Abstract WP47: Quantitative Analysis of Hemorrhage and Infarct Volume for Predicting Outcome in Cerebral Venous Sinus Thrombosis

Abstract

Background and purpose: Intracerebral hemorrhage (ICH) and venous infarction are known complications that independently predict poor outcome in patients with cerebral venous sinus thrombosis (CVST). We aimed to study the utility of precise quantification of volumes of venous infarction and ICH using digitalized images in refining our prediction of discharge outcome. Methods: We retrospectively reviewed all patients with CVST admitted to University of Iowa Health Care between 2004-2014. All patients received standard of care management per available AHA/ASA recommendations. Scans were analyzed using Carestream Vue PACS software. Gradient Echo (GRE) MRI sequence was used to calculate ICH volume. When MRI was not available CT was used. Diffusion Weighted Imaging (DWI) MRI sequence was used to calculate infarction volume. Volume scorer was blinded to the clinical outcomes. Univariate and multivariate analyses were conducted. A p value ≤ 0.05 was set as statistically significant. Results: 115 patients met inclusion criteria, 76 women, mean age 42. A total of 30 (26%) had venous infarction, 34 (30%) had ICH. The median volumes were 8 cc (SD=27) for ICH and 14 cc (SD= 35) for infarction. Both ICH and infarction volumes were significantly higher among those with a poor functional outcome (mRS=3-6) [Mdn=16 cc, SD=32 for ICH; Mdn=33 cc, SD=46 for infarction] than those with good outcome [Mdn=5 cc, SD=7 for ICH; Mdn=7 cc, SD=162 for infarction], p= .029 and p=.027 respectively. The correlation between ICH volume and mRS was r=.43, p=.012 and the corresponding correlation for infarction volume was r=.62, p=.002. Regressions testing the effects of ICH and infarction volumes on discharge mRS showed that both significantly contributed (p<.001 & p=.004 respectively). These effects remained significant when controlling for the effects of age, gender, presence of deep infarct, and altered mental status at presentation (p=.001 and p=.013 respectively). Conclusion: Quantitative measurement of Infarction and ICH volumes on admission leads to a more accurate prediction of outcome at discharge in CVST patients. Its combination with clinical variables might signify an additional prognostication tool for identification of high risk patients refractory to standard therapy.