RPS 1901 - Exploring fatty liver disease

Author Block: M. Zbroja-Putowska, M. A. Kuczyńska, A. Brodzisz, P. Krawiec, M. M. Wozniak; Lublin/PL
Purpose: The aim of the study is to evaluate the use of multiparametric liver ultrasound in the diagnosis and monitoring of liver steatosis and fibrosis.
Methods or Background: Forty-five pediatric patients (20 W, 25 M) with increased liver laboratory parameters (AST, ALT) were included in the study. They were subjected to multiparametric liver ultrasound with evaluation of SWE (Emedian [kPa]), attenuation (ATI [dB/cm/MHz]) and dispersion (SWD [(m/s)/kHz]). All ultrasound examinations were performed with a Canon Aplio i800, with a convex i8CX1 probe from intercostal access; the sampling gate was positioned 1-2 cm from the liver pouch. Min. 5 measurements for all parameters, determining the median and the IQR coefficient for the median.
Results or Findings: 40 out of 45 pediatric patients examined turned out to have liver steatosis (mostly mild and moderate). In 32 of them the level of steatosis was proportional to the level of obesity. 8 patients didn’t have obesity but the liver steatosis turned out to be mild or moderate, 1 of them had fibrosis (F1/F2). These patients were taking steroid medications. After one year, 8 patients had follow up examination. In 6 patients liver steatosis and fibrosis levels have decreased, in 2 patients steatosis maintainted on the same level – these patients didn’t follow the recommendations.
Conclusion: Liver parenchymal steatosis remains asymptomatic for a long time, posing a serious clinical problem. Multiparametric liver assessment provides a rapid, accessible means of reliably evaluating not only the degree of parenchymal steatosis, but also the risk of progression or transformation to cirrhosis. Thus, mpUSG now has the potential to become the most important non-invasive diagnostic tool in the screening assessment and monitoring of patients with metabolic syndromes as well as those undergoing hepatotoxic treatment.
Limitations: Small group
Funding for this study: n/a
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information:

Author Block: N. Seneviratne, C. Fang, D. Quinlan, P. S. Sidhu; London/UK
Purpose: Quantitative ultrasound (QUS) techniques—including tissue attenuation imaging (TAI), tissue scatter-distribution imaging (TSI), and ultrasound derived fat fraction (USFF)—are emerging non-invasive tools for quantifying hepatic steatosis. We assessed their diagnostic accuracy against histological steatosis grade and compared them with the Controlled Attenuation Parameter (CAP).
Methods or Background: One hundred and twenty patients with chronic liver disease were prospectively recruited for contemporaneous liver biopsy and QUS imaging. Reference was histological steatosis grade (S0-3) scored by consensus by liver pathologists. Diagnostic accuracy of USFF, TSI, TAI and CAP was evaluated using ROC analysis and Youden optimal cut-offs. Multivariable ordinal logistic regression examined independent predictors of steatosis grade.
Results or Findings: The final cohort comprised 115 patients (57 men, 58 women; median age 45 years; median BMI 27.8 kg/m2). Histological steatosis grades were S0=65, S1=28, S2=14, S3=8. USFF, TSI and TAI correlated strongly with histological grade (ρ=0.81, 0.76 and 0.75, all P<0.001), while BMI showed a moderate correlation (ρ=0.56, P<0.001). Age and gender were not significant. USFF achieved excellent AUCs of 0.95 for S≥1 and 0.94 for S≥2, with optimal cut-offs of 8.8% and 15.4%. At these thresholds, sensitivity/specificity were 88%/89% (S≥1) and 91%/85% (S≥2). TAI and TSI showed similar accuracy (AUCs 0.89–0.94) with cut-offs of 0.83 and 89.2 (S≥1), 0.91 and 93.1 (S≥2). In a subset (N=61), CAP showed comparable accuracy (AUCs 0.93, 0.94). On multivariate analysis, USFF (OR 18.0, P<0.001), TSI (OR 12.8, P<0.001) and TAI (OR 8.4, p<0.001) remained independent predictors; BMI was also significant in the USFF model (OR 1.94, P=0.038).
Conclusion: QUS-derived USFF, TSI and TAI provide excellent diagnostic accuracy for histological steatosis, performing at least as well as CAP.
Limitations: Possibility of sampling error at liver biopsy. Relatively small high steatosis grade patients.
Funding for this study: Samsung Medison
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: Research Ethics [REC] reference: 22/NW/0401, IRAS Project ID: 212836

