Cerebral venous thrombosis: Treatment and prognosis

INTRODUCTION — Cerebral venous thrombosis (CVT) is an uncommon but serious disorder. Clinical manifestations can include headache, papilledema, visual loss, focal or generalized seizures, focal neurologic deficits, confusion, altered consciousness, and coma.

Many cases have been linked to inherited and acquired thrombophilias, pregnancy, puerperium, infection, and malignancy. Infarctions due to CVT are often hemorrhagic and associated with vasogenic edema. (See "Cerebral venous thrombosis: Etiology, clinical features, and diagnosis".)

Treatment, which is started as soon as the diagnosis is confirmed, consists of reversing the underlying cause when known, control of seizures and intracranial hypertension, and antithrombotic therapy. Anticoagulation is the mainstay of acute and subacute treatment for CVT.

This topic will review the prognosis and treatment of CVT. Other aspects of this disorder are discussed separately. (See "Cerebral venous thrombosis: Etiology, clinical features, and diagnosis".)

ACUTE ANTITHROMBOTIC MANAGEMENT — While the overall aim of treatment for CVT is to improve outcome, the immediate goals treatment for CVT are [1-4]:

●To recanalize the occluded sinus/vein

●To prevent the propagation of the thrombus, namely to the bridging cerebral veins

●To treat the underlying prothrombotic state, in order to prevent venous thrombosis in other parts of the body, particularly pulmonary embolism, and to prevent the recurrence of CVT

The main treatment option to achieve these goals is anticoagulation, most commonly using either heparin or low molecular weight heparin (LMWH), as discussed in the next section.

Initial anticoagulation

Management for most patients — For most patients with CVT, we recommend anticoagulation with subcutaneous LMWH or intravenous heparin for adults with symptomatic CVT who have no contraindication.

The presence of hemorrhagic venous infarction, intracerebral hemorrhage, or isolated subarachnoid hemorrhage are not contraindications for anticoagulant treatment in CVT. The evidence summarized below (see 'Efficacy' below) suggests that subcutaneous LMWH is more effective than unfractionated heparin (UFH) and is at least as safe. Therefore, we prefer subcutaneous LMWH unless the patient is clinically unstable or invasive interventions such as lumbar puncture or surgery are planned or there is a contraindication to LMWH, such as kidney failure.

Treatment for children during the acute phase of CVT is similar to that for adults, but the evidence is weaker since there are no large randomized trials in this age group [5].

Treatment regimens for patients with CVT and heparin-induced thrombocytopenia (HIT) are discussed separately. (See "Management of heparin-induced thrombocytopenia".)

The duration of antithrombotic treatment is reviewed below. (See 'Long-term anticoagulation' below.)

Management for patients with COVID-19 vaccine-associated thrombosis — Case series have described instances of CVT associated with thrombocytopenia in patients who are between 5 and 30 days post-vaccination with an adenovirus-vector ChAdOx1 nCov-19 (AstraZeneca COVID-19) or Ad26.COV2.S (Janssen COVID-19) vaccine [6-11].

These events are due to vaccine-induced autoantibodies against a PF4 platelet antigen, similar to those found in patients with HIT [7]. Treatment for most patients includes anticoagulation (eg, a direct oral anticoagulant, fondaparinux, argatroban) and intravenous immune globulin [12]. Platelet transfusions are typically reserved for cases of clinical relevant bleeding or preoperatively for procedures associated with a high bleeding risk (eg, neurosurgery) [13,14]. (See "COVID-19: Vaccines", section on 'Thrombosis with thrombocytopenia' and "COVID-19: Vaccine-induced immune thrombotic thrombocytopenia (VITT)", section on 'Management'.)

Efficacy — Although definitive evidence of effectiveness is lacking, there is a general consensus that anticoagulation with UFH or LMWH is appropriate treatment for acute CVT, based on available data on efficacy as well as rapid onset of effect and reversibility [3]. As an example, more than 80 percent of the patients in the International Study on Cerebral Vein and Dural Sinus Thrombosis (ISCVT) were treated with anticoagulation [15].

Two randomized controlled trials of anticoagulation in acute CVT (table 1) have been published [1,16]. Both have methodologic problems, most importantly their modest sample size.

●The Berlin trial of intravenous heparin versus placebo was stopped prematurely because of excess mortality in the placebo arm [16]. Patients randomized to the heparin arm had significantly better outcomes on a nonvalidated composite CVT severity scale than those in the placebo group. The average length from onset of symptoms to anticoagulation treatment, four weeks, was exceptionally long.

●The Dutch trial of subcutaneous nadroparin versus placebo enrolled 60 patients but excluded those who needed lumbar punctures for the relief of increased intracranial pressure [1]. More patients treated with LMWH followed by oral anticoagulation had a favorable outcome than controls, but the difference between the groups was not statistically significant (table 1). Despite randomization, an imbalance at baseline may have favored the placebo group, as there were more cases with isolated intracranial hypertension in the placebo group and more patients with infarcts in the nadroparin group.

A meta-analysis of these two trials found that anticoagulant treatment compared with placebo was associated with a pooled relative risk of death of 0.33 (95% CI 0.08-1.21) and a risk of death or dependency of 0.46 (95% CI 0.16-1.31) [17]. While these data suggest that anticoagulant treatment for CVT may be associated with a reduced risk of death or dependency, the results did not achieve statistical significance.

