3. Post-Acute Management of Cerebral Venous Thrombosis and Person-centered Care

Most patients with CVT are transitioned to oral anticoagulation. Vitamin K antagonists (VKA) have traditionally been recommended by older guidelines (Ferro et al. 2017); however, a recent emerging body of evidence, including three small randomized trials (Connor et al. 2020; Ferro et al. 2019; Field et al. 2023) and a large retrospective observational study (Yaghi et al. 2022b) suggest that, in selected patients, DOACs may also be appropriate, which is also reflected in the updated AHA/ASA Scientific Statement on CVT (Saposnik et al. 2024). In those studies, markers of efficacy, including rates of recurrent CVT and peripheral VTE and venous recanalization are similar between patients treated with DOAC and VKA. Safety events including major extracranial and intracranial hemorrhage were low overall in the randomized trials; the large retrospective ACTION-CVT study found that rates of hemorrhage were lower in patients treated with DOAC compared to VKA as part of their routine clinical care (Yaghi et al. 2022b).

DOACs are not the treatment of choice, however, for patients with known antiphospholipid antibody syndrome (APLAS). Multiple small trials examining DOAC versus warfarin for prevention of venous and arterial thromboembolic events in individuals with APLAS were stopped early due to safety concerns with an excess of thromboembolic events in the DOAC arm (Pengo et al. 2018; Woller et al. 2022), or were completed with similar findings (Ordi-Ros et al. 2019).

Rates of functional independence after CVT are high (85-90%). However, a number of studies have identified high rates of persisting issues with cognition, mood, fatigue and headache impacting quality of life. While retrospective studies report high rates of persisting symptoms months and years following CVT, (Hiltunen et al. 2016; Koopman et al. 2009) participants in the recent prospective SECRET trial found that these symptoms continued to improve on average over one year of follow-up (Field et al. 2023). Late seizures affect approximately 10% of patients with CVT (Sánchez van Kammen et al. 2020).

The needs, goals and experiences of people with lived experience of CVT are highly individualized. People with lived experience highlighted that approaches to ongoing management of CVT and other post-acute sequelae of CVT need to be person-centered and involve an interdisciplinary team as required. People with lived experience emphasized the importance of ongoing mental health support, as well as addressing “invisible” impairments such as cognitive impairment, visual changes, and headaches.

Throughout rehabilitation and recovery, people with lived experience highlighted the value of feedback from their healthcare provider on progress as well as areas in which progress may be slower or lacking. This is helpful to better understand their impairments and recovery post-stroke. They also emphasized the importance of ongoing follow-up, as rehabilitation and recovery needs and goals change over time. Peer support was identified as helpful for people with lived experience, in particular those who are at similar life stages and with other shared experiences. Person-centred support for return to work or school was very important among PWLE. Individuals with CVT also emphasized the importance of support and information for family members and caregivers.

1. Health promotion efforts that contribute to the prevention of stroke and CVT in all communities (integrated with existing chronic disease prevention initiatives) must be established. 
2. Improved communication and transition planning between all stages and settings of care and ensuring that primary care team members are fully informed on the goals of care, prevention therapies initiated by the healthcare providers during first assessments (e.g., in the emergency department), follow-up appointments for further investigations and long-term management. 
3. Coordinated efforts among stakeholders such as the Heart and Stroke Foundation, public health agencies, ministries of health and care providers across the continuum to produce patient, family and caregiver education materials with consistent information and messages on risk factor management.
4. Coordinated processes for ensuring access to and awareness of educational materials, programs, activities and other media related to risk factor management by healthcare professionals, patients and caregivers, including promotion of educational material and effective dissemination mechanisms.

System Indicators: 

1. Annual occurrence rates for CVT in each province and territory. 
2. Number of individuals who experience CVT who are admitted to hospital annually.

Process Indicators: 

3. Number of individuals who are readmitted to hospital within 30 days, 90 days and one year following initial admission for CVT, stratified by reason for readmission. 
4. Number of visits to acute care post-discharge for individuals with CVT.
5. Number of follow-up investigations performed on individuals with CVT, stratified by age and sex (especially repeat CT/CTVs in young people) within first year following diagnosis. 
6. Proportion of individuals with CVT who receive a referral for rehabilitation assessment at discharge and at one year.
7. Proportion of individuals with CVT who receive referrals to subspecialists, stratified by specialty (e.g., Ophthalmology, Hematology) in hospital and at 30 days following discharge.

