PURPOSE: To evaluate the benefits and risks of repeat cerebral angiogr
aphy in patients with subarachnoid hemorrhage of unknown cause. MATERI
ALS AND METHODS: Findings from 391 angiographic examinations (323 init
ial, 68 repeat) were retrospectively reviewed in 323 patients with non
traumatic subarachnoid hemorrhage to determine the cause of bleeding a
nd the frequency of complications with neurologic deficit. RESULTS: At
the initial angiographic examination, aneurysms were found in 195 pat
ients (60.4%), and arteriovenous malformations were found in 11 patien
ts (3.4%). Sixty-six patients with negative findings at initial examin
ation underwent repeat cerebral angiography. Three additional aneurysm
s were detected, all of which could be seen retrospectively on the ini
tial angiogram. In 63 patients, the cause of bleeding remained unexpla
ined. Complications of the 391 angiographic examinations were definiti
ve neurologic deficits in one patient (0.2%) and transient deficits in
seven patients (1.8%). CONCLUSION: Repeat angiography is not necessar
y in patients with subarachnoid hemorrhage if technically good, carefu
lly evaluated digital subtraction angiography was performed initially.
Repeat angiography seems to be justified only when the initial examin
ation is technically inadequate, when vasospasm is present, or if furt
her bleeding occurs.