The value of contrast-enhanced ultrasound combined with microflow imaging in predicting microvascular invasion of hepatocellular carcinoma before operation

Abstract

OBJECTIVE:

To evaluate the preoperative predictive value of contrast-enhanced ultrasound (CEUS) combined with microflow imaging (MFI) in microvascular invasion (MVI) of hepatocellular carcinoma (HCC).

METHODS:

In our study, 80 patients with HCC were analyzed retrospectively. According to the gold standard of postoperative pathology, the patients were divided into MVI positive group (n = 39) and MVI negative group (n = 41). we were to analyze the correlation between CEUS and MVI in combination with MFI, to identify independent risk factors for the occurrence of MVI positive, and to analyze the predictive efficacy of every independent risk factor and their combination in preoperative prediction of MVI.

RESULTS:

In our study, 80 patients were enrolled, including 39 patients in the MVI-positive group and 41 patients in the MVI-negative group, with a MVI-positive rate of 48.8%. By univariate analysis and multivariate analysis, it was found that there were statistically significant differences in enhancement range extension, start time of wash out and CEUS-MFI between the two groups, which were independent risk factors for MVI-positive. The combination of three independent risk factors is more effective than single one in predicting MVI of HCC.

CONCLUSIONS:

CEUS combined with MFI is feasible for the preoperative prediction of MVI in HCC, and can provides meaningful help for individualized clinical treatment.

Keywords

Hepatocellular carcinoma (HCC)microvascular invasion (MVI)contrast-enhanced ultrasound (CEUS)

1 Introduction

Primary liver cancer is the sixth most common cancer in the world, and hepatocellular carcinoma (HCC) accounts for about 80% of it. HCC is one of the leading causes of cancer-related deaths worldwide, and related morbidity and mortality continue to increase in many countries [1]. At present, the treatment of HCC shows the characteristics of multi-disciplinary participation and multi-methods. However, even after active treatment, the postoperative recurrence rate is still high. The microvascular invasion (MVI) has been proved to be a high risk factor for recurrence and poor prognosis after treatment, and is considered to be an important pathological mechanism [2]. The diagnosis of MVI depends on the pathological examination of postoperative specimens, which is delayed to a certain extent, so the prediction before operation is very important.

Contrast-enhanced ultrasound (CEUS) can sensitively and visually observe the lesions in real time, and can provide key features such as tumor enhancement, clearance and hemodynamic changes. Compared with contrast enhanced computed tomography (CT) or magnetic resonance imaging (MRI), the side effect of ultrasound contrast agent is lower, and the scope of use not limited by liver and kidney function. Therefore, CEUS has higher diagnosis rate and lower risk in liver diagnosis [3]. CEUS has been recognized as a diagnostic tool for space-occupying lesions of the liver, and may have a wider range of applications in clinical diagnosis [4].

The CEUS combined with MFI includes CEUS and MFI based on CEUS (CEUS-MFI). The CEUS-MFI can observe the flow of contrast medium microbubbles in the blood vessels of the lesions in real time and continuously, and can clearly show the smaller blood vessels and their course in the lesion. At present, there are few reports on the application of CEUS-MFI to predict the MVI in HCC. The purpose of our study is to retrospectively analyze the correlation between CEUS-MFI and MVI in HCC, and to evaluate the value of CEUS combined with MFI in predicting MVI in HCC.

2 Materials and methods

2.1 Study population

The clinical and ultrasonographic data of patients with HCC confirmed by pathology who underwent liver tumor resection in our hospital from January 2019 to May 2022 were analyzed retrospectively. The inclusion criteria were as follows: (1) Postoperative pathology confirmed that the tumor was HCC and showed MVI-positive or MVI-negative. (2) There was no treatment before liver tumor resection, including ablation and chemoembolization. (3) The two-dimensional ultrasound, CEUS and CEUS-MFI examination were performed in our hospital within two weeks before liver tumor resection, and complete imaging data have been retained. The exclusion criteria were as follows: (1) Preoperative imaging examination showed definite portal vein tumor thrombus. (2) Loss of CEUS image data. (3) Poor image quality of CEUS. (4) Incomplete or missing clinical information. All patients signed informed consent form before receiving CEUS examination. Finally, 80 patients were included, and the general clinical data and ultrasonic imaging data of the patients were collected.

