Liver stiffness measured with two-dimensional shear wave elastography comparable to histopathology falls dominantly on the severe liver fibrosis

Abstract

BACKGROUND:

Two-dimensional shear-wave elastography (2D-SWE) has been used for years for liver assessment of patients with chronic hepatitis B (CHB), but its effectiveness remains unclear in different populations and using different ultrasound systems.

OBJECTIVE:

This study investigated the effectiveness of 2D-SWE in evaluating liver fibrosis in patients with CHB.

METHODS:

A prospective investigation was conducted after approval by the institutional ethics committee, with 116 out of 133 patients with CHB referred for liver biopsy included and 50 patients with healthy livers selected as controls. Assessment with 2D-SWE of liver stiffness measurement (LSM) was compared with histopathological results. Cutoff values for LSM were set to determine the degree of fibrosis, and area under the receiver operating characteristic (AUROC) curve, sensitivity, and specificity were calculated.

RESULTS:

The optimal LSM cutoff for differentiating healthy livers from livers with CHB and any liver fibrosis was 6.485 kPa, with an AUROC of 0.927, sensitivity of 94%, and specificity of 19.8%. The optimal LSM cutoff values for F1, F2, F3, and F4 were 6.19 kPa, 6.485 kPa, 7.46 kPa, and 9.62 kPa, respectively, with corresponding AUROCs of 0.516, 0.625, 0.779, and 0.881, respectively. Comparisons of AUROCs between F1 and F3, F1 and F4, F2 and F3, and F2 and F4 were all significantly different (P = 0.0001, P < 0.0001, P = 0.0139, and P = 0.0003, respectively); comparisons of AUROCs between F1 and F2 and between F3 and F4 were not significantly different (P = 0.1232 and P = 0.2462, respectively). Comparisons of LSMs between healthy livers and F0 and between healthy livers and a combination of F0 and F1 were significantly different (P = 0.002 and P = 0.001, respectively). Comparisons of LSMs between F1 and F2 and between F3 and F4 were not significantly different (P = 0.233 and P = 0.072, respectively). Other comparisons between fibrosis score groups were significantly different (F1 and F3, P = 0.003; F1 and F4, P = 0.007; F2 and F3, P = 0.013; F2 and F4, P = 0.015).

CONCLUSION:

2D-SWE using a specific diagnostic ultrasound system is effective for the assessment of severe liver fibrosis and cirrhosis, but is limited in diagnosing mild liver fibrosis.

Keywords

Chronic hepatitis B liver fibrosis two-dimensional shear wave elastography (2D-SWE)

1 Introduction

Chronic hepatitis B (CHB) is a common diagnosis globally, with the highest prevalence occurring in Africa and Asia [1, 2]. CHB may progress to advanced liver fibrosis, cirrhosis, portal hypertension, liver failure, hepatocellular carcinoma, and death [2]. To improve the ability to slow or stop the progression of this liver disease and reduce CHB-related mortality, continued efforts are needed to enhance early identification of infected individuals through targeted screening and monitor and treat those at risk for CHB complications [2, 3]. Implementation of liver fibrosis staging can provide information on proper treatment and improve prognosis, though it is challenging. If the liver fibrosis staging is not accurate, it will lead to misdiagnosis and inadequate treatment. There are several methods used to screen and evaluate CHB, including serum biomarkers, medical imaging, and histopathological study of liver biopsy [2 –11]. While computed tomography (CT) for the evaluation of intermediate stages of liver fibrosis (F1–F3) shows poor diagnostic performance, contrast-enhanced CT combined with liver surface nodularity quantification can improve the accuracy of liver fibrosis staging, though due to radiation exposure and the side effects of contrast agents, it is not used routinely or repeatedly for the majority of patients [10]. Diffusion-weighted magnetic resonance imaging, contrast-enhanced magnetic resonance imaging, and magnetic resonance elastography have some advantages for the detection and staging of liver fibrosis in CHB, but they are time-consuming, expensive, and limited by contraindications; these techniques are thus only employed for some patients [11]. Ultrasonography (US) has been used routinely for decades for the surveillance of CHB due to its noninvasiveness and convenience, but it is limited in evaluating early-stage liver fibrosis [6 –8]. Tissue sampling study has been the most widely accepted method for the diagnosis and monitoring of CHB and is still considered the golden standard. However, performing a liver biopsy carries the risk of complications and a considerable probability of sampling error due to the very small area of liver tissue examined. As it is an invasive approach, it is not recommended for repeated use.

