Skip to main content
Insights into Imaging
Download PDF

Quantitative US fat fraction for noninvasive assessment of hepatic steatosis in suspected metabolic-associated fatty liver disease

Insights into Imaging volume 15, Article number: 159 (2024) Cite this article

Abstract

Objectives

To evaluate the agreement between quantitative ultrasound system fat fraction (USFF) and proton magnetic resonance spectroscopy (1H-MRS) and the diagnostic value of USFF in assessing metabolic-associated fatty liver disease (MAFLD).

Methods

The participants with or suspected of MAFLD were prospectively recruited and underwent 1H-MRS, USFF, and controlled attenuation parameter (CAP) measurements. The correlation between USFF and 1H-MRS was assessed using Pearson correlation coefficients. The USFF diagnostic performance for different grades of steatosis was evaluated using receiver operating characteristic curve analysis (ROC) and was compared with CAP, visual hepatic steatosis grade (VHSG).

Results

A total of 113 participants (mean age 44.79 years ± 13.56 (SD); 71 males) were enrolled, of whom 98 (86.73%) had hepatic steatosis (1H-MRS ≥ 5.56%). USFF showed a good correlation (Pearson r = 0.76) with 1H-MRS and showed a linear relationship, which was superior to the correlation between CAP and 1H-MRS (Pearson r = 0.61). The USFF provided high diagnostic performance for different grades of hepatic steatosis, with ROC from 0.84 to 0.98, and the diagnostic performance was better than that of the CAP and the VHSG. The cut-off values of the USFF were different for various grades of steatosis, and the cut-off values for S1, S2, and S3 were 12.01%, 19.98%, and 22.22%, respectively.

Conclusions

There was a good correlation between USFF and 1H-MRS. Meanwhile, USFF had good diagnostic performance for hepatic steatosis and was superior to CAP and VHSG. USFF represents a superior method for noninvasive quantitative assessment of MAFLD.

Critical relevance statement

Quantitative ultrasound system fat fraction (USFF) accurately assesses liver fat content and has a good correlation with magnetic resonance spectroscopy (1H-MRS) for the assessment of metabolic-associated fatty liver disease (MAFLD), as well as for providing an accurate quantitative assessment of hepatic steatosis.

Key Points

  • Current diagnostic and monitoring modalities for metabolic-associated fatty liver disease have limitations.

  • USFF correlated well with 1H-MRS and was superior to the CAP.

  • USFF has good diagnostic performance for steatosis, superior to CAP and VHSG.

Graphical Abstract

Introduction

Metabolism-associated fatty liver disease (MAFLD) is a prevalent condition affecting over one-third of the world’s population [1]. Hepatic steatosis, also known as fatty liver disease, is a condition characterized by the accumulation of excess fat in the liver. It can be diagnosed through histological examination of liver biopsy specimens, blood biomarkers or scores, or imaging modalities such as ultrasound or magnetic resonance imaging (MRI). Hepatic steatosis is a common finding in individuals with overweight or obese (body mass index, BMI > 25 kg/m2), and metabolic dysfunction (including hypertriglyceridemia, hypercholesterolemia, increased waist circumference, insulin resistance, and systemic hypertension) [2]. MAFLD is not a single lesion, but rather a spectrum of progressive stages of liver disease, starting with simple steatosis and progressing to steatohepatitis, liver fibrosis, cirrhosis, and hepatocellular carcinoma (HCC) [3]. There are several important complications associated with MAFLD that can lead to increased mortality including cardiovascular disease (CVD), non-hepatic malignancies, lung disease, chronic kidney disease (CKD), cognitive impairment, and complications of type 2 diabetes mellitus (T2DM). CVD is the leading cause of death in these patients [4]. Due to the global increase in obesity and diabetes mellitus, the prevalence of MAFLD and its complications are on the rise [5]. Therefore, it is crucial to accurately identify hepatic steatosis in the clinical setting for timely diagnosis and treatment.

While liver biopsy remains the current gold standard for diagnosing MAFLD, it is an invasive procedure that carries risks of complications like bleeding and infection [6]. Moreover, the small amount of liver tissue obtained during a biopsy (1/50,000 of the liver) can result in sampling variation, making histological evaluation challenging [7]. Therefore, there is a critical need for multiple assessment methods, particularly noninvasive ones, to accurately diagnose and monitor hepatic steatosis [8]. MRI-based methods, including MRI proton density fat fraction (MRI-PDFF) and proton magnetic resonance spectroscopy (1H-MRS), can accurately and reproducibly assess hepatic steatosis with an area under the curve (AUC) for identifying hepatic steatosis as high as 0.99 [9], but these are not widely used due to their high cost and low accessibility [10, 11]. Ideally, a modality should be noninvasive, accurate, cost-effective, and provide a point of care and clinical assessment of the degree of hepatic steatosis in patients with MAFLD [12].

