Thromb Haemost
DOI: 10.1055/s-0040-1713400
Letter to the Editor
Georg Thieme Verlag KG Stuttgart · New York

Role of von Willebrand Factor and ADAMTS-13 in the Pathogenesis of Thrombi in SARS-CoV-2 Infection: Time to Rethink

Nuccia Morici
1  Intensive Cardiac Care Unit and De Gasperis Cardio Center, ASST Grande Ospedale Metropolitano Niguarda, Milan, Italy
2  Department of Clinical Sciences and Community Health, Università degli Studi di Milano, Milan, Italy
,
Maurizio Bottiroli
3  Department of Anesthesiology, ASST Grande Ospedale Metropolitano Niguarda, Milan, Italy
,
Roberto Fumagalli
3  Department of Anesthesiology, ASST Grande Ospedale Metropolitano Niguarda, Milan, Italy
4  Università degli Studi Milano-Bicocca, Milan, Italy
,
Claudia Marini
1  Intensive Cardiac Care Unit and De Gasperis Cardio Center, ASST Grande Ospedale Metropolitano Niguarda, Milan, Italy
,
Marco Cattaneo
5  Dipartimento di Scienze della Salute, Università Degli Studi di Milano, Milan, Italy
6  Unità di Medicina 2, ASST Santi Paolo e Carlo, Milan, Italy
› Author Affiliations
Further Information

Publication History

09 May 2020

20 May 2020

Publication Date:
23 June 2020 (online)

There is growing evidence supporting the idea that vascular occlusions in pulmonary and systemic circulation are among the most severe and common causes of poor outcome in COVID-19 patients.[1] [2] [3] [4] [5] [6] [7] These vascular events appear to be mostly caused by local formation of thrombi, rather than by venous thromboembolism, which likely form as a consequence of a thromboinflammatory process, triggered by viral infection-induced endothelial damage and cytokine storm, with consequent activation of hemostasis leading to thrombus formation and boosting inflammation further.[8] Among the several hemostasis players that are implicated in thromboinflammation, von Willebrand factor (VWF) has a leading role.[9] [10] [11] VWF plays two important roles in normal hemostasis: it carries factor VIII and mediates platelet-vessel wall and platelet-to-platelet interaction, especially at high shear, through its binding to the platelet membrane glycoprotein (GP) Ib and GPIIb/IIIa.[12] VWF is a multimeric protein that is released into the circulation from endothelial stores in a highly thrombogenic form, characterized by the presence of ultralarge multimers. Under normal circumstances, these ultralarge multimers are cleaved by the protease ADAMTS-13 (a disintegrin and metalloprotease with a thrombospondin type 1 motif, member 13) and, consequently, the high thrombogenicity of released VWF is reduced.[12] [13] [14] Conditions that are associated with congenital or autoantibody-induced ADAMTS-13 deficiency are characterized by increased incidence of thrombotic complications, which may be very severe and potentially fatal, as in thrombotic thrombocytopenic purpura (TTP).[15] [16] Acquired deficiency of ADAMTS-13 is associated also with systemic disorders, including severe inflammatory diseases and sepsis.[9] [10] [17] Considering that COVID-19 is a severe inflammatory disease, it is plausible that it is associated with acquired ADAMTS-13 deficiency and, hence, increased thrombogenicity of VWF.

To test this hypothesis, we measured ADAMTS-13, VWF levels, and other relevant tests in plasma from six patients admitted to the intensive care unit of ASST Grande Ospedale Metropolitano Niguarda in Milano, Italy. Patients' consent was obtained on admission. All tests were performed by means of commercial kits: in particular, VWF assays and ADAMTS-13 measurement were done by HemosIL AcuStar (Werfen, Milano, Italy). Main clinical characteristics and laboratory findings of the patients are reported in [Table 1]. Their mean ( ± standard deviation) age was 62 ± 5 years. Two patients had cardiovascular risk factors ([Table 1]). No thrombotic events were diagnosed in 5 patients, while subclavian and axillary veins thrombosis was detected in one; two patients had bleeding events: rectorrhagia and a bleeding complication after a tracheotomy procedure.

Table 1

Clinical characteristics and main laboratory examinations of the six included patients

