Prophylaxis for deep venous thrombosis in neurosurgical oncology: review of 2779 admissions over a 9-year period.

Object: Management of disease in patients undergoing neurosurgical treatment for tumors requires balancing the competing risks of hemorrhage and thrombosis. The authors compared the incidence of clinically apparent deep venous thrombosis (DVT) and pulmonary embolism (PE) in patients admitted for treatment of intracranial and spinal tumors at three institutions. At the public hospital (the Royal North Shore Hospital [RNSH]) nonsequential calf compression was used, and at the other two private neurosurgery services sequential calf compression with low-molecular-weight heparin was used in patients undergoing spinal surgery. All patients wore compression stockings and underwent follow-up scanning after surgery.

Methods: The authors identified from their neurosurgery databases 2779 discharges of patients with tumor from the RNSH and private hospitals between January 1, 1995 and December 31, 2003. Patient admissions were relatively well matched for age, sex, duration of stay, and tumor type. For patients who underwent spinal surgery, the incidence of DVT was higher in the RNSH (2.6% of admissions) than in private hospitals, where no case of DVT was seen (p = 0.02). The incidence of PE was higher in admissions patients who had been treated for cranial tumors in the RNSH (2.9%) than in those treated in the private hospitals (1.3%, p = 0.01). Possible reasons for these discrepancies include a higher proportion of ambulatory patients before and after surgery in the group treated at private hospitals. More emergency and semi-emergency surgery was performed in the RNSH than in the private hospitals. Nevertheless, fewer patients discharged from the RNSH had undergone surgery, which is a known risk factor for DVT.

Conclusions: Patients with intracranial tumors had a higher incidence of PE than those with spinal tumors. The incidence of DVT and PE was higher in patients admitted to the public hospital. The incidence of DVT in patients undergoing spinal procedures was lower when low-molecular-weight heparin was used judiciously, but the incidence of PE in patients undergoing cranial procedures was lower with the private hospital protocol, which did not include prophylaxis with anticoagulating agents.