Comparison of thermal clearance measurement of regional cerebral blood flow with radiolabelled microspheres.

Abstract
A thermal clearance technique for measuring cerebral blood flow is described and compared with the radiolabelled microsphere technique. The thermal technique involves measurement of the rewarming curve generated after bolus infusion of 4-5 ml of ice-cold saline into the common carotid artery with a subdural thermistor placed on the parietal cortex. Evaluation of the biexponential decay curves obtained with this technique demonstrated a close correlation with total hemispheric, parietal, and parietal gray blood flow determined by simultaneous microsphere measurement. Despite significant correlations (p < 0.001), scatter in the data produced a broad 95% confidence interval, thus making interpretation of blood flow with the thermal clearance technique impossible. Furthermore, instrumentation with the thermal probe, which required opening of the dura, blunted the cerebral blood flow response to hypercapnia. We conclude that the major limitations of the thermal clearance technique include 1) nonhomogeneous clearance function, 2) significant variability, and 3) depression of CO2 reactivity. These limitations must be addressed before this technique can be used reliably in the laboratory.