Determination of time-course change rate for arterial xenon using the time course of tissue xenon concentration in xenon-enhanced computed tomography

In calculating tissue blood flow (TBF) according to the Fick principle, time-course information on arterial tracer concentration is indispensable and has a considerable influence on the accuracy of calculated TBF. In TBF measurement by xenon-enhanced computed tomography (Xe-CT), nonradioactive xenon gas is administered by inhalation as a tracer, and end-tidal xenon is used as a substitute for arterial xenon. There has been the assumption that the time-course change rate for end-tidal xenon concentration ( K e ) and that for arterial xenon concentration ( K a ) are substantially equal. Respiratory gas sampling is noninvasive to the patient and K e can be easily measured by exponential curve fitting to end-tidal xenon concentrations. However, it is pointed out that there would be a large difference between K e and K a in many cases. The purpose of this work was to develop a method of determining the K a value using the time course of tissue xenon concentration in Xe-CT. The authors incorporated K a into the Kety autoradiographic equation as a parameter to be solved, and developed a method of least-squares to obtain the solution for K a from the time-course changes in xenon concentration in the tissue. The authors applied this method of least-squares to the data from Xe-CT abdominal studies performed on 17 patients; the solution for K a was found pixel by pixel in the spleen, and its K a map was created for each patient. On the one hand, the authors obtained the average value of the K a map of the spleen as the calculated K a ( K a c a l c ) for each patient. On the other hand, the authors measured K a ( K a m e a s ) using the time-course changes in CT enhancement in the abdominal aorta for each patient. There was a good correlation between K a c a l c and K a m e a s ( r = 0.966 , P < 0.0001 ), and these two K a values were close to each other ( K a c a l c = 0.935 × K a m e a s + 0.089 ) . This demonstrates that K a c a l c would be close to the true K a value. Accuracy of TBF by Xe-CT can be improved with use of the average value of the K a map of an organ like the spleen that has a single blood supply (only arterial inflow).