REVIEWS AND COMMENTARY• STATEMENTS AND GUIDELINES

Update to the Society of Radiologists in Ultrasound Liver

Elastography Consensus Statement
Richard G. Barr, MD, PhD  •  Stephanie R. Wilson, MD  •  Deborah Rubens, MD  •  Guadalupe Garcia-Tsao,
MD  •  Giovanna Ferraioli, MD
From the Department of Radiology, Northeastern Ohio Medical University, Rootstown, Ohio (R.G.B.); Department of Radiology, University of Calgary, Calgary, Canada
(S.R.W.); Departments of Imaging Science, Oncology, and Biomedical Engineering, University of Rochester Medical Center, Rochester, NY (D.R.); Section of Digestive
Diseases, Department of Medicine, Yale University, New Haven, Conn (G.G.T.); and Ultrasound Unit, Department of Clinical Sciences and Infectious Diseases, Fondazi-
one IRCCS Policlinico San Matteo, University of Pavia, Pavia, Italy (G.F.). Received October 31, 2019; revision requested December 11; revision received April 2, 2020;
accepted April 23. Address correspondence to R.G.B., Southwoods Imaging, 7623 Market St, Youngstown, OH 44512 (e-mail: [email protected]).

Conflicts of interest are listed at the end of this article.

Radiology 2020; 00:1–12 • https://doi.org/10.1148/radiol.2020192437 • Content codes:

This multidisciplinary update of the Society of Radiologists in Ultrasound consensus statement on liver elastography incorporates the
large volume of new information available in the literature since the initial publication. The recommended procedure for acquiring
stiffness measurements is reviewed. There has been substantial improvement in the acoustic radiation force impulse (ARFI)
technology—most notably the addition of a quality assessment of the shear wave propagation. Due to the efforts of the
Quantitative Imaging Biomarkers Alliance, or QIBA, the variability of liver stiffness measurements between systems had decreased.
There are now effective treatments for hepatitis B and hepatitis C, and follow-up after effective treatment should be based on the
use of the delta change of the value obtained at viral eradication or suppression. Because the detection of compensated advanced
chronic liver disease (cACLD) is very important, the new guidelines are made based on the probability of cACLD for given stiffness
values. The panel recommends a vendor-neutral rule of four for interpretation for ARFI techniques. This new method simplifies
interpretation of liver stiffness results and is more clinically relevant.
© RSNA, 2020

T his statement is an update produced by the Society of

Radiologists in Ultrasound (SRU). Authors include the
clinical members of the original statement and comprise
This led to a need to update the SRU recommendations
on the use of ARFI SWE for the assessment of fibrosis in
patients with diffuse liver disease, as a guide for performing
society representatives and hepatologists with expertise in and interpreting the examination, taking into account the
liver elastography in the United States and the European interim technology advances and published studies.
Union. The revision process involved identifying a panel Chronic liver disease is a world-wide problem. It can
leader (R.G.B.), who then selected relevant previous panel- be due to a wide range of inciting factors. Its major conse-
ists to participate in the update. The panel chair and co- quence is increasing deposition of fibrous tissue within the
chair (G.F.) created a preliminary draft with recommended liver leading to the development of cirrhosis, which in turn
updates, which were reviewed by the panel. Consensus was may give rise to portal hypertension, hepatic insufficiency,
obtained iteratively after successive reviews and revisions and hepatocellular carcinoma. The stage of liver fibrosis is
and finalized after review by the SRU Executive Board. important to determine the prognosis, for surveillance, for
The use of shear-wave elastography (SWE) for the non- prioritization for treatment, and even to determine the po-
invasive assessment of liver fibrosis has grown rapidly, and tential for reversibility (1,2,7–9). The spectrum of fibrosis
substantial new information regarding disease-specific liver is a continuum, and patients with a higher stage of liver
stiffness is available since the publication of the consensus fibrosis (stage F3–F4) are at risk for clinical complications
statement of the SRU in September 2015 (1,2). Vibration- (eg, ascites, variceal hemorrhage, hepatic encephalopathy).
controlled transient elastography has been available for For patients with severe fibrosis or liver cirrhosis who are
almost 20 years and has a large body of literature (3–5). asymptomatic, the term “compensated advanced chronic
Acoustic radiation force impulse (ARFI) techniques, both liver disease” (cACLD) has been proposed (10,11). In
point SWE (pSWE) and two-dimensional (2D) SWE have patients with cACLD, the degree of portal hypertension
been available for almost 10 years. Currently, several ven- is predictive of decompensation and/or death (10,11). A
dors implement ARFI technology (both pSWE and 2D portal pressure (as assessed by means of the hepatic venous
SWE, which are described in detail elsewhere [2,6]) in pressure gradient) of 10 mm Hg or higher (normal, 3–5
their US equipment and provide suggestions for optimal mm Hg)—a threshold that is designated “clinically signifi-
technique and assessment of data quality. Since publica- cant portal hypertension” (CSPH)—has been associated
tion of the previous guidelines, several additional vendors with an almost four-fold higher risk of decompensation
have introduced ARFI techniques, and the development compared with lower pressures (12).
of quality or confidence maps have led to the ability to as- Many clinical guidelines recommend the use of non-
sess the quality of the results. With excellent, less-expensive invasive tests for the detection and staging of liver fibro-
treatments for both hepatitis C and hepatitis B, these pa- sis (3,5,13,14). Although biopsy is historically the refer-
tients are being treated regardless of the liver stiffness value. ence standard for staging fibrosis, it is imperfect, with

This copy is for personal use only. To order printed copies, contact [email protected]
Update to the Society of Radiologists in Ultrasound Liver Elastography Consensus Statement

prevalence of 25% with imaging estimation (19). NAFLD ranges

Abbreviations from simple steatosis to nonalcoholic steatohepatitis, which may
ARFI = acoustic radiation force impulse, cACLD = compensated ad- progress to liver fibrosis and cirrhosis with its complications.
vanced chronic liver disease, CSPH = clinically significant portal hyper-
tension, IQR = interquartile range, NAFLD = non-alcoholic fatty liver Although there is no specific therapy for nonalcoholic steatohep-
disease, pSWE = point SWE, SRU = Society of Radiologists in Ultra- atitis, lifestyle modifications have been associated with a decrease
sound, SWE = shear wave elastography, 2D = two-dimensional in fibrosis and portal hypertension (20,21), and identification of
cACLD allows for screening and surveillance of varices and
Summary
hepatocellular carcinoma. Therefore, the availability of non-
To follow-up patients, the consensus suggests using the delta changes
of liver stiffness over time instead of the absolute values, using as a invasive tools to exclude or diagnose cACLD in these patients
baseline value in case of viral hepatitis that obtained after viral eradi- is of the utmost importance.
cation or suppression.

