Abstract T MP67: Different Hematoma Volume Threshold to Predict Favorable Outcome in Basal Ganglia and Thalamic Hemorrhage
Background and Purpose: Hematoma volume is the strongest predictor of outcome in intracerebral hemorrhage (ICH). Historically, basal ganglia (BG) and thalamic ICH have been aggregated as “deep” ICH in the prior studies. We assessed whether BG and thalamic ICH have different hematoma volume threshold that is associated with 3-month functional outcome.
Design/Methods: From a prospective observational cohort study of ICH patients between July 2011 and May 2014, we compared the hematoma volume of groups that achieved favorable outcome (modified Rankin Scale of 0-2) and unfavorable outcome (modified Rankin Scale of 3-6) at 3 months in BG and thalamic ICH patients. Based on the hematoma size analyses, four different ICH Score models were tested for their accuracy in predicting favorable 3-month outcome, using each component of the original ICH Score except for the hematoma size threshold: ≥30 mL (model 1), ≥15 mL (model 2), ≥10 mL (model 3), and ≥5 mL (model 4).
Results: A total of 108 deep ICH patients (60 BG, 48 thalamic) with available 3-month outcome data were studied. For BG ICH, median hematoma volume for favorable outcome group was 10.7 mL [6.2-14.4] and unfavorable outcome group was 40.1 mL [21.0-86.3] (p<0.001). For thalamic ICH, median hematoma volume for favorable outcome group was 2.7 mL [1.2-9.9] and unfavorable outcome group was 9.8 mL [5.4-20.4] (p=0.002). For BG ICH, an area under the ROC curve for the different models of ICH Score was 0.85 (model 1), 0.88 (model 2), 0.84 (model 3), and 0.80 (model 4). For thalamic ICH, an area under the ROC curve was 0.86 (model 1), 0.87 (model 2), 0.85 (model 3), and 0.89 (model 4).
Conclusions: Basal ganglia and thalamic ICH have different hematoma volume threshold that is associated with favorable 3-month outcome. Hematoma volume threshold of 15 mL for BG ICH and 5 mL for thalamic ICH may enhance the ICH Score for predicting favorable functional outcome.
Author Disclosures: K. Nakagawa: None. S.L. King: None. M.A. Vento: None. M.M. Ing: None. T.S. Stern: None. D.N. Dittrich: None. T.B. Seto: None.
This research has received full or partial funding support from the American Heart Association, Western States - Alaska, Arizona, California, Hawaii, Idaho, Montana, Nevada, Oregon, Utah, Washington.
- © 2015 by American Heart Association, Inc.