Thirty-two inbred weanling puppies were divided into four groups to study the effect on cerebral blood flow and metabolism of different hypothermic strategies for cerebral protection similar to those used during cardiac operations in infancy. All animals were cooled to 18 ° C. The animals in the hypothermic control group were immediately rewarmed. One group underwent 30 minutes of hypothermic circulatory arrest at 18 ° C; another group had 90 minutes of hypothermic circulatory arrest at 18 ° C, and the final group had low-flow cardiopulmonary bypass (25 ml/kg per minute) at 18 ° C for 90 minutes. All animals had preoperative and postoperative neurologic and behavioral evaluation and extensive intraoperative monitoring of cerebral blood flow, cerebral vascular resistance, and oxygen and glucose uptake and metabolism: quantitative electroencephalography was also monitored before, during, and after operation, but those results are reported separately. Two animals in the 90-minute arrest group died, and all the survivors showed evidence of clinical, neurologic, and behavioral impairment on postoperative day 1, with residual abnormalities in all but one animal on day 6. In contrast, the survivors in all the other groups showed no significant clinical or behavioral sequelae. Cerebral metabolism was reduced only to 32% to 40% of baseline values at 18 ° C in all groups, although systemic metabolism was only 16% of normal. Cerebral metabolism returned promptly to baseline in all groups during rewarming and remained at baseline levels throughout the 8 hours of follow-up. Cerebral blood flow showed marked hyperemia in the hypothermic arrest groups during rewarming but then significant reductions below baseline values in all groups except the controls at 2 and 4 hours after the operation, lasting as late as 8 hours after the operation in the 90-minute arrest group. Cerebral vascular resistance showed increases in all groups at 2 and 4 hours after the operation, which persisted in the 90-minute arrest group at 8 hours. Cerebral metabolism was maintained at baseline levels despite postoperative decreases in cerebral blood flow and increases in cerebral vascular resistance by increases in oxygen and glucose extraction. The result was very low sagittal sinus oxygen saturations in all groups, most marked in the 90-minute arrest groups, which had a saturation of only 24% 8 hours after the operation. Our data show a severe, prolonged disturbance in cerebral blood flow and cerebral vascular resistance after 90 minutes of hypothermic circulatory arrest at 18 ° C, which correlates with clinical evidence of cerebral injury. The presence of similar but milder inappropriate hemodynamic responses in animals after 30 minutes of arrest at 18 ° C, and even after prolonged low-flow cardiopulmonary bypass at 18 ° C, suggest that strategies to improve cerebral protection during hypothermia--including use of colder temperatures--need to be explored. ( J THORAC CARDIOVASC SURG 1995; 109: 959-75).