Department of Biomedical Imaging and Radiological Science, National Yang-Ming Medical University, Taiwan.
Br J Radiol. 2010 Dec;83(996):1023-8. doi: 10.1259/bjr/21476692. Epub 2010 Apr 22.
The purpose of this study was to determine the percentage of signal intensity loss (PSIL) threshold for the characterisation of focal liver lesions among patients with chronic liver disease. 55 nodules in 49 patients with chronic liver disease who underwent ferucarbotran-enhanced MR studies were included. Among the 49 patients, 40 had liver cirrhosis and 9 had chronic hepatitis. 8 haemangiomas, 3 focal nodular hyperplasia, 9 dysplastic nodules and 12 well, 19 moderately and 4 poorly differentiated hepatocellular carcinomas (HCCs) were revealed. The PSIL, signal-to-noise ratio and contrast-to-noise ratio of each lesion type were calculated. The diagnostic performance of PSIL on ferucarbotran-enhanced T(2) weighted images (PSIL(T2WI)) and T(2) weighted fat-suppression images (PSIL(FS-T2WI)) that characterised hepatic tumours was compared with receiver operating characteristic (ROC) analysis. Using ROC analysis, the diagnostic performance of PSIL(FS-T2WI) was superior to that of PSIL(T2WI) (p = 0.01). The mean PSIL(FS-T2WI) of the benign lesions was significantly higher than that of HCC (p<0.001), and the mean PSIL(FS-T2WI) of well-differentiated HCC was significantly higher than that of moderately/poorly differentiated HCCs (p = 0.001). With a PSIL(FS-T2WI) threshold of 40% in lesions characterising ferucarbotran-enhanced FS-T2WI, the sensitivity, specificity, accuracy, positive predictive value and negative predictive value were 88.6%, 95%, 90.9%, 96.9% and 82.6%, respectively. In conclusion, with ferucarbotran-enhanced FS-T2WI, a PSIL(FS-T2WI) threshold of 40% for characterising focal liver nodules among patients with chronic liver disease is recommended. It is useful for distinguishing HCC from benign nodules.
本研究旨在确定慢性肝病患者铁氟羧酸增强磁共振研究中局灶性肝脏病变特征化的信号强度损失（PSIL）阈值百分比。纳入了 49 例慢性肝病患者的 55 个结节，这些患者接受了铁氟羧酸增强磁共振研究。在这 49 例患者中，40 例患有肝硬化，9 例患有慢性肝炎。8 个肝血管瘤、3 个局灶性结节性增生、9 个不典型增生结节和 12 个高分化、19 个中分化和 4 个低分化肝细胞癌（HCC）被发现。计算了每种病变类型的 PSIL、信噪比和对比噪声比。比较了 PSIL 在铁氟羧酸增强 T2 加权图像（PSIL（T2WI））和 T2 加权脂肪抑制图像（PSIL（FS-T2WI））上对肝肿瘤的特征化的诊断性能，并进行了受试者工作特征（ROC）分析。使用 ROC 分析，PSIL（FS-T2WI）的诊断性能优于 PSIL（T2WI）（p=0.01）。良性病变的平均 PSIL（FS-T2WI）明显高于 HCC（p<0.001），高分化 HCC 的平均 PSIL（FS-T2WI）明显高于中/低分化 HCC（p=0.001）。当病变特征化铁氟羧酸增强 FS-T2WI 的 PSIL（FS-T2WI）阈值为 40%时，灵敏度、特异性、准确性、阳性预测值和阴性预测值分别为 88.6%、95%、90.9%、96.9%和 82.6%。总之，使用铁氟羧酸增强 FS-T2WI，建议在慢性肝病患者中为局灶性肝脏结节特征化设定 PSIL（FS-T2WI）阈值为 40%。它有助于区分 HCC 与良性结节。