Key Results Protocol for ARFI SWE Acquisition

n The variability between consecutive liver stiffness acquisitions, as- The patient preparation, imaging technique, and measurement
sessed by means of the interquartile range–to-median ratio, is the recommendations for ARFI SWE (both pSWE and 2D SWE)
most important quality criterion; when this ratio is higher than
30% for measurements given in kilopascals or higher than 15% are the same, and the recommended protocol in the original
for measurements given in meters per second, the accuracy of the SRU consensus is unchanged and similar to the European Fed-
technique is reduced. eration of Societies for Ultrasound in Medicine and Biology
n Given the large overlap of stiffness values for mild-to-moderate and World Federation for Ultrasound in Medicine and Biology
fibrosis, the Society of Radiologists in Ultrasound continues to guidelines (3,5). The protocol includes obtaining measurements
recommend a low cutoff value below which there is a high prob-
ability of no or mild fibrosis and recommends a high cut-off value between the ribs in the right upper quadrant, instructing the
above which there is a high probability of compensated advanced patient to fast for at least 4 hours, imaging the patient in a supine
chronic liver disease (cACLD). or slight left lateral decubitus position (not more than 30°) with
n Because the overlap of liver stiffness values between METAVIR their right hand above their head, obtaining measurements in
scores is as large if not larger than the difference between vendors, a neutral breath hold, placing the transducer perpendicular to
separate cut-off values for each vendor are not required.
the liver capsule and the measurement box parallel to the liver
n The panel recommends a vendor-neutral “rule of four” for the
acoustic radiation force impulse techniques in the viral causes capsule, and taking measurements 1.5–2.0 cm from the liver
and non-alcoholic fatty liver disease (NAFLD): Liver stiffness less capsule to avoid reverberation artifact. A brief outline of how to
than or equal to 5 kPa (1.3 m/sec) has high probability of being perform the examination is included in Table 1.
normal, liver stiffness less than 9 kPa (1.7 m/sec), in the absence Because B-mode is used to track the shear waves, high-quality
of other known clinical signs, rules out cACLD, and values greater
than 13 kPa (2.1 m/sec) are highly suggestive of cACLD; in some B-mode imaging is required. Images should be free of artifacts.
patients with NAFLD, the cut-off values for cACLD may be lower Several studies have shown that operators require only a short
and follow-up or additional testing in those with values between 7 period of training to perform reliable liver stiffness measure-
and 9 kPa is recommended.
ments; however, the reproducibility of liver stiffness measure-
n For pediatric patients with liver disease or congenital heart disease ments over time is higher for expert operators than for novice
with Fontan surgery, it is expert opinion that each patient becomes
his or her own control, and the stiffness delta changes over time operators (22–24).
should be used to evaluate the efficacy of the treatment or the pro-
gression of disease.
Quality Criteria
The recommended quality criteria include the number of
considerable interobserver variability and k values varying from required acquisitions and the interquartile range (IQR)–to-
0.5 to 0.9 in the literature (15,16). It should be emphasized, median ratio (subsequently referred to as IQR/M). Further-
however, that histologic examination of liver specimens does more, some vendors provide a quality or confidence factor for
provide information on inflammation that is not yet possible measurements obtained with 2D SWE. Some vendors also
to evaluate with US. Despite this benefit, the use of noninvasive provide an assessment of the quality of each measurement for
tests is favored due to the need for longitudinal monitoring and pSWE. Each vendor has recommendations for use of their
to safely extend screening to larger populations. quality criteria.
There are many different causes of chronic liver disease world-
wide. Chronic viral hepatitis (hepatitis C in the West, hepatitis B Obtaining Measurements
in the East) remains a major risk factor. Although the incidence Measurements should be obtained in areas of high quality,
of cACLD may be lower because of the advent of highly effective which is determined by a high amplitude of the shear waves, a
interferon-free antiviral therapies, staging of liver fibrosis is still normal shear-wave propagation, and a linear slope of the time
necessary before treatment because patients with cACLD require of the peak and distance from ARFI pulse of the displacement
continued surveillance for hepatocellular carcinoma and/or vari- curves. Each vendor provides a confidence or quality number
ces even after the clearance of the virus (17,18). or map that combines these factors into one number for clini-
A rising cause of chronic liver disease worldwide is nonalco- cal use. Figure 1 demonstrates various methods used to assess
holic fatty liver disease (NAFLD). NAFLD is currently the most the quality of an image. If the quality is poor in most of the im-
common liver disease in the United States, with a worldwide age, a measurement should not be obtained from that image.

2 radiology.rsna.org  n  Radiology: Volume 00: Number 0— 2020

 Barr et al

Table 1: Recommendations for Performing Liver Stiffness Measurements with the ARFI Technique

Recommendations
1. Patients should fast at least 4 hours before the examination
2. Measurement should be taken at an intercostal space with the patient in the supine or slight lateral decubitus (30°) position with right
arm in extension
3. Measurements should be taken at neutral breathing during a breath hold
4. Measurement should be taken at least 15–20 mm below liver capsule in pSWE
5. The 2D SWE region of interest can be positioned closer to the liver capsule, if reverberation artifacts are avoided; however, the measure-
ment box should be positioned at least 15–20 mm below the liver capsule
6. Results can be reported in meters per second or in kilopascals
7. In most systems, the maximum ARFI push pulse is at 4–4.5 cm from the transducer, which is the optimal location for obtaining mea-
surements. In most systems, the ARFI push pulse is attenuated by 6–7 cm, limiting adequate shear wave generation
8. Major potential confounding factors include liver severe inflammation indicated by AST and/or ALT elevation greater than five times
upper normal limits, obstructive cholestasis, liver congestion, acute hepatitis, and infiltrative liver disease (these all lead to overestimation
of the stage of fibrosis)
9. Ten measurements should be obtained with pSWE, and the final result should be expressed as the median together with the IQR/M
10. Fewer than 10 measurements with pSWE can be obtained (at least five); however, the IQR/M should be within the recommended range
11. For 2D SWE, five measurements should be obtained when the manufacturer’s quality criteria are available, and the final result should be
expressed as the median together with the IQR/M
12. The most important reliability criterion is an IQR/M of 30% of the 10 measurements (pSWE) or five measurements (2D SWE) for
kilopascals and 15% for measurements in velocity (in meters per second)
13. Adequate B-mode liver imaging is a prerequisite for point and 2D SWE as shear waves are tracked with B-mode
Note.—ALT = alanine aminotransferase, ARFI = acoustic radiation force impulse, AST = aspartate aminotransaminase, IQR/M = inter-
quartile range–to-median ratio, pSWE = point SWE, SWE = shear-wave elastography, 2D = two-dimensional.

Number of Measurements Cut-off Values

Cut-off values for fibrosis staging vary across US systems from
pSWE.—Ten measurements are still recommended; however,
different vendors; however, the variance has decreased due to
studies have shown that there is no loss in accuracy with five
the efforts of the Quantitative Image Biomarker Alliance, or
measurements when the quality criterion of IQR/M is fulfilled
QIBA (29,30). QIBA (an RSNA organization with vendors,
(25–28). In the study by Fang et al (25), six measurements were
scientists, members of the U.S. Food and Drug Administra-
recommended; however, when only the values obtained with a
tion, and clinicians) developed standardized phantoms that the
high reliability (IQR/M, ≤30%) were considered, there was no
vendors have used to standardize their measurements. The dif-
difference between five and six measurements.
ference between various system measurements increases as liver
Two-dimensional SWE.—The measurement area is larger stiffness increases. The difference in cut-off values is greatest as
than that with pSWE, and thus each value is an average of patients exceed the threshold of cACLD (31).
several measurements. Hence, five measurements are adequate Given the large overlap of stiffness values for mild-to-mod-
if a quality assessment is provided by the manufacturer. If a erate fibrosis, the SRU continues to recommend a low cut-off
quality assessment is not available, 10 measurements are value below which there is a high probability of no or mild fi-
recommended. brosis and recommends a high cut-off value above which there
is a high probability of cACLD. In this update, a new cut-off
IQR/M Values value to rule out CSPH has been added on the basis of some re-
Studies have shown that the level of variability between con- cent studies (32–35). The consensus panel also divides the liver
secutive acquisitions, assessed by means of the IQR/M, is the stiffness values between no or minimal disease and cACLD into
most important quality criterion. When this ratio is higher two categories. For these middle liver stiffness values, confirma-
than 30% (for measurements given in kilopascals), the accu- tion with an additional test may be needed to rule in or rule out
racy of the technique is reduced (3,25,27). It is important to cACLD. From a clinical perspective, it is more important to rule
note that the IQR/M for measurements reported in kilopas- in or rule out significant disease than it is to provide an exact stage
cals should be 30% or less, whereas that for measurements by using the METAVIR scoring system. Because of the large liver
reported in meters per second (shear wave speed) should be stiffness value overlap of METAVIR scores (1), which is greater
15% or less as the conversion of meters per second to kilopas- than the measurement variability between vendors (31), separate
cals is nonlinear. If the IQR/M values are greater than 30% cut-off values for each vendor are not required. Based on some
in kilopascals or 15% in meters per second, the measurement published studies and mirroring the Baveno VI consensus confer-
of liver stiffness should be judged as unreliable. ence (10,11), that is, the so-called “rule of five” (5, 10, 15, 20 kPa)

