Introduction: Computer tomographic angiography (CTA) is currently a standard diagnostic modality in initial testing for patients that present with non-traumatic subarachnoid hemorrhage. However, most institutions would still perform digital subtraction angiography (DSA) to further evaluate aneurysms or to confirm a negative CTA. The objective of this study was to evaluate whether CTA could reliably diagnose vascular pathology.
Methods: We performed a retrospective review of all spontaneous subarachnoid hemorrhages at LSUHSC-Shreveport between January 2010 and December 2011. All CTA results where compared with DSA results.
Results: Out of 109 patients with spontaneous subarachnoid hemorrhage, 18 patients had conflicting reports on DSA when compared with CTA. 71 patients had positive findings on CTA while 38 patients had an initial negative CTA. 16 patients with negative CTA on admission had a subsequent DSA finding of cerebral aneurysms. Of the 2 patients with conflicting positive findings, a small second aneurysm was missed. All missed aneurysms were < 5mm, with exception of 2 which were 7mm at maximum width. Only one patient had vasospasm on initial DSA. 9 patients had Fisher grade 4 on CT, 4 patients had grade 3, 3 patients had grade 2, and 2 patients had grade 1. All were saccular aneurysms. CTA had 84.4% sensitivity 91.6% specificity, 97.8% positive predictive value (PPV), and 57.9% negative predictive value (NPV) in diagnosing vascular pathology.
Conclusions: CTA has a high positive predictive value in detecting aneurysms but a low negative predictive value. Severity of subarachnoid hemorrhage seen on CT could affect the accuracy of CTA; whereas, vasospasm did not seem to play a role in that effect. In addition, small aneurysms, less than 5mm, were more likely to be missed. It is beneficial to perform digital subtraction angiography, even if the initial CTA was negative in spontaneous SAH patients.
Patient Care: Will assist in determining which SAH patients will need further invasive digital subtraction angiography after obtaining CTA results.
Learning Objectives: Understanding the accuracy and reliability of CTA for diagnosing cerebrovascular pathology