Accuracy in Identifying the Source of Subarachnoid Hemorrhage in the Setting of Multiple Intracranial Aneurysms

BACKGROUND: Subarachnoid hemorrhage cases with multiple cerebral aneurysms frequently demonstrate a hemorrhage pattern that does not definitively delineate the source aneurysm. In these cases, rupture site is ascertained from angiographic features of the aneurysm such as size, morphology, and location. OBJECTIVE: To examine the frequency with which such features lead to misidentification of the ruptured aneurysm. METHODS : Records of patients who underwent surgical clipping of a ruptured aneurysm at our institution between 2004 and 2014 and had multiple aneurysms were retrospectively reviewed. A blinded neuroendovascular surgeon provided the rupture source based on the initial head computed tomography scans and digital subtraction angiography images. Operative reports were then assessed to confirm or refute the imaging-based determination of the rupture source. RESULTS: One hundred fifty-one patients had multiple aneurysms. Seventy-one patients had definitive hemorrhage patterns on initial computed tomography scans and 80 patients had nondefinitive hemorrhage patterns. Thirteen (16.2%) of the cases with nondefinitive hemorrhage patterns had discordance between the imaging-based determination of the rupture source and intraoperative findings of the true ruptured aneurysm, yielding an imperfect positive predictive value of 83.8%. Of all multiple aneurysm cases with subarachnoid hemorrhage treated by surgical or endovascular means at our institution, 4.3% (13 of 303) were misidentified. CONCLUSION: Morphological features cannot reliably be used to determine rupture site in cases with nondefinitive subarachnoid hemorrhage patterns. Microsurgical clipping, confirming obliteration of the ruptured lesion, may be preferentially indicated in these patients unless, alternatively, all lesions can be contemporaneously and safely treated with endovascular embolization.