Point Shear Wave Elastography by ElastPQ for Fibrosis Screening in Patients with NAFLD: A Prospective, Multicenter Comparison to Vibration-Controlled Elastography

Keywords

Abdomen - Gastrointestinal Tract - Point of care - Physics and Technology

Introduction

The prevalence of nonalcoholic fatty liver disease (NAFLD) in the general population is 20–30% and may be up to 70% in the obese population and 90% in individuals with diabetes mellitus [1] [2] [3]. Since epidemiological data show a further increase in the prevalence of obesity and diabetes mellitus, the number of NAFLD patients is also expected to increase in the forthcoming years [4]. NAFLD progresses to nonalcoholic steatohepatitis (NASH) in 25% of cases within 3 years, and NASH already accounts for 12% of all liver transplantations in Europe [3]. Fibrosis, more than steatosis and inflammatory activity, is associated with portal hypertension and subsequent adverse long-term outcomes, such as hepatocellular carcinoma (HCC) in NAFLD [2] [5] [6]. Thus, reliable noninvasive techniques for the assessment of liver fibrosis risk are of increasing clinical importance. Simple laboratory-based scores (such as the FIB-4 score) can be used to identify patients at risk for fibrosis in low-prevalence populations. However, a sequential algorithm of laboratory-based scores followed by liver elastography techniques is suggested to identify patients with NAFLD-associated fibrosis [7].

Liver biopsy is the gold standard for diagnosing and staging liver fibrosis and can be performed via the transcutaneous or transjugular route [9, 8]. However, it is an imperfect gold standard [8] and associated with periprocedural risks and costs. Ultrasound-based techniques for estimating liver fibrosis offer the advantages of lower-cost, low-risk, quick, and point-of-care assessment. Vibration-controlled transient elastography (VCTE) is a well-validated method of assessing liver fibrosis [7] [9]. However, it requires a dedicated device that cannot perform conventional ultrasound needed in patients with chronic liver disease to screen for HCC and other liver-related complications. In contrast, acoustic radiation force impulse (ARFI)-based techniques, either point shear wave elastography (pSWE) or two-dimensional SWE, can be included in regular modern ultrasound devices. The diagnostic accuracy and cutoffs for VCTE for different fibrosis stages, portal hypertension, and esophageal varices have been established in multiple liver disease etiologies, including NAFLD [6] [7]. The use of VCTE in NAFLD has been recommended in national and international guidelines [10]. However, VTCE also has relevant limitations: despite the introduction of the XL-probe, failed and unreliable measurements in obese patients remain a problem [11] and measurements cannot be reliably performed in the presence of perihepatic ascites.

pSWE-Elast-PQ (EPQ) is the pSWE technique implemented in Philips ultrasound systems. Its use for liver fibrosis assessment has been validated in various etiologies [12] [13]. In contrast to VCTE, the region of interest (ROI) placement and, therefore, the depth of measurement can be chosen on the B-mode ultrasound image, which allows for more advantageous localization (e.g., avoiding larger blood vessels or other interfering structures). Since abdominal ultrasound devices are needed to screen for liver-related complications, pSWE is potentially more widely and readily available than VCTE. However, reliable cutoffs for ruling in or ruling out significant fibrosis and advanced fibrosis in patients with NAFLD have not yet been sufficiently established.

Therefore, we set up a prospective European multicenter study to assess the performance of EPQ for ruling out significant fibrosis and ruling in advanced fibrosis in patients with (suspected) NAFLD using VCTE as a reference standard, exploring factors, which might influence liver stiffness as measured by EPQ and to provide a clinical algorithm comprised of elastography and the laboratory-based noninvasive FIB-4 tests [14].

Patients and methods

Study centers and study population

Patients with suspected or diagnosed NAFLD were prospectively recruited at 4 European centers. We included patients for whom the clinical diagnosis of NAFLD was made and who underwent ultrasound evaluation and elastography to assess fibrosis and steatosis. Patients with viral liver disease (as assessed by HBsAg and HCVAb), alcohol intake, hepatic malignancies, and heart failure were excluded.

