Treatment and prevention of venous thromboembolism in patients with brain tumors

INTRODUCTION — Treatment and prevention of venous thromboembolism (VTE) in patients with primary and metastatic brain tumors is complicated by two conflicting issues. Patients with brain tumors have a substantial risk for developing VTE due to a hypercoagulable state, neurosurgical procedures, and often leg paresis. However, there is concern that antithrombotic agents can precipitate hemorrhage into the tumor with neurologic worsening.

The balance between these issues is discussed here. The overall risk, diagnosis, and treatment of VTE in patients with malignancy, as well as risk and treatment of anticoagulant-associated intracerebral hemorrhage, are reviewed separately. (See "Anticoagulation therapy for venous thromboembolism (lower extremity venous thrombosis and pulmonary embolism) in adult patients with malignancy" and "Risks and prevention of bleeding with oral anticoagulants" and "Reversal of anticoagulation in intracranial hemorrhage".)

PRE-ANTICOAGULATION RISK ASSESSMENT

General principles — The principles of managing acute VTE in patients with brain tumors are the same as in other patients with and without cancer. Anticoagulation is the cornerstone of treatment, and treatment decisions balance the risk of bleeding complications due to anticoagulation with the risks of untreated and recurrent VTE.

The risks of untreated VTE are substantial. Mortality associated with untreated pulmonary embolism is approximately 30 percent, usually due to recurrent embolism. Untreated, symptomatic deep vein thrombosis (DVT) carries a 50 percent risk of pulmonary embolism. (See "Epidemiology and pathogenesis of acute pulmonary embolism in adults", section on 'Prognosis' and "Overview of the treatment of proximal and distal lower extremity deep vein thrombosis (DVT)", section on 'Patients at low risk of bleeding'.)

Patients with cancer have higher than usual rates of recurrent VTE as well as a higher risk of bleeding with anticoagulant therapy compared with the general population. Therapy can be further complicated by comorbidities, procedures, and medication interactions (table 1). (See "Anticoagulation therapy for venous thromboembolism (lower extremity venous thrombosis and pulmonary embolism) in adult patients with malignancy".)

For patients with brain tumors, an individual assessment of bleeding risk with anticoagulant therapy takes into account the following:

●General risk factors for bleeding at any site (table 2). (See "Overview of the treatment of proximal and distal lower extremity deep vein thrombosis (DVT)", section on 'Assessing bleeding risk' and "Risks and prevention of bleeding with oral anticoagulants", section on 'Risk factors for bleeding' and "Risks and prevention of bleeding with oral anticoagulants", section on 'Intracranial'.)

●Risk factors for intracranial bleeding specifically, which may be influenced by the type of brain tumor, history of prior intracranial hemorrhage, recent craniotomy, and certain therapies. Each of these factors is discussed individually below.

Evidence-based guidelines on the treatment of acute VTE in cancer patients provide variable amounts of detail and specificity regarding brain tumors [1-4]. Our approach is generally consistent with these guidelines except as noted in the topic. Our recommendations for anticoagulation assume that treatment is in accordance with patient preferences with regard to goals of care and life expectancy.

Type of brain tumor — Brain tumors have varying propensities to hemorrhage spontaneously. In general, malignant tumors are more likely to bleed than benign tumors, and certain cancer types are associated with higher risk than others.

●Primary brain tumors – The rate of symptomatic bleeding into high-grade gliomas (eg, glioblastoma) is estimated at approximately 1 to 12 percent in the absence of antithrombotic therapy [5-9]. In observational studies, the risk of hemorrhage is several percentage points higher in patients receiving anticoagulation. (See 'Safety' below.)

Spontaneous bleeding into benign or low-grade primary brain tumors occurs less commonly, although studies are lacking to provide precise estimates of the frequency of symptomatic intracranial hemorrhage. Among benign or low-grade tumors, pituitary adenomas may be especially prone to spontaneous intratumoral hemorrhage [7].

●Brain metastases – Brain metastases from melanoma, choriocarcinoma, thyroid carcinoma, and renal cell carcinoma have particularly high propensities for spontaneous hemorrhage [10], while brain metastases from certain other primary tumors (eg, breast) generally do not bleed spontaneously. In one study, the risk of measurable intracranial hemorrhage over a one-year interval was fourfold higher in patients with brain metastases from renal cell carcinoma or melanoma compared with lung cancer [11]. The cumulative incidence of significant hemorrhage (defined as size >10 mL, symptomatic, or requiring surgical intervention) in the absence of anticoagulation was 37 percent in patients with melanoma or renal cell cancer and 19 percent in patients with non-small cell lung cancer.