Conventional ultrasound (US) is the most common imaging modality used to assess hepatic steatosis because of its safety, noninvasiveness, and low cost. Moreover, US techniques demonstrated good reproducibility [13]. However, conventional US has limitations in terms of quantitative accuracy and reproducibility due to operator dependence [14]. Furthermore, accurately grading the degree of steatosis in conventional US is challenging, which may lead to underdiagnosis of low-grade steatosis [15]. Therefore, multiple methods have been investigated to extract quantitative information from the US to improve steatosis screening [16,17,18,19,20,21], each with advantages and disadvantages. For example, steatosis can be assessed by the hepatorenal index, which is a computer-assisted measurement of the hepatic and renal ultrasound echo intensity ratio (H/R) to measure the degree of hepatic steatosis [22]. However, displaying the right kidney and liver on the same ultrasound image can be challenging for inexperienced radiologists, particularly when examining obese patients or those who have had renal surgery resulting in renal morphological changes or split liver transplantation. This can make measuring the H/R ratio difficult [18]. The controlled attenuation parameter (CAP) method used by FibroScan® is the first approved and most widely used attenuation-based technique [19, 23]. The hepatic steatosis classification correlates with the measurement of the US attenuation signal in dB/m by a series of algorithms ranging from 100–400 dB/m, known as CAP [24]. The diagnostic accuracy of CAP has been extensively validated [25], with more than 10% of steatosis being differentiated by CAP. However, CAP has some disadvantages. It does not allow for the simultaneous assessment of hepatic morphological changes and has a high rate of measurement failure, making it ineffective in patients with a BMI [26]. To address the limitations of CAP in visualizing the liver, several techniques have been developed. These include attenuation imaging (ATI) [27], attenuation factor™ (ATT), ultrasound-guided attenuation parameter (UGAP) [14], tissue attenuation imaging (TAI), and tissue scattering distribution imaging (TSI) [11]. These noninvasive quantitative techniques may be appropriate for large-scale screening and for repeated measurements of liver fat content during longitudinal follow-up. This allows for dynamic monitoring of disease progression and shows promising potential for clinical use.

Quantitative ultrasound (QUS) techniques such as TAI and TSI have been applied to quantify liver fat with high intra- and intergroup reproducibility and agreement [28,29,30]. Meanwhile, a fat fraction (FF) estimator for QUS has been developed (USFF) that is able to convert directly from QUS to the corresponding liver fat fraction [31]. Studies have shown that 1H-MRS is more effective than conventional US in detecting various grades of hepatic steatosis, particularly mild hepatic steatosis [32]. The aim of this study is to explore the correlation between USFF and 1H-MRS, and to evaluate the advantages and disadvantages of USFF compared to conventional visual hepatic steatosis grade (VHSG) and CAP in diagnosing different grades of hepatic steatosis. Additionally, reference thresholds for different grades of steatosis will be provided.

Materials and methods

This prospective, single-institution, cross-sectional study was conducted in Fudan University Zhongshan Hospital. Our institutional review board approved this prospective study (approval number: B2021-0921R). All participants provided written informed consent. The authors had control over the data and information submitted for publication. The study was conducted in accordance with the Declaration of Helsinki.

Study participants and design

Study participants were recruited consecutively and prospectively between August 2022 and May 2023 by one radiologist (Y.L.Z. with 10 years of abdominal ultrasound experience) from the Center for the Study of MAFLD at Zhongshan Hospital, Fudan University. The diagnostic criteria for MAFLD followed the international expert consensus and clinical practice guidelines [2, 33]. Inclusion criteria for this study were: (a) age > 18 years; (b) clinical suspicion of MAFLD, i.e., the existence of fatty liver based on imaging or even blood biomarkers/score, like fatty liver index (FLI) and one of the following three conditions: overweight/obesity (BMI ≥ 23 kg/m2), type 2 diabetes mellitus, and metabolic dysfunction. Metabolic dysfunction is defined as the existence of at least two risk factors: (1) Waist circumference: ≥ 90 cm (male) and 80 cm (female); (2) Blood pressure: ≥ 130/85 mmHg or receiving treatment for lowering blood pressure; (3) Blood triglycerides: ≥ 1.7 mmol/L or receiving lipid-lowering drugs; (4) Plasma high-density lipoprotein (HDL) cholesterol: < 1.0 and 1.3 mmol/L for men and women, respectively, or receiving lipid-regulating drugs; (5) Pre-diabetes: fasting blood glucose 5.6–6.9 mmol/L or 2 h postprandial blood glucose 7.8–11.0 mmol/L or glycated hemoglobin of 5.7%–6.4%; (6) insulin resistance index assessed by the homeostasis model: ≥ 2.5; and (7) blood ultrasensitive C-reactive protein: ≥ 2 mg/L. Patients with contraindications to MRI and inability to hold their breath for >3 s during US examination were excluded. Demographic and clinical data were recorded for all participants. All participants underwent USFF, CAP, and 1H-MRS assessment on the same day or within 14 days when possible. The inclusion and exclusion criteria and study flow chart are shown in Fig. 1. No participants underwent any treatment prior to the imaging examination.

Fig. 1
figure 1

Flow chart of participants included and excluded from this study

Full size image

B-mode ultrasound measurements

The R10 Prestige US system (Samsung Medison Co. Ltd., Hongcheon, Korea) was used to perform the abdominal scans using a CA 1–7 S convex probe (1–7 MHz). Other clinical information about the participants was kept confidential by the radiologist. Characteristics of fatty liver on the gray-scale US included parenchymal echogenicity enhancement, attenuation of far-field echoes, increased hepatorenal contrast, decreased visibility of the diaphragm line and intrahepatic portal vein wall, and decreased visibility of deep liver structures [25]. Subsequently, VHSG was assessed by a radiologist using gray-scale US images obtained during B-mode US examination of the liver parenchyma [26]. The scoring system is as follows: 0, no fatty liver; 1, mild fatty liver; 2, moderate fatty liver; and 3, severe fatty liver. Mild steatosis was defined as increased echogenicity of the liver compared with that of the right renal cortex. Moderate steatosis refers to hepatic echogenicity greater than that of the right renal cortex with insignificant perihepatic echogenicity. Severe steatosis was characterized by increased hepatic echogenicity that obscures diaphragmatic echogenicity.

