Cerebral vein thrombosis after ChAdOx1 nCov‐19 vaccination: Long‐term outcome of four patients

Vaccineinduced immune thrombotic thrombocytopenia (VITT) is a rare prothrombotic syndrome associated with an adenoviral vectorbased vaccine ChAdOx1 CoV19 vaccine (AstraZeneca, University of Oxford) or Ad26.COV2.S vaccine (Janssen; Johnson & Johnson). VITT is characterized by thrombosis, often at unusual sites, particularly cerebral venous sinus thrombosis (CVST) with thrombocytopenia, low fibrinogen, and significantly raised Ddimer level.1 A large UKbased study reported an overall mortality rate of 22%, rising to 73% in patients with platelet count less than 30 × 109/L and intracerebral hemorrhage (ICH).2 Early recognition and prompt treatment reduced mortality of VITT from approximately 50% in the first series reported in April 2021 to 22% by June 2021 in the United Kingdom and more recently to about 5% observed in Australia, probably due to earlier recognition and prompt treatment.3 The pathophysiology of VITT remains incompletely defined but appears to be an immunemediated process, similar to that of heparininduced thrombocytopenia (HIT). It is associated with the development of IgG antibodies directed against platelet factor 4 (PF4) but without exposure to heparin. Thrombosis can be venous or arterial or both and often develops at multiple sites. Laboratory markers can demonstrate a consumptive coagulopathy, with thrombocytopenia, hypofibrinogenemia, and grossly elevated Ddimers. Notably, PF4 antibodies are most reliably detected using ELISA while chemiluminescence immunoassays assays produce falsenegative results. CVST has been reported to be the predominant thrombotic manifestation, whereas it is rare in HIT. In a recent systematic review of 23 case series of HIT including 1220 patients with HIT, only 27 had CSVT(1.6%)4 compared to 50% of patients with VITT (110/220).2 The presence of ICH is associated with high mortality in patients with CVST, reaching 73% in patients with VITTassociated CVST.2 However, longterm outcome data of the patients presenting with CVST associated with VITT are lacking. We report a singlecenter experience of beyond 1year (14 months) outcomes of four female patients, aged 41– 46 years, who developed CVST with or without thrombosis at other sites due to VITT during May 7– 24, 2021, in London, United Kingdom (Table S1). These four patients represented all patients admitted to our hospitals with VITT during this period. The institutional review board approved this study, and informed written consent was obtained from all patients. Each presented with headache and neurological deficit. Two patients had subarachnoid hemorrhage. A functional assay to demonstrate heparinindependent plateletactivating antiPF4 antibodies without additional PF4 was performed for the diagnosis using wholeblood impedance aggregometry in addition to antiPF4 ELISA. Aggregation of donor platelets after incubation with serum from the patients was measured in the presence of low (0.43 IU/ml) or high (100 IU/ml) concentrations of unfractionated heparin (UFH) and in the absence of added heparin (saline). Serum from a healthy donor and serum from a patient with confirmed classical HIT were also tested. Serum from patients with classical HIT showed marked platelet aggregation with lowdose UFH but not with high dose UFH. Serum from all four patients with VITT showed higher (spontaneous) platelet aggregation in the absence of heparin (with saline) compared to serum from patient with classical HIT and the healthy control. When serum from patients with VITT was incubated with lowor highdose UFH, there was a variable response in platelet aggregation.5 All four patients received a uniform management approach with immediate transfer to an intensive care unit with combined hyperacute stroke unit and neurosurgical presence on initial presentation. Urgent plasma exchange (PLEX) with Octaplas was initiated within 24 h and continued for 5 days. Alongside PLEX, intravenous (IV) immunoglobulin at 1 g/kg in two divided doses and highdose steroids (1 g IV methylprednisolone followed by 20 mg dexamethasone IV/oral for 4 days) were started irrespective of the platelet count, as the benefit of the steroids outweighs the side effects related to a short course of steroids. Anticoagulation was initiated with argatroban at presentation despite the low platelet count due to concern of the progressive CVST. Target activated partial thromboplastin time ratio was 1.5– 2.0 for the first 48 h, increased to 2.0– 2.5 in the absence of further bleeding or deterioration on repeat imaging in those with subarachnoid hemorrhage. There was no progression of ICH with argatroban, and platelet count gradually improved following the start of IV IgG and highdose steroids5 (Figure 1A– D). A single dose of rituximab (375 mg/m2) was given to two patients (Figure 1A,B). Patient 1 was given rituximab in addition to PLEX, highdose steroids, and IV IgG, as she presented with severe thrombocytopenia and subarachnoid hemorrhage, with a history of being semiconscious for 96 h with neurological features

