Association of Chronic Low-grade Inflammation With Risk of Alzheimer Disease in ApoE4 Carriers | Dementia and Cognitive Impairment | JAMA Network Open | JAMA Network
[Skip to Navigation]
Sign In
Figure 1.  Cumulative Rates of Dementia, Alzheimer Disease (AD), and Mortality Based on ApoE Alleles and Chronic Low-grade Inflammation
Cumulative Rates of Dementia, Alzheimer Disease (AD), and Mortality Based on ApoE Alleles and Chronic Low-grade Inflammation

A, Cumulative incidence of AD, dementia, and mortality. Individuals were divided into ApoE2, ApoE3, and ApoE4 genotypes. To define the absence and presence of chronic low-grade inflammation, C-reactive protein (CRP) cutoff levels at 2 measurement points are used to represent severity. The incident rates of AD, dementia, and mortality between those without and with different severity levels of chronic low-grade inflammation were compared by using the χ2 test. To convert CRP to nanomoles per liter, multiply by 9.524. aP < .05. bP < .01. cP < .001. dP = .08. B, Cumulative incidence of AD. Individuals were divided into ApoE2, ApoE3, and ApoE4 genotypes for CRP concentrations of 3 mg/L or lower, 3 mg/L or higher, and 8 mg/L or higher. The AD incident rates were compared among the 3 ApoE subgroups by using the χ2 test for each level of CRP concentration.

Figure 2.  Kaplan-Meier Analysis for Survival Free of Alzheimer Disease (AD), Dementia, and Mortality in the Context of ApoE Alleles and Chronic Low-grade Inflammation
Kaplan-Meier Analysis for Survival Free of Alzheimer Disease (AD), Dementia, and Mortality in the Context of ApoE Alleles and Chronic Low-grade Inflammation

A, AD-free probability for ApoE2 (P = .32). B, AD-free probability for ApoE3 (P = .14). C, AD-free probability for ApoE4 (P = .009). D, Dementia-free probability for ApoE2 (P = .74). E, Dementia-free probability for ApoE3 (P = .06). F, Dementia-free probability for ApoE4 (P = .001). G, Survival probability for ApoE2 (P = .18). H, Survival probability for ApoE3 (P < .001). I, Survival probability for ApoE4 (P = .19). C-reactive protein (CRP) cutoff level of 10 mg/L or higher at a minimum of 2 time points was used to define chronic low-grade inflammation.