Radiology: Volume 00: Number 0— 2020  n  radiology.rsna.org 3

Update to the Society of Radiologists in Ultrasound Liver Elastography Consensus Statement

Figure 1:  (a) Image obtained with point shear-wave elastography (pSWE) system (ElastPQ; Philips, Bothell, Wash). A standard deviation (Std) of 30% or less of the
mean value is indicative of an acquisition of good quality. In this case, the standard deviation is 1.06/9.90 or 10.7%. When the signal-to-noise ratio of an acquisition is
very low, the mean value is not shown. (b) Image obtained with pSWE (SWM; Hitachi, Tokyo, Japan). “VsN” is a reliability index that indicates the percentage of effec-
tive push-track sequences. When the signal-to-noise ratio of an acquisition is very low, the mean value is not shown. A good acquisition has a VsN of at least 50%. In this
case, the VsN measurements are all above 66%. (c) Image obtained with pSWE (VTQ; Siemens, Mountain View, Calif). The system automatically filters out the measure-
ments that are not good. In these cases, the numeric value of shear-wave speed is replaced by an “XXX” sequence. (d) Images obtained with two-dimensional (2D)
shear-wave elastography (SWE) (EQI, Philips). The color-coded confidence map (left) is an evaluation of the quality of the acquired signals. The confidence threshold
(CT) is set at 60%: Areas of low quality (red) are filtered out and left blank on the color-coded image of liver stiffness assessment (right); the yellow color on the confidence
map is a warning, that is, it indicates that the acquisition in that area is not the highest quality (Fig 1 continues).

for the staging of liver fibrosis with vibration-controlled transient For other causes such as alcoholic hepatitis, primary biliary
elastography, the consensus panel proposes a vendor-neutral “rule cirrhosis, Wilson disease, autoimmune hepatitis, sclerosing chol-
of four” (5, 9, 13, 17 kPa) for the ARFI techniques for viral etiolo- angitis, and drug-induced liver disease, there is insufficient data
gies and NAFLD: Liver stiffness of 5 kPa (1.3 m/sec) or less has to make a conclusion.
high probability of being normal; liver stiffness less than 9 kPa (1.7 Table 2 summarizes these cut-off value recommendations and
m/sec), in the absence of other known clinical signs, rules out cA- provides them in both kilopascals and meters per second. For
CLD; values between 9 kPa (1.7 m/sec) and 13 kPa (2.1 m/sec) are those who would like a value to rule out significant fibrosis, most
suggestive of cACLD but may need further test for confirmation; studies that used ARFI (pSWE and 2D SWE) suggest that a liver
and values greater than 13 kPa (2.1 m/sec) are highly suggestive of stiffness value of less than 7 kPa (1.5 m/sec) can help rule out
cACLD. There is a probability of CSPH with liver stiffness values significant fibrosis.
greater than 17 kPa (2.4 m/sec), but additional patient testing may With vibration-controlled transient elastography, the ala-
be required. In some patients with NAFLD, the cut-off values for nine aminotransferase–adapted cut-off values of liver stiffness
cACLD may be lower and follow-up or additional testing in those reportedly improved the staging of liver fibrosis in patients
with values between 7 and 9 kPa is recommended. with chronic hepatitis B in a single study (36). The consensus

4 radiology.rsna.org  n  Radiology: Volume 00: Number 0— 2020

 Barr et al

Figure 1 (continued):  (e) Images obtained with 2D SWE (STE; Mindray, Shen-
zhen, China). Two quality criteria are provided: the motion stability (M-STB) index,
which is indicated by stars (with the highest stability shown with five green stars), and
the reliability (RLB) map, which goes from purple to green—with the latter indicating
the highest reliability. The stars are an indicator of motion during the acquisition. When
there are fewer than four stars, there is significant motion during the acquisition and
that frame should not be used for liver stiffness measurement. (f) Images obtained
with 2D SWE (Aplio; Canon, Tochigi, Japan). The system filters out values with a low
signal-to-noise ratio, and these areas are left blank. The proprietary quality parameter
is the propagation map (right). A proper propagation map is displayed with parallel
lines, with the intervals between the lines constant. The propagation map is used to
guide placement of the measurement box. Image on left is velocity map. (g) Images
obtained with 2D SWE (SSI; SuperSonic, Aix-en-Provence, France). Values with a
low signal-to-noise ratio are filtered out. The stability index (SI) is an indicator of tem-
poral stability, and it is displayed while positioning the measurement box (Q-Box). An
acquisition of good quality should have a stability index greater than 90%. Top image
is velocity map, and bottom image is B-mode image.

Table 2: Recommendation for Interpretation of Liver Stiffness Values Obtained with ARFI Techniques in Patients with Viral Hepa-
titis and NAFLD

Liver Stiffness Value Recommendation

5 kPa (1.3 m/sec) High probability of being normal
,9 kPa (1.7 m/sec) In the absence of other known clinical signs, rules out cACLD. If there are known clinical
signs, may need further test for confirmation
9–13 kPa (1.7–2.1 m/sec) Suggestive of cACLD but need further test for confirmation
.13 kPa (2.1 m/sec) Rules in cACLD
.17 kPa (2.4 m/sec) Suggestive of CSPH
Note.—ARFI = acoustic radiation force impulse, cACLD = compensated advanced chronic liver disease, CSPH = clinically significant
portal hypertension, NAFLD = non-alcoholic fatty liver disease.

panel therefore does not recommend alanine aminotransfer- (1,3,5). These conditions include, but are not limited to, acute
ase–adapted cut-offs until additional publications confirm its hepatitis, liver inflammation, transaminitis flares with ala-
usefulness. The updated World Federation of Ultrasound in nine aminotransferase value more than five times the up-
Medicine and Biology guidelines provide a detailed review of per limit of normal, obstructive cholestasis, hepatic con-
the literature for several of the causes that progress to chronic gestion, and infiltrative liver diseases such as amyloidosis,
liver disease and associated confounding factors (3). lymphoma, or extramedullary hematopoiesis. Other factors
may also affect liver stiffness measurement, such as post-
Confounding Factors prandial hyperemia or intense physical exercise. In all these
There are several clinical conditions in which an increase conditions, however, stiffness values within the normal
of liver stiffness unrelated to liver fibrosis can be observed range exclude significant liver fibrosis.