Ultrasound elastography, which has been categorized into four groups as strain elastography, transient elastography, acoustic radiation force impulse (ARFI) imaging, and shear wave speed measurement and imaging using acoustic radiation force impulse excitation in the light of the differences in the calibration of physical quantity, the excitation method, and the method of displaying the calibrated quantity, and can be used for the assessment of stiffness of human soft tissues and early stage liver fibrosis [7 , 12–16]. Studies have shown that two-dimensional shear-wave elastography (2D-SWE) has better performance than transient elastography and point-shear-wave elastography [13 –16]. Previous studies have indicated that inter-operator error in SWE measurement can be mitigated by training, though other factors affect assessment accuracy, including different ultrasound systems, hepatic inflammation, patient somatometric or demographic characteristics, and patient cooperation, and reports on the impact of these factors are inconsistent across studies [17 –20]. The 2D-SWE has been widely integrated into diagnostic ultrasound systems, but sufficient validation of its use in different ultrasound systems has not been established. Therefore, further investigation is useful to gain information before this technique is extensively applied. The purpose of this study was to investigate the effectiveness of 2D-SWE in the assessment of liver fibrosis in patients with CHB using a specific diagnostic ultrasound system.

2 Patients and methods

2.1 Study population

A prospective study on the efficacy of 2D-SWE for the evaluation of liver fibrosis was conducted. Of the 133 eligible consecutive patients with CHB who were referred for ultrasound-guided liver biopsy between January 2017 and June 2021, 116 were included in the study and 17 were excluded. The diagnosis of CHB and related diseases were referral according to the American Association for the Study of Liver Diseases (AASLD) guideline for the diagnosis of liver disease and Update on prevention, diagnosis, and treatment of chronic hepatitis B: AASLD 2018 hepatitis B guidance [2]. Before the liver stiffness measurement (LSM), comprehensive clinical assessment was performed, the body mass index (BMI) was calculated according to measured body mass (kilogram, kg) and height (meter, m). An overnight fasting blood sample was drawn within 3 days of the LSM to determine the serum levels of aspartate aminotransferases, alanine transaminases, ¡ -glutamyltransferase, alkaline phosephatase, platelets, albumin, γ-globulin, serum bilirubin, prothrombin time, and so on. Before biopsy, patients were required usually to undergo CHB related laboratory tests if appliable, ultrasound examination and ultrasound elastography. Patients were excluded if they had not performed ultrasound elastography, with a history of decompensated cirrhosis, liver tumor, obesity or BMI > 26 kg/m², hepatic steatosis, diabetes or hyperlipidemia, biliary duct stones or obstruction, or other serious diseases that may affect the 2D-SWE measurement. Between July 2020 and February 2021, 239 consecutive patients with benign thyroid nodules or hyperhidrosis undergoing preoperative ultrasound examination of the liver were considered for enrollment in the control group. Those who had larger liver focal lesions (> 2 cm) or abnormal liver echogenicity, abnormal liver function tests (aspartate aminotransferases, alanine transaminases, ϒ-glutamyltransferase, alkaline phosephatase, albumin, γ-globulin, serum bilirubin, etc), HBsAg+, obesity or BMI > 26 kg/m², diabetes, hyperlipidemia, liquor or wine drinker, or heart or kidney diseases were excluded, for a final selection of 50 patients with healthy livers. Table 1 shows the demographic and clinical characteristics of the patients.

Table 1
Demographic and clinical characteristics of the patients

Characteristic Reference standard Patients with healthy liver (n = 50) Patients with chronic hepatitis B (n = 116) P value

