Risk of Anticoagulants and Antiplatelet Agents in Brain Tumor Patients
Patients with primary or metastatic brain tumors are predisposed to venous thromboembolism (VTE) due to an underlying hypercoagulable state, particularly in the postoperative period.1,2 The etiology of thrombosis is likely multifactorial and includes tissue factor–induced activation of coagulation, vessel wall injury from chemotherapy or radiation, and stasis as a result of surgery or neurologic deficits.3,4 Antithrombotic prophylaxis has been routinely used in patients undergoing general and orthopedic surgery and in hospitalized medical patients, but its adoption has been slower in ambulatory patients with brain tumors because of the concern about anticoagulant-associated intracranial bleeding. The risk of developing clots due to a hypercoagulable state and neurosurgical procedures needs to be balanced between the risk of precipitating a hemorrhage into the tumors of these patients. This chapter discusses the risk of anticoagulation and antiplatelet therapy in patients with brain tumors.
Risk of Venous Thromboembolism
Although the actual incidence of venous thromboembolism in patients with primary or metastatic brain tumors is unknown, early studies have suggested an incidence of 25% or higher.5 More recently a lower rate has been observed in a larger retrospective analysis of 9,489 patients with malignant gliomas, in which 7.5% of patients had a VTE event.2 Another study reported an incidence of 18% in a prospective study of patients with malignant gliomas.6 Similarly, in the most recent randomized study comparing dalteparin to placebo (PRODIGE), authors report an incidence of 17% in patients with malignant gliomas.7 Interestingly, the risk of VTE in pediatric patients with brain tumors is exceedingly low. In a study of 462 children with brain tumors, the incidence of clinical VTE was only 0.64%.8
Risk of Bleeding from Tumor
The risk of intracranial bleeding into a brain tumor, causing neurologic deterioration, raises serious concerns about using anticoagulant or antiplatelet agents in patients with known brain tumors. The risk of anticoagulant-associated intracranial bleeding within a brain tumor is difficult to assess because the rate of spontaneous tumor hemorrhage is significantly different among tumor types. Metastatic brain lesions from choriocarcinoma, melanoma, renal cell carcinoma, and thyroid cancer have a particularly high rate of spontaneous hemorrhage, whereas lung, breast, and prostate cancer have a much lower rate.9 The risk of intracranial bleeding in cancer patients with systemic disease and occult metastases raises concerns whether such patients should undergo brain imaging prior to the use of anticoagulants or antiplatelet agents. However, there are very few data to support such practice. In regard to gliomas, one study suggests that the use of anticoagulants is associated with 2 to 4% symptomatic bleeding.10
Prevention of Venous Thromboembolism in Patients with Brain Tumors
Primary prophylaxis with anticoagulants has generally not been recommended in patients with brain tumors except in the perioperative period, where the frequency of VTE is highest, likely due to immobilization and tissue-factor release from damaged blood vessels.2,11 One of the first studies looking at VTE prophylaxis in patients with brain tumors was performed at the University of Michigan. The researchers observed a low incidence of deep vein thrombosis (DVT) (1.4%) and pulmonary embolism (PE) (1.0%) in neurosurgical patients undergoing surgery while using only sequential compression devices (SCDs). However, analysis of their cohort by diagnostic category showed that 56% of their VTE complications occurred in patients with brain tumors.12 The apparent ineffectiveness of SCD prophylaxis for patients with intracranial neoplasms led the researchers to initiate a prospective study comparing the effects of preoperative administration of the low molecular weight heparin (LMWH) enoxaparin with that of SCDs alone; 68 patients completed the study, and the researchers found no statistical difference in their primary end point, which was postoperative DVT. Postoperative intracranial hemorrhage (ICH) did not occur in the patients with only SCDs, whereas five of 46 patients receiving LMWH (prophylactic dosing—30 mg twice daily) suffered a clinically significant ICH, which led to termination of the study. The researchers’ conclusions were that initiation of enoxaparin at the time of anesthesia induction increases the risk of postoperative intracranial hemorrhage.
In contrast, there is class I and II evidence supporting the use of chemical prophylaxis in patients with central nervous system (CNS) malignancies undergoing surgery. The first comes from Tel Aviv University in Israel and was a prospective, randomized, double-blinded study that looked at the safety of perioperative minidose heparin in patients undergoing brain tumor surgery.13 Fifty-five patients were treated with 5,000 units of heparin compared with 48 patients receiving placebo starting 2 hours before surgery; treatment was continued until full mobilization or 7 days, whichever came first. The authors found no statistical difference in bleeding tendencies during the intra- or postoperative periods between the two groups. They concluded that perioperative minidose heparin was safe in patients undergoing craniotomy for CNS tumors and recommended routine use of heparin in these patients for the prevention of VTE. This study is likely the most cited in regard to evidence-based medicine and the use of perioperative anticoagulants to prevent VTE complications in patients harboring CNS malignancies.