Elastography protocol

We consecutively performed liver stiffness measurements (LSM) with pSWE and VTCE on the same day in all included patients. For the pSWE, we used the EPQ module of the Philips EPIQ 7 ultrasound system (Philips Medical System, The Netherlands) with a C5–1 convex probe, and for VCTE, either the FibroScan 502 Touch or Mini 430 (Echosens, Paris, FR) with the M- or XL-Probe, as suggested by the probe selection tool. Patients fasted for at least four hours before the examination. As recommended, both measurements were taken with the patient in a supine position, with the right arm abducted [9]. Both VCTE and EPQ examinations were performed via an intercostal space on the upper right lobe of the liver. Measurements with EPQ and VCTE were performed following guideline recommendations [7]. We considered VCTE and EPQ LSM with an IQR/median ratio ≤0.3 reliable and included VCTE measures < 7.1 kPa irrespective of IQR/median [15] [16].

Statistical analysis

The mode of display of the given variables and the statistical tests used are described in the supplementary material.

Definition of significant and advanced fibrosis and cirrhosis

As VCTE was used as a diagnostic reference standard for risk assessment, published cutoffs [17] [18] [19] were used to differentiate the ≥F2/significant fibrosis group, the ≥F3/advanced fibrosis group, and the F4/cirrhosis group. These cutoffs were > 7 kPa for ≥F2/significant fibrosis, ≥10 kPa for ≥F3/advanced fibrosis and ˃ 15 kPa for F4/cirrhosis. Using the grouping described above, the area under the receiver operator characteristic curve (AUROC) and the respective 95% CI for the ability of EPQ to determine the fibrosis stage defined by VCTE were calculated. To establish cutoffs for the differentiation between fibrosis stages, the Youden Index [20] was used to derive optimal cutoffs. With this method, the optimal cutoff is chosen to maximize the sum of sensitivity (sens.) and specificity (spec.). Additional rule-in and rule-out cutoffs were chosen to maximize sensitivity and specificity. For each cutoff, the sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), as well as the false positives (FP) and false negatives (FN) for the given population were derived. To further elucidate the fibrosis stage of patients in whom significant fibrosis could not be ruled out and advanced fibrosis could not be ruled in using EPQ, published cutoffs of the FIB-4 score were used [14]. To further elucidate the factors impacting the absolute difference between paired measurements of VCTE and EPQ, we performed univariate linear modelling and multivariate modelling comprising all univariately significant variables and another model of the same kind exploring the impact of factors additional to VCTE-based liver stiffness on EPQ-liver stiffness.

Results

In total, 353 concomitant, paired VCTE, and EPQ-based liver stiffness measurements were obtained in the four participating centers. Of these, 21 (6.0%) were excluded, as they did not fulfill the reliability criterion of IQR/median ≤0.3% for either VCTE or EPQ (see Supplementary Table-ST1). Consequently, 332 (94.1%) patients were included in the analysis. Among these, 222 (66.9%) had no significant fibrosis (F0–1), 41 (12.4%) had significant fibrosis (F2), 30 (9.0%) were diagnosed with advanced fibrosis (≥F3), and 39 (11.8%) were diagnosed with cirrhosis (F4). A summary of the included patients, the calculated cutoffs, and the resulting fibrosis staging is shown in [Fig. 1]A.

Fig. 1 A Study flow chat, showing the fibrosis staging by VCTE and EPQ, as well as resulting cutoffs. B Reclassification of patients without rule-out of significant fibrosis and without rule-in of advanced fibrosis, using the FIB-4 index.

The population defining characteristics for all patients, patients without significant fibrosis, patients with significant and advanced fibrosis, and cirrhosis patients are displayed in [Table 1] and the patient characteristics by center in Supplementary Table-ST2. The patient populations were heterogeneous between centers, differing in age, sex distribution, BMI, markers of disease severity, such as laboratory values and liver stiffness, steatosis grade, and liver stiffness. The included patients were 59 [IQR: 16.5] years old and predominantly males. The median BMI was 29 [IQR: 7] kg/m² and did not increase with the fibrosis stage. The blood work was mostly obtained on the same day as the elastography [IQR: 1 day]. Platelet count, cholesterol, albumin, and the CAP decreased, while the international normalized ratio (INR), alkaline phosphatase (ALP), and aspartate aminotransferase (AST), gamma-glutamyltransferase (GGT), and FIB-4 increased with each fibrosis stage. Liver stiffness as measured with EPQ significantly increased with the fibrosis stage as defined by VCTE ([Table 1]).