Reference values

Patient 1

Patient 2

Patient 3

Patient 4

Patient 5

Patient 6

Age

63

65

54

59

65

68

Sex

Female

Male

Female

Male

Male

Male

Ethnicity

Caucasian

Caucasian

Caucasian

Caucasian

Caucasian

Hispanic

Hypertension

No

No

No

Yes

Yes

No

Diabetes

No

No

No

Yes

No

No

Prior cardiovascular disease

No

No

No

Yes

No

No

Prior antithrombotic treatments

No

No

No

No

No

No

ABO blood group

0

0

AB

0

A

0

Chronic kidney disease

No

No

yes

No

No

No

Clinical presentation

 Fever

Yes

Yes

Yes

Yes

Yes

Yes

 Cough

Yes

No

Yes

No

Yes

No

 Dyspnea

No

Yes

Yes

No

Yes

No

 Gastrointestinal disorders

No

No

No

Yes

No

No

 Syncope

No

Yes

No

Yes

No

No

 Disease severity status

Critical

Critical

Critical

Critical

Critical

Critical

 Invasive mechanical ventilation

Yes

Yes

Yes

Yes

Yes

Yes

 In-hospital death

Yes

No

No

No

Yes

No

 Cerebrovascular events

No

No

No

No

No

No

 Deep vein thrombosis

No

No

No

No

No

Yes

 Pulmonary artery occlusion

No

No

No

No

No

No

 Acute coronary syndrome

No

No

No

No

No

No

 Vascular occlusion in other districts

No

No

No

No

No

No

 Bleeding events

Yes

No

No

No

No

Yes

Peak creatinine, mg/dL

0.5–1.0

2.9

2.7

3.1

1.0

1.1

2.2

Peak blood urea nitrogen, mg/dL

18–48

249

226

220

109

53

140

FVIII, U/dL

51–147

279

316

308

229

240

259

VWF:Ag, U/dL

40–165

772

735

568

455

763

511

VWF:RCo, U/dL

41–151

451

496

470

339

496

448

VWF:CB, U/dL

45–174

612

683

502

397

623

485

ADAMTS-13, %

45–138

24

37

30

56

44

33

Anti-ADAMTS-13, U/mL

< 12

< 12

< 12

< 12

Not tested

< 12

18

IL-6, pg/mL

< 7

7,152

Not tested

Not tested

42

90

Not tested

C-reactive protein, mg/dL

0–0.5

24

45

45

30

31

19

Abbreviations: ADAMTS-13, a disintegrin and metalloprotease with a thrombospondin type 1 motif, number 13; FVIII, factor VIII; IL-6, interleukin 6; VWF:Ag, von Willebrand factor antigen; VWF:CB von Willebrand collagen-binding activity; VWF:RCo, von Willebrand factor ristocetin cofactor activity.


Factor VIII, D-dimer, fibrinogen, VWF:Ag (antigen), VWF:RCo (ristocetin cofactor activity), and VWF:CB (collagen-binding activity) were significantly increased in all patients, confirming the results of a previous study (with the exception of VWF:CB, which had not been measured).[18] These results reflect a severe inflammatory state, as all the tested proteins are known to be acute phase reactants and D-dimer increases in inflammatory conditions. The original finding of our study is that 5 of the 6 tested patients had plasma ADAMTS-13 levels lower than 45%, the lower limit of the normal range: 3 of them had ADAMTS-13 levels around 30%, possibly preventing the effective cleavage of VWF multimers, thus favoring the formation of local thrombi in the circulation. Up to one-third of patients with sepsis have been shown to have ADAMTS-13 levels < 50%,[17] while patients with septic shock and ADAMTS-13 levels < 30% had increased risk of mortality.[19] Plasma levels of antibodies against ADAMTS-13 were normal in our patients, with the exception of one, who displayed mild elevation of the antibody titer ([Table 1]). This observation reinforces the hypothesis that low ADAMTS-13 levels in inflammation and sepsis are not caused by autoimmunity, but by consumption by the large amounts of VWF released in the circulation by endothelial cells, and/or by cleavage by cell-derived proteases.[17] It is interesting to note that the VWF:RCo to VWF:Ag ratio was slightly decreased in our COVID-19 patients, as it has already been shown in another study,[18] while the VWF:CB to VWF:Ag ratio was close to unity. The correspondence between VWF activity tests and immunological determinations of VWF is probably not optimal when the amount of protein present is extremely high; the better correspondence with VWF:Ag that was displayed by VWF:CB, compared with VWF:RCo, is likely a reflection of the greater sensitivity of VWF:CB to very high molecular weight multimers,[20] which are likely increased in our patients with low ADAMTS-13 levels. Unfortunately, we could not measure VWF multimers directly.

The mild–moderate decrease in plasma ADAMTS-13 levels in COVID-19 patients is not expected to cause a thrombotic microangiopathy of similar severity as that observed in TTP, as testified also by the very mild reduction in platelet count displayed by our patients. However, it could contribute to the pathogenesis of thrombi. Indeed, the association of low ADAMTS-13 plasma levels and thrombosis has already been shown in previous studies: while the association with thrombotic microangiopathies different from TTP, myocardial infarction, and stroke is well established,[21] that with venous thromboembolism is less clear.[22] This is in accordance with the role of high molecular weight multimers of VWF in supporting platelet-vessel wall interaction especially at high shear rates that are encountered in the arterial circulation.[12] [23] These considerations open the question of what could be the most effective treatment of VWF-mediated thrombotic episodes in COVID-19. Heparin, whether unfractionated or low molecular weight, is likely the best therapeutic option to prevent venous thromboembolism and, as a matter of fat, is recommended for thromboprophylaxis in all critically ill medical patients at risk.[24] The question for COVID-19, which is characterized by a high degree of hypercoagulability,[6] [18] is whether doses of heparin higher than those generally recommended are necessary, which is being currently addressed by several randomized clinical trials (NCT04372589, NCT04367831, NCT04345848, NCT04366960). Heparin prophylaxis might also limit the severity of disseminated intravascular coagulation that is observed in COVID patients,[6] albeit not very frequently. In contrast, heparin is unlikely the ideal treatment for microthrombosis in arterioles of the lungs and other organs of COVID patients,[8] for which treatment aimed at restoring the balance between VWF and ADAMTS-13 (infusion of fresh-frozen plasma or recombinant ADAMTS-13 concentrate), at inhibiting VWF–platelet interaction (infusion of caplacizumab), or, more unspecifically, at inhibiting platelet function, could have more promising results. Other targets that play an important role in thromboinflammation should also be considered in COVID-19.[8]