Research ArticleClinical Investigations

Diabetes and Elevated Hemoglobin A1c Levels Are Associated with Brain Hypometabolism but Not Amyloid Accumulation

Rosebud O. Roberts, David S. Knopman, Ruth H. Cha, Michelle M. Mielke, V. Shane Pankratz, Bradley F. Boeve, Kejal Kantarci, Yonas E. Geda, Clifford R. Jack, Ronald C. Petersen and Val J. Lowe
Journal of Nuclear Medicine May 2014, 55 (5) 759-764; DOI: https://doi.org/10.2967/jnumed.113.132647

Abstract

Dysfunctional insulin signaling may affect brain metabolism or amyloid deposition. We investigated the associations of type 2 diabetes with amyloid accumulation measured using 11C-Pittsburgh compound B (11C-PiB) and brain hypometabolism measured using 18F-FDG PET. Methods: We studied a sample of nondemented participants from the population-based Mayo Clinic Study of Aging. All subjects underwent MR imaging, amyloid PET, and 18F-FDG PET. Alzheimer disease (AD) signature and region-of-interest (ROI) measures for 11C-PiB retention ratio and 18F-FDG ratio were measured. Diabetes was assessed from the Rochester Epidemiology Project medical records linkage system. Results: Among 749 participants (median age, 79.0 y; 56.5% men, 81.0% cognitively normal; 20.6% diabetic individuals), 18F-FDG hypometabolism (18F-FDG ratio < 1.31) in the AD signature meta-ROI was more common in diabetic individuals (48.1%) than in nondiabetic individuals (28.9%; P < 0.001). The median 18F-FDG ratio was lower in diabetic individuals than in nondiabetic individuals in the AD signature meta-ROI (1.32 vs. 1.40, P < 0.001) and in the angular (1.40 vs. 1.48, P < 0.001) and posterior cingulate gyri ROIs (1.63 vs. 1.72, P < 0.001). The odds ratio (OR) for abnormal AD signature 18F-FDG hypometabolism was elevated (2.28; 95% confidence interval [CI], 1.56–3.33) in diabetic individuals versus nondiabetic individuals after adjustment for age, sex, and education and after additional adjustment for apolipoprotein ε4 allele, glycemic level, and cognitive status (OR, 1.69; 95% CI, 1.10–2.60). However, the AD signature 11C-PiB retention ratio was similar in diabetic individuals versus nondiabetic individuals (OR, 1.03; 95% CI, 0.71–1.51; P = 0.87). In post hoc analyses in nondiabetic individuals, a 1% increase in hemoglobin A1c was associated with greater AD signature hypometabolism in cognitively normal subjects (OR, 1.93; 95% CI, 1.03–3.62; P = 0.04) and in the total cohort (OR 1.59; 95% CI, 0.92–2.75; P = 0.10). Conclusion: Diabetes and poor glycemic control in nondiabetic individuals may enhance glucose hypometabolism in AD signature regions. These factors should be investigated in longitudinal studies for their role in detecting onset of symptoms in AD.

Several studies have reported associations of type 2 diabetes with cognitive impairment, Alzheimer disease (AD), or vascular dementia (1–5). The exact mechanisms underlying the association of diabetes with AD remain incompletely understood, but studies suggest that ischemic cerebrovascular disease and hippocampal atrophy are involved (6–10). Other potential mechanisms including global and regional brain hypometabolism (6,10) and amyloid deposition have been hypothesized (11) but require further investigation. Neuropathologic studies, which provide the gold standard for AD pathology, have been inconsistent; some studies have reported greater amyloid-related pathology in diabetic brains (11) whereas others have not (3,7,12,13). Given the dysfunctional insulin signaling in type 2 diabetes, we hypothesized that abnormal cerebral glucose metabolism in diabetic individuals may lead to neuronal injury and cognitive impairment (14). PET offers the opportunity to study in vivo markers of both abnormal brain metabolism and amyloid accumulation.

The objective of our study, therefore, was to specifically examine the association of type 2 diabetes with in vivo markers of amyloid accumulation and cerebral glucose metabolism measured using 11C-Pittsburgh compound B (11C-PiB) PET and 18F-FDG PET, in a subset of participants from the population-based Mayo Clinic Study of Aging.