Table 1 Patient characteristics.

		All NAFLD N= 332 (100%)	No/Mild Fibrosis F0/F1 N= 222 (66.9%)	Significant Fibrosis F2 N= 41 (12.3%)	Severe Fibrosis F3 N= 30 (9.0%)	Cirrhosis F4 N= 39 (11.7%)	P-value
P-values were calculated using the Kruskal-Wallis test comparing the values of patients with no significant fibrosis, F2, F3, and F4. Significant p-values are bold. Abbreviations ALT: alanine aminotransferase; AST: aspartate aminotransferase; CAP: continuous attenuation parameter; EPQ: Elast PQ;F0/1: fibrosis stage 0 or 1; F2: fibrosis stage 2; F3: fibrosis stage 3; F4: fibrosis stage 4; GGT: gamma glutamyl transferase; INR: international normalized ratio; IQR: interquartile range; IQR: interquartile range; kPa: kilopascal; M: median; N: number; VCTE: vibration-controlled transient elastography
Age, years, median [IQR]		59.0 [16.5]	56.5 [19.0]	59.0 [12]	59.5 [16.0]	63.0 [10.5]	0.001
Sex, N (%)
	Female	137 (41.3)	92 (41.4)	17 (41.5)	12 (40.0)	16 (41.0)	0.999
	Male	195 (58.7)	130 (58.6)	24 (58.5)	18 (60.0)	23 (59.0)
Center, N (%)
	Pavia	27 (8.1)	24 (10.8)	1 (2.4)	0 (0.0)	2 (5.1)	<0.001
	Timișoara	57 (17.2)	39 (17.6)	6 (14.6)	8 (26.7)	4 (10.3)
	Vienna	112 (33.7)	58 (26.1)	15 (36.6)	12 (40.0)	27 (69.2)
	Zagreb	136 (41.0)	101 (45.5)	19 (46.3)	10 (33.3)	6 (15.4)
BMI, kg/m2, median [IQR]		29.0 [7.1]	28.6 [6.3]	30.7 [10.9]	26.9 [6.0]	31.2 [8.9]	0.025
Platelets, G/L, mean (SD)		224 (75)	240 (70)	228 (64)	200 (62)	149 (72)	<0.001
INR, median [IQR]		1.0 [0.2]	1.0 [0.1]	1.0 [0.2]	1.1 [0.3]	1.0 [0.20]	<0.001
Bilirubin, mg/dL, median [IQR]		0.76 [0.42]	0.76 [0.37]	0.83 [0.49]	0.76 [0.25]	0.77 [0.54]	0.913
Albumin, g/dL, median [IQR]		41.9 [8.1]	43.1 [8.4]	40.5 [7.1]	40.75 [6.9]	39.2 [5.7]	0.027
AST, U/L, median [IQR]		30 [21]	27 [18]	29 [34]	38 [35]	45 [22]	<0.001
ALT, U/L, median [IQR]		35 [38]	35 [39]	32 [33]	44 [38]	33 [22]	0.275
GGT, U/L, median [IQR]		50 [74]	40 [60]	64 [68]	73 [60]	112 [193]	<0.001
FIB-4, points, median [IQR]		1.33 [1.12]	1.17 [0.75]	1.36 [1.05]	1.81 [1.68]	3.62 [3.27]	<0.001
VCTE, kPa, median [IQR]		6 [4.33]	5 [1.8]	8.4 [1.3]	12.7 [3.1]	22.8 [16.0]	<0.001
VCTE IQR/median, %, median [IQR]		15.0 [10.0]	15.0 [10.0]	16.0 [11.0]	14.5 [7.2]	17.5 [11.6]	0.284
CAP, dB/m, median [IQR]		298 [78]	298 [73]	305 [89]	299 [70]	272 [122]	0.42
CAP IQR, dB/m, median [IQR]		29 [20]	30 [19]	29 [10]	30 [12]	30 [34]	0.855
EPQ, kPa, median [IQR]		5.2 [2.8]	4.6 [1.5]	7.0 [3.1]	9.4 [4.3]	13.6 [9.2]	<0.001
EPQ IQR, kPa, median [IQR]		1.23 [0.02]	1.24 [0.51]	1.53 [0.33]	1.77 [0.42]	2.13 [0.68]	<0.001
EPQ IQR/median, %, median [IQR]		22.9 [10.4]	22.0 [10.0]	24.6 [9.6]	25.0 [12.6]	26.0 [7.3]	0.033