Prior intracranial hemorrhage — In the general adult population, patients with a history of spontaneous intracranial hemorrhage are at increased risk for recurrent hemorrhage. The annual risk is estimated at 1 to 7 percent, and recurrence is most common within the first year following hemorrhage [12]. The most important risk factors for recurrent intracranial hemorrhage are uncontrolled hypertension, lobar location of the initial hemorrhage (as opposed to deep), and older age. In patients who require anticoagulation, guidelines suggest delaying anticoagulants for at least four weeks after onset of hemorrhage [13]. (See "Spontaneous intracerebral hemorrhage: Secondary prevention and long-term prognosis", section on 'Anticoagulation'.)

Patients with a history of tumor-associated intracranial hemorrhage are likely to be at increased risk of recurrent hemorrhage as well, although this has not been well studied, and the population is more heterogeneous. In the absence of better data, it is reasonable to consider recurrence risk estimates from patients with spontaneous intracranial hemorrhage when making risk/benefit decisions in patients with brain tumors. Estimates of risk may be further individualized based on the status of the residual tumor, if present, and the clinical circumstances at the time of the index hemorrhage, including the presence or absence of additional risk factors for bleeding (eg, thrombocytopenia, antithrombotic therapy).

Microhemorrhage within tumor — Evidence of microhemorrhage within tumors on susceptibility-weighted or proton density magnetic resonance imaging (MRI) sequences, without signal change on T1- or T2-weighted imaging, is common in patients with high-grade gliomas and brain metastases, especially melanoma. Susceptibility sequences are highly sensitive to the paramagnetic properties of hemosiderin, a breakdown product of blood. The changes may be variably described in radiology reports as susceptibility artifact, microhemorrhage, or chronic hemorrhage.

The significance of isolated susceptibility artifact within a tumor with regard to risk of clinical hemorrhage with or without anticoagulation has not been established in patients with brain tumors. While intuitively it raises concern that anticoagulation may pose a higher risk of tumor-related hemorrhage, this has not been well studied. In our clinical experience, the presence of microhemorrhage in isolation does not appear to be a contraindication in a patient who is otherwise deemed a candidate for anticoagulation.

Treatment with antiangiogenic therapy — Bevacizumab is a monoclonal antibody that binds to vascular endothelial growth factor and is used in the treatment of recurrent malignant glioma as well as other cancers. (See "Management of recurrent high-grade gliomas", section on 'Bevacizumab'.)

Bevacizumab is associated with both an increased risk of VTE and an increased risk of bleeding complications. (See "Cardiovascular toxicities of molecularly targeted antiangiogenic agents", section on 'Arterial and venous thromboembolism' and "Non-cardiovascular toxicities of molecularly targeted antiangiogenic agents", section on 'Bleeding'.)

Limited data in patients with malignant glioma who are treated with bevacizumab and anticoagulation suggest that although the risk of intracerebral hemorrhage is increased in patients receiving anticoagulants compared with those who are not, the risk-to-benefit ratio favors concurrent therapy in many cases [14,15]. (See "Management of recurrent high-grade gliomas", section on 'Side effects'.)

In a retrospective review of 38 malignant glioma patients treated with bevacizumab, VTE developed in five patients (13 percent) and was preceded four weeks before the onset of symptoms by D-dimer elevation above 865 ng/mL [16]. An existing hemiparesis conferred the highest risk for thrombotic complication.

In patients with brain metastases, there does not appear to be a significant increase in the risk of hemorrhage with use of bevacizumab. (See "Non-cardiovascular toxicities of molecularly targeted antiangiogenic agents", section on 'Intracranial bleeding'.)

THERAPEUTIC ANTICOAGULATION — Anticoagulation is the cornerstone of treatment for acute VTE for patients with and without cancer, including most patients with intracranial tumors.

Historically, patients with primary or metastatic brain tumors and VTE were often managed with the placement of inferior vena cava (IVC) filters rather than with anticoagulation because of concern over an increased risk of symptomatic intratumoral hemorrhage [17-20]. However, practice has evolved, and anticoagulation is now generally preferred in many cases because the incidence of complications with IVC filters is higher than originally thought, and the risk of intracranial hemorrhage secondary to anticoagulation, although elevated overall compared with no anticoagulation [21,22], is not as high as once feared.

