Measurement of activated rCBF by the 133Xe inhalation technique: a comparison of total versus partial curve analysis.

Abstract

SUMMARY An initial assessment of the differential sensitivity of total versus partial curve analysis in estimating task related focal changes in cortical blood flow measured by the 133 Xe inhalation technique was accomplished by comparing the patterns during the performance of two sensorimotor tasks by normal subjects. The validity of these patterns was evaluated by comparing them to the activation patterns expected from activation studies with the intra-arterial technique and the patterns expected from neuropsychological research literature. Subjects were 10 young adult nonsmoking healthy male volunteers. They were administered two tasks having identical sensory and cognitive components but different response requirements (oral versus manual). The regional activation patterns produced by the tasks varied with the method of curve analysis. The activation produced by the two tasks was very similar to that predicted from the research literature only for total curve analysis. To the extent that the predictions are correct, these data suggest that the 133 Xe inhalation technique is more sensitive to regional flow changes when flow parameters are estimated from the total head curve. The utility of the total head curve analysis will be strengthened if similar sensitivity is demonstrated in future studies assessing normal subjects and patients with neurological and psychiatric disorders.Stroke Xe in the mouth, sinuses and pharynx is one of the more serious of these methodological problems. Obrist et al 2 3 tried to reduce the influence of these artifacts by fitting a model to the subset of data collected after the concentration of xenon in end-expired air has fallen below 20 percent of its maximum value. However, their simulation studies showed that, in the absence of scattered radiation, estimates for K, (gray matter clearance) match the true value when model fitting is started near time zero and decline as the start fit time is increased. Thus, although the effects of scattered radiation are minimized, data corresponding to the inhalation phase (during which most of the relevant information on gray matter flow occurs) are discarded. Hazelrig et al. 4 have recently proposed a time domain correction method to adjust for the presence of these artifacts, thereby allowing utilization of the entire head curve in the estimation of the blood flow parameters.One of our long term goals is to determine whether analysis of the total head curve can improve the sensitivity of the l33 Xe inhalation technique in demonstrating focal activation of cortical flow relative to focal changes in flow estimated by the partial curve method. The purpose of this study was to make an initial assessFrom the

Analysis of inhalation rCBF data.

“…Subsequent studies of 10 subjects by this same group showed that the xenon-133 inhalation technique is more sensitive to regional flow changes when parameters are estimated from the total head curve. 10 These results provide additional evidence for total head curve analysis using Equation 10 against partial analysis with the DSF approach.…”

“…Fitting all head data with a two-compartment, one-artifact model containing P 3 and At as unknowns (Equation 10) results in both a statistically and a clinically significant reduction in the variance of k, compared to DSF, a fixed At, and no APA term.…”

“…These investigators performed whole-curve analysis with a model that was essentially the same as Equation 10. Subsequent studies of 10 subjects by this same group showed that the xenon-133 inhalation technique is more sensitive to regional flow changes when parameters are estimated from the total head curve.…”

Comparison of Quantitative Autoradiographic and Xenon-133 Clearance Methods: Correlation of Gray and White Matter Cerebral Blood Flow with Compartmental Blood Flow Indices

“…The intracarotid xenon-133 clearance technique permits repeated measurements of CBF in either humans or animals using the external monitoring of a freely diffusible tracer (Hoedt-Rasmussen et aI., 1966; Lassen and Ingvar, 1972;Fitch et aI., 1978;Hetz et aI., 1984 xenon-133 clearance techniques are still widely used in humans today (Obrist et aI., 1975;Wyper et aI., 1976;Leli et a\., 1985). However, uncertainty remains concerning the interpretation of the bicom partmental model used in the analysis of most clearance curves (Iliff et aI., 1974;Marcus et aI., 1981).…”