Mapping neuroinflammation with diffusion-weighted MRI: randomized crossover study.

Abstract

Background: The pathophysiology of neuroinflammation in psychiatric conditions remains poorly understood, highlighting the need for non-invasive tools that can measure neuroinflammation in vivo. We explored advanced diffusion-weighted MRI techniques for detection of low-level neuroinflammation induced by typhoid vaccine, with potential applications to psychiatric disorders.

Methods: Twenty healthy volunteers (10 males, median age 34, range 18-44 years) participated in a randomized, placebo-controlled, crossover design study. Participants underwent MRI before and after receiving placebo or vaccine in alternating sessions, separated by a washout period. Diffusion tensor (multi-shell and single-shell), diffusion kurtosis, and neurite orientation density and dispersion imaging parameter maps were generated. Probabilistic tractography investigated differences in tract volume, fractional anisotropy (FA), and mean diffusivity (MD) of the tracts. Thirteen tracts and 15 regions were analyzed using a region-of-interest approach entered into linear mixed models to evaluate treatment effects.

Results: A treatment effect was observed on white matter tracts derived from XTRACT, with a global reduction in MD (p=.040). White matter tracts-of-interest showed increased axial kurtosis (p<.001) while grey matter regions-of-interest demonstrated increased mean and radial kurtosis (both p=.038). Additionally, several correlations were found between the inflammatory marker interleukin (IL)-6 and diffusion parameters.

Conclusions: Our findings demonstrate that diffusion-weighted MRI may be sensitive to inflammation-induced microstructural changes in the brain. Future studies should integrate complementary techniques and clinical assessments to deepen our understanding of inflammatory pathophysiology and its implications for health outcomes in clinical populations.