Radiology: Volume 00: Number 0— 2020  n  radiology.rsna.org 5

Update to the Society of Radiologists in Ultrasound Liver Elastography Consensus Statement

Table 3: Recommendations for Performing Spleen Stiffness Measurements with the ARFI Technique

Recommendations
1. Patients should fast at least 4 hours before the examination (56)
2. Measurement should be taken at an intercostal space with the patient in supine position with left arm in extension
3. Measurements should be taken during breath hold at neutral breathing (57)
4. Measurement should be taken at least 15 mm below spleen capsule with pSWE and reverberation artifacts avoided with 2D SWE. The
region of interest should be placed perpendicular to the splenic surface
5. Results can be reported in meters per second or kilopascals
6. In most systems, the maximum ARFI push pulse is at 4–4.5 cm from the transducer, which is the optimal location for obtaining mea-
surements. In most systems, the ARFI push pulse is attenuated by 6–7 cm, limiting adequate shear wave generation
7. Ten measurements should be obtained with pSWE, and the final result should be expressed as the median together with the IQR/M
8. For 2D SWE, five measurements should be obtained, and the final result should be expressed as the median together with the IQR/M
9. The most important reliability criteria is a IQR/M of 30% of the recommended measurements for kilopascals and 15% for meters
per second
Note.—ARFI = acoustic radiation force impulse, IQR/M = interquartile range–to-median ratio, pSWE = point SWE, SWE = shear-wave
elastography, 2D = two-dimensional.

of liver stiffness values over time should be used instead of the ab-
solute values (37–40,42). Thus, every patient becomes his or her
own control. Because there is an approximately 10% variability of
the measurements within a vendor and between vendors (29,30),
a clinically significant change should be considered when the delta
change is greater than 10%. The panel recommends using the
same equipment for follow-up studies. In patients with chronic vi-
ral hepatitis who are successfully treated, the baseline liver stiffness
should be that obtained after viral eradication or suppression. Ap-
plying this rule, liver stiffness assessment can be suitable for evalu-
ating all clinical conditions leading to an increase of liver stiffness,
independent of the disease etiology including nonfibrotic causes of
liver stiffness increase, such as congestive heart failure.

Spleen Stiffness
It has been reported that liver stiffness correlates with the se-
verity of liver fibrosis up to the threshold of CSPH, defined as
Figure 2:  Image obtained with two-dimensional (2D) shear-wave elastogra- an increase in hepatic venous pressure gradient greater than
phy (SWE) demonstrates area of increased stiffness (red and teal, arrows) due to 10 mm Hg (43). In patients with CSPH, the strength of the
reverberation artifact. The reverberation artifact occurs below the liver capsule in correlation between liver stiffness and fibrosis decreases, prob-
both point SWE (pSWE) and 2D SWE. In pSWE, the artifact is not seen; therefore, ably due to an increasing role played by extrahepatic factors,
it is important to obtain measurements at least 1.5 cm below the liver capsule to
avoid the artifact. This area should be avoided when placing the measurement
mainly the increase in portal venous inflow, as portal hyperten-
box for liver stiffness measurements. sion progresses (10,44). The acquisition technique is the same
as that for liver, except the measurements are taken between the
left ribs with the patient in a supine or slight right lateral posi-
Follow-up tion. It is the opinion of the expert panel that adequate studies
In patients with chronic hepatitis B virus or hepatitis C virus have not been performed to provide cut-off values at this time.
who have been successfully treated with antiviral drugs, the A review of the existing literature is provided below. In patients
cut-offs obtained in viremic patients should not be used be- with chronic liver disease, splenic measurements should only
cause a rapid decline of stiffness values has been observed in be taken in patients with cACLD as significant portal pressures
these patients, likely due to the decrease of liver inflammation are not expected at lower levels of fibrosis.
(3,5). When liver cirrhosis is evident with B-mode findings, CSPH is predictive of the development of complications of
elastography should not be used to rule out the disease because cirrhosis, including variceal rupture and death. However, it is
a value in the low range of liver stiffness may only indicate a also present in about 50%–60% of patients with compensated
successful response to antiviral treatment. cirrhosis without gastroesophageal varices (12,45). It appears
On the basis of results of both prospective and retrospective that spleen stiffness shows better correlation with portal pressure
studies with more than 1000 patients (37–41), the delta change than does liver stiffness (46). Portal hypertension leads to splenic

6 radiology.rsna.org  n  Radiology: Volume 00: Number 0— 2020

 Barr et al

Figure 3:  (a) Artifacts occur around large blood vessels and bile ducts. These artifacts are not seen in point shear-wave elastog-
raphy (SWE), and therefore measurements should be obtained at least 5 mm from these structures. In two-dimensional SWE, these
artifacts can be identified and avoided. Image on right is velocity map, and image on left is quality map. Arrows indicate artifacts.
Depending on the vendor, artifacts may not be color-coded or appear as areas of increased stiffness (teal). These areas should be
avoided when placing the measurement box. (b) Shear-wave propagation occurs in all directions perpendicular to the acoustic
radiation force impulse (ARFI) pulse. Therefore, artifacts from a blood vessel just out of the image plane can also produce artifacts.
Velocity image (right) shows artifacts in teal (white arrows). These artifacts are most likely from vessels just out of the image plane. The
measurement box should not include these areas. Black arrows point to teal areas at the deep part of the image. These are artifacts
from the ARFI pulse strength decreased due to attenuation, leading to weak shear waves that make it difficult to obtain accurate esti-
mates of shear-wave speed. Note that the quality map (left) in this case suggests high quality throughout the field of view. The quality
map does not identify all artifacts, and both the quality map and velocity map should be evaluated for artifacts.

congestion, increasing splenic stiffness. In fact, portal hyperten- However, there are differences in cut-off values between studies,
sion may cause splenic fibrosis (47). and the level of evidence is still too low to recommend spleen stiff-
In healthy individuals, the spleen is stiffer than the liver. Several ness in the diagnostic work-up of patients with cirrhosis.
studies, most of which were performed with vibration-controlled For ARFI-based techniques, limited studies suggest that
transient elastography, have shown that, in patients with portal abdominal wall thickness and splenic longitudinal diam-
hypertension, spleen stiffness is more reliable than liver stiffness eter are independent predictors of successful spleen stiffness
for assessing the risk of CSPH and esophageal varices (46,48–50). measurement (51,52). The feasibility of performing spleen

Radiology: Volume 00: Number 0— 2020  n  radiology.rsna.org 7

Update to the Society of Radiologists in Ultrasound Liver Elastography Consensus Statement

Figure 4:  Images from two-dimensional shear-wave elastography. Image on left is confidence map, and image on right is veloc-
ity map. When the acoustic radiation force impulse pulse is not perpendicular to the liver capsule, artifacts occur. In this case, the liver
capsule (dashed white line) is not parallel to the transducer (solid white line) or the field-of-view box (red line). The heterogeneous
stiffness measurements in the field of view are due to artifacts occurring because the three lines are not parallel.

