Associations Between Serum Inflammatory Markers and Hippocampal Volume in a Community Sample

Mike F. Schmidt, Kevin B. Freeman, Beverly G. Windham, Michael E. Griswold, Iftikhar J. Kullo, Stephen T. Turner, Thomas H. Mosley

Research output: Contribution to journalArticlepeer-review

10 Scopus citations


Objectives: To quantify associations between inflammatory biomarkers and hippocampal volume (HV) and to examine effect modification according to sex, race, and age. Design: Cross-sectional analyses using generalized estimating equations to account for familial clustering; standardized β-coefficients adjusted for age, sex, race, and education. Setting: Community cohorts in Jackson, Mississippi and Rochester, Minnesota. Participants: The Genetic Epidemiology Network of Arteriopathy study. Measurements: C-reactive protein (CRP), interleukin-6 (IL-6), and soluble tumor necrosis factor receptors 1 (sTNFR-1) and 2 (sTNFR-2) from peripheral blood were measured in a sample of 773 non-Hispanic whites (61% women, aged 60.2 ± 9.8) and 514 African Americans (70% women, aged 63.9 ± 8.1) who also underwent brain magnetic resonance imaging. Biomarkers were standardized and compared according to sex, race and age with HV. Results: In the full sample, higher sTNFR-1 and sTNFR-2 were associated with smaller HV. Each standard deviation (SD) increase in sTNFR-1 was associated with 59.1 mm3 (95% confidence interval (CI) = −101.4 to −16.7 mm3) smaller HV and each SD increase in sTNFR-2 associated with 48.8 mm3 (95% CI = −92.2 to −5.3 mm3) smaller HV. Relationships were stronger for sTNFR-2 in men (HV = −116.6 mm3 for each SD increase, 95% CI = −201.0 to −32.1) than women (HV = −26.0 per SD increase, 95% CI = −72.4–20.5) and sTNFR-1 in non-Hispanic whites (HV = −84.7 mm3 per SD increase, 95% CI = −142.2 to −27.1) than African Americans (HV = −14.1 mm3 per SD increase, 95% CI = −78.3–50.1). Associations between IL-6 or CRP and HV were not supported. Conclusion: Higher levels of sTNFRs were associated cross-sectionally with smaller hippocampi. Longitudinal data are needed to determine whether these biomarkers may help to identify risk of late-life cognitive impairment.

Original languageEnglish (US)
Pages (from-to)1823-1829
Number of pages7
JournalJournal of the American Geriatrics Society
Issue number9
StatePublished - Sep 1 2016
Externally publishedYes


  • TNF α
  • hippocampus
  • inflammation

ASJC Scopus subject areas

  • Geriatrics and Gerontology


Dive into the research topics of 'Associations Between Serum Inflammatory Markers and Hippocampal Volume in a Community Sample'. Together they form a unique fingerprint.

Cite this