Association of Steroid Use with Deep Venous Thrombosis and Pulmonary Embolism in Neurosurgical Patients: A National Database Analysis.

Abstract

Venous thromboembolism (VTE) is a major preventable cause of morbidity and mortality in hospitalized patients and is a widely accepted measure for quality of care. Prolonged corticosteroid therapy, which is common in neurosurgical patients, has been associated with VTE. Using a national database, we sought to determine whether corticosteroid use for >10 days was an independent risk factor for deep venous thrombosis (DVT) and pulmonary embolism (PE). The well-validated American College of Surgeons National Surgical Quality Improvement Program database was queried to evaluate the rates of VTE during the period 2006-2013 in patients undergoing neurosurgical procedures. A multivariate regression model was constructed to assess the effect of prolonged corticosteroid use on the occurrence of PE and DVT by postoperative day 30. Of 94,620 patients identified, 565 (0.60%) developed PE and 1057 (1.12%) developed DVT within 30 days after surgery. In the multivariate model, patients receiving corticosteroids were significantly more likely to have PE (odds ratio = 1.47, 95% confidence interval = 1.13-1.90, P = 0.004) and DVT (odds ratio = 1.55, 95% confidence interval = 1.28-1.87, P < 0.001). Other factors independently associated with development of PE and DVT included the presence of malignancy, longer hospitalization, certain infections (including pneumonia and urinary tract infections), and stroke with a neurologic deficit. In the neurosurgical population, prolonged courses of corticosteroids are associated with an increased risk of developing postoperative DVT and PE, even when controlling for potential confounders.