USFF data acquisition

During each examination, the radiologist performed at least five data acquisitions at the same location in the right hepatic lobe by using the lateral intercostal approach. Participants were positioned in a supine position with the right arm in maximum abduction. Prior to the first data acquisition, the system settings were adjusted for each participant to optimize visualization of the right lobe of the liver and to identify areas of liver parenchyma with localized fat deposits and large blood vessels to be avoided. The settings were kept constant for subsequent examinations. Each acquisition was performed by the operator pressing a button once to record a B-mode image and TAI and TSI values. Prior to each data acquisition, participants paused for breath after shallow inspiration. R2 values were used to optimize the accuracy of ROI. The operator’s goal is to obtain R2 values > 0.80 for all measurements. TAI is reported in dB/cm/MHz, while TSI is reported in arbitrary units [25]. After complete measurement of TAI and TSI values, the system automatically outputs the corresponding USFF values.

CAP data acquisition

FibroScan 502 Touch (Echosens, Paris, France) was used for CAP measurements. All CAP measurements were performed using a standardized protocol [17] by a gastroenterologist with more than 5 years of experience, who was blinded to the other results. The scans were performed on the same day as the US. The subjects were placed in the supine position with the right arm maximally abducted, and measurements of the right hepatic lobe were made through the intercostal space. The selection of the same intercostal area for QUS measurements ensures that the measurement area is as consistent as possible. The rationale for the CAP has been illustrated elsewhere [17]. The CAP was taken ten times, and the median and IQR values were calculated and recorded.

1H-MRS data acquisition

Participants underwent 1H-MRS examinations using an Avanto 1.5 T MR scanner (Siemens, Erlangen, Germany) within 1 week before/after the US scan. Participants were examined in a supine position and underwent breathing exercises prior to the examination. To localize spectroscopically collected voxels, sagittal, coronal, and axial sections covering the entire liver were collected. A single 8 cubic centimeter (2 × 2 × 2 cm) voxel was placed within the right lobe, avoiding major vascular structures and subcutaneous adipose tissue. After shimming the body of interest, a proton spectrum was acquired using a body coil. Proton spectra were acquired using a point-resolved spectroscopy (PRESS) sequence with the following parameters: repetition time = 1500 ms, echo time = 135 ms. Signal intensities of the water peak at 4.8 ppm (Sw) and the fat peak at 1.4 ppm (Sf) were measured at 1.4 ppm and 1.4 ppm, respectively. The percentage of liver fat was calculated using the formula 100 × Sf/(Sf + Sw), where Sf and Sw represent the signal intensities of the fat peak at 1.4 ppm and the water peak at 4.8 ppm, respectively. All MRS analysis results were analyzed by radiologists whose information was inter-blinded. The guidelines recommend MRI-based techniques, such as MRI-PDFF and 1H-MRS, as the gold standard for quantifying liver fat. Therefore, we used MRS as the gold standard and classified hepatic steatosis into four grades based on previous studies: S0 (1H-MRS < 5.56%), S1 (1H-MRS = 5.56–12.7%), S2 (1H-MRS = 12.7–18.9%), and S3 (1H-MRS ≥ 18.9%) [34, 35].

Statistical analysis

Continuous variables were expressed as means and standard deviations, and categorical variables were expressed as counts and percentages. Pearson correlation coefficients were calculated between USFF, CAP, and 1H-MRS for all participants. Bland-Altman analysis with 95% limits of agreement was used to test the correlation between USFF and MRS. Linear regression analysis was used to evaluate linear regression slope, intercept, and R2. A two-tailed t-test or Mann–Whitney U-test was performed for each quantitative USFF, CAP to determine whether the means of the USFF, CAP were statistically different between groups of participants. The significance level was set at p < 0.05. Receiver operating characteristic curve analysis (ROC) was used to compare the performance of USFF, CAP, VHSG in detecting varying degrees of hepatic steatosis of NAFLD. For each ROC analysis, the area under the ROC curve, sensitivity, specificity, positive and negative predictive values (NPVs) were calculated. AUCs were compared using the method described by DeLong et al [36]. All statistical analyses were performed using commercially available software (SPSS version 26.0; IBM Corporation, Armonk, NY, USA; and GraphPad Prism version 8.0; GraphPad Software Corporation, San Diego, CA, USA).

Results

Patient characteristics

In total, 113 prospectively recruited participants (mean age, 44.79 ± 13.56 years; 42 women (37.17%), 71 men (62.83%), mean BMI, 26.83 ± 5.57) with or suspected of having MAFLD, with both 1H-MRS, CAP and USFF measurements, were included in this study. The mean 1H-MRS was 16.45% ± 10.50% (range, 1.42%–57.70%). 98 (86.72%) of the participants had MAFLD (defined as 1H-MRS ≥ 5.56%) and 65 (57%) of the participants had 1H-MRS of 12.7% or more. The mean USFF was 18.81% ± 7.76% (range, 1.86%–46.60%). The mean CAP was 307.4 ± 51.51 dB/m (range, 175–400 dB/m). The baseline demographic, biochemical, histological, and imaging data of the participants were summarized in Table 1.

