Objective: Deep hypothermic circulatory arrest for neonatal heart surgery poses the risk of brain damage. Several studies suggest that pH-stat management during cardiopulmonary bypass improves neurologic outcome compared with alpha-stat management. This study compared neurologic outcome in a survival piglet model of deep hypothermic circulatory arrest between alpha-stat and pH-stat cardiopulmonary bypass, Methods: Piglets were randomly assigned to alpha-stat (n = 7) or pH-stat (n = 7) cardiopulmonary bypass, cooled to 19 degreesC brain temperature, and subjected to 90 minutes of deep hypothermic circulatory arrest. After bypass rewarming/reperfusion, they survived 2 days. Neurologic outcome was assessed by neurologic performance (0-95, 0 = no deficit and 95 = brain death) and functional disability scores, as well as histopathology. Arterial pressure, blood gas, glucose, and brain temperature were recorded before, during, and after bypass. Results: All physiologic data during cardiopulmonary bypass were similar between groups (pH-stat vs alpha-stat) except arterial pH (7.06 +/- 0.03 vs 7.43 +/- 0.09, P < .001) and arterial PCO2 (98 <plus/minus> 8 vs 36 +/- 8 mm Hg, P < .001). No differences existed in duration of cardiopulmonary bypass or time to extubation. Performance was better in pa-stat versus alpha-stat management at 24 hours (2 <plus/minus> 3 vs 29 +/- 17, P = 0.004) and 48 hours (1 +/- 2 vs 8 +/- 9, P = .1). Also, functional disability was less severe with pH-stat management at 24 hours (P = .002) and 48 hours (P = .053). Neuronal cell damage was less severe with pH-stat versus alpha-stat in the neocortex (4% +/- 2% vs 15% +/- 7%, P < .001) and hippocampal CA1 region (11% <plus/minus> 5% vs 33% +/- 25%, P = .04), but not in the hippocampal CA3 region (3% +/- 5% vs 16% +/- 23%, P = .18) or dentate gyrus (1% +/- 1% vs 3% +/- 6%, P = .63). Conclusions: pH-stat cardiopulmonary bypass management improves neurologic outcome with deep hypothermic circulatory arrest compared with alpha-stat bypass. The mechanism of protection is not related to hemodynamics, hematocrit, glucose, or brain temperature.