Age (year) – 35 (27.7–44.3) 37.5 (32.0–47.0) 0.097

Gender

Male (n) – 31 82 0.282

Female (n) – 19 34

Body mass index (Kg/m²) 18.5–25 22.6 (21.2–25.1) 23.5 (21.5–25.4) 0.063

HBsAg (+/–) – – + –

Anti-HBsAg (+/–) + + – –

DNA of HBV (IU/mL) 0–1000 N/A 50 cases < 1000 –

52 cases 1000–2000

14 cases > 2000

Aspartate aminotransferase (U/L) 5–40 26 (18–34) 30.5 (24.0–40.0) <0.001

Alanine aminotransferase (U/L) 5–40 31 (20–36) 45 (28.5–59) 0.108

Γ-Glutamine aminotransferase (U/L) 7–45 31 (18–46) 35 (19–58) 0.006

Alkaline phosephatase (U/L) 35–134 89 (32–112) 74 (64.5–90) 0.061

Total bilirubin (μmol/L) 1.7–21 8.3 (3.2–11.5) 12.5 (9.2–18.6) 0.054

Direct bilirubin (μmol/L) 1.7–7.1 3.9 (2.8–4.2) 4.3 (3.1–6.8) 0.141

Indirect bilirubin (μmol/L) 1.7–17 10.6 (3.4–10.7) 9.2 (5.9–11.8) 0.173

Albumin (g/L) 34–50 45 (39.1–51.2) 40.4 (38.8–43.5) 0.057

γ-Globulin (g/L) 20–40 26 (21.5–30.6) 30.4 (27.5–33.1) 0.062

Platelet count (10⁹/L) 100–300 203 (138–234) 228 (186–280) 0.064

Prothrombin time (s) 9.8–13.2 11.3 (10.8–11.6) 11.5 (10.7–12.5) 0.013

Characteristic	Reference standard	Patients with healthy liver (n = 50)	Patients with chronic hepatitis B (n = 116)	P value
Age (year)	–	35 (27.7–44.3)	37.5 (32.0–47.0)	0.097
Gender
Male (n)	–	31	82	0.282
Female (n)	–	19	34
Body mass index (Kg/m²)	18.5–25	22.6 (21.2–25.1)	23.5 (21.5–25.4)	0.063
HBsAg (+/–)	–	–	+	–
Anti-HBsAg (+/–)	+	+	–	–
DNA of HBV (IU/mL)	0–1000	N/A	50 cases < 1000	–
			52 cases 1000–2000
			14 cases > 2000
Aspartate aminotransferase (U/L)	5–40	26 (18–34)	30.5 (24.0–40.0)	<0.001
Alanine aminotransferase (U/L)	5–40	31 (20–36)	45 (28.5–59)	0.108
Γ-Glutamine aminotransferase (U/L)	7–45	31 (18–46)	35 (19–58)	0.006
Alkaline phosephatase (U/L)	35–134	89 (32–112)	74 (64.5–90)	0.061
Total bilirubin (μmol/L)	1.7–21	8.3 (3.2–11.5)	12.5 (9.2–18.6)	0.054
Direct bilirubin (μmol/L)	1.7–7.1	3.9 (2.8–4.2)	4.3 (3.1–6.8)	0.141
Indirect bilirubin (μmol/L)	1.7–17	10.6 (3.4–10.7)	9.2 (5.9–11.8)	0.173
Albumin (g/L)	34–50	45 (39.1–51.2)	40.4 (38.8–43.5)	0.057
γ-Globulin (g/L)	20–40	26 (21.5–30.6)	30.4 (27.5–33.1)	0.062
Platelet count (10⁹/L)	100–300	203 (138–234)	228 (186–280)	0.064
Prothrombin time (s)	9.8–13.2	11.3 (10.8–11.6)	11.5 (10.7–12.5)	0.013

Note: Data are medians, data in parentheses are interquartile range (25th–75th percentile).

2.2 Image acquisition

A Mindray Resona 7 ultrasound system (Shenzhen Mindray Bio-Medical Electronics Co., Ltd., Shenzhen, China) equipped with a curvilinear transducer with 5–1 MHz (SC5-1U) range was used to acquire US images, color Doppler flow imaging, 2D-SWE images, and Young’s modulus from SWE. After the examination of the liver parenchyma, vascularity, and biliary ducts using gray-scale US and color Doppler flow imaging, 2D-SWE was performed. All patients were instructed to fast for at least 8 hours before the 2D-SWE assessment. The measurement of Young’s modulus was performed with the patient in the supine position with the right arm in maximal abduction. Patients were instructed to suspend respiration during LSM, avoiding deep inspiration or expiration. Samples of LSM were made at segments 5 and 8 of the right lobe of the liver, with a depth of 1–4 cm beneath the liver capsule. The acoustic beam was directed vertically to the liver parenchyma through the right intercostal space to exclude large vessels and artifacts. Gel was applied to the skin and the transducer was placed on the gel without compression. The examination mode was adjusted to sound touch elastography and the color-coded map of the liver on the monitor screen was set to a fixed measurement dimension of 4 cm×3 cm. When the quality-control “M-STB index” stars in the upper right portion of the monitor screen all turned green denoting technical competence, the screen froze, a round region of interest (ROI) of 2 cm in diameter was placed within the study box over a homogeneous region of the color-coded map, and the data were obtained (Figs. 1 –3). Up to five measurements of Young’s modulus in different ROIs were obtained for each patient, and the median value was calculated for statistical analysis [8]. Each measurement was performed during a separate breath suspension. US, color Doppler flow imaging, and 2D-SWE images and Young’s modulus of the liver parenchyma, liver veins and artery, portal vein, biliary systems were saved in the Picture Archiving and Communication Systems (PACS). A sonologist (physician) with 21 years of US experience, 7 years of ultrasonic elastography experience, and 2 years of 2D-SWE experience performed US and 2D-SWE.

Fig. 1

Two-dimensional shear wave elastography (2D-SWE) measurement of the right hepatic lobe in a healthy woman with normal parameters of biomedical tests for the liver, kidney and blood. The measurement is performed inside the 2-cm-diameter circle positioned in an area of homogeneous color. The mean, minimum, maximum and standard deviation are displayed. The mean liver stiffness was determined as 4.76±0.44 kPa.

