Department of diagnostic and interventional Radiology and Neuroradiology, University Hospital Essen, Essen, Germany,
Eur J Nucl Med Mol Imaging. 2014 Jan;41(1):79-88. doi: 10.1007/s00259-013-2560-2. Epub 2013 Oct 2.
The aim of this study was to evaluate the positron emission tomography (PET) component of [(18)F]choline PET/MRI and compare it with the PET component of [(18)F]choline PET/CT in patients with histologically proven prostate cancer and suspected recurrent prostate cancer.
Thirty-six patients were examined with simultaneous [(18)F]choline PET/MRI following combined [(18)F]choline PET/CT. Fifty-eight PET-positive lesions in PET/CT and PET/MRI were evaluated by measuring the maximum and mean standardized uptake values (SUVmax and SUVmean) using volume of interest (VOI) analysis. A scoring system was applied to determine the quality of the PET images of both PET/CT and PET/MRI. Agreement between PET/CT and PET/MRI regarding SUVmax and SUVmean was tested using Pearson's product-moment correlation and Bland-Altman analysis.
All PET-positive lesions that were visible on PET/CT were also detectable on PET/MRI. The quality of the PET images was comparable in both groups. Median SUVmax and SUVmean of all lesions were significantly lower in PET/MRI than in PET/CT (5.2 vs 6.1, p<0.05 and 2.0 vs 2.6, p<0.001, respectively). Pearson's product-moment correlation indicated highly significant correlations between SUVmax of PET/CT and PET/MRI (R=0.86, p<0.001) as well as between SUVmean of PET/CT and PET/MRI (R=0.81, p<0.001). Bland-Altman analysis revealed lower and upper limits of agreement of -2.77 to 3.64 between SUVmax of PET/CT vs PET/MRI and -1.12 to +2.23 between SUVmean of PET/CT vs PET/MRI.
PET image quality of PET/MRI was comparable to that of PET/CT. A highly significant correlation between SUVmax and SUVmean was found. Both SUVmax and SUVmean were significantly lower in [(18)F]choline PET/MRI than in [(18)F]choline PET/CT. Differences of SUVmax and SUVmean might be caused by different techniques of attenuation correction. Furthermore, differences in biodistribution and biokinetics of [(18)F]choline between the subsequent examinations and in the respective organ systems have to be taken into account.
本研究旨在评估经组织学证实的前列腺癌和疑似复发性前列腺癌患者中[（18）F]胆碱 PET/MRI 的正电子发射断层扫描（PET）成分，并将其与[（18）F]胆碱 PET/CT 的 PET 成分进行比较。
36 例患者在联合[（18）F]胆碱 PET/CT 后接受[（18）F]胆碱 PET/MRI 检查。采用基于感兴趣区（VOI）分析的最大和平均标准化摄取值（SUVmax 和 SUVmean）测量 PET/CT 和 PET/MRI 中 58 个 PET 阳性病变，应用评分系统确定两种 PET/CT 和 PET/MRI 图像的质量。采用 Pearson 积矩相关和 Bland-Altman 分析检验 PET/CT 和 PET/MRI 之间关于 SUVmax 和 SUVmean 的一致性。
PET/CT 上可见的所有 PET 阳性病变在 PET/MRI 上也可检测到。两组 PET 图像质量相当。所有病变的 SUVmax 和 SUVmean 在 PET/MRI 中均显著低于 PET/CT（5.2 比 6.1，p<0.05 和 2.0 比 2.6，p<0.001）。Pearson 积矩相关表明 SUVmax 之间具有高度显著相关性，PET/CT 和 PET/MRI（R=0.86，p<0.001）以及 SUVmean 之间的相关性，PET/CT 和 PET/MRI（R=0.81，p<0.001）。Bland-Altman 分析显示，SUVmax 的一致性界限在 PET/CT 与 PET/MRI 之间为-2.77 至 3.64，SUVmean 的一致性界限在 PET/CT 与 PET/MRI 之间为-1.12 至 2.23。
PET/MRI 的 PET 图像质量与 PET/CT 相当。SUVmax 和 SUVmean 之间存在高度显著的相关性。在[（18）F]胆碱 PET/MRI 中，SUVmax 和 SUVmean 均显著低于[（18）F]胆碱 PET/CT。SUVmax 和 SUVmean 的差异可能是由于不同的衰减校正技术引起的。此外，在随后的检查和各自的器官系统中，[（18）F]胆碱的生物分布和生物动力学之间存在差异，需要加以考虑。
