Department of Neurosurgery, Graduate School of Biomedical & Health Sciences, Hiroshima University, Hiroshima, Japan.
Eur J Radiol. 2013 Apr;82(4):658-63. doi: 10.1016/j.ejrad.2012.11.037. Epub 2013 Jan 10.
We assessed whether a high b-value DWI at b=4000s/mm(2) would discriminate the histopathological differentiation of the tumor grade of meningiomas, and also focused on the relationship between radiologic features and the tumor grade.
We acquired DWI at 3T with b=1000 and b=4000s/mm(2) in 77 patients (42, 31 and 4 patients were WHO grades I (G1), II (G2), and III (G3), respectively). The apparent diffusion coefficient (ADC) was measured by placing multiple regions of interest (ROIs) on ADC maps. The ADC values of each tumor were determined preoperatively from several ROIs, and expressed as the minimum (ADCMIN), mean (ADCMEAN), and maximum absolute values (ADCMAX). We evaluated the relationship between ADCs and histological findings, and assessed the radiologic features such as tumor location, tumor size, presence/absence of peritumoral edema, shape of the tumor, presence/absence of bone destruction or hyperplasia, status of contrast enhancement, presence/absence of calcification and cyst.
ADCs of the meningiomas were inversely correlated with the histological grade of meningiomas. According to results of the discriminant analysis, the apparent log likelihood value was greatest for ADCMIN at b=4000. Furthermore, only the ADCMIN value at b=4000 was significantly correlated with the histological grade of meningiomas when we performed a multiple logistic regression analysis to identify the significant independent factors such as shape of tumor, presence/absence of bone destruction, status of contrast enhancement, presence/absence of cyst and ADCMIN at b=4000.
A meningioma with a low ADCMIN at a high b-value might imply a high-grade meningioma.
我们评估高 b 值 DWI（b=4000s/mm²）是否能区分脑膜瘤的组织病理学分化程度，并重点关注影像学特征与肿瘤分级之间的关系。
我们在 77 例患者（42 例、31 例和 4 例分别为 WHO 分级 I（G1）、II（G2）和 III（G3））中使用 3T 磁共振仪获取 b=1000 和 b=4000s/mm² 的 DWI。通过在 ADC 图上放置多个感兴趣区（ROI）来测量表观扩散系数（ADC）。术前从多个 ROI 确定每个肿瘤的 ADC 值，并表示为最小 ADC（ADCMIN）、平均 ADC（ADCMEAN）和最大绝对值 ADC（ADCMAX）。我们评估了 ADC 与组织学发现之间的关系，并评估了肿瘤位置、肿瘤大小、是否存在瘤周水肿、肿瘤形状、是否存在骨破坏或增生、对比增强状态、是否存在钙化和囊肿等影像学特征。
脑膜瘤的 ADC 值与脑膜瘤的组织学分级呈负相关。根据判别分析的结果，在 b=4000 时 ADCMIN 的表观对数似然值最大。此外，当我们进行多变量逻辑回归分析以确定肿瘤形状、骨破坏、对比增强状态、囊肿存在与否和 b=4000 时 ADCMIN 等有意义的独立因素时，只有 b=4000 时的 ADCMIN 值与脑膜瘤的组织学分级显著相关。
在高 b 值时 ADCMIN 较低的脑膜瘤可能提示为高级别脑膜瘤。