Venous thromboembolic complications following meningioma resection is well documented in the literature and led neurosurgeons at the University of California, San Francisco (UCSF) to look at the use and safety of postoperative enoxaparin in these patients.14 In a retrospective case-control study of 86 patients with intracranial meningiomas who underwent craniotomy, the authors identified 24 patients who started receiving enoxaparin within 48 hours after surgery and compared this treatment with a group of 62 patients who received no pharmacological prophylaxis. The authors found that enoxaparin did not increase the incidence of ICH following surgical resection, and the incidence of VTE complications was 0% in the LMWH group compared with 4.8% in the non-enoxaparin group, which was not statistically significant.14 The authors concluded that there was no significant increase in postoperative hemorrhage in patients who received LMWH postoperatively compared with those who did not. This study is somewhat limited by its retrospective nature and small numbers.
Long-Term Venous Thromboembolism Prophylaxis in Brain Tumor Patients
The previous studies focused on VTE prevention during the perioperative period and did not address long-term VTE prophylaxis. Extended use of the LMWH dalteparin as primary prophylaxis in patients with newly diagnosed malignant gliomas was recently assessed in the PRODIGE trial.7 This study was a blinded, randomized, placebo-controlled trial that enrolled 186 patients to receive either dalteparin or placebo (within 4 weeks of diagnosis/surgery) for 6 months, with an option to continue the study medication for up to 12 months. The incidence of clinically evident VTE at 6 months was lower among those receiving dalteparin (11 versus 17% with placebo), but was not statistically significant. Additionally, ICHs were more common in patients treated with LMWH compared with placebo (5% versus 1%), which also was not statistically significant. This trial confirmed the substantial risk of VTE complications in patients with malignant gliomas. Although there were trends toward reduction of VTE and increased ICH, the trial did not have sufficient statistical power to enable drawing definitive conclusions about the risk/benefit of long-term anticoagulation in these patients. Similar to these findings, in a phase 1 safety trial involving 40 patients with malignant gliomas, clinicians at Duke University reported that one patient developed a DVT and another developed ICH following daily tinzaparin administration (another LMWH) for a median 5 months.15 A larger phase 2/3 study is on the way.
More studies look at the general question of VTE prophylaxis in neurosurgical patients compared with neurosurgical patients harboring a brain tumor. Management for brain tumor patients must be put into this context. In general, it is assumed that the incidence of VTE is higher among patients undergoing surgery for a brain tumor compared with other neurosurgical procedures; nonetheless, safety and efficacy of anticoagulation can still be assessed and extrapolated from neurosurgical patients who do not have brain tumors. Eight randomized clinical trials (RCTs) have been published that evaluate either heparin or LMWH in patients undergoing elective cranial neurosurgery12,13,16–21; two studies have already been discussed, and three meta-analyses reviewing this subject have been published.22–24 The most recent review and meta-analysis by Hamilton and coworkers24 is the most comprehensive and analyzes these eight RCTs. Six of the RCTs evaluated either heparin or LMWH versus a control (placebo) group and included 1,170 patients. Five of the six trials found a significant reduction in the risk of VTE complications (symptomatic or asymptomatic) with perioperative heparin/LMWH prophylaxis. The pooled risk ratio for VTE with heparin use was 0.58 from the six RCTs. ICH was more common in those receiving anticoagulation prophylaxis, but five of the six RCTs found no statistical increase in ICH in patients receiving perioperative anticoagulation. The one study showing a significant increase in ICH among anticoagulated patients, from Dickinson and coworkers12 at the University of Michigan, was previously discussed. Hamilton and coworkers’24 meta-analysis predicts that for every 1,000 patients who receive heparin prophylaxis, 91 VTE events will be prevented, but seven patients will have a major ICH. They conclude that for patients undergoing elective cranial neurosurgery, heparin reduces the risk of VTE but may also increase the bleeding risk, and the balance only slightly favors heparin prophylaxis.
