pmc.ncbi.nlm.nih.gov

Role of the Neuropathology of Alzheimer Disease in Dementia in the Oldest-Old

. Author manuscript; available in PMC: 2011 Apr 5.

Published in final edited form as: Arch Neurol. 2008 Sep;65(9):1211–1217. doi: 10.1001/archneur.65.9.1211

Abstract

Background

Neuritic plaques (NPs) and neurofibrillary tangles (NFTs) in the brain, especially in the hippocampus, entorhinal cortex, and isocortex, are hallmark lesions of Alzheimer disease and dementia in the elderly. However, this association has not been extensively studied in the rapidly growing population of the very old.

Objective

To assess the relationship between estimates of cognitive function and NP and NFT pathologic conditions in 317 autopsied persons aged 60 to 107 years.

Design

We studied the relationship between severity of dementia and the density of these characteristic lesions of Alzheimer disease in young-old, middle-old, and oldest-old persons. The relationship of the severity of dementia as measured by the Clinical Dementia Rating scale to the density of NPs and NFTs was then assessed in each age group.

Participants

Three hundred seventeen brains of persons aged 60 years and older were selected to have either no remarkable neuropathological lesions or only NP and NFT lesions. Brains with any other neuropathological conditions, either alone or in addition to Alzheimer disease findings, were excluded. The study cohort was then stratified into the youngest quartile (aged 60–80 years), middle 2 quartiles (aged 81–89 years), and oldest quartile (aged 90–107 years).

Results

While the density of NPs and NFTs rose significantly by more than 10-fold as a function of the severity of dementia in the youngest-old group, significant increases in the densities of NPs and NFTs were absent in the brains of the oldest-old. This lack of difference in the densities of NPs and NFTs was due to reduced lesion densities in the brains of oldest-old persons with dementia rather than to increased density of these lesions in the brains of nondemented oldest-old persons.

Conclusions

These findings suggest that the neuropathological features of dementia in the oldest-old are not the same as those of cognitively impaired younger-old persons and compel a vigorous search for neuropathological indices of dementia in this most rapidly growing segment of the elderly population.

Neuritic plaques (NPs) and neurofibrillary tangles (NFTs), mainly in the neocortex, the entorhinal cortex, and the hippocampus, represent the hallmark neuropathological lesions of Alzheimer disease (AD). Most studies have found the densities of NPs and NFTs to be associated with the severity of cognitive deficits, although the relative contributions of NPs and NFTs to cognitive dysfunction continue to be debated.¹^–⁸ The different neuropathological diagnostic systems in common use, such as the Khachaturian criteria,⁹ the Consortium to Establish a Registry for Alzheimer Disease (CERAD) criteria,¹⁰ and the National Institute on Aging–Reagan Institute criteria,¹¹ use the density of these lesions and their association with dementia to arrive at a diagnosis of AD with varying degrees of certainty.

The pivotal studies that have laid the foundation for our understanding of the relationship between lesion density and cognitive impairment have been conducted most frequently in individuals dying in the seventh through ninth decades of life.¹²^,¹³ However, US Census Bureau data project that the number of Americans older than 85 years will increase more rapidly than any other group, quadrupling from 4.4 million in 2001 to 19.3 million by 2050.¹²^–¹⁴ Of these 19.3 million, more than 8 million are predicted to suffer from dementia,¹⁵^–¹⁸ whereby in addition to personal and caregiver suffering, the inflation-unadjusted cost of care in this population will rise to more than $33 billion.¹⁹ Furthermore, a recent study²⁰ has shown that even after controlling for physical disorders, 5-year mortality rate in persons aged 95 years and older is significantly higher in demented individuals than in cognitively intact persons (96.0% vs 73.0%, respectively). Dementia was a stronger predictor of mortality in this population than cardiovascular disease, cancer, or being male. These observations underscore the necessity for understanding the relationship between cognitive impairment and the neuropathologies of dementia in the oldest-old.

The number of studies of the neuropathological correlates of cognitive dysfunction in persons who are aged 85 years and older are few,²^,²¹^,²² and they often focus on cases that meet clinical or neuropathological criteria for AD.¹⁵^,²³^–²⁶ However, the validity of the neuropathological criteria for diagnosing AD in this age group can be questioned. The CERAD criteria, for example, treat all persons older than 75 years uniformly, suggesting an underlying assumption of uniformity of the interaction of clinical and pathological indices in persons older than 75. Studies of the relationships between direct measures of cognitive function and estimates of lesion density are rarer.²^,²⁷^–³⁰ In general, qualitative and quantitative studies have found the neuropathological correlates of cognitive impairment in AD to be more variable and often less profound in persons older than 85 or 90 years than in younger persons⁷^,²³^,²⁸^,³⁰; but note that Nelson et al² present opposing results.

