Cutoff Values for Diagnosing Hepatic Steatosis Using Contemporary MRI-Proton Density Fat Fraction Measuring Methods

Development of a Deep Learning Model for Classification of Hepatic Steatosis from Clinical Standard Ultrasound

OBJECTIVE

Early detection and monitoring of hepatic steatosis can help establish appropriate preventative measures against progression to more advanced disease. We aimed to develop a deep learning (DL) program for classification of hepatic steatosis from standard-of-care grayscale ultrasound (US) images.

METHODS

In this single-center retrospective study, we utilized grayscale US images from January 1, 2010, to October 23, 2022, labeled with magnetic resonance imaging (MRI) proton density fat fraction (MRI-PDFF) to develop a DL multi-instance program for differentiating normal (S0) from steatotic liver (S1/2/3) and normal/mild steatosis (S0/1) from moderate/severe steatosis (S2/3). Diagnostic performances were assessed with area under the receiver operating characteristic curves (AUC), sensitivity, specificity and balanced accuracy with 95% confidence interval (CI).

RESULTS

A total of 403 patients with 403 US exams were included: 171 (42%) were normal (S0: MRI-PDFF <5%), 154 (38%) had mild steatosis (S1: MRI-PDFF 5-17.4%), 29 (7%) had moderate steatosis (S2: MRI-PDFF >17.4%-22.1%) and 49 (12%) had severe steatosis (S3: MRI-PDFF >22.1%). The dataset was split to include 322 patients in train/validation and 81 patients in a holdout test set (kept blind). The S0 versus S1/2/3 model achieved 81.3% (95% CI 72.1-90.5) AUC, 81.1% (70.6-91.6) sensitivity, 71.4% (54.7-88.2) specificity and 76.3% (66.4-86.2) balanced accuracy. The S0/1 versus S2/3 model achieved 95.9% (89-100) AUC, 87.5% (71.3-100) sensitivity, 96.9% (92.7-100) specificity and 92.2% (83.8-100) balanced accuracy. A multi-class model achieved a sensitivity of 71.4% (54.7-88.2) for S0, 67.6% (52.5-82.7) for S1 and 87.5% (71.3-100) for S2/3; specificity for the same model was 81.1% (70.6-91.6) for S0, 77.3% (64.9-89.7) for S1 and 96.9% (92.7-100) for S2/3.

CONCLUSION

Our DL program offered high sensitivity and accuracy in detecting and categorizing hepatic steatosis from standard-of-care ultrasound.

Even moderate liver fat accumulation below conventional fatty liver cutoffs is linked to multiple metabolomic alterations and gestational dysglycemia in Asian women of reproductive age

BACKGROUND

It is not clear if conventional liver fat cutoff of 5.56% weight which has been used for identifying fatty liver in western populations is also applicable for Asians. In Asian women of reproductive age, we evaluate the optimum metabolic syndrome (MetS)-linked liver fat cutoff, the specific metabolomic alterations apparent at this cutoff, as well as prospective associations of preconception liver fat levels with gestational dysglycemia.

METHODS

Liver fat (measured by magnetic resonance spectroscopy), MetS, and nuclear magnetic resonance (NMR)-based plasma metabolomic profiles were assessed in 382 Asian women, who were planning to conceive. Ninety-eight women went on to become pregnant and received an oral glucose tolerance test at week 26 of gestation.

RESULTS

The optimum liver fat cutoff for diagnosing MetS was 2.07%weight. Preconception liver fat was categorized into Low (liver fat < 2.07%), Moderate (2.07% ≤ liver fat < 5.56%), and High (liver fat ≥ 5.56%) groups. Individual MetS traits showed worsening trends, going from Low to Moderate to High groups. Multiple plasma metabolomic alterations, previously linked to incident type 2 diabetes (T2D), were already evident in the Moderate group (adjusted for ethnicity, age, parity, educational attainment, and BMI). Both a cross-sectional multi-metabolite score for incident T2D and mid-gestational glucose area under the curve showed increasing trends, going from Low to Moderate to High groups (p < 0.001 for both). Gestational diabetes incidence was 2-fold (p = 0.23) and 7-fold (p < 0.001) higher in the Moderate and High groups relative to the Low group.