Fig. 2

2D-SWE measurement of the right hepatic lobe in a patient with CHB and biopsy confirmed fibrosis stage F3. The measurement is performed inside the 2-cm-diameter circle positioned in an area of homogeneous color. The mean, minimum, maximum and standard deviation are displayed. The mean liver stiffness was determined as 9.21±1.48 kPa.

Fig. 3

2D-SWE measurement of the right hepatic lobe in a patient with CHB and biopsy confirmed fibrosis stage F2. The measurement is performed inside the 2 cm-diameter circle positioned in an area of homogeneous color. The mean, minimum, maximum and standard deviation are displayed. The mean liver stiffness was determined as 8.54±1.19 kPa.

2.3 Histopathological assessment

Within three days following 2D-SWE assessment, a percutaneous liver biopsy was performed under US-guidance, using a 16-gauge Magnum needle (MG1522/BARD Magnum, C.R. Bard Inc., Murray Hill, NJ, USA). The biopsy sites were in the right lobe of the liver, and in a few cases performed extra sample in other places. The formalin was used to fix the biopsy specimens (15–22 mm long), sections were performed from paraffin embedded specimens, stained with hematoxylin-eosin, and then counterstained with stain reticulin or Masson. Two experienced liver pathologists who were blinded to the results of 2D-SWE, but not to the clinical and laboratory tests of the patient analyzed the specimens. The METAVIR scoring system was used for the analysis of liver fibrosis, as follows: score of F0, no fibrosis; F1, portal fibrosis without septa; F2, portal fibrosis and few septa; F3, numerous septa without cirrhosis; and F4, cirrhosis [22]. Substantial fibrosis and severe fibrosis were defined as a score of F2 and F3, respectively; and hepatic steatosis was defined as lipid droplets present in at least 30%of hepatocytes [22]. Histological inflammatory activity (A) of the liver was graded as four grades: A0 indicates no activity, A1 indicates mild activity, A2 indicates moderate activity, and A3 indicates severe activity.

2.4 Statistical analysis

Continuous data were expressed as means and standard deviations for normally distributed variables and as counts and medians (interquartile range, IQR; 25th–75th percentile) for non-normally distributed variables. Categorical measurements were summarized as counts. Independent samples t-test and Wilcoxon rank-sum test were used for comparisons between LSMs of patients with healthy livers and patients with liver fibrosis, LSMs of patients with different liver fibrosis scores, as well as other variables, with reference to normally or non-normally distributed variables. Intraclass correlation coefficients (ICCs) were used to assess the absolute agreement of mean LSMs using two consecutive examination results. Commonly established ICC cut-off values were used; values between 0 and less than 0.40 were considered poor, values between 0.40 and less than 0.60 were considered fair, values between 0.60 and less than 0.75 were considered good, and values between 0.75 and 1.0 were considered excellent [23]. Calculation of 95%confidence intervals (CIs) was performed as appropriate. Receiver operating characteristic curves for LSMs were built, and area under the receiver operating characteristics (AUROC) curve and the 95%CIs of the AUROC values were calculated to determine the accuracy of LSM. An optimal cut-off value was determined for LSM to determine the degree of fibrosis using the Youden index (the maximum combined values of sensitivity and specificity-1). Correlation coefficients between fibrosis score, LSM, and histological inflammatory activity were determined by nonparametric Spearman’s rho test. Correlation coefficients between 0–0.25 means none or slight correlation, 0.25–0.50 fair to moderate; 0.50–0.75 moderate to good and 0.75–1 almost perfect. The statistical tests were two-sided, and a P value less than 0.05 was considered a statistically significant difference. The SPSS version 25 (IBM Corp., Armonk, NY, USA) and MedCalc 15.2.2 (MedCalc Software, Mariakerke, Belgium) was used for all statistical analyses.