Limited data suggest that LMWH is more effective than UFH and at least as safe for the treatment of CVT:

●In an open-label trial, 66 adults with CVT were randomly assigned to treatment with LMWH or UFH [18]. In-hospital mortality was significantly lower in the LMWH group (0 versus 19 percent). At three months, the proportion of patients with complete recovery was greater for the LMWH group (88 versus 63 percent), but the difference was not statistically significant. Small numbers limit the strength of these findings.

●In a nonrandomized case-control study, a greater proportion of adult patients treated with LMWH (n = 119) compared with UFH (n = 302) were independent at six months (92 versus 84 percent; adjusted odds ratio 2.4, 95% CI 1.0-5.7) [19]. Treatment with LMWH was also associated with slightly lower rates of mortality (6 versus 8 percent) and new intracranial hemorrhage (10 versus 16 percent), but these outcomes were not statistically significant.

●In a single-center double-blind trial conducted in Iran, 52 cases of CVT were randomly assigned to treatment with LMWH or UFH. There was no difference between the treatment groups in neurological deficits, disability, and mortality [20].

Risk of new intracranial hemorrhage — Anticoagulants appear to be safe to use in adult patients with CVT who have associated intracranial hemorrhage, either intracerebral (such as hemorrhagic venous infarction) or subarachnoid [21]. In the Berlin and Dutch trials, 34 of 79 patients (43 percent) had an intracerebral hemorrhage at baseline [1,16]. None of the patients randomized to heparin developed a new intracerebral hemorrhage. In contrast, a new intracerebral hemorrhage developed in three patients randomized to placebo. Case series have also reported relatively low risks of intracranial hemorrhage (<5 percent) and systemic hemorrhage (<2 percent), and such hemorrhages did not influence outcome [22-25]. These findings are in accordance with the hypothesis that hemorrhage in CVT is caused by the probable mechanism of venous outflow blockage and very high intradural and intravenous pressure, leading to both rupture of venules and to hemorrhagic transformation of venous infarctions.

Similarly, observational data from case series and subgroup analyses of controlled trials suggest that anticoagulant therapy is safe in children with CVT [5,26-32].

Endovascular treatment — For selected adults and children with CVT who develop progressive neurologic worsening despite adequate anticoagulation with subcutaneous LMWH or intravenous heparin, endovascular thrombolysis or mechanical thrombectomy may be a treatment option at centers experienced with these methods.

The potential utility of endovascular treatment was described in a 2015 systematic review that identified 42 studies and 185 patients with CVT who were treated with mechanical thrombectomy [33]. Many of the patients were severely ill; pretreatment intracerebral hemorrhage was present in 60 percent and stupor or coma in 47 percent. A variety of devices were used, including the AngioJet rheolytic catheter, balloon angioplasty, stents, and microsnares; concurrent local thrombolysis was used in 71 percent. Overall, a good outcome was reported for 84 percent of patients; mortality was 12 percent. New or worsened intracerebral hemorrhage affected 10 percent. A high recanalization rate (95 percent; 21 percent partial) was achieved.

However, a randomized controlled trial (TO-ACT) [34] failed to show benefit of endovascular treatment (thrombectomy with or without chemical thrombolysis) over anticoagulation in patients with acute CVT and at least one risk factor for clinical deterioration (coma, mental status disturbance, CVT involving the deep venous system, intracerebral hemorrhage). However, as the trial had a modest sample size and was prematurely stopped for futility, the possibility of a small treatment effect in some patients cannot be excluded.

Among nearly 50,000 patients with CVT from the United States Nationwide Inpatient Sample 2004 to 2014, mortality was higher for patients with CVT treated with endovascular approaches than medical therapy with anticoagulation, even after adjusting for age, severity of symptoms, and the burden of complications (eg, intracranial hemorrhage, venous infarction, and cerebral edema; odds ratio 1.96, 95% CI 1.6-3.3) [35]. In addition, a 2010 systematic review of 15 studies including 156 patients revealed that despite endovascular treatment, there is a death rate of 9 percent and that local thrombolysis is complicated by a non-negligible rate of major bleeding (10 percent), including 8 percent intracranial hemorrhages, 58 percent of which were fatal [36].

Guideline recommendations — Consensus guidelines support the use of anticoagulation for the acute treatment of CVT in adults and children.

Guidelines for acute treatment in adults include:

●The 2017 European Stroke Organization guidelines for the diagnosis and treatment of cerebral venous thrombosis, endorsed by the European Academy of Neurology, recommend heparin at therapeutic dosage to treat adult patients with acute CVT, including those with an intracerebral hemorrhage at baseline [37]. The guidelines suggest using LMWH instead of UFH. No recommendation is made regarding thrombolysis for acute CVT, except that patients who have a pretreatment low risk of poor outcome (eg, absence of coma, mental status disturbance, thrombosis of the deep venous system, intracranial hemorrhage, and malignancy) should not be exposed to aggressive treatments such as thrombolysis.

