PET activation studies employing the autoradiographic technique and 15O-water or 15O-butanol use the difference between images acquired under baseline conditions and during activation to detect focal changes in cerebral blood flow which occur upon stimulus presentation. Typically, the activating task or baseline conditions are maintained throughout the entire imaging period. Simulations of the kinetics of these freely diffusible tracers suggest there may be an advantage to switching between activation and baseline conditions during the course of the study which results in images which maximize the difference signal rather than seeking to quantitate blood flow. We examine the potential of these switched protocols to increase signal-to-noise (S/N) in PET activation studies.
Methods: We examined S/N in activation studies using both-standard and switched paradigms with a simple switched protocol and dynamic three-dimensional PET data from human subjects. With tracer kinetic simulations, we investigated the sensitivity of the S/N gain to factors such as the shape of the input function, the time at which the conditions are switched and the magnitude of the activation.
Results: In human studies of activation sites in the visual cortex, primary motor and premotor areas, S/N improvements of 20%-30% were detected using the switched paradigms. Simulations show that this gain is virtually independent of activation magnitude and that there is a broad time window of 20 sec for making the switch between conditions. To obtain the highest S/N gain, a rapid bolus injection is required.
Conclusion: Switched paradigms have the potential to significantly increase S/N in PET activation studies. In human studies, the S/N increase averaged 25% which is equivalent to increasing the number of counts collected by 50%. Switched paradigms can be used to maximize the difference signal in many activation studies, and do not preclude the absolute quantitation of blood flow using the standard autoradiographic technique.