Table 1 Demographic, biochemical, histological, and imaging data of the participants
Full size table

Correlation between USFF and 1H-MRS

The scatterplot in Fig. 2a, b illustrates the distribution of USFF values according to the grade of liver steatosis. The mean USFF values of all participants with 1H-MRS = S1 (16.00% ± 4.08%; range, 5.79%–22.44%) were significantly higher than those of controls without steatosis (7.42% ± 2.73%; range, 1.86%–11.98%; p < 0.0001). In addition, participants with steatosis with 1H-MRS = S3 had significantly higher USFF values (24.77% ± 6.86%; range, 11.23%–46.60%, p < 0.0028) compared to other participants in the hepatic steatosis group (Fig. 2a, b and Table 2).

Fig. 2
figure 2

Correlation of USFF with 1H-MRS and CAP with 1H-MRS. a The degree of hepatic steatosis was graded according to the 1H-MRS results, and the distribution of USFF in different grades. t-test: S1-S0: p < 0.0001; t-test: S2-S1: p = 0.0024; t-test: S3-S2: p = 0.0028. b USFF values and 1H-MRS values corresponding to each participant. c Bland-Altman plot shows the difference between USFF and the 1H-MRS. d USFF vs. 1H-MRS scatterplot, Person correlation analysis, and linear regression line. e The degree of hepatic steatosis was graded according to the 1H-MRS results, and the distribution of CAP in different grades. t-test: S1-S0: p < 0.0012; t-test: S2-S1: p = 0.0057; t-test: S3-S2: p = 0.0687. f CAP vs. 1H-MRS scatterplot, Person correlation analysis, and the linear regression line

Full size image
Table 2 The hepatic fat content of all participants was graded based on 1H-MRS results
Full size table

The agreement between USFF and MRS is shown on Bland-Altman plots and scatter plots (Fig. 2c). The mean deviation between USFF and 1H-MRS was −2.29 ± 6.84%. The 95% coincidence range was −15.71% and 11.12%. The USFF showed a strong correlation with the 1H-MRS (r = 0.76, 95% confidence interval (CI): 0.82, 0.90; p < 0.001). In the linear range, linear regression analysis of USFF versus 1H-MRS showed a slope of 0.58 (95% CI: 0.49, 0.68), an intercept of 9.24 (95% CI: 7.40, 11.08), and an R2 of 0.58 (Fig. 2d).

Correlation between CAP and 1H-MRS

The distribution of CAP in different grades of hepatic steatosis was shown in Fig. 2e, and the results indicated that CAP was significantly different in varying degrees of hepatic steatosis. In addition, CAP was correlated with 1H-MRS (r = 0.61, 95% CI: 0.48, 0.71; p < 0.001). However, the correlation was poorer compared with USFF. CAP had some linear relationship with MRS. Within the linear range, linear regression analysis of USFF with 1H-MRS showed a slope of 2.78 (95% CI: 0.56, 0.85), an intercept of 262.1 (95% CI: −0.11, 1.63), and an R2 of 0.37. However, when the 1H-MRS ≥ 20%, the CAP tended to saturate without a significant linear relationship (Fig. 2f).

Comparison between USFF, CAP, and VHSG in the detection of 1H-MRS ≥ S1

Table 3 summarizes the performance of assessing steatosis by USFF, CAP, and VHSG based on 1H-MRS. The AUCs of USFF, CAP, and VHSG used to assess steatosis based on 1H-MRS ≥ S1 were 0.98 (95% CI: 0.96–1.00), 0.88 (95% CI: 0.80–0.96) and 0.85 (95% CI: 0.75–0.95), respectively (Fig. 3a, b and Table 3). In the receiver operating characteristic analysis, USFF had the highest diagnostic performance with a cut-off value of 12.01% and sensitivity and specificity of 92.86% and 100% respectively. In addition, CAP had the highest diagnostic performance with a cut-off value of 254.5 dB/m and sensitivity, and specificity of 94.79% and 62.50%, respectively. Also, hepatic steatosis was present when the VHSG ≥ 1, with sensitivity, and specificity of 98.98%, and 46.67%, respectively. When assessing hepatic steatosis based on 1H-MRS ≥ S1, USFF had significantly higher diagnostic performance than CAP and VHSG (p = 0.024 and p = 0.004, respectively).