●The 2014 American Heart Association (AHA)/American Stroke Association (ASA) guidelines for the prevention of stroke state that anticoagulation is reasonable for patients with acute CVT, even in selected patients with intracranial hemorrhage [38]. The 2011 AHA/ASA guidelines for the diagnosis and management of CVT conclude that initial anticoagulation with adjusted-dose UFH or weight-based LMWH in full anticoagulant doses is reasonable, followed by vitamin K antagonists, regardless of the presence of intracerebral hemorrhage [39].

Guidelines for acute treatment in children include:

●The 2012 American Academy of Chest Physicians (ACCP) guidelines recommends initial anticoagulation with UFH or LMWH, followed by LMWH or vitamin K antagonist treatment (ie, warfarin) for a minimum of three months for children with CVT but without significant intracerebral hemorrhage [40]. Anticoagulation for an additional three months is suggested if there is still cerebral sinovenous occlusion or ongoing symptoms (the latter presumably meaning new venous infarcts or increased intracranial pressure) after three months of therapy.

For children with CVT who have significant intracerebral hemorrhage, the ACCP suggests either initial anticoagulation as for children without hemorrhage or radiologic monitoring of the thrombosis at five to seven days and anticoagulation if thrombus extension is noted at that time [40]. The ACCP suggests thrombolysis, thrombectomy, or surgical decompression only in children with severe CVT in whom there is no improvement with initial anticoagulation therapy.

●The 2019 AHA/ASA scientific statement for the management of stroke in neonates and children endorses anticoagulation for children with CVT [41]. Antithrombotic selection should be individualized, guided by patient circumstances, and informed by a multidisciplinary consensus approach particularly when CVT is associated with hemorrhagic infarction, otitis media/mastoiditis, head trauma, or cranial surgery. Surveillance vascular neuroimaging is recommended to guide the duration of anticoagulation. Endovascular intervention is an option for rare circumstances when there is sudden clinical deterioration or a high risk of mortality.

For neonatal patients with CVT, anticoagulation with LMWH or heparin may be considered, particularly those with clinical deterioration or evidence of thrombus extension on serial imaging [41]. Serial imaging at five to seven days should be considered to exclude propagation when a decision is made not to be anticoagulated.

OTHER ACUTE MANAGEMENT ISSUES — Major problems that may require intervention in the acute phase of CVT include elevated intracranial pressure, brain swelling, and seizures.

Elevated intracranial pressure and herniation — In the acute phase, elevated intracranial pressure (ICP) may arise from single or multiple large hemorrhagic lesions, infarcts, or brain edema. Elevated ICP or space-occupying lesions may cause transtentorial herniation and death.

●General recommendations to control acutely elevated ICP should be followed, including elevating the head of the bed, intensive care unit admission, mild sedation as needed, administering osmotic therapy (mannitol or hypertonic saline), hyperventilation to a target partial pressure of carbon dioxide (PaCO₂) of 30 to 35 mmHg, and ICP monitoring [21,42]. (See "Evaluation and management of elevated intracranial pressure in adults" and "Elevated intracranial pressure (ICP) in children: Clinical manifestations and diagnosis".)

●In patients with impending herniation due to unilateral hemispheric lesion, hemicraniectomy can be lifesaving. Retrospective data from a registry and systematic reviews suggest that death can be prevented and a good functional outcome can be achieved [43,44].

There is no good evidence to support ventricular shunting as a treatment for acute hydrocephalus or impending brain herniation in the acute phase of CVT [45].

In patients with sustained ICP elevation, successful treatment of intracranial hypertension can prevent visual failure and resolve headache. A prospective study in 59 patients with CVT presenting with isolated intracranial hypertension noted a complete recovery in over 90 percent [46]; patients had a variety of interventions, and most had a therapeutic lumbar puncture (LP). However, there are no studies specifically evaluating therapeutic LP for elevated ICP or isolated intracranial hypertension in patients with CVT. Data from the International Study on Cerebral Vein and Dural Sinus Thrombosis (ISCVT) did not show differences in the outcomes in patients who had diagnostic LP [15]. European guidelines state that therapeutic LP may be considered in patients with CVT and signs of intracranial hypertension, because of a potential beneficial effect on visual loss and/or headache, whenever its safety profile is acceptable [37].

Although glucocorticoids, in particular intravenous dexamethasone, are prescribed in many centers, they are not recommended for treating CVT in the absence of an underlying inflammatory disorder such as Behçet disease or systemic lupus erythematosus [37,47-49]. No randomized clinical trials have been performed to evaluate their efficacy for CVT, but available evidence suggests they are ineffective. This conclusion is supported by a study that analyzed data from the observational ISCVT cohort of 642 patients with CVT (including 150 patients treated with glucocorticoids) using case-control methods that failed to demonstrate any benefit of glucocorticoids, even for patients with parenchymal lesions [50].

Seizures — For patients with CVT who have both seizures and focal cerebral supratentorial lesions such as edema or infarction on admission head computed tomography (CT) or brain magnetic resonance imaging (MRI) lesions, we recommend seizure prophylaxis with antiseizure medication.

In patients with CVT, recurrent seizures are more likely to develop in those who present with seizures and in those with supratentorial brain lesions (focal edema or ischemic or hemorrhagic infarcts) on admission brain imaging [51]. The risk of developing seizures after CVT diagnosis is very low in patients who do not have these risk factors. (See "Cerebral venous thrombosis: Etiology, clinical features, and diagnosis", section on 'Seizures'.)