Given the potential ambiguity of relying on clinical or neuropathological diagnoses of AD in the oldest-old, the current study assessed the relationship between estimates of cognitive function and NP and NFT pathologic conditions in 317 autopsied persons aged 60 to 107 years. These cases included persons who had either no significant neuropathological features or only NP and NFT neuropathological features. Persons with non-AD clinical conditions known to lead to cognitive impairment (eg, schizophrenia, major depression) and persons with non-AD neuropathologies as defined by CERAD¹⁰^,³¹ (eg, significant cerebrovascular lesions, Lewy bodies with or without NPs and NFTs) were excluded. The 317-member study cohort was then stratified by age at death into quartiles, and the association of NPs and NFTs with cognitive impairment was assessed in the youngest (youngest-old), middle 2 (middle-old), and oldest (oldest-old) age quartiles.

METHODS

SUBJECTS

All antemortem and postmortem procedures were approved by the Mount Sinai School of Medicine and Jewish Home and Hospital institutional review boards. Postmortem brains, donated by the next of kin of deceased residents of the Jewish Home Hospital and other facilities participating in studies of aging and early dementia, were received over a period of 25 years by the Mount Sinai School of Medicine Department of Psychiatry Brain Bank. Analyses were based on the brains of 317 consecutively autopsied donors who met the inclusion criteria described later. Research staff blind both to the hypotheses tested and neuropathological findings performed neuropsychological tests, reviewed detailed medical records, and, whenever possible, conducted in-depth interviews with caregivers to obtain information about antemortem medical, neurological, psychiatric, functional, and cognitive status. Donors were included in the study if they met CERAD¹⁰ neuropathological criteria for normal brain, definite AD, probable AD, or possible AD only, as described previously.¹^,⁵

DEMENTIA STAGING

The Clinical Dementia Rating (CDR) scale assesses cognitive and functional impairments associated with dementia and classifies subjects as nondemented (CDR score of 0), questionably demented (CDR score of 0.5), or with increasing levels of severity of dementia (CDR score of 1 to 5).³²^–³⁵ A previously described¹^,⁵ multistep approach that included independent ratings by research psychometricians (V.H., M.S.-B., D. Purohit, D. Perl, G.T.L., M.M., H.T.G.) and a consensus conference that included evaluation of all available neuropsychological test results and caregiver interviews was applied to the assignment of CDR on the basis of stable (nonagonal) cognitive and functional status during the last 6 months of life.

NEUROPATHOLOGICAL ASSESSMENT

The neuropathological assessment procedures used have been described extensively¹^,⁴^,⁵ and followed CERAD consensus recommendations.¹⁰^,³¹ Standardized representative blocks from the neocortex (superior and midfrontal gyrus, orbital cortex, superior temporal gyrus, inferior parietal lobule of the parietal cortex, and calcarine cortex), basal ganglia with basal fore-brain, amygdala, hippocampus (rostral and caudal levels with adjacent entorhinal, parahippocampal, and inferior temporal cortex), rostral cingulated gyrus, corpus striatum, thalamus, midbrain, pons, medulla, cerebellar vermis, and lateral cerebellar hemisphere were examined using hematoxylin-eosin, modified Bielschowsky, and modified thioflavin S stains. Persons with Lewy body formation in the substantia nigra or locus ceruleus underwent anti-ubiquitin or α-synuclein staining of representative cerebral cortical sections. Neuropathological assessments were conducted blindly (D. Purohit and D. Perl) to all clinical and psychometric data. Every person was evaluated for the extent of neuropathological lesions using the CERAD neuropathological battery¹⁰^,³¹ for the extent of NPs and NFTs using a 4-point scale (0 indicates none; 1, sparse; 3, moderate; and 5, frequent). In addition, quantitative data regarding the density of NPs that were collected in 5 cortical regions using previously published methods¹^,⁴^,⁵ were expressed as mean plaque density per square millimeter for that region: the midfrontal gyrus (Brodmann area 9), orbital frontal cortex (Brodmann area 45/47), superior temporal gyrus (Brodmann area 21/22), inferior parietal lobule (Brodmann area 39), and calcarine cortex (Brodmann area 17). The primary rating variables used in the analyses reported here were densities of NPs and NFTs in the entorhinal cortex, hippocampus, amygdala, and the neocortical regions. Density estimates of NPs and NFTs from the subcortical fields did not substantively contribute to the results described¹^,⁴^,⁵ and were not included in analyses.

STRATIFICATION OF THE STUDY COHORT INTO AGE GROUPS

Quartile age ranges were calculated for the entire study cohort (60–107 years of age). Subjects were assigned to 3 groups consisting of the youngest age quartile (young-old; mean age, 71.4 years; range, 60–80 years), the middle 2 age quartiles (middle-old; mean age, 84.9 years; range, 81–89 years), and the oldest age quartile (oldest-old; mean age, 94.4 years; range, 90–107 years) (Table 1). Quartiles were defined by first calculating the median of the ordered age range of the study cohort and then dividing each half by the median of that half. The 3 groups differed in their composition by sex. As expected from population trends, the proportion of women in the 2 older age groups was significantly greater (χ² > 6.2; P < .02 for all) than in the youngest age quartile.

