Journal of Computer Assisted Tomography 7(1):51—58, February © 1983 Raven Press, New York Design of Steady-State Positron Emission Tomography Protocols for Neurobehavioral Studies: CO 1 O and ‘ 9 Ne Kimberlee J. Kearfott, David A. Rottenberg, and Bruce T. Volpe Abstract: Although the [‘ 8 F]-2-fluoro-2-deoxyglucose positron emission to mographic technique for measuring regional glucose metabolic rate has been successfully employed for neurobehavioral studies, the long (>30 mm) equili bration time required may complicate the interpretation of experimental re sults. Positron emission tomography neurobehavioral protocols employing the continuous inhalation of CO’ 5 0 and 19 Ne were developed for measuring re gional cerebral blood flow during multiple control and stimulation periods. Timing, lung absorbed dose, statistical accuracy, and resolution were consid ered. Studies with ‘ 9 Ne require shorter equilibration and stimulation times than do CO’ 5 0 studies but entail higher absorbed doses and yield poorer imaging statistics. Key Words: Emission computed tomography—Neurobehavioral studies—Radioisotopes—Cerebral blood vessels, flow dynamics—Radiation dose. Positron emission tomography (PET) offers the possibility of obtaining regional information about brain function in vivo. Functional activation (e.g., perception, cognition, volition) is associated with increased neuronal metabolism (1). Increases in re gional glucose metabolic rate (rCMRglu) during vi sual, tactile, and aural stimulation have been dem onstrated by PET studies using [‘ 8 Fj-2-fluoro-2- deoxyglucose (‘ 8 FDG) (2). Phelps et al. (3) observed a correlation between the complexity of a visual stimulus and the percentage increase in rCMRglu. Neurobehavioral studies using ‘ 8 FDG involve continuous stimulation for a 20—40-mm period. Furthermore, the relative “metabolic importance” of stimulation during this period is proportional to the area under the plasma 18 FDG curve (4). Repeat closely spaced 18 FDG studies are difficult to obtain because of the long half-life of 18 F (110 mm); abso lute quantitation of ‘ 8 FDG images requires regional determination of Sokoloff’ s “lumped constant” (5) and four model rate constants for both diseased and normal tissue. A detailed sensitivity analysis of the 18 FDG technique has been published (4). From the Department of Neurology, Memorial Sloan- Kettering Cancer Center (K. 3. Kearfott and D. A. Rottenberg) and the Department of Neurology, Cornell University Medical College (B. T. Volpe), New York, NY. Address correspondence and reprint requests to Dr. Kearfott at Department of Neurology, Memorial Sloan-Kettering Cancer Center, 1275 York Avenue, New York, NY 10021. In normal human subjects, changes in regional cerebral blood flow (rCBF) faithfully reflect changes in regional cerebral metabolism (6,7). Steady-state measurements of rCBF during neuro behavioral stimulation offer the advantages of relative ease of performance and lack of dynamic imaging constraints. Steady-state PET techniques have been devel oped for quantitatively determining rCBF using 15 0-labeled CO 2 (8—10). The use of short-lived ra dionuclides for steady-state PET measurements permits accurate image quantitation (11) and the se rial (control versus stimulation) studies required for meaningful neurobehavioral investigations. Repro ducible activation studies were carried out using 77 Kr (half-life 1.24 h) and PET (12), but the sensitiv ity of such studies, as well as that of their single photon counterparts using slmKr (13), is limited by the low solubility of krypton in lung blood. Neon- 19, a positron emitter with a half-life of 17.6 s, has been produced in quantities sufficient for PET imaging (14,15). However, the limited solubiity of neon in blood and the difficulty of handling an ultra-short-lived isotope may restrict its use as a blood-flow tracer to studies with PET instruments of high sensitivity and to subjects in which high lung absorbed doses are permissible. This paper will examine the steady-state model kinetics, PET statistics, and critical organ absorbed doses for both C0 15 0 and 19 Ne measurements of 51