Aneurysms and MR angiography.

The magnetic resonance angiography (MRA) study reported by Dr Curnes and his associates (1) is in some senses astonishing, given the skepticism with which MRA for intracranial aneurysms has been viewed by neurosurgeons in many institutions. When mentioning that "MRA may be useful as a screening test for aneurysms larger than 3 mm in size" based on work by Ross (2) , even this more restrained use of MRA has met with considerable skepticism. The problem of the intracranial aneurysm not seen on conventional angiography in a patient with subarachnoid hemorrhage (SAH) has been with us since Egas-Moniz. It may be useful to review the current clinical assessment of intracranial aneurysms. Approximately 75% of patients with spontaneous SAH will be determined to have an underlying intracranial aneurysm. Five percent will be found to have an intracranial arteriovenous malformation (3). SAH can be associated with deficiencies of each of the known coagulation factors and can also occur with therapeutic anticoagulation . Uncommon medical diseases such as fibromuscular dysplasia or moya moya can be the source of the bleeding. In sickle cell anemia with SAH, as many as 50% of the patients will have an associated aneurysm in addition to the sickle cell occlusive disease (3) . Collagen vascular disease such as systemic lupus erythematosus or polyarteritis nodosa can cause intracranial bleeding. Drug abuse that produces hypertension such as of cocaine, amphetamine, and phencyclidine ("angel dust") can cause SAH. There are a few sources of SAH that may not show up on angiograms. These are designated "cryptic malformations." Small arteriovenous malformations, thrombosed or partially thrombosed , can lead to significant SAH. Other cerebral malformations (cavernous hemangioma, venous angioma) almost never produce significant SAH (3, 4). Cavernous hemangiomas may bleed recurrently , but the hemorrhage is usually confined locally. There is one type of SAH in which the source of bleeding is usually never found . This is peri-