2.2 Instruments and methods

The ultrasound instrument is PhilipS-EPIQ7 (probe C5-1), equipped with MFI software and set the Mechanical index (MI) to 0.06 for CEUS inspection. The contrast medium was SonoVue (Milan, Italy) sulfur hexafluoride microbubble contrast medium.

The target lesion was identified by two-dimensional ultrasound. For multiple lesions, the largest diameter was selected as the target lesion. The section with the largest diameter of the target lesion was selected as the observation section, and then CEUS mode and CEUS-MFI mode were activated. The ultrasound display adopts the dual format display mode that displays both CEUS and CEUS-MFI. The ultrasound contrast agent was added into 5.0 ml normal saline to make microbubble suspension, which was injected into the median cubital vein, and then injected into the normal saline 5 ml flush tube. The injection starts timing at the same time, and full video recording of ultrasound images, not less than 120 s.

The contents observed by contrast-enhanced ultrasound and CEUS-MFI for target lesions are as follows: (a) Enhancement range extension (yes/no), defined as an increase in the maximum diameter of the tumor in the enhancement range compared with that measured by two-dimensional ultrasound. (b) Enhancement mode in arterial phase (uniform/non-uniform). (c) Tumor shape in arterial phase (regular/irregular). (d) Intralesional arterial vessels (presence/absence). (e) Degree of wash out(no-mild washout/significant washout), significant regression is defined as complete removal of Tumor with hypoechoic within 2 minutes after injection of contrast medium; mild regression is defined as the degree of enhancement of the lesion is lower than that of the hepatic parenchyma but not completely removed (there is still some enhancement in the lesion). (f) Start time of wash out. (g) CEUS-MFI (I Type/ II Type/ III Type), Type I is defined as only punctate blood flow signal in the focus, Type II defined as small branched blood flow signal in the focus, Type III defined as messy and thick blood flow signal in the focus.

2.3 Statistical methods

SPSS26.0 is used to analyze the statistical data in our study. Univariate analysis: measurement data (normal distribution) were analyzed by independent sample t-test; measurement data (non-normal distribution) were analyzed by Mann Whitney U test; counting data and classification data were analyzed by variance test or Fisher test. Taking MVI as the dependent variable, the statistically significant variables in univariate analysis were included in binary Logistic regression analysis, and the independent risk factors with positive MVI were screened. A diagnostic model for predicting MVI positive with single and combined independent risk factors was established, receive operating characteristic (ROC) curve was made and the area under the curve (AUC) was calculated. P < 0.05 indicates that the difference is statistically significant.

3 Results

3.1 General clinical data and pathological features

80 patients with HCC were included in our study. According to the postoperative pathological results, these patients were divided into two groups: MVI-positive group 39 cases (48.8%) and MVI-negative group 41 cases (51.2%). In terms of sex, age, and presence of hepatitis, there was no significant difference between MVI positive group and MVI negative group (P > 0.05), as shown in Table 1.

Table 1
Univariate analysis of clinical and ultrasonic characteristics between MVI positive group and negative group