MATERIALS AND METHODS

Study Participants

The study design and methodology have previously been published (15). Briefly, participants were Olmsted County, Minnesota, residents aged 70–89 y on October 1, 2004, who were randomly selected from an enumeration of the population using the Rochester Epidemiology medical records linkage system and invited to participate in the Mayo Clinic Study of Aging (16). Starting in 2008, additional subjects were recruited continually to maintain the sample size. Participants were invited to undergo MR imaging of the brain in 2005 and PET imaging in 2006. All subjects signed a written informed consent form that included the Health Insurance Portability and Accountability Act authorization to use and disclose protected health information. The Mayo Clinic and Olmsted Medical Center Institutional Review Boards approved all study protocols.

Assessment of Cognitive Status

Each participant underwent a nurse or study coordinator interview, a physician evaluation, and cognitive testing by a psychometrist. The interview included questions about memory (participant) and the Clinical Dementia Rating Scale (17) and Functional Activities Questionnaire (informant) (18). The physician evaluation included the Short Test of Mental Status (19) and a neurologic examination. The cognitive testing used 9 tests to assess performance in memory (3 tests), executive function (2 tests), language (2 tests), and visuospatial skills (2 tests) domains. Cognitive domain scores were computed from age-adjusted test scores that were scaled (mean, 10; SD, 3), then summed and scaled tests within a domain to allow comparisons across domains (15,20,21).

Mild cognitive impairment (MCI) criteria were defined as cognitive concern (subject, informant, nurse, or physician), impairment in one cognitive domain or more, essentially normal functional activities, and no dementia (15,21,22). Dementia was based on the Diagnostic and Statistical Manual of Mental Disorders, 4th ed. (DSM-IV) (23) criteria. Subjects who performed within the range for cognitively normal (CN) subjects from the population and did not meet criteria for MCI or dementia were considered CN (15,21,22).

Ascertainment of Type 2 Diabetes

Type 2 diabetes was abstracted from the medical records (16) as treatment for diabetes (oral antidiabetic agents, insulin), or a fasting blood glucose of 126 mg/dL or more reported two or more times, or a physician diagnosis (24).

Other Covariates

Date of birth and years of education were assessed by interview. Hemoglobin A1c (HBA1c) level at enrollment was assessed from the medical record (16). Apolipoprotein (APOE) genotype was performed at baseline, and blood glucose was measured just before PET imaging.

18F-FDG PET and 11C-PiB PET Acquisition

MR imaging was performed at 3T with a 3-dimensional volumetric T1-D-MPRAGE (magnetization-prepared rapid gradient-echo) sequence previously described (25). PET images were acquired using a PET/CT scanner operating in 3-dimensional mode (26). A CT image was obtained for attenuation correction. The 11C-PiB PET scan consisted of four 5-min dynamic frames acquired 40–60 min after injection (27). Subjects were injected with 18F-FDG 1 h after the 11C-PiB scan and imaged after 30–38 min, for an 8-min image acquisition of four 2-min dynamic frames (27,28). An in-house fully automated image-processing pipeline was used for quantitative image analysis for 11C-PiB and 18F-FDG. Each subject’s PET images were registered to his or her MR images, which had been labeled with our parcellation atlas (27). An AD signature cortical 11C-PiB PET meta-region-of-interest (meta-ROI) retention ratio was calculated from the median uptake over voxels in the prefrontal, orbitofrontal, parietal, temporal, anterior cingulate, and posterior cingulate/precuneus ROIs normalized to the cerebellar gray matter ROI of the atlas (26). An AD signature 18F-FDG PET ratio was calculated in a similar manner based on glucose metabolic rates from an Alzheimer signature meta-ROI and consisted of the average bilateral angular gyri, posterior cingulate/precuneus, and inferior temporal cortical ROIs from both hemispheres normalized to pons uptake (29–31).

