Quantitative Multiple Tracer Autoradiography: Considerations in Optimizing Precision and Accuracy

In early autoradiographic studies of cerebral function, only single radionuclide experiments were generally performed. Animals were given quantities of a radiolabeled tracer, an autoradio­ graph representing the distribution of the tracer was produced, and one functional parameter was measured. When the relationships between two or more parameters were of interest, studies were performed on paired series of animals. The primary problem with such paired series is the masking of significant differences between the groups caused by variations of data within the groups. Reports for such commonly measured pa­ rameters as local cerebral glucose metabolic rate (LCMRglu) and local cerebral blood flow (LCBF) generally show standard deviations of 15-25% of the values for individual structures in normal an­ imals (Sokoloff et aI., 1977; Sakurada et aI., 1978). Causes of this data variability include ( 1) physiolog­ ical interanimal variability, (2) errors in times of ar­ terial samples, (3) intrastructural variability, (4) variations in tissue thickness, and (5) variations in uniformity of film response. To circumvent these problems, techniques whereby two or more parameters can be measured simultaneously have been recently developed. In these techniques, two or more radiolabeled tracers are administered simultaneously. Two or more au­ toradiographs are then produced, each representing more or less the distribution of one of the tracers. Such multiple radionuclide techniques can poten­ tially eliminate or substantially reduce the major sources of variability of single label comparison studies. As two or more parameters are compared