Abstract 2424: Who Bleeds? Predicting Intracranial Bleeding in Patients with Acute Ischemic Stroke and an Indication for Anticoagulation

Abstract

Introduction: Many patients with acute ischemic stroke also have an indication for anticoagulation. A major concern in these patients is whether they are at increased risk for intracranial bleeding, and if so, what factors further increase their risk. We evaluated stroke patients with an indication for anticoagulation to determine what factors are associated with increased risk of intracranial bleeding. Methods: Patients presenting with acute ischemic stroke and an indication for anticoagulation were included in analysis. Data were collected in the following categories: demographics (age, race, gender), medical profile (blood pressure, glycemic control, statin use/LDL, renal function [divided into GFR categories- 0: GFR ≥60, 1: GFR 30-59, or 2: GFR <30], antiplatelet use, anticoagulation status), and stroke characteristics (type of stroke, time from onset of symptoms to anticoagulation, NIH Stroke Scale). Stroke volume, gradient echo positivity, and evidence of intracranial bleeding on head CT and MRI were also assessed. Hemorrhages were classified as: hemorrhagic conversion (petechiae) versus intracranial hemorrhage (a space occupying lesion), and symptomatic versus asymptomatic (based on subjective clinical worsening noted by the primary neurology team). Using stepwise regression analysis we determined which factors were associated with increased risk of intracranial bleeding (with any intracranial bleeding as the primary outcome). Results: 143 patients (mean age 63.5 years) met criteria and were included in analysis. 117 were placed on anticoagulation. The most common indications for anticoagulation were atrial fibrillation (35.6%), deep vein thrombosis (27.3%), presence of a hypercoagulable state (18.2%), and pulmonary embolism (17.5%). The difference in bleeding rates between those placed on anticoagulation and those treated with an antiplatelet agent was not statistically significant (25.6% versus 23.1%, χ2=0.785), but all of the intracranial hemorrhages (n=8) and symptomatic bleeds (n=9) occurred in the anticoagulated group. Age (OR= 1.31 per 10 year increment, 95% CI 0.98-1.74), volume (OR= 1.13 per 10 cc’s, 95% CI 1.05-1.21), and worsening GFR category (OR= 1.81, 95% CI 1.01-3.26) were predictors of intracranial bleeding. Odds of hemorrhage was predicted by exponentiating the equation: -3.823563 + (0.0120706)*(Volume) + (0.5939482)*(GFR Category) + (0.0266442)*(Age). Probability of hemorrhage can then be calculated. Conclusions: Probability of intracranial bleeding in patients with acute ischemic stroke and an indication for anticoagulation can be calculated based on an individual’s age, stroke volume, and GFR. This score can be used to assist with prognosis and clinical management. While anticoagulation does not appear to increase risk of bleeding, it does tend to result in larger, more symptomatic bleeds.