Objective: To determine the yield of repeated angiography in patients with a non-perimesencephalic subarachnoid haemorrhage (SAH) and a negative first cerebral angiogram.
Design: Retrospective.
Method: All diagnostic data of patients with a spontaneous SAH admitted to the Department of Neurology, St. Elisabeth Hospital, Tilburg, the Netherlands, in the period 1 January 1992-30 June 2000 were analysed. Patients with a perimesencephalic haemorrhage on a CT-scan were excluded and follow-up was completed. A negative angiogram was considered false-negative, if an aneurysm was shown on a repeat angiogram or after a rebleed. These angiograms were reviewed.
Results: A total of 333 patients with a spontaneous SAH were registered. Of these, 249 patients had one or more angiograms made, which resulted in 59 first angiograms being negative (24%). A total of 36 patients had a non-perimesencephalic SAH (26 women and 10 men; mean age: 54 years (range: 25-77)). In 25 of these 36 patients, angiography was repeated revealing 9 aneurysms. Four patients suffered from a rebleed after a previous negative angiogram. Altogether, in 13 of these 36 patients the first negative angiogram was false-negative (36%). In 5 of the 9 patients with a positive repeat angiogram, the first angiogram had been incorrectly assessed as negative.
Conclusion: Of the 36 patients with a non-perimesencephalic subarachnoid haemorrhage and a negative angiogram, 13 were revealed to have an aneurysm. Nine of these 13 aneurysms were demonstrated on a repeat angiogram. Technical and interpretation factors appeared to play an important role in missing an aneurysm on a cerebral angiogram.