Comparison of CTA to DSA in determining the etiology of spontaneous ICH

Background and Purpose: To compare the efficacy of computed tomographic angiography (CTA) to that of digital subtraction angiography (DSA) in the detection of secondary causes of intracerebral hemorrhage (ICH). Methods: Between January 2001 and February 2007 there were 286 patients that had both CTA and DSA for intracranial hemorrhage of all types. Those with primarily subarachnoid hemorrhage or recent trauma were excluded. Fifty-five patients formed the study cohort. Three reviewers independently analyzed the CTAs in a blinded protocol and classified them based on presence or absence of a secondary etiology. Results were compared with the reference standard DSA and kappa values determined for interobserver variability. Results: The overall sensitivity, specificity, positive predictive value, negative predictive value and accuracy of CTA were 89%, 92%, 91%, 91% and 91%, respectively. Kappa value for interobserver agreement ranged from 0.78 to 0.89. Two of four dural arteriovenous fistulas (dAVF) were missed on CTA by all three reviewers. Conclusion: CTA is nearly as effective as DSA at determining the cause of secondary intracerebral hemorrhage, but with a lower sensitivity for dAVFs. This supports the use of CTA as the first screening test in patients presenting with spontaneous ICH.