for performing spleen stiffness measurement is pre-

sented in Table 3.
With use of pSWE, investigators in one study
reported a higher incidence of esophageal variceal
bleeding in patients with a spleen stiffness value of
at least 39 kPa (3.64 m/sec); no bleeding occurred
in patients with spleen stiffness less than 36 kPa
(3.48 m/sec) (58). With use of 2D SWE, other
investigators showed that CPSH is unlikely in pa-
tients with spleen stiffness less than 26.6 kPa (3.0
Figure 5:  Suggested reporting format for liver stiffness measurements. cACLD = compen- m/sec) (35). Algorithms that combine liver stiffness
sated advanced chronic liver disease, IQR = interquartile range, NAFLD = non-alcoholic fatty
and spleen stiffness, or platelets count, have been
liver disease, SWE = shear wave elastography, SRU = Society of Radiologists in Ultrasound,
2D = two-dimensional. proposed (59).
In a multicenter study in which liver stiffness
and spleen stiffness were available in 109 patients
stiffness measurement was evaluated by Procopet et al (53) undergoing hepatic venous pressure gradient measurement, liver
in 88 patients undergoing hepatic venous pressure gradient stiffness of 16.0 kPa (2.3 m/sec) or less and spleen stiffness of
measurement for portal hypertension. The overall success rate 21.7 kPa (2.7 m/sec) or less were able to help rule out CSPH,
of obtaining an accurate measurement, defined as the system whereas liver stiffness values greater than 29.5 kPa (3.2 m/sec)
being able to estimate a stiffness value, was 66%. In that series, and spleen stiffness values greater than 35.6 kPa (3.5 m/sec) were
the patients with failure of spleen stiffness had higher body able to help rule in CSPH (specificity, .92%). In patients with
mass index (mean, 28.3 kg/m2 6 5.0 vs 25.2 kg/m2 6 3.7; liver stiffness of 38.0 kPa (3.6 m/sec) or less, a splenic stiffness
P = .002) and smaller spleen (mean bipolar diameter, 11.8 cm greater than 27.9 kPa (3.2 m/sec) ruled in CSPH. This algo-
6 2.7 vs 14.2 cm 6 4.0; P , .0001). In a series composed rithm had a sensitivity of 89.2% and a specificity of 91.4% to
of 313 consecutive patients who underwent liver stiffness and rule in CSPH (41). However, in a series of 191 patients (60), this
spleen stiffness measurements on the same day (52), the suc- algorithm has not been validated: Specificity and positive predic-
cess rate of spleen stiffness measurement was 80% in patients tive value were 52% and 83%, respectively.
with splenomegaly. Technical success of spleen stiffness mea- Interestingly, it has been reported that patients with hepatis
surements was 78% in another small series (54), including 54 C virus hepatitis successfully treated with antiviral drugs show a
patients with cirrhosis who either had low-grade esophageal rapid decline of liver stiffness but not of spleen stiffness because
varices or were without esophageal varices at upper endoscopy. there is not an immediate effect on portal hypertension. Spleen
Normal values of spleen stiffness with ARFI-based tech- stiffness is more accurate in assessing portal hypertension in this
niques in published studies range from 20.5 kPa (2.6 m/sec) setting. Therefore, the risk of variceal hemorrhage remains in the
to 24.4 kPa (2.85 m/sec) (52,53,55). The suggested procedure short term (61).

8 radiology.rsna.org  n  Radiology: Volume 00: Number 0— 2020

 Barr et al

Table 4: Summary of Recommendations

Protocol for acquisition: As reported in Table 1, the most important criterion is IQR/M 30% for values in kilopascals and 15% for values
in meters per second. In pediatric patients, the same protocol must be used
Protocol for 2D SWE acquisition in children who are unable to hold their breath: The consensus panel suggests recording a 2D SWE cine
loop for up to 30 seconds if real-time 2D SWE is available, reviewing it, and choosing the image that demonstrates the most stable pattern
for the stiffness measurement. No more than one image should be chosen in each recorded cine loop
Cut-off values: “rule of four” (5, 9, 13, 17 kPa) for the ARFI techniques for viral causes and NAFLD (Table 2)
NAFLD and rare diseases in pediatric patients: The number of published pediatric studies of NAFLD remains low, and the cutoff values for
staging liver fibrosis varies between studies. It is expert opinion that each patient becomes his or her own control, using the stiffness delta
changes over time to evaluate the efficacy of the treatment or the progression of disease—remembering that the measurement reflects stiff-
ness and not fibrosis
Follow-up: The use the delta changes of LS values over time should be used instead of the absolute values. In patients with chronic viral
hepatitis who are successfully treated, the baseline LS stiffness should be that obtained after viral eradication or suppression. A clinically
significant change should be considered when the delta change is greater than 10%. Applying this rule, LS assessment can be suitable for
evaluating all clinical conditions leading to an increase of LS, independent of the disease cause including nonfibrotic causes of LS increase
(eg, congestive heart failure)
Spleen stiffness: It appears that spleen stiffness is better correlated with portal pressure than LS. However, there are differences in cut-off values
between studies and the level of evidence is still low to recommend spleen stiffness in the diagnostic work-up of patients with cirrhosis
Reporting: The report should include the system vendor name, the SWE technique (pSWE or 2D SWE), the probe used, the number of
acquisitions, the IQR/M, and conclusions (Fig 5)
Note.—ARFI = acoustic radiation force impulse, IQR/M = interquartile range–to-median ratio, LS = liver stiffness, NAFLD = non-alcohol-
ic fatty liver disease, pSWE = point SWE, SWE = shear-wave elastography, 2D = two-dimensional.

Pediatric Patients The mean normal shear-wave velocity value ranges from 1.07
The use of a noninvasive technique for staging liver fibrosis is of to 1.16 m/sec (66–68).
great interest because it may avoid liver biopsy, which, in addi- For liver disease associated with cystic fibrosis, autoimmune
tion to its well-known complications, is particularly stressful for hepatitis, biliary atresia and the Kasai procedure, or congeni-
pediatric patients. In the pediatric age group, NAFLD is the most tal heart disease with Fontan surgery or even NAFLD or viral
common cause of chronic liver disease. A 2015 meta-analysis hepatitis, it is expert opinion that each patient becomes his or
(62) determined that the pooled mean prevalence of NAFLD in her own control, using the stiffness delta changes over time to
the United States was 7.6% in the general U.S. pediatric popula- evaluate the efficacy of the treatment or the progression of dis-
tion and that it reached 34.2% in obese children. In one study of ease—remembering that the measurement reflects stiffness and
347 children suspected of having NAFLD who were identified not fibrosis. Results must always be interpreted considering
through screening in primary care and referral to pediatric gas- transaminase values and clinical condition.
troenterology, advanced fibrosis was present in 17% of 193 chil-
dren diagnosed with NAFLD at liver biopsy. Conversely, in 242 Steatosis Assessment
consecutive adolescents undergoing bariatric surgery, the preva- Liver fat content has also been evaluated by using US-based
lence of NAFLD was 58.8%, and 6% of the cohort had definite methods. Several studies have demonstrated proof of concept.
nonalcoholic steatohepatitis. Fibrosis was mild: 81% had none, Although there is insufficient evidence at this time to provide
while 18% had stage 1 or 2 fibrosis (63,64). recommendations regarding the use of US-based methods in
The use of noninvasive techniques in this population is par- this setting, early work suggests that these methods will be
ticularly appealing. However, the number of published pediatric clinically useful (69–73).
studies of NAFLD to date remains low and the cut-off values for
staging liver fibrosis vary between studies (65). Artifacts
For liver stiffness assessment, the procedure used for adults Artifacts are common in ARFI-based techniques and can sig-
should be adopted. In children who are unable to hold their nificantly change the liver stiffness value. It is important to
breath, the consensus panel suggests recording a 2D SWE cine recognize and avoid these artifacts (eg, liver capsule reverbera-
loop for up to 30 seconds if real-time 2D SWE is available, tion artifact [Fig 2], ARFI push artifacts, artifacts from blood
reviewing it, and choosing the image demonstrating the most vessels [Fig 3], and the artifact that occurs when the transducer
stable pattern for the stiffness measurement. No more than one is not parallel to the liver capsule [Fig 4]). Most systems now
image should be chosen in each recorded cine loop. have a confidence map or quality map that helps identify most
For ARFI-based techniques, most published studies have artifacts. However, none of the confidence maps or quality
shown that age has no significant influence on liver stiffness maps depict all artifacts and knowledge of artifacts is crucial
values (66–68). However, there is not enough literature at this for obtaining accurate liver stiffness values. Although a detailed
time for the panel to recommend the rule of four for NAFLD in discussion of artifacts is beyond the scope of this article, it is
pediatric patients. available elsewhere (74–77).