Absolute contraindications — Absolute contraindications to anticoagulation include acute (<48 hours) intracranial hemorrhage, uncontrolled malignant hypertension, severe coagulopathy, severe platelet dysfunction, severe thrombocytopenia, inherited bleeding disorder, and high-risk invasive intracranial procedure within the last 7 to 14 days [1]. (See "Overview of the treatment of proximal and distal lower extremity deep vein thrombosis (DVT)", section on 'Patients at high risk of bleeding'.)

Relative contraindications — Most brain tumors do not pose a prohibitive risk of hemorrhage in patients with VTE who are otherwise appropriate candidates for anticoagulation, and the treatment approach is the same as in other patients with cancer. (See "Overview of the treatment of proximal and distal lower extremity deep vein thrombosis (DVT)" and "Anticoagulation therapy for venous thromboembolism (lower extremity venous thrombosis and pulmonary embolism) in adult patients with malignancy".)

The primary exceptions are patients with a history of intratumoral hemorrhage and those with tumors that have a high rate of spontaneous intracranial hemorrhage (ie, metastases from melanoma, choriocarcinoma, thyroid carcinoma, and renal cell carcinoma). In such patients, decisions must be individualized, as each of these factors represents a relative contraindication to anticoagulation that must be weighed against the risk of untreated VTE. (See 'Pre-anticoagulation risk assessment' above.)

Selected patients may be reasonable candidates for anticoagulation, such as those with surgically resected disease or otherwise effectively treated brain metastases [11]. Anticoagulation can also be considered even in patients with cancers that have a greater propensity for hemorrhage (ie, melanoma, choriocarcinoma, thyroid carcinoma, and renal cell carcinoma), provided that there is no evidence of acute intracranial hemorrhage on imaging and the potential benefits outweigh the risks. In a meta-analysis of retrospective studies, the risk of intracerebral hemorrhage was not significantly elevated in patients with brain metastases from melanoma or renal cell carcinoma who were treated with warfarin and/or low molecular weight (LMW) heparin [21].

We generally avoid anticoagulation in patients with intratumoral hemorrhage within the past four weeks or a remote history of a clinically significant intracranial hemorrhage. We do not consider the presence of microhemorrhage on MRI to be a contraindication to anticoagulation for VTE, in the absence of additional risk factors for bleeding. (See 'Prior intracranial hemorrhage' above and 'Microhemorrhage within tumor' above.)

Agent selection — Based on accumulating safety data in brain tumor patients as well as convenience, our approach to agent selection in patients who require anticoagulation for VTE is as follows:

●For most patients, we suggest use of a direct oral anticoagulant (DOAC) rather than LMW heparin or warfarin. (See 'Direct oral anticoagulants' below.)

●For patients who cannot take a DOAC due to reasons such as cost, we suggest LMW heparin rather than warfarin for maintenance. (See 'Low molecular weight heparin' below.)

●For patients with severe renal insufficiency (eg, creatinine clearance <30 mL/minute), options include either renally-dosed LMW heparin with monitoring of anti-factor Xa activity or warfarin. (See 'Low molecular weight heparin' below and 'Warfarin' below.)

●Thrombolysis is contraindicated in patients with an intracranial malignancy. (See "Approach to thrombolytic (fibrinolytic) therapy in acute pulmonary embolism: Patient selection and administration".)

For most patients, a DOAC or LMW heparin are appropriate for both immediate (ie, first five days) and continued anticoagulation. We prefer to use unfractionated heparin initially in patients who are clinically unstable and in patients at very high risk of bleeding, based on the short half-life of heparin and the ability to reverse rapidly with protamine sulfate. (See "Heparin and LMW heparin: Dosing and adverse effects", section on 'Unfractionated heparin'.)

There are no randomized trials directly comparing DOACs, LMW heparin, and/or warfarin anticoagulation specifically in patients with brain tumors. However, a growing body of observational data in these patients suggests that DOACs have similar efficacy and lower major bleeding risk compared with LMW heparin (see 'Direct oral anticoagulants' below). In addition, indirect evidence drawn from randomized trials in cancer patients (mostly without brain metastases) suggests that DOACs are associated with similar or lower VTE recurrence rates and similar or slightly increased major bleeding rates compared with LMW heparin, and that LMW heparin is more effective for recurrent VTE than warfarin with similar rates of bleeding. These data are reviewed in detail separately. (See "Anticoagulation therapy for venous thromboembolism (lower extremity venous thrombosis and pulmonary embolism) in adult patients with malignancy".)

