Evaluation of Liver Fibrosis Using Shear Wave Elastography: An Overview

Abstract

All kinds of chronic liver disease can progress into liver fibrosis, and the stage of liver fibrosis is an important prognostic factor. Therefore, assessment of liver fibrosis is of importance for the management of the chronic liver disease. Although liver biopsy is considered the standard method, its invasive nature limits clinical use. In this regard, shear wave-based ultrasound elastography has been emerged as a noninvasive method to evaluate liver fibrosis. Among various techniques, transient elastography (TE) has been the most extensively used and validated method. TE provides good diagnostic performance in staging liver fibrosis. In addition to TE, point shear wave elastography (pSWE) and two-dimensional SWE (2D-SWE) have been developed as another noninvasive method, and also reported good diagnostic performance in staging liver fibrosis. Although TE, pSWE, and 2D-SWE show good performance in assessing liver fibrosis, concurrent inflammatory activity and/or hepatic congestion are important limitations in the current elastography technique.

1. Introduction

Chronic liver disease is a major healthcare problem worldwide, and various etiologies including viral hepatitis caused by hepatitis B virus (HBV) or hepatitis C virus (HCV), alcohol abuse, and non-alcoholic fatty liver disease (NAFLD) can induce chronic liver disease [1]. Moreover, chronic liver disease is an evolving and dynamic process, progressing into liver fibrosis [2, 3, 4]. When appropriate management is not given, liver injury and fibrosis can continuously progress, eventually leading to the development of liver cirrhosis, portal hypertension, hepatic insufficiency as well as hepatocellular carinoma (HCC) which can increase morbidity and mortality [5, 6]. In addition, the stage of liver fibrosis is associated with the risk of HCC development and liver-related mortality. Therefore, information regarding the stage of liver fibrosis is important for both surveillance and personalized treatment [7, 8, 9]. Owing to the dynamic and evolving nature, liver fibrosis would be reversible under the adequate management, especially in early stage of the disease. In contrast, liver cirrhosis is generally considered as an irreversible process [10, 11, 12, 13]. Therefore, evaluation, as well as detection of liver fibrosis in the early stage, is of importance for the management of the chronic liver disease.

For the assessment of liver fibrosis, liver biopsy with histopathologic examination has been used as the reference standard method [14]. In addition, histopathologic examination enables the evaluation of concurrent inflammatory activity in the liver, in addition to the assessment of liver fibrosis. However, liver biopsy has several important drawbacks limiting its clinical use. First, liver biopsy is an invasive procedure that can cause potentially lethal complications, such as bleeding. Due to the invasive nature, repeated biopsy for the monitoring of liver fibrosis during the disease course in the same patient can hardly be performed in clinical practice [15]. The small sample volume of liver biopsy, generally 1/50000th of total liver parenchyma, is another important limitation. When the distribution of liver fibrosis is heterogeneous, a small volume with sampling variability of liver biopsy can lead to either overestimation or under-estimation of liver fibrosis [16, 17]. Another important limitation of liver biopsy is considerable inter-reader variability, and the reported kappa value among the different pathologists varies from 0.5 to 0.9 [18, 19]. Therefore, there has been a continuous need for a reliable and noninvasive methods for the evaluation of liver fibrosis in clinical practice, and tremendous effort has been made to develop non-invasive diagnostic methods for the assessment of liver fibrosis [13]. In this regard, shear wave based ultrasound elastography has been developed and introduced as an accurate noninvasive diagnostic method for the evaluation of liver fibrosis. After the introduction of transient elastography (TE) which was the first commercially available liver elastography technique, various ultrasound-based shear wave elastography methods including point shear wave elastography (pSWE) and two-dimensional shear wave elastography (2D-SWE) have been introduced in clinical practice and reported a good diagnostic performance in assessing liver fibrosis [20, 21].