Table 3 Comparison of the diagnostic performance between USFF, CAP, and VHSG in the detection of varying degrees of hepatic steatosis
Full size table
Fig. 3
figure 3

Comparison of the performance of USFF and CAP in detecting different grades of steatosis. a Receiver operating characteristic curves used to detect ≥ S1 (1H-MRS ≥ 5.56%) hepatic steatosis for USFF. b Receiver operating characteristic curves used to detect ≥ S1 (1H-MRS ≥ 5.56%) hepatic steatosis for CAP. c Receiver operating characteristic curves used to detect ≥ S2 (1H-MRS ≥ 12.7%) hepatic steatosis for USFF. d Receiver operating characteristic curves used to detect ≥ S2 (1H-MRS ≥ 12.7%) hepatic steatosis for CAP. e Receiver operating characteristic curves used to detect ≥S3 (1H-MRS ≥ 18.9) hepatic steatosis for USFF. f Receiver operating characteristic curves used to detect ≥ S3 (1H-MRS ≥ 18.9) hepatic steatosis for CAP. Pairwise comparison of receiver operating characteristic curves: ≥ S1: USFF-CAP: p = 0.024; USFF-VHSG: p = 0.004; CAP-VHSG: p = 0.190; ≥ S2: USFF-CAP: p = 0.231; USFF-VHSG: p < 0.001; CAP-VHSG: p = 0.001; ≥S3: USFF-CAP: p = 0.433; USFF-VHSG: p = 0.004; CAP-VHSG: p = 0.020

Full size image

Comparison between USFF, CAP, and visual score in the detection of 1H-MRS ≥ S2

The AUCs of USFF, CAP, and visual scores for the detection of hepatic steatosis (1H-MRS ≥ S2) were 0.89 (95% CI: 0.83–0.95), 0.85 (95% CI: 0.85–0.91), and 0.67 (0.58–0.78). The cut-off values for the detection of hepatic steatosis (1H-MRS ≥ S2) were USFF of 19.98%, CAP of 277 dB/m, and VHSG of 2, respectively (the corresponding sensitivity, and specificity are shown in Fig. 3c, d and Table 3). When assessing hepatic steatosis based on 1H-MRS ≥ S2, USFF had significantly higher diagnostic performance than VHSG (p < 0.001).

Comparison between USFF, CAP, and visual score in the detection of 1H-MRS ≥ S3

The study found that for 1H-MRS ≥ S3 the diagnostic performance of USFF was not significantly different from that of CAP (p = 0.433), but was significantly higher than that of VHSG (p = 0.004). The AUC for USFF was 0.84, while for CAP and VHSG it was 0.79 and 0.72, respectively (Fig. 3e, f and Table 3). When the cut-off value of USFF was 22.22%, the diagnostic sensitivity and specificity were 67.50% and 95.83%, respectively. When the cut-off value of CAP was 305.5 dB/m, its sensitivity and specificity were 87.18% and 69.44%, respectively. When the cut-off value of VHSG was 3, its sensitivity and specificity were 25% and 89%, respectively.

Discussion

MAFLD has become the most common liver disease and a significant cause of liver disability and death worldwide [33]. Therefore, it is extremely important to discern and accurately quantify the extent of hepatic steatosis and fat content during the initial stage [31]. Therefore, we sought to investigate the correlation between the USFF technique and 1H-MRS in a well-characterized prospective cross-sectional observational study of participants with or suspected of having MAFLD, along with a comparison of different imaging modalities. The study found that the USFF technique correlates well with 1H-MRS and was superior to other imaging techniques in identifying various grades of hepatic steatosis. This provides a cost-effective and convenient method for identifying, monitoring, and predicting outcomes for patients with MAFLD.

Related research has discovered that both CAP and USFF are capable of quantitatively assessing hepatic steatosis and are directly correlated with the severity of hepatic steatosis [25, 31]. Nevertheless, their utilization of distinct imaging principles and calculation methodologies results in varying quantitative parameters and diagnostic efficacy. Recognizing these discrepancies is crucial for both clinical practice and research purposes. Numerically, USFF showed a good correlation with 1H-MRS (r = 0.76). The most widely used US attenuation technique, CAP, had a moderate correlation coefficient with 1H-MRS (r = 0.61), similar to previous studies [37]. Comparing with CAP, USFF had a better correlation and demonstrated a better linear relationship with 1H-MRS. However, when MRI 1H-MRS ≥ 20%, CAP showed a saturation effect and hence had a poorer linear relationship with 1H-MRS. This suggested that the USFF technique was more accurate than CAP in higher grades of steatosis, and therefore, further comparison of their diagnostic performance was extremely important.

Our results showed that USFF had excellent diagnostic performance (AUC = 0.98) for diagnosing hepatic steatosis (1H-MRS ≥ S1). It was significantly higher than the CAP and subjective visual scores (0.88 and 0.85; p = 0.024 and p = 0.004, respectively). Given its excellent diagnostic performance, USFF was expected to be a noninvasive and accurate assessment tool for liver fat quantification in patients with MAFLD. It was also expected to be a potential diagnostic enrichment biomarker for identifying patients who could benefit from drug trials. High assessment failure rates have been reported in MAFLD-related drug trials due to participants not meeting MRI-derived PDFF criteria, regardless of their preselected CAP results [38]. USFF had a positive predictive value (PPV) of ≥ S1 (100%) for the diagnosis, which may help to reduce assessment failure rates in MAFLD-related drug trials.

