Intracranial Hemorrhage in Patients with Primary Brain Tumors Treated with Therapeutic Enoxaparin: A Matched Cohort Study

BACKGROUND

Venous thromboembolism occurs in approximately 15-30% of individuals with primary brain tumors (e.g. glioblastoma, astrocytoma, and oligodendroglioma). Spontaneous intracranial hemorrhage (ICH) is also a frequent complication of primary brain tumors thus complicating the decision to administer therapeutic anticoagulation. Therapeutic anticoagulation does not appear to increase the risk of intracranial hemorrhage among patients with solid tumor brain metastases but whether anticoagulation is safe to administer to patients with primary brain tumors is less clear. The aim of this study was to determine the rate of intracranial hemorrhage associated with therapeutic enoxaparin for treatment of venous thromboembolism in patients with primary brain tumors compared to those patients with brain tumors not exposed to therapeutic anticoagulation.

METHODS

A 1:1 matched, cohort study was performed using a large hospital-based online medical record database (CQ2) linking ICD-9 codes with prescription medication records, cases were initially identified based on coding for primary brain tumors, venous thromboembolism, and prescription of enoxaparin. Matched controls were identified using a "round-robin" algorithm that ranked controls according to a scoring formula based on successful match for year of diagnosis, age, and gender. A blinded review of radiographic imaging was performed and intracranial hemorrhages were categorized as trace, measurable, and significant. Measurable intracranial hemorrhages were those defined as greater than 1 mL in volume and "significant" intracranial hemorrhages were defined as greater than 10 mL in volume, symptomatic (defined as focal neurologic deficit, headache, nausea, or change in cognitive function), or required surgical intervention. Time-to-event statistical analysis was performed using a competing risk analysis to account for death from any cause as an absorbing competing risk. Statistical comparison of event rates between cases and controls was performed using Fine and Gray competing risk regression.

RESULTS

A total of 100 patients with primary brain tumors were included in the study. The most common diagnosis was glioblastoma (85%), followed by anaplastic oligodendroglioma (8%), and anaplastic astrocytoma (7%). The two cohorts were well matched for age (60 years old), gender (65% male), and types of treatment received (99% radiation, 34% stereotactic radiosurgery, and 71% surgical resection). There was no statistical difference in the rate of measurable intracranial hemorrhage for the group of patients who received therapeutic enoxaparin at any point following the diagnosis of glioma compared to those who did not receive anticoagulation (subdistribution ratio hazard ratio 1.09, 95% CI 0.53-2.22). The 1-year cumulative incidence of measurable hemorrhage among those who were treated with enoxaparin was 23.6% compared with 20.0% in the control group (Gray's test P=0.48). The 1-year cumulative incidence of significant hemorrhage was 13.1% in those receiving enoxaparin compared with 6.0% in controls (sHR 1.45, 95% CI .47-4.65, P=0.68). The median survival was similar for the enoxaparin (1.56 years) and controls (1.63 years, Log rank P=0.81).

CONCLUSION

Intracranial hemorrhage is common in patients with primary brain tumors. In this matched cohort analysis utilizing a blinded radiologic review, the administration of therapeutic low molecular weight heparin did not significantly increase the risk of intracranial hemorrhage in the setting of glioma and venous thromboembolism. In patients with primary brain tumors, the diagnosis of venous thromboembolism treated with therapeutic enoxaparin did not impact overall survival.