Patient-oriented outcome and experience indicators: 

8. CVT mortality rates across provinces and territories, including in-hospital or 30-day rate and one-year rate (KQI). 
9. Quality of life rating at 30 and 90 days for individuals with CVT, using a validated tool.
10. Proportion of individuals with CVT unable to return to work or school following index event.
11. Proportion of individuals with CVT who experience adverse effects that impact daily living – e.g., changes in mood, fatigue, cognition, ongoing headaches at 30 and 90 days following diagnosis.

Measurement Notes

1. Validated tools should be used to measure quality of life, mood and cognitive changes and other impacts.
2. For Indicator #5: Due to the smaller incidence rates of CVT compared to other stroke types, numerators and denominators may become very small when looking at multiple sub-categories of complications and stratifying by age and sex. Larger grouping variables may be required.

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 team or Heart & Stroke. The reader is encouraged to review these resources and tools critically and implement them into practice at their discretion.

Healthcare provider information

• CSBPR Cerebral Venous Thrombosis Module: Appendix Four Antiphospholipid Antibody Testing Flowsheet 
• Heart & Stroke: Signs of stroke
• CVT Consortium
• Heart & Stroke: FAST Signs of Stroke...what are the other signs?
• Heart & Stroke: Post-Stroke Checklist
• Heart & Stroke: Virtual Stroke Care Implementation Toolkit
• Heart & Stroke: Taking Action for Optimal Community and Long-Term Stroke Care (TACLS) A Resource for Healthcare Providers
• Stroke Engine

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

• Heart & Stroke: Cerebral Venous Thrombosis Infographic
• Heart & Stroke: Signs of stroke
• CVT Consortium
• Heart & Stroke: FAST Signs of Stroke...what are the other signs?
• Heart & Stroke: Your stroke journey
• Heart & Stroke: Post-Stroke Checklist
• Heart & Stroke: Enabling Self-management Following Stroke: A Checklist for Patients, Families and Caregivers
• Heart & Stroke: A family guide to pediatric stroke
• Heart & Stroke: Stroke in young adults: a resource for patients and families
• Heart & Stroke: Secondary Prevention Infographic
• Heart & Stroke: Rehabilitation and Recovery Infographic
• Heart & Stroke: Transitions and Community Participation Infographic
• Heart & Stroke: Virtual Healthcare Checklist
• Heart & Stroke: Online and peer support
• Stroke Engine

Evidence Table and Reference List

Factors related to clinical decision-making for anticoagulation

Please see Section 2.1, “Antithrombotic management,” for discussion of the evidence related to timing of transition from parenteral to oral anticoagulation. 

