Comparative evaluation of different methods for calculation of cerebral blood flow (CBF) in nonanesthetized rabbits.

The measurement of cerebral blood flow (CBF) by the extracranial detection of the radioactivity of 133Xe injected into an internal carotid artery has proved to be of considerable value for the investigation of cerebral circulation in conscious rabbits. Methods are described for calculating CBF from the curves of clearance of 133Xe, and include exponential analysis (two-component model), initial slope, and stochastic method. The different methods of curve analysis were compared in order to evaluate the fitness with the theoretical model. The initial slope and stochastic methods, compared with the biexponential model, underestimate the CBF by 35% and 46% respectively. Furthermore, the validity of recording the clearance curve for 10 min was tested by comparing these CBF values with those obtained from the whole curve. CBF values calculated with the shortened procedure are overestimated by 17%. A correlation exists between the "10 min" CBF values and the CBF calculated from the whole curve; in spite of that, the values are not accurate for limited animal populations or for single animals. The extent of the two main compartments into which the CBF is divided was also measured. There is no correlation between CBF values and the extent of the relative compartment. This fact suggests that these two parameters correspond to different biological entities.