Statistical Analyses

We defined an abnormal AD signature meta-ROI 11C-PiB retention ratio (standardized uptake value ratio) as greater than 1.50 and an abnormal AD signature meta-ROI 18F-FDG ratio as less than 1.31, corresponding to the 90% sensitivity in AD dementia (28,32). For 11C-PiB PET, for which higher biomarker values are worse, the cutpoint corresponding to 90% sensitivity was at the 10th percentile of the AD distribution. For 18F-FDG PET, for which lower biomarker values are worse, the cutpoint was at the 90th percentile of the AD distribution. We examined the association of diabetes with abnormal AD signature 11C-PiB PET ratio and 18F-FDG PET ratio using logistic regression models adjusted for age and sex. Given that 18F-FDG ratio varies with cognitive status, we performed analyses for the total sample, MCI, and CN (30) We examined potential confounding and effect modification (interaction) by APOE ε4 allele, blood glucose at PET imaging, and cognitive status. For comparison, we selected non-AD regions of the primary sensorimotor regions and global whole brain (based on 35 ROIs), computed unadjusted and glucose-adjusted means of 18F-FDG ratio, and examined their associations with diabetes using linear regression models. We conducted post hoc analyses in nondiabetes to determine whether glycemic control (HBA1c) was also associated with hypometabolism. All analyses were conducted using SAS (version 9.3; SAS). P values were considered significant at an α less than 0.05 (2-tailed tests).

RESULTS

Table 1 shows the characteristics of 749 participants without dementia who underwent PET imaging (2006–2012). The frequency of an abnormal AD signature meta-ROI 18F-FDG ratio was significantly higher among diabetic individuals; the median was lower in diabetic individuals than in nondiabetic individuals in both CN and MCI subjects (Fig. 1, top) and in the total sample (not shown). 18F-FDG hypometabolism was found in all 3 subregions of the AD signature meta-ROI. Unadjusted means for AD signature 18F-FDG ratio for sensorimotor regions and whole brain and were significantly different in diabetic individuals versus nondiabetic individuals, but glucose-adjusted means did not differ (Supplemental Table 1; supplemental materials are available at http://jnm.snmjournals.org).

View this table:
TABLE 1

Characteristics of Participants by Type 2 Diabetes

FIGURE 1.
FIGURE 1.

Box plots for 18F-FDG and 11C-PiB retention ratio in AD signature regions by diabetes mellitus (DM) in CN and MCI subjects. ● = women; ○ = men.

In the total sample, AD signature meta-ROI 18F-FDG ratio was significantly lower in younger diabetic individuals than in nondiabetic individuals (Figs. 2A and 2B), in men than in women (Figs. 2C and 2D), and in APOE ε4 noncarriers than in carriers (Figs. 2E and 2F); the findings were consistent for CN and MCI (not shown). In representative scans, AD signature 18F-FDG hypometabolism was absent in subjects with no diabetes and no MCI (Fig. 3A), minimal in nondiabetic individuals with MCI (Fig. 3B), and worst in diabetic individuals with MCI (Figs. 3C and 3D).

FIGURE 2.
FIGURE 2.

Box plots of 18F-FDG ratio in AD signature regions in total sample by diabetes (DM) status stratified by age (70–79 y [A] and 80–89 y [B]), sex (women [C] and men [D]), and apolipoprotein ε4 allele (ε4 carrier [E] and ε4 noncarrier [F]). ● = women; ○ = men.

FIGURE 3.
FIGURE 3.

18F-FDG scans for participants without diabetes (CN [A] and MCI [B]) and with diabetes (both with MCI [C and D]). Each image set represents 3-dimensional stereotactic surface projections of regional hypometabolism normalized to age-appropriate healthy subjects and displayed as z score; upper boundary as 7.0 (red) and lower boundary as 0.0 (black). z score of 1.5 (medium blue) or greater would be considered significant.

By contrast, the proportion of diabetic individuals versus nondiabetic individuals with an abnormal AD signature meta-ROI 11C-PiB ratio did not differ (Table 1). The median AD signature meta-ROI 11C-PiB ratio was similar for diabetic individuals versus nondiabetic individuals (Table 1) among CN and MCI (Fig. 1, bottom).

Association of Diabetes with AD Signature 18F-FDG Uptake

Diabetes was significantly associated with AD signature meta-ROI 18F-FDG hypometabolism of the range seen in AD dementia (Table 2). The odds ratio (OR) decreased slightly after adjustment for glucose but remained significant for the total sample and MCI cases for the fully adjusted models. Diabetes was associated with decreased 18F-FDG ratio in the sensorimotor ROIs and the whole brain, but the associations were not significant after adjustment for glucose (Supplemental Table 2). There were no significant interactions of diabetes with age, sex, or APOE ε4; the OR for diabetes and hypometabolism was significant for subjects with glucose levels less than 126 mg/dL (OR, 1.72; 95% confidence interval, 1.11–2.69; P = 0.02) (Supplemental Table 3).

