Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
AJNR Am J Neuroradiol. Author manuscript; available in PMC 2010 August 13.
Published in final edited form as:
PMCID: PMC2890033

Effects of age on the glucose metabolic changes in mild cognitive impairment


Background and Purpose

Decreased glucose metabolism in the temporal and parietal lobes on [18F]fluorodeoxyglucose (FDG) PET is recognized as an early imaging marker for the Alzheimer’s disease (AD) pathology. Our objective was to investigate the effects of age on FDG PET findings in aMCI.


25 patients with aMCI at 55–86 years of age (median = 73), and 25 age and gender matched cognitively normal (CN) subjects underwent FDG PET. SPM5 was used to compare the FDG uptake in aMCI-old (>73 years) and aMCI-young (>73 years) patients to CN subjects. The findings in the aMCI-old patients were independently validated in a separate cohort of 10 aMCI and 13 CN subjects older than 73 years of age.


The pattern of decreased glucose metabolism and gray matter atrophy in the medial temporal, posterior cingulate, precuneus, lateral parietal and temporal lobes in aMCI-young subjects was consistent with the typical pattern observed in AD. The pattern of glucose metabolic changes in aMCI-old subjects was different, predominantly involving the frontal lobes and the left parietal lobe. Gray matter atrophy in aMCI-old subjects was less pronounced than the aMCI-young subjects involving the hippocampus and the basal forebrain in both hemispheres


Pathological heterogeneity may be underlying the absence of AD-like glucose metabolic changes in older compared to younger aMCI patients. This may be an important consideration for the clinical use of temporoparietal hypometabolism on FDG PET as a marker for early diagnosis of AD in aMCI.


Neuroimaging markers of Alzheimer’s disease (AD) pathology would allow new therapies to be developed more quickly, and increase the probability of success in clinical trials 1. An important clinical group for early diagnosis and treatment of AD are people with amnestic mild cognitive impairment (aMCI), who are at a higher risk of developing AD than their cognitively normal peers 2. Decreased glucose metabolism on [18F]fluorodeoxyglucose (FDG) PET is a sensitive marker for AD related pathological changes in the brain. The risk of developing AD is higher in patients with aMCI who have the typical pattern of AD related glucose metabolic changes on FDG PET than aMCI patients who do not have these changes 37.

Glucose metabolism on FDG PET is primarily decreased in the temporal and parietal cortex in AD, consistent with the regional pattern of neurofibrillary pathological involvement, neurodegeneration, and atrophy in AD 8. The pattern of decreased glucose metabolism however, appears to depend on the age of onset in AD 9, 10. Patients with early onset AD have the typical pattern of glucose metabolic changes in the temporal and parietal lobes, whereas these changes are either less pronounced or absent in AD patients older than 73–75 years of age 9, 10. Furthermore, the diagnostic accuracy of FDG PET was higher in AD patients at an average age of 58 compared to patients at an average age of 76 11.

Although FDG PET is being recognized as an important early diagnostic marker for AD pathology in aMCI, it is unknown whether age modifies the pattern of glucose metabolic changes in aMCI. Our objective was to investigate the effects of age on the pattern of FDG PET changes and to compare this to the pattern of cortical atrophy on MRI in aMCI.



We recruited 25 consecutive patients with aMCI from the Mayo Clinic AD Research Center (ADRC) /AD Patient Registry (ADPR) to participate in the FDG PET study 12. The ages of the aMCI subjects ranged from 55 to 86. From the same ADRC/ADPR cohort, we recruited 25 cognitively normal (CN) subjects who were matched on age and gender to the aMCI subjects and underwent FDG PET during the same time period. This study was approved by the Mayo Institutional Review Board, and informed consent for participation was obtained from every subject or an appropriate surrogate.

Individuals participating in the ADRC /ADPR studies undergo clinical examinations, structural brain MRI, routine laboratory tests, and a battery of neuropsychological tests. At the completion of the evaluation, a consensus committee meeting is held involving the behavioral neurologists, neuropsychologists, nurses and the geriatrician who evaluated the subjects to assign a clinical diagnosis to the participant.

