OBJECTIVE Thromboembolic stroke is likely to occur in patients with a restricted cerebral blood flow reserve. Our aims were to determine (1) whether symptomatic patients had any significant hemodynamic restriction ipsilateral to carotid occlusive disease compared with patients whose carotid stenosis is asymptomatic and (2) whether patients with carotid occlusive disease have impaired cerebral perfusion reserve compared with control subjects. METHODS We compared cerebral blood flow and collateral capacity using the 133Xe inhalation method and acetazolamide test in symptomatic (n = 10) and asymptomatic (n = 10) patients who had a high-grade internal carotid artery stenosis (range, 70% to 99%). Results were compared with those from 10 healthy control subjects. RESULTS Mean baseline cerebral blood flow was 40.29 +/- 1.38 mL/100 g per minute on the symptomatic side in symptomatic patients versus 45.20 +/- 2.53 mL/100 g per minute on the lesion side in asymptomatic patients (control subjects, 46.91 +/- 2.11 mL/100 g per minute in the right hemisphere versus 46.17 +/- 1.93 mL/100 g per minute in the left). There was no statistical difference between patients in symptomatic and asymptomatic groups versus control subjects (P > .10). Mean cerebral blood flow increase after acetazolamide was in the same range in symptomatic (52.89 +/- 2.54 mL/100 g per minute) and asymptomatic (56.22 +/- 3.35 mL/100 g per minute) patients (P > .10), and no difference was observed regarding control subjects (54.25 +/- 2.94 mL/100 g per minute; P > .10). Three asymptomatic and two symptomatic patients and three control subjects had no significant cerebral blood flow increase after acetazolamide. CONCLUSIONS An additional hemodynamic factor in thromboembolic ischemia related to severe unilateral carotid stenosis might be an unusual finding in patients without apparent hemodynamic induction of symptoms. The lack of significant variation in postacetazolamide cerebral blood flow in some patients and control subjects implies that this procedure may be inconsistent in assessing the cerebral perfusion reserve in the individual case.