Table 1.  Demographic, Longitudinal CRP Measures, and Incident AD in ApoE Genotypes in the Framingham Heart Study Population
Demographic, Longitudinal CRP Measures, and Incident AD in ApoE Genotypes in the Framingham Heart Study Population
Table 2.  Cox Proportional Hazards Regression Models for the Risk of Chronic Low-grade Inflammation on the Incidence of Dementia, Alzheimer Disease, and Mortality
Cox Proportional Hazards Regression Models for the Risk of Chronic Low-grade Inflammation on the Incidence of Dementia, Alzheimer Disease, and Mortality
Table 3.  General Linear Regression Analyses of Chronic Low-grade Inflammation Effect and the Interaction Effect Between ApoE4 and Chronic Low-grade Inflammation on Brain Volumes
General Linear Regression Analyses of Chronic Low-grade Inflammation Effect and the Interaction Effect Between ApoE4 and Chronic Low-grade Inflammation on Brain Volumes
1.
Strittmatter  WJ, Saunders  AM, Schmechel  D,  et al.  Apolipoprotein E: high-avidity binding to beta-amyloid and increased frequency of type 4 allele in late-onset familial Alzheimer disease.  Proc Natl Acad Sci U S A. 1993;90(5):1977-1981. doi:10.1073/pnas.90.5.1977PubMedGoogle ScholarCrossref
2.
Tanzi  RE.  A genetic dichotomy model for the inheritance of Alzheimer’s disease and common age-related disorders.  J Clin Invest. 1999;104(9):1175-1179. doi:10.1172/JCI8593PubMedGoogle ScholarCrossref
3.
Stephensen  CB, Gildengorin  G.  Serum retinol, the acute phase response, and the apparent misclassification of vitamin A status in the third National Health and Nutrition Examination Survey.  Am J Clin Nutr. 2000;72(5):1170-1178. doi:10.1093/ajcn/72.5.1170PubMedGoogle ScholarCrossref
4.
Desikan  RS, Schork  AJ, Wang  Y,  et al; Inflammation working group, IGAP and DemGene Investigators.  Polygenic overlap between C-reactive protein, plasma lipids, and Alzheimer disease.  Circulation. 2015;131(23):2061-2069. doi:10.1161/CIRCULATIONAHA.115.015489PubMedGoogle ScholarCrossref
5.
Song  IU, Chung  SW, Kim  YD, Maeng  LS.  Relationship between the hs-CRP as non-specific biomarker and Alzheimer’s disease according to aging process.  Int J Med Sci. 2015;12(8):613-617. doi:10.7150/ijms.12742PubMedGoogle ScholarCrossref
6.
O’Bryant  SE, Waring  SC, Hobson  V,  et al.  Decreased C-reactive protein levels in Alzheimer disease.  J Geriatr Psychiatry Neurol. 2010;23(1):49-53. doi:10.1177/0891988709351832PubMedGoogle ScholarCrossref
7.
Sundelöf  J, Kilander  L, Helmersson  J,  et al.  Systemic inflammation and the risk of Alzheimer’s disease and dementia: a prospective population-based study.  J Alzheimers Dis. 2009;18(1):79-87. doi:10.3233/JAD-2009-1126PubMedGoogle ScholarCrossref
8.
Royall  DR, Al-Rubaye  S, Bishnoi  R, Palmer  RF.  Few serum proteins mediate APOE’s association with dementia.  PLoS One. 2017;12(3):e0172268. doi:10.1371/journal.pone.0172268PubMedGoogle ScholarCrossref
9.
Wilson  PW, Nam  BH, Pencina  M, D’Agostino  RB  Sr, Benjamin  EJ, O’Donnell  CJ.  C-reactive protein and risk of cardiovascular disease in men and women from the Framingham Heart Study.  Arch Intern Med. 2005;165(21):2473-2478. doi:10.1001/archinte.165.21.2473PubMedGoogle ScholarCrossref
10.
Kannel  WB, Feinleib  M, McNamara  PM, Garrison  RJ, Castelli  WP.  An investigation of coronary heart disease in families: the Framingham offspring study.  Am J Epidemiol. 1979;110(3):281-290. doi:10.1093/oxfordjournals.aje.a112813PubMedGoogle ScholarCrossref
11.
Rost  NS, Wolf  PA, Kase  CS,  et al.  Plasma concentration of C-reactive protein and risk of ischemic stroke and transient ischemic attack: the Framingham study.  Stroke. 2001;32(11):2575-2579. doi:10.1161/hs1101.098151PubMedGoogle ScholarCrossref
12.
Seshadri  S.  Elevated plasma homocysteine levels: risk factor or risk marker for the development of dementia and Alzheimer’s disease?  J Alzheimers Dis. 2006;9(4):393-398. doi:10.3233/JAD-2006-9404PubMedGoogle ScholarCrossref
13.
DeCarli  C, Massaro  J, Harvey  D,  et al.  Measures of brain morphology and infarction in the Framingham Heart Study: establishing what is normal.  Neurobiol Aging. 2005;26(4):491-510. doi:10.1016/j.neurobiolaging.2004.05.004PubMedGoogle ScholarCrossref
14.
Jefferson  AL, Himali  JJ, Beiser  AS,  et al.  Cardiac index is associated with brain aging: the Framingham Heart Study.  Circulation. 2010;122(7):690-697. doi:10.1161/CIRCULATIONAHA.109.905091PubMedGoogle ScholarCrossref
15.
DeCarli  C, Reed  T, Miller  BL, Wolf  PA, Swan  GE, Carmelli  D.  Impact of apolipoprotein E epsilon4 and vascular disease on brain morphology in men from the NHLBI twin study.  Stroke. 1999;30(8):1548-1553. doi:10.1161/01.STR.30.8.1548PubMedGoogle ScholarCrossref
16.
Miklossy  J.  Chronic inflammation and amyloidogenesis in Alzheimer’s disease—role of spirochetes.  J Alzheimers Dis. 2008;13(4):381-391. doi:10.3233/JAD-2008-13404PubMedGoogle ScholarCrossref
17.
Marottoli  FM, Katsumata  Y, Koster  KP, Thomas  R, Fardo  DW, Tai  LM.  Peripheral inflammation, apolipoprotein E4, and amyloid-β interact to induce cognitive and cerebrovascular dysfunction.  ASN Neuro. 2017;9(4):1759091417719201. doi:10.1177/1759091417719201PubMedGoogle ScholarCrossref
18.
Yarchoan  M, Louneva  N, Xie  SX,  et al.  Association of plasma C-reactive protein levels with the diagnosis of Alzheimer’s disease.  J Neurol Sci. 2013;333(1-2):9-12. doi:10.1016/j.jns.2013.05.028PubMedGoogle ScholarCrossref
19.
Schuitemaker  A, Dik  MG, Veerhuis  R,  et al.  Inflammatory markers in AD and MCI patients with different biomarker profiles.  Neurobiol Aging. 2009;30(11):1885-1889. doi:10.1016/j.neurobiolaging.2008.01.014PubMedGoogle ScholarCrossref
20.
Licastro  F, Pedrini  S, Davis  LJ,  et al.  Alpha-1-antichymotrypsin and oxidative stress in the peripheral blood from patients with probable Alzheimer disease: a short-term longitudinal study.  Alzheimer Dis Assoc Disord. 2001;15(1):51-55. doi:10.1097/00002093-200101000-00007PubMedGoogle ScholarCrossref
21.
Dik  MG, Jonker  C, Comijs  HC,  et al.  Memory complaints and APOE-epsilon4 accelerate cognitive decline in cognitively normal elderly.  Neurology. 2001;57(12):2217-2222. doi:10.1212/WNL.57.12.2217PubMedGoogle ScholarCrossref
22.
Jefferson  AL, Massaro  JM, Wolf  PA,  et al.  Inflammatory biomarkers are associated with total brain volume: the Framingham Heart Study.  Neurology. 2007;68(13):1032-1038. doi:10.1212/01.wnl.0000257815.20548.dfPubMedGoogle ScholarCrossref
23.
Gu  Y, Manly  JJ, Mayeux  RP, Brickman  AM.  An inflammation-related nutrient pattern is associated with both brain and cognitive measures in a multiethnic elderly population.  Curr Alzheimer Res. 2018;15(5):493-501. doi:10.2174/1567205015666180101145619PubMedGoogle ScholarCrossref
24.
Kamer  AR, Pirraglia  E, Tsui  W,  et al.  Periodontal disease associates with higher brain amyloid load in normal elderly.  Neurobiol Aging. 2015;36(2):627-633. doi:10.1016/j.neurobiolaging.2014.10.038PubMedGoogle ScholarCrossref
25.
Wilson  D, Peters  R, Ritchie  K, Ritchie  CW.  Latest advances on interventions that may prevent, delay or ameliorate dementia.  Ther Adv Chronic Dis. 2011;2(3):161-173. doi:10.1177/2040622310397636PubMedGoogle ScholarCrossref
26.
Wang  TJ, Nam  BH, Wilson  PW,  et al.  Association of C-reactive protein with carotid atherosclerosis in men and women: the Framingham Heart Study.  Arterioscler Thromb Vasc Biol. 2002;22(10):1662-1667. doi:10.1161/01.ATV.0000034543.78801.69PubMedGoogle ScholarCrossref
27.
D’Mello  C, Swain  MG.  Liver-brain interactions in inflammatory liver diseases: implications for fatigue and mood disorders.  Brain Behav Immun. 2014;35:9-20. doi:10.1016/j.bbi.2013.10.009PubMedGoogle ScholarCrossref
28.
Jeukendrup  AE, Vet-Joop  K, Sturk  A,  et al.  Relationship between gastro-intestinal complaints and endotoxaemia, cytokine release and the acute-phase reaction during and after a long-distance triathlon in highly trained men.  Clin Sci (Lond). 2000;98(1):47-55. PubMedGoogle ScholarCrossref
29.
Gordienko  AI.  Levels of serum antibodies to enterobacterial lipopolysaccharides and their relationship with concentration of C-reactive protein in diabetes mellitus patients  [in Ukrainian].  Ukr Biochem J. 2015;87(3):98-106. doi:10.15407/ubj87.03.098PubMedGoogle ScholarCrossref
30.
Monnet  E, Lapeyre  G, Poelgeest  EV,  et al.  Evidence of NI-0101 pharmacological activity, an anti-TLR4 antibody, in a randomized phase I dose escalation study in healthy volunteers receiving LPS.  Clin Pharmacol Ther. 2017;101(2):200-208. doi:10.1002/cpt.522PubMedGoogle ScholarCrossref
31.
Zhao  Y, Cong  L, Lukiw  WJ.  Lipopolysaccharide (LPS) accumulates in neocortical neurons of Alzheimer’s disease (AD) brain and impairs transcription in human neuronal-glial primary co-cultures.  Front Aging Neurosci. 2017;9:407. doi:10.3389/fnagi.2017.00407PubMedGoogle ScholarCrossref
32.
ADAPT-FS Research Group.  Follow-up evaluation of cognitive function in the randomized Alzheimer’s Disease Anti-inflammatory Prevention Trial and its follow-up study.  Alzheimers Dement. 2015;11(2):216-225. doi:10.1016/j.jalz.2014.03.009PubMedGoogle ScholarCrossref
Limit 200 characters
Limit 25 characters
Conflicts of Interest Disclosure

Identify all potential conflicts of interest that might be relevant to your comment.

Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.

Err on the side of full disclosure.

If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.

Not all submitted comments are published. Please see our commenting policy for details.

Limit 140 characters
Limit 3600 characters or approximately 600 words
    1 Comment for this article
    EXPAND ALL
    Immune Tolerance Disorders
    Paul Nelson, M.D., M.S. | Family Health Care, P.C.
    Finally, here is a glimmer of clear evidence for viewing the occurrence of Senile Dementia as a disorder of immune tolerance. Can we add it to the list of possible "disruptive processes" underlying the unstable-health associated with a disorder of immune tolerance? THINK: perinatal morbidity, initiating phase of new onset asthma, chronic obstructive pulmonary disease, polymyalgia rheumatica and temporal arteritis? The Nobel Prize winning team of Doctors Peter Medawar and Macfarlane Burnet at the UK University of Manchester (1940s-1960s) should be proud that their description of immune tolerance may become a broadly pervasive theme for a person's unstable-health.
    CONFLICT OF INTEREST: None Reported
    READ MORE
    Original Investigation
    Geriatrics
    October 19, 2018

    Association of Chronic Low-grade Inflammation With Risk of Alzheimer Disease in ApoE4 Carriers

    Author Affiliations
    • 1Department of Pharmacology and Experimental Therapeutics, Boston University School of Medicine, Boston, Massachusetts
    • 2Department of Anatomy and Neurobiology, Boston University School of Medicine, Boston, Massachusetts
    • 3Department of Epidemiology, Boston University School of Public Health, Boston, Massachusetts
    • 4Alzheimer’s Disease Center, University of California Davis Medical Center, Sacramento
    • 5Department of Neurology, Boston University School of Medicine, Boston, Massachusetts
    • 6Framingham Heart Study, Boston University School of Medicine, Boston, Massachusetts
    • 7Department of Psychiatry, Boston University School of Medicine, Boston, Massachusetts
    • 8Department of Pathology and Laboratory Medicine, Boston University School of Medicine, Boston, Massachusetts
    • 9Department of Pathology, Veterans Affairs Boston Healthcare System, Boston, Massachusetts
    • 10Alzheimer’s Disease Center, Boston University School of Medicine, Boston, Massachusetts
    • 11Department of Medicine, Boston University School of Medicine, Boston, Massachusetts
    • 12Department of Biostatistics, Boston University School of Public Health, Boston, Massachusetts
    JAMA Netw Open. 2018;1(6):e183597. doi:10.1001/jamanetworkopen.2018.3597
    Key Points español 中文 (chinese)

    Question  Is interaction of chronic low-grade inflammation and the apolipoprotein E genotype associated with development of Alzheimer disease?

    Findings  In this population-based cohort study, chronic low-grade inflammation, assessed by longitudinal C-reactive protein measurements, was associated with an increased risk of Alzheimer disease only in ApoE4 carriers.

    Meaning  Clinical follow-up and treatment of high levels of C-reactive protein may be beneficial for prevention of Alzheimer disease in ApoE4 carriers.

    Abstract

    Importance  The association between peripheral inflammatory biomarkers and Alzheimer disease (AD) is not consistent in the literature. It is possible that chronic inflammation, rather than 1 episode of inflammation, interacts with genetic vulnerability to increase the risk for AD.

    Objective  To study the interaction between the apolipoprotein E (ApoE) genotype and chronic low-grade inflammation and its association with the incidence of AD.

    Design, Setting, and Participants  In this cohort study, data from 2656 members of the Framingham Heart Study offspring cohort (Generation 2; August 13, 1971-November 27, 2017) were evaluated, including longitudinal measures of serum C-reactive protein (CRP), diagnoses of incident dementia including AD, and brain volume. Chronic low-grade inflammation was defined as having CRP at a high cutoff level at a minimum of 2 time points. Statistical analysis was performed from December 1, 1979, to December 31, 2015.

    Main Outcomes and Measures  Development of AD and brain volumes.

    Results  Of the 3130 eligible participants, 2656 (84.9%; 1227 men and 1429 women; mean [SD] age at last CRP measurement, 61.6 [9.5] years) with both ApoE status and longitudinal CRP measurements were included in this study analysis. Median (interquartile range) CRP levels increased with mean (SD) age (43.3 [9.6] years, 0.95 mg/L [0.40-2.35 mg/L] vs 59.1 [9.6] years, 2.04 mg/L [0.93-4.75 mg/L] vs 61.6 [9.5] years, 2.21 mg/L [1.05-5.12 mg/L]; P < .001), but less so among those with ApoE4 alleles, followed by ApoE3 then ApoE2 genotypes. During the 17 years of follow-up, 194 individuals (7.3%) developed dementia, 152 (78.4%) of whom had AD. ApoE4 coupled with chronic low-grade inflammation, defined as a CRP level of 8 mg/L or higher, was associated with an increased risk of AD, especially in the absence of cardiovascular diseases (hazard ratio, 6.63; 95% CI, 1.80-24.50; P = .005), as well as an increased risk of earlier disease onset compared with ApoE4 carriers without chronic inflammation (hazard ratio, 3.52; 95% CI, 1.27-9.75; P = .009). This phenomenon was not observed among ApoE3 and ApoE2 carriers with chronic low-grade inflammation. Finally, a subset of 1761 individuals (66.3%) underwent brain magnetic resonance imaging, and the interaction between ApoE4 and chronic low-grade inflammation was associated with brain atrophy in the temporal lobe (β = –0.88, SE = 0.22; P < .001) and hippocampus (β = –0.04, SE = 0.01; P = .005), after adjusting for confounders.

    Conclusions and Relevance  In this study, peripheral chronic low-grade inflammation in participants with ApoE4 was associated with shortened latency for onset of AD. Rigorously treating chronic systemic inflammation based on genetic risk could be effective for the prevention and intervention of AD.

    Introduction

    The apolipoprotein E4 (ApoE4 [OMIM 107741]) allele is the major genetic risk factor for late-onset Alzheimer disease (AD).1 However, not all ApoE4 carriers develop AD, even among those older than 90 years.2 It is likely that a complex interaction of genetic vulnerabilities with environmental risk factors lead to AD and identifying such factors could be beneficial for the prevention of AD. One such interacting factor could be sustained or frequent systemic inflammations, as infections of the respiratory, gastrointestinal, and urinary tract systems are common in elderly individuals.

