Stable xenon CT cerebral blood flow measurements computed by a single compartment--double integration model in normal aging and dementia.

Programs have been developed, using a single compartmental analysis model, that provide rapid computer derivation for values of both local cerebral blood flows (LCBF) and local tissue: blood partition coefficients (L lambda) using inhaled stable xenon gas (Xes) as the indicator. These programs are planned for utilization of raw data points derived from serial computed tomography scans made between the 3rd and 9th min after 35% Xes inhalation, while concentrations of end-tidal Xes (PEXes) are concurrently monitored. Double integration and least squares fitting permitted estimation of corresponding arterial (assumed from PEXes) and tissue Xes concentrations at each scanning interval during Xes saturation. Derived estimates for L lambda and LCBF values for each region of interest mathematically approximate values saturated to infinity. The method described can be utilized for any freely diffusible indicator. Cross-sectional analysis of results obtained in 13 normal healthy volunteers between 20 and 80 years of age showed no age-related changes in L lambda values but progressive decreases in blood flow of cortex, basal ganglia, and white matter with advancing age. In senile dementia of Alzheimer's type (SDAT, n = 8) L lambda values were likewise found to be unchanged, but cortical and thalamic gray matter LCBF values were significantly reduced compared to age-matched normal volunteers.