In addition, previous studies have not explored and compared the performance of USFF and CAP for the diagnosis of moderate-to-severe hepatic steatosis (1H-MRS ≥ S2 and 1H-MRS ≥ S2). Early identification of moderate-to-severe grade steatosis was more clinically valuable, contributing to the recognition and early intervention of metabolic-associated steatohepatitis (MASH) and to the assessment of the efficacy of related treatments. The results showed that USFF had a good diagnostic performance (AUC = 0.89). It was significantly higher than the VHSG (0.85 and 0.67, respectively; p < 0.001 and p = 0.004). At the same time, its performance was not inferior to CAP for recognizing moderate to severe steatosis. Meanwhile, although the optimal cut point for CAP to diagnose MAFLD varied, USFF was directly correlated with the 1H-MRS results and provided a more intuitive response to the degree of hepatic steatosis. This suggested that the USFF technique was an efficient, convenient, and highly accurate tool for the early recognition of MASH. This study also provided reference cut-off values for different grades of hepatic steatosis, which were 12.01%, 19.98%, and 22.22%, corresponding to S1, S2, and S3, respectively.

Our study had several limitations. First, participants were skewed toward steatosis; the proportion of participants without steatosis was 13.3% (15 of 113), which does not characterize the distribution of real-world populations and may have contributed to the high cut-off value for the diagnosis of hepatic steatosis (1H-MRS ≥ 5.56%). To improve the accuracy of our diagnostic thresholds, it is necessary to include additional healthy volunteers without steatosis. Second, our study was a single-center study with a study population skewed toward mild and moderate steatosis, which may have led to some selection bias. Further multicenter validation is needed. In addition, due to the small sample size, this study failed to assess potential confounders such as inflammation or fibrosis, thus more studies are needed in the future. Finally, compared to MRI-PDFF, 1H-MRS can only analyze a portion of the liver parenchyma, while MRI-PDFF can scan and analyze the entire organ. Therefore, MRS is susceptible to specimen error. Additionally, 1H-MRS can only analyze the fat content of liver parenchyma and does not reflect the degree of inflammation and fibrosis of the liver, unlike liver biopsy. Therefore, in future studies, we will investigate the role of multiparametric ultrasound in identifying fibrosis and inflammation using liver biopsy as the gold standard. Meanwhile, USFF is similar to 1H-MRS as it analyses a portion of the liver parenchyma rather than the whole organ and therefore has some sampling error. It is also susceptible to respiratory effects, and some older patients may be unable to hold their breath well, resulting in measurement errors.

In conclusion, our study has demonstrated that quantitative USFF offers a strong correlation with 1H-MRS, enabling more accurate identification of hepatic steatosis than both CAP and VHSG, and is expected to be a valuable tool in the assessment of MAFLD and MASH.

Data availability

The datasets used or analyzed during the current study are available from the corresponding author upon reasonable request.

Abbreviations

1H-MRS:

Proton magnetic resonance spectroscopy

AUC:

Area under the curve

BMI:

Body mass index

CAP:

Controlled attenuation parameter

CI:

Confidence interval

MAFLD:

Metabolic-associated fatty liver disease

NPV:

Negative predictive value

PPV:

Positive predictive value

QUS:

Quantitative ultrasound

ROC:

Area under the receiver operating characteristic curve

TAI:

Tissue attenuation imaging

TSI:

Tissue scatter distribution imaging

USFF:

Quantitative ultrasound system fat fraction

VHSG:

Visual hepatic steatosis grade

References

  1. Chan KE, Koh TJL, Tang ASP et al (2022) Global prevalence and clinical characteristics of metabolic-associated fatty liver disease: a meta-analysis and systematic review of 10 739 607 individuals. J Clin Endocrinol Metab 107:2691–2700

    Article  PubMed  Google Scholar 

  2. Eslam M, Newsome PN, Sarin SK et al (2020) A new definition for metabolic dysfunction-associated fatty liver disease: an international expert consensus statement. J Hepatol 73:202–209

    Article  PubMed  Google Scholar 

  3. Friedman SL, Neuschwander-Tetri BA, Rinella M, Sanyal AJ (2018) Mechanisms of NAFLD development and therapeutic strategies. Nat Med 24:908–922

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  4. Pal SC, Mendez-Sanchez N (2023) Screening for MAFLD: who, when and how? Ther Adv Endocrinol Metab 14:20420188221145650

    Article  PubMed  PubMed Central  Google Scholar 

  5. Quek J, Chan KE, Wong ZY et al (2023) Global prevalence of non-alcoholic fatty liver disease and non-alcoholic steatohepatitis in the overweight and obese population: a systematic review and meta-analysis. Lancet Gastroenterol Hepatol 8:20–30

    Article  PubMed  Google Scholar 

  6. Han A, Byra M, Heba E et al (2020) Noninvasive diagnosis of nonalcoholic fatty liver disease and quantification of liver fat with radiofrequency ultrasound data using one-dimensional convolutional neural networks. Radiology 295:342–350

    Article  PubMed  Google Scholar 

  7. Ratziu V, Charlotte F, Heurtier A et al (2005) Sampling variability of liver biopsy in nonalcoholic fatty liver disease. Gastroenterology 128:1898–1906

    Article  PubMed  Google Scholar 

  8. Romero-Gomez M, Cortez-Pinto H (2017) Detecting liver fat from viscoelasticity: How good is CAP in clinical practice? The need for universal cut-offs. J Hepatol 66:886–887

    Article  PubMed  Google Scholar 

  9. Runge JH, Smits LP, Verheij J et al (2018) MR spectroscopy-derived proton density fat fraction is superior to controlled attenuation parameter for detecting and grading hepatic steatosis. Radiology 286:547–556