Table 1.

Demographic Characteristics of the Study Cohort

CDR Score	No. of Subjects	Mean (SD) Age, y	Male/Female, No.	Mean (SD) PMI	No. of Subjects by Race
W	B	H	O
Youngest-Old Group (Age Range, 60–80 y)
0	35	69.4 (0.96)	18/17	19.0 (2.5)	25	6	3	1
0.5	9	70.8 (2.15)	5/4	8.1 (2.6)	8	1	0	0
1.0	7	74.3 (1.64)	2/5	8.8 (2.9)	4	1	2	0
2.0	6	76.3 (1.69)	4/2	18.9 (9.1)	3	1	2	0
3.0	17	76.1 (1.07)	7/10	11.3 (2.7)	13	2	1	1
4.0	6	67.3 (1.26)	4/2	16.7 (5.8)	3	2	0	1
5.0	15	69.1 (1.48)	12/3	7.8 (1.8)	13	1	1	0

Middle-Old Group (Age Range, 81–89 y)
0	11	84.7 (0.85)	1/10	8.4 (1.7)	11	0	0	0
0.5	9	85.7 (0.78)	3/6	12.8 (5.4)	8	1	0	0
1.0	12	84.7 (0.84)	6/6	8.5 (2.5)	10	2	0	0
2.0	12	83.9 (0.51)	3/9	12.1 (3.8)	9	1	1	1
3.0	23	85.1 (0.48)	4/19	9.8 (2.7)	19	4	0	0
4.0	18	84.0 (0.52)	8/10	11.2 (3.8)	15	2	1	0
5.0	21	86.0 (0.48)	4/17	3.9 (0.5)	19	1	1	0

Oldest-Old Group (Age Range, 90–107 y)
0	7	96.4 (1.36)	2/5	8.3 (2.3)	7	0	0	0
0.5	8	92.7 (0.59)	4/4	7.7 (2.6)	5	1	1	1
1.0	10	95.8 (1.49)	1/9	11.6 (5.0)	10	0	0	0
2.0	16	94.1 (1.07)	3/13	10.8 (2.9)	15	0	0	1
3.0	36	93.4 (0.60)	6/30	7.2 (1.3)	29	5	2	0
4.0	21	94.1 (0.67)	4/17	12.5 (4.9)	19	2	0	0
5.0	18	96.6 (0.78)	2/16	5.7 (1.0)	16	1	1	0
All ages and CDRs	317	84.4 (0.58)	103/214	10.5 (0.7)	261	34	16	6

STATISTICAL ANALYSES

The densities of NPs and NFTs in the different brain regions and the summary variables described earlier were analyzed by factorial analyses of variance using Statistica (version 6.1; Stat-Soft Inc, Tulsa, Oklahoma). Separate analyses of variance tests were run to test the associations of age quartile and CDR at death with NP and NFT density in each of the predefined brain regions described earlier. Tukey’s honestly significant difference tests were used for post hoc analyses. In each post hoc analysis, the density of NPs or NFTs in subjects with a CDR score of 0 in each age group was compared with subjects with CDR scores of 0.5, 1, 2, 3, 4, or 5.

RESULTS

The 3 age groups did not differ from each other with respect to distribution of APOE4 genotype (χ² < 1.62, Ps > .10). The principal results of this study are summarized in Figure 1 and Table 2. This figure shows the fold change in the aggregate density of NFTs (top panel) and NPs (bottom panel) in persons with severe cognitive deficits (CDR=5) relative to cognitively intact persons in the 3 age groups. The results of analyses based on the specific regions of the brain examined were similar to those shown for all of the examined regions in aggregate. Analyses of variance for mean NP density revealed significant effects of age groups (F_2,296=6.9; P =.002), CDR groups (F_6,296=19.9; P < .001), and age group by CDR interaction (F_12,296=1.76; P =.046). The density of NPs was significantly increased (Ps < .02) in subjects with CDR scores of 2, 3, 4, and 5 in the youngest-old age group and in subjects with CDR scores of 5 in the middle-old age group, but no CDR-associated significant differences in NP density were detected in the oldest-old group (Ps > .7). A similar pattern of significant associations and interactions was noted for the density of NFTs in these same groups: age groups (F_2,297=4.1; P =.02), CDR (F_6,296=27.6; P < .001), and age-by-group interaction (F_12,296=2.8; P =.002). Post hoc analyses revealed that while subjects in the young-old and middle-old age groups with CDR scores of 2, 3, 4, and 5 evidenced significantly greater (Ps < .02) densities of NFTs, the density of NFTs in none of the cognitively impaired groups in the oldest-old age group differed significantly (Ps > .68) from the cognitively unimpaired (CDR= 0) subjects. Similar patterns of NP and NFT density as a function of age grouping and CDR were revealed when the densities of NPs and NFTs were examined in individual brain regions (individual statistical analyses not shown). As expected, the occipital cortex showed less NP and NFT involvement. Significant differences from the CDR 0 group in the young-old cohort emerged only in severe dementia (CDR=4 or 5; Ps < .01), and NP density increases were significant at only CDR 5 in the intermediate age group (P=.04). Although age group by CDR sample sizes was too small for meaningful statistical analysis, nominally the results described earlier held true when NP and NFT density scores were analyzed for men and women separately. Despite the lack of statistically significant differences in NP and NFT density between cognitively intact and cognitively impaired oldest-old subjects, there were modest and generally significant correlations between CDR and NP and NFT densities in this group (average r for all brain regions studied=0.26;Ps < .046), although the correlations between NP densities in the entorhinal cortex and NFT densities in the entorhinal cortex and hippocampus did not reach statistically significant levels (rs < 0.14; Ps > .14). In contrast to these modest correlations in the oldest-old, the densities of NPs and NFTs in the youngest-old group were strongly correlated with CDR in all brain regions (average r=0.67; Ps < .001). The correlations between these variables for the subjects in the middle-old group were intermediate (average r=0.4, Ps < .02).