The operational definition of aMCI was based on clinical judgment through a careful history from the patient and preferably a collateral source without reference to MRI using the Petersen criteria for aMCI 2: 1) memory complaint, preferably corroborated by an informant; 2) objective memory impairment; 3) normal general cognitive function; 4) intact activities of daily living; 5) not demented. Patients with structural abnormalities that could impair cognitive function such as tumor, subdural hematoma, contusion from a previous head trauma, infarctions (>1cm in largest diameter), as well as addictions, psychiatric diseases or treatments that would have an effect on cognitive function were excluded.

CN subjects were people recruited from the community but were evaluated in the same manner as patients with MCI. The cognitively normal group did not have any neurological or psychiatric conditions, did not have a cognitive complaint, had a normal neurological and neurocognitive exam, and were not taking psychoactive medications in doses that would affect cognition. Patients with depression (according to the DSM IV criteria) were not excluded.

Based on the results from the study subjects, we recruited an independent sample of age and gender matched aMCI (n=13) and CN (n=10) subjects from a population based cohort: Mayo Clinic Study of Aging using similar imaging methods. We describe the reasons for recruiting this validation sample in the results section.

PET and MRI Acquisitions

All subjects underwent FDG PET and MRI studies within 110 days of the clinical evaluation. FDG PET scans were performed on the General Electric (GE) Advance PET scanner (GE, Milwaukee, WI) operating in three-dimensional mode. The subjects were injected with FDG in a dimly lit room with minimal auditory stimulation. After a 30 minute uptake period, PET acquisition was followed by a transmission scan. Emission data was reconstructed into a 128×128 matrix, 256 mm field of view (FOV). The pixel size was 2.00 mm with 4.25 mm slice thickness. All subjects underwent MRI studies on a 1.5 T scanner (Signa;GE Medical Systems, Milwaukee, WI). T1 weighted three-dimensional volumetric spoiled gradient-recalled echo sequence with 124 continuous partitions in coronal plane, 1.6 mm slice thickness, a 24 × 18.5 cm. field of view, 192 views, and 25° flip angle was acquired for registration and segmentation of FDG PET images. A fluid attenuated inversion recovery (FLAIR) pulse sequence in axial plane with TR/TI/TE=16,000/2,600/140ms, 256×160 matrix, one repetition, 22cm FOV, 3 mm interleaved images of the whole head was used for the assessment for cerebrovascular lesions.

Assessment of Cerebrovascular lesions

Cerebrovascular lesions were identified and rated on FLAIR images because they contribute to cognitive impairment in MCI 13. A radiologist (KK) blinded to all clinical information assessed the white matter hyperintensities (WMH), hemispheric cortical and lacunar infarctions. WMH volume was estimated by visually comparing the subject’s FLAIR images to a bank of ten FLAIR image templates with increasing WMH volumes (from 1 cm3 to 100 cm3) determined with an automated image segmentation algorithm 13. The WMH volume estimation algorithm was previously validated against quantification using automated image segmentation of the WMH volume 14. Because cortical infarctions >1cm. in largest diameter was one of the exclusion criteria, we only evaluated hemispheric cortical infarctions ≤ 1cm in largest diameter. Subcortical infarctions were defined as discrete subcortical lesions >3 mm in diameter with intensity that is equivalent to CSF on FLAIR images and accompanying hyperintense gliotic rim 13.

