Saksena Mansi, Harisinghani Mukesh, Hahn Peter, Kim John, Saokar Anuradha, King Benjamin, Weissleder Ralph
Division of Abdominal Imaging, Massachusetts General Hospital and Harvard Medical School, White 270, 55 Fruit Street, Boston, MA 02114, USA.
AJR Am J Roentgenol. 2006 Dec;187(6):W582-8. doi: 10.2214/AJR.05.0873.
This study was performed to empirically evaluate T2-weighted fast spin-echo, moderately T2*-weighted gradient-refocused echo (GRE), and heavily T2*-weighted GRE sequences to determine which sequence is the most effective for nodal characterization on lymphotropic nanoparticle-enhanced MRI (LNMRI).
The study included 65 patients who had proven primary cancer and were scheduled for either surgical lymph node dissection or imaging-guided lymph node biopsy. All patients underwent LNMRI using T2-weighted fast spin-echo, moderately T2*-weighted GRE, and heavily T2*-weighted GRE sequences. Unequivocal correlation of histopathology and MRI could be made in 140 nodes and only these were included in the analysis. Two blinded reviewers performed qualitative analysis of the nodes. Alternative free-response receiver operating characteristic (ROC) curves with a continuous rating scale were plotted for each sequence for both reviewers and the diagnostic accuracy of fast spin-echo T2-weighted and GRE T2*-weighted images were compared by calculating the area under the curve (A(Z)). A two-tailed Student's t test was performed to test the significance (p < 0.05) of the differences between the ROC curves derived from the three sequences.
Irrespective of reviewer experience, T2*-weighted sequences showed better nodal characterization when compared with T2-weighted sequences. For both reviewers, there was a statistically significant difference between the A(Z) for T2- and the two T2*-weighted sequences (p < 0.05). Neither reviewer showed a statistically significant difference between the two T2*-weighted sequences.
GRE T2*-weighted sequences are superior for nodal characterization on LNMRI to fast spin-echo T2-weighted sequences. Imaging protocols for LNMRI should include fast spin-echo T2-weighted imaging for anatomic localization, but characterization of nodes should be based on their appearance on contrast-enhanced T2*-weighted images. The T2*-weighted images acquired with dual TE values, one of which is intermediate and the other longer, improve nodal characterization.
本研究旨在通过实证评估T2加权快速自旋回波序列、中度T2加权梯度重聚回波(GRE)序列和重度T2加权GRE序列,以确定哪种序列在亲淋巴纳米颗粒增强磁共振成像(LNMRI)中对淋巴结特征的显示最为有效。
该研究纳入了65例已确诊原发性癌症且计划进行手术淋巴结清扫或影像引导下淋巴结活检的患者。所有患者均接受了使用T2加权快速自旋回波序列、中度T2加权GRE序列和重度T2加权GRE序列的LNMRI检查。140个淋巴结的组织病理学与MRI结果具有明确的相关性,仅将这些淋巴结纳入分析。两名盲法阅片者对淋巴结进行定性分析。为每位阅片者针对每个序列绘制具有连续评分量表的替代自由反应式接收者操作特征(ROC)曲线,并通过计算曲线下面积(A(Z))比较快速自旋回波T2加权图像和GRE T2*加权图像的诊断准确性。进行双尾Student t检验以检验源自三个序列的ROC曲线之间差异的显著性(p < 0.05)。
无论阅片者经验如何,与T2加权序列相比,T2加权序列在淋巴结特征显示方面表现更佳。对于两位阅片者而言,T2加权序列与两个T2加权序列的A(Z)之间均存在统计学显著差异(p < 0.05)。两位阅片者均未发现两个T2*加权序列之间存在统计学显著差异。
在LNMRI中,GRE T2加权序列在淋巴结特征显示方面优于快速自旋回波T2加权序列。LNMRI的成像方案应包括用于解剖定位的快速自旋回波T2加权成像,但淋巴结特征的显示应基于其在对比增强T2加权图像上的表现。采用双TE值获取的T2*加权图像,其中一个TE值为中等,另一个为较长值,可改善淋巴结特征的显示。