CONCLUSIONS

In Asian women of reproductive age, moderate liver fat accumulation below the conventional fatty liver cutoff was not metabolically benign and was linked to gestational dysglycemia. The newly derived cutoff can aid in screening individuals before adverse metabolic phenotypes have consolidated, which provides a longer window for preventive strategies.

Advanced MRI techniques in abdominal imaging

Magnetic resonance imaging (MRI) is a crucial modality for abdominal imaging evaluation of focal lesions and tissue properties. However, several obstacles, such as prolonged scan times, limitations in patients' breath-hold capacity, and contrast agent-associated artifacts, remain in abdominal MR images. Recent techniques, including parallel imaging, three-dimensional acquisition, compressed sensing, and deep learning, have been developed to reduce the scan time while ensuring acceptable image quality or to achieve higher resolution without extending the scan duration. Quantitative measurements using MRI techniques enable the noninvasive evaluation of specific materials. A comprehensive understanding of these advanced techniques is essential for accurate interpretation of MRI sequences. Herein, we therefore review advanced abdominal MRI techniques.

Evaluation of MRI proton density fat fraction in hepatic steatosis: a systematic review and meta-analysis

BACKGROUND

Amidst the global rise of metabolic dysfunction-associated steatotic liver disease (MASLD), driven by increasing obesity rates, there is a pressing need for precise, non-invasive diagnostic tools. Our research aims to validate MRI Proton Density Fat Fraction (MRI-PDFF) utility, compared to liver biopsy, in grading hepatic steatosis in MASLD.

METHODS

A systematic search was conducted across Embase, PubMed/Medline, Scopus, and Web of Science until January 13, 2024, selecting studies that compare MRI-PDFF with liver biopsy for hepatic steatosis grading, defined as grades 0 (< 5% steatosis), 1 (5-33% steatosis), 2 (34-66% steatosis), and 3 (> 66% steatosis).

RESULTS

Twenty-two studies with 2844 patients were included. The analysis showed high accuracy of MRI-PDFF with AUCs of 0.97 (95% CI = 0.96-0.98) for grade 0 vs ≥ 1, 0.91 (95% CI = 0.88-0.93) for ≤ 1 vs ≥ 2, and 0.91 (95% CI = 0.88-0.93) for ≤ 2 vs 3, diagnostic odds ratio (DOR) from 98.74 (95% CI = 58.61-166.33) to 23.36 (95% CI = 13.76-39.68), sensitivity and specificity from 0.93 (95% CI = 0.88-0.96) to 0.76 (95% CI = 0.63-0.85) and 0.93 (95% CI = 0.88-0.96) to 0.89 (95% CI = 0.84-0.93), respectively. Likelihood ratio (LR) + ranged from 13.3 (95% CI = 7.4-24.0) to 7.2 (95% CI = 4.9-10.5), and LR - from 0.08 (95% CI = 0.05-0.13) to 0.27 (95% CI = 0.17-0.42). The proposed MRI-PDFF threshold of 5.7% for liver fat content emerges as a potential cut-off for the discrimination between grade 0 vs ≥ 1 (p = 0.075).

CONCLUSION

MRI-PDFF is a precise non-invasive technique for diagnosing and grading hepatic steatosis, warranting further studies to establish its diagnostic thresholds.

CLINICAL RELEVANCE STATEMENT

This study underscores the high diagnostic accuracy of MRI-PDFF for distinguishing between various grades of hepatic steatosis for early detection and management of MASLD, though further research is necessary for broader application.

KEY POINTS

MRI-PDFF offers precision in diagnosing and monitoring hepatic steatosis. The diagnostic accuracy of MRI-PDFF decreases as the grade of hepatic steatosis advances. A 5.7% MRI-PDFF threshold differentiates steatotic from non-steatotic livers.