2. Principle of shear wave elastography

Elastography is an imaging technique measuring a tissue mechanical characteristic such as elasticity, that was firstly described by Ophir et al. [22]. Tissue elasticity is defined as the resistance to the deformation of a certain tissue against applied stress [15], and stiff tissue is more resistant to the deformation than soft tissue in given applied stress. For the superficial organs such as the breast and thyroid, tissue elasticity can be measured by using strain elastography. In strain elastography, stress to tissue is directly applied by manual compression of an ultrasound transducer, and then the degree of tissue deformation after compression is measured by ultrasound imaging [22]. Manual compression works fairly well for superficial organs, and therefore, strain elastography is a useful technique for the evaluation of breast or thyroid lesion, providing information regarding tissue stiffness [23]. However, it is very challenging to induce stress to deeper located organs by manual compression such as the liver, limiting the application of strain elastography to the liver [24]. For deeper located organs such as the liver, the stress can be employed by acoustic radiation force impulse (ARFI) or mechanical push pulse to generate a shear wave within the target tissue [15]. Since shear wave propagation velocity is related to tissue elasticity and the shear wave velocity is faster in stiff tissue than in soft tissue, measurement of shear wave velocity generated by either ARFI or mechanical push pulse leads to the quantitative assessment of tissue elasticity [23]. Given that, the type of ultrasound-based shear wave elastography for the liver can be determined by following two factors: 1) how to generate shear wave within the liver tissue?; and 2) how to measure the velocity of generated shear wave within the liver tissue?. Based on these two factors, currently, there are three available ultrasound-based shear wave elastography techniques for the liver: 1) one-dimensional transient elastography (TE); 2) point shear wave elastography (pSWE), and 3) two-dimensional shear wave elastography (2D-SWE) [23]. The characteristics of these three elastography techniques are summarized in Table 1 and Figure 1.

	Excitation method	Frequency of generated shear wave	Shear wave velocity measurement direction	Measurement area	Placement of region of interest	Reported parameter
TE	Mechanical push pulse	50 Hz	Parallel to excitation	Small	Restricted, no guidance	Young modulus (kPa)
pSWE	ARFI, single focal location	Wideband (100–500 Hz)	Perpendicular to ARFI application	Small	Flexible under B-mode guidance	Young modulus (kPa) or shear wave velocity (m/s)
2D-SWE	ARFI, multiple focal zones	Wideband (100–500 Hz)	Perpendicular to ARFI application	Medium	Flexible under B-mode guidance	Young modulus (kPa) or shear wave velocity (m/s)

Table 1.

Characteristics of currently available ultrasound based shear wave elastography techniques for the liver.

Note: TE, transient elastography; pSWE, point shear wave elastography; 2D-SWE, two-dimensional shear wave elastography; ARFI, acoustic radiation force impulse; kPa, kilopascal.

Figure 1.

Currently available ultrasound-based shear wave elastography methods for the liver. (a) Transient elastography (TE). In TE, B-mode images of the liver are not provided, and thus the measurement area cannot be selected. Ten valid measurements were performed for this patient, and the IQR/M value was 6%, indicating reliable measurement result. (b) Point shear wave elastography (pSWE) (Virtual touch quantification, Siemens Acouson S2000). The measurement box is placed within liver parenchyma 2.5 cm apart from liver capsule. Since pSWE provides B-mode images of the liver simultaneously, the placement of measurement box is undertaken under the B-mode image guidance, avoiding large hepatic vessels or areas showing artifact. (c) Two-dimensional shear wave elastography (2D-SWE) (Aixplorer, Supersonic Imagine). The size of measurement box of 2D-SWE is larger than that of pSWE, and placed within liver parenchyma under the B-mode image guidance. 2D-SWE can also provide color-coded elastogram, superimposed on B-mode image of the liver.