The correlation between reliable measurements of VCTE and EPQ was high, with a Pearson’s R = 0.87 (95% CI: 0.83–0.89, p < 0.0001). This is shown in a scatterplot of the reliable measurements of VCTE and EPQ in [Fig. 2].

Fig. 2 Pearson correlation of reliable liver stiffness measurements by VCTE and EPQ, with CI 95% shown, on a log-10 scale.

The concordance between VCTE and EPQ was high, as shown by Lin's CCC = 0.792 (95% CI: 0.623–0.889). To display the effect of increasing liver stiffness on the concordance of the two compared methods, a Bland-Altmann-Leh plot is provided in [Fig. 3]. It shows that the numeric difference between absolute liver stiffness as measured by VCTE and EPQ increased with increasing liver stiffness leading to a converse monotonous decrease in the Pearson’s correlation coefficient ([Fig. 4]). Multivariate analysis to predict the absolute difference in liver stiffness as measured by EPQ and VCTE revealed that higher VCTE, EPQ-based liver stiffness, BMI and EPQ IQR are associated with a larger absolute difference in measured liver stiffness (compare Supplementary Table-ST3). A second analysis aimed at factors that might impact EPQ-based liver stiffness, in addition to VCTE-based liver stiffness, showed an association of VCTE IQR and probe with EPQ-based liver stiffness (see Supplementary Table-ST4).

Fig. 3 Bland-Altman-Leh plot comparing the difference between VCTE and EPQ measurements for a given mean of a measurement with lines of average difference and two 95% CI lines, with the Lin’s concordance coefficient shown on a pseudo-log10 scale.

Fig. 4 Correlation of VCTE with EPQ at different VCTE liver stiffnesses. The figure shows the Pearson correlation coefficient of the LSM by VCTE and EPQ (y-axis) at the liver stiffness as determined by VCTE +/- 10 kPa (x-axis). This does not indicate the correlation at this point but illustrates how increasing liver stiffness affects the correlation.

To assess the accuracy of EPQ for fibrosis risk assessment in NAFLD, we calculated the AUROC values for the detection of significant fibrosis and advanced fibrosis, as well as respective rule-in and rule-out cutoffs. EPQ can identify significant fibrosis with an AUROC of 0.94 (95% CI: 0.910–0.969,). The optimal (Youden) cutoff was determined at ≥6.5 kPa (≥1.47 m/s; sens.: 80%, spec.: 95%), the specific rule-in cutoff at ≥6.8 kPa (≥1.51 m/s; sens.: 77%, spec.: 97%), and the sensitive rule-out cutoff at <6 kPa (<1.41 m/s; sens.: 86%, spec.: 89%). The AUROC for advanced fibrosis was 0.949 (95% CI: 0.919–0.979). The optimal cutoff was ≥6.9 kPa (≥1.52 m/s; sens.: 88%, spec.: 89%), while the respective rule-in and rule-out cutoffs were ≥10.9 kPa (≥1.91 m/s; sens.: 61%, spec.: 99%) and <7 kPa (<1.53 m/s; sens.: 86%, spec.: 90%). EPQ detected cirrhosis with an AUROC of 0.949 (95% CI: 0.91–0.989). The optimal rule-in and rule-out cutoffs were ≥10.4 kPa (≥1.86 m/s; sens.: 87%, spec.: 94%), ≥15.6 kPa (≥2.28 m/s; sens.: 36%, spec.: 99%), and <8 kPa (<1.63 m/s; sens.: 90%, spec.: 90%), respectively. These cutoffs and the corresponding measures are summarized in [Table 2].