Vitamin K antagonists have been the longstanding guideline-recommended treatment for secondary prevention after CVT (Ferro et al. 2017). Recently, multiple observational studies and small randomized trials have compared efficacy and safety of DOACs against warfarin for CVT. The RE-SPECT CVT trial randomized 120 individuals with CVT 1:1 to six months with dabigatran 150 mg bid, versus warfarin, target INR 2.0 - 3.0 (Ferro et al. 2019). There were no VTE recurrences in either group at 6 months. There were two major GI hemorrhages in the dabigatran group and one symptomatic intracranial hemorrhage in the warfarin group. There were no differences between groups with respect to recanalization. The pediatric EINSTEIN-Jr trial randomized 114 children with CVT 2:1 to three months of 20 mg equivalent dosing of rivaroxaban versus standard-of-care anticoagulation with either VKA, target INR 2.0 - 3.0, or low molecular-weight heparin (Connor et al. 2020). There was one recurrent VTE in the comparator group. Rates of recanalization were similar between groups. There were 5 clinically relevant non-major bleeding events in the rivaroxaban group and one symptomatic intracranial hemorrhage in the comparator group. The SECRET feasibility trial randomized 50 individuals with CVT 1:1 to a minimum of six months with rivaroxaban 20 mg daily versus standard-of-care anticoagulation with either warfarin, target INR 2.0 - 3.0, or low molecular-weight heparin (Field et al. 2023). At six months, there was one recurrent VTE in the rivaroxaban group. There was one symptomatic intracranial hemorrhage and two clinically relevant non-major bleeding events in the rivaroxaban group. There were no VTE recurrences, major or clinically relevant non-major bleeding events in the comparator group. There were no differences between groups with respect to recanalization. ACTION-CVT was a large retrospective international study comparing safety and efficacy of DOACs versus VKA prescribed to individuals with CVT as part of their routine clinical care (Yaghi et al. 2022b). Apixaban was the most commonly prescribed DOAC (67%), followed by rivaroxaban (18%) and dabigatran (14%), or other or multiple DOACs (3%). Rates of recurrent VTE did not differ between groups (aHR 0.94, 95% CI 0.15 - 1.73). There was a lower risk of major hemorrhage in the DOAC group (aHR 0.35, 95% CI 0.15 - 0.82), primarily driven by a lower risk of ICH. There were no differences in rates of partial or complete recanalization at a median of 345 days (IQR 140-720). A recent systematic review including three randomized trials and 16 observational studies comparing DOACs to VKAs found similar risks between groups with respect to VTE recurrence, major hemorrhage and complete recanalization (Yaghi et al. 2022a). Overall, the literature support using DOACs as an alternative to VKA in lower-risk individuals with CVT. DOACs are now included as an option for anticoagulation in CVT in the updated AHA/ASA guidelines (Saposnik et al. 2024).

Certain groups are excluded, or underrepresented, in studies to date comparing DOACs to warfarin. DOACs are contraindicated in pregnancy and breastfeeding and thus this population has been excluded by design. In addition to those who were pregnant, both RE-SPECT CVT and ACTION-CVT excluded individuals with malignancy, central nervous system infection, trauma and pregnancy. ACTION-CVT and SECRET excluded individuals with known antiphospholipid antibody syndrome. While malignancy-, infection- or trauma-associated CVT were not exclusion criteria for SECRET, these groups are not well-represented in the study. EINSTEIN-Jr had a large proportion of participants with CVT associated with head and neck infection (65%); less so major head trauma (7%) or active cancer (8%). No specific concerns related to use of DOACs were identified in these groups. The role of DOACs in malignancy-associated CVT remains an area of controversy, although randomized trials from the general (i.e. non-CVT) VTE literature suggest that DOAC is an acceptably safe and efficacious alternative to LMWH in malignancy-associated VTE (Agnelli et al. 2020; McBane et al. 2020; Raskob et al. 2018; Schrag et al. 2023; Young et al. 2018).

Clinical trials comparing DOACs to warfarin for prevention of venous and arterial events in antiphospholipid antibody syndrome (APLAS) have demonstrated an excess of thromboembolic events with DOAC. The TRAPS trial, which compared rivaroxaban to warfarin in 120 individuals with high-risk APLAS (triple-positive with a history of previous arterial or venous thromboembolism) was stopped early due to an excess or arterial events in the rivaroxaban arm without any events in the warfarin arm (Pengo et al. 2018). A Spanish trial also examining rivaroxaban versus warfarin in 190 individuals with thrombotic APLAS found an excess of arterial thromboembolic events with rivaroxaban that was nearly double that of the warfarin group (Ordi-Ros et al. 2019). The ASTRO-APS trial, which was limited due to protocol modifications and slow recruitment, compared apixaban (initially 2.5 mg bid later changed to 5 mg bid) against dose-adjusted warfarin in 48 patients with definite (42%), likely (25%) or historical (33%) APLAS (Woller et al. 2022). There was a high rate of arterial thromboembolic events in the apixaban group compared to none in the warfarin group and the study was terminated having recruited one-quarter of its target sample size. 

Duration of anticoagulation

In those without a permanent indication for anticoagulation following CVT, including antiphospholipid antibody syndrome, active malignancy, or major-risk hereditary thrombophilia, the optimal duration of anticoagulation for CVT is not known. 

Previous AHA/ASA and European guidelines for the management of CVT recommend the initial use of parental heparin followed by transition to oral vitamin K antagonists (VKA) for 3-12 months in the context of transient risk factors, or indefinitely in the context of chronic major risk factors for thrombosis or recurrent VTE (Einhäupl et al. 2010; Ferro et al. 2017). Previous surveys of physician practices, and the recent Canadian SECRET randomized trial, suggest that most patients without an indication for permanent therapy are currently treated for 6-12 months (Coutinho et al. 2011b; Field et al. 2017).