FDG PET Analysis

The automated anatomic labeling (AAL) atlas 15 was modified in-house to contain the following labeled regions of interest (ROI) : right and left parietal lobe that included the posterior cingulate gyrus and precuneus, temporal lobe and pons. These regions were chosen based on previous reports showing decreased FDG uptake in these regions in patients with pathologically proven AD 1618 and pons was chosen as an internal reference ROI 6. The high-resolution T1 weighted single-subject brain image 15 with atlas labels was normalized to the custom template generated from 200 AD and 200 CN subjects 19 using the unified segmentation method in SPM5 20, giving a set of labels corresponding to the custom template space. The resulting ROI labels were then imported into the Analyze (Mayo Clinic, Rochester, MN) ROI tool, and hand-edited by a trained technician (MMS), reviewed by a radiologist (KK), and saved to produce a more accurate set of ROI labels corresponding to the custom template, which we refer to as the “custom template AAL” atlas. The FDG PET image volume of each subject was co-registered to his/her own T1-weighted MRI scan, using a 6-degrees of freedom (DOF) affine registration with mutual information cost function. Each subject’s MRI scan was then spatially normalized to the custom template using the unified segmentation model of SPM5 20, giving a discrete cosine transformation (DCT), say Gi, which normalizes the MRI of subject i to the custom template. Then for each subject, the inverse transformation (Gi−1) was applied to the custom template AAL atlas in order to warp the atlas labels to the subject's native anatomical space. Atlas-based parcellation of FDG PET images into ROIs was therefore performed in subject space. For each subject, the native-space segmented grey and white matter probability maps generated from the unified segmentation routine, were combined to create a binary brain mask. Each subject's brain mask was then multiplied by the subject-specific warped atlas, to generate a custom atlas for each subject, parcellated into the aforementioned ROIs. This step was performed in order to minimize including non-brain regions in statistics of all ROIs 21, 22. Partial volume correction for tissue and cerebrospinal fluid (CSF) compartments was applied using the two compartment model 21. Statistics on image voxel values were extracted from each labeled cortical ROI in the atlas. FDG ratio images were calculated by dividing the median value in each target cortical ROI value by the median value in the pontine ROI of the atlas 6. A temporoparietal cortical FDG uptake summary measure (temporoparietal FDG) was formed by combining the left and right parietal lobe (including posterior cingulate gyrus and precuneus) and temporal lobe ratio values for each subject. The temporoparietal FDG was computed as the ratio of the median FDG uptake of the contributing ROIs, to median FDG uptake of the pontine ROI. Temporal and parietal regions were chosen to derive the temporoparietal FDG measure, because decreased glucose metabolism in these regions is a sensitive marker for AD related pathology 22.

Statistical parametric mapping (SPM)5 was used to evaluate FDG uptake on a voxel-wise basis 20. For each subject, all voxels in the subject’s FDG PET image were divided by the median FDG uptake of the pontine ROI to form uptake ratio images. Then for each subject, i, spatial normalization of the FDG PET uptake ratio image to custom template space was performed using the DCT normalization parameters, Gi, obtained as described above. Finally, smoothed with a Gaussian kernel with full width at half maximum of 8mm. Using these smoothed images, voxel-wise FDG uptake differences between groups were assessed in SPM5. Statistical maps of group differences were displayed at a significance value of P < 0.005, uncorrected for multiple comparisons.

Voxel Based Morphometry (VBM)

First, a custom template and tissue probability maps were created in SPM5, using the T1 weighted MR images of the 25 aMCI and 25 CN subjects 23. The custom template and tissue probability maps were created by first normalizing and segmenting the 50 scans using the unified segmentation model in SPM5 with the standard Montreal Neurological Institute (MNI) template and tissue probability maps, followed by a clean up step which uses a hidden Markov random field model to increase the accuracy of the individual subject tissue probability maps, and finally averaging the normalized subject tissue probability maps. All subject images were then normalized and segmented using the unified segmentation model and the custom tissue probability maps, followed by the hidden Markov random field clean up step. Jacobian modulation was applied to compensate for the effect of spatial normalization and to restore the original absolute grey matter volume in the segmented gray matter images. These modulated images were then smoothed with a 6 mm full width at half maximum (FWHM) smoothing kernel. Gray matter differences between groups were assessed using a two-sided T-test within the general linear model framework of SPM. Group differences were displayed at a significance value of P<0.005, uncorrected for multiple comparisons. Note that the same significance threshold was used for the FDG-PET analysis.