3 Results

The demographic and clinical characteristics of the patients are listed in Table 1. The ICC of mean LSMs obtained in two consecutive examinations was 0.958 (95%CI: 0.950–0.965). The optimal LSM cutoff value for differentiating patients with healthy livers from patients with CHB and any liver fibrosis was 6.485 kPa. The AUROC of this value was 0.927 (95%CI: 0.890–0.964), the sensitivity was 94%, and the specificity was 19.8%. Correlation coefficients for fibrosis score and LSM, fibrosis score and histological inflammatory activity, and histological inflammatory activity and LSM were 0.526, 0.537, and 0.427, respectively (P < 0.001 for all). The optimal LSM cutoff values for the evaluation of F1, F2, F3, and F4 were 6.19 kPa, 6.485 kPa, 7.46 kPa, and 9.62 kPa, respectively, with corresponding AUROCs of 0.516, 0.625, 0.799, and 0.881, respectively (Table 4). Comparisons of AUROCs between F1 and F3, F1 and F4, F2 and F3, and F2 and F4 were all significantly different (P = 0.0001, P < 0.0001, P = 0.0139, and P = 0.0003, respectively); comparisons of AUROCs between F1 and F2 and between F3 and F4 were not significantly different (P = 0.1232 and P = 0.2462, respectively). Comparisons of LSMs between healthy livers and F0 and between healthy livers and a combination of F0 and F1 were significantly different (P = 0.002 and P = 0.001, respectively). Comparisons of LSMs between F1 and F2 and between F3 and F4 were not significantly different (P = 0.233 and P = 0.072, respectively). Other comparisons between fibrosis group scores were all significantly different (F1 and F3, P = 0.003; F1 and F4, P = 0.007; F2 and F3, P = 0.013; F2 and F4, P = 0.015). The distribution of liver fibrosis score and corresponding histological inflammatory activity is listed in Table 2, and the distribution and comparison of LSMs of different patients and fibrosis scores are listed in Table 3.

Table 4
Diagnostic performance of 2D-SWE to discriminate different stages of fibrosis

Diagnostic indexes F0 F1 F2 F3 F4

Cut-off values(kPa) 5.405 6.19 6.485 7.46 9.62

Sensitivity (%) 100 87 78 88 78.6

Specificity (%) 14.4 35.7 48.8 67.4 88.8

AUC (95%CI) 0.458 0.516 0.625 0.799 0.881

(0.328–0.587) (0.417–0.615) (0.536–0.715) (0.728–0.870) (0.805–0.956)

Diagnostic indexes	F0	F1	F2	F3	F4
Cut-off values(kPa)	5.405	6.19	6.485	7.46	9.62
Sensitivity (%)	100	87	78	88	78.6
Specificity (%)	14.4	35.7	48.8	67.4	88.8
AUC (95%CI)	0.458	0.516	0.625	0.799	0.881
	(0.328–0.587)	(0.417–0.615)	(0.536–0.715)	(0.728–0.870)	(0.805–0.956)

Note: F: Fibrosis score; AUC: Aera under the receiver operating characteristic curve; 95%CI: 95%confidential interval.

Table 2

Distribution of liver fibrosis score and corresponding histological of inflammatory activity

Fibrosis score (METAVIR)	Histological inflammatory activity
	A0	A1	A2	A3
F0 (n = 13)	1	11	1	0
F1 (n = 23)	0	18	4	1
F2 (n = 42)	0	16	25	1
F3 (n = 25)	0	5	12	8
F4 (n = 13)	0	2	7	4
Total (n = 116)	1	52	49	14

Table 3

Distribution and comparisons of liver stiffness measurements among different patients and fibrosis scores

Item	Liver stiffness measurement (kPa)
Patients with healthy liver (n = 50)	5.53, (4.97–6.12) [3.87–6.91]
Patients with chronic hepatitis B (n = 116)	7.61, (6.68–9.52) [5.09–24.20]
P value	<0.001
Fibrosis score (METAVIR)
F0 (n = 13)	6.74, (5.41–9.18), [5.82–8.25]
F1 (n = 23)	6.91, (6.32–7.54), [5.55–15.45]
F2 (n = 42)	7.57, (6.58–9.49), [5.09–12.72]
F3 (n = 25)	8.90, (7.61–10.63), [6.40–12.60]
F4 (n = 13)	10.07, (9.21–14.05), [6.84–24.20]
P value	F0 vs. F1 = 0.532, F0 vs. F2 = 0.088, F0 vs. F3 < 0.001, F0 vs. F4 = 0.003, F1 vs. F2 = 0.286, F1 vs. F3 = 0.003, F1 vs. F4 = 0.005, F2 vs. F3 = 0.009, F2 vs. F4 = 0.010, F3 vs. F4 = 0.066.

Note: Data are medians, data in parentheses and brackets are interquartile range (25th–75th percentile) and minimal and maximal range, respectively.

4 Discussion

Liver fibrosis is defined as excessive extracellular matrix deposition in response to chronic injury based on complex interactions between matrix-producing hepatic stellate cells and an abundance of liver-resident and -infiltrating cells [24]. CHB-related cirrhosis is a diffuse pathological process, with characteristics of liver fibrosis and the conversion of normal liver architecture into structurally disruptive nodules [24]. Increased liver stiffness is an important feature of the pathological changes that occur in CHB-related diseases, and measurement of liver stiffness using ultrasonic elastography may reveal liver fibrosis [13, 21]. Accurate and early prediction of fibrosis stages enable the implementation of prophylactic treatment of progressive liver disease.