This approach, however, diverges somewhat from current recommendations around management of DVT/PE from the general VTE literature. For DVT/PE, the first 3-6 months following VTE are considered as the “primary treatment” phase, with subsequent ongoing therapy termed “secondary prevention.” In the general VTE literature, events that are provoked in the context of a transient risk factor receive 3-6 months of primary treatment without secondary prevention. Events provoked by a chronic risk factor, however, may be indefinitely anticoagulated given the net clinical benefit of long-term versus short-term anticoagulation in this population, with decision-making based on a follow-up period of approximately 2 years. Overall, risk of bleeding is increased approximately two-fold, while relative rates of VTE recurrence and mortality are reduced by 20-30% and 75%, respectively.

How these results should be extrapolated to the CVT population, however, is unclear. The CVT population includes a high proportion of younger women with transient sex-specific provoking risk factors, including oral contraceptives and the puerperium. Outside of high-risk thrombophilias and those with a history of recurrent events, overall risks of recurrent CVT and other VTE appear to be low (Shu et al. 2022). However, certain groups, including those with unprovoked events, men, and heterozygotes for genetic thrombophilias such as Factor V Leiden and prothrombin gene mutation, (Palazzo et al. 2017; Pires et al. 2019) may have a higher risk. Estimated risks of recurrence are somewhat variable. A review of 4 observational studies from the American Society of Hematology estimated an overall rate of VTE recurrence of 38/1000 over the first year (Dentali et al. 2012a; Martinelli et al. 2010; Miranda et al. 2010; Palazzo et al. 2017). The recent retrospective ACTION-CVT study found a recurrence of 51 recurrent VTE (de novo recurrent CVT and peripheral VTE) per 1000 patient-years (Shu et al. 2022). There are additional inconsistencies in the literature around whether risk of recurrence is heightened in the first year as compared to subsequent years, (Shu et al. 2022) versus a more linear increase over subsequent years (Palazzo et al. 2017).

Additional information regarding duration of shorter versus longer durations of primary anticoagulation will be brought forth by the ongoing EXCOA trial comparing 3-6 months versus 12 months of anticoagulation (Miranda et al. 2018).

Refer to Section 3.2, “Role of Routine Follow-up Vascular Neuroimaging” around the role of repeat neuroimaging as it relates to duration of anticoagulation. 

CVT workup: cancer screening

There is no strong evidence to date to suggest that individuals with CVT should receive enhanced cancer screening. A recent Danish population-based study with a median follow-up of 6.2 years found that overall, rate of incident cancer was not significantly higher in individuals with a diagnosis of CVT. Of 811 patients with CVT, 43 had an incident cancer diagnosis over time, rates that were similar to another recent Swedish population-based study (Sipilä et al. 2022). Risks of incident cancer compared to the general population were elevated over the first year following diagnosis, however, these high rates were driven by a small number of cases overall. The authors estimated that the number of patients to be screened in the six months after CVT to detect one additional incident cancer was 85.5 (95% CI 55.3 - 188.2) overall, 122.1 (95% CI, 71.7 - 411.2) in patients aged 18-54 years and 47.5 (26.2 - 258.8) in those aged >55 years (Skajaa et al. 2023). These numbers needed to screen are similar to those of DVT/PE.

Testing for Janus Kinase V617F mutations in people with CVT without signs or symptoms of myeloproliferative neoplasm (MPN) remains an area of controversy (Xavier et al. 2011). Larger cohorts of mostly unselected CVT patients from Italy, India and Israel report rates of 5.6-6.5% of the JAK2 V617F mutation (De et al. 2012; Lamy et al. 2017; Passamonti et al. 2012; Simaan et al. 2023). All series note that many diagnosed with the mutation at the time of their CVT did not meet criteria for MPN diagnosis.

