Kim Jung Hoon, Eun Hyo Won, Kim Young Jae, Han Joon Koo, Choi Byung Ihn
Department of Radiology and Institute of Radiation Medicine, Seoul National University College of Medicine, 101 Daehang-no, Chongno-gu, Seoul, 110-744, Republic of Korea,
Abdom Imaging. 2013 Oct;38(5):1106-14. doi: 10.1007/s00261-013-0011-y.
To investigate staging accuracy of MR for pancreatic neuroendocrine neoplasms (PNETs) and imaging findings according to the tumor grade.
Our study consisted of 39 patients with PNET G1 (n = 24), PNET G2 (n = 12), and pancreatic neuroendocrine carcinoma (PNEC) (n = 3). All underwent preoperative MRI. Two radiologists retrospectively reviewed MR findings including tumor margin, SI on T2WI, enhancement patterns, degenerative change, duct dilation, and ADC value. They also assessed T-stage, N-stage, and tumor size. Statistical analyses were performed using Chi square tests, ROC analysis, and Fisher's exact test.
Specific findings for PNEC or PNET G2 were ill-defined borders (P = 0.001) and hypo-SI on venous- and delayed-phase (P = 0.016). ADC value showed significant difference between PNET G1 and G2 (P = 0.007). The Az of ADC value for differentiating PNET G1 from G2 was 0.743. Sensitivity and specificity were 70% and 86%. Accuracy for T-staging was 77% (n = 30) and 85% (n = 33), and for N-staging was 92% (n = 36) and 87% (n = 34) with moderate agreement. T-stage showed significant difference according to tumor grade (P < 0.001), although there was no significant difference in tumor size or N-stage.
Ill-defined borders and hypo-SI on venous- and delayed-phase imaging are common findings of higher grade PNET, and ADC value is helpful for differentiating PNET G1 from G2. MR is useful for preoperative evaluation of T-, N-stage. Tumor size of PNET and T-stage showed significant difference according to tumor grade.
探讨磁共振成像（MR）对胰腺神经内分泌肿瘤（PNETs）的分期准确性以及根据肿瘤分级的影像学表现。
我们的研究纳入了39例患者，其中PNET G1（n = 24）、PNET G2（n = 12）和胰腺神经内分泌癌（PNEC）（n = 3）。所有患者均接受了术前MRI检查。两名放射科医生回顾性分析了MR表现，包括肿瘤边界、T2加权像（T2WI）上的信号强度（SI）、强化模式、退行性改变、导管扩张和表观扩散系数（ADC）值。他们还评估了T分期、N分期和肿瘤大小。采用卡方检验、ROC分析和Fisher精确检验进行统计学分析。
PNEC或PNET G2的特异性表现为边界不清（P = 0.001）以及静脉期和延迟期低信号（P = 0.016）。ADC值在PNET G1和G2之间存在显著差异（P = 0.007）。ADC值区分PNET G1和G2的曲线下面积（Az）为0.743。敏感性和特异性分别为70%和86%。T分期的准确率分别为77%（n = 30）和85%（n = 33），N分期的准确率分别为92%（n = 36）和87%（n = 34），一致性中等。T分期根据肿瘤分级显示出显著差异（P < 0.001），尽管肿瘤大小或N分期无显著差异。
边界不清以及静脉期和延迟期成像低信号是高级别PNET的常见表现，ADC值有助于区分PNET G1和G2。MR对T、N分期的术前评估有用。PNET的肿瘤大小和T分期根据肿瘤分级显示出显著差异。
