Comparison of external lung monitoring with end-tidal air detection using the xenon-133 inhalation method

When the 133Xe inhalation method is employed for measuring regional cerebral blood flow, the arterial 133Xe concentration is usually approximated by the end-tidal air concentration. However, this approximation may be invalid in the presence of certain lung pathologies or when the breathing pattern is irregular. Jaggi and Obrist, using an intravenous injection of 133Xe, suggested that the counts detected by an external lung probe could provide an alternative estimate for arterial blood concentration once the noise produced by 133Xe in superficial tissues is removed from the signal. A mathematical model, based on hypotheses similar to theirs is presented here together with a new computational procedure for removing the noise. Results from normal rest studies on ten healthy young males indicate that the approximations for arterial blood concentration obtained from end-tidal air and from corrected lung counts are not equivalent when 133Xe is administered by inhalation. The concentration-time curves have different shapes, and these differences are reflected in blood flow values computed by head channel. However, there is no effect on comparisons between homologous regions of the left and right hemispheres