Data are limited regarding the effectiveness of seizure prophylaxis with antiseizure medications in patients with CVT [21,52]. In the ISCVT cohort, early seizures (those occurring within two weeks after CVT diagnosis) were observed in the following patient subgroups, comparing those not treated with antiseizure medication versus those treated with antiseizure medication [51]:

●In patients with no supratentorial lesion and no seizure at presentation, early seizure occurred in 5 of 197 (2.5 percent) not on antiseizure medications versus 0 of 11 (0 percent) on antiseizure medications.

●In those with no supratentorial lesion who presented with seizure, early seizures occurred in 1 of 14 (7 percent) not on antiseizure medications and 0 of 35 (0 percent) on antiseizure medications.

●In patients with a supratentorial lesion but no seizure at presentation, early seizures occurred in 11 of 134 (8 percent) not on antiseizure medications and 1 of 35 (3 percent) on antiseizure medications (odds ratio [OR] 0.3, 95% CI 0.04-2.6).

●In patients with a supratentorial lesion who presented with seizure, early seizures occurred in 24 of 47 (51 percent) not on antiseizure medications and 1 of 148 (<1 percent) on antiseizure medications (OR 0.006, 95% CI 0.001-0.05).

Thus, antiseizure medications prophylaxis appears to be associated with a reduced risk of early seizures in patients with CVT. The risk reduction was statistically significant for patients in the highest risk group (those with a supratentorial lesion and seizure at presentation) [51]. The strength of this study is limited by its observational and retrospective design, but it represents the largest experience in the literature.

Based upon these data, we recommend seizure prophylaxis only for patients with both seizures at presentation and supratentorial lesions such as edema, infarction, or hemorrhage on admission head CT or brain MRI. Prophylaxis is not clearly required for a single early symptomatic seizure with CVT in the absence of supratentorial lesion, as there is often no seizure recurrence. Furthermore, seizure prophylaxis is not recommended for patients who have focal cerebral lesions without seizures [37,39].

When antiseizure medication prophylaxis is used, valproate or levetiracetam is preferable to phenytoin because they have fewer pharmacologic interactions with oral vitamin K antagonist anticoagulants (eg, warfarin) [53]. General recommendations for the selection of antiseizure medications are discussed separately. (See "Overview of the management of epilepsy in adults", section on 'Antiseizure medication therapy' and "Seizures and epilepsy in children: Initial treatment and monitoring", section on 'Selection of an antiseizure medication'.)

The duration of antiseizure medication therapy is discussed separately (See 'Seizure prevention' below.)

Infection and inflammation — Antibiotic treatment is mandatory whenever there is meningitis or other intracranial infection or an infection of a neighboring structure, such as otitis or mastoiditis. For associated inflammatory diseases such as Behçet disease, lupus, or vasculitis, treatment with glucocorticoids may be necessary.

MANAGEMENT AFTER THE ACUTE PHASE — The subacute phase of CVT often involves decisions regarding the duration of anticoagulation and antiseizure medication use. In addition, there may be long-term complications including headaches, visual loss, cognitive impairment, and psychiatric disturbances.

Long-term anticoagulation — The aim of continuing anticoagulation after the acute phase is to prevent CVT recurrence, which affects 2 to 7 percent of patients, and to prevent extracerebral venous thrombosis, which occurs in up to 5 percent of patients with CVT, mainly from deep venous thrombosis of the limbs or pelvis, and/or pulmonary embolism [15]. (See 'Recurrence' below.)

Selection of anticoagulant — For most adults with CVT, we suggest anticoagulation with warfarin or a direct oral anticoagulant after the acute phase. Direct oral anticoagulants may be preferable for most patients, based upon the lower burden of blood monitoring and dose adjustments, fewer drug interactions, and lack of dietary restrictions when compared with warfarin. When warfarin is used, the dose should be adjusted to an international normalized ratio (INR) target of 2.5 (acceptable range: 2 to 3).

Although high-quality evidence is limited, either warfarin or a direct oral anticoagulant appears to be safe and effective to prevent recurrent CVT and other forms of venous thromboembolism (VTE) in patients with CVT. A systematic review of 19 studies that included nearly 2000 patients with CVT found rates of recurrent thromboembolism and intracranial hemorrhage were similar for patients treated with vitamin K antagonists or direct oral anticoagulants [54]. However, certainty of the data is limited by the observational nature of many of the included studies, risk of selection and treatment biases, as well as varied outcome parameters. In the open-label RE-SPECT CVT trial, 120 patients with CVT of mild to moderate severity were randomly assigned to dabigatran (150 mg twice daily) or warfarin (titrated to a target INR of 2 to 3) for a period of 24 weeks [55]. Patients with coma, major trauma, central nervous system infections, or active cancer were excluded. During the study period, there were no recurrent venous thromboembolic events in either treatment group. Major bleeding was limited to intestinal bleeding in one patient assigned to dabigatran and subdural hemorrhages in two patients assigned to warfarin. European guidelines, which were published in 2017 prior to the RE-SPECT CVT trial, do not recommend using direct oral anticoagulants for the prevention of recurrent venous thrombosis after CVT [37].