2.1 Transient elastography

The FibroScan system (Echosens, Paris, France), which is TE system, was the first commercially available ultrasound-based shear wave elastography system for the liver [25]. The FibroScan probe contains both a mechanical vibrating device and an ultrasound transducer [23]. When the mechanical vibrating device part of FibroScan probe employs a 50 Hz mechanical impulse to the skin surface, the shear wave is generated and propagated within the liver tissue [15]. The generated shear wave within the liver tissue by mechanical push pulse applied to the skin surface is traced by an ultrasound transducer for the measurement of shear wave velocity. Then, liver stiffness can be calculated by measured shear wave velocity. The frequency of generated shear wave within liver tissue by mechanical push pulse in TE is 50 Hz. Although TE is an ultrasound-based technique, it is impossible to provide B-mode images of the liver in TE system, and therefore, TE is performed without direct B-mode image guidance [23]. The size of the measurement area of TE is approximately 1 cm width × 4 cm length, which is >100 times larger than the tissue volume assessed by a liver biopsy [26, 27]. There are several available probes for TE, and M probe with an operating center frequency of 3.5 MHz is used for the standard examination [15]. Since TE applies a mechanical push pulse to the skin surface for the generation of shear wave within the liver tissue, the presence of ascites and obesity limiting the shear wave generation by mechanical push pulse would be a drawback. In addition, M probe would have a limited ultrasound penetration for obese patients, hampering the accurate measurement of shear wave velocity. To overcome this limitation of M probe, XL probe with a lower operating frequency (2.5 MHz for XL probe vs. 3.5 MHz for M probe) enabling measurement at a greater depth (35–75 mm for XL probe vs. 25–65 mm for M probe) is introduced. Using XL probe, accurate and reliable measurement can be possible for obese patients.

2.2 Point shear wave elastography (pSWE)

In contrast to TE which uses a mechanical push pulse to generate a shear wave within the liver tissue, pSWE uses ARFI technique to induce stress and to generate a shear wave within the liver tissue. When ARFI is delivered in the liver tissue, the longitudinal waves along with the plane of applied ARFI are generated. At the same time, a portion of longitudinal waves is converted to shear waves within the liver tissue, and propagate perpendicular to the plane of longitudinal waves [28]. The frequency of generated shear wave by applied ARFI is wideband, ranging from 100 to 500 Hz. In pSWE, the velocity of the shear wave generated by ARFI is measured, which is either directly reported in meters per second or changed to Young’s modulus E in kilopascal for the estimation of tissue elasticity [27]. Under the assumption of incompressibility, shear wave velocity can be converted to Young’s modulus E by the following equation: E (kilopascal) = 3ρc², where c is the measured shear wave velocity in meter per second and ρ is the tissue density, assumed to be 1 of water [15]. Unlike TE, pSWE can be performed using a conventional ultrasound probe equipped with standard diagnostic ultrasound machine [27]. Therefore, pSWE can provide B-mode images of the liver simultaneously during the examination, enabling the selection of a uniform area of liver parenchyma without any large vessels, focal lesions, or artifacts where the shear wave velocity will be measured [23]. Given that, the accuracy and measurement reliability of pSWE are expected to be higher than those of TE. In addition, since the shear wave is generated by ARFI which is introduced inside the liver parenchyma, pSWE would be less affected by the presence of ascites and obesity than TE [9, 29, 30].

2.3 Two-dimensional shear wave elastography (2D-SWE)

2D-SWE is the newest ultrasound-based shear wave elastography technique, which also utilizes ARFI. In contrast to the pSWE which introduces ARFI in a single focal location, 2D-SWE uses multiple focused ultrasound push pulses to create multiple focal zones interrogated in rapid succession, faster than shear wave speed [23]. Those multiple push pulses in 2D-SWE generate a near cylindrical shear wave cone, allowing the real-time tracing of shear waves in 2D to measure the velocity of induced shear wave or Young’s modulus E [23, 31]. The same as the pSWE, the frequency of generated shear wave by multiple push pulses in 2D-SWE is wideband, ranging from 100 to 500 Hz. Since 2D-SWE utilizes the conventional ultrasound probe for standard diagnostic imaging, it can also provide B-mode images of the liver simultaneously, and real-time visualization of a color-coded quantitative elastogram can be superimposed on a B-mide image. This merit of 2D-SWE allows the operator to obtain both anatomical and tissue stiffness information [20]. Currently, most of the major ultrasound vendors provide their own shear wave elastography technique for the liver, either form of pSWE or 2D-SWE.