    C-reactive protein (CRP) is an immune system response to toxins or injuries in systemic inflammation, while CRP levels increase with age.3 Although multiple AD-related genes are associated with the level of CRP,4 the association between blood CRP levels and risk of AD are not conclusive in the literature,5-7 with studies showing both low and high levels of CRP in patients with AD. Since AD is a chronic disease characterized by neurodegeneration in the brain, chronic low-grade inflammation, either sustained or frequently episodic, may be a risk factor for AD. However, most studies to date rely on one-time measurements of CRP and thus do not distinguish between a condition of acute inflammatory reaction followed by recovery and a condition of chronic inflammation without complete recovery or frequent episodic inflammation. Since preclinical studies suggest that CRP plays a role in ApoE4 leading to AD,4,8 we thus hypothesized that the association of chronic elevated CRP levels with the risk of AD would be different across ApoE genotypes. This hypothesis prompted our study that includes longitudinal measures of high levels of CRP as a biomarker of chronic low-grade inflammation to determine the risk for development of AD.

    The Framingham Heart Study is a large population-based, multigeneration cohort with long and intensive follow-up that includes multiple measurements of serum CRP taken during a 2-decade period.9 The purpose of this study was to determine if and how peripheral CRP levels are associated with the onset of AD in the context of ApoE genotypes. Included in this study is the Framingham Heart Study Generation 2 cohort enrolled in 1971, who had up to 3 CRP measurements between 1979 and 2001. Chronic low-grade inflammation was defined as meeting specified cutoff levels of plasma CRP in at least 2 measurements taken years apart. We examined the association between chronic low-grade inflammation, and risk of a diagnosis of dementia ,including AD, and brain volumes, stratified by ApoE genotype.

    Methods
    Study Design and Participants

    The Framingham Heart Study is a single-site, community-based, prospective cohort study in Framingham, Massachusetts. The design and selection criteria of the Framingham Heart Study offspring cohort (Generation 2) have been previously described.10 The source population is 3130 participants, who were 20 years or older at the second health examination (1979-1983), had baseline CRP measured during that examination, and consented to use of their genetic information (ie, ApoE genotype). Excluded were 404 individuals who did not have another CRP measurement at the sixth (1995-1998) or seventh (1998-2001) health examinations and 46 individuals with the ApoE2/4 genotype. In addition, 24 individuals with prevalent dementia at the time of each CRP measurement were excluded. Thus, the final study sample consisted of 2656 individuals (eFigure 1A in the Supplement), whose data were used for the primary analyses to examine the association between ApoE, CRP level, and risk of AD. Data on a subset of 1785 individuals, who also underwent brain magnetic resonance imaging (MRI) after the seventh examination (1999-2011) (eFigure 1B in the Supplement), were used for secondary analyses to examine the association between CRP, ApoE, and AD-related changes in brain structure. Cardiovascular disease status at examination 7 was used as a covariate and was represented as a dichotomous variable (yes or no), determined by the presence of the following conditions: myocardial infarction, angina pectoris, coronary insufficiency, congestive heart failure, and intermittent claudication. Written informed consent was obtained from all study participants and the study protocol was approved by the Institutional Review Board of Boston University Medical Campus. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline. This study was monitored by a National Heart, Lung, and Blood Institute Observational Study Monitoring Board and followed their guidelines.

    CRP Measurement

    During the clinic visits, blood samples were drawn, under fasting condition, from the antecubital vein while participants were supine. Serum aliquots were frozen at –20°C after the initial phlebotomy and subsequently thawed for measurement of high-sensitivity CRP. Both CRP concentrations at examinations 2 and 6 were performed in the Framingham Heart Study laboratory using a previously described enzymatic immunoassay11 (Hemagen Diagnostics Inc). The CRP measurement at examination 7 was conducted by a Dade Behring BN100 nephelometer. The Pearson product moment correlation between both techniques was 0.98 and the CRP quartile assignment was identical.9

    To define chronic low-grade inflammation, we used the following 2 criteria: (1) since a CRP level lower than 3 mg/L (to convert to nanomoles per liter, multiply by 9.524) is considered normal in a clinical setting, we defined low-grade inflammation as any CRP measurement above 3 mg/L and used different cutoff levels to indicate severity; and (2) chronic inflammatory status was defined as having at least 2 longitudinal CRP measurements above the stipulated cutoff levels. Participants with only 1 CRP measurement equal to or higher than 3 mg/L were not considered as having chronic low-grade inflammation.

    Diagnoses of Dementia Including AD

    Beginning in 1979, all Generation 2 participants have been followed up for incident dementia. The Mini-Mental State Examination (MMSE) was administered beginning at the fifth health examination (ie, 1991-1995) to monitor change in cognitive status. A performance drop in Mini-Mental State Examination score of 3 or more points from the immediately preceding examination or 5 or more points across all examinations would indicate a change in cognitive status that warranted review by a dementia diagnostic panel consisting of at least 1 neurologist and 1 neuropsychologist. Furthermore, from 1999 to 2005, all surviving Generation 2 participants were invited for an in-depth cognitive examination, which also screened for incident cognitive impairment that warranted review by the dementia diagnostic panel. Consensus diagnostic procedures have been previously described.12 Incidence of dementia, AD, and person-time accruement after the last longitudinal CRP measurement was used for analyses.

    Brain Measurements

    The brain MRI was conducted beginning March 1999; only data acquired from brain MRI scans after the measurement of the last CRP (eg, examinations 6 or 7) were used. The brain MRI protocol has been reported in detail elsewhere.13,14 A Siemens 1-T MR machine (Siemens Medical) with a T2-weighted double spin-echo coronal imaging sequence was used. A central laboratory blinded to demographic and clinical information processed the digital information on brain images and quantified the brain data with a custom-written computer program operating on a UNIX, Solaris platform (Sun Microsystems).

    The semiautomated segmentation protocol for quantifying total cranial volume, total cerebral brain volume, frontal lobar brain volume, parietal lobe brain volume, temporal lobe brain volume, and hippocampal volume has been described elsewhere,15 as have the interrater reliabilities for these methods. For segmentation of white matter hyperintensities from other brain tissues, the first and second images from T2 sequences were summed and a log-normal distribution was fitted to the summed data. A segmentation threshold for white matter hyperintensities was determined as 1 SD in pixel intensity greater than the mean of the fitted distribution of brain parenchyma. The units for the brain volumes, including total cerebral brain volume, frontal lobar brain volume, temporal lobe brain volume, hippocampal volume, and white matter hyperintensities, were computed as the percentage of total cranial volume. Each image set underwent rigorous quality control assessment that includes assessment of the original acquisition quality as well as the quality of the image processing. Moreover, each of the analysts was highly trained to maintain rigorous precision, with intraclass (analyst) coefficients above 90% for all analyses.