Utilizing fully-automated 3D organ segmentation for hepatic steatosis assessment with CT attenuation-based parameters

OBJECTIVES

To investigate the clinical utility of fully-automated 3D organ segmentation in assessing hepatic steatosis on pre-contrast and post-contrast CT images using magnetic resonance spectroscopy (MRS)-proton density fat fraction (PDFF) as reference standard.

MATERIALS AND METHODS

This retrospective study analyzed 362 adult potential living liver donors with abdominal CT scans and MRS-PDFF. Using a deep learning-based tool, mean volumetric CT attenuation of the liver and spleen were measured on pre-contrast (liver(L)_pre and spleen(S)_pre) and post-contrast (L_post and S_post) images. Agreements between volumetric and manual region-of-interest (ROI)-based measurements were assessed using the intraclass correlation coefficient (ICC) and Bland-Altman analysis. Diagnostic performances of volumetric parameters (L_pre, liver-minus-spleen (L-S)_pre, L_post, and L-S_post) were evaluated for detecting MRS-PDFF ≥ 5% and ≥ 10% using receiver operating characteristic (ROC) curve analysis and compared with those of ROI-based parameters.

RESULTS

Among the 362 subjects, 105 and 35 had hepatic steatosis with MRS-PDFF ≥ 5% and ≥ 10%, respectively. Volumetric and ROI-based measurements revealed ICCs of 0.974, 0.825, 0.992, and 0.962, with mean differences of -4.2 HU, -3.4 HU, -1.2 HU, and -7.7 HU for L_pre, S_pre, L_post, and S_post, respectively. Volumetric L_pre, L-S_pre, L_post, and L-S_post yielded areas under the ROC curve of 0.813, 0.813, 0.734, and 0.817 for MRS-PDFF ≥ 5%; and 0.901, 0.915, 0.818, and 0.868 for MRS-PDFF ≥ 10%, comparable with those of ROI-based parameters (0.735-0.818; and 0.816-0.895, Ps = 0.228-0.911).

CONCLUSION

Automated 3D segmentation of the liver and spleen in CT scans can provide volumetric CT attenuation-based parameters to detect and grade hepatic steatosis, applicable to pre-contrast and post-contrast images.

CLINICAL RELEVANCE STATEMENT

Volumetric CT attenuation-based parameters of the liver and spleen, obtained through automated segmentation tools from pre-contrast or post-contrast CT scans, can efficiently detect and grade hepatic steatosis, making them applicable for large population data collection.

KEY POINTS

• Automated organ segmentation enables the extraction of CT attenuation-based parameters for the target organ. • Volumetric liver and spleen CT attenuation-based parameters are highly accurate in hepatic steatosis assessment. • Automated CT measurements from pre- or post-contrast imaging show promise for hepatic steatosis screening in large cohorts.

Development and validation of MRI-PDFF cutoffs for living liver donor eligibility assessment

Hepatic steatosis (HS) criteria for living donor liver transplantation (LDLT) donor eligibility should be based on large droplet fat as per Banff consensus recommendations. We aimed to establish magnetic resonance imaging proton density fat fraction cutoffs for HS assessment in potential LDLT donors. This retrospective study included consecutive potential LDLT donors who underwent MRI and liver biopsy between 2013 and 2023 at 2 tertiary institutions, each as development (n = 3062; 2015 men; median [IQR] age of 32 [25-38] y) and external validation (n = 472; 287 men; 35 [26-44] y) data sets. Proton density fat fraction (PDFF) was measured using dedicated MRI sequences. Histologic HS, defined as a large droplet fat fraction, was used as the reference standard. Dual PDFF cutoffs aimed at 95% sensitivity or 95% specificity, for diagnosing histologic HS of ≥10%, ≥20%, ≥30%, and ≥40%, were determined in the development data set using 10-fold cross-validation. The cutoffs were then validated in the external validation data set. The equation for estimating histologic HS from PDFF was also derived using linear regression. The PDFF cutoffs for histologic HS of ≥10%, ≥20%, ≥30%, and ≥40%, targeting 95% sensitivity, were 3.7%, 5.5%, 8.0%, and 10.0%, respectively. External validation demonstrated high sensitivities ≥97.9% with specificities ranging from 60.9% to 95.1%. The PDFF cutoffs targeting 95% specificity were 6.3%, 8.0%, 9.1%, and 10.1%, respectively. External validation rendered high specificities ranging from 88.5% to 95.3%, with sensitivities ranging from 76.6% to 100%. For diagnosing histologic HS ≥30%, which is the most prevalently used threshold for LDLT donor eligibility assessment, the PDFF cutoffs achieved sensitivities and specificities of over 90%. The equation of (Histologic HS = -2.95 + 1.93 × PDFF) was derived.