3. Measurement protocol and reliability criteria

Regarding patient preparation, ultrasound-based shear wave elastography techniques including TE, pSWE, and 2D-SWE share the same recommended protocols [6, 32, 33]. Since the amount of portal flow can affect the result of liver stiffness measurement obtained by shear wave elastography, fasting for at least 4 hours before the examination is recommended for patients who undergo shear wave elastography examination to minimize the effect of portal flow. The liver stiffness measurement using shear wave elastography is usually performed in either supine or slightly left lateral decubitus position (not more than 30 degrees) with the right arm extended above the head to obtain the optimal sonic window via the stretching of the intercostal muscles [6, 34]. It has been known that both deep inspiration and deep expiration can have an influence on the result of liver stiffness measurement using shear wave elastography, and therefore, the neutral breath-hold is recommended for shear wave elastography examination to minimize the effect of breath-hold status. In addition to the aforementioned protocols for patient preparation, current guidelines for both pSWE and 2D-SWE have several recommendations for imaging acquisitions since pSWE as well as 2D-SWE provide B-mode images of the liver simultaneously, and the measurement area of pSWE and 2D-SWE can be selected under the real-time B-mode imaging guidance [6, 32, 33]. The transducer should be placed perpendicular to the liver capsule to ensure proper generation and propagation of the shear wave. The measurement box for both pSWE and 2D-SWE is placed parallel to the liver capsule, and the upper edge of the measurement box should be placed 1.5 to 2.0 cm apart from the liver capsule to minimize the effect of reverberation artifact which is generally seen in the area adjacent to the liver capsule. In most currently available ultrasound systems, the ARFI pulse reaches the maximum intensity at 4.0 to 4.5 cm apart from the transducer and is attenuated by 6.0–7.0 cm [6]. Given that, the area located at 4.0 to 4.5 cm apart from the transducer would be the optimal location for liver stiffness measurement. Since B-mode image is utilized to trace the shear wave in both pSWE and 2D-SWE, high-quality B-mode images without artifacts should be acquired for accurate and reliable liver stiffness measurement. The recommended protocols for both patient preparation and imaging acquisition are summarized in Table 2.

	Recommendation	Aim
Patient preparation	Fasting for at least 4 hours before examination	To minimize effect of portal flow
	Position: supine or slight left lateral decubitus (not more than 30°) with right arm extended above the head	To obtain optimal sonic window via stretching of the intercostal muscles
	Neutral breath hold, neither deep inspiration nor expiration	To minimize effect of breath-hold status
Imaging acquisition for pSWE and 2D-SWE	Transducer placed perpendicular to the liver capsule	To ensure proper shear wave generation
	Upper portion of measurement box placed at least 1.5–2.0 cm apart from liver capsule	To minimize effect of reverberation artifact
	Ideal location of measurement box: 4–4.5 cm apart from the transducer	To maximize intensity of ARFI pulse

Table 2.

Recommendation for patient preparation and imaging acquisition.

Note: pSWE, point shear wave elastography; 2D-SWE, two-dimensional shear wave elastography; ARFI, acoustic radiation force impulse.

Regarding the acquisition number of liver stiffness measurements using TE, ten valid measurements are recommended. In addition, the interquartile range (IQR)-to-median ratio of ten valid measurements (subsequently referred to as IQR/M) is usually used as the quality criteria: IQR/M equal to or less than 30% indicates reliable measurement results [6, 32, 33]. According to the result of a study including 13,369 TE examinations using M probe [35], the failure rate of obtaining valid liver stiffness measurement and unreliable measurements rates was 3.1% of cases and 15.8% of cases, respectively. Regarding the contributory factors for failed and/or unreliable measurements of TE was body mass index [15, 35], and high body mass index was significantly associated with the failed and/or unreliable measurements. With the introduction of XL probe for TE examination, the reliability of liver stiffness measurements using TE has been improved, especially for NAFLD patients [36, 37, 38, 39, 40]. Regarding the measurement reproducibility of TE, excellent inter-reader agreement with the intraclass coefficient (ICC) of 0.98 was reported in a cohort of 188 patients having chronic HCV infection [41].