Fold change in NFT (A) and NP (B) density in cognitively intact (CDR=0) vs severely impaired (CDR=5.0) subjects in representative brain regions as a function of age grouping. CDR indicates clinical dementia rating; NFT, neurofibrillary tangle; and NP, neuritic plaque.

Table 2.

Summary of NP and NFT Densities in the 3 Age Groups as a Function of Dementia Severity

CDR Score	No. of Subjects	Mean (SD)
Age, y	Summary NP Density	Summary NFT Density
Youngest-Old Age Group
0	35	69.37 (5.66)	1.09 (2.53)	2.06 (2.51)
0.5	9	70.78 (6.46)	4.11 (8.81)	5.22 (9.01)
1.0	7	74.29 (4.35)	6.87 (12.14)	9.43 (11.59)
2.0	6	76.33 (4.13)	16.03 (14.89)	21.67 (8.02)
3.0	17	76.06 (4.39)	11.12 (10.56)	15.29 (9.94)
4.0	6	67.33 (3.08)	19.58 (14.67)	26.83 (4.49)
5.0	15	69.13 (5.74)	22.84 (11.79)	27.00 (6.44)

Middle-Old Age Group
0	11	84.73 (2.83)	2.17 (2.53)	3.82 (2.75)
0.5	9	85.67 (2.35)	6.00 (4.98)	6.67 (3.77)
1.0	12	84.75 (2.90)	7.79 (7.07)	8.83 (6.39)
2.0	12	83.92 (1.78)	9.26 (8.25)	15.92 (9.43)
3.0	23	85.09 (2.31)	7.24 (7.13)	16.00 (8.80)
4.0	18	84.00 (2.22)	12.50 (7.86)	17.89 (7.37)
5.0	21	86.00 (2.19)	13.79 (10.16)	18.19 (10.35)

Oldest-Old Age Group
0	7	96.43 (3.60)	2.40 (1.69)	9.57 (4.47)
0.5	8	92.75 (1.67)	2.81 (4.50)	6.75 (4.86)
1.0	10	95.80 (4.71)	7.91 (4.13)	9.00 (4.99)
2.0	16	94.06 (4.28)	7.64 (5.90)	12.81 (6.71)
3.0	37	93.35 (3.63)	7.33 (5.31)	14.00 (6.28)
4.0	21	94.10 (3.06)	9.93 (8.47)	16.43 (9.04)
5.0	18	96.56 (3.31)	10.32 (10.37)	17.06 (8.49)

The failure to observe significant differences in NP and NFT densities between cognitively intact and cognitively impaired oldest-old subjects could have arisen if cognitively intact oldest-old subjects had higher densities of NPs and NFTs than the youngest-old group. However, the density of NPs and NFTs in cognitively intact (CDR 0) subjects in the 3 age groups did not differ significantly from each other (Ps > .60; Table 2). This observation held true irrespective of whether analyses were based on aggregate summary estimates of NP and NFT densities of the brain regions examined, or whether separate analyses were performed for each assessed brain region. In contrast, NP and NFT density scores in demented (CDR=5) oldest-old persons were significantly lower (Ps < .02) than those of persons in the youngest-old group.

It can be argued that the lower densities of NPs and NFTs in the brains of oldest-old persons who are demented (CDR=5) may be due to the reduced life expectancy of the oldest-old. Therefore, NPs and NFTs may have fewer years to accumulate maximally in the oldest-old relative to young-old persons. To test this possibility, we reasoned that the influence of duration of illness would be less in persons who were only moderately demented. We compared the density of NPs and NFTs in young-old vs oldest-old persons who were cognitively intact with a composite group of persons whose CDR scores ranged from 0.5 (questionable dementia) to 2 (moderate dementia). These analyses (CDR F_1,136=17.9, P ≤ .001; Figure 2) showed that in contrast to the youngest-old (Ps < .002), densities of NPs and NFTs did not rise significantly in the oldest-old with questionable to moderate dementia.

