BACKGROUND AND PURPOSE: Cerebrovascular autoregulation is frequently measured in head-injured patients. We attempted to validate 4 bedside methods used for assessment of autoregulation. METHODS: PET was performed at a cerebral perfusion pressure (CPP) of 70 and 90 mm Hg in 20 patients. Cerebral blood flow (CBF) and cerebral metabolic rate for oxygen (CMRo2) were determined at each CPP level. Patients were sedated with propofol and fentanyl. Norepinephrine was used to control CPP. During PET scanning, transcranial Doppler (TCD) flow velocity in the middle cerebral artery was monitored, and the arterio-jugular oxygen content difference (AJDo2) was measured at each CPP. Autoregulation was determined as the static rate of autoregulation based on PET (SROR(PET)) and TCD (SROR(TCD)) data, based on changes in AJDo2, and with 2 indexes based on the relationship between slow waves of CPP and flow velocity (mean velocity index, Mx) and between arterial blood pressure and intracranial pressure (pressure reactivity index, PRx) RESULTS: We found significant correlations between SROR(PET) and SROR(TCD) (r2=0.32; P<0.01) and between SROR(PET) and PRx (r2=0.31; P<0.05). There were no significant associations between PET data and autoregulation as assessed by changes in AJDo2. Global CMRo2 was significantly lower at the higher CPP (P<0.01). CONCLUSIONS: Despite some variability, SROR(TCD) and PRx may provide useful approximations of autoregulation in head-injured patients. At least with our methods, CMRo2 changes with the increase in CPP; hence, flow-metabolism coupling may affect the results of autoregulation testing.