Table 2 Elast-PQ cutoffs.

Stage	Cutoff	AUROC	EPQ cutoff, stiffness (kPa)	EPQ cutoff, shear wave (m/s)	Sensitivity, %	Specificity, %	PPV, %	NPV, %	FP, N	FN, N
EPQ cutoffs for the diagnosis of significant fibrosis (≥F2), advanced fibrosis (≥F3), and cirrhosis (F4). Abbreviations: AUROC: area under the receiver operator characteristics; F2: fibrosis stage 2; F3: fibrosis stage 3; F4: fibrosis stage 4; FN: false negative; FP: false positive; kPa: kilopascal; NPV: negative predictive value; PPV: positive predictive value
≥F2	Youden	0.940 (95%CI: 0.910–0.969, Power: 1)	≥6.5	≥1.47	80	95	90	91	10	21
	Rule-in		≥6.8	≥1.51	77	97	92	89	7	25
	Rule-out		<6.0	<1.41	86	89	80	92	24	16
≥F3	Youden	0.949 (95%CI: 0.919–0.979, Power: 1)	≥6.9	≥1.52	88	89	69	97	28	8
	Rule-in		≥10.9	≥1.91	60	99	93	96	3	27
	Rule-out		<7.0	<1.53	86	90	71	96	25	9
F4	Youden	0.949 (95%CI: 0.91–0.989, Power: 1)	≥10.4	≥1.86	87	94	65	98	18	5
	Rule-in		≥15.6	≥2.28	36	99	93	92	1	25
	Rule-out		<8	<1.63	90	90	54	99	30	4

Using the established EPQ cutoffs, we could not rule-out significant fibrosis and could not rule-in advanced fibrosis in some patients. Here we applied a FIB-4 cutoff of ˂ 1.45 to rule-out significant fibrosis and a cutoff of 3.45 to rule in advanced fibrosis [14].The FIB-4 (available in 69/73, 94.5%) allowed a further stratification of patients in the EPQ gray zone, since significant fibrosis could be ruled out in 32 (46.4%) patients, and advanced fibrosis could be ruled in in 8 (11.6%). Finally, after considering EPQ and FIB-4 results, only 33 (9.9%) patients remained in the indeterminate "gray zone" (Fig-1B).

Discussion

While different noninvasive tools for the diagnosis and staging of NAFLD are available, ultrasound-based methods are a very attractive screening option since the first suspicion of NAFLD is often based on detecting a hyperechogenic liver texture on abdominal ultrasound. Subsequent pSWE of the liver – readily available in mid-and high-end ultrasound devices – could represent a valuable one-stop assessment for diagnosing hepatic steatosis and further risk stratification according to fibrosis grade. It is crucial to identify significant liver fibrosis and cirrhosis as early as possible to initiate effective measures to prevent disease progression and HCC screening. Furthermore, pSWE may allow for noninvasive, longitudinal monitoring of the effects of lifestyle modification and potential future medical therapies for NASH in the academic and routine clinical setting.

As of today, among all the available pSWE techniques, only Virtual Touch Quantification (VTQ) has been extensively evaluated for use in NAFLD/NASH: A 2015 ARFI meta-analysis [23] reported only moderate sensitivity (80.2%) and specificity (85.2%). pSWE is not recommended as an endpoint in patients with NAFLD by the 2017 update to the EFSUMB guidelines [9], while the 2018 update to the WFUMB guidelines on the use of liver shear wave elastography recommends SWE – a term that includes VCTE, pSWE, and two-dimensional SWE – for liver stiffness measurements in NAFLD, in particular, to rule-out advanced fibrosis and to select patients for further investigation [7]. A more recent meta-analysis (2020) on pSWE (VTQ) including 1,147 patients reported a better diagnostic value of VTQ for significant fibrosis (≥F2; AUROC: 0.89; sens.: 85%, spec.: 83%) and for cirrhosis (F4; AUROC: 0.94; sens.: 90%, spec.: 95%) in NAFLD patients [24].

