Thromboses and COVID-19: reducing inflammation in addition to thromboprophylaxis

The COVID-19 pandemic is an unprecedented global health-care emergency, with high mortality in patients who develop COVID-19 pneumonia. These patients have a prothrombotic state with both venous and arterial thrombi occurring despite thromboprophylaxis. Prothrombotic mechanisms are multifactorial, with immune activation leading to an acute phase response, resulting in elevated plasma coagulation factors (particularly fibrinogen). Other features include platelet hyperreactivity, the effects of hypoxia, formation of neutrophil extracellular traps, and complement activation. Although very high circulating D-dimer concentrations are observed in patients with COVID-19, there is little evidence of disseminated intravascular coagulation, as thrombocytopenia and hypofibrinogenaemia are not present and screening clotting times are not prolonged. Many mechanisms driving thromboses in patients with COVID-19 have been suggested, including inflammatory activation of endothelial cells. We believe that the pathogenesis of thrombosis in patients with COVID-19 pneumonia shares similarities with that in patients with Behçet's syndrome.

Behçet's syndrome is a multisystem vasculitis, most commonly characterised by recurrent orogenital ulcers and uveitis. Vascular involvement affects 10–30% of people with Behçet's syndrome, causing mainly superficial or deep venous thrombosis. Vascular wall inflammation, rather than a hypercoagulable state, is the main cause of thromboses in patients with Behçet's syndrome. Hence, treatment guidelines endorse immunosuppression (including steroids and tumour necrosis factor blockade) and discourage the use of anticoagulation, mainly due to the perceived risk of bleeding from covert pulmonary arterial aneurysms, which is not present in patients with COVID-19.¹ Although pulmonary emboli are described in patients with Behçet's syndrome and patients with COVID-19, this term could be misleading, as segmental and subsegmental changes seen on CT pulmonary angiograms might not be caused by emboli but by immunothrombosis or in-situ thrombosis due to local inflammation. There are histological similarities in the two conditions. In patients with Behçet's syndrome, thrombi are tightly adherent to the vessel wall, and some thrombus casts in patients with COVID-19 have been shown to conform to the pulmonary artery vasculature (suggesting in-situ anatomical origin) and to occur without an overt distal embolic source, such as deep venous thrombosis.² Therefore, pulmonary inflammation is likely to drive thrombosis in both patients with Behçet's syndrome and patients with COVID-19.

Thromboprophylaxis reduces the risk of venous thromboembolism for unwell, immobile, hospitalised patients by approximately 50%. Pharmacological thromboprophylaxis improves survival and has become standard of care in patients with COVID-19 pneumonia, but thrombotic complications still occur at high rates. We hypothesise that an anti-inflammatory strategy, in addition to thromboprophylaxis, might be beneficial in patients with COVID-19 pneumonia to reduce the burden of immunothrombosis.

The therapeutic potential of targeting inflammation to reduce thromboses was shown by the Canakinumab Anti-inflammatory Thrombosis Outcomes Study (CANTOS),³ a placebo-controlled trial that was published before the COVID-19 pandemic. This trial showed that targeting interleukin-1β significantly reduced the frequency of recurrent thrombotic events (eg, myocardial infarction, stroke, and cardiovascular death), particularly in patients with greater reductions in high-sensitivity C-reactive protein.³ Although the CANTOS patient population had arterial thromboses only (and are therefore not directly comparable to patients with COVID-19), the results might signal a therapeutic opportunity for patients with COVID-19. A meta-analysis of clinical trials in critically ill patients with COVID-19 showed that systemic corticosteroids were associated with a decreased 28-day all-cause mortality,⁴ although the effect of corticosteroids on haemostasis has not yet been reported. A systematic review of glucocorticoid use in patients with inflammatory conditions showed a reduction in the concentration of procoagulant factors (von Willebrand factor and fibrinogen), whereas the plasminogen activator inhibitor-1 (an antifibrinolytic protein) concentration increased .⁵ We therefore eagerly await data on the effect of immunomodulatory approaches on thrombotic outcomes in patients with COVID-19.