Department of Diagnostic and Interventional Radiology and Neuroradiology, University Hospital Essen, University of Duisburg-Essen, Essen, Germany
Department of Diagnostic and Interventional Radiology, University Hospital Dusseldorf, University of Dusseldorf, Dusseldorf, Germany.
J Nucl Med. 2014 Dec;55(12):1930-5. doi: 10.2967/jnumed.114.146886. Epub 2014 Nov 13.
The aim of this study was to assess the diagnostic benefit of diffusion-weighted imaging (DWI) in an (18)F-FDG PET/MR imaging protocol for whole-body staging of women with primary or recurrent malignancies of the pelvis.
Forty-eight patients with a primary pelvic malignancy or suspected recurrence of a pelvic malignancy were included in our study. All patients underwent a whole-body (18)F-FDG PET/MR imaging examination that included DWI. Two radiologists separately evaluated the PET/MR imaging datasets without DWI followed by a second interpretation with DWI. First, both readers identified all primary tumors, as well as lymph node and distant metastases. In a second session, PET and DWI data were assessed qualitatively. Image interpretation comprised lesion conspicuity defined as visual lesion-to-background contrast (4-point ordinal scale) and diagnostic confidence (3-point ordinal scale) for all tumors. The results from histopathologic examination and cross-sectional imaging follow-up (≥6 mo) were used as the reference standard. Statistical analysis was performed to assess the significance of differences between obtained values.
Among the 122 suspected lesions seen, 98 (80.3%) were considered malignant. PET/MR imaging without DWI had a sensitivity, specificity, positive predictive value, negative predictive value, and diagnostic accuracy of 92.9%, 87.5%, 96.8%, 75.0%, and 91.8%, respectively, for the detection of malignant lesions. PET/MR imaging with DWI had slightly higher values (94.9%, 83.3%, 95.9%, 80.0%, and 92.6%, respectively), but the difference was not significant (P > 0.05). In the qualitative assessment of lesion-to-background contrast, PET had significantly (P < 0.05) higher values (3.79 ± 0.58) than DWI (3.63 ± 0.77). Furthermore, significantly (P < 0.05) higher scores were found for diagnostic confidence using PET (2.68 ± 0.64) for the determination of malignant lesions, when compared with DWI (2.53 ± 0.69).
DWI in PET/MR imaging has no diagnostic benefit for whole-body staging of women with pelvic malignancies. The omission of DWI for staging or restaging gynecologic cancer may significantly reduce examination times, thus increasing patient comfort without a relevant decrease in diagnostic competence.
本研究旨在评估扩散加权成像（DWI）在（18）F-FDG PET/MR 全身分期成像方案中对原发性或复发性盆腔恶性肿瘤患者的诊断获益。
本研究纳入了 48 例原发性盆腔恶性肿瘤或疑似盆腔恶性肿瘤复发的患者。所有患者均行全身（18）F-FDG PET/MR 成像检查，包括 DWI。两名放射科医生分别在不使用 DWI 的情况下评估 PET/MR 成像数据集，然后在使用 DWI 的情况下进行第二次解释。首先，两位读者均识别出所有原发性肿瘤、淋巴结和远处转移。在第二次会议中，对 PET 和 DWI 数据进行定性评估。图像解释包括病变显示度（4 分有序量表）和对所有肿瘤的诊断信心（3 分有序量表）。组织病理学检查和横断面成像随访（≥6 个月）的结果被用作参考标准。采用统计学分析评估获得值之间差异的显著性。
在 122 个可疑病变中，98 个（80.3%）被认为是恶性的。不使用 DWI 的 PET/MR 成像对恶性病变的检测灵敏度、特异性、阳性预测值、阴性预测值和诊断准确率分别为 92.9%、87.5%、96.8%、75.0%和 91.8%。使用 DWI 的 PET/MR 成像略有更高的值（分别为 94.9%、83.3%、95.9%、80.0%和 92.6%），但差异无统计学意义（P＞0.05）。在病变与背景对比的定性评估中，PET 的数值（3.79±0.58）明显高于 DWI（3.63±0.77）（P＜0.05）。此外，在使用 PET 确定恶性病变时，诊断信心的评分明显更高（2.68±0.64），与 DWI（2.53±0.69）相比（P＜0.05）。
DWI 对盆腔恶性肿瘤患者的全身分期无诊断获益。对于妇科癌症的分期或再分期，省略 DWI 可能会显著缩短检查时间，从而增加患者的舒适度，而不会对诊断能力产生明显影响。
