Objective: Selective cerebral perfusion (SCP) affords brain protection superior to hypothermic circulatory arrest (HCA) for prolonged aortic arch procedures. Optimal pH strategy for HCA is controversial; for SCP it is unknown. We compared pH strategies during SCP in a survival pig model. Methods: Twenty juvenile pigs (26+/-2.4kg), randomized to alpha-stat (n=10) or pH-stat (n=10) management, underwent cooling to 20 degrees C on cardiopulmonary bypass (CPB) followed by 90min of SCP at 20 degrees C. SCP was conducted with a mean pressure of 50mmHg and hematocrit of 22.5%. Using fluorescent microspheres and sagittal sinus blood sampling, cerebral blood flow (CBF) and oxygen metabolism (CMRO(2)) were assessed at the following time points: baseline, after 30min cooling (20 degrees C), 30min of SCP, 90min of SCP, 15min post-CPB and 2h post-CPB. Visual evoked potentials (VEP) were assessed at baseline and monitored for 2h during recovery. Neurobehavioral recovery (10=normal) was assessed in a blinded fashion for 7 postoperative days. Results: There were no significant differences between the groups at baseline. CBF was significantly higher at the end of cooling, and after 30 and 90min of SCP in the pH-stat group (P=0.02, 0.007, 0.03). CMRO(2) was also higher with pH-stat (P=0.06, 0.04, 0.10). Both groups showed prompt return to values close to baseline after rewarming (P=ns). VEP suggested a trend towards improved recovery in the alpha-stat group at 2h post-CPB, P=0.15. However, there were no significant differences in neurobehavioral score: (alpha-stat versus pH-stat) median values 7 and 7.5 on day 1; 9 and 9 on day 4, and 10 and 10 on day 7. Conclusions: These data suggest that alpha-stat management for SCP provides more effective metabolic suppression than pH-stat, with lower CBF. Clinically, the better preservation of cerebral autoregulation during alpha-stat perfusion should reduce the risk of embolization.