Reader agreement and accuracy of ultrasound features for hepatic steatosis

Qualitative assessment of hepatic steatosis on modern grayscale ultrasound: more accurate than previously thought?

OBJECTIVE

To assess the diagnostic performance of standardized qualitative assessment of hepatic steatosis on grayscale ultrasound.

METHODS

This retrospective, single-center, multi-case, multi-reader study included 200 patients with ultrasound examinations of the liver. Three readers assessed hepatic steatosis based on a standardized system of 3 ultrasound features: presence of increased fine echoes, visualization of right hemidiaphragm, and visualization of portal triads, assigning a four-grade category (normal, mild, moderate, or severe). Magnetic resonance imaging proton density fat fraction (MRI-PDFF) was used as reference standard. Binary discrimination (normal vs. steatosis) was summarized with binary area under the curve (AUC), sensitivity, and specificity. Discrimination across four categories was performed with pairwise comparisons. Reader differences were tested with the Obuchowski-Rockette-Hillis model. Inter-reader agreement was calculated with Gwet's agreement coefficient (AC).

RESULTS

Of the 200 patients, 27% (54/200) had normal liver (MRI-PDFF < 5%), 35% (70/200) had mild steatosis (MRI-PDFF ≥ 5-17.4%), 15% (29/200) had moderate steatosis (MRI-PDFF > 17.4-22.1%), and 24% (47/200) had severe steatosis (MRI-PDFF > 22.1%). Median time interval between ultrasound and MRI exams was 4 days (IQR 1-28). Sensitivity, specific, and binary AUC for readers 1/2/3 were 90%/82%/94%, 65%/82%/54%, and 0.87/0.85/0.88 with no statistically significant difference between readers (p = 0.46). Four-class category analysis showed excellent performance of ultrasound to distinguish extreme categories (AUC > 0.95 for normal vs. severe). Inter-reader agreement was substantial (Gwet's AC 0.63) for steatosis category assignment and moderate to substantial (Gwet's AC 0.55-0.71) for ultrasound features.

CONCLUSION

Contrary to popular belief, qualitative ultrasound assessment of hepatic steatosis is accurate in detecting and grading steatosis when evaluation criteria are standardized.

Ultrasound evaluation of chronic liver disease

Chronic liver disease is a world-wide epidemic. Any etiology that causes inflammation in the liver will lead to chronic liver disease. Presently, the most common inciting factor worldwide is steatotic liver disease. Recent advances in ultrasound imaging provide a multiparametric ultrasound methodology of diagnosing, staging, and monitoring treatment of chronic liver disease. Elastography has become a standard of care technique for the evaluation of liver fibrosis. Quantitative ultrasound allows for determination of the degree of fatty infiltration of the liver. Portal hypertension is the most important factor in determination of liver decompensation. B-mode findings combined with Doppler, and elastography techniques provide qualitative and quantitative methods of determining clinically significant portal hypertension. A newer method using contrast enhanced ultrasound may allow for a non-invasive quantitative estimation of the portal pressures. This paper reviews the use of multiparametric ultrasound in the evaluation of chronic liver disease including conventional B-mode ultrasound, Doppler, elastography and quantitative ultrasound for estimation of liver fat. The recent guidelines are presented and advised protocols reviewed.

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.

Hepatic steatosis: Qualitative and quantitative sonographic assessment in comparison to histology

Introduction

Globally, B-mode ultrasound is the most common modality used for the diagnosis of hepatic steatosis. We aimed to assess the correlation between qualitative liver ultrasound parameters, attenuation imaging (ATI) and histopathology-diagnosed steatosis grade obtained from liver biopsy. Our secondary aim was to examine the interobserver variability of qualitative ultrasound features.

Methods

A retrospective cohort study was performed which included adult patients (age ≥ 18 years) who had same-day liver ultrasound, ATI and liver biopsy for grading hepatic steatosis severity between 2018 and 2022. The qualitative US features for hepatic steatosis were independently scored by three radiologists and interobserver variability was examined. Histologic steatosis grade, ATI and qualitative ultrasound parameters were compared.