Fold change in neurofibrillary tangle (NFT) (A) and neuritic plaque (NP) (B) density in cognitively intact (clinical dementia rating [CDR]=0) vs questionably impaired (CDR 0.5) subjects in representative brain regions as a function of age grouping.

COMMENT

The results of this study show that the nearly universally replicated rise in NP and NFT densities that is observed in persons who are 60 to 80 years of age and transition from normal cognition to dementia is absent in similarly demented nonagenarians and centenarians.

This observation is consistent with previous findings that have pointed to the greater incidence of dementia in the oldest-old in the absence of clear neuropathological sequelae²³^–²⁵^,³⁶^,³⁷ or divergence of regions of vulnerability and changes in the neuropathological phenotype of cognitively impaired nonagenarians and centenarians.⁷ Other studies have noted the presence of hallmark AD lesions in the brains of very old persons with no obvious dementia.²⁷^,³⁸ In contrast, a recent study² has suggested that the relationship between NFTs and cognitive impairment may be more profound in persons older than 90 years than in younger cohorts. Given the overall similarity of the procedures used in that study and those described here and by others,²¹ the reasons for the differences in results are unclear. One consequence of this changing or inconsistent neuropathological phenotype is to raise questions concerning the age-associated algorithms, formal or informal, that we apply to the diagnosis of AD. As mentioned previously, most neuropathological criteria for the diagnosis of AD incorporate an underlying assumption of an age-associated and dementia-independent rise in NP and NFT features. Therefore, the “burden of proof” for a diagnosis of AD rises with age such that greater and greater densities of NPs and/or NFTs are required for the diagnosis. This study raises the possibility of a need to reconsider or refine the application of a monotonic age-associated scale to the way that we think about the neuropathological features of AD and weigh the utility of considering the influence of age on NP and NFT density as an inverted U-shaped or non-monotonic function.

Alternatively, it may be heuristically advantageous to think of the association of different underlying neurobiological mechanisms with cognitive impairment and dementia in relatively young elderly persons vs the oldest-old. The generally weak association of NPs and NFTs with dementia in the oldest-old have led some to challenge the importance of these lesions to dementia in the oldest-old³⁰ and others to suggest that microvascular pathological correlates may be an important determinant of cognitive dysfunction in the oldest-old.³⁹ Abnormalities of synaptic structure and function represent another one of many possible neurobiological functions that may be better correlates of dementia in the oldest-old. Recent studies have extended the now classic⁴⁰ and often replicated observations⁴¹ of the strong association of indices of synaptic loss or dysfunction with cognitive impairment in AD to the oldest-old.²¹

It may be argued that dementia in the oldest-old is associated with other known pathological correlates of dementia such as large-vessel cerebrovascular disease and Lewy body disease.⁴²^,⁴³ Although these and other neuropathological correlates of dementia may well be associated with dementia in the oldest-old in the general population, such an association is an unlikely explanation for the current study. In this study we specifically excluded from the analyses the brains of donors that had lesions other than NPs and NFTs; however, the overall conclusions of the study were not changed when 110 persons with comorbid cerebrovascular disease (CERAD diagnostic criteria 7), 86 of whom (78.0%) were in the oldest-old age quartile, were included in the study cohort and the data reanalyzed (data not shown). Another factor of potential significance is the level of education of the participating subjects and their cognitive reserve.⁴⁴ Data on years of education were available on a relatively small subset of the study subjects (39.4%) owing to lack of detailed knowledge of distant histories by informants and lost historical educational records. However, there were no differences between the 3 age groups with respect to the proportion of subjects on whom data on years of education were available (χ²< 0.4, P > .55), nor were there any differences between the 3 age groups in their levels of education (mean,13.2 years, t tests< 0.6, Ps> .57). Although the incompleteness of the available data on level of education in the study cohort demands a high level of caution, the similarity of the groups with respect to education suggests that level of education and perhaps cognitive reserve⁴⁴ are unlikely strong determinants of the lack of association of NP and NFT densities with cognitive deficits in the oldest-old group studied here.

United States and world population trends show that by 2050 the population of persons living over the age of 85 years is estimated to rise to 19.3 million, and some 40.0% to 50.0% of these elderly will experience dementia.¹⁵^–¹⁸ The results of the current study suggest that the neuropathological correlates of dementia in the oldest-old are likely to be different than the NP and NFT lesions that we have come to associate most directly with dementia in the youngest-old or in the middle-old. This observation may have significant consequences for how we think about dementia in the oldest-old, how we consider the criteria for AD diagnosis, how we research the neurobiological substrates of dementia in the aged, and the therapeutic avenues that we use to treat dementia in the rapidly growing oldest-old subgroup of the elderly. One speculative yet intriguing possibility for studies of the neurobiological substrates of dementia suggested by these results is that the study of the neurobiological substrates of dementia in the oldest-old could unmask core physiological and molecular substrates of dementia that may be otherwise overshadowed by NP and NFT neuropathological features in younger persons. Irrespective of speculations, it is clear that our current knowledge of the substrates of dementia in the oldest-old is rudimentary at best and that much more research will be required to address the cognitive health needs of this rapidly growing segment of the population.

