Parametric Images of Standardized Uptake Values using P-18-FDG Attenuation Corrected Whole Body PET

Background and Purpose : Standardized uptake value(SUV) has been used as a quantitative index for differentiating benign and malignant tumors with F-18-FDG PET In this study, we produced whole body parametric images of SUV(WBPIS) by body weight normalization, and validated the values by comparison with SUV's calculated with regional scans. Subjects and Methods : Whole body scans were followed by regional scans sequentially on 23 patients. In whole body study, transmission and emission scans were acquired for 2 minutes and 6 minutes for each bed position, respectively. In regional study, transmission and emission scans were acquired for 20 minutes. Measured and segmented/ smoothed attenuation correction were applied using these 2 min transmission scans in whole body studies. The effects of attenuation correction on SUVs were evaluated quantitatively using F-18 filled cylindrical phantom. The mean and peak SUVs obtained from WBPIS were compared with SUVs of the regional scans. Results : In phantom studies, with any method of attenuation correction using regional or whole body studies of phantom, SUVs were nearly consistent. In whole body scan, SUV obtained using measured attenuation correction method was a little higher than SUV of regional scan. SUV obtained using segmented/smoothed attenuation correction method was a little lower. In patient studies, WBPIS using segmented/smoothed attenuation correction method was much smoother and more readable. SUVs of WBPIS obtained with both methods of attenuation correction were well correlated with SUVs of regional scans(r=0.9). SUVs of WBPIS with measured attenuation correction method were 5% lower than SUVs of regional scans. SUVs of WBPIS with segmented/smoothed attenuation correction method were 10% lower than SUVs of regional scans. The differences of SUVs of WBPIS by the two attenuation correction methods were relatively small compared with the possible differences derived from biological characteristics of tumors. Conclusion : We concluded that WBPIS could be useful in the quantification of tumor as well as in localization of whole body lesions, which were often outside the field of view in regional scan. WBPIS made using segmented/smoothed attenuation correction method could be used in clinical routines and SUVs from attenuation corrected F-18-FDG PET could be used interchangeably with SUVs of regional studies.