Results

Ninety patients were included; 67% female with a median age of 54 (IQR 39-65) years. The radiologist's overall impression had the highest correlation (very strongly correlated) with histologic steatosis grade (r = 0.82, P < 0.001). ATI coefficient and all qualitative ultrasound B-mode features except for liver echotexture and focal fat sparing were strongly correlated with histologic steatosis grade (r ≥ 0.70, P < 0.001). Most qualitative ultrasound features had good agreement between observers (Kappa statistic 0.61-1.0, P < 0.001), (Kendall coefficient 0.92, P < 0.001).

Conclusion

The examined qualitative ultrasound parameters and ATI had good-excellent performance for diagnosing clinically significant hepatic steatosis; however, the radiologist's overall impression had the best correlation with histologic steatosis grade. Our findings suggest an ongoing role for qualitative liver ultrasound assessment of hepatic steatosis despite the emergence of newer quantitative measures.

Early and accurate diagnosis of steatotic liver by artificial intelligence (AI)-supported ultrasonography

OBJECTIVES

Steatotic liver disease is the most frequent chronic liver disease worldwide. Ultrasonography (US) is commonly employed for the assessment and diagnosis. Few information is available on the possible use of artificial intelligence (AI) to ameliorate the diagnostic accuracy of ultrasonography.

MATERIALS AND METHODS

An AI-based algorithm was developed using a dataset of US images. We prospectively enrolled 134 patients for algorithm validation. Patients underwent abdominal US and Proton Density Fat Fraction MRI scans (MRI-PDFF), assumed as reference technique. The hepatorenal index was manually calculated (HRIM) by 4 operators. An automatic hepatorenal index (HRIA) was obtained by the algorithm. The accuracy of HRIA to discriminate steatosis grades was evaluated by ROC analysis using MRI-PDFF cut-offs.

RESULTS

Overweight was 40 % of subjects (BMI 26.4 kg/cm2). The median HRIA was 1.11 (IQR 0.32) and the average of 4 manually calculated HRIM was 1.08 (IQR 0.26), with a 15 % inter-operator variability. Both HRIA (R = 0.79, P < 0.0001) and HRIM (R = 0.69, P < 0.0001) significantly correlated with liver fat percentage (MRI-PDFF). According to MRI-PDFF, 32 % of enrolled subjects had steatosis. Discrimination capacity by AUC between patient with steatosis and patient without steatosis was better for HRIA than HRIM (AUC: 0.87 vs. 0.82, respectively). ROC analysis showed an AUC = 0.98 for HRIA with 1.64 cut-off in distinguishing between mild and moderate/severe groups.

CONCLUSIONS

The use of AI improves accuracy and speed of ultrasonography in the diagnosis of liver steatosis. Further studies should evaluate the routine use of this technique in the management of liver steatosis at high cardio-metabolic risk.

Ultrasonic Sound Speed Estimation for Liver Fat Quantification: A Review by the AIUM-RSNA QIBA Pulse-Echo Quantitative Ultrasound Initiative

Non-alcoholic fatty liver disease (NAFLD) is a significant cause of diffuse liver disease, morbidity and mortality worldwide. Early and accurate diagnosis of NALFD is critical to identify patients at risk of disease progression. Liver biopsy is the current gold standard for diagnosis and prognosis. However, a non-invasive diagnostic tool is desired because of the high cost and risk of complications of tissue sampling. Medical ultrasound is a safe, inexpensive and widely available imaging tool for diagnosing NAFLD. Emerging sonographic tools to quantitatively estimate hepatic fat fraction, such as tissue sound speed estimation, are likely to improve diagnostic accuracy, precision and reproducibility compared with existing qualitative and semi-quantitative techniques. Various pulse-echo ultrasound speed of sound estimation methodologies have been investigated, and some have been recently commercialized. We review state-of-the-art in vivo speed of sound estimation techniques, including their advantages, limitations, technical sources of variability, biological confounders and existing commercial implementations. We report the expected range of hepatic speed of sound as a function of liver steatosis and fibrosis that may be encountered in clinical practice. Ongoing efforts seek to quantify sound speed measurement accuracy and precision to inform threshold development around meaningful differences in fat fraction and between sequential measurements.