View this table:
TABLE 2

Association of Type 2 Diabetes with Abnormal 18F-FDG PET Ratio* in AD Signature Meta-ROI

Association of Diabetes with AD Signature 11C-PiB

In contrast to unadjusted estimates, the OR for abnormal AD signature meta-ROI 11C-PiB was significantly reduced in subjects with MCI after adjustment for age, sex, and education but not after adjustment for APOE ε4 allele (Table 3).

View this table:
TABLE 3

Associations of Type 2 Diabetes with Abnormal 11C-PiB PET Retention Ratio* in AD Signature Meta-ROI

HbA1c and AD Signature 18F-FDG Hypometabolism

In CN nondiabetic individuals, the OR for hypometabolism in the AD signature meta-ROI increased with each 1% increase in HBA1c after adjustment for age, sex, and education (OR, 1.93; 95% CI, 1.03–3.62; P = 0.04) and after additional adjustment for APOE ε4 allele (OR, 1.89; 95% CI, 1.00–3.57; P = 0.049).

DISCUSSION

In this sample of elderly persons, diabetes was associated with 18F-FDG hypometabolism. This association persisted in the AD signature region for the total sample and MCI cases even after controlling for serum glucose but not in the sensorimotor cortex. By contrast, diabetes was not associated with increased 11C-PiB retention ratio (amyloid accumulation) in AD signature meta-ROIs for the total sample, or for CN subjects, but the OR was significantly reduced in subjects with MCI in models that did not include APOE ε4.

Glucose hypometabolism in diabetic individuals suggests that diabetes affects cerebral energy metabolism independently of amyloid deposition, possibly through mechanisms that cause neuronal injury (33). This hypometabolism in CN diabetic individuals suggests that neuronal injury is present even before symptomatic evidence of cognitive impairment. The loss of significance in the CN group after adjustment for glucose level, however, suggests that the effect of diabetes may be small in CN or below the threshold for sustained effect, such that a larger sample of CN is required to detect a significant difference. Alternately, the findings in CN may be due to a more generalized diabetes effect rather than a specific AD meta-ROI effect. The persistence of significant associations for the total sample and for MCI cases, even after adjustment for glucose, suggests the effect may be real. The stronger association for MCI subjects suggests that hypometabolism may be in the causal pathway for cognitive impairment in diabetic individuals. Finally, the absence of an effect in multivariable models for the sensorimotor cortex suggests that the associations are not entirely explained by a competitive inhibition of labeled glucose uptake by elevated serum glucose in diabetic individuals.

Our subgroup analyses showed stronger associations of diabetes with hypometabolism for APOE ε4 noncarriers, younger subjects, and men. The stronger OR for APOE ε4 noncarriers suggests that APOE ε4 allele effects might share some mechanistic commonalities with diabetes, that are independent of APOE ε4–related amyloidosis (34). The stronger association in 70- to 79-y-olds may relate to earlier diabetes onset, age-related differences in diabetes effects, nonparticipation in older subjects, or potential survival bias. The stronger OR in men suggests etiologic differences as observed for amnestic MCI in our previous studies (21,35). Follow-up of the cohort will provide greater power to detect significant interactions.

Our findings for HBA1c in nondiabetic individuals support our hypothesis that altered glycemic control is associated with neuronal injury and are consistent with other studies. In CN adults with newly diagnosed prediabetes or diabetes, increasing insulin resistance was associated with greater 18F-FDG hypometabolism in AD regions (36), and higher fasting glucose levels in CN nondiabetic individuals were associated with 18F-FDG hypometabolism (37). These findings suggest that markers of prediabetes should be examined for early detection of brain hypometabolism.

The analyses reported here did not use partial-volume correction. When we applied partial-volume correction, our findings changed somewhat. Significant associations persisted with adjustment for age- and sex-adjusted models in AD signature regions. However, glucose-adjusted models were not significant. This difference in results may mean that the 18F-FDG changes we observed may in part reflect loss of brain volume due either to diabetes or to other etiology. Alternately, we recognize that if diabetes and related hyperglycemia contributes to structural and metabolic changes, controlling for blood glucose may result in overcontrolling.

The absence of an association of diabetes with 11C-PiB retention ratio in AD signature regions in CN subjects is consistent with findings from several studies (3,7,12,13,38). In one study, diabetes was associated with cerebral infarctions but not with AD pathology (7); another study found no difference in AD pathology among diabetic individuals and nondiabetic individuals (3). A third study reported that diabetic APOE ε4 carriers had a lower amyloid β load than nondiabetic individuals (12). In contrast, one study reported that diabetic APOE ε4 carriers had greater AD pathology (11), and another reported greater plaque density with longer duration of diabetes (3).