CVT workup: hypercoagulability testing

Practices and recommendations around hypercoagulability testing following VTE continue to evolve. The recent guidelines from the American Society of Hematology for thrombophilia testing for management of venous thromboembolism included a conditional recommendation for patients with CVT where anticoagulation would otherwise be discontinued (Middeldorp et al. 2023). This recommendation was based on an estimate that, based on an annual recurrent risk of 38/1000/year (Dentali et al. 2012a; Martinelli et al. 2010; Miranda et al. 2010; Palazzo et al. 2017), that a strategy of testing for thrombophilia followed by indefinite anticoagulation in patients with thrombophilia, and stopping anticoagulation in patients without thrombophilia, would result in 18 (range 14-23)/1000 fewer recurrent VTE compared to a no-testing strategy.

Recommendations related to hypercoagulability testing for specific conditions have been published previously by the International Society on Thrombosis and Haemostasis and the International Society for Laboratory Hematology (Baker et al. 2020; Barbhaiya et al. 2023; Devreese et al. 2014; Marlar et al. 2021; Tripodi et al. 2020).

CVT workup: antiphospholipid antibody testing

Please see Section 3.1, Factors related to clinical decision-making for anticoagulation, for a summary of the evidence related to warfarin versus DOACs in people who have a diagnosis of APLAS. 

Given that a diagnosis of APLAS would affect decisions around choice of antithrombotic, it is reasonable to test for antiphospholipid antibodies in patients with CVT who do not have a previous diagnosis. In the absence of a history of APLAS or clinical suspicion of APLAS, however, it is reasonable to proceed with anticoagulation as if the patient did not have APLAS, without waiting for test results, and then altering the anticoagulant accordingly if testing is positive. The prevalence of APLAS-associated CVT is not known. A systematic review cited rates of 6-17% of positive antiphospholipid antibody testing in previous series, (Silvis et al. 2016) although definitions varied between studies and up to 5% of healthy individuals are noted to have antiphospholipid antibodies (Dabit et al. 2022).

Role of routine follow-up vascular neuroimaging

The literature comparing diagnostic accuracy of contrast-enhanced CT venography and contrast-enhanced MR venography to non-enhanced neuroimaging mainly relates to diagnosis of CVT, and not recanalization on follow-up imaging. One small study comparing contrast-enhanced versus time-of-flight MR venography in 6 patients with CVT undergoing follow-up imaging found that 10/15 venous segments on contrast-enhanced MRV with peripheral enhancement and a central non-enhancing filling defect showed peripheral continuous channel morphology on time-of-flight MR venography. However, the study was cross-sectional and did not compare modalities with respect to assessing recanalization from baseline scans (Leach et al. 2007).

Whether degree of venous recanalization should inform duration of anticoagulation remains an area of uncertainty (Aguiar de Sousa et al. 2020; Aguiar de Sousa et al. 2018b; Ferro et al. 2022; Kim et al. 2023). Although a subset of clinicians will modify their duration of anticoagulation based on the degree of venous recanalization on repeat neuroimaging, (Field et al. 2017) it is unclear if this strategy is beneficial. An early prospective neuroimaging study found that 68% of patients experienced partial, and 4%, full recanalization, after one week of anticoagulation. By day 90, 95% of patients had partial (41%) or complete (54%) recanalization (Aguiar de Sousa et al. 2020). A substudy of the ACTION-CVT study found that 88.2% of patients had partial (48.5%) or complete (39.7%) recanalization. Of those patients who had complete recanalization, 59% were noted to be fully recanalized by three months, with 15.4% of additional patients achieving complete recanalization by 6 months and 16.7% by 12 months (Salehi Omran et al. 2023).

A recent prospective neuroimaging study noted that anticoagulated patients with a diagnosis of CVT who had at least partial recanalization within the first 8 days of treatment had fewer new non-hemorrhagic lesions and less extension of pre-existing non-hemorrhagic lesions. However, this was not associated with a reduction in headache or improved functional outcomes at day 90 (Aguiar de Sousa et al. 2020). Most recanalization literature focuses on later recanalization, past the three-month mark. A recent meta-analysis of observational data found that complete or partial venous recanalization was associated with an improved odds of a favourable functional outcome as compared with no recanalization as well as lower risk of recurrence and less presence of common headache (Aguiar de Sousa et al. 2018b). However, there was significant heterogeneity between studies, and the directionality of the association between recanalization and outcomes remains uncertain. The authors note that in meta-regression, the relationship between recanalization and functional outcomes was modified by sex. A prospective substudy of RESPECT-CVT found no association between recanalization and functional outcome (Ferro et al. 2022). Whether recanalization status is causally related to other post-CVT sequelae, including visual loss, cognition or development of dural arteriovenous fistulae (dAVF), is not known.    

