Comparison of 133 xenon ventilation equilibrium scan (XV) and 99m technetium transmission (TT) scan for use in regional lung analysis by 2D gamma scintigraphy in healthy and cystic fibrosis lungs.

BACKGROUND Quantification of particle deposition in the lung by gamma scintigraphy requires a reference image for location of regions of interest (ROIs) and normalization to lung thickness. In various laboratories, the reference image is made by a transmission scan ((57)Co or (99m)Tc) or gas ventilation scan ((133)Xe or (81)Kr). There has not been a direct comparison of measures from the two methods. METHODS We compared (99m)Tc transmission scans to (133)Xe equilibrium ventilation scans as reference images for 38 healthy subjects and 14 cystic fibrosis (CF) patients for their effects on measures of regional particle deposition: the central-to-peripheral ratio of lung counts (C/P); and ROI area versus forced vital capacity. Whole right lung ROI was based on either an isocontour threshold of three times the soft tissue transmission (TT) or a threshold of 20% of peak xenon ventilation counts (XV). We used a central ROI drawn to 50% of height and of width of the whole right lung ROI and placed along the left lung margin and centered vertically. RESULTS In general, the correlation of normalized C/P (nC/P) between the two methods was strong. However, the value of nC/P was significantly smaller for the XV method than the TT method. Regression equations for the relationship of nC/P between the two methods were, for healthy subjects, y=0.75x+0.61, R(2)=0.64 using rectangular ROIs and y=0.76x+0.45, R(2)=0.66 using isocontour ROIs; and for CF patients, y=0.94x+0.46, R(2)=0.43 and y=0.85x+0.42, R(2)=0.41, respectively. CONCLUSIONS (1) A transmission scan with an isocontour outline in combination with a rectangular central region to define the lung borders may be more useful than a ventilation scan. (2) Close correlation of nC/Ps measured by transmission or gas ventilation should allow confident comparison of values determined by the two methods.