The absence of an association of diabetes with AD signature 11C-PiB ratio and the significantly lower AD signature 11C-PiB in diabetic MCI subjects are inconsistent with the hypothesis that diabetes contributes to increased amyloid deposition (39,40). However, they are consistent with other studies (41). In the Baltimore Longitudinal Study of Aging, diabetes-related parameters were not associated with AD pathology (41). It is possible that diabetes may have greater effects on neuronal injury and brain atrophy than on amyloid accumulation in elderly persons (8,9). The null association may also be due to interactions of timing of onset of vascular pathology with amyloid deposition in diabetic individuals. By contrast, subjects with diabetes and 11C-PiB deposition occurring independently at earlier ages may progress more rapidly from MCI to dementia and may be less likely to be captured in our cohort because we did not initially enroll demented subjects to imaging studies. The absence of a significant difference in AD signature 11C-PiB ratio for MCI diabetic individuals and nondiabetic individuals (Fig. 1, lower) versus the significant association observed in multivariable models suggests possible confounding; the association was significant after adjustment for age, sex, and education but not APOE ε4 allele. These findings indicate the need to account for potential confounders when comparing 18F-FDG ratio across regions.

One potential limitation of our study is that because of the cross-sectional design, we cannot assess causality. The absence of associations of diabetes with AD signature 18F-FDG ratio in APOE ε4 carriers and with AD signature 11C-PiB retention ratio could be due to underrepresentation of diabetic individuals, lower participation of diabetic individuals with more severe or longer disease duration, or lower frequency of APOE ε4 allele in this older age group. HBA1c was assessed at enrollment and not at the time of imaging; however, the findings support our hypothesis that impaired glucose metabolism affects brain metabolism. Ongoing recruitment should increase the number of diabetic individuals in our sample. Finally, more than 98% of our cohort is of northern European ancestry, thus our findings should be confirmed in other ethnicities.

Our study has several important strengths. Participants were randomly selected from the population; thus, our findings are less subject to selection or volunteer bias. The diagnosis of diabetes was reliably ascertained using the medical records linkage system and not subject to recall bias as occurs with self-reporting. We had a large sample size of subjects who had undergone PET imaging, which provides more reliable estimates than studies of small sample size.

CONCLUSION

In elderly persons, diabetes and impaired glucose metabolism in CN might contribute to neuronal injury. These factors should be examined in future longitudinal studies for their potential role in detecting onset of AD symptoms.

DISCLOSURE

The costs of publication of this article were defrayed in part by the payment of page charges. Therefore, and solely to indicate this fact, this article is hereby marked “advertisement” in accordance with 18 USC section 1734. This work was supported by NIH grants U01 AG006786, P50 AG16574, K01 AG028573, K01 MH68351, R01 AG011378, and R01 AG041851; the Robert H. and Clarice Smith and Abigail Van Buren Alzheimer’s Disease Research Program; the Center for Individualized Medicine of Mayo Clinic; the Minnesota Partnership for Biotechnology and Medical Genomics; GE Healthcare; the Elsie and Marvin Dekelboum Family Foundation; the Rochester Epidemiology Project (R01 AG034676); Abbvie Health Economics and Outcomes Research; the Walter S. and Lucienne Driskill Foundation; Cephalon, Inc.; Allon Therapeutics, Inc.; the NIA; the Alzheimer’s Association; the Mangurian Foundation; the Alexander Family Alzheimer's Disease Research Professorship; Siemens Molecular Imaging; the MN Partnership for Biotechnology and Medical Genomics; and the Leukemia & Lymphoma Society. Dr. Knopman is deputy editor for Neurology; on the data safety monitoring boards for Lilly Pharmaceuticals, Lundbeck Pharmaceuticals, and the DIAN study; a consultant to TauRx Pharmaceuticals through 2012; and an investigator for clinical trials by Baxter and Elan Pharmaceuticals, TauRx. Dr. Kantarci is on the data safety monitoring board for Takeda Global Research & Development Center, Inc. Dr. Boeve is a consultant to GE Healthcare and receives publishing royalties for The Behavioral Neurology of Dementia (Cambridge University Press, 2009). Dr. Petersen is on the scientific advisory boards for Pfizer, Inc., Janssen Alzheimer Immunotherapy, Elan Pharmaceuticals, and GE Healthcare; is a CME lecture for Novartis, Inc.; and receives royalties for Mild Cognitive Impairment (Oxford University Press, 2003). Dr. Jack is a consultant for Janssen, Bristol-Meyer-Squibb, General Electric, Siemens, and Johnson and Johnson and an investigator for Allon and Baxter, Inc., clinical trials. Dr. Lowe is on the scientific advisory boards for Bayer Schering Pharma and GE Healthcare. No other potential conflict of interest relevant to this article was reported.