Acknowledgments

Funding/Support: This study was supported by grants AG02219 and M01RR00071 from the National Institutes of Health and by the Dextra Baldwin McGonagale Foundation.

Footnotes

Financial Disclosure: None reported.

Author Contributions: Drs Haroutunian and Lesser had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Haroutunian, Purohit, Perl, and Lesser. Acquisition of data: Haroutunian, Wysocki, Purohit, Perl, Maroukian, and Grossman. Analysis and interpretation of data: Haroutunian, Schnaider-Beeri, Schmeidler, Perl, Libow, Maroukian, and Grossman. Drafting of the manuscript: Haroutunian. Critical revision of the manuscript for important intellectual content: Haroutunian, Schnaider-Beeri, Schmeidler, Wysocki, Purohit, Perl, Libow, Lesser, Maroukian, and Grossman. Statistical analysis: Haroutunian and Schmeidler. Obtained funding: Haroutunian, Purohit, and Perl. Administrative, technical, and material support: Haroutunian, Schnaider-Beeri, Wysocki, Purohit, Perl, and Libow. Study supervision: Haroutunian, Perl, and Grossman.

References

1.Haroutunian V, Davies P, Vianna C, Buxbaum JD, Purohit DP. Tau protein abnormalities associated with the progression of Alzheimer disease type dementia. Neurobiol Aging. 2007;28(1):1–7. doi: 10.1016/j.neurobiolaging.2005.11.001. [DOI] [PubMed] [Google Scholar]
2.Nelson PT, Jicha GA, Schmitt FA, et al. Clinicopathologic correlations in a large Alzheimer disease center autopsy cohort: neuritic plaques and neurofibrillary tangles “do count” when staging disease severity. J Neuropathol Exp Neurol. 2007;66(12):1136–1146. doi: 10.1097/nen.0b013e31815c5efb. [DOI] [PMC free article] [PubMed] [Google Scholar]
3.Morris JC, Storandt M, Miller JP, et al. Mild cognitive impairment represents early-stage Alzheimer disease. Arch Neurol. 2001;58(3):397–405. doi: 10.1001/archneur.58.3.397. [DOI] [PubMed] [Google Scholar]
4.Haroutunian V, Purohit DP, Perl DP, et al. Neurofibrillary tangles in nondemented elderly subjects and mild Alzheimer disease. Arch Neurol. 1999;56 (6):713–718. doi: 10.1001/archneur.56.6.713. [DOI] [PubMed] [Google Scholar]
5.Haroutunian V, Perl DP, Purohit DP, et al. Regional distribution of neuritic plaques in the nondemented elderly and subjects with very mild Alzheimer disease. Arch Neurol. 1998;55(9):1185–1191. doi: 10.1001/archneur.55.9.1185. [DOI] [PubMed] [Google Scholar]
6.Braak H, Braak E. Frequency of stages of Alzheimer-related lesions in different age categories. Neurobiol Aging. 1997;18(4):351–357. doi: 10.1016/s0197-4580(97)00056-0. [DOI] [PubMed] [Google Scholar]
7.von Gunten A, Kovari E, Rivara CB, Bouras C, Hof PR, Giannakopoulos P. Stereologic analysis of hippocampal Alzheimer’s disease pathology in the oldest-old: evidence for sparing of the entorhinal cortex and CA1 field. Exp Neurol. 2005;193(1):198–206. doi: 10.1016/j.expneurol.2004.12.005. [DOI] [PubMed] [Google Scholar]
8.Bussière T, Gold G, Kovari E, et al. Stereologic analysis of neurofibrillary tangle formation in prefrontal cortex area 9 in aging and Alzheimer’s disease. Neuroscience. 2003;117(3):577–592. doi: 10.1016/s0306-4522(02)00942-9. [DOI] [PubMed] [Google Scholar]
9.Khachaturian ZS. Diagnosis of Alzheimer’s disease. Arch Neurol. 1985;42(11):1097–1105. doi: 10.1001/archneur.1985.04060100083029. [DOI] [PubMed] [Google Scholar]
10.Mirra SS, Heyman A, McKeel D, et al. The Consortium to Establish a Registry for Alzheimer Disease (CERAD), part II: standardization of the neuropathologic assessment of Alzheimer’s disease. Neurology. 1991;41(4):479–486. doi: 10.1212/wnl.41.4.479. [DOI] [PubMed] [Google Scholar]
11.National Institute on Aging and Reagan Institute Working Group on Diagnostic Criteria for the Neuropathological Assessment of Alzheimer’s Disease. Consensus recommendations for the postmortem diagnosis of Alzheimer’s disease. Neurobiol Aging. 1997;18(suppl 4):S1–S2. [PubMed] [Google Scholar]