Safety

Direct oral anticoagulants — The available data and our accumulating clinical experience suggest that DOACs are an effective, convenient, and probably safer alternative to LMW heparin for brain tumor patients who require anticoagulation [23-29]. Nonetheless, all of the data on comparative risk of intracranial hemorrhage in this patient population are retrospective, and in the absence of randomized trials, we acknowledge ongoing uncertainty for this endpoint.

Evidence in patients with glioblastoma includes a single-center retrospective cohort study of 121 patients diagnosed with VTE who were treated with a DOAC (n = 33) or LMW heparin (n = 88) [27]. At 30 days, the incidence of clinically relevant intracranial hemorrhage was nonsignificantly lower in patients treated with a DOAC compared with LMW heparin (0 versus 9 percent, p = 0.11). At six months, among 107 patients who were maintained on their initial anticoagulant, the cumulative incidence of clinically relevant intracranial hemorrhage was significantly lower in the DOAC group (0 versus 24 percent, p = 0.001), with four fatal hemorrhages in the LMW heparin group. Rates of recurrent VTE were similar between groups (0 and 4 percent for DOACs and LMW heparin, respectively, p = 0.55). Patient characteristics appeared balanced between groups with the exception of time since most recent surgery, which was shorter in the LMW heparin group, and concurrent use of an antiplatelet agent, which was nonsignificantly higher in the DOAC group.

Other retrospective data in patients with brain metastases and various primary brain tumors have shown similar results favoring DOACs. In a retrospective study of patients with primary (n = 67) or metastatic (n = 105) brain tumors with VTE who were anticoagulated with either a DOAC or LMW heparin, the 12-month incidence of major intracranial hemorrhage was lower with DOACs in patients with primary brain tumors (0 versus 18 percent) as well as brain metastases (11 versus 18 percent), although the difference was not statistically significant for brain metastases [24]. In both the primary and metastatic brain tumor comparisons, patients in the DOAC group were more likely to have baseline hypertension, chronic kidney disease, and concomitant aspirin use. In a second retrospective study that included 96 patients with brain metastases treated with either a DOAC or LMW heparin, the adjusted risk of any intracranial hemorrhage was similar between DOAC- and LMW heparin-treated patients (HR 0.98, 95% CI 0.28-3.40) [25]. In a third study that included 52 brain tumor patients treated with a DOAC, the risk of any major bleeding was higher in association with LMW heparin, and the risk of intracranial hemorrhage was nonsignificantly lower with DOACs [26].

In trials comparing DOACs with LMW heparin for initial treatment of cancer-associated VTE, most or all patients did not have brain metastases at baseline. Available data on intracranial hemorrhage and brain metastases in these trials are reviewed below:

●Edoxaban, an oral factor Xa inhibitor, was compared with dalteparin in 1050 patients with active cancer and VTE, of which 74 patients had a primary or metastatic brain tumor [30]. The trial showed a nonsignificant reduction in the rate of recurrent thrombosis in the edoxaban arm but a higher rate of major bleeding, mostly from gastrointestinal bleeding and predominantly in patients with primary gastrointestinal malignancies. The rate of serious bleeding, which included intracranial hemorrhage, was equal between the groups.

●Rivaroxaban, an oral factor Xa inhibitor, was compared with dalteparin in 406 patients with active cancer and VTE, but only three patients had a brain tumor [31]. Rivaroxaban was associated with relatively low VTE recurrence but higher clinically relevant nonmajor bleeding compared with dalteparin. No central nervous system (CNS) bleeds were reported on study.

●Apixaban, another oral factor Xa inhibitor, was compared with dalteparin in 287 cancer patients with acute symptomatic or incidental VTE [32]. This trial enrolled eight patients with brain tumors, but the number of patients with brain metastases was not explicitly reported. VTE recurrence rate was significantly lower in the apixaban arm, and the rate of major bleeding was similar in the two treatment groups. In a larger trial comparing apixaban versus dalteparin, there were no brain tumor patients, and major bleeding rates were similar between groups [33]. There was one fatal intracranial hemorrhage in the dalteparin group.

