This study on dogs determined whether the requirement for enflurane anesthesia was different pre-versuspostcardiopulmonary bypass (CPB). Male mongrel dogs (n = 16) were anesthetized with enflurane in oxygen. Tracheal intubation was performed, monitors placed, and end-tidal enflurane concentration measuredviaa Puritan-Bennett Anesthesia Agent Monitor.,® MAC was determined by the tail-clamp method. CPB was then initiated using aortoatrial (n = 6, group 1) or femoral artery-vein (n = 4, group 2) cannulation or none (n = 6, group 3, control). CPB was maintained for 1 h using a bubble oxygenator, a crystalloid prime, and flows of approximately 70–80 ml/kg with a mean systemic pressure maintained between 50–70 mmHg. Following separation from CPB, MAC was again determined. The reduction in enflurane MAC following CPB was 30.1 ± 21.5% (mean ± SD;P< 0.05vs.pre-CPB) in group 1 but there was a wide range of reduction produced (3.8–58.8%). The degree of MAC reduction (19.8 ± 8.6%;P< 0.05vs.pre-CPB) produced by CPB in group 2 was much less variable in degree (range 13.0–32.4%) but did not differ from group 1. Although pre-versuspost-CPB mean systemic pressure fell from 83 ± 13 to 69 ± 15 mmHg (P< 0.05), this is above the level likely to produce a reduction in MAC. No other significant hemodynamic changes were observed. Temperature pre-versuspost-CPB was not different. The degree of hemodilution and acid-base disturbances are unlikely to be the explanation. In group 3, MAC did not differ significantly when measured repeatedly over 8.7 ± 1.5 h (2.13 ± 0.26% at the beginning of the experiment vs. 2.14 ± 0.35% at the end), although the gradient between inspired and expired concentration was significantly reduced (0.15 ± 0.05 to 0.06 ± 0.05%;P< 0.05). This suggests that contamination of end-tidal gas by dead-space gas is also an unlikely explanation for our results. The authors conclude that normothermic CPB alters the requirements for enflurane anesthesia in dogs.