73% in patients with platelet count less than 30 × 10 9 /L and intracerebral hemorrhage (ICH). 2 Early recognition and prompt treatment reduced mortality of VITT from approximately 50% in the first series reported in April 2021 to 22% by June 2021 in the United Kingdom and more recently to about 5% observed in Australia, probably due to earlier recognition and prompt treatment. 3 The pathophysiology of VITT remains incompletely defined but appears to be an immune-mediated process, similar to that of heparin-induced thrombocytopenia (HIT). It is associated with the development of IgG antibodies directed against platelet factor 4 (PF4) but without exposure to heparin. Thrombosis can be venous or arterial or both and often develops at multiple sites. Laboratory markers can demonstrate a consumptive coagulopathy, with thrombocytopenia, hypofibrinogenemia, and grossly elevated D-dimers. Notably, PF4 antibodies are most reliably detected using ELISA while chemiluminescence immunoassays assays produce false-negative results. CVST has been reported to be the predominant thrombotic manifestation, whereas it is rare in HIT. In a recent systematic review of 23 case series of HIT including 1220 patients with HIT, only 27 had CSVT(1.6%) 4 compared to 50% of patients with VITT (110/220). 2 The presence of ICH is associated with high mortality in patients with CVST, reaching 73% in patients with VITT-associated CVST. 2 However, long-term outcome data of the patients presenting with CVST associated with VITT are lacking.
We report a single-center experience of beyond 1-year (14 months) outcomes of four female patients, aged 41-46 years, who developed CVST with or without thrombosis at other sites due to VITT during May 7-24, 2021, in London, United Kingdom (Table S1).
These four patients represented all patients admitted to our hospitals with VITT during this period. The institutional review board approved this study, and informed written consent was obtained from all patients. Each presented with headache and neurological deficit.
Two patients had subarachnoid hemorrhage. A functional assay to demonstrate heparin-independent platelet-activating anti-PF4 antibodies without additional PF4 was performed for the diagnosis using whole-blood impedance aggregometry in addition to anti-PF4 ELISA. Aggregation of donor platelets after incubation with serum from the patients was measured in the presence of low (0.43 IU/ml) or high (100 IU/ml) concentrations of unfractionated heparin (UFH) and in the absence of added heparin (saline). Serum from a healthy donor and serum from a patient with confirmed classical HIT were also tested. Serum from patients with classical HIT showed marked platelet aggregation with low-dose UFH but not with high dose UFH. Serum from all four patients with VITT showed higher (spontaneous) platelet aggregation in the absence of heparin (with saline) compared to serum from patient with classical HIT and the healthy control. When serum from patients with VITT was incubated with low-or high-dose UFH, there was a variable response in platelet aggregation. 5 All four patients received a uniform management approach with immediate transfer to an intensive care unit with combined hyperacute stroke unit and neurosurgical presence on initial presentation. Urgent plasma exchange (PLEX) with Octaplas was initiated within 24 h and continued for 5 days. Alongside PLEX, intravenous (IV) immunoglobulin at 1 g/kg in two divided doses and high-dose steroids (1 g IV methylprednisolone followed by 20 mg dexamethasone IV/oral for 4 days) were started irrespective of the platelet count, as the benefit of the steroids outweighs the side effects related to a short course of steroids. Anticoagulation was initiated with argatroban at presentation despite the low platelet count due to concern of the progressive CVST. Target activated partial thromboplastin time ratio was 1.5-2.0 for the first 48 h, increased to 2.0-2.5 in the absence of further bleeding or deterioration on repeat imaging in those with subarachnoid hemorrhage. There was no progression of ICH with argatroban, and platelet count gradually improved following the start of IV IgG and high-dose steroids 5 ( Figure 1A-D). A single dose of rituximab (375 mg/m 2 ) was given to two patients ( Figure 1A There was no demonstrable antispike antibody response following the ChAdOx1 nCoV-19 vaccine, although two had received rituximab at least 10 days following vaccination. However, all subsequently showed a good response following mRNA BNT162b2 (Pfizer-BioNTech) vaccination 12-16 weeks after the diagnosis of VITT (Table S2). Clinically symptomatic thrombosis or increase in anti-PF4-heparin IgG ELISA did not occur in any patient year, which represents a distinct cause of stroke primarily affecting young adults. 8 Factors that can contribute CVST in general are multiple, including those associated with VTE and specific local causes such as local infections, trauma, and brain tumors. 9 Overall death and dependency rate following CVST is around 15%. 10 However, presence of ICH is generally associated with a high mortality in patients with CVST. Around 40%-60% of patients with CVST can have evidence of ICH including parenchymal/subdural hematomas and subarachnoid hemorrhages. 11 The frequency of ICH was around 36% (40/110) in patients with CVST associated with VITT, and mortality was (73%). 2 Evidence on management of CVST in general is limited, and initial anticoagulation with either UFH or low-molecular-weight heparin is currently recommended in patients without VITT, irrespective of the presence of ICH. 10 Nonheparin parenteral anticoagulation such as argatroban was used for patients with VITT, especially those presenting with ICH. However, there was no significant difference in the mortality among patients who received heparin following VITT The main limitation of this study is that it is based on only four patients. However, the study reports an excellent outcome in all four patients.
The four cases of VITT presenting with CSVT in this case series represented an excellent outcome likely attributable to an aggressive treatment approach with multidisciplinary team involvement from the outset.

AUTH O R CO NTR I B UTI O N S
DJA designed the study, collected and interpreted the data, drew some of the figures, and wrote and reviewed the manuscript. CCT collected the data, drew some of the figures, and wrote the manuscript. NA and SR collected the data and prepared the radiological figures. ML collected the data and reviewed the manuscript. All authors reviewed and approved the final version of the manuscript.

FU N D I N G I N FO R M ATI O N
DJA is funded by MRC UK (MR/V037633/1).