Validation of Diagnostic Thresholds for Compensated Advanced Chronic Liver Disease Using Supersonic Shear Imaging

Background The Society of Radiologists in Ultrasound (SRU) has proposed thresholds for acoustic radiation force impulse techniques to diagnose compensated advanced chronic liver disease (cACLD). However, the diagnostic performance of these thresholds has not been extensively validated. Purpose To validate the SRU thresholds in patients with chronic liver disease who underwent supersonic shear imaging and, if suboptimal diagnostic performance is observed, to identify optimal values for diagnosing cACLD. Materials and Methods This retrospective single-center study included high-risk patients with chronic liver disease who had liver stiffness (LS) measurements and had undergone endoscopy or liver biopsy between January 2018 and December 2021. Patients were randomly allocated to test and validation sets. cACLD was defined as varices at endoscopy and/or severe fibrosis or cirrhosis at liver biopsy. The diagnostic performance of the SRU guidelines was evaluated, and optimal threshold values were identified using receiver operating characteristic (ROC) curve analysis. Results A total of 1180 patients (median age, 57 years [IQR, 50-64 years]; 761 men), of whom 544 (46%) had cACLD, were included. With the SRU recommended thresholds of less than 9 kPa and greater than 13 kPa in the test set (n = 786), the sensitivity and specificity for ruling out and ruling in cACLD were 81% (303 of 374 patients; 95% CI: 77, 85) and 92% (380 of 412 patients; 95% CI: 89, 94), respectively. In ROC curve analysis, the identified optimal threshold values were less than 7 kPa and greater than 12 kPa, showing 91% sensitivity (340 of 374 patients; 95% CI: 88, 93) for ruling out cACLD and 91% specificity (373 of 412 patients; 95% CI: 87, 93) for ruling in cACLD, respectively. In the validation set (n = 394), the optimal thresholds showed 91% sensitivity (155 of 170 patients; 95% CI: 86, 95) and 92% specificity (206 of 224 patients; 95% CI: 88, 95). Conclusion Compared with the SRU guidelines, the dual LS threshold values of less than 7 kPa and greater than 12 kPa were better for diagnosing cACLD. © RSNA, 2024 Supplemental material is available for this article. See also the editorial by Barr in this issue.

Diagnosis and assessment of disease severity in patients with nonalcoholic fatty liver disease

Nonalcoholic fatty liver disease (NAFLD) includes simple steatosis, nonalcoholic steatohepatitis (NASH), fibrosis, and eventually cirrhosis and hepatocellular carcinoma (HCC). The diagnosis of NAFLD is based on the detection of excess fat disposition in the liver, which is the first step to trigger further evaluation of NAFLD, including necroinflammation and fibrosis. In this review, we discuss non-invasive biomarkers and imaging tools that are currently and potentially available for different features (steatosis, necroinflammation and fibrosis) and disease severity assessment of NAFLD. In the past 2 decades, advances in non-invasive tests of fibrosis have transformed the management of NAFLD. Blood and imaging biomarkers have already been evaluated in multiple studies for the diagnosis of fibrosis and cirrhosis. Among the various histological features of NAFLD, the degree of fibrosis has the strongest correlation with liver-related morbidity and mortality. Non-invasive tests of fibrosis have been shown to predict liver-related outcomes, both in the general population and among patients with NAFLD. What is lacking, however, is good data to support the use of non-invasive tests as monitoring and response biomarkers. With the conclusion of several large phase 3 studies in the next few years, the availability of paired liver biopsy, non-invasive test and clinical outcome data will likely advance the field and shed light on new biomarkers and the way to use various non-invasive tests in a longitudinal manner.