In the retrospective ACTION-CVT observational study that included 845 anticoagulated patients with CVT of mild to moderate severity, clinical and radiographic outcomes were assessed at a median follow-up of 345 days for patients treated with warfarin (52 percent), a direct oral anticoagulant (33 percent), or both at different times (15 percent) [56]. Patients with CVT associated with pregnancy, antiphospholipid syndrome, and cancer were excluded. Compared with warfarin, treatment with a direct oral anticoagulant was associated with similar risks of recurrent venous thrombosis and death, as well as similar rates of recanalization on follow-up imaging. In addition, treatment with a direct oral anticoagulant was associated with a lower risk of major hemorrhage (adjusted hazard ratio 0.35, 95% CI 0.15-0.82).

Special populations of patients with CVT require a different approach:

●Malignancy – For patients with malignancy who require long-term anticoagulation and who do not have chronic kidney failure (creatinine clearance <30 mL/minute), low molecular weight heparin (LMWH) is generally preferred rather than warfarin or direct oral anticoagulants, but oral anticoagulation is preferred over no therapy. (See "Anticoagulation therapy for venous thromboembolism (lower extremity venous thrombosis and pulmonary embolism) in adult patients with malignancy".)

●Antiphospholipid syndrome – For patients with an antiphospholipid antibody syndrome and CVT, clinical efficacy and safety data suggest warfarin is preferred over direct oral anticoagulants. (See "Management of antiphospholipid syndrome".)

●Kidney failure – Patients with severe chronic kidney failure should not receive direct oral anticoagulants; vitamin K antagonists (eg, warfarin) are preferred for oral therapy.

●Pregnancy – For most patients who require long-term anticoagulation during pregnancy (with the exception of patients with mechanical heart valves), heparins are safer than other anticoagulants. Warfarin and direct oral anticoagulants are contraindicated. (See "Use of anticoagulants during pregnancy and postpartum".)

●Children – Several anticoagulants have been used for children with CVT including vitamin K antagonists, unfractionated heparin, or LMWH [5]. Rivaroxaban or dabigatran can also be used to prevent recurrent venous thrombotic events after acute CVT, depending on patient and parent preferences and drug and dietary interactions.

Rivaroxaban was compared with standard anticoagulation (LMWH or vitamin K antagonist) in children with CVT in an exploratory substudy [57] of the open-label EINSTEIN-Jr trial [58], which assigned patients with VTE to bodyweight-adjusted rivaroxaban 20 mg equivalent dose or standard anticoagulation for a period of three months. Partial or complete recanalization rates were similar in both groups and occurred in roughly 75 percent. There was one recurrent VTE (in a child assigned to standard anticoagulation). One subdural hematoma occurred among children assigned to standard anticoagulation (n = 41) and five clinically relevant (nonmajor extracranial) bleeding events in those who received rivaroxaban (n = 73). A subgroup analysis of the EINSTEIN-Jr trial, including children with CVT and an associated head or neck infection administered therapeutic anticoagulants, showed that generally they had low risks of bleeding and thrombotic complications, including those who had surgical interventions with delay or interruption of anticoagulation [59].

In a subgroup analysis of the DIVERSITY trial, a single-arm trial of dabigatran in children with VTE, few children with CVT developed recurrent VTE or experienced major or clinically relevant nonmajor bleeding when receiving prophylaxis with dabigatran [60].

Duration of anticoagulation — After the acute phase of CVT, we suggest continuing anticoagulation for a minimum period of three months and up to 12 months. However, there is no definitive evidence regarding the optimal duration of anticoagulant therapy specifically for reducing the risk of recurrent CVT [37]. A reasonable approach may be to stratify the duration of anticoagulant therapy according to the individual prothrombotic risk as follows [39,61]:

●For patients with a provoked CVT associated with a transient risk factor (table 2), anticoagulation is continued for three to six months.

●For patients with an unprovoked CVT, anticoagulation is continued for 6 to 12 months.

●For patients with recurrent CVT, VTE after CVT, or a first CVT with a severe thrombophilia (ie, homozygous prothrombin gene G20210A variant, homozygous factor V Leiden genetic variant, deficiencies of protein C, protein S, or antithrombin, combined thrombophilia defects, or antiphospholipid syndrome), anticoagulation may be continued indefinitely.

Aspirin — We generally do not use aspirin or other antiplatelet medications for long-term management of patients after CVT, unless a separate indication for therapy is present. The benefit of aspirin or other antiplatelet agents after CVT has not been established in controlled trials or observational studies, and current guidelines make no recommendation about aspirin in this setting [37,39].

Seizure prevention — Patients who experience a seizure after a hemispheric CVT are typically started on antiseizure medication treatment, while patients who do not are generally not started on antiseizure medication prophylaxis. (See 'Seizures' above.)

The optimal duration of antiseizure medication treatment after CVT is unknown. For patients with CVT and associated parenchymal brain lesions who present with one or more seizures in the acute phase, antiseizure medications should be continued until seizure-free for a defined duration (eg, one year).

The risk of epilepsy after CVT ranges from 5 to 11 percent of patients [15,62-64]. The risk is higher in those with seizures in the acute phase, with hemorrhagic parenchymal lesions, and who survived a decompressive hemicraniectomy [63,64]. Late-onset seizures indicate an even higher risk of epilepsy. In one study of 123 CVT patients with late seizures (occurring more than seven days after CVT diagnosis), 70 percent had a recurrent seizure during a mean two-year follow-up [64].

