Dexamethasone Suppresses Endotheliopathy and Endothelial-Induced Coagulopathy in COVID-19

BACKGROUND: Endotheliopathy and coagulopathy are known complications of COVID-19, with a significant association with mortality. Although dexamethasone is the standard of care for patients with severe COVID-19, its precise mode of action remains elusive. We aim to investigate the functional consequences of dexamethasone treatment on COVID-19–associated procoagulant endotheliopathy. METHODS: First, during the 7 days after hospitalization, we measured several endothelial and coagulopathy biomarkers in a prospective cohort of patients with COVID-19 with acute respiratory distress syndrome (ARDS) who were either treated or not with both dexamethasone and therapeutic UFH (unfractionated heparin). Second, we developed an in vitro thrombin generation assay on cultured human endothelial cells to measure the ability of stimulated endothelial colony-forming cells to activate coagulation in normal plasma, which is expressed as an endogenous thrombin potential (ETP). RESULTS: Among the cohort of 44 ARDS COVID-19 patients, 23 patients treated with dexamethasone and therapeutic UFH had significantly decreased von Willebrand Factor, Ang-2 (angiopoietin-2), soluble E-selectin, and d -dimer levels over 7 days. To differentiate the effect of UFH and dexamethasone on endotheliopathy, we used the thrombin generation assay and showed that endothelial colony-forming cell stimulation with dexamethasone but not UFH—in addition to a cocktail of proinflammatory cytokines (to mimic the cytokine storm of severe COVID-19)—significantly decreased ETP in comparison to proinflammatory cytokines only. Moreover, in another cohort of 331 patients with COVID-19 of varying severity, the endothelial colony-forming cell stimulation with the plasma of 87 ARDS patients showed significantly higher ETP (1260 nmol/L per minute [interquartile range, 1140–1260]) compared with 75 non-ARDS patients (1024 nmol/L per minute [interquartile range, 915–1200]; P <0.001). Finally, 94 dexamethasone-treated ARDS patients had significantly lower ETP (962.8 nmol/L per minute [interquartile range, 782.9–1112]) in contrast to 75 nondexamethasone-treated ARDS patients (1,260 nmol/L per minute [interquartile range, 1140–1359]; P <0.001). ETP could help predict in-hospital mortality in a Kaplan-Meier estimator analysis ( P =0.0002). CONCLUSIONS: Our data suggest that dexamethasone protects against COVID-19 endothelium-induced coagulopathy. These findings are in line with the decreased prevalence of venous thrombosis among hospitalized patients with COVID-19 treated with dexamethasone.