Intracerebral hemorrhage: toward physiological imaging of hemorrhage risk in acute and chronic bleeding.

Despite improvements in management and prevention of intracerebral hemorrhage (ICH), there has been little improvement in mortality over the last 30 years. Hematoma expansion, primarily during the first few hours is highly predictive of neurological deterioration, poor functional outcome, and mortality. For each 10% increase in ICH size, there is a 5% increase in mortality and an additional 16% chance of poorer functional outcome. As such, both the identification and prevention of hematoma expansion are attractive therapeutic targets in ICH. Previous studies suggest that contrast extravasation seen on CT Angiography (CTA), MRI, and digital subtraction angiography correlates with hematoma growth, indicating ongoing bleeding. Contrast extravasation on the arterial phase of a CTA has been coined the CTA Spot Sign. These easily identifiable foci of contrast enhancement have been identified as independent predictors of hematoma growth, mortality, and clinical outcome in primary ICH. The Spot Sign score, developed to stratify risk of hematoma expansion, has shown high inter-observer agreement. Post-contrast leakage or delayed CTA Spot Sign, on post contrast CT following CTA or delayed CTA respectively are seen in an additional ∼8% of patients and explain apparently false negative observations on early CTA imaging in patients subsequently undergoing hematoma expansion. CT perfusion provides an opportunity to acquire dynamic imaging and has been shown to quantify rates of contrast extravasation. Intravenous recombinant factor VIIa (rFVIIa) within 4 h of ICH onset has been shown to significantly reduce hematoma growth. However, clinical efficacy has yet to be proven. There is compelling evidence that cerebral amyloid angiopathy (CAA) may precede the radiographic evidence of vascular disease and as such contribute to microbleeding. The interplay between microbleeding, CAA, CTA Spot Sign and genetic composition (ApoE genotype) may be crucial in developing a risk model for ICH.