J Nucl Med. 2007; 48 (Supplement 2):177P
General Clinical Specialties: Gastroenterology
Hepatobiliary and Pancreas |
Functional mapping of Tc-99m GSA regional hepatic clearance from whole hepatic clearance and static liver SPECT: Calculation of whole hepatic clearance from dynamic planar image
Noriyuki Shuke1,
Atsutaka Okizaki1,
Junichi Sato1,
Hiroyuki Sugimori1 and
Tamio Aburano1
1 Department of Radiology, Asahikawa Medical College, Asahikawa, Japan
602
Objectives: Functional mapping of Tc-99m GSA hepatic clearance (GSA Cl) is useful for quantitative evaluation of regional hepatic function. It could be done by allotting whole GSA CL to each voxel in proportional to voxel count of a static SPECT image, as a simple alternative to voxel by voxel dynamic SPECT analysis. To make this method more clinically applicable, utilization of planar dynamic image for calculating whole GSA CL was proposed and compared with the method based on dynamic SPECT analysis. Methods: A total of 30 patients with chronic liver disease (Age: 69±9, M/F: 26/4) were studied with Tc-99m GSA dynamic SPECT (20 rotations / 20 min). Functional maps of regional GSA CL were generated by 2 methods. As a rigorous method, Patlak plot was done to calculate voxel GSA CL, using voxel liver time-activity curves (TACs) and cardiac blood pool TAC as an input function. As a simple alternative method, whole GSA CL was calculated from whole liver TAC and cardiac blood pool TAC obtained from dynamic planar images constituted from serial anterior projection images of dynamic SPECT by using Patlak plot. Then voxel GSA CL was determined by allotting whole GSA CL to each voxel in proportional to voxel count of static SPECT image generated by merging dynamic SPECT data (16-20 min). To compare the methods, whole GSA CL obtained from the voxel by voxel dynamic SPECT analysis was compared with that obtained from dynamic planar images. Results: Whole GSA CL (ml/min) obtained by the voxel by voxel dynamic SPECT analysis ranged from 55 to 476 (272±111), and those obtained by the analysis of dynamic planar images ranged from 58 to 451(248±102), showing good linear correlation (X / Y: dynamic planar / dynamic SPECT, Y=1.03 X + 18, R2=0.883, SEE =39, P < 0.00001). In Bland-Altman plot, mean of the difference (dynamic SPECT – dynamic planar) was 24.6 ml/min and no proportional error was observed. Conclusions: GSA CL calculated from dynamic planar image was comparable to that obtained by the dynamic SPECT analysis. Utilizing this method for linear allotment of whole GSA CL to static SPECT image, functional mapping of Tc-99m GSA regional hepatic clearance could be done without dynamic SPECT data acquisition.