Footnotes

  • Published online Mar. 20, 2014.

REFERENCES

  1. 1.↵
    1. Akomolafe A,
    2. Beiser A,
    3. Meigs JB,
    4. et al
    . Diabetes mellitus and risk of developing Alzheimer disease: results from the Framingham Study. Arch Neurol. 2006;63:1551–1555.
    OpenUrlCrossRefPubMed
  2. 2.
    1. Arvanitakis Z,
    2. Wilson RS,
    3. Bienias JL,
    4. Evans DA,
    5. Bennett DA
    . Diabetes mellitus and risk of Alzheimer disease and decline in cognitive function. Arch Neurol. 2004;61:661–666.
    OpenUrlCrossRefPubMed
  3. 3.↵
    1. Janson J,
    2. Laedtke T,
    3. Parisi JE,
    4. O'Brien P,
    5. Petersen RC,
    6. Butler PC
    . Increased risk of type 2 diabetes in Alzheimer disease. Diabetes. 2004;53:474–481.
    OpenUrlAbstract/FREE Full Text
  4. 4.
    1. MacKnight C,
    2. Rockwood K,
    3. Awalt E,
    4. McDowell I
    . Diabetes mellitus and the risk of dementia, Alzheimer's disease and vascular cognitive impairment in the Canadian Study of Health and Aging. Dement Geriatr Cogn Disord. 2002;14:77–83.
    OpenUrlCrossRefPubMed
  5. 5.↵
    1. Luchsinger JA,
    2. Tang MX,
    3. Stern Y,
    4. Shea S,
    5. Mayeux R
    . Diabetes mellitus and risk of Alzheimer's disease and dementia with stroke in a multiethnic cohort. Am J Epidemiol. 2001;154:635–641.
    OpenUrlAbstract/FREE Full Text
  6. 6.↵
    1. Debette S,
    2. Seshadri S,
    3. Beiser A,
    4. et al
    . Midlife vascular risk factor exposure accelerates structural brain aging and cognitive decline. Neurology. 2011;77:461–468.
    OpenUrlCrossRef
  7. 7.↵
    1. Arvanitakis Z,
    2. Schneider JA,
    3. Wilson RS,
    4. et al
    . Diabetes is related to cerebral infarction but not to AD pathology in older persons. Neurology. 2006;67:1960–1965.
    OpenUrlCrossRef
  8. 8.↵
    1. Marchant NL,
    2. Reed BR,
    3. Sanossian N,
    4. et al
    . The aging brain and cognition: contribution of vascular injury and aβ to mild cognitive dysfunction. JAMA Neurol. 2013;70:488–495.
    OpenUrlCrossRefPubMed
  9. 9.↵
    1. Chui HC,
    2. Zarow C,
    3. Mack WJ,
    4. et al
    . Cognitive impact of subcortical vascular and Alzheimer's disease pathology. Ann Neurol. 2006;60:677–687.
    OpenUrlCrossRefPubMed
  10. 10.↵
    1. Cherbuin N,
    2. Sachdev P,
    3. Anstey KJ
    . Higher normal fasting plasma glucose is associated with hippocampal atrophy: the PATH Study. Neurology. 2012;79:1019–1026.
    OpenUrlCrossRefPubMed
  11. 11.↵
    1. Peila R,
    2. Rodriguez BL,
    3. Launer LJ
    . Type 2 diabetes, APOE gene, and the risk for dementia and related pathologies: the Honolulu-Asia Aging Study. Diabetes. 2002;51:1256–1262.
    OpenUrlAbstract/FREE Full Text
  12. 12.↵
    1. Alafuzoff I,
    2. Aho L,
    3. Helisalmi S,
    4. Mannermaa A,
    5. Soininen H
    . Beta-amyloid deposition in brains of subjects with diabetes. Neuropathol Appl Neurobiol. 2009;35:60–68.
    OpenUrlCrossRefPubMed
  13. 13.↵
    1. Heitner J,
    2. Dickson D
    . Diabetics do not have increased Alzheimer-type pathology compared with age-matched control subjects: a retrospective postmortem immunocytochemical and histofluorescent study. Neurology. 1997;49:1306–1311.
    OpenUrlCrossRef
  14. 14.↵
    1. Duarte AI,
    2. Moreira PI,
