5. Considerations related to Cerebral Venous Thrombosis in Special Circumstances

The management of CVT occurring in certain contexts may be different from typical approaches to managing acute symptomatic CVT. The optimal management of CVT associated with head trauma, and CVT found incidentally on scans performed for other indications, is not clear. Head trauma is a well-documented risk factor for CVT, although the prevalence is not well-characterized. A US-based health services study found that 11% of CVT cases were associated with trauma (Otite et al. 2020). Skull fractures involving a venous sinus are at higher risk (Bokhari et al. 2020). Regarding incidental CVT found on scans for other indications, one Canadian study estimated that 11% of CVT cases at a single centre were incidental findings (Zhou et al. 2022).

The management of CVT occurring in association of Vaccine-induced Thrombosis with Thrombocytopenia (VITT) is distinct from the management of non-VITT-associated CVT. VITT is a very rare autoimmune complication of adenovirus vector-based COVID vaccinations affecting 1/26500 to 1/1273000 individuals with first doses of the AstraZeneca vaccine (ChAdOx1 nCoV-19) administered (Pai 2022). CVT was a common complication seen with VITT; immunomodulation is a cornerstone of management.

Data from the earlier part of the COVID-19 pandemic suggest that there is a heightened risk of CVT associated with recent COVID infection. Management, however, does not differ from that of non-COVID-associated CVT.

1. Integration of care across all disciplines for people with CVT to efficiently manage appointments and ensure coordination of care, especially during transition from inpatient to outpatient and community-based care.
2. Support for ongoing research into diagnosis and management for individuals with CVT from a range of causes.

System Indicators: 

1. Number of individuals who experience CVT who are admitted to hospital annually. 
2. Proportion of people who experience a CVT related to major trauma or other primary diagnoses.
   

Process Indicators: 

3. Proportion of incidentally-diagnosed CVT referred for hematology assessment.
4. Proportion of individuals with CVT who have follow-up with a stroke specialist.
   

Patient-oriented outcome and experience indicators:

5. Mortality rates for individuals with other health conditions (e.g., COVID-19, VITT) who experience a CVT related to that condition (stratified by comorbidity).

Measurement Notes

None

Resources and tools listed below that are external to Heart & Stroke and the Canadian Stroke Best Practice Recommendations may be useful resources for stroke care. However, their inclusion is not an actual or implied endorsement by the Canadian Stroke Best Practices writing group. The reader is encouraged to review these resources and tools critically and implement them into practice at their discretion.

Healthcare provider information

• CVT Consortium
• Stroke Engine

Information for individuals with lived experience of stroke, including family, friends and caregivers

• Heart & Stroke: Cerebral Venous Thrombosis Infographic
• Heart & Stroke: Your stroke journey
• CVT Consortium
• Heart & Stroke: Post-stroke checklist
• Heart & Stroke: Online and peer support
• Stroke Engine

Evidence Table and Reference List

Trauma-associated CVT

Head trauma is a well-documented risk factor for CVT, although rates are challenging to ascertain from observational CVT cohorts, which may focus primarily on individuals presenting with a diagnosis of new symptomatic CVT. A recent Canadian single-center study found that one-quarter of 289 CVT cases identified over a 10-year period through discharge diagnosis coding and validated through chart review were associated with trauma (Zhou et al. 2022) and an US-based study using State Inpatient data from New York and Florida found that 11.3% of cases identified between 2006-2016 were associated with a comorbid code for trauma (Otite et al. 2020).

Estimates for rates of CVT complicating head trauma are evolving as use of routine vascular imaging continues to increase. One Chinese single-centre study of 240 consecutive patients with moderate-to-severe closed traumatic brain injury found a CVT on CT venography or MR venography in 16.7%. (Li et al. 2015). Injuries crossing a venous sinus, specifically skull fracture or epidural hematoma, were found to be independent risk factors for CVT. A meta-analysis focusing specifically on patients with skull fracture reported a pooled frequency of 26.2% (Bokhari et al. 2020).