12.Blessed G, Tomlinson BE, Roth M. The association between quantitative measures of dementia and of senile change in the cerebral grey matter of elderly subjects. Br J Psychiatry. 1968;114(512):797–811. doi: 10.1192/bjp.114.512.797. [DOI] [PubMed] [Google Scholar]
13.Tomlinson BE, Blessed G, Roth M. Observations on the brains of demented old people. J Neurol Sci. 1970;11(3):205–242. doi: 10.1016/0022-510x(70)90063-8. [DOI] [PubMed] [Google Scholar]
14.US Census Bureau, Population Division. [Accessed June 13, 2008];US Population projections: national population projects: II, detailed files. http://www.census.gov/population/www/projections/natdet-D1A.html.
15.Hebert LE, Scherr PA, Bienias JL, Bennett DA, Evans DA. Alzheimer disease in the US population: prevalence estimates using the 2000 census. Arch Neurol. 2003;60(8):1119–1122. doi: 10.1001/archneur.60.8.1119. [DOI] [PubMed] [Google Scholar]
16.von Strauss E, Viitanen M, De Ronchi D, Winblad B, Fratiglioni L. Aging and the occurrence of dementia: findings from a population-based cohort with a large sample of nonagenarians. Arch Neurol. 1999;56(5):587–592. doi: 10.1001/archneur.56.5.587. [DOI] [PubMed] [Google Scholar]
17.Hebert LE, Scherr PA, Beckett LA, et al. Age-specific incidence of Alzheimer’s disease in a community population. JAMA. 1995;273(17):1354–1359. [PubMed] [Google Scholar]
18.Miech RA, Breitner JC, Zandi PP, Khachaturian AS, Anthony JC, Mayer L. Incidence of AD may decline in the early 90s for men, later for women: the Cache County study. Neurology. 2002;58(2):209–218. doi: 10.1212/wnl.58.2.209. [DOI] [PubMed] [Google Scholar]
19.Langa KM, Chernew ME, Kabeto MU, et al. National estimates of the quantity and cost of informal caregiving for the elderly with dementia. J Gen Intern Med. 2001;16(11):770–778. doi: 10.1111/j.1525-1497.2001.10123.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
20.Börjesson-Hanson A, Gustafson D, Skoog I. Five-year mortality in relation to dementia and cognitive function in 95-year-olds. Neurology. 2007;69(22):2069–2075. doi: 10.1212/01.wnl.0000280464.59322.af. [DOI] [PubMed] [Google Scholar]
21.Head E, Corrada MM, Kahle-Wrobleski K, et al. Synaptic proteins, neuropathology, and cognitive status in the oldest-old. Neurobiol Aging. doi: 10.1016/j.neurobiolaging.2007.10.001. [published online ahead of print November 12, 2007] [DOI] [PMC free article] [PubMed] [Google Scholar]
22.Kawas CH, Corrada MM. Alzheimer’s and dementia in the oldest-old: a century of challenges. Curr Alzheimer Res. 2006;3(5):411–419. doi: 10.2174/156720506779025233. [DOI] [PMC free article] [PubMed] [Google Scholar]
23.Polvikoski T, Sulkava R, Myllykangas L, et al. Prevalence of Alzheimer’s disease in very elderly people: a prospective neuropathological study. Neurology. 2001;56(12):1690–1696. doi: 10.1212/wnl.56.12.1690. [DOI] [PubMed] [Google Scholar]
24.Silver MH, Newell K, Brady C, Hedley-White ET, Perls TT. Distinguishing between neurodegenerative disease and disease-free aging: correlating neuropsychological evaluations and neuropathological studies in centenarians. Psychosom Med. 2002;64(3):493–501. doi: 10.1097/00006842-200205000-00014. [DOI] [PubMed] [Google Scholar]
25.Silver M, Newell K, Hyman B, Growdon J, Hedley-Whyte ET, Perls T. Unraveling the mystery of cognitive changes in old age: correlation of neuropsychological evaluation with neuropathological findings in the extreme old. Int Psychogeriatr. 1998;10(1):25–41. doi: 10.1017/s1041610298005122. [DOI] [PubMed] [Google Scholar]
26.Ebly EM, Parhad IM, Hogan DB, Fung TS. Prevalence and types of dementia in the very old: results from the Canadian Study of Health and Aging. Neurology. 1994;44(9):1593–1600. doi: 10.1212/wnl.44.9.1593. [DOI] [PubMed] [Google Scholar]
27.Giannakopoulos P, Hof PR, Kovari E, Vallet PG, Herrmann FR, Bouras C. Distinct patterns of neuronal loss and Alzheimer’s disease lesion distribution in elderly individuals older than 90 years. J Neuropathol Exp Neurol. 1996;55(12):1210–1220. doi: 10.1097/00005072-199612000-00004. [DOI] [PubMed] [Google Scholar]