General recommendations for the selection and withdrawal of antiseizure medications are discussed separately. (See "Overview of the management of epilepsy in adults", section on 'Antiseizure medication therapy' and "Overview of the management of epilepsy in adults", section on 'Discontinuing antiseizure medication therapy' and "Seizures and epilepsy in children: Initial treatment and monitoring", section on 'Selection of an antiseizure medication'.)

Headaches — Chronic headaches have been reported to occur in more than half of patients with prior CVT [65]. Headaches severe enough to require bed rest or hospital admission afflict 14 percent of patients with CVT [15]. Repeated brain MRI and magnetic resonance venography (MRV) are necessary to exclude the rare case of recurrent CVT. MRV may depict stenosis of a previously occluded sinus [66,67].

Lumbar puncture may be necessary to exclude chronically elevated intracranial pressure (ICP) if headache persists and brain MRI and MRV are normal. In such cases, therapeutic options to treat elevated ICP include acetazolamide (500 mg twice daily) or topiramate (for patients who cannot tolerate acetazolamide), but efficacy is unproven [39]. Additional options if severe headache associated with increased intracranial hypertension persists include repeated lumbar punctures, a lumboperitoneal shunt, or stenting of sinus stenosis [68-70], but efficacy after CVT is also unproven.

Visual loss — Severe visual loss due to CVT is fortunately a rare event [71-73]. Nevertheless, elevated intracranial pressure must be rapidly ruled out and managed accordingly if visual acuity decreases during follow-up and is not explained by ocular causes. Fenestration of the optic nerve sheath has also been used to relieve pressure and prevent optic nerve atrophy, but efficacy is not established [39,74].

Because of the potential for visual loss caused by severe or long-standing elevation of intracranial pressure, serial assessment of visual fields and visual acuity is recommended for children with CVT during follow-up, particularly during the first year [39,42]. It is reasonable to do the same for adults with visual complaints, chronic headaches, or papilledema.

Cognitive and psychiatric complications — Despite the apparent general good recovery in most patients with CVT, approximately one-half of the survivors feel depressed or anxious, and minor cognitive or language deficits may preclude them from resuming their previous jobs [75,76]. Patients should be reassured of the very low level of risk of recurrence of CVT and be encouraged to return to previous occupations and hobbies. In some cases, antidepressants may be necessary.

Subsequent pregnancy — We suggest prophylaxis with LMWH during pregnancy and puerperium for those with a previous history of CVT to reduce the risk of recurrent CVT and other venous thromboembolic events, in accord with guidelines from the European Stroke Organization and the American Heart Association/American Stroke Association [37,39,77].

For pregnant patients with a history of CVT, we suggest temporary prophylactic anticoagulation with subcutaneous LMWH throughout pregnancy and continuing up to eight weeks postpartum. (See "Deep vein thrombosis and pulmonary embolism in pregnancy: Prevention".)

Pregnancy and the puerperium are known risk factors for CVT (see "Cerebral venous thrombosis: Etiology, clinical features, and diagnosis", section on 'Risk factors and associated conditions'). The absolute risk of complications during subsequent pregnancy among those who have a history of CVT appears to be low, although the relative risks of recurrent CVT and noncerebral VTE are quite elevated compared with the general population. Supporting evidence comes from a 2016 systematic review of 13 observational studies evaluating the frequency of CVT or noncerebral VTE associated with pregnancy and the puerperium in those with a history of previous CVT [78]. The following observations were reported:

●Recurrent CVT occurred in 2 of 217 pregnancies (9 per 1000 pregnancies, 95% CI 3-33 per 1000), a rate that was more than 80-fold higher than the previously reported incidence in the general population.

●Noncerebral VTE occurred in 5 of 186 pregnancies (27 per 1000, 95% CI 12-61 per 1000), a rate that was approximately 16-fold higher than the incidence previously described in the general population.

●Spontaneous abortion occurred in 33 of 186 pregnancies (18 percent, 95% CI 13-24 percent), a rate similar to the estimated rate in the general population. (See "Pregnancy loss (miscarriage): Terminology, risk factors, and etiology", section on 'Incidence'.)

●There was no significant difference in the rate of spontaneous abortion for patients treated or not treated with antithrombotic therapy (11 versus 19 percent).

Thus, based upon the available evidence, a history of CVT, including pregnancy- or puerperium-related CVT, is not a contraindication for future pregnancy.

Patients should be advised not to become pregnant while on warfarin because of its teratogenic effects and increased risk of fetal bleeding. (See "Use of anticoagulants during pregnancy and postpartum", section on 'Already taking warfarin'.)

Oral contraceptives — Because it is a risk factor for CVT, female patients with prior CVT should be informed about the risks of combined estrogen-progestin hormonal contraception and advised against its use [37,42]. Risk is associated with the dose of ethinyl estradiol (less risk with doses <50 mcg). The type of progestin is also associated with VTE risk; in general, the lowest risk is seen with combined oral contraceptives that contain a second-generation progestin such as levonorgestrel. These risks are discussed in greater detail separately. (See "Combined estrogen-progestin contraception: Side effects and health concerns", section on 'Venous thromboembolism'.)

PROGNOSIS — CVT can result in death or permanent disability but usually has a favorable prognosis.