    Statistical Analysis

    Statistical analysis was performed from December 1, 1979, to December 31, 2015. Analyses were performed using SAS software, version 9.3 (SAS Institute) and the R statistical environment (The R Foundation for Statistical Computing Platform 2017). We performed univariate analyses to describe baseline characteristics of the final sample population, stratified by ApoE genotype (ApoE2-2/2 or 2/3; ApoE3-3/3; ApoE4-3/4 or 4/4). Means and SDs were determined and analysis of variance tests were conducted on variables with normal distribution, Mann-Whitney tests were performed on variables with a skewed distribution using the median (interquartile range), and χ2 tests were used for categorical variables using number and percentage.

    To establish the temporal sequence between chronic low-grade inflammation and the development of AD, we excluded individuals who did not have ApoE genotype measured, had an ApoE2/4 genotype, did not have longitudinal CRP measurements, and/or had dementia at the time of or prior to the last measurement of CRP, leaving a total sample of 2656 (eFigure 1 in the Supplement). Cox proportional hazards regression models were used to examine chronic low-grade inflammation and the development of AD, dementia, or mortality after adjusting for age, sex, educational level, and cardiovascular disease; the interaction between ApoE4 and chronic low-grade inflammation was examined as well. Kaplan-Meier survival analyses were performed to compare the onset of AD, dementia, or death among ApoE genotypes.

    The secondary analysis focused on 1761 individuals who underwent a brain MRI after their last longitudinal CRP measurement (eFigure 1 in the Supplement). Multivariate linear regression was also used to study the association between chronic low-grade inflammation and total and regional brain volume measures, controlling for total brain volume, after adjusting for age (age groups, 20-29 years; 30-39 years, 40-49 years, and ≥50 years), sex, educational level, ApoE4 status, and the time between the last CRP measurement and brain MRI scan. The interaction of the CRP status and ApoE4 allele was also examined in these regression models for each brain measure. Given multiple testing and the need to minimize the rate of false positives, P values were adjusted using a conventional Bonferroni correction threshold of .005 in 2-sided t tests (eTable in the Supplement).

    Results
    Association of Median Levels of CRP With Age in the Context of ApoE Genotypes

    The 2656 individuals who completed at least 2 CRP measurements and did not have dementia at the time of their last CRP measurement were a mean (SD) age of 61.4 (9.4) years (eFigure 1 in the Supplement). Participants were further categorized into the following 3 groups: ApoE2 (n = 364), ApoE3 (n = 1729), and ApoE4 (n = 532) (Table 1). There were no differences in age, sex, educational levels, and prevalence of cardiovascular diseases among the ApoE subgroups.

    There was a positive association between increasing mean (SD) age and higher median (interquartile range) CRP levels (43.3 [9.6] years, 0.95 mg/L [0.40-2.35 mg/L] vs 59.1 [9.6] years, 2.04 mg/L [0.93-4.75 mg/L] vs 61.6 [9.5] years, 2.21 mg/L [1.05-5.12 mg/L]; P < .001) and a higher proportion of older participants had CRP levels of 3 mg/L or higher across all ApoE genotypes (Table 1). ApoE4 carriers consistently had lower levels of CRP than did ApoE2 and ApoE3 carriers at each of the examinations. Overall, age was positively associated with chronic low-grade systemic inflammation, while ApoE4 was negatively associated (Table 1).

    Association of the Interaction of ApoE4 and Chronic Low-Grade Inflammation With Increased Risk of AD

    The mean (SD) follow-up from the last CRP measurement was approximately 14.1 (4.2) years and there were no significant differences in follow-up time among the 3 ApoE groups (Table 1). Of the 194 individuals (7.3%) who developed dementia, 152 (78.4%) were diagnosed with AD. Using χ2 analyses, we tested whether the association between ApoE genotype and chronic low-grade inflammation status with the risk for incident dementia including AD and mortality changed when using different CRP cutoff levels (Figure 1). Those with both the ApoE4 allele and chronic low-grade inflammation demonstrated a CRP level–dependent pattern that was linked to increased risk of AD and dementia (Figure 1A). This phenomenon was not observed in ApoE2 and ApoE3 carriers. ApoE4 carriers with CRP levels of 3 mg/L or less had a onefold to twofold increase of risk of AD compared with ApoE3 and ApoE2 carriers with similar CRP levels (37 of 410 [9.0%] vs 58 of 1174 [4.9%] vs 8 of 245 [3.3%]; P = .003; Table 1 and Figure 1B). For CRP levels of 3 mg/L or higher, ApoE4 carriers had a twofold to threefold increase of risk of AD compared with ApoE3 and ApoE2 carriers (18 of 132 [13.6%] vs 25 of 573 [4.4%] vs 6 of 122 [4.9%]; P = .001). For CRP levels of 8 mg/L or higher, ApoE4 carriers had a fivefold to 10-fold increase of risk of AD risk compared with ApoE3 and ApoE2 carriers (7 of 27 [25.9%] vs 7 of 132 [5.3%] vs 1 of 42 [2.4%]; P = .002). In contrast, as expected, chronic low-grade inflammation was positively associated with or showed a positive trend with mortality rates across all ApoE genotypes (Figure 1).

    We next used Cox proportional hazards regression analyses, adjusted for age, sex, educational level, cardiovascular disease, and ApoE4 status. Although chronic low-grade inflammation, based on a CRP level of 8 mg/L or higher, 9 mg/L or higher, or 10 mg/L or higher, was not found to be associated with AD or dementia, the interaction between ApoE4 and chronic low-grade inflammation was associated with AD or dementia (CRP level ≥9 mg/L: hazard ratio, 3.47; 95% CI, 1.10-10.94; P = .03) or tended to be positively associated with AD. To determine if cardiovascular diseases may account for this association, we excluded individuals with cardiovascular diseases and found that, in the absence of cardiovascular diseases, the association was significantly stronger (hazard ratio, 6.89; 95% CI, 1.74-27.30; P = .006) (Table 2). In addition, after stratifying participants by ApoE4 status, we found that chronic low-grade inflammation was significantly associated with AD risk only in ApoE4 carriers (hazard ratio, 4.70; 95% CI, 1.83-12.04; P = .001), but not in ApoE2 or ApoE3 carriers. Again, chronic low-grade inflammation was positively associated with mortality across all ApoE genotypes, but the interaction between ApoE4 and chronic low-grade inflammation was not associated with mortality (Table 2).

    Interaction of ApoE4 and Chronic Low-grade Inflammation and Latency of Onset of AD

    Using Kaplan-Meier analysis, we found that individuals with ApoE4 and chronic low-grade inflammation, defined as having a CRP cutoff level of 10 mg/L or more in at least 2 examinations, was more strongly associated with onset of dementia as well as AD compared with ApoE4 carriers without this level of inflammation. In comparison, chronic low-grade inflammation status did not affect risk of dementia including AD in ApoE2 carriers. Results for ApoE3 carriers fell between findings for ApoE2 and ApoE4 carriers, showing marginal influences of chronic low-grade inflammation on risk of AD. We also tested cutoff CRP levels of 8 mg/L or higher and 9 mg/L or higher to examine severity of chronic low-grade inflammation and found that they were all linked to onset of AD in the Kaplan-Meier analysis for carriers of ApoE4, but not for carriers of ApoE2 and ApoE3 (eFigure 2 in the Supplement). Again, chronic low-grade inflammation was significantly associated with a low survival rate in ApoE3 carriers and showed this same trend in ApoE2 and ApoE4 carriers (Figure 2).