Statistical Analysis

Wilcoxon rank sum tests were used for statistical analysis comparing the demographic and clinical features of the clinical groups within age strata. Various types of cerebrovascular lesion load were compared with Kruskal-Wallis or Fisher’s exact tests in the four groups. Age effects on WMH were tested using Spearman’s rank-order correlation. Spearman’s rank-order correlation was used to test for a relationship between age, as a continuous measure, and FDG PET uptake within clinical groups. Additionally, a linear regression model was fit with main effects for age and clinical diagnosis group as well as an interaction term between age and clinical group. This interaction term was used to assess different associations with FDG PET and age among the clinical groups.


The temporoparietal FDG was plotted against age in patients with aMCI and cognitively normal subjects in Figure 1. The temporoparietal FDG was lower in aMCI subjects than CN subjects. However, the difference between CN and aMCI subjects decreased with increasing age as demonstrated in Figure 1. When age was treated as a continuous variable, a trend of increasing temporoparietal FDG with increasing age was identified only in the aMCI group although this did not reach statistical significance. Spearman’s rank-order correlation between the temporoparietal FDG and age is estimated to be ( −0.15, P=0.46) among CN and (+0.32, P=0.12) among aMCI subjects. We saw evidence of a difference in the relationship of age with temporoparietal FDG by clinical diagnosis (p=0.10). In particular, there was some evidence that in aMCI subjects, younger age was on average associated with lower temporoparietal FDG but in CN subjects mean FDG remained unchanged with age.

Figure 1
Scatter plot of the global-cortical-to-pons FDG ratio versus patient age for CN (panel a) and aMCI (panel b) subjects. The lines indicate the estimated mean as a function of age using a nonparametric loess scatter plot smoother. The shaded regions represent ...

For visualization of the age effects on the regional pattern of glucose metabolic changes in aMCI patients by voxel-based analysis and for practical utility, we took the median age 73 (interquartile range 69–78) as cut-off and divided the aMCI subjects and the age and gender matched CN subjects into two groups: CN or aMCI-young (age≤73) and CN or aMCI-old (age>73). In order to validate the unexpected finding of no difference in FDG between old CN and old aMCI, we analyzed an independent sample of age and gender matched aMCI (n=13) and CN (n=10) subjects (age>73) who were recruited from a population based cohort: Mayo Clinic Study of Aging using similar imaging methods. Mayo Clinic Study of Aging is an epidemiological study on 2,000 non-demented persons aged 70–89 years recruited in 2004 through a random selection process in Olmsted County, Minnesota 24. A population based validation sample was available only for the a-MCI-old group, but not for the aMCI-young group because of the age range of the Mayo Clinic Study of Aging cohort.

Demographic aspects of the study groups, Mini Mental State Examination (MMSE) Scores and Clinical Dementia Rating (CDR) sum of boxes scores are listed in Table 1. On average, age and gender were similar between CN-young compared to MCI-young and CN-old compared to MCI-old groups in both the study sample and the validation sample. Years of education was not different among all of the groups, although a trend of a higher education level was observed in CN-old subjects compared to CN young subjects (p>0.08) in the study sample. CN-old subjects in the study sample tended to have lower MMSE than CN-young (P=0.01). There was no difference in cognitive function measured with MMSE and CDR sum of boxes between aMCI-young and aMCI-old subjects in the study and the validation samples (p>0.87).

Table 1
Patient Characteristics

The temporoparietal FDG measure was lower in the aMCI –young group compared to the CN-young group (p=0.01). However, no difference in temporoparietal FDG was observed among the CN-old and aMCI-old groups (p=0.67) in the study sample and the CN-old and aMCI-old groups (p=0.52) in the validation sample (Figure 2). This was due to the fact that FDG uptake increased with age in aMCI.

Figure 2
Box plots with individual data points for the global-cortical-to-pons FDG summary measure by age and diagnosis for study subjects from the Mayo community (ADPR) and (ADRC) samples (panel a) and an independent population-based validation sample from the ...