Ultrasound-derived fat fraction for detection of hepatic steatosis and quantification of liver fat content

PURPOSE

To compare ultrasound (US) and US-derived fat fraction (UDFF) with magnetic resonance proton density fat fraction (MRI-PDFF) for the detection of hepatic steatosis and quantification of liver fat content.

MATERIALS AND METHODS

Between October and December 2022, 149 patients scheduled for an abdominal MRI agreed to participate in this study and underwent MRI-PDFF, US and UDFF. Inclusion criteria were: (a) no chronic liver disease or jaundice; (b) no MRI motion artifacts; (c) adequate liver examination at US. Exclusion criteria were: (a) alcohol abuse, chronic hepatitis, cirrhosis, or jaundice; (b) MRI artifacts or insufficient US examination. The median of 10 MRI-PDFF and UDFF measurements in the right hepatic lobe was analyzed. UDFF and MRI-PDFF were compared by Bland-Altman difference plot and Pearson's test. Sensitivity, specificity, positive and negative predictive values, accuracy, and area under the receiver-operator curve (AUC-ROC) of US and UDFF were calculated using an MRI-PDFF cut-off value of 5%. p values ≤ 0.05 were statistically significant.

RESULTS

122 patients were included (61 men, mean age 60 years, standard deviation 15 years). The median MRI-PDFF value was 4.1% (interquartile range 2.9-6); 37.7% patients had a median MRI-PDFF value ≥ 5%. UDFF and MRI-PDFF had high agreement (p = 0.11) and positive correlation (⍴ = 0.81, p < 0.001). UDFF had a higher diagnostic value than US for the detection of steatosis, with AUC-ROCs of 0.75 (95% CI 0.65, 0.84) and 0.53 (95% CI 0.42, 0.64), respectively.

CONCLUSIONS

UDFF reliably quantifies liver fat content and improves the diagnostic value of US for the detection of hepatic steatosis.

Application of attenuation imaging for the detection of fatty liver in potential liver donors

PURPOSE

In living donor liver transplantation (LDLT), fatty liver adversely affects the outcome in donors or in recipients. The attenuation imaging (ATI) may be valuable for detecting fatty liver in potential liver donors. We aimed to investigate the role of ATI in screening liver donors.

METHOD

In this prospective study, potential liver donors undergoing MR examination, including proton MR spectroscopy (1H-MRS), were enrolled between January 2020 and December 2021 (study identifier: KCT0004486). All participants underwent ATI examinations to assess fatty liver disease. The reference standard for fatty liver was the hepatic fat fraction (HFF) on 1H-MRS, with 8% defined as the threshold for significant fatty liver. The correlation between attenuation coefficient (AC) and HFF was evaluated using Spearman's correlation coefficient. The diagnostic performance of AC for the detection of fatty liver disease was evaluated using receiver operating characteristic curve analysis.

RESULTS

A total of 102 participants (median age, 37 [range, 20-61] years; 65 men) were enrolled. Nineteen participants (18.6%) had significant fatty liver on 1H-MRS. AC significantly correlated with HFF on 1H-MRS (ρ = 0.674, P < 0.001), and was significantly higher in patients with HFF on 1H-MRS ≥ 8% than in patients with HFF on 1H-MRS < 8% (0.76 vs. 0.59, P < 0.001). By using the cutoff value of 0.66, the area under the curve of AC for the detection of significant fatty liver on 1H-MRS was 0.923 (95% confidence interval [CI]: 0.853-0.967), with sensitivity, specificity, and negative predictive values of 89.5% (95% CI: 66.9-98.7%), 83.1% (95% CI: 73.3-90.5%), and 97.2% (95% CI: 90.3-99.2%), respectively.

CONCLUSIONS

ATI showed good diagnostic performance with a high negative predictive value for the detection of significant fatty liver among potential liver donors.