Radiology: Volume 00: Number 0— 2020  n  radiology.rsna.org 9

Update to the Society of Radiologists in Ultrasound Liver Elastography Consensus Statement

Reporting 2. How should the use of elastography change the screening

The report should include the system vendor name, the SWE interval in patients at risk for hepatocellular carcinoma?
technique (pSWE or 2D SWE), the probe used, the number of
Author contributions: Guarantors of integrity of entire study, R.G.B., G.F.; study
acquisitions, the IQR/M, and conclusions. Conclusions should concepts/study design or data acquisition or data analysis/interpretation, all authors;
use the rule of four detailed earlier (Table 2). An example of a manuscript drafting or manuscript revision for important intellectual content, all
authors; approval of final version of submitted manuscript, all authors; agrees to
report is shown in Figure 5. A summary of recommendations
ensure any questions related to the work are appropriately resolved, all authors;
is given in Table 4. literature research, all authors; clinical studies, S.R.W.; experimental studies, D.R.;
and manuscript editing, all authors
Future Directions
Disclosures of Conflicts of Interest: R.G.B. Activities related to the present
The development of new US techniques that will provide a article: institution received equipment grants from Philips Ultrasound, Siemens
measurement of liver steatosis and dispersion imaging (ie, Ultrasound, Canon Ultrasound, Mindray Ultrasound, Samsung Ultrasound, and
evaluating the change in stiffness values by varying the ARFI GE Medical. Activities not related to the present article: receives payment for
board membership at Samsung Ultrasound; receives payment for lectures includ-
frequency) are also being evaluated as a method to assess ing service on speakers bureaus from Philips Ultrasound, Siemens Ultrasound,
inflammation. This is extremely important to differentiate Canon Ultrasound, and Mindray Ultrasound; receives royalties from Thieme
simple steatosis, a benign condition, from nonalcoholic ste- Publishers; receives payment for development of educational presentations from
Philips Ultrasound and Siemens Ultrasound; receives travel/accommodations/
atohepatitis. However, evidence available for these techniques meeting expenses unrelated to activities listed from Philips Ultrasound, Siemens
is not yet at a level where recommendations can be given. Ultrasound, Canon Ultrasound, and Mindray Ultrasound. Other relationships:
Other US techniques that do not use vibration-controlled disclosed no relevant relationships. S.R.W. Activities related to the present article:
disclosed no relevant relationships. Activities not related to the present article:
transient elastography or ARFI technology techniques are institution received partial research support from Samsung; received payment
being evaluated for liver stiffness evaluation (78). for lectures including service on speakers bureaus from Philips. Other relation-
ships: disclosed no relevant relationships. D.R. disclosed no relevant relationships.
G.G.T. disclosed no relevant relationships. G.F. Activities related to the present
Future Research Questions article: disclosed no relevant relationships. Activities not related to the present
article: receives payment for lectures including service on speakers bureaus from
Canon Medical Systems, Hitachi, Mindray Bio-Medical Electronics, and Philips
Basic Questions Healthcare. Other relationships: disclosed no relevant relationships.
1. What are the sources of variability between commer-
cial SWE systems? In particular, how does the ARFI References
frequency component affect measures of stiffness? 1. Barr RG, Ferraioli G, Palmeri ML, et al. Elastography Assessment of Liver
Fibrosis: Society of Radiologists in Ultrasound Consensus Conference State-
2. Should we measure in more than one location? ment. Ultrasound Q 2016;32(2):94–107.
3. What are appropriate tissue-mimicking phantom materi- 2. Barr RG, Ferraioli G, Palmeri ML, et al. Elastography Assessment of Liver
als for the liver? Fibrosis: Society of Radiologists in Ultrasound Consensus Conference State-
ment. Radiology 2015;276(3):845–861.
4. Will liver dispersion be helpful in evaluating inflamma- 3. Ferraioli G, Wong VW, Castera L, et al. Liver Ultrasound Elastography: An
tion and/or steatosis? Update to the World Federation for Ultrasound in Medicine and Biology
Guidelines and Recommendations. Ultrasound Med Biol 2018;44(12):2419–
2440.
Clinical Questions 4. Ferraioli G, Filice C, Castera L, et al. WFUMB guidelines and recommen-
1. How different are cut-offs depending on the cause of dations for clinical use of ultrasound elastography: Part 3: liver. Ultrasound
chronic liver disease? Med Biol 2015;41(5):1161–1179.
5. Dietrich CF, Bamber J, Berzigotti A, et al. EFSUMB Guidelines and Recom-
2. How can US elastography complement hepatic venous mendations on the Clinical Use of Liver Ultrasound Elastography, Update
pressure measurements in the assessment of portal hyper- 2017 (Long Version). Ultraschall Med 2017;38(4):e16–e47.
tension and in the assessment of changes in portal venous 6. Shiina T, Nightingale KR, Palmeri ML, et al. WFUMB guidelines and
recommendations for clinical use of ultrasound elastography: Part 1: basic
pressure in patients with liver disease? principles and terminology. Ultrasound Med Biol 2015;41(5):1126–1147.
3. Inflammation and congestion are important processes to 7. Ellis EL, Mann DA. Clinical evidence for the regression of liver fibrosis. J
document in the evolution of liver disease. Histologic as- Hepatol 2012;56(5):1171–1180 [Published correction appears in J Hepatol
2014;60(2):468–469.]
sessment of biopsy specimens can only be used to identify 8. Marcellin P, Gane E, Buti M, et al. Regression of cirrhosis during treatment
the cellular component of inflammation and is essentially with tenofovir disoproxil fumarate for chronic hepatitis B: a 5-year open-label
blind to the fluid component. Quantitative elastography, follow-up study. Lancet 2013;381(9865):468–475.
9. Zoubek ME, Trautwein C, Strnad P. Reversal of liver fibrosis: From fiction
conversely, seems to be sensitive to the effects of the fluid to reality. Best Pract Res Clin Gastroenterol 2017;31(2):129–141.
component of inflammation. How can this capability be 10. Augustin S, Pons M, Maurice JB, et al. Expanding the Baveno VI criteria
exploited for diagnostic purposes? for the screening of varices in patients with compensated advanced chronic
liver disease. Hepatology 2017;66(6):1980–1988.
4. Can we use elastography and measures of loss modulus 11. de Franchis R; Baveno VI Faculty. Expanding consensus in portal hyperten-
to differentiate nonalcoholic or alcoholic steatohepati- sion: Report of the Baveno VI Consensus Workshop: Stratifying risk and
tis from simple steatosis? individualizing care for portal hypertension. J Hepatol 2015;63(3):743–752.
12. Ripoll C, Groszmann R, Garcia-Tsao G, et al. Hepatic venous pressure gradi-
ent predicts clinical decompensation in patients with compensated cirrhosis.
Follow-up of Patients Gastroenterology 2007;133(2):481–488.
1. What is a minimal clinically important difference in stiff- 13. European Association for Study of Liver; Asociacion Latinoamericana para
el Estudio del Higado. EASL-ALEH Clinical Practice Guidelines: Non-
ness measurements over time? How often should these invasive tests for evaluation of liver disease severity and prognosis. J Hepatol
measures be obtained? 2015;63(1):237–264.