US Quantification of Liver Fat: Past, Present, and Future

Fatty liver disease has a high and increasing prevalence worldwide, is associated with adverse cardiovascular events and higher long-term medical costs, and may lead to liver-related morbidity and mortality. There is an urgent need for accurate, reproducible, accessible, and noninvasive techniques appropriate for detecting and quantifying liver fat in the general population and for monitoring treatment response in at-risk patients. CT may play a potential role in opportunistic screening, and MRI proton-density fat fraction provides high accuracy for liver fat quantification; however, these imaging modalities may not be suited for widespread screening and surveillance, given the high global prevalence. US, a safe and widely available modality, is well positioned as a screening and surveillance tool. Although well-established qualitative signs of liver fat perform well in moderate and severe steatosis, these signs are less reliable for grading mild steatosis and are likely unreliable for detecting subtle changes over time. New and emerging quantitative biomarkers of liver fat, such as those based on standardized measurements of attenuation, backscatter, and speed of sound, hold promise. Evolving techniques such as multiparametric modeling, radiofrequency envelope analysis, and artificial intelligence-based tools are also on the horizon. The authors discuss the societal impact of fatty liver disease, summarize the current state of liver fat quantification with CT and MRI, and describe past, currently available, and potential future US-based techniques for evaluating liver fat. For each US-based technique, they describe the concept, measurement method, advantages, and limitations. © RSNA, 2023 Online supplemental material is available for this article. Quiz questions for this article are available through the Online Learning Center.

Ultrasound-based hepatic fat quantification: current status and future directions

Non-alcoholic fatty liver disease (NAFLD) is a spectrum of disease from fatty accumulation (steatosis), necro-inflammation though to fibrosis. It is of increasing global prevalence as a hepatic manifestation of the metabolic syndrome. Although accurate histopathology and magnetic resonance imaging techniques for hepatic fat quantification exist, these are limited by invasiveness and availability, respectively. Ultrasonography is potentially ideal for assessing and monitoring hepatic steatosis given the examination is rapid and readily available. Traditional ultrasound methods include qualitative B-mode for imaging markers, such as increased hepatic parenchymal echogenicity compared to adjacent renal cortex are commonplace; however, there is acknowledged significant interobserver variability and they are suboptimal for detecting mild steatosis. Recently quantitative ultrasound metrics have been investigated as biomarkers for hepatic steatosis. These methods rely on changes in backscatter, attenuation, and speed of sound differences encountered in a steatotic liver. Prospective studies using quantitative ultrasound parameters show good diagnostic performance even at low steatosis grades and in NAFLD. This review aims to define the clinical need for ultrasound-based assessments of liver steatosis, to describe briefly the physics that underpins the various techniques available, and to assess the evidence base for the effectiveness of the techniques that are available commercially from various ultrasound vendors.

Hepatic steatosis: Ultrasound assessment using attenuation imaging (ATI) with liver biopsy correlation

INTRODUCTION

Hepatic steatosis duration and severity are risk factors for liver fibrosis and cardiometabolic disease. We assessed the diagnostic accuracy of attenuation imaging (ATI), compared with histologic hepatosteatosis grading in adults with varied suspected liver pathologies.

METHODS

Liver biopsy was performed on 76 patients (51 women, 25 men) with non-malignant diffuse parenchymal liver disease, within 4 weeks of multiparametric liver ultrasound including attenuation imaging (ATI). Skin-liver capsule distance (SCD) and body mass index (BMI) were measured. Histologic steatosis was graded none (S0), mild (S1), moderate (S2) or severe (S3). We compared histology and sonographic parameters.

RESULTS

The median patient age was 50.5 (range 18-83) years and BMI 28.9 kg/m² (interquartile range 24.0-33.3). The distribution of histologic steatosis grade was S0 (44%), S1(17%), S2(30%) and S3(9%). Median ATI value for each biopsy steatosis grade was 0.60 (IQR: 0.52-0.65), 0.65 (IQR: 0.6-0.71), 0.83 (IQR: 0.74-0.90) and 0.90 (IQR: 0.82-1.01) dB/cm/MHz for S0, S1, S2 and S3, respectively. The AUC of ATI for detection of any steatosis (S1-S3) and moderate to severe steatosis (S2-S3) was 0.85 (95% CI: 0.75-0.91) and 0.91 (95% CI: 0.83-0.99) with cut-offs of 0.55 and 0.62 dB/cm/MHz. ATI threshold of 0.74 dB/cm/MHz was able to discriminate between S0-S1 and S2-3 with accuracy, CI and kappa statistic of 0.8889, 0.65-0.98 and 0.7534.

