To determine whether the absolute value for the stump pressure might be a useful index of symmetrical cerebral blood flow (CBF), and to examine correlations with the stump pressure ratio (initial mean stump pressure/preocclusion mean arterial pressure), fifty candidates for ICA injury or permanent occlusion were evaluated preoperatively. Each was continuously monitored for mean stump pressure and arterial pressure before, during (for a total of 20 min), and after balloon test occlusion. During the occlusion, CBF was measured by 99 m Tc-hexamethyl-propyleneamine oxime (99 m Tc-HMPAO) single photon emission computed tomography (SPECT). The stump pressure and the stump pressure ratio were then compared with the results of 99 m Tc-HMPAO SPECT. Patients who failed to tolerate even brief periods of carotid occlusion and showed asymmetric decreases in CBF on SPECT were divided into high and moderate risk groups. Those with no significant changes in CBF on the occluded site formed the minimum risk group. Mean stump pressure was over 50 mmHg in 10 of a total of 25 patients in the high and moderate risk groups, and below 50 mmHg in 5 of the 25 patients in the minimum risk group. The stump pressure ratio did not exceed 56% in any but two patients in the high and moderate risk groups, and values were at least 60% in all patients of the minimum risk group. Decrease of CBF in two moderate risk group cases was localized in the posterior circulation. Difference in symmetrical CBF between the stump pressure ratio vs. the absolute value of mean stump pressure were statistically significant (p < 0.01, Fisher's Exact Test). Maintenance of a stump pressure ratio of 60% or more during test occlusion may be a more useful index for a good collateral circulation than any absolute value for mean stump pressure.