The recommended acquisition number of liver stiffness measurements using pSWE is also ten valid measurements. The same as the TE, the result with IQR/M equal to or less than 30% for measurement given in kilopascals is considered a reliable result. Regarding the 2D-SWE, the area for liver stiffness measurement is larger than pSWE, and thus, each liver stiffness measurement value is actually an average value of several measurements [6]. In addition, several manufacturers provide quality assessment methods for their 2D-SWE systems such as propagation map, stability index, and reliable measurement index [6]. Given that, the current guideline recommends five measurements for 2D-SWE when a quality assessment method is provided by the manufacturer. However, when a quality assessment method is not available, ten measurements for 2D-SWE are recommended, the same as the TE or pSWE [6]. IQR/M for measurement given in kilopascals is also used as the quality criteria for 2D-SWE, the same as the TE or pSWE. Result with IQR/M equal to or less than 30% of five or ten measurements given in kilopascals indicates reliable measurement results. It has been reported that when IQR/M for measurement given in kilopascals was higher than 30%, the accuracy of liver stiffness value obtained from shear wave elastography was reduced [33, 42, 43]. According to the result of a study comparing pSWE and 2D-SWE in 79 patients at the same day [44], the failure rate was 1.3% for pSWE and 5.1% for 2D-SWE, respectively. The overall intra-reader agreement was higher for pSWE than 2D-SWE (ICC of 0.915 for pSWE vs. ICC of 0.829 for 2D-SWE, P < 0.001). In addition, intra-reader reproducibility between liver stiffness measurements by using 2D-SWE performed in the same participant on different days was higher for the experienced operator than novice operator (ICC of 0.84 for experienced reader vs. 0.65 for novice reader) [45], indicating that reader experience has an influence on the measurement reliability. Ferraioli et al. also reported that the liver stiffness measurement by using pSWE was affected by operator experience [46]. Given that, operators doing pSWE and/or 2D-SWE examinations need to be properly trained and to follow the recommendations for patient preparation and imaging acquisition [15].

4. Diagnostic performance for staging liver fibrosis

Liver fibrosis is the result of chronic liver injury and is defined as an abnormal and excessive deposition of collagen and other extracellular matrix components in the liver [9, 47]. Essentially, any kind of chronic liver disease caused by HBV or HCV infection, alcohol abuse, and NAFLD lead to steatosis, inflammation with necrosis in response to an injury [9]. Without appropriate management, these liver cell injury continuously progresses, eventually developing liver cirrhosis. Information regarding the liver fibrosis stage is beneficial for the prediction of prognosis, personalized follow-up, and treatment decisions. For example, antiviral therapy for HBV or HCV infection might be guided by the information regarding the liver fibrosis stage [48, 49]. Therefore, an accurate assessment of the liver fibrosis stage is an important step for chronic liver disease management. For this purpose, liver biopsy with histopathologic examinations using various staging systems including Ishak, METAVIR, and Batts-Ludwig systems has been traditionally used as the standard reference method [18, 50]. However, liver biopsy is limited for widespread application in clinical practice, mainly due to its invasive nature. To overcome the limitation of liver biopsy, ultrasound-based shear wave elastography techniques including TE, pSWE, and 2D-SWE have been emerged as noninvasive methods for the evaluation of liver fibrosis and reported a good diagnostic performance.

