A striking clinical feature of illness from SARS-CoV-2 is a marked increase in thrombotic and microvascular complications, or COVID-19–associated coagulopathy (CAC).
Dr. Alan P. Lyss
A new study suggests endothelial cell injury plays a major role in the pathogenesis of CAC, and blood levels of soluble thrombomodulin correlate with mortality.
George Goshua, MD, of Yale University, New Haven, Conn., presented this study as a late-breaking abstract at the virtual annual congress of the European Hematology Association.
Dr. Goshua cited past research showing CAC to be highly prevalent among hospitalized patients. Venous thromboembolism was found in 17% to 69% of patients, despite thromboprophylaxis.1-4 Arterial thrombosis has been seen in 3.6% to 4.0% of patients,1-3 and autopsy findings have shown microvascular thrombosis in as many as 87% of patients.5-7
For their study, Dr. Goshua and colleagues assessed endothelial cell damage, platelet activation, and hemostatic and fibrinolytic cascade effects of CAC.
The investigators measured markers of endothelial cell injury and platelet activation, plasminogen activation inhibitor 1 (PAI-1), and coagulation factors in stable and critically ill patients hospitalized with COVID-19. In addition, the team sought to identify biomarkers of mortality in hospitalized patients.
Dr. Goshua and colleagues studied 68 adults hospitalized for suspected COVID-19 – 48 in the ICU and 20 outside the ICU. Patients in the ICU received mechanical ventilation, while the non-ICU patients required supplemental oxygen (≤3 L/min per nasal cannula).
There were more men than women (69% vs. 31%) in the ICU population but not in the non-ICU population (40% vs. 60%). There were no statistically significant differences in age or comorbid conditions between the ICU and non-ICU patients.
Results and interpretation
Consistent with augmentation of the coagulation cascade – and as expected – D-dimer and thrombin-antithrombin levels were high in both the ICU and non-ICU populations, but levels were significantly higher (P < .001) among the ICU patients.
Endogenous anticoagulants (antithrombin and proteins C and S) and fibrinolytic enzymes (alpha 2-antiplasmin) were preserved, verifying that CAC is distinct from disseminated intravascular coagulation. Classic fibrinolysis did not occur, as PAI-1 was high in ICU and non-ICU patients, and lysis-30 was normal in nearly all ICU patients (96%).
Von Willebrand factor antigen and activity levels and factor VIII levels were markedly elevated in non-ICU and ICU patients, but they were significantly higher (P < .001) in the ICU cohort. This supports the hypothesis that endothelial cell damage and platelet activation play major roles in CAC.
Similarly, soluble P-selectin, which is shed from endothelial cells and platelets, was dramatically elevated in ICU patients in comparison with controls and non-ICU patients (P < .001 for both comparisons).
Levels of soluble thrombomodulin, which is released from endothelial cells, were not significantly different in ICU patients and controls. However, given thrombomodulin’s significant role in the coagulation cascade, Dr. Goshua and colleagues plotted receiver operating curves to see if soluble thrombomodulin levels were predictive of mortality.
The results showed that soluble thrombomodulin correlated with the probability of survival, both overall and in ICU patients. Soluble thrombomodulin levels greater than 3.26 ng/mL were associated with significantly worse survival in all patients (P = .0087) and ICU patients (P = .0309).