    Association of the Interaction of ApoE4 and Chronic Low-grade Inflammation With Brain Atrophy

    The above findings suggest that chronic systemic inflammation may make the brain more vulnerable to AD. To test this hypothesis, we used brain volumetric measures acquired after the last CRP measurement from 1761 participants. The mean (SD) time from the first CRP measurement to the brain MRI scan was 24.0 (3.1) years, mean (SD) time from the second CRP measurement to the brain MRI scan was 8.5 (3.1) years, and mean (SD) time from the third CRP measurement to the brain MRI scan was 5.6 (3.0) years. The brain volumes of those without (n = 1647) and with (n = 114) chronic low-grade inflammation, defined by cutoff CRP levels of 8 mg/L or higher at a minimum of 2 examinations, were compared (eTable in the Supplement). With 1 exception, we found no differences in brain regions between those without and with chronic low-grade inflammation after Bonferroni corrections. There was a significant difference between the 2 groups for total white matter volume. These associations held for multivariate linear regressions that adjusted for age, sex, educational level, ApoE4 status, and the time between CRP and brain MRI measures (Table 3). The interaction of ApoE4 and chronic low-grade inflammation, however, was negatively associated with the regions associated with AD pathologic characteristics (eg, temporal lobe brain volume: β = –0.78, SE = 0.25; P = .01) and persisted after adjusting for age and sex (temporal lobe brain volume: β = –0.85, SE = 0.22; P < .001) as well as after adjusting for age, sex, time to brain MRI, educational level, and cardiovascular disease (temporal lobe brain volume: β = –0.88, SE = 0.22; P < .001). This same association was found for hippocampal volume in the model adjusted for age, sex, time to brain MRI, educational level, and cardiovascular disease (β = –0.04, SE = 0.01; P = .005). No other significant associations were found for other brain regions.

    Discussion

    C-reactive protein is a biomarker of low-grade inflammation. To our knowledge, this is the first study to use longitudinal measurements of CRP to define a chronic condition of low-grade inflammation at baseline and demonstrate that ApoE4 interacting with chronic low-grade inflammation increased the risk of AD and shortened the latency for developing AD (Figure 1 and Figure 2). Since it is well documented that infection and inflammation are common in elderly individuals, and preclinical studies have reported that inflammation induced AD pathologic characteristics in mice who only carried ApoE4,16,17 our findings may explain why ApoE4 carriers have increased risk for AD at an old age and suggest that treating chronic low-grade inflammation may delay the onset of AD in ApoE4 carriers.

    The strength of this study was its longitudinal follow-up for incident cases of dementia that was preceded by multiple measurements of CRP with an interval between CRP measurements of 6 to 16 years (Table 1). This study offsets the limitation of earlier studies that relied on onetime measurement of CRP to study the development of AD that has resulted in reports of positive,5 negative,6,18 and no association7,19-21 between CRP and AD. Although these studies cannot distinguish between those who had an elevated CRP level and then recovered vs those who had a sustained or multiple episodically elevated CRP levels, based on our data analyses we hypothesized that genetic vulnerability for AD might be associated with a long-term low-grade inflammatory condition, albeit either sustained or episodic.

    Although genetic risk factors such as ApoE4 for AD are present across the lifespan, disease onset does not occur until later in life.2 Our findings suggest that chronic low-grade inflammation interacts with ApoE4 to accelerate the onset of AD in a pattern dependent on the CRP level (Figure 1 and Figure 2; eFigure 2 in the Supplement). Although ApoE2 carriers had higher levels of CRP with increasing age than did ApoE3 and ApoE4 carriers (Table 1), high CRP levels were not associated with risk of AD among ApoE2 carriers (Figure 1 and Figure 2). It is probable that chronic low-grade inflammation linked with ApoE4 puts the brain into a vulnerable state for the development of AD (Table 3), but the brain of ApoE2 carriers is resilient to the influence of chronic low-grade inflammation on the development of AD. This possibility is consistent with other studies that report other proinflammatory factors associated with brain atrophy.22,23 Another study found that periodontal disease, another common inflammatory condition in elderly individuals, is associated with a higher brain amyloid load detected by amyloid positron emission tomographic scan in healthy elderly individuals.24 Together, these studies suggest a possible link between systemic infection and AD pathologic characteristics in the brain in humans. We propose that if chronic low-grade inflammation is detected through follow-up CRP measurements and is treated among elderly individuals who are ApoE4 carriers, the onset of AD can be delayed or even prevented, since studies have reported that delaying onset by 5 years can reduce the risk for AD by nearly 50%.25

    The association between chronic low-grade inflammation and risk of AD for ApoE4 carriers became even more significant in the absence of cardiovascular diseases (Table 2). Since CRP levels are linked to cardiovascular disease,26 which is also a risk factor for AD, our results indicate that chronic low-grade inflammation may play an early role leading to AD in ApoE4 carriers4,8 independent from cardiovascular diseases. Acute inflammatory reaction to infection or injury is a physiological process that is a defense mechanism of the body and is marked by an elevation of CRP levels; however, chronic low-grade inflammation is a pathologic process that may lead to chronic diseases such as cardiovascular disease.26 Although chronic low-grade inflammation was linked to high rates of mortality across all ApoE genotypes, an increased risk of AD was found only in ApoE4 carriers (Table 2, Figure 1, and Figure 2).

    The mechanism for the interaction between ApoE4 and a high sustained level of CRP that leads to an increased risk of AD is unknown. Both ApoE and CRP are produced mainly by the liver, implying a liver-brain inflammation axis for the pathogenesis of AD. A liver-brain inflammation axis has been proposed to cause abnormal clinical symptoms in brain diseases, including mood diseases, cognition diseases, and neurovegetative signs,27 but its association with AD remains unclear. Gram-negative bacteria often cause infection in elderly individuals, including in the gastrointestinal, respiratory, and urinary systems. Injection of the gram-negative bacterial cell wall component lipopolysaccharides can increase levels of CRP in the blood,28-30 implying that the level of CRP could be a biomarker after attacks from bacteria endotoxin lipopolysaccharides. Systemic administration of lipopolysaccharides into AD mouse models induced AD pathologic characteristics in the brain.16 It has been shown that lipopolysaccharide challenge is linked to cerebrovascular pathologic findings and increased amyloid burden in an ApoE4 AD mouse model, but not in a non-ApoE4 AD mouse model.17 One recent study found that the amount of lipopolysaccharides in the brains of humans with AD are twofold higher compared with control brains.31

    Limitations

    Although a strength of this study is the longitudinal measurements of CRP, a limitation is that there were not more frequent, preferably annual, CRP measures; thus, it is possible that some cases of sustained inflammatory status may have been misclassified or missed. It is also possible that we have underestimated the interactive association of ApoE4 and chronic low-grade inflammation with AD. Furthermore, the Framingham Heart Study cohort lacks ethnic diversity and thus these findings lack generalizability to nonwhite populations.