CONCLUSION

We found a good correlation between ATI and steatosis grade. The most accurate discrimination was between none to mild (S0-1) and moderate to severe (S2-3) steatosis.

Correlation between hepatorenal index and attenuation imaging for assessing hepatic steatosis

INTRODUCTION

Hepatic steatosis screening is required to assess high-risk populations, identify those for intervention, monitor response and prevent disease progression and complications. Liver biopsy and magnetic resonance imaging proton density fat fraction are current gold standards, but are limited by biopsy risk factors, patient tolerance and cost. Non-invasive, cost-effective, semi-quantitative and quantitative ultrasound assessment exists. The aim of this study was to assess the correlation between the semi-quantitative hepatorenal index (HRI) to assess hepatic steatosis using the quantitative attenuation imaging (ATI) as a reference standard, in adults with varied suspected liver pathologies.

METHODS

Data were collected prospectively between April 2019 and March 2020 at a tertiary institution on any patient >18 years referred to US assessment of suspected liver pathology. The only exclusion criteria were absent or invalid HRI or ATI measurements. Three hundred fifty eight patients were included.

RESULTS

There was a significant weak positive correlation between HRI and ATI (r = 0.351, P < 0.001) and between HRI steatosis grade (SG) and ATI SG (r = 0.329, P < 0.001), using previously established cut-off values. With ATI as the reference standard, there was no significant correlation between HRI and hepatic steatosis within steatosis grades, nor for no (SG = 0) or any (SG > 0) hepatic steatosis.

CONCLUSIONS

Our study in a typical heterogeneous clinical population suggests the semi-quantitative HRI is of limited use in hepatic steatosis imaging. As HRI is the objective measure of the subjective brightness (B)-mode assessment, this imaging feature may not be as reliable as previously thought. Quantitative ATI may be the preferred non-invasive technique for hepatic steatosis assessment.

Conventional ultrasound for diagnosis of hepatic steatosis is better than believed

BACKGROUND

Hepatic steatosis is a condition frequently encountered in clinical practice, with potential progression towards fibrosis, cirrhosis, and hepatocellular carcinoma. Detection and staging of hepatic steatosis are of most importance in nonalcoholic fatty liver disease (NAFLD), a disease with a high prevalence of more than 1 billion individuals affected. Ultrasound (US) is one of the most used noninvasive imaging techniques used in the diagnosis of hepatic steatosis. Detection of hepatic steatosis with US relies on several conventional US parameters, which will be described. US is the first-choice imaging in adults at risk for hepatic steatosis. The use of some scoring systems may add additional accuracy especially in assessing the severity of hepatic steatosis.

SUMMARY

In the presented paper, we discuss screening and risk stratification, ultrasound features for diagnosing hepatic steatosis, B-mode criteria, focal fatty patterns and Doppler features of the hepatic vessels, and the value of the different US signs for the diagnosis of liver steatosis including classifying the severity of steatosis using different US scores. Limitations of conventional B-mode and Doppler features in the evaluation of hepatic steatosis are also discussed, including those in grading and assessing the complications of steatosis, namely fibrosis and nonalcoholic steatohepatitis.

KEY MESSAGES

Ultrasound is the first-line imaging examination for the screening and follow-up of patients with liver steatosis. The use of some scoring systems may add additional accuracy in assessing the severity of steatosis. Conventional B-mode and Doppler ultrasound have limitations in grading and assessing the complications of steatosis.

Accurate and generalizable quantitative scoring of liver steatosis from ultrasound images via scalable deep learning

BACKGROUND

Hepatic steatosis is a major cause of chronic liver disease. Two-dimensional (2D) ultrasound is the most widely used non-invasive tool for screening and monitoring, but associated diagnoses are highly subjective.