    3. Oliveira CR
    . Insulin in central nervous system: more than just a peripheral hormone. J Aging Res. 2012;2012:384017.
    OpenUrlPubMed
  15. 15.↵
    1. Roberts RO,
    2. Geda YE,
    3. Knopman DS,
    4. et al
    . The Mayo Clinic Study of Aging: design and sampling, participation, baseline measures and sample characteristics. Neuroepidemiology. 2008;30:58–69.
    OpenUrlCrossRefPubMed
  16. 16.↵
    1. St. Sauver JL,
    2. Grossardt BR,
    3. Yawn BP,
    4. Melton LJ III.,
    5. Rocca WA
    . Use of a medical records linkage system to enumerate a dynamic population over time: the Rochester Epidemiology Project. Am J Epidemiol. 2011;173:1059–1068.
    OpenUrlAbstract/FREE Full Text
  17. 17.↵
    1. Morris JC
    . The Clinical Dementia Rating (CDR): current version and scoring rules. Neurology. 1993;43:2412–2414.
    OpenUrlCrossRefPubMed
  18. 18.↵
    1. Pfeffer RI,
    2. Kurosaki TT,
    3. Harrah CH Jr.,
    4. Chance JM,
    5. Filos S
    . Measurement of functional activities in older adults in the community. J Gerontol. 1982;37:323–329.
    OpenUrlAbstract/FREE Full Text
  19. 19.↵
    1. Kokmen E,
    2. Smith GE,
    3. Petersen RC,
    4. Tangalos E,
    5. Ivnik RC
    . The short test of mental status. Correlations with standardized psychometric testing. Arch Neurol. 1991;48:725–728.
    OpenUrlCrossRefPubMed
  20. 20.↵
    1. Ivnik RJ,
    2. Malec JF,
    3. Smith GE,
    4. et al
    . Mayo's older Americans normative studies: WAIS-R, WMS-R and AVLT norms for ages 56 through 97. Clin Neuropsychol. 1992;6:(suppl):1–30, 49–82, 83–104.
    OpenUrl
  21. 21.↵
    1. Petersen RC,
    2. Roberts RO,
    3. Knopman DS,
    4. et al
    . Prevalence of mild cognitive impairment is higher in men: the Mayo Clinic Study of Aging. Neurology. 2010;75:889–897.
    OpenUrlCrossRef
  22. 22.↵
    1. Petersen RC
    . Mild cognitive impairment as a diagnostic entity. J Intern Med. 2004;256:183–194.
    OpenUrlCrossRefPubMed
  23. 23.↵
    American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders: DSM-IV. 4th ed. Washington, DC: American Psychiatric Association; 1994.
  24. 24.↵
    Report of the expert committee on the diagnosis and classification of diabetes mellitus. Diabetes Care. 2003;26:S5–S20.
    OpenUrlCrossRefPubMed
  25. 25.↵
    1. Jack CR Jr.,
    2. Bernstein MA,
    3. Fox NC,
    4. et al
    . The Alzheimer's Disease Neuroimaging Initiative (ADNI): MRI methods. J Magn Reson Imaging. 2008;27:685–691.
    OpenUrlCrossRefPubMed
  26. 26.↵
    1. Lowe VJ,
    2. Kemp BJ,
    3. Jack CR Jr.,
    4. et al
    . Comparison of 18F-FDG and PiB PET in cognitive impairment. J Nucl Med. 2009;50:878–886.
    OpenUrlAbstract/FREE Full Text
  27. 27.↵
    1. Jack CR Jr.,
    2. Lowe VJ,
    3. Senjem ML,
    4. et al
    . 11C PiB and structural MRI provide complementary information in imaging of Alzheimer's disease and amnestic mild cognitive impairment. Brain. 2008;131:665–680.
    OpenUrlAbstract/FREE Full Text
  28. 28.↵
    1. Knopman DS,
    2. Jack CR Jr.,
    3. Wiste HJ,
    4. et al
    . Short-term clinical outcomes for stages of NIA-AA preclinical Alzheimer disease. Neurology. 2012;78:1576–1582.
    OpenUrlCrossRefPubMed
  29. 29.↵
    1. Landau SM,
    2. Harvey D,
    3. Madison CM,