EPQ has previously been evaluated for use in other etiologies [13] but only in a small group of NAFLD/NASH patients against liver biopsy [21]. VCTE was found to be superior to EPQ for fibrosis screening, indicating a diagnostic performance for EPQ for significant fibrosis (≥F2) with an AUROC: 0.74 (sens.: 78.1%, spec.: 61.4%) and for cirrhosis (F4) with an AUROC: 0.72 (sens.: 66.7%, spec: 93.2%). However, this study included only a very limited number of patients with F0 and F4 fibrosis, which might have impacted this study's ability to assess the "true" diagnostic accuracy in NAFLD fibrosis staging. Furthermore, the authors state that the histology specimens, on average, did not fulfill recommended standards, introducing potential sampling bias [21]. Therefore, although we applied liver stiffness measurement by VCTE instead of liver biopsy as a diagnostic gold standard, our study provides important results regarding the concordance of EPQ and VCTE in a large cohort of prospectively recruited NAFLD patients from four European centers. While the largest group of patients was classified as having no significant fibrosis, the significant and advanced fibrosis and cirrhosis groups were sufficiently powered, thereby reflecting the clinical spectrum of NAFLD seen at specialized centers.

Furthermore, VCTE is known to sometimes fail in patients with a higher BMI. Because obese patients in whom VCTE failed or delivered unreliable results, despite using the XL-probe, were excluded from the analysis, the present study cannot provide information on the accuracy of EPQ in patients in the higher BMI range.

We found a high correlation between VCTE and EPQ (Pearson's correlation coefficient: 0.87) as well as a high concordance (CCC: 0.792). Therefore, EPQ can correctly identify significant and advanced fibrosis and cirrhosis, as expressed by respective AUROCs of ≥F2: 0.940, ≥F3: 0.949, and F4: 0.949. Optimal Youden's index-derived cutoffs for stages of fibrosis were defined at ≥6.5 kPa (≥1.47 m/s) for significant fibrosis (≥F2), at ≥6.9 kPa (≥1.52 m/s) for advanced fibrosis (≥F3), and at >10.4 kPa (>1.86 m/s) for cirrhosis (F4). To further increase clinical applicability, we also provide specific rule-in and specific rule-out cutoffs for the clinically relevant NAFLD-fibrosis stages. The subsequent use of the FIB-4 score on the 69/73 patients in the EPQ gray zone, where the FIB-4 was available, could rule-out significant fibrosis in 32 (9.6%) patients and rule-in advanced fibrosis in 8 (2.4%), finally leaving only 33 (9.9%) patients with an indeterminate fibrosis stage. This algorithm combining noninvasive EPQ liver stiffness measurement and broadly available blood tests of fibrosis is a valuable tool for noninvasive fibrosis risk discrimination.

In concordance with previous studies [27, 26], we noted an increase in the discrepancy in numerical values between VCTE and EPQ above 20–25 kPa. To explore the factors impacting the discrepancy between VCTE and EPQ, we performed a regression modeling of factors that may impact the absolute difference between VCTE and EPQ. Next to higher VCTE-LSM values, higher BMI and higher IQR of EPQ were significantly associated with a discrepancy between VCTE-LSM vs. EPQ-LSM in multivariate analysis. Still, liver stiffness values above 20 kPa would indicate advanced chronic liver disease (ACLD) using both VCTE and EPQ, and the divergence above this range does not question the ability to detect ACLD. However, since EPQ may not be in concordance with VCTE within high stiffness ranges, useful cutoffs or relevant delta/changes in EPQ stiffness to monitor improvement or worsening of liver cirrhosis/portal hypertension remain to be explored. Importantly, EPQ was still useful for assessing the fibrosis risk in NAFLD and predicting clinical outcomes [22] [25] [26]. Importantly, our cohort was acquired at tertiary centers, as evident from the high proportion of patients with advanced fibrosis and cirrhosis. Therefore, the cutoffs derived here might not be directly applicable to the primary care setting.