In our study, the ICC of mean LSMs was 0.958 (95%CI: 0.950–0.965), indicating that 2D-SWE had excellent agreement between measurements. This is consistent with the study by Petzold et al. [25]. Our study’s findings that the comparisons of LSMs between F1 and F2 and between F3 and F4 were not significantly different while comparisons of LSMs among other stages were significantly different indicate that the LSM cannot accurately differentiate between F1 and F2 or F3 and F4 fibrosis stages. Comparisons of AUROCs between F2 and F3 and between F2 and F4 were significantly different while comparisons of AUROCs between F1 and F2 and between F3 and F4 were not significantly different, indicating that the LSM can distinguish mild and substantial fibrosis from severe fibrosis and cirrhosis but has poor performance discriminating between mild and substantial fibrosis and between severe fibrosis and cirrhosis. Our study results are consistent with the study of Gao et al, which found that 2D-SWE was not effective in distinguishing F0–1 from F2 fibrosis [26].

The determination of normal liver stiffness is an important reference for the rating of abnormal liver stiffness. In our study, the median LSM of healthy livers was 5.53 kPa (range, 3.87 to 6.91 kPa), and the IQR was 4.97–6.12 kPa, which is similar to the study by Serra et al, which found that the LSM of healthy subjects was 4.65±1.15 kPa, with a range of 2.66–6.80 kPa, and that the optimal cut-off value in differentiating healthy subjects from chronic liver patients with any fibrosis was 5.47 kPa [27]. Our results also agree with Petzold et al., [25] who stated that healthy liver LSMs ranged from 3.62 to 7.02 kPa.

In our study, the optimal LSM cutoff value for differentiating patients with healthy livers from CHB patients with any liver fibrosis was 6.485 kPa, which is higher than the value of 5.47 kPa reported by Serra et al. [27]. The AUROC of our LSM cutoff value for differentiating healthy livers from CHB-related liver fibrosis was 0.927, the sensitivity was 94%, and the specificity was 19.8%. In our study, the median LSM values of patients with CHB were as follows: 6.91 kPa for F1, 7.57 kPa for F2, 8.90 kPa for F3, and 10.07 kPa for F4, which differ from the 6.7 kPa for F1, 6.33 kPa for F2, 9.2 kPa for F3, and 13.7 kPa for F4 reported by Huang et al. [28].

In our study, LSM had a positive correlation with fibrosis stage (Spearman’s correlation, 0.526; P < 0.001); the optimal cut-off values of LSM for the evaluation of F1, F2, F3, and F4 were 6.19 kPa, 6.485 kPa, 7.46 kPa, and 9.62 kPa, respectively, with corresponding AUROCs of 0.516, 0.625, 0.799, and 0.881, respectively, which differed from multiple previous studies. Huang et al. [28] reported a Spearman’s correlation for liver fibrosis stage and LSM of 0.708 and optimal cutoff values for predicting F2, F3, and F4 of 7.05 kPa, 9.45 kPa, and 11.1 kPa, respectively, with corresponding AUROCs of 0.835, 0.916, and 0.881, respectively. Serra et al. [27] found a Spearman’s correlation for liver fibrosis stage and LSM of 0.628, with AUROCs for F1, F2, F3, and F4 of 0.724, 0.857, 0.946, and 0.935, respectively. Zhuang et al. [29] reported that optimal LSM cutoff values for F2, F3, and F4 were 7.6 kPa, 9.2 kPa, and 10.4 kPa, respectively, and the corresponding AUROCs were 0.97, 0.96, and 0.98, respectively. The cutoff values obtained by Herrmann et al. [30] of 8.1 kPa for diagnosing severe fibrosis and 11.5 kPa for diagnosing cirrhosis were higher than our results. Our optimal cutoffs are similar to the study by Serra et al., [27] which found that optimal cutoffs from 2D-SWE in diagnosing F≥1, F≥2, F≥3, and F = 4 were close to each other [27]. Our AUROCs for F2, F3, and F4 were 0.625, 0.799, and 0.881, respectively, and were much lower than those found by Zeng et al., [31] with AUROCs of 0.851, 0.975, and 0.972 in the diagnosis of F2, F3, and F4, respectively, and lower than some of the previously described studies [27 –30].

We noted that although F0 indicates no presence of liver fibrosis, there were 13 patients with F0 who presented with significantly higher LSMs than those with healthy livers. We believe the reason for this finding was related to the presence of liver inflammatory activity, as 12 of them had grade 1 inflammatory activity, one had grade 2 inflammatory activity, and only one had no inflammatory activity. This may be supported by Peschel et al., who reported that the LSMs of patients with chronic HCV declined in parallel with the decrease in systemic inflammatory parameters during treatment with direct-acting antivirals [32]. The finding that 2D-SWE could not accurately distinguish between F1 and F2 or between F3 and F4 in our study may be in part due to sources of measurement variability, including breath-holding cooperation, disease heterogeneity, and body habitus [17 –20].