The body of literature examining secondary injury attributable to CVT after head trauma is small, and with methodological limitations. Rates of venous infarction and edema reported in three studies including adults were highly variable (5-46%) (Netteland et al. 2022). One study reported rates of secondary ICH attributable to the CVT in 11% of 73 patients. Members of the writing group note that this high rate is at odds with our collective clinical experience (Netteland et al. 2020). The aforementioned systematic review examined the available evidence for use of anticoagulation or specific treatment regimens did not find specific comparative studies but noted that the majority of studies including adults included a subset of patients treated with anticoagulation, (Netteland et al. 2022) although the writing group notes that this includes patients treated with both prophylactic as well as therapeutic doses. In the absence of supportive evidence for a particular strategy, case-by-case collaborative management is recommended.

Incidentally diagnosed CVT

With increased use of routine vascular neuroimaging, incidental CVT may be diagnosed more frequently. A single-centre Canadian study found that 11% of CVT cases identified between 2008-2018 using administrative data and verified through chart review were new incidental diagnoses. In the general VTE literature, the majority of prognostic studies on incidental VTE focus on populations with cancer. One registry including a non-cancer population (n=68 incidental, 1501 symptomatic) found that 90-day VTE recurrence was similar after incidental versus symptomatic VTE (1.5% vs. 2.3%, HR 1.02, 95% CI 0.30-3.42) (Spirk et al. 2021). Thus, assessment for suitability for anticoagulation is warranted. Ophthalmological assessment should be considered given that patients may be unaware of visual deficits complicating CVT.

COVID-19-associated CVT

COVID-19 has been associated with increased risk of CVT in both community- and hospital-based cohorts. Community-based studies have cited incidence rates with SARS-CoV-2 infection that are substantially higher than baseline incidence rates, though estimates have varied widely. A study using US-based administrative data reported a rate of 42.8 (95% CI 28.5 - 64.2) per million within the first two weeks of infection (Taquet et al. 2021); a Singapore-based study examining radiologically-confirmed CVT diagnosed within 6 weeks of SARS-CoV2 infection estimated an incidence rate of 83.3 (95% CI 30.6 - 181.2) per one hundred thousand person-years based on the total number of SARS-CoV2 infections reported in Singapore over the same time period (Tu et al. 2022). Estimates from hospitalized patients are also variable. One study using US hospital-based administrative data estimated an overall rate of 231 per million person-years in patients hospitalized with COVID-19 (95% CI 152 - 351) (McCullough-Hicks et al. 2022); a meta-analysis of case series estimated a rate of CVT in hospitalized patients with COVID-19 that was 0.08% (95% CI 0.01% - 0.5%) (Baldini et al. 2021). Mortality rates with COVID-19-associated CVT may be higher than with non-COVID-associated CVT, although reporting biases cannot be excluded and small numbers and limited details make it challenging to ascertain if heightened mortality is due to worse CVT severity versus other medical issues (Siegler et al. 2023).

Vaccination

A history of CVT is not a contraindication to receiving mRNA vaccinations against COVID-19 or vaccinations against other infections. One observational study of 62 patients with a history of CVT receiving COVID-19 vaccination (69% Pfizer, 11% Moderna, 11% AstraZeneca ChAdOx1 and 9% Janssen Ad26.COV2.S found no thrombotic recurrences within 30 days of vaccination (95% CI 0.0 - 5.8%) (Gil-Díaz et al. 2022). In the general population, most studies have not found an increase in the risk of CVT following mRNA COVID-19 vaccination (Cari et al. 2021; Houghton et al. 2022; Simpson et al. 2021); one UK population-based study identified a small increased risk of CVT associated with mRNA vaccination on the order of 1 per 500000 doses (Hippisley-Cox et al. 2021; Nicholson et al. 2022). One retrospective study from the Mayo Clinic Health system that also examined risk associated with 10 common non-COVID-19 vaccines (n=771805 doses) found no difference in risk of CVT in the 30 days pre- versus post-vaccination (Pawlowski et al. 2021). 

VITT-associated CVT

Vaccine-induced immune thrombotic thrombocytopenia (VITT) was first identified as an entity in 2021, occurring as a rare complication after adenovirus vector-based vaccination against COVID-19 (ChAdOx1 nCoV-19 [AstraZeneca-Oxford] and Ad26.COV2.S [Janssen/Johnson+Johnson]). Antibodies directed against platelet factor-4 (PF4) were soon identified in association with the disorder. 