Bevacizumab
Bevacizumab, a monoclonal antibody that binds to vascular endothelial growth factor, is Food and Drug Administration (FDA) approved for the management of recurrent glioblastoma and has been used experimentally in the treatment of other brain tumors. It presents special considerations with regard to prophylaxis for VTE, as it is associated with both an increased risk of bleeding into a primary brain tumor and with VTE. The risk of ICH has limited its use in patients requiring anticoagulation for venous thrombosis. Two studies have looked at the safety of concurrent administration of bevacizumab and anticoagulation. The first was a small retrospective analysis of 21 patients who received therapeutic-dose anticoagulation (LMWH or warfarin—not perioperative) and bevacizumab.25 No serious ICH occurred, and only three small, nonsymptomatic hemorrhages were identified, compared with seven patients who had symptomatic ICH who were not anticoagulated. In contrast, a recent observational study of 64 patients receiving concurrent bevacizumab and anticoagulation found seven patients with ICH (11%) compared with 218 patients not receiving anticoagulation, where only two patients had serious ICH (1%).26 Because bevacizumab use is increasing, larger studies will be necessary to determine if concurrent anticoagulation is safe.
Recommendations for Venous Thromboembolism Prophylaxis in Patients with Brain Tumors
After reviewing the available data on VTE prophylaxis, there is no universal agreement that patients should receive anticoagulation or for how long. The two studies most applicable for patients with brain tumors provide contradictory results. Dickinson and coworkers12 found a statistical increase in ICH when LMWH (prophylactic dosing) was initiated prior to surgery. However, Constantini et al’s13 study did not find an increase in ICH when low-dose heparin was initiated before surgery. Does this difference reflect a difference in the two drugs, a difference in methodology, or a difference in patient selection? The answer remains unclear. The largest body of evidence on this topic is provided by the recent meta-analysis.24 Its conclusions are probably most accurate, and suggest that pharmacological prophylaxis does reduce the risk of VTE but also with a trend toward increased bleeding risk with a balance only slightly favoring anticoagulation.
Accordingly, at our institution we recommend that prophylactic anticoagulation be initiated only during the perioperative period in patients with brain tumors, as there is not yet statistically significant evidence to support long-term use in ambulatory patients. It is our practice to initiate heparin prophylaxis (5,000 mg twice daily) the day after surgery, as long as postoperative imaging is negative for bleeding or other complications that may necessitate return to the operating room, and anticoagulation is continued until discharge.
Treatment of Venous Thromboembolism in Patients with Brain Tumors
Patients with brain tumors and a DVT or PE have been historically treated with inferior vena cava (IVC) filters rather than anticoagulation due to the concern over an increased risk of anticoagulant-associated intracranial bleeding. However, anticoagulation is now the first-line therapy because the incidence of IVC filter complications is higher than originally thought, and the risk of hemorrhage secondary to anticoagulation is not as high as expected.27–29 In one study of 42 such patients, 57% developed IVC filter thrombosis, recurrent DVT, or postthrombotic syndrome.29
Warfarin Anticoagulation for Venous Thromboembolism in Patients with Brain Tumors
Several retrospective case-control studies have suggested that, outside the perioperative period, anticoagulation with warfarin is safe as long as the levels are carefully controlled. In one report, two of 103 patients (1.9%) of patients with a malignant glioma receiving warfarin for VTE developed a symptomatic intratumoral hemorrhage, compared with 2.2% of 272 malignant glioma patients without VTE or anticoagulation.10 In a series of 51 patients with metastatic brain tumors, three (6%) developed ICH. However, two of the patients had supratherapeutic warfarin levels.30 Although randomized comparisons are not available, data from these case studies suggest that warfarin is reasonably safe to use in patients with VTE and brain tumors.
Low Molecular Weight Heparin for Venous Thromboembolism in Patients with Brain Tumors
No studies directly report the use of full-therapeutic dosing of LMWH for treatment of VTE in patients with brain tumors or neurosurgical patients. Several studies already mentioned looked at LMWH for VTE prophylaxis in neurosurgical patients; it appears to be relatively safe according to the recent meta-analysis by Hamilton and coworkers.24 Additionally, there are no randomized studies comparing warfarin and LMWH anticoagulation specifically in patients with brain tumors. In the large randomized CLOT study comparing warfarin and LMWH in 673 patients with systemic cancer and VTE, LMWH was found to be more effective than warfarin in reducing the risk of recurrent thromboembolism without increasing the risk of bleeding. However, only 34 patients had primary brain tumors in this study.31
Recommendations for Venous Thromboembolism Treatment in Patients with Brain Tumors
To summarize, the management of venous thromboembolism in patients with brain tumors requires a balance between the effectiveness of treatment and the risk of ICH or other side effects. All experts agree that VTE complications need to be treated, and treatments include oral anticoagulation with warfarin, LMWH, or placement of an IVC filter. Based on the available data, which is mostly class III evidence and expert opinion, we suggest that LMWH rather than warfarin be used to treat VTE in patients with brain tumors because data from the CLOT trial found increased effectiveness in preventing recurrent thromboembolism, combined with the lower likelihood of developing supratherapeutic levels with weight-based LMWH administration. Patients harboring untreated metastatic brain tumors that have an increased risk of hemorrhage (melanoma, choriocarcinoma, renal cell carcinoma, and thyroid cancer) should have an IVC filter be placed until the tumors have been treated.