In this study, LSM had a moderate positive correlation with fibrosis score (0.526) and a moderate positive correlation with histological inflammatory activity (0.427), while fibrosis score had a relatively higher positive correlation with histological inflammatory activity (0.537), indicating that LSM is not a precision value, it is a range value, and the value can be affected by histological inflammatory activity and other factors [26]. Such information may support that the LSM of F0 and F1 were approximate values in this study, or incongruence values, as described by Huang [28].

Our study has limitations. First, the patients were relatively fewer for different groups of fibrosis score. However, it did not cause marked impact on the overall statistical analysis. Second, although we used standardized protocols, the patients’ cooperation was not able to control, which might cause untoward effect on the measurement.

Collectively, 2D-SWE using a specific diagnostic ultrasound system may acquire and deliver useful information for the noninvasive assessment of substantial liver fibrosis, with AUROCs of 0.799 and 0.881 for stage F3 and F4 fibrosis, respectively, but its ability to identify mild liver fibrosis remains limited. In the future, further validation will be completed based on more samples and relevant factors to implement an effective assessment of 2D-SWE for liver fibrosis.

Conflict of interest

The authors declare they have no conflict of interests.

References

WHO. Global hepatitis report, 2017 [cited 2021 June 30]. Available from: https://www.who.int/hepatitis/publications/global-hepatitis-report2017/en/

Terrault

, Lok

ASF

, McMahon

, Chang

, Hwang

, Jonas

, et al. Update on prevention, diagnosis, and treatment of chronic hepatitis B: AASLD hepatitis B guidance. Hepatology. 2018;67(4):1560–99.

Kisseleva

, Brenner

. Molecular and cellular mechanisms of liver fibrosis and its regression. Nat Rev Gastroenterol Hepatol. 2021;18(3):151–66.

Inoue

, Tanaka

. Novel biomarkers for the management of chronic hepatitis B. Clin Mol Hepatol. 2020;26(3):261–79.

Takanori

, Masatoshi

, Yoji

, Kuzuya

, Honda

, Hayashi

, et al. Utility and limitations of noninvasive fibrosis markers for predicting prognosis in biopsy-proven Japanese non-alcoholic fatty liver disease patients. J Gastroenterol Hepatol. 2019;34(1):207–14.

Colli

, Fraquelli

, Andreoletti

, Marino

, Zuccoli

, Conte

. Severe liver fibrosis or cirrhosis: accuracy of US for detection: analysis of 300 cases. Radiology. 2003;227(1):89–94.

, Tu

, Liang

. Patchy echogenicity of the liver in patients with chronic hepatitis B does not indicate poorer elasticity. Ultrasonography. 2019;38(4):327–35.

Zheng

, Guo

, Zeng

, Huang

, Zheng

, Ren

, et al. Two dimensional shear-wave elastography and conventional US: the optimal evaluation of liver fibrosis and cirrhosis. Radiology. 2015;275(1):290–300.

Pickhardt

, Malecki

, Kloke

, Lubner

. Accuracy of Liver Surface Nodularity Quantification on MDCT as a Noninvasive Biomarker for Staging Hepatic Fibrosis. AJR Am J Roentgenol. 2016;207(6):1194–99.

10.

Kromrey

, Le Bihan

, Ichikawa

, Motosugi

. Diffusion-weighted MRI-based Virtual Elastography for the Assessment of Liver Fibrosis. Radiology. 2020;295(1):127–35.

11.

Goodman

. Grading and staging systems for inflammation and fibrosis in chronic liver diseases. J Hepatol. 2007;47(4):598–607.

12.

Shiina

, Nightingale

, Palmeri

, Hall

, Bamber

, Barr

, et al. WFUMB guidelines and recommendations for clinical use of ultrasound elastography: Part basic principles and terminology. Ultrasound Med Biol. 2015;41(5):1126–47.

13.

Dietrich

, Bamber

, Berzigotti

, Bota

, Cantisani

, Castera

, et al. EFSUMB Guidelines and Recommendations on the Clinical Use of Liver Ultrasound Elastography, Update 2017 (Long Version). Ultraschall Med. 2017;38(4):e16–e47.

14.

Ren

, Xia

, Ni

, Zhan

, Zhou

. Analysis of three ultrasound elastography techniques for grading liver fibrosis in patients with chronic hepatitis B. Radiol Med. 2018;123(10):735–41.

15.

Zheng

, Xia

, Ren

, Zhan

, Zheng

, Chen

. A study of spleen shear-wave elastography in indirect prediction of liver fibrosis in patients with chronic hepatitis B. Clin Hemorheol Microcirc. 2020;76(1):63–72.