Early deterioration and death — Approximately 5 percent of patients die in the acute phase of the disorder [79,80]. Most of the early deaths are a consequence of CVT. A systematic review found that mortality rates among patients with CVT declined since the 1960s; increased detection of less severe cases with advances in neuroimaging along with improved hospital care may have accounted for some or all of the decreased mortality [81].

In the International Study on Cerebral Vein and Dural Sinus Thrombosis (ISCVT) that evaluated 624 patients (age >15 years) with CVT, 27 patients (4.3 percent) died during hospitalization for the acute phase, including 21 patients (3.4 percent) who died within 30 days from symptom onset [79].

Predictors of mortality at 30 days in the ISCVT were as follows [79]:

●Depressed consciousness

●Altered mental status

●Thrombosis of the deep venous system

●Right hemisphere hemorrhage

●Posterior fossa lesions

Early mortality in children with CVT is similar to that in adults. In a European cohort of 396 children with CVT (median age 5.2 years), death in the first two weeks after presentation occurred in 12 patients (3 percent) [82].

The main cause of acute death with CVT is transtentorial herniation secondary to a large hemorrhagic lesion [79]. Other causes of early death include herniation due to multiple lesions or to diffuse brain edema, status epilepticus, medical complications, and pulmonary embolism [2].

Long-term outcome — Mortality after the acute phase of CVT is predominantly related to underlying conditions. The main ISCVT report [15] performed a meta-analysis of seven prospective series and found that acute CVT was associated with a 15 percent overall death or dependency rate at the end of follow-up, which varied from 3 to 78 months.

In the ISCVT study, complete recovery at the end of follow-up (median 16 months) was noted in 79 percent of the entire cohort of patients, while death was the outcome in 8 percent [15].

Predictors of poor long-term prognosis in the ISCVT were as follows [15]:

●Central nervous system infection

●Any malignancy

●Thrombosis of the deep venous system

●Hemorrhage on head CT or MRI

●Glasgow coma scale score <9 on admission (table 3)

●Mental status abnormality

●Age >37 years

●Male sex

The CVT risk score was designed to estimate the functional prognosis at six months after CVT onset; the score was derived using data from the original ISCVT cohort of 624 patients and validated in two smaller cohorts [83]. The score is tallied as follows:

●Presence of malignancy – 2 points

●Coma on admission – 2 points

●Thrombosis involving the deep venous system – 2 points

●Mental status disturbance on admission – 1 point

●Male sex – 1 point

●Intracranial hemorrhage on admission – 1 point

A CVT risk score ≥3 was associated with a poor outcome, defined as a modified Rankin Scale (table 4) score of >2 (dependency or death), with a high sensitivity but poor specificity (96 and 14 percent, respectively) [83].

In the ISCVT, complete recovery at six months was significantly more common for female than male patients (81 versus 71 percent), while dependency or death was less likely for females than males (12 versus 20 percent) [84]. These differences were driven entirely by the more favorable outcome for the subgroup of females who had sex-specific risk factors (mainly oral contraceptives, pregnancy, or puerperium) for CVT. Females without sex-specific risk factors had outcomes similar to males. (See "Cerebral venous thrombosis: Etiology, clinical features, and diagnosis", section on 'Epidemiology'.)

Intracerebral hemorrhage present at the time of CVT diagnosis was identified in 245 patients (39 percent) of the ISCVT cohort [85]. In this subgroup with early intracerebral hemorrhage, predictors of poor prognosis at six months were older age, male sex, thrombosis of the deep cerebral venous system or of the right lateral sinus, and a motor deficit [85].

By contrast, several studies have found that a good outcome after CVT is predicted when symptoms of intracranial hypertension are the only manifestations of CVT at the time of diagnosis [46,86]. In patients with CVT presenting with isolated intracranial hypertension, a subgroup analysis of the ISCVT cohort found that poor outcome was associated with a longer diagnostic delay [87].

Recurrence — The risk of recurrent CVT is approximately 2 to 4 percent, while the risk of recurrent venous thromboembolism (VTE) in other locations after CVT ranges from 4 to 7 percent. In the ISCVT study, which evaluated 624 patients with CVT over a median 16 months of follow-up, 14 (2 percent) had a recurrent CVT and 27 (4 percent) had other thrombotic events during follow-up [15]. The risks of recurrent CVT or any venous thrombosis were 1.5 and 4.1 per 100 person-years, respectively [88].

Observational data also suggest longer-term CVT recurrence rates appear to be lower than rates of recurrent VTE. In a retrospective cohort study of 706 patients with a first CVT who were followed for 6 to 297 months (median 40 months), CVT recurred in 31 patients (4 percent), and VTE in a different site occurred in 46 patients (7 percent) [89]. In another study that prospectively followed 145 patients with CVT for a median of six years after discontinuation of anticoagulant therapy, a recurrent CVT developed in five patients (3 percent), and other types of VTE developed in another 10 patients (6 percent) [90]. The risks of recurrent CVT or other types of VTE were 0.5 and 2.0 per 100 person-years, respectively.

