Department of Radiology and Center for Imaging Sciences, Samsung Medical Center, Sungkyunkwan University School of Medicine, 81 Irwon-ro, Gangnam-gu, Seoul, 06351, Republic of Korea.
Department of Radiology, Seoul National University Hospital, Seoul National University College of Medicine, 101 Daehangno, Jongno-gu, Seoul, 03080, Republic of Korea.
Eur Radiol. 2022 Dec;32(12):8507-8517. doi: 10.1007/s00330-022-08900-6. Epub 2022 Jun 16.
To evaluate the diagnostic accuracy of perfluorobutane contrast-enhanced ultrasonography (CEUS) for hepatocellular carcinoma (HCC) and to explore how accuracy can be improved compared to conventional diagnostic criteria in at-risk patients.
A total of 123 hepatic nodules (≥ 1 cm) from 123 at-risk patients who underwent perfluorobutane CEUS between 2013 and 2020 at three institutions were retrospectively analyzed. Ninety-three percent of subjects had pathological results, except benign lesions stable in follow-up images. We evaluated presence of arterial phase hyperenhancement (APHE), washout time and degree, and Kupffer phase (KP) defects. KP defects are defined as hypoenhancing lesions relative to the liver in KP. HCC was diagnosed in two ways: (1) Liver Imaging Reporting and Data System (LI-RADS) criteria defined as APHE and late (≥ 60 s)/mild washout, and (2) APHE and Kupffer (AK) criteria defined as APHE and KP defect. We explored grayscale features that cause misdiagnosis of HCC and reflected in the adjustment. Diagnostic performance was compared using McNemar's test.
There were 77 HCCs, 15 non-HCC malignancies, and 31 benign lesions. An ill-defined margin without hypoechoic halo on grayscale applied as a finding that did not suggest HCC. Regarding diagnosis of HCC, sensitivity of AK criteria (83.1%; 95% confidence interval [CI]: 72.9-90.7%) was higher than that of LI-RADS criteria (75.3%; 95% CI: 64.2-84.4%; p = 0.041). Specificity was 91.3% (95% CI: 79.2-97.6%) in both groups.
On perfluorobutane CEUS, diagnostic criteria for HCC using KP defect with adjustment by grayscale findings had higher diagnostic performance than conventional criteria without losing specificity.
• Applying Kupffer phase defect instead of late/mild washout and adjusting with grayscale findings can improve the diagnostic performance of perfluorobutane-enhanced US for HCC. • Adjustment with ill-defined margins without a hypoechoic halo for features unlikely to be HCC decreases false positives for HCC diagnosis using the perfluorobutane-enhanced US. • After adjustment with grayscale findings, the sensitivity and accuracy of the APHE and Kupffer criteria were higher than those of the LI-RADS criteria; specificity was 91.3% for both.
评估全氟丁烷对比增强超声（CEUS）在肝细胞癌（HCC）诊断中的准确性，并探讨在高危患者中与常规诊断标准相比如何提高准确性。
回顾性分析 2013 年至 2020 年在三所机构接受全氟丁烷 CEUS 的 123 例高危患者的 123 个肝结节（≥1cm）。93%的患者有病理结果，除了在随访图像中稳定的良性病变。我们评估了动脉期高增强（APHE）、洗脱时间和程度以及库普弗（KP）缺陷。KP 缺陷定义为 KP 中相对于肝脏的低增强病变。HCC 通过两种方式诊断：（1）肝脏成像报告和数据系统（LI-RADS）标准定义为 APHE 和晚期（≥60s）/轻度洗脱，以及（2）APHE 和 Kupffer（AK）标准定义为 APHE 和 KP 缺陷。我们探讨了导致 HCC 误诊的灰度特征，并在调整中反映出来。使用 McNemar 检验比较诊断性能。
共有 77 例 HCC、15 例非 HCC 恶性肿瘤和 31 例良性病变。在灰度上应用无明确边界且无低回声晕的特征不提示 HCC。关于 HCC 的诊断，AK 标准的敏感性（83.1%；95%置信区间[CI]：72.9-90.7%）高于 LI-RADS 标准（75.3%；95%CI：64.2-84.4%；p=0.041）。两组的特异性均为 91.3%（95%CI：79.2-97.6%）。
在全氟丁烷 CEUS 上，使用 KP 缺陷并通过灰度特征调整的 HCC 诊断标准比不丧失特异性的常规标准具有更高的诊断性能。
使用 KP 缺陷代替晚期/轻度洗脱，并通过灰度特征进行调整，可提高全氟丁烷增强超声诊断 HCC 的诊断性能。
通过对不太可能为 HCC 的特征进行无明确边界且无低回声晕的调整，可以减少使用全氟丁烷增强超声诊断 HCC 的假阳性。
调整灰度特征后，APHE 和 Kupffer 标准的敏感性和准确性均高于 LI-RADS 标准；特异性均为 91.3%。
