Association between an inflammatory biomarker score and future dementia diagnosis in the population-based UK Biobank cohort of 500,000 people

Mekli, Krisztina, Lophatananon, Artitaya, Maharani, Asri ORCID: https://orcid.org/0000-0002-5931-8692, Nazroo, James Y and Muir, Kenneth R (2023) Association between an inflammatory biomarker score and future dementia diagnosis in the population-based UK Biobank cohort of 500,000 people. PLoS One, 18 (7). e0288045. ISSN 1932-6203

Preview

Published Version Available under License Creative Commons Attribution. Download (507kB) | Preview

Abstract

This study was designed to investigate the relationship between a systematic inflammatory biomarker measure, concurrent and later cognitive performance, and future dementia risk. The literature has reported the potential involvement of inflammation in cognitive performance as well as Alzheimer’s Disease, but not consistently. We used a population-based cohort of 500,000 people in the UK and assessed the association between a composite inflammatory biomarker and cognitive performance measures across five domains measured concurrently and 4–13 years later, taking advantage of the large sample size. We also assessed the same biomarker’s association with dementia diagnosis 3–11 years later in the initially dementia-free sample. We report small but significant associations between elevated biomarker levels and worsened cognitive performance at baseline for four cognitive tasks (OR = 1.204, p<0.001 for Prospective memory, β = -0.366, p<0.001 for Fluid intelligence, β = 8.819, p<0.001 for Reaction time, and β = -0.224, p<0.001 for Numeric memory), comparing the highest quartile of the biomarker to the lowest. We also found that for one measure (Pairs matching) higher biomarker levels were associated with fewer errors, i.e. better performance (β = -0.096, p<0.001). We also report that the 4th quartiles of the baseline biomarker levels were significantly associated with cognitive task scores assessed years later on the p< = 0.002 level, except for the Pair matching test, for which none of the quartiles remained a significant predictor. Finally, the highest biomarker quartile was significantly associated with increased dementia risk compared to the lowest quartile (HR = 1.349, p<0.001). A case-only analysis to assess disease subtype heterogeneity suggested probable differences in the association with the highest biomarker quartile between vascular dementia and Alzheimer disease subtypes (OR = 1.483, p = 0.055). Our results indicate that systemic inflammation may play a small but significant part in dementia pathophysiology, especially in vascular dementia.