We found statistically significant age effects on the WMH load. WMH load increased with increasing age in the CN (r=0.52, p= 0.008) and aMCI (r=0.43; p=0.03) subjects. Although subcortical infarctions were twice as frequent in aMCI-old and CN-old compared to aMCI-young and CN-young groups and cortical infarctions (≤ 1cm in diameter) were twice as frequent in aMCI-old compared to aMCI-young, there were no statistically significant differences among groups most likely because of the small number of infarctions and lack of power (Table 2).

Table 2
Cerebrovascular lesions by group within the study sample

Voxel-based analysis using SPM5 demonstrated that compared to age matched CN subjects, aMCI-young patients had decreased glucose metabolism in the precuneus, posterior cingulate gyrus, medial temporal lobe, lateral temporal and parietal lobes in both hemispheres, consistent with the typical pattern of FDG PET changes observed in AD (Figure 3). The regional decrease in FDG uptake in aMCI-old patients compared to age matched CN subjects however was different from the AD-like pattern observed in aMCI-young subjects, primarily involving the orbitofrontal and prefrontal cortex in both hemispheres, and the left lateral parietal lobe (Figure 4).

Figure 3
Voxel-based analysis of the FDG PET uptake-to-pons ratio in aMCI-young patients compared to age and gender matched CN-young subjects (P<0.005 uncorrected).
Figure 4
Voxel-based analysis of the FDG PET uptake-to-pons ratio in aMCI-old patients compared to age and gender matched CN-old subjects (P<0.005 uncorrected).

The pattern of gray matter atrophy in aMCI-young patients mirrored the pattern of reduction in glucose metabolism in this group involving mainly the precuneus, medial temporal lobe, lateral temporal and parietal lobes in both hemispheres and some involvement of the prefrontal cortex (Figure 5). On the other hand, the cortical atrophy was far less significant in aMCI-old subjects mainly involving the hippocampus and the basal forebrain in both hemispheres and a few clusters were present in the prefrontal cortices (Figure 6).

Figure 5
Pattern of grey matter reduction in aMCI-young patients compared to age and gender matched CN-young subjects demonstrated in red and yellow (P<0.005 uncorrected).
Figure 6
Pattern of grey matter reduction in aMCI-old patients compared to age and gender matched CN-old subjects demonstrated in red and yellow (P<0.005 uncorrected).


Our data demonstrated that the effects of age on the pattern of glucose metabolism in aMCI are in agreement with the previous observations in AD 911. The median age of 73 was taken as the cut-off to classify subjects into young and old groups in order to determine the regional differences in glucose metabolic changes in aMCI-old and aMCI-young subjects. The dicotomization of the aMCI cohort was also necessary for determining the practical clinical utility of FDG PET for early diagnosis of AD in aMCI. Furthermore, absence of temporoparietal and posterior cingulate cortex hypometabolism have been previously demonstrated in late onset AD (onset >73–75 years of age) 9, 10.

Patients with aMCI who were younger than or at 73 years of age or less had decreased glucose metabolism in the posterior cingulate gyrus, lateral temporal and parietal cortex compared to age and gender matched controls. This regional pattern of decreased glucose metabolism is consistent with the typical pattern observed in pathologically proven AD 1618, and in patients with aMCI who progress to AD 3, 5, 25. However, when aMCI-old patients were compared to age and gender matched CN subjects, glucose metabolism was mainly reduced in the frontal lobes and the left lateral parietal lobe. Although, the reduction in lateral parietal lobe glucose metabolism can be considered as a partial expression of the AD pattern, the differences in glucose metabolic patterns between old and young aMCI can not be ascribed to differences in cognitive performance level, as performance was the same in these two groups. Decreased glucose metabolism in the frontal cortex has been observed in aMCI patients who did not progress to AD after an average of 12–24 months 5, 25. Whether frontal lobe hypometabolism represents a low likelihood of progression to AD in the aMCI-old subjects in our cohort will be investigated.

There may be several explanations for the observed age effects on FDG PET changes in aMCI compared to age matched CN subjects. An obvious explanation may be that the CN-old group may have had a higher frequency of incipient AD pathology than the CN-young group thus diminishing the difference in glucose uptake between aMCI-old and CN-old subjects in brain areas typically associated with AD. However, while aging is the primary risk factor for dementia, we did not identify age effects on the temporoparietal FDG uptake in CN subjects (Figure 1), which suggest that the absence the AD pattern of FDG uptake differences among the MCI-old and CN-old groups is less likely to be related to a high frequency of incipient AD in the CN-old group.

