Deep hypothermic circulatory arrest(DHCA), in which systemic temperatures of 2$0^{circ}C$ or less are used to allow temporary cessation of the circulation, is an useful adjunct in cardiac surgery. Because man in natural circumstances is never exposed to the extreme hypothermic condition, however, one of the controversial aspects is appropriate blood gas management($alpha$STAT versus PH-STAT) during DHCA. This study aims to compare $alpha$STAT with PH-STAT management for control of blood gases in experimental cardiopulmonary bypass(CPB) circuits with a membrane oxygenator. Fourteen young pigs were assigned to one of two strategies of gas manipulation. After a median sternotomy, CPB was established. Core cooling was initiated and continued until nasopharyngeal temperature fell below 2$0^{circ}C$. The flow rate was set at 2,500 ml/min. Once their temperatures were below 2$0^{circ}C$, the animals were subjected to circulatory arrest for 40mins. During cooling, blood gas was maintained according to either $alpha$$alpha$STAT or pH-STAT strategies. After DHCA, the body was rewarmed to normal body temperature. Arterial blood gases were measured before the onset of CPB, before cooling, before DHCA, at the point of 27$^{circ}C$ during re-warming, on completion of re-warming. Cooling time was significantly shorter in $alpha$-STAT than PH-STAT strategy, while there was no significant differences in rewarming time between two groups. Carbon dioxide was added between 5.5 and 3.0% in PH-STAT, while no carbon dioxide was added in $alpha$STAT management. Amounts of oxygen administration were gradually lowered as temperature decreased. In this way, criteria of PH, PaCO, and PaO adjustments were satisfied in both $alpha$STAT and PH-STAT management groups.