4.1 Transient elastography

Since TE was the first approved and commercially available ultrasound-based elastography technique for the liver, there have been a lot of studies including meta-analyses reporting the diagnostic performance of TE in assessing liver fibrosis stage for chronic liver disease patients with various etiologies. Currently, TE is the most widely used and extensively validated elastography technique for liver stiffness measurement. Regarding the detection of advanced fibrosis and liver cirrhosis originated from HBV or HCV infection by using TE, early studies reported an excellent diagnostic performance with areas under the receiver operating characteristic curve (AUROCs) of 0.88–0.99 [51, 52, 53, 54, 55, 56, 57]. Several meta-analyses also reported the excellent diagnostic capability of TE to detect liver cirrhosis with AUROCs of 0.93–0.96, better than those for diagnosing moderate fibrosis (F2-F4) with AUROCs ranging from 0.83 to 0.88 [58, 59, 60, 61, 62]. The reported cut-off liver stiffness value was 7.0–7.9 kPa for the detection of moderate fibrosis (F2-F4) and 11.3–15.6 kPa for the diagnosis of cirrhosis (F4) [58, 59, 60, 63]. In addition to the HBV and HCV infection, TE also showed a good diagnostic performance in assessing liver fibrosis for NAFLD patients. However, the application of TE for NAFLD patients is challenging, mainly due to the high failure rate and poor measurement reliability in obese patients, especially when standard M probe is used. The reported rate of unreliable and/or failed measurement of TE for NAFLD patients ranged from 3.8% to 50.0% [38, 64, 65]. According to the result of a meta-analysis including 854 NAFLD patients with individual data, the reported pooled sensitivity and specificity of TE using the standard M probe was 79% and 75% to detect F2-F4, 85% and 82% to detect F3–4, and 92% and 92% to detect F4, respectively [66]. The AUROCs of TE ranged from 0.79–0.87 for detection of F2-F4, 0.76–0.98 for detection of F3-F4, and 0.91–0.99 for the diagnosis of F4, respectively, in NAFLD patients [15]. The introduction of XL probe for obese patients has improved the measurement reliability of TE [67].

4.2 Point shear wave elastography (pSWE) and two-dimensional shear wave elastography (2D-SWE)

Since both pSWE and 2D-SWE have become commercially available more lately than TE, the number of studies and the amount of data are less than those of TE. Thus, the level of evidence for the diagnostic performance of pSWE or 2D-SWE in assessing the liver fibrosis stage is usually lower than that for TE.

Regarding the pSWE, several early studies reported a high accuracy for liver fibrosis staging in both HBV patients [68, 69, 70, 71] and HCV patients [72, 73, 74, 75]. For example, a study using pSWE done by Sporea et al. reported an AUROCs of 0.91 for detecting F3-F4 stage fibrosis and 0.94 for detecting cirrhosis (F4), respectively, in 274 patients having chronic HCV infection [72]. A meta-analysis including 21 studies containing 2691 individual data with chronic HBV or HCV infections showed an AUROCs of 0.88 for the detection of F2-F4, and 0.91 for the diagnosis of cirrhosis, respectively [76]. pSWE also provides a good diagnostic performance in diagnosing liver fibrosis stage for NAFLD patients, and the reported AUROCs of pSWE to detect liver cirrhosis (F4) was greater than 0.97 [77, 78, 79, 80]. When pSWE was compared to TE for NAFLD patients in assessing liver fibrosis stage, there was no significant difference in diagnostic capability between the two elastography methods, although pSWE provided a significantly higher rate of reliable measurement [81].

In addition to pSWE, 2D-SWE also provides excellent diagnostic performance in assessing the liver fibrosis stage for patients having chronic HBV or HCV infection [20, 82, 83, 84]. However, since 2D-SWE is the newest elastography method, it has been less validated than TE or pSWE. A meta-analysis including seven studies using 2D-SWE in assessing liver fibrosis stage showed an AUROCs of 0.91 for detection of F2-F4 stage fibrosis and 0.95 for the diagnosis of liver cirrhosis (F4) [85]. In addition, recent studies reported that 2D-SWE showed a significant better diagnostic capability in detecting both F3-F4 stage fibrosis and cirrhosis (F4) than TE [86, 87]. The same as the chronic HBV or HCV patients, 2D-SWE is less well-validated for NAFLD patients than TE or pSWE. Several prospective studies showed a good diagnostic performance of 2D-SWE in detecting liver cirrhosis for NAFLD patients with AUROCs ranging from 0.88 to 0.95 [88, 89, 90].