MVI (–) (n = 41) MVI (+) (n = 39) z/c² P

Age 58.65±12.45 55.08±10.38 1.394 0.167

Gender 2.941 0.086

Male 28 33

Female 13 6

Hepatitis 2.018 0.155

Yes 28 32

No 13 7

Number of nodules 19.550 <0.001

One 26 13

Two 9 2

Multiple 6 24

Enhancement range extension 18.420 <0.001

No 34 14

Yes 7 25

Enhancement mode at arteria phase 4.334 0.037

Non-uniform 25 32

Uniform 16 7

Tumor shape in arterial phase 14.410 <0.001

Irregular 11 27

Regular 30 12

Intralesional arterial vessels 10.000 0.002

No 32 17

Yes 9 22

Time to washout 102.0(52.5,140.0) 44.0(38.0,60.0) 4.094 <0.001

Enhancement intensity at portal phase 5.894 0.028

Hypo-enhancement 16 23

Iso-enhancement 25 14

Hyper-enhancement 0 2

Degree of wash out 0.404 0.525

No-mild wash out 27 23

Significant wash out 14 16

CEUS-MFI 19.214 <0.001

Type I 18 5

Type II 18 12

Type III 5 22

	MVI (–) (n = 41)	MVI (+) (n = 39)	z/c²	P
Age	58.65±12.45	55.08±10.38	1.394	0.167
Gender			2.941	0.086
Male	28	33
Female	13	6
Hepatitis			2.018	0.155
Yes	28	32
No	13	7
Number of nodules			19.550	<0.001
One	26	13
Two	9	2
Multiple	6	24
Enhancement range extension			18.420	<0.001
No	34	14
Yes	7	25
Enhancement mode at arteria phase			4.334	0.037
Non-uniform	25	32
Uniform	16	7
Tumor shape in arterial phase			14.410	<0.001
Irregular	11	27
Regular	30	12
Intralesional arterial vessels			10.000	0.002
No	32	17
Yes	9	22
Time to washout	102.0(52.5,140.0)	44.0(38.0,60.0)	4.094	<0.001
Enhancement intensity at portal phase			5.894	0.028
Hypo-enhancement	16	23
Iso-enhancement	25	14
Hyper-enhancement	0	2
Degree of wash out			0.404	0.525
No-mild wash out	27	23
Significant wash out	14	16
CEUS-MFI			19.214	<0.001
Type I	18	5
Type II	18	12
Type III	5	22

3.2 Univariate analysis of ultrasonic characteristics between the two groups

In term of portal clearance, there was no significant difference between MVI-positive group and MVI-negative group (P > 0.05). In term of lesion enhancement expanded, enhancement of size, shape, intralesional arterial vessels, start time of wash out, degree of wash out and CEUS-MFI, there were significant differences between MVI-positive group and MVI-negative group (P < 0.05), as shown in Table 1.

3.3 Screening of independent risk factors for MVI-positive by multivariate analysis

The statistical differences in the results of univariate analysis were included in the binary Logistic regression analysis, and the results were as follows: enhancement range extension, start time of wash out, and CEUS-MFI were independent risk factors for the occurrence of MVI positive, as shown in Table 2.

Table 2
Multivariate Logistic regression analysis of MVI positive incidence

β SE OR 95% CI P

Number of tumors 0.705 0.473 2.023 0.801–5.113 0.136

Enhancement range extension 1.975 0.957 7.204 1.105–46.980 0.039

Tumor shape in arterial phase –1.463 0.977 0.232 0.034–1.572 0.134

Enhancement mode in arterial phase 0.632 1.257 1.881 0.160–22.11 0.615

Intralesional arterial vessels –0.291 0.919 0.748 0.123–4.526 0.752

Start time of wash out –0.031 0.011 0.970 0.948–0.992 0.007

Enhancement intensity at portal phase 0.260 0.915 1.296 0.216–7.795 0.777

CEUS-MFI 1.580 0.645 4.855 1.370–17.20 0.014

	β	SE	OR	95% CI	P
Number of tumors	0.705	0.473	2.023	0.801–5.113	0.136
Enhancement range extension	1.975	0.957	7.204	1.105–46.980	0.039
Tumor shape in arterial phase	–1.463	0.977	0.232	0.034–1.572	0.134
Enhancement mode in arterial phase	0.632	1.257	1.881	0.160–22.11	0.615
Intralesional arterial vessels	–0.291	0.919	0.748	0.123–4.526	0.752
Start time of wash out	–0.031	0.011	0.970	0.948–0.992	0.007
Enhancement intensity at portal phase	0.260	0.915	1.296	0.216–7.795	0.777
CEUS-MFI	1.580	0.645	4.855	1.370–17.20	0.014

Table 3

Single factor and combination to predict the positive efficacy of MVI

	AUC	P	95% CI	Truncation value	Accuracy	Sensitivity	Specificity degree
Enhancement range extension	0.735	<0.001	0.622–0.848		0.738	0.641	0.829
Start time of wash out	0.765	<0.001	0.660–0.871	54	0.688	0.692	0.756
CEUS-MFI	0.819	<0.001	0.724–0.914		0.725	0.564	0.878
Combination of the three	0.942	<0.001	0.892–0.991		0.838	0.821	0.854

3.4 Predictive efficacy of single and combined independent risk factors for MVI-positivity

ROC curve analysis results show that, the AUC values of single independent risk factors (enhancement range extension, start time of wash out, and CEUS-MFI) predicting MVI positive occurrence were respectively 0.735 (P < 0.001), 0.765 (P < 0.001), and 0.819 (P < 0.001), and the AUC value of MVI-positive predicted by the combination of three independent risk factors was 0.942 (P < 0.001). The result show that the combination of these three independent risk factors can effectively predict MVI-positive.