DOACs are convenient because they do not require injections or drug level monitoring. US Food and Drug Administration (FDA)-approved antidotes or specific reversal agents for the DOACs are now available, though routine coagulation tests cannot be used to determine the degree of anticoagulation. (See "Management of bleeding in patients receiving direct oral anticoagulants" and "Anticoagulation therapy for venous thromboembolism (lower extremity venous thrombosis and pulmonary embolism) in adult patients with malignancy".)

Low molecular weight heparin — The best available estimate of the risk of major hemorrhage in brain tumor patients treated with LMW heparin is derived from a retrospective case control study of 364 patients with cancer-associated VTE, half of whom had a primary or metastatic brain tumor [34]. Anticoagulation consisted of extended-duration therapeutic LMW heparin in approximately 90 percent of patients. With a median follow-up of 6.7 months, the incidence of major bleeding was similar in patients with and without brain tumors (8.6 versus 5.0 per 100 patient-years). The rate of intracranial hemorrhage in patients with brain tumors was 4.4 percent; seven of nine hemorrhages were symptomatic and none were fatal. There were no intracranial hemorrhages in the control group.

Most but not all smaller studies have found similar results:

●In a retrospective matched cohort study of 293 patients with brain metastases in which approximately one-third of the patients received therapeutic enoxaparin, there was no significant difference in the cumulative incidence of intracranial hemorrhage at one year in patients treated with enoxaparin compared with controls who were not (19 versus 21 percent for measurable hemorrhage, 21 versus 22 percent for significant hemorrhage, and 44 versus 37 percent for total hemorrhages) [11]. The risk of hemorrhage was fourfold higher in patients with melanoma or renal cell carcinoma compared with lung cancer (adjusted hazard ratio [HR] 3.98, 95% CI 2.41-6.57), but the risk was not affected by anticoagulation status.

●In a retrospective study of 64 patients with glioblastoma who were diagnosed with VTE, 93 percent of patients were treated with anticoagulation, mostly LMW heparin [35]. The majority of patients were treated for >6 months. Ten patients (16 percent) developed complications due to anticoagulation, including three patients with brain hemorrhage (4.7 percent) and three with extracranial bleeding.

●A retrospective cohort study that included 50 patients with a primary brain tumor (mostly glioblastoma) treated with enoxaparin for VTE identified a higher rate of intracranial hemorrhage (14.7 versus 2.5 percent not treated with anticoagulation) [36]. Neither hypertension nor use of an antiangiogenic agent were risk factors for hemorrhage; older age at diagnosis approached statistical significance.

●A larger retrospective study in 220 patients with high-grade glioma found a similarly high risk of measurable intracranial hemorrhage in patients treated with LMW heparin for VTE (17 percent one-year cumulative incidence), which was not statistically different than the cumulative incidence in patients not receiving anticoagulation, both with VTE (9 percent) and without VTE (13 percent) [9]. Approximately two-thirds of the hemorrhages were associated with symptoms.

Warfarin — Studies of warfarin in patients with brain tumors also indicate that the risk of tumor-associated intracranial hemorrhage may not be significantly increased in patients with primary or metastatic brain tumors if the degree of anticoagulation with warfarin is carefully monitored [6,17,18,37-40].

Dosing — Dosing for DOACs, LMW heparin, unfractionated heparin, and fondaparinux in the treatment of VTE is discussed separately. (See "Venous thromboembolism: Initiation of anticoagulation".)

Duration of therapy — Patients with acute VTE should be treated for at least three to six months. Extended or lifelong therapy is suggested in most patients with glioblastoma and other forms of active cancer, including those receiving antitumor therapy, as well as in patients with recurrent VTE, provided the bleeding risk is low and no clinically relevant complications from anticoagulation have occurred. (See "Anticoagulation therapy for venous thromboembolism (lower extremity venous thrombosis and pulmonary embolism) in adult patients with malignancy".)

Duration of therapy in patients without active malignancy (eg, patients with low-grade gliomas or benign brain tumors such as meningioma) should be individualized based on an assessment of risk of recurrence, risk of bleeding, and patient values and preferences. (See "Selecting adult patients with lower extremity deep venous thrombosis and pulmonary embolism for indefinite anticoagulation".)