AIM

To develop a scalable deep learning (DL) algorithm for quantitative scoring of liver steatosis from 2D ultrasound images.

METHODS

Using multi-view ultrasound data from 3310 patients, 19513 studies, and 228075 images from a retrospective cohort of patients received elastography, we trained a DL algorithm to diagnose steatosis stages (healthy, mild, moderate, or severe) from clinical ultrasound diagnoses. Performance was validated on two multi-scanner unblinded and blinded (initially to DL developer) histology-proven cohorts (147 and 112 patients) with histopathology fatty cell percentage diagnoses and a subset with FibroScan diagnoses. We also quantified reliability across scanners and viewpoints. Results were evaluated using Bland-Altman and receiver operating characteristic (ROC) analysis.

RESULTS

The DL algorithm demonstrated repeatable measurements with a moderate number of images (three for each viewpoint) and high agreement across three premium ultrasound scanners. High diagnostic performance was observed across all viewpoints: Areas under the curve of the ROC to classify mild, moderate, and severe steatosis grades were 0.85, 0.91, and 0.93, respectively. The DL algorithm outperformed or performed at least comparably to FibroScan control attenuation parameter (CAP) with statistically significant improvements for all levels on the unblinded histology-proven cohort and for "= severe" steatosis on the blinded histology-proven cohort.

CONCLUSION

The DL algorithm provides a reliable quantitative steatosis assessment across view and scanners on two multi-scanner cohorts. Diagnostic performance was high with comparable or better performance than the CAP.

Clinical evaluation of grayscale and linear scale hepatorenal indices for fatty liver quantification: a prospective study of a native Chinese population

BACKGROUND AND AIMS

Hepato-renal index (HRI) has been investigated extensively in various clinical studies. New linear scale HRI (LS-HRI) is proposed as an alternative to conventional grayscale HRI (GS-HRI) that suffers from lack of a widely accepted cut-off value for differentiation of fatty from normal livers. To investigate the diagnostic performance of conventional GS-HRI and new LS-HRI for a relatively large Chinese population with NAFLD using a well-established ultrasonographic fatty liver indicator (US-FLI) as the reference standard for steatosis grades.

MATERIALS AND METHODS

A total of 106 patients with various stages of NAFLD were prospectively enrolled. All ultrasound images for these patients were first acquired by a highly experienced ultrasound doctor and their US-FLI scores then obtained by the same doctor. Both GS-HRI and LS-HRI values were measured off-line by two additional ultrasound doctors. Four steatosis grades were determined from US-FLI scores for steatosis detection and staging.

RESULTS

Inter-observer agreements for both GS-HRI and LS-HRI were excellent with the respective concordance correlation coefficient (CCC) of 0.900 for GS-HRI and 0.822 for LS-HRI. A linear correlation to US-FLI for LS-HRI (R = 0.74) was substantially superior to that for GS-HRI (R = 0.46). LS-HRI had a sensitivity of 85.9% and a specificity of 96.3% to differentiate steatosis from the normal liver (AUROC: 95.5%) while GS-HRI had a sensitivity of 85.9% and a specificity of 92.6% to distinguish steatosis from the normal liver (AUROC: 94.7%).

CONCLUSIONS

Both GS-HRI and LS-HRI measurements are reproducible between two ultrasonographic clinicians and are evidently effective for steatosis detection.

Liver Fat Assessment in Multiview Sonography Using Transfer Learning With Convolutional Neural Networks

OBJECTIVES

To develop and evaluate deep learning models devised for liver fat assessment based on ultrasound (US) images acquired from four different liver views: transverse plane (hepatic veins at the confluence with the inferior vena cava, right portal vein, right posterior portal vein) and sagittal plane (liver/kidney).