Fig. 1

A, CEUS-MFI showed several punctated blood flow signals in the tumor. B, CEUS-MFI showed several thin line blood flow signals in the tumor. C, CEUS-MFI showed that there were coarse and irregular blood flow signals in the tumor. D, The pathological map of MVI-positive of HCC: a large number of cancer cells can be seen in the tumor, and clusters of cancer cells can be seen in the microvessels adjacent to the tumor, which is MVI-positive (HE×200).

Fig. 2

Single independent risk factor and joint prediction of ROC curve of MVI.

4 Discussion

The MVI means that clusters of cancer cells can be seen in the vascular lumen lined with endothelial cells under the microscope [5]. HCC is a blood-rich tumor with a dual blood supply system, and the positive rate of MVI is 34.6–70.4% [6]. MVI is considered to be an important pathological mechanism of the increased malignant degree of HCC and cancer cells destroying the surrounding normal liver tissue, and to a large extent leading to intrahepatic metastasis and recurrence after hepatectomy [7]. At present, the diagnosis of MVI depends on the pathological examination of the specimens after hepatectomy, which has a serious lag in the formulation of patients’ treatment plans. Therefore, it is of great significance to predict the MVI of HCC before operation.

Angiogenesis is an important part of the tumor microenvironment, which is very important for tumor growth. Morphological changes can reflect different stages of tumor development [8]. The formation and development of liver cancer are related to tumor neovascularization. The guidelines CEUS perfusion could be an important diagnostic tool for the detection of tumor lesions by changes of dynamic microvascularization [9]. CEUS-MFI is a new microvascular imaging technique, which is the superposition of MFI and CEUS. It can not only display the perfusion of contrast medium in the focus in real-time but also show the morphological characteristics such as the number and distribution of microvessels in the tumor according to the formation and flow of contrast medium microbubbles. CEUS-MFI has the characteristics of high spatial resolution, polar motion artifacts, and high frame rate [10]. Recently, some scholars have found that this technique is better than MFI in displaying microvessels in liver tumors, and reduces the influence of tumor size, location, and depth [11]. In our study, 81.5% of the tumors in the MVI-positive group showed messy and thick blood flow signal on CEUS-MFI. The statistical results showed that the vascular morphology shown by CEUS-MFI was closely related to MVI and was an independent risk factor for MVI positive. This is consistent with the results of Han H et al. [10] Our study thinks that the reason for the above manifestations may be due to the high malignant degree and invasive ability of cancer cells in this kind of HCC tumors, invading normal tissues and blood vessels around the tumor focus, and leading to intrahepatic metastasis, vascular tumor thrombus, hemodynamic changes in and around the focus, and compensatory increase of neovascularization. CEUS-MFI is a new ultrasonic imaging technology, which can more accurately display the blood flow inside the tumor, and to a certain extent, can improve the diagnostic efficiency of tumor and the analysis efficiency of tumor biological behavior.

CEUS can real-time and dynamically display the whole process of local liver lesions and hepatic parenchyma microcirculation perfusion changes, which can improve the ability of diagnosis and differential diagnosis, and indirectly reflect the biological characteristics of tumors, such as vascular invasion, tissue differentiation, and so on [12]. In our study, the incidence of MVI-positive was higher in tumors with more than 3 lesions, and it was found that the number of lesions was an independent risk factor for MVI-positive. This is consistent with the relevant research results [13]. Regarding the enlargement range extension HeY et al. found that the enlargement of tumor size was closely related to the biological behavior of HCC, such as MVI, Ki-67, and tumor tissue differentiation [14]. The results of our study confirmed found that it was an independent risk factor for the occurrence of MVI-positive. We believe that the cancer cells in the MVI-positive tumors are easy to invade the surrounding liver tissue and blood vessels, but at the initial stage of invasion, the surrounding liver parenchyma is not completely invaded, so it is difficult to detect this change by two-dimensional ultrasound, but by CEUS to observe the perfusion changes of microcirculation around the focus, we can find that the liver parenchyma and blood vessels around the focus have been invaded. As a result, the scope of the focus was enlarged before and after CEUS.

In the CEUS, the lesions of HCC usually showed hyper-enhancement at arterial phase, followed by gradual washing out of the portal vein and late stage. B.Schellhaas et al believe that the hyper-enhancement at arterial phase is a key feature in the diagnosis of HCC [15]. The main patterns of enhancement of HCC were entirety heterogeneous enhancement and entirety homogeneous enhancement [16]. However, hepatocellular adenomas (HCA) usually also show Hyper-enhancement at the arterial phase. The guidelines point out that the arterial phase of HCA is mainly rapid, complete, Uniform, peripherally dominated filling, and the portal phase and delayed phase show iso-enhancement [17]. This feature can be seen to distinguish HCA from HCC. In our study, all lesions showed hyper-enhancement at arterial phase, and there was no correlation with MVI (P < 0.05), which was consistent with the results of relevant scholars [18, 19].

