Table 1.

Consensus recommendations on coagulation assays and associated thrombosis biomarkers in COVID-19

Category of biomarker	Measure	LoE supporting usefulness in COVID-19-related thrombosis			Summary of evidence supporting usefulness	Consensus recommendation	LoEa
		Prognosis	Diagnosis	Management
Acute-phase proteins	C-reactive protein	✓ LoE 1; large, prospective studies including meta-analyses	⨯	✓ LoE 1	Significant data that levels of C-reactive protein are increased in patients with COVID-19 and are associated with disease severity, occurrence of VTE and mortality	Routine use to guide prognosis and disease severity assessment; also useful to assess risk of VTE	1
	IL-6	⨯ LoE 1 to guide prognosis but not for prediction of thrombosis	⨯	⨯	Several studies show that IL-6 levels are dramatically increased in patients with COVID-19 and are associated with disease severity and mortality	Useful to guide prognosis and disease severity assessment, including in addition to C-reactive protein measurement, but no convincing data that relate directly to thrombosis	5
	d-dimer	✓ LoE 1; large, prospective studies including meta-analyses showing a link to thrombosis	✓ LoE 1; meta-analyses showing that d-dimer levels are higher in patients with VTE than in those without VTE	⨯ LoE 3; conflicting data; largest prospective study showed a benefit of anticoagulation, regardless of d-dimer level, but magnitude of benefit was greater in patients with a higher d-dimer level	Significant evidence available that d-dimer levels are elevated in patients with COVID-19 and are associated with disease severity and adverse outcomes, including mortality; various thresholds of d-dimer level have been used to guide anticoagulant management, but the benefit of this approach is uncertain	Routine use to guide prognosis and assessment of disease severity and risk of VTE	1
						Management of anticoagulation on the basis of d-dimer levels is not currently recommended	3
Markers of coagulation	Prothrombin	⨯ Abnormal only in very severe disease	⨯	✓ Useful only to assess anticoagulant effect	Not useful as a marker of disease severity or prognosis	Routine measurement is not recommended on the basis of current evidence (except to assess anticoagulant effect, as needed)	3
	aPTT	⨯	⨯	✓ Useful only to assess anticoagulant effect	Not useful as a marker of disease severity or prognosis	Routine measurement is not recommended on the basis of current evidence (except to assess anticoagulant effect, as needed)	3
	Anti-Xa	⨯	⨯	✓ Clear potential for management of LMWH dose, but prospective study and validation needed	Not useful as a marker of COVID-19 severity or prognosis	Routine measurement is not recommended on the basis of current evidence	4
	Thrombin generation	⨯	⨯	⨯	Not useful as a marker of COVID-19 severity or prognosis; not useful to guide management	Routine measurement is not recommended on the basis of current evidence	4
	Viscoelastic assays	✓ LoE 3; some conflicting results, but preliminary data indicate that fibrin clot strength might be associated with prognosis, including thrombotic events	⨯	⨯ Inadequate response, as indicated by prolonged reaction time, has been demonstrated with prophylactic anticoagulation, but no convincing data that tailored management can improve outcomes	Elevated levels of platelet–fibrin clot strength, fibrinogen and fibrin clot strength in patients with COVID-19 compared with patients with pneumonia; preliminary data suggest the use of TEG to personalize antiplatelet or antithrombotic therapy to improve outcomes, but more data are needed before implementing in routine practice	Potential is evident, but insufficient evidence to recommend routine use	3
Markers of fibrinolysis	Viscoelastic assays	✓ LoE 3; case–control studies, mainly in the ICU, showing that hypofibrinolysis is associated with thrombotic complications	⨯	⨯	Marked hypofibrinolysis documented in patients with COVID-19 in the ICU, but data from patients not in the ICU are limited; hypofibrinolysis detected by prolonged clot lysis time and increased maximum clot firmness is associated with thrombotic events and adverse prognosis; no data that this finding can help to guide management	Might be useful in critically ill patients to guide prognosis and predict the risk of thrombosis; potential is evident, but insufficient evidence to recommend routine use	3
	PAI-1, tPA, TAFI, thrombo-modulin	⨯ Levels are associated with disease severity, but not with occurrence of thrombosis; mechanistic link to thrombosis	⨯	⨯	Elevated levels are associated with disease severity; PAI-1, tPA and thrombomodulin levels might be associated with thrombotic events and prognosis, but studies have so far been too small to draw definitive conclusions	Routine measurement is not recommended on the basis of current evidence	5
Markers of endothelial dysfunction	vWF and ratio of vWF antigen to ADAMTS13	⨯✓ LoE 3; abundant case–control studies showing usefulness to guide prognosis, but not to predict risk of thrombosis LoE 5; mechanistic link to thrombosis biomarkers	⨯	⨯	Increased levels of vWF antigen and activity and increased ratio of vWF antigen to ADAMTS13 reported in patients with COVID-19, which correlate with disease acuity and mortality; no data that these tests can help to guide management	Useful to guide prognosis, but not directly related to thrombotic events; not useful to guide management	3
Extracellular vesicles	Extracellular vesicles	✓ LoE 4; predictive of illness severity and a few case–control studies showing association with thrombosis; mechanistic link to thrombosis biomarkers	⨯	⨯	Various cut-offs used and various extracellular vesicles measured; upregulation of extracellular vesicle volume, mean particle size and extracellular vesicle tissue factor activity are all correlated with disease severity and thrombosis; no data that these markers can be used to diagnose thrombotic events or guide management	Potential is evident, but the markers are highly heterogeneous; measurement is not recommended on the basis of current evidence	4
Novel soluble biomarkers	NETs	✓ LoE 5; observational studies show an association with disease severity, but not with thrombotic events; NETs seen in tissue samples; mechanistic link to markers of thrombosis such as d-dimer	⨯	⨯	Potentially of use in detecting severe versus non-severe COVID-19, but not in predicting thrombotic risk; management according to NET parameters has not been examined	Measurement is not recommended on the basis of current evidence	5
	Complement factors	⨯ Associated with disease severity but not with occurrence of thrombosis	⨯	⨯	Potentially of use in detecting severe COVID-19; longer-term prognostic utility unknown; management according to complement factor levels has not been examined	Measurement is not recommended on the basis of current evidence	4–5
	ACE2	⨯	⨯	⨯	Discrimination of COVID-19 severity not shown	Measurement is not recommended on the basis of current evidence	5
	Calprotectin	✓ LoE 3; retrospective, observational studies show an association with thrombosis and critical illness	⨯	⨯	Potentially of use in detecting severe COVID-19 and assessing the risk of thrombosis; management according to calprotectin levels has not been examined	Potential is evident, but prospective study and validation are needed; routine measurement is not recommended	3

ACE2, angiotensin-converting enzyme 2; ADAMTS13, a disintegrin and metalloproteinase with thrombospondin motifs 13; aPTT, activated partial thromboplastin time; COVID-19, coronavirus disease 2019; ICU, intensive care unit; LMWH, low-molecular-weight heparin; LoE, level of evidence; NET, neutrophil extracellular trap; PAI-1, plasminogen activator inhibitor 1; TAFI, thrombin-activatable fibrinolysis inhibitor; TEG, thromboelastography; tPA, tissue plasminogen activator; VTE, venous thromboembolism; vWF, von Willebrand factor. ^aThe level of evidence to support measurement as a biomarker of thrombosis is based on the scoring system of the Oxford Centre for Evidence-Based Medicine Levels of Evidence 2 (ref.²⁵).