Role of inferior vena cava filter — While IVC filters should not be routinely inserted in patients with acute deep vein thrombosis (DVT), they remain in use for patients deemed to have a prohibitive risk of bleeding with anticoagulation and life-threatening thrombus burden [1].

The efficacy of IVC filters for prevention of recurrent pulmonary embolism in patients with brain tumors is not well characterized. In the general population, observational data indicate that rates of recurrent pulmonary embolism are low (<5 percent in most series). (See "Overview of the treatment of proximal and distal lower extremity deep vein thrombosis (DVT)", section on 'Inferior vena cava filter'.)

In small studies, the use of IVC filters in patients with brain tumors has been associated with higher risk of recurrent VTE and variable complication rates [17,37]. In an older series of 42 such patients, 12 percent had recurrent pulmonary emboli and 57 percent developed IVC or filter thrombosis, recurrent DVT, or post-thrombotic syndrome [17]. A more modern series reported a 30 percent rate of recurrent VTE in 60 patients with glioblastoma who had an IVC filter placed, some of whom also received anticoagulation [41]. (See "Placement of vena cava filters and their complications".)

PRIMARY PREVENTION (VTE PROPHYLAXIS)

Incidence and risk factors — Patients with primary or metastatic brain tumors, as well as malignancies at other sites, have a latent hypercoagulable state that predisposes to thromboembolism, particularly in the postoperative period [42-44]. (See "Risk and prevention of venous thromboembolism in adults with cancer" and "Cancer-associated hypercoagulable state: Causes and mechanisms".)

Estimates of the incidence of VTE consistently show increased relative risk among patients with cancer compared with the general population, particularly in patients with glioblastoma [45]. In prospective studies of patients with malignant glioma, the observed incidence of symptomatic VTE ranges from 17 to 26 percent [5,46-48].

Although there is clustering of VTE events in the postoperative period following craniotomy and during intensive chemotherapy, the risk persists throughout the clinical course [44]. Risk factors for VTE in patients with brain tumors include [16,35,43,44,49-52]:

●Older age (≥60 years)

●Low performance status

●Obesity

●Glioblastoma histology and absence of an isocitrate dehydrogenase type 1 or 2 (IDH1/2) mutation

●Large tumor size (eg, >5 cm) and subtotal resection

●Use of steroids, bevacizumab, and/or chemotherapy

●Recent neurosurgery (within the past two months)

●Leg paresis or plegia

●A or AB blood type

For adults with diffuse gliomas, a VTE prediction tool has been developed based on a cohort of 258 adults with newly-diagnosed grade 2 to 4 diffuse glioma, in which the incidence of VTE was 18 percent [53]. Multivariable modeling verified many of the above risk factors and was used to generate a 10-item tool, which predicted high versus low VTE risk with moderate accuracy in validation cohorts (areas under the curve [AUCs] ranging from 0.63 to 0.84). A web-based version of the tool is available [54]. However, since it is based on relatively low AUCs and has not been prospectively validated, it should not be used as a stand-alone tool to make clinical decisions about preventive treatment.

Inpatients and postoperative patients — Perioperatively, the frequency of VTE in brain tumor patients is approximately 10 to 15 percent [43,55,56]. The use of pneumatic compression stockings combined with low molecular weight (LMW) heparin or unfractionated subcutaneous heparin started preoperatively and resumed 24 to 48 hours after surgery is effective and relatively safe in neurosurgical patients [57-61]. These measures are generally continued until the patient resumes ambulation.

Outpatients — Primary prophylaxis with anticoagulants is not generally recommended in ambulatory patients with brain tumors except in the perioperative period [52,62].

The extended use of LMW heparin (dalteparin) as primary prophylaxis of VTE in patients with newly diagnosed malignant glioma was assessed in the PRODIGE trial [5]. The trial was terminated prematurely because of the unavailability of placebo. Overall, 186 patients were randomly assigned to six months of treatment with dalteparin or placebo, with an option to continue on study medication for up to 12 months. The incidence of clinically evident VTE at six months was lower among those receiving dalteparin (11 versus 17 percent with placebo), but the difference was not statistically significant. Intracranial hemorrhages were more common in patients treated with dalteparin (5 versus 1 percent with placebo at 12 months; hazard ratio [HR] 4.2, 95% CI 0.48-36).