A second possible explanation is that the aMCI-young patients had more severe AD pathology than the aMCI-old patients. If so, we would need to explain the fact that the cognitive performance of the two groups was similar. Age of onset has been a significant predictor of neurofibrillary tangles and neuritic plaques, and synaptic density in patients with AD; with early onset AD patients having more neurofibrillary tangles and neuritic plaques and lower synaptic density than late onset AD patients 26. On the basis of these findings, it is possible that aMCI-young patients with a greater AD pathology and AD related glucose metabolic changes may be compensating more efficiently than the aMCI-old patients in whom the pathological involvement and the FDG PET changes were less severe, but the cognitive function was similar to aMCI-young patients.

A third possible explanation for our findings is that a higher frequency of non-AD dementia related pathologies that increase in prevalence with age may be contributing to the cognitive impairment in aMCI-old patients more significantly than in aMCI-young patients. Although AD is the most common pathology underlying aMCI, other dementia pathologies encountered in aMCI include cerebrovascular disease, and Lewy body pathology 2731. Aging is a major risk factor for cerebrovascular disease and Lewy body pathology, as it is for AD, and mixed pathologies account for a majority of the dementia cases in the community 32. For example in our cohort, the WMH load, which contributes to the cognitive impairment in MCI, increased with increasing age in both CN and aMCI subjects. Thus, it is possible that the absence of a typical AD-like pattern of glucose metabolism in the aMCI-old group may be related to a greater pathological heterogeneity in the aMCI-old compared to the aMCI-young patients 33, 34.

VBM showed medial temporal lobe atrophy in both the aMCI-young and old subjects compared to the age matched CNs, which is a feature of early AD pathology and aMCI 19. However, the pattern of the neocortical atrophy differed significantly between the two age groups. The temporoparietal cortex and precuneus atrophy in the aMCI-young patients was similar to the pattern of decreased glucose metabolism in this group on FDG-PET. On the other hand, atrophy, which was mainly confined to the medial temporal lobe and basal forebrain in the aMCI-old patients with some involvement of the prefrontal cortex, was different from the pattern of decreased glucose metabolism in the aMCI-old group. This topographical dissociation between the decrease in glucose metabolism and atrophy points to a heterogeneity in the imaging findings of the older aMCIs.

The association between the pathological features of Alzheimer's disease and dementia have recently been found to be stronger in younger than in older old persons, demonstrating a greater heterogeneity in clinical expression of AD pathology in older than in younger elderly. The differences in AD pathological burden between demented and non-demented elderly narrowed down with increasing age 34. In the current study, aMCI-young patients had greater neocortical atrophy and greater reduction in temporoparietal glucose metabolism than aMCI-old patients when compared to the age matched controls. Taken together, the age effects on imaging findings in aMCI agree with the findings at autopsy 34, showing that the AD-like differences in glucose metabolism and gray matter volume between CN and aMCI may become less significant with increasing age in aMCI.

We optimized the power for detecting differences among groups by comparing aMCI subjects to age and gender matched CN subjects. Because our study sample was recruited from the community (ADPR) and dementia (ADRC) clinics, we confirmed the observed age effects in the aMCI-old subjects in a population based sample. Age 73 was used as a cut-off based on the age distribution of our cohort. We recognize that a larger study could reveal a slightly different age above which FDG PET looses its ability to distinguish aMCI from CN.


The data we present suggest that age effects the glucose metabolic changes in aMCI, and agree with the observations in AD 9, 10. The typical temporoparietal hypometabolic pattern of AD characterizes younger aMCI patients (age ≤73), but not the older aMCI patients (age >73). It is important to account for age when using temporoparietal hypometabolism on FDG PET as a marker for early diagnosis of AD in aMCI.