At present, there is no strong evidence to suggest routine follow-up neuroimaging to screen for dAVF after CVT. A substudy of the RE-SPECT CVT study found that out of 112 patients, none developed dAVF at six months on repeat MR venography (Ferro et al. 2020). A retrospective study from the international CVT consortium found that out of 1218 patients, 2.4% were found to develop dAVF on neuroimaging performed at a median of 6 months (IQR 5-12). Risk factors for dAVF included male sex, older age and late presentation with CVT (>30 days following symptom onset) (Lindgren et al. 2022).

Management of other post-acute sequelae of CVT

Although rates of functional independence after CVT are high, survivors are noted to have reduced quality of life, with a high prevalence of residual symptoms related to headache, depression, fatigue and cognitive impairment. In the Canadian SECRET trial, 72% of participants were functionally independent (modified Rankin 0 - 2) at the time of their diagnosis. However, mean baseline assessments were indicative of mild-moderate depression, substantial-severe impact of headache, substantial fatigue and impaired cognitive performance. On average, participants experienced improvements in all patient-centered metrics over time between baseline and day 180 and at day 365. Other retrospective studies suggest that reduced participation may persist in many survivors. A retrospective study from China including CVT patients who were employed or in school prior to their index event found that 42% had not returned at six months. Aphasia, cognitive impairment and recurrent CVT were independent predictors for an inability to return to previous activities (Liu et al. 2023).

The definition of “late seizures” in the CVT literature generally refers to those experiencing an event at seven days following presentation or later. A combined prospective and retrospective study from the International Cerebral Venous Thrombosis Consortium found that 11% of 1127 patients were noted to have late seizures, median time to onset being 5 months (IQR 1 - 16 months) (Sánchez van Kammen et al. 2020). ASM was prescribed in 45% of patients prior to the first late seizure and in 94% following the first late seizure. Of those with late seizures, 70% were noted to have later recurrence, with a median time to recurrence of 1 month (IQR 0 - 8). However, the authors did not specify what proportion of those with later recurrence were no longer taking ASM. Independent risk factors for late seizures included acute seizures or status epilepticus at presentation, intracranial hemorrhage, subdural hematoma specifically, and decompressive hemicraniectomy. A meta-analysis including the aforementioned study and three smaller studies identified similar risk factors for late seizures (Gasparini et al. 2022). 

Sex, gender and other equity-related considerations

Considerations related to choice of anticoagulant in women who are pregnant or breastfeeding are summarized in Section 4.1, CVT and pregnancy. Warfarin is a known teratogen and individuals with the potential to become pregnant should be counselled to use effective contraception while on this medication. 

Determining duration of primary anticoagulation or strategies around secondary prevention anticoagulation may involve shared decision-making between patients and clinicians. Clinicians are encouraged to consider the potential for heavy menstrual bleeding in discussions with patients who menstruate and to take a thorough menstrual history as well as a complete blood count and ferritin at baseline. Refer to section 4.1, clinical considerations, anticoagulation and heavy menstrual bleeding for further summary of the existing evidence around assessment and management of heavy menstrual bleeding.

In a substudy of the ACTION-CVT study, Black race was an independent risk factor for recurrent VTE (HR 2.13, 95% CI 1.14-3.98), both overall in a sensitivity analysis examining events occurring off oral anticoagulation (HR 2.59, 95% CI 1.17-5.75). Provoking factors were not significantly different between individuals with Black race versus non-Black race, with lower rates of Factor V Leiden and prothrombin gene mutations in individuals with Black race. Black race was associated with lower rates of follow-up with INR checks in warfarin-treated patients. The authors concluded that, similar to what has been reported in the overall stroke literature, (Towfighi et al. 2023) that the increased risk associated with Black race was likely due to social determinants of health, including disparities in access to care, structural racism and socioeconomic inequities (Shu et al. 2022).