10 radiology.rsna.org  n  Radiology: Volume 00: Number 0— 2020

 Barr et al

14. European Association for the Study of the Liver. EASL 2017 Clinical Prac- 38. Pons M, Simón-Talero M, Millán L, et al. Basal values and changes of liver
tice Guidelines on the management of hepatitis B virus infection. J Hepatol stiffness predict the risk of disease progression in compensated advanced
2017;67(2):370–398. chronic liver disease. Dig Liver Dis 2016;48(10):1214–1219.
15. Goodman ZD. Grading and staging systems for inflammation and fibrosis 39. Taniguchi T, Ohtani T, Kioka H, et al. Liver Stiffness Reflecting Right-Sided
in chronic liver diseases. J Hepatol 2007;47(4):598–607. Filling Pressure Can Predict Adverse Outcomes in Patients With Heart
16. Pavlides M, Birks J, Fryer E, et al. Interobserver Variability in Histologic Failure. JACC Cardiovasc Imaging 2019;12(6):955–964.
Evaluation of Liver Fibrosis Using Categorical and Quantitative Scores. Am 40. Bachofner JA, Valli PV, Kröger A, et al. Direct antiviral agent treatment of
J Clin Pathol 2017;147(4):364–369. chronic hepatitis C results in rapid regression of transient elastography and
17. Singal AG, Lim JK, Kanwal F. AGA Clinical Practice Update on Inter- fibrosis markers fibrosis-4 score and aspartate aminotransferase-platelet ratio
action Between Oral Direct-Acting Antivirals for Chronic Hepatitis C index. Liver Int 2017;37(3):369–376.
Infection and Hepatocellular Carcinoma: Expert Review. Gastroenterology 41. Jansen C, Bogs C, Verlinden W, et al. Shear-wave elastography of the liver
2019;156(8):2149–2157. and spleen identifies clinically significant portal hypertension: A prospective
18. Jakab SS, Garcia-Tsao G. Screening and Surveillance of Varices in Patients multicentre study. Liver Int 2017;37(3):396–405.
With Cirrhosis. Clin Gastroenterol Hepatol 2019;17(1):26–29 [Published 42. Jansen C, Möller P, Meyer C, et al. Increase in liver stiffness after transjugular
correction appears in Clin Gastroenterol Hepatol 2019;17(5):1009.]. intrahepatic portosystemic shunt is associated with inflammation and predicts
19. Younossi ZM, Koenig AB, Abdelatif D, Fazel Y, Henry L, Wymer M. Global mortality. Hepatology 2018;67(4):1472–1484.
epidemiology of nonalcoholic fatty liver disease-Meta-analytic assessment of 43. Berzigotti A. Non-invasive evaluation of portal hypertension using ultrasound
prevalence, incidence, and outcomes. Hepatology 2016;64(1):73–84. elastography. J Hepatol 2017;67(2):399–411.
20. Nobili V, Carter-Kent C, Feldstein AE. The role of lifestyle changes in the 44. Vizzutti F, Arena U, Romanelli RG, et al. Liver stiffness measurement
management of chronic liver disease. BMC Med 2011;9(1):70. predicts severe portal hypertension in patients with HCV-related cirrhosis.
21. Schuppan D, Surabattula R, Wang XY. Determinants of fibrosis progression Hepatology 2007;45(5):1290–1297.
and regression in NASH. J Hepatol 2018;68(2):238–250. 45. Garcia-Tsao G, Abraldes JG, Berzigotti A, Bosch J. Portal hypertensive
22. Ferraioli G, Tinelli C, Zicchetti M, et al. Reproducibility of real-time bleeding in cirrhosis: Risk stratification, diagnosis, and management: 2016
shear wave elastography in the evaluation of liver elasticity. Eur J Radiol practice guidance by the American Association for the study of liver diseases.
2012;81(11):3102–3106. Hepatology 2017;65(1):310–335.
23. Hudson JM, Milot L, Parry C, Williams R, Burns PN. Inter- and intra- 46. Colecchia A, Montrone L, Scaioli E, et al. Measurement of spleen stiffness to
operator reliability and repeatability of shear wave elastography in the liver: evaluate portal hypertension and the presence of esophageal varices in patients
a study in healthy volunteers. Ultrasound Med Biol 2013;39(6):950–955. with HCV-related cirrhosis. Gastroenterology 2012;143(3):646–654.
24. Ferraioli G, Tinelli C, Lissandrin R, et al. Point shear wave elastography 47. Mejias M, Garcia-Pras E, Gallego J, Mendez R, Bosch J, Fernandez M. Rel-
method for assessing liver stiffness. World J Gastroenterol 2014;20(16): evance of the mTOR signaling pathway in the pathophysiology of splenomegaly
4787–4796. in rats with chronic portal hypertension. J Hepatol 2010;52(4):529–539.
25. Fang C, Jaffer OS, Yusuf GT, et al. Reducing the Number of Measurements 48. Manatsathit W, Samant H, Kapur S, et al. Accuracy of liver stiffness,
in Liver Point Shear-Wave Elastography: Factors that Influence the Number spleen stiffness, and LS-spleen diameter to platelet ratio score in detection
and Reliability of Measurements in Assessment of Liver Fibrosis in Clinical of esophageal varices: Systemic review and meta-analysis. J Gastroenterol
Practice. Radiology 2018;287(3):844–852. Hepatol 2018;33(10):1696–1706.
26. Ferraioli G, De Silvestri A, Reiberger T, et al. Adherence to quality criteria 49. Ma X, Wang L, Wu H, et al. Spleen Stiffness Is Superior to Liver Stiffness
improves concordance between transient elastography and ElastPQ for for Predicting Esophageal Varices in Chronic Liver Disease: A Meta-Analysis.
liver stiffness assessment-A multicenter retrospective study. Dig Liver Dis PLoS One 2016;11(11):e0165786.
2018;50(10):1056–1061. 50. Sharma P, Kirnake V, Tyagi P, et al. Spleen stiffness in patients with cirrhosis
27. Ferraioli G, Maiocchi L, Lissandrin R, et al. Accuracy of the ElastPQ in predicting esophageal varices. Am J Gastroenterol 2013;108(7):1101–1107.
Technique for the Assessment of Liver Fibrosis in Patients with Chronic 51. Balakrishnan M, Souza F, Muñoz C, et al. Liver and Spleen Stiffness
Hepatitis C: a “Real Life” Single Center Study. J Gastrointestin Liver Dis Measurements by Point Shear Wave Elastography via Acoustic Radiation
2016;25(3):331–335. Force Impulse: Intraobserver and Interobserver Variability and Predictors of
28. Roccarina D, Iogna Prat L, Buzzetti E, et al. Establishing Reliability Criteria Variability in a US Population. J Ultrasound Med 2016;35(11):2373–2380.
for Liver ElastPQ Shear Wave Elastography (ElastPQ-SWE): Comparison 52. Cho YS, Lim S, Kim Y, Sohn JH, Jeong JY. Spleen Stiffness Measurement Using
Between 10, 5 and 3 Measurements. Ultraschall Med 2019 Oct 8 [Epub 2-Dimensional Shear Wave Elastography: The Predictors of Measurability and
ahead of print]. the Normal Spleen Stiffness Value. J Ultrasound Med 2019;38(2):423–431.
29. Hall TJ, Milkowski A, Garra B, et al. RSNA/QIBA: Shear wave speed as 53. Procopet B, Berzigotti A, Abraldes JG, et al. Real-time shear-wave elastog-
a biomarker for liver fibrosis staging. Ultrasonics Symposium (IUS), 2013. raphy: applicability, reliability and accuracy for clinically significant portal
2013; 397–400. hypertension. J Hepatol 2015;62(5):1068–1075.
30. Palmeri M, Nightingale K, Fielding S, et al. RSNA QIBA ultrasound shear 54. Lucchina N, Recaldini C, Macchi M, et al. Point Shear Wave Elastography
wave speed Phase II phantom study in viscoeastic media. In: 2015 IEEE of the Spleen: Its Role in Patients with Portal Hypertension. Ultrasound
Ultrasonics Symosium International, 2015. Med Biol 2018;44(4):771–778.
31. Ferraioli G, De Silvestri A, Lissandrin R, et al. Evaluation of Inter-System Vari- 55. Ferraioli G, Tinelli C, Lissandrin R, et al. Ultrasound point shear wave
ability in Liver Stiffness Measurements. Ultraschall Med 2019;40(1):64–75. elastography assessment of liver and spleen stiffness: effect of training on
32. Morishita N, Hiramatsu N, Oze T, et al. Liver stiffness measurement by repeatability of measurements. Eur Radiol 2014;24(6):1283–1289.
acoustic radiation force impulse is useful in predicting the presence of 56. Kjærgaard M, Thiele M, Jansen C, et al. High risk of misinterpreting liver
esophageal varices or high-risk esophageal varices among patients with and spleen stiffness using 2D shear-wave and transient elastography after a
HCV-related cirrhosis. J Gastroenterol 2014;49(7):1175–1182 [Published moderate or high calorie meal. PLoS One 2017;12(4):e0173992.
correction appears in J Gastroenterol 2015;50(6):705.]. 57. Karlas T, Lindner F, Tröltzsch M, Keim V. Assessment of spleen stiffness
33. Park Y, Kim SU, Park SY, et al. A novel model to predict esophageal varices using acoustic radiation force impulse imaging (ARFI): definition of ex-
in patients with compensated cirrhosis using acoustic radiation force impulse amination standards and impact of breathing maneuvers. Ultraschall Med
elastography. PLoS One 2015;10(3):e0121009. 2014;35(1):38–43.
34. Attia D, Schoenemeier B, Rodt T, et al. Evaluation of Liver and Spleen 58. Takuma Y, Nouso K, Morimoto Y, et al. Portal Hypertension in Patients
Stiffness with Acoustic Radiation Force Impulse Quantification Elastography with Liver Cirrhosis: Diagnostic Accuracy of Spleen Stiffness. Radiology
for Diagnosing Clinically Significant Portal Hypertension. Ultraschall Med 2016;279(2):609–619.
2015;36(6):603–610. 59. Colecchia A, Ravaioli F, Marasco G, et al. A combined model based on spleen
35. Jansen C, Bogs C, Verlinden W, et al. Algorithm to rule out clinically stiffness measurement and Baveno VI criteria to rule out high-risk varices in
significant portal hypertension combining Shear-wave elastography of liver advanced chronic liver disease. J Hepatol 2018;69(2):308–317.
and spleen: a prospective multicentre study. Gut 2016;65(6):1057–1058. 60. Elkrief L, Ronot M, Andrade F, et al. Non-invasive evaluation of portal
36. Zeng J, Zheng J, Jin JY, et al. Shear wave elastography for liver fibrosis in hypertension using shear-wave elastography: analysis of two algorithms
chronic hepatitis B: Adapting the cut-offs to alanine aminotransferase levels combining liver and spleen stiffness in 191 patients with cirrhosis. Aliment
improves accuracy. Eur Radiol 2019;29(2):857–865. Pharmacol Ther 2018;47(5):621–630.
37. Vergniol J, Boursier J, Coutzac C, et al. Evolution of noninvasive tests of 61. Verlinden W, Francque S, Michielsen P, Vanwolleghem T. Successful anti-
liver fibrosis is associated with prognosis in patients with chronic hepatitis viral treatment of chronic hepatitis C leads to a rapid decline of liver stiffness
C. Hepatology 2014;60(1):65–76. without an early effect on spleen stiffness. Hepatology 2016;64(5):1809–1810.

