Study Objective: To evaluate the effects of controlled hypotension combined with hemodilution on human middle cerebral artery flow velocity (Vmca) by transcranial Doppler ultrasonography. Design: Randomized prospective study. Setting: inpatient surgery at Nagasaki Rosai Hospital. Patients: Thirty American Society of Anesthesiologists physical status I and II patients scheduled for total hip arthroplasty. Interventions: Anesthesia was maintained with nitrous oxide-oxygen (N2O-O-2) and sevoflurane during normocapnia. Hemodilution was carried out after induction of anesthesia, in which blood was withdrawn then replaced with the same amount of hydroxyethyl starch to achieve a final hematocrit level of 32% (group A = mild hemodilution group, N = 15) or 24% (group B = moderate hemodilution group, N = 15). In both groups, controlled hypotension was induced with prostaglandin E, to maintain mean arterial pressure at approximately 55 mm Hg for 80 minutes. Measurements and Main Results: Vmca and blood gas were measured before hemodilution, after hemodilution, 80 minutes after starting hypotension,and 60 minutes after recovery from hypotension. Vmca significantly increased in group A (+122%) and group B (+156%) after each hemodilution. In group B, Vmca was significantly greater than baseline values at 80 minutes after starting hypotension (+135%) and 60 minutes after recovery from hypotension (+140%). Conclusion: The combination of moderate hemodilution, such as hematocrit value of 24%, and prostaglandin El-induced hypotension would not impair middle cerebral artery flow during sevoflurane-N2O-O-2 anesthesia during normocapnia. (c) 2005 Elsevier Inc. All rights reserved.