Abstract Body: Background
As of mid-2025, over 100 million people have had COVID-19 and in the United States, 1.2 million have died. Black Americans continue to experience higher rates of severe COVID-19 (requiring hospitalization or causing death) than White Americans. Biological mechanisms underlying severe COVID-19 remain unclear, although evidence suggests endothelial dysfunction contributes to disease severity.

Aims
We studied associations of pre-pandemic endothelial biomarkers with severe COVID-19 and evaluated differences in associations between Black and White Americans.

Methods
We employed a case only design among Black and White participants of the REasons for Geographic and Racial Differences in Stroke (REGARDS) study who had COVID-19 between 2020 and 2021. Endothelial biomarkers (E-selectin, P-selectin, factor VIII, ICAM-1, and VCAM-1) were measured from stored serum collected in 2013-2016. The outcome was severe COVID-19 defined as hospitalization or death (adjudication rate 96.2%). Non-severe COVID-19 (the reference group) was defined as self-reported symptomatic COVID-19, positive SARS-CoV2 testing, and no hospitalization or death. Logistic regression was used to estimate odds ratios of severe COVID-19 by SD higher of each biomarker. Restricted cubic splines were used to visualize odds ratios of severe COVID-19.

Results
Among the 415 participants with COVID-19, 47% were male, 33% were Black, and the mean (SD) age was 60 (7) years. Each SD higher ICAM-1 and log VCAM-1, but not other biomarkers, was associated with 30% higher odds of severe COVID-19 (Table). Associations were similar across racial groups (Table). Restricted cubic splines showed that the association of VCAM-1 and severe COVID-19 was non-linear whereas the association of ICAM-1 and severe COVID-19 was linear (Figure).

Conclusions
Pre-pandemic endothelial dysfunction, reflected by higher ICAM-1 and VCAM-1, was associated with higher odds of severe COVID-19. These findings suggest that longstanding endothelial dysfunction predisposes to more severe COVID-19, and highlights ICAM-1 and VCAM-1 as potential therapeutic targets. Replication in larger studies is needed to confirm these results.