowledge G. Stennis Watson, PhD, for his assistance in interpreting neuropsychological data and Andrew David for his editorial assistance.

References

1. Ebly EM, Hogan DB, Parhad IM. Cognitive impairment in the nondemented elderly. Results from the Canadian Study of Health and Aging. Arch Neurol. 1995;52:612–619. [PubMed]
2. Flicker C, Ferris SH, Reisberg B. Mild cognitive impairment in the elderly: predictors of dementia. Neurology. 1991;41:1006–1009. [PubMed]
3. Bowen J, Teri L, Kukull W, et al. Progression to dementia in patients with isolated memory loss. Lancet. 1997;349:763–765. [PubMed]
4. Lyketsos CG, Toone L, Tschanz J, et al. Population-based study of medical comorbidity in early dementia and “cognitive impairment, no dementia (CIND)”: association with functional and cognitive impairment: The Cache County Study. Am J Geriatr Psychiatry. 2005;13:656–664. [PubMed]
5. Tuokko H, Frerichs R, Graham J, et al. Five-year follow-up of cognitive impairment with no dementia. Arch Neurol. 2003;60:577–582. [PubMed]
6. Winblad B, Palmer K, Kivipelto M, et al. Mild cognitive impairment--beyond controversies, towards a consensus: report of the International Working Group on Mild Cognitive Impairment. Journal of Internal Medicine. 2004;256:240–246. [PubMed]
7. Petersen RC. Mild cognitive impairment as a diagnostic entity. Journal of Internal Medicine. 2004;256:183–194. [PubMed]
8. Petersen RC, Doody R, Kurz A, et al. Current concepts in mild cognitive impairment. Archives of Neurology. 2001;58:1985–1992. [PubMed]
9. Jicha GA, Parisi JE, Dickson DW, et al. Neuropathologic outcome of mild cognitive impairment following progression to clinical dementia. Arch Neurol. 2006;63:674–681. [PubMed]
10. Markesbery WR, Schmitt FA, Kryscio RJ, et al. Neuropathologic substrate of mild cognitive impairment. Archives of Neurology. 2006;63:38–46. [PubMed]
11. Petersen RC, Parisi JE, Dickson DW, et al. Neuropathologic features of amnestic mild cognitive impairment. Arch Neurol. 2006;63:665–672. [PubMed]
12. Bennett DA, Schneider JA, Bienias JL, et al. Mild cognitive impairment is related to Alzheimer disease pathology and cerebral infarctions. Neurology. 2005;64:834–841. [PubMed]
13. Larson EB, Kukull WA, Teri L, et al. University of Washington Alzheimer's Disease Patient Registry (ADPR): 1987−1988. Aging-Clinical & Experimental Research. 1990;2:404–408. [PubMed]
14. American Psychiatric Association . Diagnostic and Statistical Manual of Mental Disorders. Revised Third Edition. American Psychiatric Association; Washington, DC: 1987.
15. McKhann G, Drachman D, Folstein M, et al. Clinical diagnosis of Alzheimer's disease: report of the NINCDS-ADRDA Work Group under the auspices of Department of Health and Human Services Task Force on Alzheimer's Disease. Neurology. 1984;34:939–944. [PubMed]
16. Folstein MF, Folstein SE, McHugh PR. Mini-Mental State: a practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res. 1974;12:189–198. [PubMed]
17. Reitan RM. Validity of the trail-making test as an indication of organic brain damage. Percept Mot Skill. 1958;8:271–276.
18. Wechsler D. Wechsler Adult Intelligence Scale-Revised Manual. Psychological Corp; New York, NY: 1981.
19. Wechsler D. A standardized memory scale for clinical use. J Psychol. 1945;19:87–95.
20. Coblentz JM, Mattis S, Zingesser LH, et al. Presenile dementia. Clinical aspects and evaluation of cerebrospinal fluid dynamics. Archives of Neurology. 1973;29:299–308. [PubMed]
21. Morris JC, Heyman A, Mohs RC, et al. The Consortium to Establish a Registry for Alzheimer's Disease (CERAD). Part I. Clinical and neuropsychological assessment of Alzheimer's disease. Neurology. 1989;39:1159–1165. [PubMed]
22. Lim A, Tsuang D, Kukull W, et al. Clinico-neuropathological correlation of Alzheimer's disease in a community-based case series. Journal of the American Geriatrics Society. 1999;47:564–569. [PubMed]
23. Braak H, Braak E. Neuropathological stageing of Alzheimer-related changes. Acta Neuropathologica. 1991;82:239–259. [PubMed]
24. Mirra SS, Heyman A, McKeel D, et al. The Consortium to Establish a Registry for Alzheimer's Disease (CERAD). Part II. Standardization of the neuropathologic assessment of Alzheimer's disease. Neurology. 1991;41:479–486. [PubMed]
25. Knopman DS, Parisi JE, Salviati A, et al. Neuropathology of cognitively normal elderly. J Neuropathol Exp Neurol. 2003;62:1087–1095. [PubMed]
26. Riekse RG, Leverenz JB, McCormick W, et al. Effect of vascular lesions on cognition in Alzheimer's disease: a community-based study. J Am Geriatr Soc. 2004;52:1442–1448. [PMC free article] [PubMed]
27. Leverenz JB, Hamilton R, Tsuang DW, et al. Empiric Refinement of the Pathologic Assessment of Lewy-Related Pathology in the Dementia Patient. Brain Pathol. 2008 [PMC free article] [PubMed]
28. McKeith IG, Dickson DW, Lowe J, et al. Diagnosis and management of dementia with Lewy bodies: third report of the DLB Consortium. Neurology. 2005;65:1863–1872. [PubMed]
29. Leverenz JB, Yu CE, Montine TJ, et al. A novel progranulin mutation associated with variable clinical presentation and tau, TDP43 and alpha-synuclein pathology. Brain. 2007;130:1360–1374. [PubMed]
30. Barnes LL, Schneider JA, Boyle PA, et al. Memory complaints are related to Alzheimer disease pathology in older persons. Neurology. 2006;67:1581–1585. [PMC free article] [PubMed]
31. Jorm AF, Masaki KH, Davis DG, et al. Memory complaints in nondemented men predict future pathologic diagnosis of Alzheimer disease. Neurology. 2004;63:1960–1961. [PubMed]
32. Tabert MH, Manly JJ, Liu X, et al. Neuropsychological prediction of conversion to Alzheimer disease in patients with mild cognitive impairment. Arch Gen Psychiatry. 2006;63:916–924. [PubMed]
33. Sarazin M, Berr C, De Rotrou J, et al. Amnestic syndrome of the medial temporal type identifies prodromal AD: a longitudinal study. Neurology. 2007;69:1859–1867. [PubMed]
34. Cooper DB, Lacritz LH, Weiner MF, et al. Category fluency in mild cognitive impairment: reduced effect of practice in test-retest conditions. Alzheimer Dis Assoc Disord. 2004;18:120–122. [PubMed]