Association of C-reactive protein with brain micro- and macro-structure among older adult men

The process by which aging leads to increased risk for Alzheimer's disease and related dementias is not entirely understood, but one hypothesized contributor is the occurrence of low-grade inflammation in older age. Associations between peripheral C-reactive protein (CRP), a marker of systemic inflammation, and brain structure have been widely studied, but fewer studies have examined CRP in relation to diffusion measures, particularly using newer techniques such as restriction spectrum imaging (RSI). In the current study, we examined how high sensitivity CRP (hsCRP) relates to diffusion metrics and global brain tissue volumes among a group of older adult men.

We analyzed a sample of 372 cognitively unimpaired men from VETSA, who were assessed at average age 67 for plasma hsCRP and underwent diffusion and structural brain imaging. Linear mixed models examined associations of hsCRP with global and regional measures of restricted normalized directional (RND) and free normalized isotropic (FNI) diffusion in white matter and hindered normalized total diffusion (HNT) and FNI diffusion in gray matter derived from RSI. Similarly, the relationship of hsCRP to global and regional fractional anisotropy (FA) in white matter and mean diffusivity (MD) in white and gray matter was examined. Finally, we examined hsCRP relationships with global gray and white matter volumes as well as global abnormal white matter (AWM; white matter hyperintensities), to attempt quasi-replication of previous findings.

Higher hsCRP was associated with lower global white matter RND, with several tract-level associations. hsCRP was also associated with greater entorhinal cortex FNI. Conventional DTI metrics showed no associations with hsCRP. In structural analyses, higher hsCRP was associated with lower global gray matter volume but not white matter volume or abnormalities.

In this sample of older males, higher hsCRP was associated with differences in white matter microstructure measured using multi-shell RSI metrics and with lower global gray matter volume. Conventional DTI metrics showed few associations with hsCRP. These findings suggest that systemic inflammation may be reflected in subtle differences in brain microstructure and macrostructure and highlight the potential value of more sensitive multi-shell diffusion approaches for detecting inflammation-related brain differences in aging populations.