VCTE has been found to have lower sensitivity and specificity in NAFLD-associated fibrosis for distinguishing lower fibrosis stages [27]. Subsequently, using VCTE as a reference standard for fibrosis staging may have led to overly optimistic results since VCTE and EPQ are based on the same diagnostic principle (i.e., elastography), which may indicate a higher concordance as compared to using liver biopsy as a reference. Furthermore, fibrosis staging by VCTE is not regarded as the gold standard, and thus, the comparator of VCTE-based LSM represents a limitation of this study. However, by using VCTE as a reference standard, we ensured that the results of EPQ-based LSM could be assessed in the context of VCTE as the currently most widely used liver stiffness measurement technique. In addition, this strategy also allowed us to recruit a larger number of NAFLD patients.

Ultimately, the EPQ cutoffs identified in our study only suggest the risk of having the respective NAFLD-associated fibrosis stages. However, a previous study using histopathological evaluation of liver biopsy – as the (imperfect) diagnostic gold standard – reported similar cutoff values for fibrosis staging as in our study [21], which suggests the validity of our EPQ NAFLD fibrosis stage cutoffs. Importantly, in our diagnostic algorithm, we specifically focused on the clinically relevant questions of ruling out significant fibrosis and ruling in advanced fibrosis/ cirrhosis.

Future studies on the utility of EPQ to assess the risk of compensated advanced chronic liver disease (cACLD) – validated by the diagnostic gold standards, i.e., varices detection by endoscopy, HVPG ≥6 mmHg, or liver biopsy showing F3/F4 fibrosis – are desirable.

In conclusion, this prospective, multinational study showed an excellent diagnostic accuracy of EPQ-pSWE liver stiffness measurements for assessing fibrosis risk in a large cohort of NAFLD patients. We found a divergence between absolute values of VCTE and EPQ, which monotonously increases above 20 kPa, but does not impact the ability of EPQ to stratify NAFLD fibrosis risk. Importantly, we provide optimized EPQ cutoffs to rule-in and rule-out significant fibrosis, advanced fibrosis, and cirrhosis in patients with NAFLD in daily clinical practice.