Radiology: Volume 00: Number 0— 2020  n  radiology.rsna.org 11

Update to the Society of Radiologists in Ultrasound Liver Elastography Consensus Statement

62. Anderson EL, Howe LD, Jones HE, Higgins JP, Lawlor DA, Fraser A. The 71. Jeon SK, Lee JM, Joo I, et al. Prospective Evaluation of Hepatic Steatosis
Prevalence of Non-Alcoholic Fatty Liver Disease in Children and Adolescents: Using Ultrasound Attenuation Imaging in Patients with Chronic Liver
A Systematic Review and Meta-Analysis. PLoS One 2015;10(10):e0140908. Disease with Magnetic Resonance Imaging Proton Density Fat Fraction
63. Xanthakos SA, Jenkins TM, Kleiner DE, et al. High Prevalence of Nonal- as the Reference Standard. Ultrasound Med Biol 2019;45(6):1407–1416.
coholic Fatty Liver Disease in Adolescents Undergoing Bariatric Surgery. 72. Bae JS, Lee DH, Lee JY, et al. Assessment of hepatic steatosis by using at-
Gastroenterology 2015;149(3):623–634.e8. tenuation imaging: a quantitative, easy-to-perform ultrasound technique.
64. Schwimmer JB, Newton KP, Awai HI, et al. Paediatric gastroenterology Eur Radiol 2019;29(12):6499–6507.
evaluation of overweight and obese children referred from primary care 73. Tada T, Iijima H, Kobayashi N, et al. Usefulness of Attenuation Imaging with
for suspected non-alcoholic fatty liver disease. Aliment Pharmacol Ther an Ultrasound Scanner for the Evaluation of Hepatic Steatosis. Ultrasound
2013;38(10):1267–1277. Med Biol 2019;45(10):2679–2687.
65. Dietrich C, Sirli R, Ferraioli G, et al. Current Knowledge in Ultrasound-Based 74. Bouchet P, Gennisson JL, Podda A, Alilet M, Carrié M, Aubry S. Artifacts
Liver Elastography of Pediatric Patients. Appl Sci (Basel) 2018;8(6):944. and Technical Restrictions in 2D Shear Wave Elastography. Ultraschall Med
66. Eiler J, Kleinholdermann U, Albers D, et al. Standard value of ultrasound 2018 Dec 21 [Epub ahead of print].
elastography using acoustic radiation force impulse imaging (ARFI) in healthy 75. Bruce M, Kolokythas O, Ferraioli G, Filice C, O’Donnell M. Limitations
liver tissue of children and adolescents. Ultraschall Med 2012;33(5):474–479. and artifacts in shear-wave elastography of the liver. Biomed Eng Lett
67. Hanquinet S, Courvoisier D, Kanavaki A, Dhouib A, Anooshiravani M. 2017;7(2):81–89.
Acoustic radiation force impulse imaging-normal values of liver stiffness in 76. Dubinsky TJ, Shah HU, Erpelding TN, Sannananja B, Sonneborn R,
healthy children. Pediatr Radiol 2013;43(5):539–544. Zhang M. Propagation Imaging in the Demonstration of Common
68. Matos H, Trindade A, Noruegas MJ. Acoustic radiation force impulse imag- Shear Wave Artifacts. J Ultrasound Med 2019;38(6):1611–1616.
ing in paediatric patients: normal liver values. J Pediatr Gastroenterol Nutr 77. Shin HJ, Kim MJ, Yoon CS, et al. Motion effects on the measurement
2014;59(6):684–688. of stiffness on ultrasound shear wave elastography: a moving liver fibrosis
69. Ferraioli G, Maiocchi L, Raciti MV, et al. Detection of Liver Steatosis With phantom study. Med Ultrason 2018;1(1):14–20.
a Novel Ultrasound-Based Technique: A Pilot Study Using MRI-Derived 78. Ormachea J, Parker KJ, Barr RG. An initial study of complete 2D shear
Proton Density Fat Fraction as the Gold Standard. Clin Transl Gastroenterol wave dispersion images using a reverberant shear wave field. Phys Med Biol
2019;10(10):e00081. 2019;64(14):145009.
70. Dioguardi Burgio M, Imbault M, Ronot M, et al. Ultrasonic Adaptive Sound
Speed Estimation for the Diagnosis and Quantification of Hepatic Steatosis:
A Pilot Study. Ultraschall Med 2019;40(6):722–733.

12 radiology.rsna.org  n  Radiology: Volume 00: Number 0— 2020