A balloon test occlusion of the internal carotid artery was performed in 11 patients with internal carotid artery aneurysms. Tolerance by patients was assessed by a combination of clinical examination; angiography; electroencephalography; 99mTc-hexamethylpropyleneamine oxime (99mTc-HMPAO) single-photon emission computed tomography (SPECT) with relative quantification; and, in four patients, 99mTc-HMPAO SPECT with absolute quantification of cerebral blood flow. During test occlusion, angiography showed a patent circle of Willis in all patients. No patient developed new clinical findings or electroencephalographic changes. The SPECT studies of five patients in whom 99mTc-HMPAO was injected during test occlusion demonstrated changes from their baseline SPECT studies. The internal carotid artery was permanently occluded in two of these patients, neither of whom became symptomatic because of the occlusion. Three patients who demonstrated no changes between baseline and test occlusion SPECT studies underwent permanent occlusion of the internal carotid artery without incident, and postoperative SPECT images were unchanged from baseline. Our preliminary results suggest that patients who have no changes between baseline and test occlusion 99mTc-HMPAO SPECT studies should have adequate collateral circulation to sustain cerebral blood flow after occlusion of the internal carotid artery if no thromboembolic episodes occur. In contrast, a patient's tolerance of permanent occlusion cannot be consistently and reliably predicted if there are changes between baseline and test occlusion SPECT studies. In these patients, absolute quantitation of cerebral blood flow is important. Greater numbers of patients are required to confirm these initial results.