The PRODIGE trial confirmed the substantial risk of VTE in patients with malignant gliomas. Although there were trends toward reduction of VTE and increase in intracranial hemorrhage, the trial did not have adequate statistic power at the time of termination to draw definitive conclusions about the benefits versus risks of long-term anticoagulation.

A small phase II trial involving 40 patients with malignant gliomas treated with the LMW heparin tinzaparin 4500 international units subcutaneously daily for a median of approximately five months reported symptomatic central nervous system (CNS) hemorrhage in one patient (2.5 percent) and deep vein thrombosis (DVT) in one patient (2.5 percent) [63].

For the prevention of VTE in at-risk patients with brain tumors, long-term treatment with daily aspirin (eg, 325 mg/day orally) may offer partial protection and is easy to administer, although neither the safety nor efficacy of this approach has been well defined in this population [40]. For general VTE prevention, aspirin appears to have some activity, but it is clearly less than that afforded by anticoagulation [64]. (See "Selecting adult patients with lower extremity deep venous thrombosis and pulmonary embolism for indefinite anticoagulation".)

Aspirin therapy may be considered for primary prevention of VTE in brain tumor patients who are ≥60 years old, have large tumors or substantial leg weakness, or are undergoing chemotherapy. However, the effect of aspirin in inhibiting platelet function in patients at risk for chemotherapy-induced thrombocytopenia must be weighed against the limited evidence of benefit in preventing VTE.

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: Anticoagulation" and "Society guideline links: Primary brain tumors".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Beyond the Basics topics (see "Patient education: Warfarin (Beyond the Basics)" and "Patient education: Deep vein thrombosis (DVT) (Beyond the Basics)")

SUMMARY AND RECOMMENDATIONS

●General principles – The general principles of managing acute venous thromboembolism (VTE) in patients with brain tumors are the same as in other patients with and without cancer. Anticoagulation is the cornerstone of treatment, and treatment decisions balance the risk of bleeding complications due to anticoagulation with the risks of untreated and recurrent VTE. (See 'Introduction' above.)

●Pretreatment risk assessment – The risk of bleeding in patients with brain tumors is determined by considering both general risk factors for bleeding with anticoagulant therapy (table 2 and table 1) and risk of intracranial hemorrhage related to the tumor and any antitumor therapies. (See 'Pre-anticoagulation risk assessment' above.)

Absolute contraindications to anticoagulation in patients with and without brain tumors include acute intracranial hemorrhage (<48 hours), uncontrolled malignant hypertension, severe coagulopathy, severe platelet dysfunction or thrombocytopenia, inherited bleeding disorder, and recent intracranial surgery. (See 'Absolute contraindications' above.)

●Therapeutic anticoagulation – Most brain tumors do not pose a prohibitive risk of hemorrhage in patients with VTE who are otherwise appropriate candidates for anticoagulation. Potential exceptions include brain metastases from melanoma, choriocarcinoma, thyroid carcinoma, and renal cell carcinoma, which have a higher rate of spontaneous intratumoral hemorrhage than other tumor types, and patients with a history of intratumoral hemorrhage. Decisions in these patients should be individualized. (See 'Relative contraindications' above.)

•Agent selection – In patients with brain tumors who are selected to receive anticoagulation for VTE, we suggest a direct oral anticoagulant (DOAC) rather than low molecular weight (LMW) heparin or warfarin (Grade 2C). For patients who cannot take a DOAC due to cost, we suggest LMW heparin rather than warfarin for maintenance (Grade 2C). For patients with severe renal insufficiency, options include renally-dosed LMW heparin with monitoring of anti-factor Xa activity or warfarin. (See 'Agent selection' above.)

•Duration – VTE in low-grade glioma and benign tumors should be treated for three to six months, whereas long-term anticoagulation is generally indicated for patients with glioblastoma and other active malignancies. (See 'Duration of therapy' above.)

●Role of inferior vena cava (IVC) filter – IVC filters should not be routinely inserted in brain tumor patients with acute deep vein thrombosis (DVT). However, they remain in use for patients deemed to have a prohibitive risk of bleeding with anticoagulation and life-threatening thrombus burden. (See 'Role of inferior vena cava filter' above.)

●Prophylactic anticoagulation – Although patients with primary or metastatic brain tumors are at relatively high risk for the development of VTE, we suggest that these patients not be anticoagulated prophylactically, except in the postoperative period (Grade 2B). (See 'Primary prevention (VTE prophylaxis)' above.)