Risk factors for CVT recurrence in adults include the following [88-91]:

●History of prior VTE

●Polycythemia/thrombocythemia

●Clinical history or laboratory evidence of thrombophilia

●Male sex

●Black race

Data on CVT recurrence and risk factors in children are limited. In the European cohort of 396 children with CVT (median age five years) followed for a median of 36 months, recurrent venous thrombosis occurred in 22 children at a median of six months, including CVT in 13 children (3 percent) [82]. There were no recurrences of CVT among children younger than 25 months. Factors independently associated with recurrent cerebral and systemic venous thrombosis in children were nonadministration of anticoagulant therapy before relapse (hazard ratio [HR] 11.2, 95% CI 3.4-37.0), persistent occlusion on repeat venous imaging (HR 4.1, 95% CI 1.1-14.8), and heterozygosity for the prothrombin (factor II) G20210A variant (HR 4.3, 95% CI 1.1-16.2) [82].

Recanalization — Most patients with CVT achieve some degree of cerebral vein and sinus recanalization. A 2018 systematic review and meta-analysis identified 19 studies, mostly retrospective, that reported recanalization rates for adult patients with CVT who were treated with anticoagulation [92]. The overall recanalization rate was 85 percent among 818 patients who received follow-up imaging, with partial recanalization achieved by 35 percent and complete recanalization by 49 percent of patients. In the few studies with available data, the recanalization rate increased with time, and approximately three-quarters of patients had achieved recanalization at three months. Positive predictors of recanalization were thrombosis of the superior sagittal sinus and female sex; negative predictors of recanalization were multiple thromboses, hormonal therapy, older age, and lack of identified risk factors for CVT.

The meta-analysis also found that recanalization was associated with functional recovery [92]. A favorable outcome, defined as a modified Rankin Scale score (table 4) of 0 (no symptoms) or 1 (no significant disability despite symptoms), was achieved in 319 of 357 (89 percent) patients with recanalization and in 42 of 59 (71 percent) patients without recanalization; the pooled odds ratio for a favorable outcome with recanalization was 3.3 (95% CI 1.2-8.9). In subsequent prospective cohort study of 68 patients with newly diagnosed CVT treated with anticoagulation and followed with MRI and magnetic resonance venography, early venous recanalization (confirmed by imaging on day 8 after starting anticoagulation) was associated with both regression and a reduced risk of enlargement of nonhemorrhagic lesions, including those with venous infarction [93].

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Stroke in adults" and "Society guideline links: Stroke in children".)

SUMMARY AND RECOMMENDATIONS

●Acute anticoagulation – For adults with symptomatic cerebral venous thrombosis (CVT), with or without hemorrhagic venous infarction, we recommend initial anticoagulation therapy with subcutaneous low molecular weight heparin (LMWH) or intravenous heparin (Grade 1C). For children with CVT, with or without significant secondary hemorrhage, we suggest initial anticoagulation therapy with subcutaneous LMWH or intravenous heparin (Grade 2C). (See 'Initial anticoagulation' above.)

●Management of acute complications – Complications that require intervention during the acute phase of CVT include elevated intracranial pressure, brain swelling, seizures, and infection. (See 'Other acute management issues' above.)

•Measures to control acutely increased intracranial pressure and impending herniation, including decompressive surgery, may be required in patients with CVT.

•For patients with CVT who have both seizures at presentation and focal cerebral supratentorial lesions (eg, edema, infarction, or hemorrhage on admission computed tomography or magnetic resonance imaging), we recommend seizure prophylaxis with an antiseizure medication (Grade 1B).

For patients with a single early symptomatic seizure due to CVT in the absence of a supratentorial cerebral lesion, the benefit of seizure prophylaxis is uncertain due to low likelihood of recurrence. Seizure prophylaxis is avoided for those who have focal cerebral lesions without seizures. (See 'Seizures' above.)

•Antibiotic treatment is mandatory whenever there is meningitis or other intracranial infection or an infection of a neighboring structure, such as otitis or mastoiditis.

●Selection and duration of anticoagulation – After the acute phase of CVT, we suggest anticoagulation for most patients with either a direct oral anticoagulant or warfarin for 3 to 12 months (Grade 2C). Exceptions include patients with comorbid malignancy (for whom LMWH is preferred), antiphospholipid syndrome or chronic kidney disease (for whom warfarin is preferred), and those who are pregnant (for whom heparins are preferred). (See 'Long-term anticoagulation' above and 'Selection of anticoagulant' above.)

It is reasonable to continue anticoagulation for three to 6 months for patients with a provoked CVT associated with a transient risk factor (table 2) and for 6 to 12 months for patients with an unprovoked CVT. Indefinite oral anticoagulation is reserved for patients with recurrent CVT, extracerebral venous thromboembolism after CVT, or those with an associated severe thrombophilia. (See 'Duration of anticoagulation' above.)

●Managing the risk of recurrence – For pregnant patients with a history of CVT, we suggest temporary prophylactic anticoagulation with subcutaneous LMWH throughout pregnancy and continuing up to eight weeks postpartum (Grade 2C). For adolescent and adult females with a history of CVT, we advise not using combined oral contraceptives.

●Prognosis – CVT is associated with a good outcome in close to 80 percent of patients. Approximately 5 percent of patients die in the acute phase of the disorder, and longer-term mortality is nearly 10 percent. The main cause of acute death with CVT is neurologic, most often from brain herniation. After the acute phase, most deaths are related to underlying disorders such as cancer. (See 'Prognosis' above.)