Grant Support: Supported by Paul Beeson Career Development Awards in Aging K23-AG030935, NIH Roadmap Multidisciplinary Clinical Research Career Development Award KL2 RR024151 (NIH/NCRR), NIH/NIA-AG11378, AG06786, and AG16574, the Alexander Family, and by the Robert H. and Clarince Smith and Abigail Van Buren Alzheimer’s Disease Research program.

Abbreviation Key

Alzheimer’s disease
Amnestic mild cognitive impairment
Cognitively normal
[18F]fluorodeoxyglucose (FDG) Positron emission tomography
Statistical parametric mapping
Alzheimer’s Disease Research Center and Alzheimer’s Disease Patient Registry
automated anatomic labeling
Region of interest
White matter hyperintensities
Voxel based morphometry


1. Thal LJ, Kantarci K, Reiman EM, et al. The role of biomarkers in clinical trials for Alzheimer disease. Alzheimer Dis Assoc Disord. 2006;20:6–15. [PMC free article] [PubMed]
2. Petersen RC, Stevens JC, Ganguli M, et al. Practice parameter: early detection of dementia: mild cognitive impairment (an evidence-based review). Report of the Quality Standards Subcommittee of the American Academy of Neurology. Neurology. 2001;56:1133–1142. [see comment] [PubMed]
3. Chetelat G, Desgranges B, de la Sayette V, et al. Mild cognitive impairment: Can FDG-PET predict who is to rapidly convert to Alzheimer's disease? Neurology. 2003;60:1374–1377. [PubMed]
4. De Santi S, de Leon MJ, Rusinek H, et al. Hippocampal formation glucose metabolism and volume losses in MCI and AD. Neurobiology of Aging. 2001;22:529–539. [PubMed]
5. Drzezga A, Lautenschlager N, Siebner H, et al. Cerebral metabolic changes accompanying conversion of mild cognitive impairment into Alzheimer's disease: a PET follow-up study. European Journal of Nuclear Medicine & Molecular Imaging. 2003;30:1104–1113. [PubMed]
6. Minoshima S, Giordani B, Berent S, et al. Metabolic reduction in the posterior cingulate cortex in very early Alzheimer's disease. Annals of Neurology. 1997;42:85–94. [PubMed]
7. Nestor PJ, Fryer TD, Smielewski P, et al. Limbic hypometabolism in Alzheimer's disease and mild cognitive impairment. Annals of Neurology. 2003;54:343–351. [PubMed]
8. Mosconi L, Brys M, Glodzik-Sobanska L, et al. Early detection of Alzheimer's disease using neuroimaging. Exp Gerontol. 2007;42:129–138. [PubMed]
9. Kim EJ, Cho SS, Jeong Y, et al. Glucose metabolism in early onset versus late onset Alzheimer's disease: an SPM analysis of 120 patients. Brain. 2005;128:1790–1801. [PubMed]
10. Sakamoto S, Ishii K, Sasaki M, et al. Differences in cerebral metabolic impairment between early and late onset types of Alzheimer's disease. J Neurol Sci. 2002;200:27–32. [PubMed]
11. Matsunari I, Samuraki M, Chen WP, et al. Comparison of 18F-FDG PET and optimized voxel-based morphometry for detection of Alzheimer's disease: aging effect on diagnostic performance. J Nucl Med. 2007;48:1961–1970. [PubMed]
12. Petersen RC, Kokmen E, Tangalos E, et al. Mayo Clinic Alzheimer's Disease Patient Registry. Aging-Clinical & Experimental Research. 1990;2:408–415. [PubMed]
13. Kantarci K, Petersen RC, Przybelski SA, et al. Hippocampal volumes, proton magnetic resonance spectroscopy metabolites, and cerebrovascular disease in mild cognitive impairment subtypes. Arch Neurol. 2008;65:1621–1628. [PMC free article] [PubMed]
14. Murray ME, Shiung MM, Weigand SD, et al. Mayo MRI visual grading scale: Validation studies. Abstracts of the Annual Meeting of the Society for Neuroscience; San Diego, CA. 2007. http://sfnscholaronecom/
