Blood-Brain barrier disruption in long COVID and cognitive correlates: A cross-sectional MRI study.

Abstract

Disruption of the blood-brain barrier (BBB) may contribute to neuropsychiatric symptoms observed in Long COVID (LC). Using a non-contrast magnetic resonance imaging (MRI) technique, we investigated BBB permeability in individuals with LC and its relationship to cognitive function. We hypothesized that LC individuals would show greater BBB permeability than recovered individuals, and that higher permeability would correlate with poorer cognition. Ninety-seven participants meeting the 2024 NASEM definition of LC with at least one neuropsychiatric symptom and 31 recovered controls completed an MRI scan and cognitive testing. BBB permeability was assessed using water-extraction-with-phase-contrast-arterial-spin-tagging (WEPCAST) MRI, which estimates the permeability-surface-area product (PS) of arterially labeled water entering the brain. Cognitive performance was summarized into eight factor scores derived from principal components analysis. Compared to controls, the LC group was older (M = 47 vs. 39 years, P = 0.003), less educated (P = 0.02), more likely female (P = 0.04), and had higher hospitalization rates for COVID-19 (P = 0.02). PS was significantly elevated in the LC group after adjusting for age and sex (B = 18.59, SE = 6.11, β = 0.28, P = 0.003). No significant group differences were found in cerebral blood flow, extraction fraction (E), or brain volume. Within the LC group, higher PS was associated with poorer motor function, but not with other cognitive domains. These findings indicate subtle but persistent BBB disruption in LC individuals over two years post-infection, with a potential link to motor dysfunction. This supports prior evidence of BBB changes following severe COVID-19 and suggests that BBB integrity may be a long-term biomarker of neuropsychiatric complications in LC.