Purpose

To assess the diagnostic performance of the LR-M criteria of Contrast-Enhanced Ultrasound Liver Imaging Reporting and Data System version 2017 in differentiating intrahepatic cholangiocarcinoma (ICC) from hepatocellular carcinoma (HCC) in patients with and without risk factors for HCC.

Methods

Fifty-four ICC in patients with risks and 55 ICC in patients without risks and matched control cases of HCC with and without risks (n = 59 and n = 55, respectively) were enrolled. The enhanced features of the lesions were retrospectively analyzed according to LR-M criteria. The diagnostic performances including the area under the receiver operating characteristic curve (AUC), sensitivity, and specificity of LR-M criteria were assessed.

Result

Peripheral rim-like hyperenhancement, early washout (< 45 or 60s), and marked washout did not differ between ICCs with and without risks, while all of these features were more common in ICCs than in HCCs (p < 0.05) no matter if patients were with and without risk factors. Using the LR-M criteria to differentiate ICC from HCC, the AUC, sensitivity, specificity, and accuracy were 0.92, 97.25%, 87.72%, and 92.38%, respectively. If early washout onset was adjusted to < 45 s, the specificity was significantly increased to 95.61% (p = 0.004) without losing sensitivity (96.33%, p = 0.945). The rate of HCCs misdiagnosed as ICCs would decrease from 12.3 to 4.4%.

Conclusion

Although the LR-M criteria showed high sensitivity in distinguishing ICCs from HCCs in patients with and without risks, the specificity would be significantly increased after adjustments to current criteria.

Key Points

• The LR-M criteria of CEUS-LI-RADS v2017 could be used for distinguishing ICC from HCC not only in patients with risk factors for HCC but also in those without risk factors.

• The diagnostic performance of differentiating ICC from HCC by using the LR-M criteria showed high AUC (0.92), high sensitivity (97.25%), intermediate specificity (87.72%), and high accuracy (92.38%).

• If the onset of early washout was adjusted to < 45 s, the specificity was significantly increased from 87.72 to 95.61% (p = 0.004) without losing sensitivity (p = 0.945), and the rate of HCCs misdiagnosed as ICCs would decrease from 12.3 to 4.4%.

目的

为了评估LR-M超声造影肝脏成像报告和数据系统2017版标准对区分肝内胆管癌（ICC）和肝细胞癌（HCC）有无肝癌危险因素的诊断性能。

方法

纳入有风险的患者中的54个ICC和无风险的患者中55个ICC，以及有或没有风险的HCC对照病例（分别为n  = 59和n  = 55）。根据LR-M标准回顾性分析病变的增强特征。评估了诊断性能，包括接收器工作特征曲线（AUC）下的面积，灵敏度和LR-M标准的特异性。

结果

在有风险和无风险的ICC之间，外周缘样过度增强，早期冲洗（<45或60s）和明显的冲洗没有差异，而所有这些特征在ICC中比在HCC中更为普遍（p  <0.05）有和没有危险因素。使用LR-M标准区分ICC和HCC，AUC，敏感性，特异性和准确性分别为0.92、97.25％，87.72％和92.38％。如果将早期冲洗开始时间调整为<45 s，则特异性会显着提高至95.61％（p  = 0.004），而不会失去敏感性（96.33％，p  = 0.945）。误诊为ICC的HCC率将从12.3％降至4.4％。

结论

尽管LR-M标准在区分有风险和无风险的患者中显示出区分ICC和HCC的高敏感性，但在调整当前标准后特异性会显着提高。

关键点

•CEUS-LI-RADS v2017的LR-M标准不仅可以用于具有HCC危险因素的患者，而且可以用于没有危险因素的患者，以区分ICC和HCC。

•使用LR-M标准将ICC与HCC区分的诊断性能显示出高AUC（0.92），高灵敏度（97.25％），中间特异性（87.72％）和高精度（92.38％）。

•如果将早期洗脱的开始时间调整为<45 s，则特异性从87.72％显着提高至95.61％（p = 0.004），而不会失去敏感性（p = 0.945），被误诊为ICC的HCC比率将从12.3降低至4.4％。