From the Departments of Radiology (B.T., B.N., C.S.J.L., C.G.A., M.D.N.) and Hematology (M.B.N., G.D.), Children's Hospital Los Angeles, 4650 Sunset Blvd, MS 81, Los Angeles, CA 90027.
Radiology. 2018 May;287(2):452-460. doi: 10.1148/radiol.2017171850. Epub 2017 Nov 30.
Purpose To determine whether whole-brain irradiation, chemotherapy, and primary brain pathologic conditions affect magnetic resonance (MR) imaging signal changes in pediatric patients independent of the administration of gadolinium-based contrast agents (GBCAs). Materials and Methods This institutional review board-approved, HIPAA-compliant study included 144 pediatric patients who underwent intravenous GBCA-enhanced MR imaging examinations (55 patients with primary brain tumors and whole-brain irradiation, 19 with primary brain tumors and chemotherapy only, 52 with primary brain tumors without any treatment, and 18 with neuroblastoma without brain metastatic disease). The signal intensities (SIs) in the globus pallidus (GP), thalamus (T), dentate nucleus (DN), and pons (P) were measured on unenhanced T1-weighted images. GP:T and DN:P SI ratios were compared between groups by using the analysis of variance and were analyzed relative to group, total cumulative number of doses of GBCA, age, and sex by using multivariable linear models. Results DN:P ratio for the radiation therapy group was greater than that for the other groups except for the group of brain tumors treated with chemotherapy (P < .05). The number of GBCA doses was correlated with the DN:P ratio for the nontreated brain tumor group (P < .0001). The radiation therapy-treated brain tumor group demonstrated higher DN:P ratios than the nontreated brain tumor group for number of doses less than or equal to 10 (P < .0001), whereas ratios in the nontreated brain tumor group were higher than those in the radiation therapy-treated brain tumor group for doses greater than 20 (P = .05). The GP:T ratios for the brain tumor groups were greater than that for the neuroblastoma group (P = .01). Conclusion Changes in SI of the DN and GP that are independent of the administration of GBCA occur in patients with brain tumors undergoing brain irradiation, as well as in patients with untreated primary brain tumors. RSNA, 2017.
目的 本研究旨在确定全脑放疗、化疗以及原发性脑部病变是否会影响接受钆对比剂（GBCA）增强磁共振成像（MRI）检查的儿科患者的 MRI 信号变化，且这种影响与 GBCA 的应用无关。
材料与方法 本研究经机构审查委员会批准，符合 HIPAA 规定，纳入 144 例接受 GBCA 增强 MRI 检查的儿科患者，包括 55 例接受全脑放疗且合并原发性脑肿瘤、19 例仅接受化疗且合并原发性脑肿瘤、52 例未接受任何治疗且合并原发性脑肿瘤、18 例无脑部转移瘤且患有神经母细胞瘤。于未增强 T1WI 上测量苍白球（GP）、丘脑（T）、齿状核（DN）和脑桥（P）的信号强度（SI）。采用方差分析比较各组间 GP/T 和 DN/P 比值，采用多变量线性模型分析组间、GBCA 总累积剂量、年龄和性别与比值的相关性。
结果 与其他组相比，放疗组的 DN/P 比值较高，除了化疗组的脑肿瘤患者（P <.05）。未接受治疗的脑肿瘤组中 GBCA 剂量数与 DN/P 比值呈正相关（P <.0001）。放疗组脑肿瘤患者接受剂量数小于或等于 10 时，其 DN/P 比值高于未接受治疗的脑肿瘤患者（P <.0001），而未接受治疗的脑肿瘤患者接受剂量数大于 20 时，其 DN/P 比值高于放疗组（P =.05）。脑肿瘤组的 GP/T 比值大于神经母细胞瘤组（P =.01）。
结论 接受脑放疗以及未接受治疗的原发性脑肿瘤的患者中，除了 GBCA 的应用外，DN 和 GP 的 SI 变化与脑肿瘤相关。
放射学学会，2017 年。
