Thrombotic risk and COVID-19 vaccines: new data and disputes


An Oxford University study reports that the incidence of a rare clotting complication, cerebral venous sinus thrombosis (CVST or CVT), is significantly higher with COVID-19 infection compared to after COVID-19 vaccination (Taquet et al. OSF preprint, 14 April 2021; The study has not been peer reviewed.

The Oxford group analysed electronic medical records for patients primarily in the U.S. and found that the incidence of CVT in the two weeks after a COVID-19 diagnosis was 39.0 cases per million compared to 4.1 cases per million after receiving the Pfizer/BioNTech or Moderna mRNA vaccines (adjusted hazard ratio 6.36). The risk with the Oxford-AstraZeneca vaccine was not determined since it has not received FDA approval. The European Medicines Agency’s estimated COVID-associated risk of CVT with the Oxford-AstraZeneca vaccine is 5.0 per million.

In the pre-COVID era, the frequency of CVT was approximately 5 cases per million, according to the American Heart Association/American Stroke Association, which represented about 1% of all strokes (Saposnik et al. Stroke 2011;42:1158-1192). The estimated mortality rate was <5% (Haghighi et al. Stroke 2012;43:262-264).

The Oxford findings were quickly disputed by Pfizer. In an email to The Scientist, the company stated that there was no evidence of a risk of CVT associated with the Pfizer/BioNTech vaccine in the company’s safety data, nor in reviews by the Centers for Disease Control and Prevention (CDC) and the UK’s Medicines and Healthcare products Regulatory Agency (Williams S. The Scientist, 16 April 2021;

It should be noted that the thrombotic complications linked to COVID vaccination are, in part, a subset of CVT that occur in the setting of thrombocytopenia, a syndrome that has been labelled vaccine-induced prothrombotic immune thrombocytopenia (VIPIT, or VITT). The CDC’s Advisory Committee on Immunization Practices has analysed six cases of CVT with thrombocytopenia (defined as < 150 x 109/L) that occurred following administration of the Janssen vaccine (Shimabukuro T et al. ACIP, 14 April 2021; free download at The estimated rate with the Janssen vaccine was 0.87 per million doses administered. CoV-2 serology was negative or not documented in all cases. The report also noted that there were three reports of CVT following administration of 84.7 million doses of the Moderna vaccine; platelet counts were normal, indicating that the cases were thrombosis, a known risk associated with vaccination, rather than VIPIT. No cases of CVT or VIPIT were recorded with the Pfizer vaccine.

One feature distinguishing VIPIT from CVT appears to be the presence of antibodies to the chemokine platelet factor 4 (PF4), first reported in a German case series of patients who received the AstraZeneca vaccine who developed thrombotic complications, which included splanchnic-vein thrombosis, pulmonary embolism and disseminated intravascular coagulation, as well as CVT (Greinacher et al. N Engl J Med, epublished 9 April 2021). The authors noted that the condition mimicked autoimmune heparin-induced thrombocytopenia. All subjects tested positive for anti-PF4 antibodies.

Two additional case series have also been published. A UK study reported on 23 patients with thrombosis and thrombocytopenia that occurred 6-24 days after receiving the first dose of the AstraZeneca vaccine (Scully et al. N Engl J Med, epublished 16 April 2021). The median age was 46 years (range 21-77 years); 61% were female. Most patients had no known thrombotic risk factors; one patient had a history of deep-vein thrombosis and one patient was on an oral contraceptive. All tested negative for CoV-2 at presentation. A total of 22 of 23 patients tested positive for antibodies to PF4. Twenty-two patients presented with thrombosis, including CVT (13 cases), pulmonary embolism (4 cases), ischemic stroke affecting the middle cerebral artery (2 cases), portal vein thrombosis (2 cases), and deep-vein thrombosis (1 case).

A Norwegian study also reported 5 cases of VIPIT that occurred in healthcare workers within 10 days of receiving the AstraZeneca vaccine (Schultz et al. N Engl J Med, epublished 9 April 2021). All five patients had high levels of IgG antibodies to PF4 on ELISA.

These results suggest that the pathophysiology of VIPIT and CVT differ, and that the risk of VIPIT should be differentiated from that of CVT or other thrombotic conditions that may develop as a complication of COVID-19 infection. It remains to be determined if components of the various vaccines induce a host immune response to PF4; why this response appears to be more common after administration of a viral vector vaccine; if anti-PF4 antibody testing would be useful to identify at-risk patients; and the possible risks when a second vaccination is administered (Cines & Bussel. N Engl J Med 2021; epublished 16 April 2021 See also correction, 20 April 2021,

Recommend to a Colleague

Related Posts

Go back to home page