Early case series of patients who were later identified as having confirmed or suspected VITT included patients with thrombocytopenia and venous and arterial thromboembolic events, but with a preponderance of CVT in particular (Klok et al. 2022). An international cohort comparing VITT-associated CVT to non-VITT-associated CVT found that the former was associated with a more fulminant course at presentation than non-VITT CVT, with higher rates of mortality, intracerebral hemorrhage and use of EVT and hemicraniectomy (Sánchez van Kammen et al. 2021b). 

Overall, VITT is extremely rare, although incidence rates have varied widely, ranging from 1/265,000 to 1/127,000 per first doses and 1/518,181 after second doses of ChAdOx1 nCoV-19 (AstraZeneca-Oxford) vaccination, respectively, and 1/263,000 Ad26.COV2.S (Janssen/Johnson+Johnson). Variable rates have been attributed to a number of factors including demographic differences between cohorts and differences in reporting structure (Klok et al. 2022; Pai 2022). Diagnostic criteria for the syndrome have evolved, but the UK Expert Hematology Panel (Pavord) criteria have been used in the Cerebral Venous Sinus Thrombosis With Thrombocytopenia Syndrome Study Group, and include: onset of symptoms 5–30 days (5–42 days if isolated DVT or PE) after COVID-19 vaccination, presence of thrombosis, thrombocytopenia (platelet count <150 × 109 cells per L), D-dimer concentration of more than 4000 FEU), and positive anti-PF4 ELISA assay (Pavord et al. 2021). Diagnosis of VITT is considered definite if all five criteria are present and probable if one is missing. Later studies have identified patients without demonstrable thrombosis who otherwise meet criteria. 

Thrombocytopenia in non-VITT CVT is unusual, with a prevalence of 8% in a study of 865 patients from the International CVT Consortium (Sánchez van Kammen et al. 2021a). The mechanism of VITT has been likened to heparin-induced thrombocytopenia, which similarly has antibodies directed against PF4, although typical HIT, unlike VITT, is less commonly complicated by CVT. A meta-analysis of HIT case series reported 1.6% with CVT from 1220 patients with HIT (Aguiar de Sousa et al. 2022). Greinacher and colleagues, supported by a combination of biophysical imaging, mouse modelling and analysis of samples from VITT patients, proposed a two-step process where (1) vaccine components form complexes with PF4, leading to exposure of an epitope (“neoantigen”) while stimulating a proinflammatory response that amplifies production of antibodies against the neoantigen. (2) After several days, there is a sufficient amount of anti-PF4 antibodies to activate platelets; granulocytes, mediated by the presence of PF4-activated platelets, are also activated to release procoagulant neutrophil extracellular traps (NETs) (Greinacher et al. 2021). The reasons why CVT and splanchnic vein thrombosis, another unusual site, were more commonly involved in VITT remains unknown. Selectively persistent procoagulant activity of NETs in central nervous system endothelial cells has been proposed, (Greinacher et al. 2021) as well as procoagulant platelet-derived microparticles (also expressed by PF4) expressing tissue factor, mediating thrombogenesis in the cerebral venous system in particular (Marchandot et al. 2021).

Most management recommendations by national and international bodies thus involved common tenets of (1) immunomodulation, with intravenous immunoglobulin recommended in particular due to selective inhibition of VITT-mediated platelet activation of the FcγRII receptor on PF4 (2) non-heparin-based anticoagulation including DOACs, fondaparinux, danaparoid or argatroban, due to the theoretical risk of worsening the HIT-like response with heparin or heparinoids, (3) supportive care, avoiding platelet transfusions when possible to reduce additional substrate for the autoimmune response (Klok et al. 2022). The prognosis of VITT has improved over time, likely due to a combination of improving awareness with associated earlier diagnosis and treatment, in addition to the establishment of management guidelines alongside evolving understanding of pathophysiology (Scutelnic et al. 2022).

High-risk scenarios

After CVT, it is not known whether targeted prophylaxis would also be suitable in other higher-risk contexts. The American Society of Hematology guidelines for management of venous thromboembolism prophylaxis issued a conditional recommendation with very low certainty for graduated compression stocking or prophylactic LMWH for individuals with a history of VTE embarking on long-distance travel >4 hours (Schünemann et al. 2018). The recommendation did not pertain specifically to individuals with a history of CVT.