Background: Subarachnoid hemorrhage (SAH) is found to have no vascular origin by initial catheter angiography in approximately 15% of cases. The most appropriate course for the type and frequency of additional diagnostic workup remains controversial.
Objective: To retrospectively assess the diagnostic yield of short-term and long-term repeat catheter angiography in the era of advanced imaging.
Methods: Between 2003 and 2011, 254 consecutive patients diagnosed with SAH had negative initial angiography. SAH was perimesencephalic (PM) in 46.5% and nonperimesencephalic (NPM) in 53.5%. Angiography was repeated at 1-week (short-term) and 6-week (long-term) intervals from the initial negative angiogram.
Results: Ten of 254 patients had a vascular source of hemorrhage on short-term follow-up angiography with a diagnostic yield of 3.9%. One hundred seventy-four patients with negative findings on the first 2 angiograms received a third angiogram, and 7 of these patients were found to have a vascular abnormality. The estimated yield of this third angiogram was 4.0%. The overall diagnostic yield of repeat angiography was 0% in the PM group and 12.5% in the NPM group. The diagnostic yield of short-term and long-term follow-up angiography in patients with NPM SAH was 7.3% and 7.8%, respectively. NPM patients were more likely to experience vasospasm and hydrocephalus requiring external ventricular drainage or cerebrospinal fluid diversion than PM patients.
Conclusion: Our results support a protocol of short-term and long-term angiographic follow-up in patients with NPM SAH and negative initial angiography. Aggressive protocols of follow-up angiography may not be necessary in patients with PM SAH.