5. Limitation of ultrasound-based shear wave elastography for the liver

Although currently available ultrasound-based shear wave elastography systems including TE, pSWE, and 2D-SWE provide an excellent diagnostic capability in assessing liver fibrosis stage and are widely used in clinical practice, ultrasound-based shear wave elastography systems have some limitations. Operators should be aware of the limitations of current ultrasound-based shear wave elastography techniques for accurate measurement of liver stiffness value as well as for the appropriate interpretation of the results. After the introduction of pSWE and 2D-SWE that can be incorporated into commercial ultrasound systems for routine B-mode imaging, many of manufacturers provide their own SWE systems for liver stiffness measurement. Therefore, inter-platform variability among the different SWE systems from the various vendors may be an issue [15]. In the view of physics, the liver stiffness measurement values obtained by different SWE systems from different vendors can not be interchangeable. Thus, vendor-specific cut-off values for the assessment of the liver fibrosis stage are needed since the frequencies of shear wave generated within the liver tissue are different among the various SWE systems from different vendors: 50 Hz for TE and wideband ranging from 100 to 500 Hz for pSWE and 2D-SWE [31, 91, 92]. However, the application of vendor-specific cut-off might be infeasible in clinical practice and it is hardly possible to follow up patients with the same SWE system during the disease course. According to the result of the study evaluating inter-observer variability of liver stiffness measurements among seven different SWE systems including TE, four pSWE methods, and two 2D-SWE methods, the overall agreement among the liver stiffness measurements performed with different SWE systems was good to excellent having ICCs ranging from 0.74 to 0.97 [93]. There would be an approximately 10% variability of the liver stiffness measurements among the different vendor SWE systems [93]. Therefore, these inter-platform variabilities should be taken into account in the application of various SWE systems from different vendors for the assessment of liver fibrosis staging.

To calculate the liver stiffness value from the measured shear wave propagation velocity, the current SWE systems assume that the tissue in that a stress is applied is purely elastic, and neglect the tissue viscosity. However, in some clinical situations, the assumption of pure tissue elasticity does not work well, leading to errors in the liver stiffness measurements. These conditions include acute hepatitis, liver inflammation with necrosis, obstructive cholangitis, hepatic congestion, and infiltrative disease such as amyloidosis or lymphoma [15], and have been known to increase tissue viscosity. When the tissue viscosity is increased by various causes, the liver stiffness values measured by SWE systems are usually higher than without those conditions, leading to the over-estimation of the liver fibrosis stage [94]. Therefore, current guidelines for liver elastography examination do not recommend the liver stiffness measurement for the assessment of liver fibrosis stage when the serum level of aspartate aminotransaminase (AST) and/or alanine aminotransaminase (ALT) is elevated greater than five times upper normal limits [15]. The assessment of the liver fibrosis stage by using liver SWE can be performed after the normalization of AST and/or ALT level to minimize the effect of liver inflammation on the results of liver stiffness measurement. In addition, tissue viscosity introduces a dependency of shear wave propagation velocity on excitation frequencies [23, 95]. Given that, more complex modeling taking tissue viscoelasticity into account is warranted to overcome the current limitation of ultrasound-based shear wave elastography for the liver.

6. Conclusion

Many studies reported an excellent diagnostic performance of ultrasound-based shear wave elastography in the evaluation of liver fibrosis and detection of liver cirrhosis. Among the various shear wave elastography techniques, TE has been the most widely used and extensively validated method for the assessment of liver fibrosis, subsequently having a higher level of evidence compared to the other elastography methods. In addition to TE, pSWE, and 2D-SWE have emerged as another noninvasive methods for the assessment of liver fibrosis. Since both pSWE and 2D-SWE utilize the conventional ultrasound probe for routine B-mode imaging equipped in standard diagnostic ultrasound machines, pSWE, and 2D-SWE can provide B-mode images of the liver simultaneously during the examination, enabling the liver stiffness measurement under the real-time B-mode image guidance. Although current ultrasound-based shear wave elastography techniques including TE, pSWE, and 2D-SWE provide an excellent diagnostic capability in assessing liver fibrosis stage, interchangeability of liver stiffness measurement results among the different SWE systems from different vendors may be an issue. In addition, the presence of concurrent liver inflammation with/without necrosis, hepatic congestion, obstructive cholestasis, and diffuse infiltrative disease in the liver, which can increase the tissue viscosity, is another limitation of the current liver elastography technique for the diagnosis of liver fibrosis and cirrhosis, leading to over-estimation of liver fibrosis stage. Therefore, operators should be aware of the limitations of current SWE systems for proper use of SWE technique in assessing liver fibrosis stage as well as for the accurate interpretation of the liver stiffness measurement results.