METHODS

US images (four separate views) were acquired from 135 participants with known or suspected nonalcoholic fatty liver disease. Proton density fat fraction (PDFF) values derived from chemical shift-encoded magnetic resonance imaging served as ground truth. Transfer learning with a deep convolutional neural network (CNN) was applied to develop models for diagnosis of fatty liver (PDFF ≥ 5%), diagnosis of advanced steatosis (PDFF ≥ 10%), and PDFF quantification for each liver view separately. In addition, an ensemble model based on all four liver view models was investigated. Diagnostic performance was assessed using the area under the receiver operating characteristics curve (AUC), and quantification was assessed using the Spearman correlation coefficient (SCC).

RESULTS

The most accurate single view was the right posterior portal vein, with an SCC of 0.78 for quantifying PDFF and AUC values of 0.90 (PDFF ≥ 5%) and 0.79 (PDFF ≥ 10%). The ensemble of models achieved an SCC of 0.81 and AUCs of 0.91 (PDFF ≥ 5%) and 0.86 (PDFF ≥ 10%).

CONCLUSION

Deep learning-based analysis of US images from different liver views can help assess liver fat.

Fatty liver disease in Living Liver Donors: A Single-Institute Experience of 220 Donors

We retrospectively reviewed 220 living liver donors, with a focus on the development of postoperative fatty liver. Data regarding demographics, comorbidities, imaging tests, operations, and biopsies were obtained from medical records. We used unenhanced CT and USG to diagnose fatty liver. Donor candidates with fatty liver underwent weight loss intervention until imaging tests no longer demonstrated any features of fatty liver. Among 220 donors, 61 were diagnosed with preoperative fatty liver. The mean BMI of these 61 donors significantly decreased from 24.9 at the first visit to 23.6 kg/m² immediately before surgery (p=0.0386). A multivariate analysis revealed the following significant risk factors for postoperative fatty liver: male sex (p=0.0033), BMI immediately before surgery (p=0.0028), and a history of treatment for preoperative fatty liver (p=0.0231). Postoperative fatty liver was often refractory to weight loss intervention. No improvement was observed in 14 of the 32 donors who had been diagnosed with fatty liver postoperatively, and 1 of the 14 donors even developed NASH. In conclusion, special attention should be paid to prevent fatty liver after surgery in male donors who show a high BMI immediately before surgery and with a history of treatment for preoperative fatty liver, and lifelong follow-up is recommended.

Susceptibility based multiparametric quantification of liver disease: Non-invasive evaluation of steatosis and iron overload

PURPOSE

To evaluate if single-voxel MR spectroscopy (MRS) of iron and fat correlates with biopsy results of hepatic steatosis and iron overload, and to compare MR-measurements with room-temperature susceptometer (RTS), ultrasound, controlled attenuation parameter (CAP) and serum ferritin.

MATERIAL AND METHODS

In this prospective study, a set of 42 patients out of 47 screened patients with several chronic liver diseases underwent MRI-examination at 1.5 T including R2-measurements by single-voxel high-speed T2-corrected multiecho spectroscopy, additional liver biopsy, abdominal ultrasound, CAP, and RTS. Routine blood and serum parameters were determined, including ferritin. Atomic absorption spectroscopy (AAS) and histologically confirmed extent of hepatic steatosis from liver biopsy were used as reference standard. For correlation of R2, RTS, CAP, ferritin, and ultrasound with results of AAS and histologically determined fat fraction of liver biopsy specimen, Spearman's and Pearson's correlation as well as receiver operating characteristics curve (ROC) analysis with cut-off values determined by maximizing Youden index was used.

RESULTS

MRS iron assessment correlated best with AAS, with a Pearson correlation coefficient of 0.715 (p < 0.001), followed by RTS 0.520 (p < 0.001), and serum ferritin 0.213 (p = 0.088, not significant). MRS fat quantification correlated best with the histological confirmed extent of steatosis hepatis with a Spearman correlation coefficient of 0.836 (p < 0.001), followed by CAP 0.604 (p < 0.001) and sonographically diagnosed steatosis 0.358 (p = 0.013).

CONCLUSION

MRS by T2-corrected multiecho single-voxel spectroscopy correlated best with histological results of hepatic fat and iron content compared to RTS, CAP, abdominal ultrasound, and ferritin. Non-invasive methods to assess hepatic fat and iron are of clinical interest for follow-up examinations of patients with chronic liver diseases, where repeated biopsy is not indicated.