DOI: 10.1093/braincomms/fcae088 ISSN: 2632-1297

Helicobacter pylori, persistent infection burden and structural brain imaging markers

May A Beydoun, Hind A Beydoun, Yi-Han Hu, Ziad W El-Hajj, Michael F Georgescu, Nicole Noren Hooten, Zhiguang Li, Jordan Weiss, Donald M Lyall, Shari R Waldstein, Dawson W Hedges, Shawn D Gale, Lenore J Launer, Michele K Evans, Alan B Zonderman
  • Neurology
  • Cellular and Molecular Neuroscience
  • Biological Psychiatry
  • Psychiatry and Mental health


Persistent infections, whether viral, bacterial or parasitic, including Helicobacter pylori (Hp) infection, have been implicated in non-communicable diseases, including dementia and other neurodegenerative diseases. In this cross-sectional study, data on 635 cognitively normal participants from the UK Biobank study (2006-2021, age range: 40-70 y) were used to examine whether Hp seropositivity (Hps) (e.g. presence of antibodies), serointensities of five Hp antigens and a measure of total persistent infection burden were associated with selected brain volumetric structural magnetic resonance imaging (MRI), (total, white, gray matter, frontal gray matter (Left/Right), white matter hyperintensity as % intracranial volume and bi-lateral sub-cortical volumes) and diffusion-weighted MRI measures (global and tract-specific bi-lateral fractional anisotropy and mean diffusivity), after an average 9-10 years of lag time. Persistent infection burden was calculated as a cumulative score of seropositivity of over 20 different pathogens. Multivariable-adjusted linear regression analyses were conducted, whereby selected potential confounders (all measures) and intracranial volume (sub-cortical volumes) were adjusted, with stratification by Alzheimer’s Disease polygenic risk score tertile when exposures were Hp antigen serointensities. Type I error was adjusted to 0.007. We report little evidence of an association between Hps or persistent infection burden with various volumetric outcomes (P > 0.007, from multivariable regression models), unlike previously reported in past research. However, Hp antigen serointensities, particularly immunoglobulin G against the Vacuolating cytotoxin A (VacA), GroEL and Outer Membrane Protein (OMP) antigens, were associated with poorer tract-specific white matter integrity (P < 0.007), with OMP serointensity linked to worse outcomes in cognition-related tracts such as the external capsule, the anterior limb of the internal capsule and the cingulum, specifically at low Alzheimer’s Disease polygenic risk. VacA serointensity was associated with greater white matter hyperintensity volume among individuals with mid-level Alzheimer’s Disease polygenic risk, while among individuals with highest Alzheimer’s Disease polygenic risk, the urease serointensity was consistently associated with reduced bi-lateral caudate volumes and the VacA serointensity was linked to reduced Right putamen volume (P < 0.007). OMP and urease were associated with larger sub-cortical volumes (e.g. left putamen and right nucleus accumbens) at middle Alzheimer’s Disease polygenic risk levels (P < 0.007). Our results shed light on the relationship between Hps, Hp antigen levels and persistent infection burden with brain volumetric structural measures. These data are important given the links between infectious agents and neurodegenerative diseases, including Alzheimer’s Disease and can be used for development of drugs and preventive interventions that would reduce the burden of those diseases.

More from our Archive