In CEUS, Xiachuan Q et al. found that the time of washing out was closely related to the degree of differentiation of lesions, and found that the earlier the clearance time, the lower the degree of differentiation of lesions [20]. In our study, we found that MVI-positive lesions began to wash out earlier, after univariate and multivariate analysis, it was found that it was an independent risk factor for predicting MVI-positive. This is consistent with the results of zhu w et al and zhou H et al. [21 –23]. We believe that the cancer cells in MVI-positive lesions are likely to be poorly differentiated and highly invasive, and their biological behavior is characterized by destroying normal liver tissue and forming tumor thrombus in blood vessels, resulting in more neovascularization and abnormal blood supply in the tumor, resulting in hemodynamic changes in the whole focus, such as the formation of arteriovenous fistula, which is characterized by shortened enhancement time in CEUS. The washing out appeared earlier. At the same time, our study also found that the lesions showed uneven non-uniform enhancement mode in arterial phase, intralesional arterial vessels could be seen in the arterial phase, and when the tumors in the portal phase showed low enhancement, the frequency of MVI-positive was high, but by multivariate analysis, it was not found that the above indexes were powerful signs for predicting the occurrence of MVI-positive. We believe that this is due to the difference in results caused by insufficient sample size and different grouping of research methods and indicators.

In our study, we found that the efficacy of the combination of three independent risk factors (enhancement range extension, start time of wash out, and CEUS-MFI) in predicting MVI was significantly higher than that of single. Therefore, our study confirmed the feasibility of the new CEUS based on MFI to predict the MVI of HCC and provided a basis for clinicians to make personalized treatment plans. Our study still has some limitations: First our study is a retrospective study from a single institution, with small sample size. Second, in our study, some of the contrast-enhanced ultrasound indexes are qualitative indexes, which may not fully reflect the perfusion characteristics of contrast media in the lesions. Third, our study still examines the operating skills of ultrasound doctors, and due to individual differences in patients, such as obesity, there may be differences between the examination results and the actual results. Fourth, our study did not exclude lesions smaller than 10 mm from the study sequence, which may affect the final study results [23].

5 Conclusion

In the CEUS based on MFI, there is a close correlation between the enhancement range extension, start time of wash out and CEUS-MFI and MVI of HCC. In addition, the perfusion process and the morphological characteristics of blood flow in HCC can be displayed at the same time, so as to provide more ultrasonic basis for clinicians to make personalized treatment plan.

Conflict of interest

The authors have no conflict of interest to report.

References

Yang

, Hainaut

, Gores

, et al. A global view of hepatocellular carcinoma: Trends, risk, prevention and management[J]. Nat Rev Gastroenterol Hepatol. 2019;16(10):589–604.

Wang

, Wu

, Cong

. Microvascular invasion predicts a poor prognosis of solitary hepatocellular carcinoma up to 2cm based on propensity score matching analysis[J]. Hepatol Res. 2019;49(3):344–54.

Jung

, Moran

, Engel

, et al. Modified contrast-enhanced ultrasonography with the new high-resolution examination technique of high frame rate contrast-enhanced ultrasound (HiFR-CEUS) for characterization of liver lesions: First results[J]. Clin Hemorheol Microcirc. 2023;83(1):31–46.

Dong

, Koch

JBH

, Löwe

, et al. VueBox^® for quantitative analysis of contrast-enhanced ultrasound in liver tumors1[J]. Clin Hemorheol Microcirc. 2022;80(4):473–86.

Bureau of Medical Administration, National Health Commission of the People’s Republic of China. Standardization for diagnosis and treatment of hepatocellular carcinoma (2022 edition) [J]. Zhonghua Gan Zang Bing Za Zhi. 1970;4:367–88.

, Zou

, Sun

, et al. Research progress of microvascular invasion in hepatocellular carcinoma[J]. Zhonghua Gan Zang Bing Za Zhi. 2022;30(8):899–904.

Cong

, Wu

. Make great efforts to improve the levels of refined diagnosis and individualized treatment for microvascular invasion of liver cancer in China[J]. Chinese Journal of Hepatobiliary Surgery. 2019(10):721–4.