    4. et al
    . Associations between cognitive, functional, and FDG-PET measures of decline in AD and MCI. Neurobiol Aging. 2011;32:1207–1218.
    OpenUrlCrossRefPubMed
  30. 30.↵
    1. Jagust WJ,
    2. Landau SM,
    3. Shaw LM,
    4. et al
    . Relationships between biomarkers in aging and dementia. Neurology. 2009;73:1193–1199.
    OpenUrlCrossRef
  31. 31.↵
    1. Landau SM,
    2. Harvey D,
    3. Madison CM,
    4. et al
    . Comparing predictors of conversion and decline in mild cognitive impairment. Neurology. 2010;75:230–238.
    OpenUrlCrossRef
  32. 32.↵
    1. Jack CR,
    2. Knopman DS,
    3. Weigand SD,
    4. et al
    . An operational approach to National Institute on Aging–Alzheimer's Association criteria for preclinical Alzheimer disease. Ann Neurol. 2012;71:765–775.
    OpenUrlCrossRefPubMed
  33. 33.↵
    1. Knopman DS,
    2. Jack CR Jr.,
    3. Wiste HJ,
    4. et al
    . Brain injury biomarkers are not dependent on β-amyloid in normal elderly. Ann Neurol. 2012;73:472–480.
    OpenUrl
  34. 34.↵
    1. Jagust WJ,
    2. Landau SM
    . Apolipoprotein E, not fibrillar beta-amyloid, reduces cerebral glucose metabolism in normal aging. J Neurosci. 2012;32:18227–18233.
    OpenUrlAbstract/FREE Full Text
  35. 35.↵
    1. Roberts RO,
    2. Geda YE,
    3. Knopman DS,
    4. et al
    . The incidence of MCI differs by subtype and is higher in men: The Mayo Clinic Study of Aging. Neurology. 2012;78:342–351.
    OpenUrlCrossRef
  36. 36.↵
    1. Baker LD,
    2. Cross DJ,
    3. Minoshima S,
    4. Belongia D,
    5. Watson GS,
    6. Craft S
    . Insulin resistance and Alzheimer-like reductions in regional cerebral glucose metabolism for cognitively normal adults with prediabetes or early type 2 diabetes. Arch Neurol. 2011;68:51–57.
    OpenUrlCrossRefPubMed
  37. 37.↵
    1. Burns CM,
    2. Chen K,
    3. Kaszniak AW,
    4. et al
    . Higher serum glucose levels are associated with cerebral hypometabolism in Alzheimer regions. Neurology. 2013;80:1557–1564.
    OpenUrlCrossRef
  38. 38.↵
    1. Petrovitch H,
    2. White LR,
    3. Izmirilian G,
    4. et al
    . Midlife blood pressure and neuritic plaques, neurofibrillary tangles, and brain weight at death: the HAAS. Honolulu-Asia Aging Study. Neurobiol Aging. 2000;21:57–62.
    OpenUrlPubMed
  39. 39.↵
    1. Qiu WQ,
    2. Folstein MF
    . Insulin, insulin-degrading enzyme and amyloid-beta peptide in Alzheimer's disease: review and hypothesis. Neurobiol Aging. 2006;27:190–198.
    OpenUrlCrossRefPubMed
  40. 40.↵
    1. Hoyer S
    . The brain insulin signal transduction system and sporadic (type II) Alzheimer disease: an update. J Neural Transm. 2002;109:341–360.
    OpenUrlCrossRefPubMed
  41. 41.↵
    1. Thambisetty M,
    2. Metter EJ,
    3. Yang A,
    4. et al
    . Glucose intolerance, insulin resistance, and pathological features of Alzheimer disease in the Baltimore Longitudinal Study of Aging. JAMA Neurol. 2013;70:1167–1172.
    OpenUrlCrossRefPubMed
  • Received for publication September 17, 2013.
  • Accepted for publication November 13, 2013.
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools
Bookmark this article

Jump to section

Related Articles

Cited By...

More in this TOC Section

Similar Articles

Keywords