16.

Xia

, Ren

, Ni

, Zhan

. A Noninvasive Method-Shear-Wave Elastography Compared with Transient Elastography in Evaluation of Liver Fibrosis in Patients with Chronic Hepatitis B. Ultrasound Q. 2019;35(2):147–52.

17.

Gilligan

, Trout

, Bennett

, Dillman

. Repeatability and Agreement of Shear Wave Speed Measurements in Phantoms and Human Livers Across 6 Ultrasound 2-Dimensional Shear Wave Elastography Systems. Invest Radiol. 2020;55(4):191–99.

18.

Ryu

, Ahn

, Yoon

, L

. Reproducibility of liver stiffness measurements made with two different 2-dimensional shear wave elastography systems using the comb-push technique. Ultrasonography. 2019;38(3):246–54.

19.

Varbobitis

, Siakavellas

, Koutsounas

, Karagiannakis

, Ioannidou

, Papageorgiou

, et al. Reliability and applicability of two-dimensional shear-wave elastography for the evaluation of liver stiffness. Eur J Gastroenterol Hepatol. 2016;28(10):1204–09.

20.

Liao

, Kuo

, Chiang

, Lin

, et al. Acoustic radiation force impulse elastography of the liver in healthy patients: test location, reference range and influence of gender and body mass index. Ultrasound Med Biol. 2015;41(3):698–704.

21.

Barr

, Ferraioli

, Palmeri

, Goodman

, Garcia-Tsao

, Rubin

, et al. Elastography Assessment of Liver Fibrosis: Society of Radiologists in Ultrasound Consensus Conference Statement. Ultrasound Q. 2016;32(2):94–107.

22.

Bedossa

, Poynard

. An algorithm for the grading of activity in chronic hepatitis C. The METAVIR Cooperative Study Group. Hepatology. 1996;24(2):289–93.

23.

Hallgren

. Computing inter-rater reliability for observational data: an overview and tutorial. Tutor Quant Methods Psychol. 2012;8(1):23–34.

24.

Tacke

, Trautwein

. Mechanisms of liver fibrosis resolution. J Hepatol. 2015;63(4):1038–9.

25.

Petzold

, Hofer

, Ellenrieder

, Neesse

, Kunsch

. Liver Stiffness Measured by 2-Dimensional Shear Wave Elastography: Prospective Evaluation of Healthy Volunteers and Patients with Liver Cirrhosis. J Ultrasound Med. 2019;38(7):1769–77.

26.

Gao

, Zheng

, Liang

, Tong

, Wang

, Wu

, et al. Liver fibrosis with two-dimensional US shear-wave elastography in participants with chronic hepatitis B: a prospective multicenter study. Radiology. 2018;289(2):407–15.

27.

Serra

, Grasso

, Conti

, Felicani

, Mazzotta

, Lenzi

, et al. A New Two-Dimensional Shear Wave Elastography for Noninvasive Assessment of Liver Fibrosis in Healthy Subjects and in Patients with Chronic Liver Disease. Ultraschall Med. 2018;39(4):432–9.

28.

Huang

, Zheng

, Zhang

, Xu

, Chen

, et al. Assessing Hepatic Fibrosis Using 2-D Shear Wave Elastography in Patients with Liver Tumors: A Prospective Single-Center Study. Ultrasound Med Biol. 2017;43(11):2522–9.

29.

Zhuang

, Ding

, Zhang

, Sun

, Xu

, Wang

, et al. Two-dimensional Shear-Wave Elastography Performance in the Noninvasive Evaluation of Liver Fibrosis in Patients with Chronic Hepatitis B: Comparison with Serum Fibrosis Indexes. Radiology. 2017;283(3):873–82.

30.

Herrmann

, de Lédinghen

, Cassinotto

, Chu

, Leung

, Ferraioli

, et al. Assessment of biopsy-proven liver fibrosis by two-dimensional shear wave elastography: An individual patient data-based meta-analysis. Hepatology. 2018;67(1):260–72.

31.

Zeng

, Zheng

, Huang

, Chen

, Liu

, Wu

, et al. Comparison of 2-D Shear Wave Elastography and Transient Elastography for Assessing Liver Fibrosis in Chronic Hepatitis B. Ultrasound Med Biol. 2017;43(8):1563–70.

32.

Peschel

, Grimm

, Buechler

, Gunckel

, Pollinger

, Aschenbrenner

, et al. Liver stiffness assessed by shearwave elastography declines in parallel with immunoregulatory proteins in patients with chronic HCV infection during DAA therapy. Clin Hemorheol Microcirc. 2021 Jun 8. doi: 10.3233/CH-211193.