van Gils A P, van den Berg R, Falke T H, Bloem J L, Prins H J, Dillon E H, van der Mey A G, Pauwels E K
Department of Diagnostic Radiology and Nuclear Medicine, Leiden University Hospital, The Netherlands.
AJR Am J Roentgenol. 1994 Jan;162(1):147-53. doi: 10.2214/ajr.162.1.8273654.
Contrast enhancement with gadopentetate dimeglumine has been advocated to increase the efficacy of MR imaging for paragangliomas of the head and neck. However, contrast media are expensive, time-consuming to use, and involve minimal but not negligible risks. The purpose of this study was to determine if the use of contrast material is warranted in patients undergoing MR imaging for the diagnosis of paragangliomas of the head and neck.
Unenhanced MR images were compared with images obtained after administration of gadopentetate dimeglumine in 23 healthy subjects and 37 patients who had a total of 71 tumors. Three combinations of sequences were reviewed independently and in a random order by four observers who had no clinical information. Combination A comprised enhanced and unenhanced T1-weighted sequences, combination B comprised unenhanced T1- and T2-weighted sequences, and combination C was a combination of all sequences. A four-point scale of certainty was used. CT, scintigraphic, angiographic, and surgicopathologic findings were used as the standard of reference. Results were subjected to alternative free-response receiver-operating-characteristic (AFROC) scoring and statistical analysis.
The mean areas under the AFROC curve for combinations A, B, and C were 0.761, 0.856, and 0.827, respectively. Mean sensitivity/specificity values after dichotomizing the scoring results were 0.73/0.94, 0.79/0.95, and 0.78/0.94 for combinations A, B, and C, respectively. The performance of combinations B and C did not differ markedly, but both combinations were significantly better than combination A. In a relatively large percentage (36%) of small postoperative tumor residues not detected on unenhanced images, however, gadopentetate dimeglumine allowed detection.
The results of this study indicate that, in general, the use of gadopentetate dimeglumine is not necessary for the detection of head and neck paragangliomas. The addition of contrast-enhanced imaging does not increase the sensitivity or specificity compared with imaging without enhancement. Only when searching for small postoperative tumor residues is the addition of gadopentetate dimeglumine warranted.
有人主张使用钆喷酸葡胺进行对比增强,以提高磁共振成像(MR)对头颈部副神经节瘤的诊断效能。然而,造影剂价格昂贵,使用耗时,且存在虽小但不可忽视的风险。本研究的目的是确定在接受MR成像以诊断头颈部副神经节瘤的患者中,使用造影剂是否必要。
对23名健康受试者和37名共有71个肿瘤的患者,将未增强的MR图像与注射钆喷酸葡胺后获得的图像进行比较。由四名不了解临床信息的观察者以随机顺序独立审查三种序列组合。组合A包括增强和未增强的T1加权序列,组合B包括未增强的T1加权和T2加权序列,组合C是所有序列的组合。采用四分制确定性量表。以CT、闪烁扫描、血管造影和手术病理结果作为参考标准。对结果进行交替自由响应接收器操作特征(AFROC)评分和统计分析。
组合A、B和C的AFROC曲线下平均面积分别为0.761、0.856和0.827。将评分结果二分法处理后的平均灵敏度/特异度值,组合A、B和C分别为0.73/0.94、0.79/0.95和0.78/0.94。组合B和C的表现没有明显差异,但这两种组合均显著优于组合A。然而,在未增强图像上未检测到的相对较大比例(36%)的术后小肿瘤残留中,钆喷酸葡胺能够检测到。
本研究结果表明,一般而言,检测头颈部副神经节瘤无需使用钆喷酸葡胺。与未增强成像相比,添加对比增强成像并未提高灵敏度或特异度。仅在寻找术后小肿瘤残留时,才有必要添加钆喷酸葡胺。