Author Block: M. Xiao, Y. Wu, N. Wang, X. Wang, Y. Chong, J. Ren, L. Wu; Guangzhou/CN
Purpose: To assess the impact of ultrasonographically assessed diffuse hepatic steatosis and metabolic factors on the risk of liver-related events (LREs) in chronic hepatitis B (CHB) patients.
Methods or Background: We conducted a retrospective cohort study of patients with CHB who attended regular follow-up visits at our institution from January 2009 to December 2022. Diffuse hepatic steatosis was evaluated using ultrasonography. The primary endpoint was the development of LREs.Kaplan-Meier curves were generated to compare LRE-free survival between groups. Propensity score matching (PSM) was performed in a 1:1 ratio to minimize confounding effects. Univariate and multivariate Cox regression models were used to identify independent predictors associated with LREs.
Results or Findings: A total of 1808 patients with CHB were included, with a median follow-up of 104 months. LREs occurred in 134 patients (7.4%), among which hepatocellular carcinoma (HCC) accounted for 74.6% (n=100), decompensated cirrhosis for 21.0% (n=28), liver transplantation for 2.2% (n=3), and liver failure-related death for 2.2% (n=3). The group with diffuse hepatic steatosis exhibited a significantly lower risk of LREs compared to those without steatosis (HR: 0.398, 95% CI: 0.232-0.682). This protective association remained significant after PSM (HR: 0.484, 95% CI: 0.250-0.940). Multivariate analysis identified type 2 diabetes (HR: 1.69) and cirrhosis (HR: 5.15) as independent risk factors for LREs, while female sex (HR: 0.56) and higher albumin levels (HR: 0.89) were associated with a reduced risk.
Conclusion: Diffuse hepatic steatosis is not a primary driver of LREs in patients with CHB, and its potential protective effect should be interpreted within an individualized clinical context. Beyond antiviral therapy, proactive management of metabolic factors—such as glycemic and blood pressure control—can effectively reduce the risk of LREs.
Limitations: This study lacks dynamic assessment of hepatic steatosis.
Funding for this study: Guangzhou Science and Technology Plan Project under Grant No: 2023A03J0728.
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: Third Affiliated Hospital of Sun Yat-sen University (Approval No. [2021]02-387-01).

Author Block: P. C. P. Joshi, A. K. K. O. Dr, I. S. Shah, A. Borse; Pune/IN
Purpose: To correlate fat quantification on ultrasound-guided attenuation parameter with MR Spectroscopy and mDIXON fat fractions in patients with non-alcoholic fatty liver disease (NAFLD).
Methods or Background: This prospective study was conducted at a tertiary care hospital in western india after obtaining ethical clearance from the institute. Thirty patients with diffuse hepatic steatosis on ultrasound / CT were included in this study. Patients with chronic alcoholism, chronic viral hepatitis and hepatocellular carcinoma were excluded. Each participant underwent UGAP ultrasound and 3T MRI (Philips) with body coil. Hepatic fat fraction was calculated using the lipid-to-water peak ratio in MR spectroscopy and fat fraction value in mDIXON imaging. UGAP parameters including attenuation coefficient were also recorded.
Results or Findings: Pearson correlation was used to assess the association between UGAP grading, MR Spectroscopy and mDIXON fat fraction. Correlation is significant at the 0.01 level. UGAP showed a strong positive correlation with both MR Spectroscopy and mDIXON results, with a correlation coefficient of 0.869 and 0.897 respectively. Notably, the mDIXON method demonstrated a higher level of agreement with MR Spectroscopy, with a correlation coefficient of 0.892.
Conclusion: UGAP shows high degree of correlation with MR spectroscopy and mDIXON imaging as a non-invasive method for the evaluation of non-alcoholic fatty liver disease. This study highlights that UGAP can be used in isolation, potentially providing a cost-effective and accessible alternative technique to MRI-based techniques in the assessment and monitoring of hepatic steatosis.
Limitations: Histological confirmation was not included in this study. Since the MR Spectroscopy fat fraction method is advised for measuring hepatic steatosis in a clinical context, it was utilized as a reference instead. Only Indian patients with NAFLD were included. Only GE scanners are compatible with the UGAP used in this study.
Funding for this study: No
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: Bharati Vidyapeeth Medical College,PUNE-411043
Institutional Ethics Committe
(DHR Reg. No:EC/NEW/INST/2022/MH/0150)

Author Block: T. Schömig¹, S. Skornitzke², L. Hieronymi¹, D. Maintz¹, N. Große Hokamp¹; ¹Cologne/DE, ²Hamburg/DE
Purpose: This study investigates reproducibility and interobserver agreement of a previously published three-material-decomposition (3MD) algorithm for hepatic fat quantification using dual-layer dual-energy CT (dlDECT). Prior research demonstrated good correlation between dlDECT and MR fat quantification. As a next step toward clinical integration, we aim to evaluate consistency and reliability of this dlDECT-based approach.
Methods or Background: In this retrospective study, we evaluated contrast-enhanced abdominal dual-layer CT scans (IQon, Philips Healthcare) of 58 patients obtained twice within 60 days (mean interval: 34.3± 20.2 days). Patients with severe changes in clinical status, such as initiation of chemotherapy, or findings affecting liver perfusion, were excluded. Fat fraction maps were generated using the published 3MD-algorithm for native, arterial, venous and delayed phases. Two radiologists independently placed five circular ROIs per scan, four in the right and one in the left hepatic lobe, avoiding vessels. Reproducibility between scans and interobserver agreement were assessed using Pearson’s correlation, Bland-Altman analysis, repeated measures ANOVA, and intraclass-coefficients (ICC).
Results or Findings: The mean hepatic fat fraction across all patients and contrast phases was 4.91%±5.91, no significant difference was found between phases (p=0.07). Averaging across phases and readers, measurements at the two time points strongly correlated (4.87%±6.04 vs. 4.94%±6.13, R²=0.78) with a low mean difference (0.34±3.09 percentage points, not significant, p=0.61), indicating good reproducibility. Interobserver agreement was excellent (ICC=0.92), with a mean difference of 0.62±1.3 percentage points, consistent across all phases.
Conclusion: The novel 3MD-algorithm enabled through dlDECT provides reproducible and consistent hepatic fat quantification with excellent interobserver agreement, supporting its potential role in clinical practice as a reliable alternative to MRI or ultrasound.
Limitations: The limitations of the study are the single-center, retrospective design, potential reader dependency of ROI-placement and limited sample size.
Funding for this study: This Work was funded by Philips Healthcare. The funding source had no involvement in study design, collection or interpretation of data.
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information:

Author Block: S. DASH, D. Gg; Puducherry/IN
Purpose: The objective of our study was to compare the performance of conventional Single Energy Computed Tomography(SECT) and Dual Energy Computed Tomography (DECT) for quantification of hepatic steatosis using Magnetic Resonance Imaging ( MRI ) Chemical shift encoded( CSE) Proton Density Fat Fraction (PDFF) as a reference standard.
Methods or Background: For this cross-sectional analytical study, we recruited 141 adult patients who underwent plain SECT, post-contrast DECT and Liver MRI. We assessed five fat quantification variables namely SECT Hounsefield Units (HU), SECT Liver Attenuation Index(LAI), DECT Fat Fraction( FF), MRI CSE PDFF, and MRS-FF. Relationships between variables were analysed using Spearman correlation, ICC, and linear regression analysis.
Results or Findings: Our results showed that DECT(r = 0.361,r2 = 0.788, p < 0.05) and SECT LAI (r = -0.38,r2 = 0.742, p < 0.05). had a weak correlation with MRI CSE PDFF, while SECT HU showed a moderate correlation (r = -0.461, r2 = 0.8, p < 0.05). There was excellent consistency between MRS FF and MRI CSE PDFF quantification methods (ICC = 0.982, p < 0.05). Our subgroup analysis showed that in patients with hepatic steatosis (MRS FF ≥ 4%), both SECT methods and DECT FF correlated better with MRI CSE PDFF compared to those without steatosis.
Conclusion: SECT fat quantification methods seem more reliable than the DECT-based fat fraction method. Among SECT methods, SECT attenuation values are more precise than SECT LAI.
Limitations: We excluded patients with clinically suspected iron overload or chronic liver disease from our study group. However, it's essential to recognize that these conditions may act as potential confounders when evaluating liver attenuation.Our study relied solely on a single DECT technology, the Twin Beam Split filter technique, which is considered an inferior DECT technique.
Funding for this study: No funding was provided.
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: The study was approved by the Ethics committee

Author Block: T. D. Özdemir, H. E. E. GÜZEL; İzmir/TR
Purpose: To compare liver and pancreatic fat quantification using two MRI-based approaches, Dual Echo and FATFRAC sequences, and to explore their relationship with clinical and biochemical markers relevant to metabolic health.
Methods or Background: We retrospectively reviewed abdominal MRI scans of 100 adult patients acquired on a 1.5T system, including both Dual Echo (in-/out-of-phase) and Dixon-based FATFRAC sequences. Regions of interest were placed in the liver, pancreas, and spleen parenchyma to calculate mean fat fractions. Clinical and laboratory data (ALT, AST, GGT, HbA1c, lipid profile) were collected for correlation analysis. Patients with advanced chronic liver disease, hemochromatosis, or diffuse liver lesions were excluded.
Results or Findings: Liver fat quantification showed strong agreement between Dual Echo and FATFRAC (r = 0.82, p < 0.001), with no significant difference in mean values.
Pancreatic fat measurements demonstrated weaker concordance (r = 0.55, p = 0.006), with FATFRAC yielding consistently higher and more stable values.
Liver fat correlated significantly with ALT, GGT, and triglycerides, whereas pancreatic fat did not show robust associations with HbA1c or lipid parameters.
Negative pancreas values occasionally occurred with Dual Echo, reflecting known susceptibility to artifacts.
Conclusion: Both Dual Echo and FATFRAC are reliable for hepatic fat assessment, with strong metabolic correlations. For pancreatic fat, FATFRAC appears more reproducible and clinically plausible, while Dual Echo may be limited by technical artifacts. These findings support the use of FATFRAC when multi-organ fat quantification is required in metabolic imaging studies.
Limitations: The study was retrospective, single-center, and lacked histopathological or MR spectroscopy validation. Pancreatic fat assessment remains technically challenging, and further prospective studies with larger cohorts and histological correlation are needed.
Funding for this study: This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: Ethical approval was obtained from İzmir City Hospital (Approval No: 2025/26, Date: 07.05.2025).