Study Objective: To investigate the effects of nicardipine-, nitroglycerin-, and prostag landine E-1-induced hypotension on cerebrovascular carbon dioxide (CO2) reactivity over a wide range of arterial CO2 tension (PaCO2) (PaCO2; range 25 to 50 mmHg). Design: Prospective, randomized study. Setting: Operating room of a university-affiliated hospital. Patients: 36 ASA physical status I and II patients without cerebrovascular disease, hypertension, or diabetes mellitus, undergoing an elective abdominal surgery. Interventions: Patients were randomly allocated to one of three groups (nicardipine-, nitroglycerin-, or prostaglandin E-1-induced hypotension group; 12 in each group). Anesthesia was induced and maintained with a bolus dose, followed by a continuous infusion of propofol (6.7 +/- 1.5 mg/kg/hr) and fentanyl (1.68 +/- 0.4 mu g/kg/hr). Deliberate hypotension of mean arterial pressure 55 to 60 mmHg was induced and maintained with a bolus dose, followed by a continuous infusion of nicardipine (6.80 +/- 0.75 mu g/kg/min), nitroglycerin (3.20 +/- 1.10 mu g/kg/min), or prostaglandin E-1 (0.103 +/- 0.052 mu g/kg/min). Measurements and Main Results: Time-averaged mean red blood cell velocity in the right middle cerebral artery (Vmca) at PaCO2 ranging from 25 to 50 mmHg was measured with transcranial Doppler ultrasonography. A minimum of six simultaneous measurements of Vmca and PaCO2 were obtained during baseline and deliberate hypotension in each patient. Absolute slope between Vmca and PaCO2 during baseline and deliberate hypotension was determined individually by linear regression analysis. Absolute slope was treated as the variable, because it yielded a significant close correlation coefficient (r > 0.95; p < 0.05). Comparisons between baseline and deliberate hypotension were made by analysis of variance for repeated measures. Mean absolute slope was significantly reduced from 1.88 +/- 0.57 cm/sec/mmHg (mean +/- SD) to 1.21 +/- 0.46 in the nicardipine group (p < 005), from 1.75 +/- 0.69 to 1.35 +/- 0.47 in the nitroglycerin group (p < 0.05), and from 1.95 +/- 0.89 to 1.33 +/- 0.70 (p < 0.05) in the prostaglandin E-1 group, respectively. Conclusion: Nicardipine-, nitroglycerin-, and prostaglandin E-1-induced hypotension attenuate the human cerebrovascular CO2 reactivity during propofol-fentanyl anesthesia. (C) 1999 by Elsevier Science Inc.