    Conclusions

    As systemic infection and inflammatory attacks are common in elderly individuals, recovery of the immune system to baseline could be critical for certain genotypes such as ApoE4 for the sequela of AD development. Evidence of chronic low-grade inflammation stage could be targeted for personalized treatment. Our findings provide initial evidence of the importance of ApoE genotype in clinical trial studies of anti-inflammatory drugs for AD. Although previous clinical trials of anti-inflammatory drugs for AD have failed,32 specifically targeting a subset of patients based on ApoE genotypes and inflammation status may be an important consideration for future clinical trial study design. Additional studies are warranted to determine whether rigorous treatment of infection and inflammation that lower CRP levels to a normal level will attenuate the risk of AD for ApoE4 carriers.

    Back to top
    Article Information

    Accepted for Publication: August 21, 2018.

    Published: October 19, 2018. doi:10.1001/jamanetworkopen.2018.3597

    Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2018 Tao Q et al. JAMA Network Open.

    Corresponding Authors: Wei Qiao Qiu, MD, PhD (wqiu67@bu.edu), and Rhoda Au, PhD (rhodaau@bu.edu), Boston University School of Medicine, 72 E Concord St, Ste R-623D, Boston, MA 02118.

    Author Contributions: Drs Tao and Ang had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Drs Tao and Ang contributed equally to the study.

    Concept and design: Tao, Ang, Qiu.

    Acquisition, analysis, or interpretation of data: All authors.

    Drafting of the manuscript: Tao, Ang, Qiu.

    Critical revision of the manuscript for important intellectual content: All authors.

    Statistical analysis: Tao, Ang, Zhang, Massaro, Qiu.

    Obtained funding: Au, Qiu.

    Administrative, technical, or material support: Ang, DeCarli, Auerbach, Devine, Au.

    Supervision: Devine, Stein, Qiu.

    Conflict of Interest Disclosures: None reported.

    Funding/Support: This work was supported by contract N01-HC-25195 from the National Heart, Lung, and Blood Institute, grant NS-17950 from the National Institute of Neurological Disorders and Stroke, and grants AG-008122, AG-16495, and AG-022476 from the National Institute on Aging.

    Role of the Funder/Sponsor: The funding sources had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

    Additional Contributions: We thank the Framingham Heart Study participants for their decades of dedication and the Framingham Heart Study staff for their hard work in collecting and preparing the data.