Yang

, Liu

, Lu

, et al. Evaluation of the vascular architecture of hepatocellular carcinoma by micro flow imaging: Pathologic correlation[J]. J Ultrasound Med. 2007;26(4):461–7.

Jung

, Wertheimer

, Putz

, et al. Contrast enhanced ultrasound (CEUS) with parametric imaging and time intensity curve analysis (TIC) for evaluation of the success of prostate arterial embolization (PAE) in cases of prostate hyperplasia[J]. Clin Hemorheol Microcirc. 2020;76(2):143–53.

10.

Kuang

, Wang

, Xiang

, et al. Application of constrast-enhanced ultrasound-micro flow imaging technique in differential diagnosis of benign and malignant renal tumors[J]. Chinese Journal of Ultrasonography. 2022;31(8):665–70.

11.

Han

, Ji

, Huang

, et al. The Preliminary Application of Simultaneous Display of CEUS and MFI in the Diagnosis of Hepatic Tumors[J]. J Ultrasound Med. 2023;42(3):729–37.

12.

, Liu

, Mou

, et al. Prognostic analysis of hepatocellular carcinoma on the background of liver cirrhosis via contrast-enhanced ultrasound and pathology[J]. Oncol Lett. 2018;3:3746–52.

13.

Zhang

, Wang

, Wei

, et al. An Eastern Hepatobiliary Surgery Hospital Microvascular Invasion Scoring System in Predicting Prognosis of Patients with Hepatocellular Carcinoma and Microvascular Invasion After R0 Liver Resection: A Large-Scale, Multicenter Study[J]. Oncologist. 2019;12:e1476–e1488.

14.

, Liu

, Mou

, et al. Prognostic analysis of hepatocellular carcinoma on the background of liver cirrhosis via contrast-enhanced ultrasound and pathology[J]. Oncol Lett. 2018;3:3746–52.

15.

Schellhaas

, Wildner

, Pfeifer

, et al. LI-RADS-CEUS-Proposal for a Contrast-Enhanced Ultrasound Algorithm for the Diagnosis of Hepatocellular Carcinoma in High-Risk Populations[J]. Ultraschall Med. 2016;37(6):627–34.

16.

Zhao

, Ji

, Chen

, et al. Contrast-enhanced ultrasound features of hepatic sarcomatoid carcinoma different from hepatocellular carcinoma[J]. Clin Hemorheol Microcirc. 2023 Dec 27.

17.

Dietrich

, Nolsøe

, Barr

, et al. Guidelines and Good Clinical Practice Recommendations for Contrast-Enhanced Ultrasound (CEUS) in the Liver-Update 2020 WFUMB in Cooperation with EFSUMB, AFSUMB, AIUM, and FLAUS[J]. Ultrasound Med Biol. 2020;46(10):2579–604.

18.

Fan

, Ding

, Mao

, et al. Enhancement patterns of small hepatocellular carcinoma (≤30 mm) on contrast-enhanced ultrasound: Correlation with clinicopathologic characteristics[J]. Eur J Radiol. 2020;132:109341.

19.

Chen

, Lu

, Zhu

, et al. Prediction of Microvascular Invasion in Combined Hepatocellular-Cholangiocarcinoma Based on Pre-operative Clinical Data and Contrast-Enhanced Ultrasound Characteristics[J]. Ultrasound Med Biol. 2022;48(7):1190–201.

20.

Xiachuan

, Xiang

, Xuebing

, et al. Predictive Value of Contrast-enhanced Ultrasound for Early Recurrence of Single Lesion Hepatocellular Carcinoma After Curative Resection[J]. Ultrason Imaging. 2019;1:49–58.

21.

Zhu

, Qing

, Yan

, et al. Can the Contrast-Enhanced Ultrasound Washout Rate Be Used to Predict Microvascular Invasion in Hepatocellular Carcinoma?[J]. Ultrasound Med Biol. 2017;43(8):1571–80.

22.

Zhou

, Sun

, Jiang

, et al. A Nomogram Based on Combining Clinical Features and Contrast Enhanced Ultrasound LI-RADS Improves Prediction of Microvascular Invasion in Hepatocellular Carcinoma[J]. Front Oncol. 2021;11:699290.

23.

Dietrich

, Potthoff

, Helmberger

, et al. Contrast-enhanced ultrasound: Liver Imaging Reporting and Data System (CEUS LI-RADS)[J]. Z Gastroenterol. 2018;56(5):499–506.