    Article  PubMed  Google Scholar 

  10. Pais R, Barritt AST, Calmus Y et al (2016) NAFLD and liver transplantation: current burden and expected challenges. J Hepatol 65:1245–1257

    Article  PubMed  PubMed Central  Google Scholar 

  11. Jung J, Han A, Madamba E et al (2022) Direct comparison of quantitative US versus controlled attenuation parameter for liver fat assessment using MRI proton density fat fraction as the reference standard in patients suspected of having NAFLD. Radiology 304:75–82

    Article  PubMed  Google Scholar 

  12. Rinella ME (2015) Nonalcoholic fatty liver disease: a systematic review. JAMA 313:2263–2273

    Article  CAS  PubMed  Google Scholar 

  13. Castera L, Friedrich-Rust M, Loomba R (2019) Noninvasive assessment of liver disease in patients with nonalcoholic fatty liver disease. Gastroenterology 156:1264–1281 e1264

    Article  PubMed  Google Scholar 

  14. Kuroda H, Oguri T, Kamiyama N et al (2023) Multivariable quantitative US parameters for assessing hepatic steatosis. Radiology 309:e230341

    Article  PubMed  Google Scholar 

  15. Paige JS, Bernstein GS, Heba E et al (2017) A pilot comparative study of quantitative ultrasound, conventional ultrasound, and MRI for predicting histology-determined steatosis grade in adult nonalcoholic fatty liver disease. AJR Am J Roentgenol 208:W168–W177

    Article  PubMed  PubMed Central  Google Scholar 

  16. Lin SC, Heba E, Wolfson T et al (2015) Noninvasive diagnosis of nonalcoholic fatty liver disease and quantification of liver fat using a new quantitative ultrasound technique. Clin Gastroenterol Hepatol 13:1337–1345.e1336

    Article  PubMed  Google Scholar 

  17. Chan WK, Nik Mustapha NR, Mahadeva S (2014) Controlled attenuation parameter for the detection and quantification of hepatic steatosis in nonalcoholic fatty liver disease. J Gastroenterol Hepatol 29:1470–1476

    Article  CAS  PubMed  Google Scholar 

  18. Shiralkar K, Johnson S, Bluth EI, Marshall RH, Dornelles A, Gulotta PM (2015) Improved method for calculating hepatic steatosis using the hepatorenal index. J Ultrasound Med 34:1051–1059

    Article  PubMed  Google Scholar 

  19. Myers RP, Pollett A, Kirsch R et al (2012) Controlled attenuation parameter (CAP): a noninvasive method for the detection of hepatic steatosis based on transient elastography. Liver Int 32:902–910

    Article  PubMed  Google Scholar 

  20. Imbault M, Faccinetto A, Osmanski BF et al (2017) Robust sound speed estimation for ultrasound-based hepatic steatosis assessment. Phys Med Biol 62:3582–3598

    Article  PubMed  Google Scholar 

  21. Jeon SK, Lee JM, Joo I et al (2019) Prospective evaluation of hepatic steatosis using ultrasound attenuation imaging in patients with chronic liver disease with magnetic resonance imaging proton density fat fraction as the reference standard. Ultrasound Med Biol 45:1407–1416

    Article  PubMed  Google Scholar 

  22. Marshall RH, Eissa M, Bluth EI, Gulotta PM, Davis NK (2012) Hepatorenal index as an accurate, simple, and effective tool in screening for steatosis. AJR Am J Roentgenol 199:997–1002

    Article  PubMed  Google Scholar 

  23. Caussy C, Alquiraish MH, Nguyen P et al (2018) Optimal threshold of controlled attenuation parameter with MRI-PDFF as the gold standard for the detection of hepatic steatosis. Hepatology 67:1348–1359

    Article  CAS  PubMed  Google Scholar 

  24. Ferraioli G, Tinelli C, Lissandrin R et al (2014) Interobserver reproducibility of the controlled attenuation parameter (CAP) for quantifying liver steatosis. Hepatol Int 8:576–581

    Article  PubMed  Google Scholar 

  25. Karlas T, Petroff D, Sasso M et al (2017) Individual patient data meta-analysis of controlled attenuation parameter (CAP) technology for assessing steatosis. J Hepatol 66:1022–1030

    Article  PubMed  Google Scholar 

  26. Eddowes PJ, Sasso M, Allison M et al (2019) Accuracy of FibroScan controlled attenuation parameter and liver stiffness measurement in assessing steatosis and fibrosis in patients with nonalcoholic fatty liver disease. Gastroenterology 156:1717–1730

    Article  PubMed  Google Scholar 

  27. Cassinotto C, Jacq T, Anselme S et al (2022) Diagnostic performance of attenuation to stage liver steatosis with MRI proton density fat fraction as reference: a prospective comparison of three US machines. Radiology 305:353–361

    Article  PubMed  Google Scholar 

  28. Dioguardi Burgio M, Imbault M, Ronot M et al (2019) Ultrasonic adaptive sound speed estimation for the diagnosis and quantification of hepatic steatosis: a pilot study. Ultraschall Med 40:722–733

    Article  PubMed  Google Scholar 

  29. Han A, Andre MP, Deiranieh L et al (2018) Repeatability and reproducibility of the ultrasonic attenuation coefficient and backscatter coefficient measured in the right lobe of the liver in adults with known or suspected nonalcoholic fatty liver disease. J Ultrasound Med 37:1913–1927