    References
    1.
    Strittmatter  WJ, Saunders  AM, Schmechel  D,  et al.  Apolipoprotein E: high-avidity binding to beta-amyloid and increased frequency of type 4 allele in late-onset familial Alzheimer disease.  Proc Natl Acad Sci U S A. 1993;90(5):1977-1981. doi:10.1073/pnas.90.5.1977PubMedGoogle ScholarCrossref
    2.
    Tanzi  RE.  A genetic dichotomy model for the inheritance of Alzheimer’s disease and common age-related disorders.  J Clin Invest. 1999;104(9):1175-1179. doi:10.1172/JCI8593PubMedGoogle ScholarCrossref
    3.
    Stephensen  CB, Gildengorin  G.  Serum retinol, the acute phase response, and the apparent misclassification of vitamin A status in the third National Health and Nutrition Examination Survey.  Am J Clin Nutr. 2000;72(5):1170-1178. doi:10.1093/ajcn/72.5.1170PubMedGoogle ScholarCrossref
    4.
    Desikan  RS, Schork  AJ, Wang  Y,  et al; Inflammation working group, IGAP and DemGene Investigators.  Polygenic overlap between C-reactive protein, plasma lipids, and Alzheimer disease.  Circulation. 2015;131(23):2061-2069. doi:10.1161/CIRCULATIONAHA.115.015489PubMedGoogle ScholarCrossref
    5.
    Song  IU, Chung  SW, Kim  YD, Maeng  LS.  Relationship between the hs-CRP as non-specific biomarker and Alzheimer’s disease according to aging process.  Int J Med Sci. 2015;12(8):613-617. doi:10.7150/ijms.12742PubMedGoogle ScholarCrossref
    6.
    O’Bryant  SE, Waring  SC, Hobson  V,  et al.  Decreased C-reactive protein levels in Alzheimer disease.  J Geriatr Psychiatry Neurol. 2010;23(1):49-53. doi:10.1177/0891988709351832PubMedGoogle ScholarCrossref
    7.
    Sundelöf  J, Kilander  L, Helmersson  J,  et al.  Systemic inflammation and the risk of Alzheimer’s disease and dementia: a prospective population-based study.  J Alzheimers Dis. 2009;18(1):79-87. doi:10.3233/JAD-2009-1126PubMedGoogle ScholarCrossref
    8.
    Royall  DR, Al-Rubaye  S, Bishnoi  R, Palmer  RF.  Few serum proteins mediate APOE’s association with dementia.  PLoS One. 2017;12(3):e0172268. doi:10.1371/journal.pone.0172268PubMedGoogle ScholarCrossref
    9.
    Wilson  PW, Nam  BH, Pencina  M, D’Agostino  RB  Sr, Benjamin  EJ, O’Donnell  CJ.  C-reactive protein and risk of cardiovascular disease in men and women from the Framingham Heart Study.  Arch Intern Med. 2005;165(21):2473-2478. doi:10.1001/archinte.165.21.2473PubMedGoogle ScholarCrossref
    10.
    Kannel  WB, Feinleib  M, McNamara  PM, Garrison  RJ, Castelli  WP.  An investigation of coronary heart disease in families: the Framingham offspring study.  Am J Epidemiol. 1979;110(3):281-290. doi:10.1093/oxfordjournals.aje.a112813PubMedGoogle ScholarCrossref
    11.
    Rost  NS, Wolf  PA, Kase  CS,  et al.  Plasma concentration of C-reactive protein and risk of ischemic stroke and transient ischemic attack: the Framingham study.  Stroke. 2001;32(11):2575-2579. doi:10.1161/hs1101.098151PubMedGoogle ScholarCrossref
    12.
    Seshadri  S.  Elevated plasma homocysteine levels: risk factor or risk marker for the development of dementia and Alzheimer’s disease?  J Alzheimers Dis. 2006;9(4):393-398. doi:10.3233/JAD-2006-9404PubMedGoogle ScholarCrossref
    13.
    DeCarli  C, Massaro  J, Harvey  D,  et al.  Measures of brain morphology and infarction in the Framingham Heart Study: establishing what is normal.  Neurobiol Aging. 2005;26(4):491-510. doi:10.1016/j.neurobiolaging.2004.05.004PubMedGoogle ScholarCrossref
    14.
    Jefferson  AL, Himali  JJ, Beiser  AS,  et al.  Cardiac index is associated with brain aging: the Framingham Heart Study.  Circulation. 2010;122(7):690-697. doi:10.1161/CIRCULATIONAHA.109.905091PubMedGoogle ScholarCrossref
    15.
    DeCarli  C, Reed  T, Miller  BL, Wolf  PA, Swan  GE, Carmelli  D.  Impact of apolipoprotein E epsilon4 and vascular disease on brain morphology in men from the NHLBI twin study.  Stroke. 1999;30(8):1548-1553. doi:10.1161/01.STR.30.8.1548PubMedGoogle ScholarCrossref
    16.
    Miklossy  J.  Chronic inflammation and amyloidogenesis in Alzheimer’s disease—role of spirochetes.  J Alzheimers Dis. 2008;13(4):381-391. doi:10.3233/JAD-2008-13404PubMedGoogle ScholarCrossref
    17.
    Marottoli  FM, Katsumata  Y, Koster  KP, Thomas  R, Fardo  DW, Tai  LM.  Peripheral inflammation, apolipoprotein E4, and amyloid-β interact to induce cognitive and cerebrovascular dysfunction.  ASN Neuro. 2017;9(4):1759091417719201. doi:10.1177/1759091417719201PubMedGoogle ScholarCrossref
    18.
    Yarchoan  M, Louneva  N, Xie  SX,  et al.  Association of plasma C-reactive protein levels with the diagnosis of Alzheimer’s disease.  J Neurol Sci. 2013;333(1-2):9-12. doi:10.1016/j.jns.2013.05.028PubMedGoogle ScholarCrossref
    19.
    Schuitemaker  A, Dik  MG, Veerhuis  R,  et al.  Inflammatory markers in AD and MCI patients with different biomarker profiles.  Neurobiol Aging. 2009;30(11):1885-1889. doi:10.1016/j.neurobiolaging.2008.01.014PubMedGoogle ScholarCrossref
    20.
    Licastro  F, Pedrini  S, Davis  LJ,  et al.  Alpha-1-antichymotrypsin and oxidative stress in the peripheral blood from patients with probable Alzheimer disease: a short-term longitudinal study.  Alzheimer Dis Assoc Disord. 2001;15(1):51-55. doi:10.1097/00002093-200101000-00007PubMedGoogle ScholarCrossref
    21.
    Dik  MG, Jonker  C, Comijs  HC,  et al.  Memory complaints and APOE-epsilon4 accelerate cognitive decline in cognitively normal elderly.  Neurology. 2001;57(12):2217-2222. doi:10.1212/WNL.57.12.2217PubMedGoogle ScholarCrossref
    22.
    Jefferson  AL, Massaro  JM, Wolf  PA,  et al.  Inflammatory biomarkers are associated with total brain volume: the Framingham Heart Study.  Neurology. 2007;68(13):1032-1038. doi:10.1212/01.wnl.0000257815.20548.dfPubMedGoogle ScholarCrossref
    23.
    Gu  Y, Manly  JJ, Mayeux  RP, Brickman  AM.  An inflammation-related nutrient pattern is associated with both brain and cognitive measures in a multiethnic elderly population.  Curr Alzheimer Res. 2018;15(5):493-501. doi:10.2174/1567205015666180101145619PubMedGoogle ScholarCrossref
    24.
    Kamer  AR, Pirraglia  E, Tsui  W,  et al.  Periodontal disease associates with higher brain amyloid load in normal elderly.  Neurobiol Aging. 2015;36(2):627-633. doi:10.1016/j.neurobiolaging.2014.10.038PubMedGoogle ScholarCrossref
    25.
    Wilson  D, Peters  R, Ritchie  K, Ritchie  CW.  Latest advances on interventions that may prevent, delay or ameliorate dementia.  Ther Adv Chronic Dis. 2011;2(3):161-173. doi:10.1177/2040622310397636PubMedGoogle ScholarCrossref
    26.
    Wang  TJ, Nam  BH, Wilson  PW,  et al.  Association of C-reactive protein with carotid atherosclerosis in men and women: the Framingham Heart Study.  Arterioscler Thromb Vasc Biol. 2002;22(10):1662-1667. doi:10.1161/01.ATV.0000034543.78801.69PubMedGoogle ScholarCrossref
    27.
    D’Mello  C, Swain  MG.  Liver-brain interactions in inflammatory liver diseases: implications for fatigue and mood disorders.  Brain Behav Immun. 2014;35:9-20. doi:10.1016/j.bbi.2013.10.009PubMedGoogle ScholarCrossref
    28.
    Jeukendrup  AE, Vet-Joop  K, Sturk  A,  et al.  Relationship between gastro-intestinal complaints and endotoxaemia, cytokine release and the acute-phase reaction during and after a long-distance triathlon in highly trained men.  Clin Sci (Lond). 2000;98(1):47-55. PubMedGoogle ScholarCrossref
    29.
    Gordienko  AI.  Levels of serum antibodies to enterobacterial lipopolysaccharides and their relationship with concentration of C-reactive protein in diabetes mellitus patients  [in Ukrainian].  Ukr Biochem J. 2015;87(3):98-106. doi:10.15407/ubj87.03.098PubMedGoogle ScholarCrossref
    30.
    Monnet  E, Lapeyre  G, Poelgeest  EV,  et al.  Evidence of NI-0101 pharmacological activity, an anti-TLR4 antibody, in a randomized phase I dose escalation study in healthy volunteers receiving LPS.  Clin Pharmacol Ther. 2017;101(2):200-208. doi:10.1002/cpt.522PubMedGoogle ScholarCrossref
    31.
    Zhao  Y, Cong  L, Lukiw  WJ.  Lipopolysaccharide (LPS) accumulates in neocortical neurons of Alzheimer’s disease (AD) brain and impairs transcription in human neuronal-glial primary co-cultures.  Front Aging Neurosci. 2017;9:407. doi:10.3389/fnagi.2017.00407PubMedGoogle ScholarCrossref
    32.
    ADAPT-FS Research Group.  Follow-up evaluation of cognitive function in the randomized Alzheimer’s Disease Anti-inflammatory Prevention Trial and its follow-up study.  Alzheimers Dement. 2015;11(2):216-225. doi:10.1016/j.jalz.2014.03.009PubMedGoogle ScholarCrossref
    ×