Objective To compare the effect of conventional and hyperoxia management strategy during deep hypothermia in patients with DeBake type 1 aortic dissection or aortic arch aneurysm undergoing total aortic arch replacement.Methods 32 adult patients undergoing total aortic arch replacement were randomly allocated to one of two groups(n=16 each):conventional(C)and hyperoxia group(H).The patients had no history of cerebral vascular disease.Left radial artery and dorsal artery of left foot were cannulated for monitoring of blood pressure of upper and lower limbs.Right internal jugular vein was cannulated for CVP monitoring and administration of drug and fluid.Anesthesia was induced with etomidate 10-15 mg,fentanyl 5-10 ?g?kg~(-1) and pancuronium 0.1 mg?kg~(-1) and maintained with fentanyl(total amount was＜20 ?g?kg~(-1)),isoflurane and pancuronium after tracheal intubation.Intermittent i.v.boluses of diazepam,sodium thiopental or propofol were given during cardiopulmonary bypass(CPB).Another catheter was inserted into right internal jugular vein eephalad until resistance was met.The tip of the catheter was at the level of mastoid process.The hyperoxia management involved the following steps:FiO_2 was gradually reduced with decreasing body temperature(T_0)from 70%(36～ 37℃)to 60%-40%(35.9-34℃),38%-30%(32-26℃),30%(26-24℃)and finally to 21%.When nasopharyngeal T_0 was reduced to 22℃ or 5-10 min before selective cerebral peffusion(SCP),FiO_2 was raised to 60%-100% to maintain PjvO_2＞20 mm Hg or SjvO_2＞60%.FiO_2 was maintained at 60%-100% during SCP until T_0 was rewarmed to 22℃,then reduced to 30%.FiO_2 was then gradually increased to 40%(when T_0 reached 28℃),to 50%-70% (34-37℃)and finally to 80%(T_0＞37℃).Blood samples were taken from jugular venous bulb and arterial port of oxygenator for determination of PjvO_2,SjvO_2 and PaO_2 before skin incision (T_1),at 15 min of CPB(T_2),10 min of SCP(T_3),5 min after descending aorta unclamping(T_4),5 min after left subclavian artery unclamping(T_5),5 min after left common carotid artery unclamping(T_6),anonymous artery unclamping(T_7),when nasopharyngeal To returned to 35℉(T_8)and 10 min after CPB was terminated(T_9).The awakening time and the duration of ICU stay(days)were recorded.Pre- and postoperative neurological examination and brain CT scan were performed.Results All patients survived the operation and were discharged from hospital.No new brain infarction occurred.Transient neurologic dysfunction occurred in 2 patients in group H and 3 patients in group C.There was a positive linear relationship between PaO_2 and PjvO_2 during deep hypothermia in group H (r=0.541,P＜0.01).The PjvO_2 and SjvO_2 were significantly higher in group H than in group C.The awakening time and the ICU stay were significantly shorter in group H than in group C.Conclusion The hyperoxia management strategy can provide clinical prognosis than the conventional management strategy during deep hypothermia for total aortic arch replacement by supplying more dissolved oxygen.