Antiplatelet Therapy in Brain Tumor Patients
Antiplatelet agents are the cornerstone for primary and secondary prevention of coronary artery disease (CAD) and cerebrovascular disease, and are especially important after percutaneous coronary interventions (PCIs). Over 2 million PCI procedures are performed annually in industrial nations, and this number continues to grow yearly, with over 90% of all procedures involving stent placement.32 As the prevalence of CAD continues to increase, so will antiplatelet use. Recent reports suggest that up to 5% of patients who undergo cardiac stent placement will also undergo some type of surgery within the first year after coronary stenting.32 This subset of patients presents surgeons, anesthetists, and cardiologists with the problem of deciding between the risk and benefits of continued antiplatelet therapy, meaning the risk of major operative bleeding versus possible coronary stent thrombosis. Because of the fear of intraoperative bleeding, it is generally accepted to stop antiplatelet therapy 5 to 10 days before surgical procedures. Unfortunately, studies that look at perioperative hemorrhage risk with antiplatelet therapy have been performed mainly in the orthopedic and cardiac surgery patients and are not generally applicable to neurosurgical patients.
No randomized controlled trials have compared the risk of bleeding in neurosurgical patients undergoing craniotomy who receive antiplatelet therapy with those who do not. Furthermore, no studies have investigated the risk of hemorrhage in patients with brain tumors who are taking antiplatelet agents. One study from the United Kingdom looks at avoidable risk factors that were associated with postoperative hematoma in neurosurgical patients. Over the 5-year period of study, 6,668 neurosurgical procedures were performed and 71 postoperative hematomas required surgical evacuation.33 Interestingly, the most frequent diagnosis leading to postoperative hematoma was meningioma. Risk factors for postoperative bleeding included low platelet count, prolonged prothrombin time, anticoagulant use that was not reversed, and heavy alcohol use; however, the most commonly associated risk factor was administration of aspirin, dipyridamole, or nonsteroidal anti-inflammatory drugs during the 2 weeks preceding the surgical procedure.33 A large meta-analysis examined the impact of low-dose aspirin in patients undergoing any surgery and found that patients on aspirin have an increase in perioperative bleeding complications of 50%; however, this did not translate into an increase in morbidity or mortality except for neurosurgical patients (referring to the previous study in the United Kingdom).34
Members of the Neuroanaesthesia Society of the United Kingdom were asked in 1997 about their policy regarding discontinuation of low-dose aspirin prior to intracranial surgery.35 Their answers were very inhomogeneous, but the majority of respondents felt that aspirin was a risk factor for hemorrhagic complications associated with intracranial procedure, but most have no specific policy regarding preoperative discontinuation. In a similar survey among neurosurgeons practicing in Germany, three quarters of the 138 neurosurgeons who responded felt that aspirin was a risk factor for hemorrhagic complications associated with intracranial procedures and more than half reported personal experiences of such problems.36 They also reported that the majority of neurosurgical facilities in Germany have distinct departmental policies concerning discontinuation of antiplatelet therapy preoperatively, an average of 7.3 days before surgery.
At our institution, we recommend stopping antiplatelet therapy 5 days prior to cranial neurosurgery. However, we recognize there is no evidence for this recommendation, and it is hoped that, in the future, well-designed studies may provide us with better guidance.
KEY POINTS
• In patients with primary or metastatic brain tumors, the risk of VTE is increased.
• In patients with brain tumors, prophylactic anticoagulation should be initiated during the perioperative period. Although there were trends toward reduction of VTE and increased intracranial hemorrhage with the use of prophylactic anticoagulation in ambulatory patients, no definitive conclusions can be drawn about the risk/benefits of long-term anticoagulation.
• VTE complications need to be treated, and treatments include oral anticoagulation with warfarin, unfractionated heparin, LMWH, or placement of an IVC filter.
• If there are no strong indications for its use during surgery, stop antiplatelet therapy at least 5 days prior to cranial surgery.