• References

• 1 Leite NC, Villela-Nogueira CA, Pannain VLN. et al. Histopathological stages of non-alcoholic fatty liver disease in type 2 diabetes: Prevalences and correlated factors. Liver Int 2011;
• 2 Younossi ZM, Koenig AB, Abdelatif D. et al. Global epidemiology of nonalcoholic fatty liver disease – Meta-analytic assessment of prevalence, incidence, and outcomes. Hepatology 2016;
• 3 Pimpin L, Cortez-Pinto H, Negro F. et al. Burden of liver disease in Europe: Epidemiology and analysis of risk factors to identify prevention policies. Journal of Hepatology 2018;
• 4 Webber L, Divajeva D, Marsh T. et al. The future burden of obesity-related diseases in the 53 WHO European-Region countries and the impact of effective interventions: A modelling study. BMJ Open 2014;
• 5 Angulo P, Kleiner DE, Dam-Larsen S. et al. Liver fibrosis, but no other histologic features, is associated with long-term outcomes of patients with nonalcoholic fatty liver disease. Gastroenterology 2015;
• 6 Semmler G, Scheiner B, Schwabl P. et al. The impact of hepatic steatosis on portal hypertension. PLoS One Public Library of Science 2019; 14: e0224506
• 7 Ferraioli G, Wong VWS, Castera L. et al. Liver Ultrasound Elastography: An Update to the World Federation for Ultrasound in Medicine and Biology Guidelines and Recommendations. Ultrasound Med Biol Elsevier USA 2018; 44: 2419-2440
• 8 Sumida Y, Nakajima A, Itoh Y. Limitations of liver biopsy and non-invasive diagnostic tests for the diagnosis of nonalcoholic fatty liver disease/nonalcoholic steatohepatitis. World J Gastroenterol 2014;
• 9 Dietrich CF, Bamber J, Berzigotti A. et al. EFSUMB Guidelines and Recommendations on the Clinical Use of Liver Ultrasound Elastography, Update 2017 (Long Version). Ultraschall in Med 2017; 38: e16-e47
• 10 Chalasani N, Younossi Z, Lavine JE. et al. The diagnosis and management of non-alcoholic fatty liver disease: Practice Guideline by the American Association for the Study of Liver Diseases, American College of Gastroenterology, and the American Gastroenterological Association. Hepatology 2012;
• 11 Myers RP, Pomier-Layrargues G, Kirsch R. et al. Feasibility and diagnostic performance of the FibroScan XL probe for liver stiffness measurement in overweight and obese patients. Hepatology 2012;
• 12 Ferraioli G, Tinelli C, Lissandrin R. et al. Point shear wave elastography method for assessing liver stiffness. World J Gastroenterol 2014;
• 13 Bucsics T, Grasl B, Ferlitsch A. et al. Point Shear Wave Elastography for Non-invasive Assessment of Liver Fibrosis in Patients with Viral Hepatitis. Ultrasound Med Biol 2018;
• 14 Sterling RK, Lissen E, Clumeck N. et al. Development of a simple noninvasive index to predict significant fibrosis in patients with HIV/HCV coinfection. Hepatology 2006;
• 15 Schwabl P, Bota S, Salzl P. et al. New reliability criteria for transient elastography increase the number of accurate measurements for screening of cirrhosis and portal hypertension. Liver Int Blackwell Publishing Ltd 2015; 35: 381-390
• 16 Ferraioli G, de Silvestri A, Reiberger T. et al. Adherence to quality criteria improves concordance between transient elastography and ElastPQ for liver stiffness assessment – A multicenter retrospective study. Dig Liver Dis 2018;
• 17 Tsochatzis EA, Gurusamy KS, Ntaoula S. et al. Elastography for the diagnosis of severity of fibrosis in chronic liver disease: A meta-analysis of diagnostic accuracy. J Hepatol Elsevier 2011; 54: 650-659
• 18 de Franchis R. Expanding consensus in portal hypertension. J Hepatol 2015; 63: 743-752
• 19 Reiberger T, Püspök A, Schoder M. et al. Austrian consensus guidelines on the management and treatment of portal hypertension (Billroth III). Wien Klin Wochenschr 2017;
• 20 Youden WJ. Index for rating diagnostic tests. Cancer 1950; 3
• 21 Leong WL, Lai LL, Nik Mustapha NR. et al. Comparing point shear wave elastography (ElastPQ) and transient elastography for diagnosis of fibrosis stage in non-alcoholic fatty liver disease. J Gastroenterol Hepatol Blackwell Publishing 2020; 35: 135-141
• 22 Semmler G, Binter T, Kozbial K. et al. Noninvasive Risk Stratification After HCV Eradication in Patients With Advanced Chronic Liver Disease. Hepatology Wiley 2020;
• 23 Liu H, Fu J, Hong R. et al. Acoustic radiation force impulse elastography for the non-invasive evaluation of hepatic fibrosis in non-alcoholic fatty liver disease patients: A systematic review & meta-analysis. PLoS One Public Library of Science 2015; 10: e0127782
• 24 Lin Y, Li H, Jin C. et al. The diagnostic accuracy of liver fibrosis in non-viral liver diseases using acoustic radiation force impulse elastography: A systematic review and meta-analysis. PLoS One Public Library of Science 2020; 15: e0227358
• 25 Garcovich M, Stasio E Di, Zocco MA. et al. Assessing Baveno VI criteria with liver stiffness measured using a new point-shear wave elastography technique (BAVElastPQ study). Liver Int 2020;
• 26 Trebicka J, Gu W, Ledinghen V De. et al. Two-dimensional shear wave elastography predicts survival in advanced chronic liver disease. Gut 2021;
• 27 Stift J, Semmler G, Walzel C. et al. Transjugular aspiration liver biopsy performed by hepatologists trained in HVPG measurements is safe and provides important diagnostic information. Dig Liver Dis 2019;
• 28 Reiberger T, Schwabl P, Trauner M. et al. Measurement of the hepatic venous pressure gradient and transjugular liver biopsy. J Vis Exp 2020; 1-16
• 29 Grgurevic I, Salkic N, Bozin T. et al. Magnitude dependent discordance in liver stiffness measurements using elastography point quantification with transient elastography as the reference test. Eur Radiol Springer Verlag; 2019; 29: 2448-2456
• 30 Ferraioli G, de Silvestri A, Lissandrin R. et al. Evaluation of Inter-System Variability in Liver Stiffness Measurements. Ultraschall in Med 2018; 40: 64-75