28.Giannakopoulos P, Hof PR, Vallet PG, Giannakopoulos AS, Charnay Y, Bouras C. Quantitative analysis of neuropathologic changes in the cerebral cortex of centenarians. Prog Neuropsychopharmacol Biol Psychiatry. 1995;19(4):577–592. doi: 10.1016/0278-5846(95)00103-3. [DOI] [PubMed] [Google Scholar]
29.Giannakopoulos P, Hof PR, Giannakopoulos AS, Herrmann FR, Michel JP, Bouras C. Regional distribution of neurofibrillary tangles and senile plaques in the cerebral cortex of very old patients. Arch Neurol. 1995;52(12):1150–1159. doi: 10.1001/archneur.1995.00540360028012. [DOI] [PubMed] [Google Scholar]
30.Prohovnik I, Perl DP, Davis KL, Libow L, Lesser G, Haroutunian V. Associations of neuropathology with severity of dementia in late-onset Alzheimer’s disease. Neurology. 2006;66(1):49–55. doi: 10.1212/01.wnl.0000191298.68045.50. [DOI] [PubMed] [Google Scholar]
31.Mirra SS, Vogel FS, Heyman A. Guide to the CERAD Protocol for the Neuropathological Assessment of Alzheimer’s Disease. Durham, NC: CERAD; 1987. [Google Scholar]
32.Fillenbaum GG, Peterson B, Morris JC Consortium to Establish a Registry for Alzheimer’s Disease. Estimating the validity of the Clinical Dementia Rating scale: the CERAD experience. Aging (Milano) 1996;8(6):379–385. doi: 10.1007/BF03339599. [DOI] [PubMed] [Google Scholar]
33.Morris JC, Ernesto C, Schafer K, et al. Clinical dementia rating training and reliability in multicenter studies: the Alzheimer’s Disease Cooperative Study experience. Neurology. 1997;48(6):1508–1510. doi: 10.1212/wnl.48.6.1508. [DOI] [PubMed] [Google Scholar]
34.Morris JC. The Clinical Dementia Rating (CDR): current version and scoring rules. Neurology. 1993;43(11):2412–2414. doi: 10.1212/wnl.43.11.2412-a. [DOI] [PubMed] [Google Scholar]
35.Dooneief G, Marder K, Tang MX, Stern Y. The Clinical Dementia Rating scale: community-based validation of “profound’ and “terminal’ stages. Neurology. 1996;46(6):1746–1749. doi: 10.1212/wnl.46.6.1746. [DOI] [PubMed] [Google Scholar]
36.Giannakopoulos P, Hof PR, Bouras C. Dementia lacking distinctive histopathology: clinicopathological evaluation of 32 cases. Acta Neuropathol. 1995;89 (4):346–355. doi: 10.1007/BF00309628. [DOI] [PubMed] [Google Scholar]
37.Crystal HA, Dickson D, Davies P, Masur D, Grober E, Lipton RB. The relative frequency of “dementia of unknown etiology” increases with age and is nearly 50% in nonagenarians. Arch Neurol. 2000;57(5):713–719. doi: 10.1001/archneur.57.5.713. [DOI] [PubMed] [Google Scholar]
38.Crystal H, Dickson D, Fuld P, et al. Clinicopathologic studies in dementia: non-demented subjects with pathologically confirmed Alzheimer’s disease. Neurology. 1988;38(11):1682–1687. doi: 10.1212/wnl.38.11.1682. [DOI] [PubMed] [Google Scholar]
39.Imhof A, Kovari E, Von Gunten A, et al. Morphological substrates of cognitive decline in nonagenarians and centenarians: a new paradigm? J Neurol Sci. 2007;257(1–2):72–79. doi: 10.1016/j.jns.2007.01.025. [DOI] [PubMed] [Google Scholar]
40.Terry RD, Masliah E, Salmon DP, et al. Physical basis of cognitive alterations in Alzheimer’s disease: synapse loss is the major correlate of cognitive impairment. Ann Neurol. 1991;30(4):572–580. doi: 10.1002/ana.410300410. [DOI] [PubMed] [Google Scholar]
41.Scheff SW, Price DA, Schmitt FA, Mufson EJ. Hippocampal synaptic loss in early Alzheimer’s disease and mild cognitive impairment. Neurobiol Aging. 2006;27(10):1372–1384. doi: 10.1016/j.neurobiolaging.2005.09.012. [DOI] [PubMed] [Google Scholar]
42.Giannakopoulos P, Gold G, Kovari E, et al. Assessing the cognitive impact of Alzheimer disease pathology and vascular burden in the aging brain: the Geneva experience. Acta Neuropathol. 2007;113(1):1–12. doi: 10.1007/s00401-006-0144-y. [DOI] [PubMed] [Google Scholar]
43.Snowden JS, Neary D, Mann DM. Frontotemporal dementia. Br J Psychiatry. 2002;180:140–143. doi: 10.1192/bjp.180.2.140. [DOI] [PubMed] [Google Scholar]
44.Stern Y. Cognitive reserve and Alzheimer disease. Alzheimer Dis Assoc Disord. 2006;20(3 suppl 2):S69–S74. doi: 10.1097/00002093-200607001-00010. [DOI] [PubMed] [Google Scholar]