15. Tzourio-Mazoyer N, Landeau B, Papathanassiou D, et al. Automated anatomical labeling of activations in SPM using a macroscopic anatomical parcellation of the MNI MRI single-subject brain. Neuroimage. 2002;15:273–289. [PubMed]
16. Hoffman JM, Welsh-Bohmer KA, Hanson M, et al. FDG PET imaging in patients with pathologically verified dementia. J Nucl Med. 2000;41:1920–1928. [PubMed]
17. Jagust W, Reed B, Mungas D, et al. What does fluorodeoxyglucose PET imaging add to a clinical diagnosis of dementia? Neurology. 2007;69:871–877. [PubMed]
18. Silverman DH, Small GW, Chang CY, et al. Positron emission tomography in evaluation of dementia: Regional brain metabolism and long-term outcome. JAMA. 2001;286:2120–2127. [see comment] [PubMed]
19. Vemuri P, Whitwell JL, Kantarci K, et al. Antemortem MRI based STructural Abnormality iNDex (STAND)-scores correlate with postmortem Braak neurofibrillary tangle stage. Neuroimage. 2008;42:559–567. [PMC free article] [PubMed]
20. Ashburner J, Friston KJ. Unified segmentation. Neuroimage. 2005;26:839–851. [PubMed]
21. Meltzer CC, Kinahan PE, Greer PJ, et al. Comparative evaluation of MR-based partial-volume correction schemes for PET. J Nucl Med. 1999;40:2053–2065. [PubMed]
22. Sun FT, Schriber RA, Greenia JM, et al. Automated template-based PET region of interest analyses in the aging brain. Neuroimage. 2007;34:608–617. [PMC free article] [PubMed]
23. Vemuri P, Gunter JL, Senjem ML, et al. Alzheimer's disease diagnosis in individual subjects using structural MR images: validation studies. Neuroimage. 2008;39:1186–1197. [PMC free article] [PubMed]
24. Roberts RO, Geda YE, Knopman DS, et al. The Mayo Clinic Study of Aging: design and sampling, participation, baseline measures and sample characteristics. Neuroepidemiology. 2008;30:58–69. [PMC free article] [PubMed]
25. Anchisi D, Borroni B, Franceschi M, et al. Heterogeneity of brain glucose metabolism in mild cognitive impairment and clinical progression to Alzheimer disease. Arch Neurol. 2005;62:1728–1733. [PubMed]
26. Bigio EH, Hynan LS, Sontag E, et al. Synapse loss is greater in presenile than senile onset Alzheimer disease: implications for the cognitive reserve hypothesis. Neuropathol Appl Neurobiol. 2002;28:218–227. [PubMed]
27. Bennett DA, Wilson RS, Schneider JA, et al. Natural history of mild cognitive impairment in older persons. Neurology. 2002;59:198–205. [PubMed]
28. 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]
29. Morris JC, Price AL. Pathologic correlates of nondemented aging, mild cognitive impairment, and early-stage Alzheimer's disease. J Mol Neurosci. 2001;17:101–118. [PubMed]
30. Petersen RC, Parisi JE, Dickson DW, et al. Neuropathologic features of amnestic mild cognitive impairment. Arch Neurol. 2006;63:665–672. [PubMed]
31. Storandt M, Grant EA, Miller JP, et al. Longitudinal course and neuropathologic outcomes in original vs revised MCI and in pre-MCI. Neurology. 2006;67:467–473. [PubMed]
32. Schneider JA, Arvanitakis Z, Bang W, et al. Mixed brain pathologies account for most dementia cases in community-dwelling older persons. Neurology. 2007;69:2197–2204. [PubMed]
33. Seo SW, Cho SS, Park A, et al. Subcortical Vascular versus Amnestic Mild Cognitive Impairment: Comparison of Cerebral Glucose Metabolism. J Neuroimaging. 2008 [PubMed]
34. Savva GM, Wharton SB, Ince PG, et al. Age, neuropathology, and dementia. N Engl J Med. 2009;360 2302-230. [PubMed]