    Article  PubMed  PubMed Central  Google Scholar 

  30. Han A, Labyed Y, Sy EZ et al (2018) Inter-sonographer reproducibility of quantitative ultrasound outcomes and shear wave speed measured in the right lobe of the liver in adults with known or suspected non-alcoholic fatty liver disease. Eur Radiol 28:4992–5000

    Article  PubMed  PubMed Central  Google Scholar 

  31. Han A, Zhang YN, Boehringer AS et al (2020) Assessment of hepatic steatosis in nonalcoholic fatty liver disease by using quantitative US. Radiology 295:106–113

    Article  PubMed  Google Scholar 

  32. Bohte AE, van Werven JR, Bipat S, Stoker J (2011) The diagnostic accuracy of US, CT, MRI and 1H-MRS for the evaluation of hepatic steatosis compared with liver biopsy: a meta-analysis. Eur Radiol 21:87–97

    Article  PubMed  Google Scholar 

  33. Eslam M, Sarin SK, Wong VW et al (2020) The Asian Pacific Association for the Study of the Liver clinical practice guidelines for the diagnosis and management of metabolic associated fatty liver disease. Hepatol Int 14:889–919

    Article  PubMed  Google Scholar 

  34. Kim JW, Lee CH, Yang Z, Kim BH, Lee YS, Kim KA (2022) The spectrum of magnetic resonance imaging proton density fat fraction (MRI-PDFF), magnetic resonance spectroscopy (MRS), and two different histopathologic methods (artificial intelligence vs. pathologist) in quantifying hepatic steatosis. Quant Imaging Med Surg 12:5251–5262

    Article  PubMed  PubMed Central  Google Scholar 

  35. Szczepaniak LS, Nurenberg P, Leonard D et al (2005) Magnetic resonance spectroscopy to measure hepatic triglyceride content: prevalence of hepatic steatosis in the general population. Am J Physiol Endocrinol Metab 288:E462–E468

    Article  CAS  PubMed  Google Scholar 

  36. DeLong ER, DeLong DM, Clarke-Pearson DL (1988) Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. Biometrics 44:837–845

  37. Ferraioli G, Maiocchi L, Raciti MV et al (2019) Detection of liver steatosis with a novel ultrasound-based technique: a pilot study using MRI-derived proton density fat fraction as the gold standard. Clin Transl Gastroenterol 10:e00081

    Article  PubMed  PubMed Central  Google Scholar 

  38. Tamaki N, Ajmera V, Loomba R (2022) Non-invasive methods for imaging hepatic steatosis and their clinical importance in NAFLD. Nat Rev Endocrinol 18:55–66

    Article  CAS  PubMed  Google Scholar 

Download references

Funding

This work is supported by the National Nature Science Foundation of China (Grant Nos. 81927801, and 82302206), Scientific Research and Development Fund of Zhongshan Hospital of Fudan University (Grant 2022ZSQD07), Shanghai Sailing Program (23YF1441600) and China Postdoctoral Science Foundation (Grant Nos. 2023TQ0073). Health Commission Health Sector Clinical Research Special Clinical Project (Grant 202340087).

Author information

Author notes

  1. Haohao Yin and Yunling Fan contributed equally to this work.

Authors and Affiliations

  1. Department of Ultrasound, Zhongshan Hospital, Institute of Ultrasound in Medicine and Engineering, Fudan University, Shanghai, 200032, China

    Haohao Yin, Yunling Fan, Jifeng Yu, Boyang Zhou, Yikang Sun, Lifan Wang, Yuli Zhu & Huixiong Xu

  2. Shanghai Institute of Medical Imaging, Fudan University, Shanghai, 200032, China

    Bing Xiong

Authors
  1. Haohao Yin
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Yunling Fan
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Jifeng Yu
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Bing Xiong
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Boyang Zhou
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Yikang Sun
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Lifan Wang
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. Yuli Zhu
    View author publications

    You can also search for this author in PubMed Google Scholar

  9. Huixiong Xu
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

The authors thank the research participants for making this study possible. Huixiong Xu: conceptualization, funding acquisition, supervision, resources, writing—review & editing. Yuli Zhu: supervision, funding acquisition, writing—review & editing. Haohao Yin, Yunling Fan, Jifeng Yu, Bing Xiong: methodology, data curation, resources, writing—original draft, writing—review & editing. Jifeng Yu, Boyang Zhou, Yikang Sun, and Lifan Wang: methodology, data curation, resources.

Corresponding authors

Correspondence to Yuli Zhu or Huixiong Xu.

Ethics declarations

Ethics approval and consent to participate

Written informed consent was waived by the Ethics Committee of the institution (Zhongshan Hospital, Fudan University), and the Ethics Committee of the institution approval was obtained (No: B2021-0921R).

Consent for publication

All participants provided written informed consent.

Competing interests

The authors declare that they have no competing interests in this section.

Additional information

Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Yin, H., Fan, Y., Yu, J. et al. Quantitative US fat fraction for noninvasive assessment of hepatic steatosis in suspected metabolic-associated fatty liver disease. Insights Imaging 15, 159 (2024). https://doi.org/10.1186/s13244-024-01728-2

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s13244-024-01728-2

Keywords