Assessment of relevant hepatic steatosis in obese adolescents by rapid fat-selective GRE imaging with spatial-spectral excitation: a quantitative comparison with spectroscopic findings

Accelerating multi-echo chemical shift encoded water-fat MRI using model-guided deep learning

PURPOSE

To accelerate chemical shift encoded (CSE) water-fat imaging by applying a model-guided deep learning water-fat separation (MGDL-WF) framework to the undersampled k-space data.

METHODS

A model-guided deep learning water-fat separation framework is proposed for the acceleration using Cartesian/radial undersampling data. The proposed MGDL-WF combines the power of CSE water-fat imaging model and data-driven deep learning by jointly using a multi-peak fat model and a modified residual U-net network. The model is used to guide the image reconstruction, and the network is used to capture the artifacts induced by the undersampling. A data consistency layer is used in MGDL-WF to ensure the output images to be consistent with the k-space measurements. A Gauss-Newton iteration algorithm is adapted for the gradient updating of the networks.

RESULTS

Compared with the compressed sensing water-fat separation (CS-WF) algorithm/2-step procedure algorithm, the MGDL-WF increased peak signal-to-noise ratio (PSNR) by 5.31/5.23, 6.11/4.54, and 4.75 dB/1.88 dB with Cartesian sampling, and by 4.13/6.53, 2.90/4.68, and 1.68 dB/3.48 dB with radial sampling, at acceleration rates (R) of 4, 6, and 8, respectively. By using MGDL-WF, radial sampling increased the PSNR by 2.07 dB at R = 8, compared with Cartesian sampling.

CONCLUSIONS

The proposed MGDL-WF enables exploiting features of the water images and fat images from the undersampled multi-echo data, leading to improved performance in the accelerated CSE water-fat imaging. By using MGDL-WF, radial sampling can further improve the image quality with comparable scan time in comparison with Cartesian sampling.

A Practical Approach to Quantitative Grayscale Ultrasound Analysis of Hepatic Steatosis in Pediatric Patients Using a Picture Archiving and Communication System-Based Tool

OBJECTIVES

To evaluate the efficacy of a picture archiving and communication system (PACS)-based ultrasound (US) quantification technique for diagnosis of hepatic steatosis in a pediatric population.

METHODS

Abdominal US images of 49 pediatric patients (≤18 years) with histopathologically proven diagnoses of hepatic steatosis (n = 17), nonsteatotic liver disease (n = 19), and a normal liver (n = 13) were retrospectively reviewed. Patient demographics, the fibrosis stage, and the steatosis grade were obtained from the database. Quantitative grayscale measurements of the echo intensity level of the liver and kidneys were performed on the US images using the PACS measuring tool. The hepatorenal ratio was obtained by dividing mean liver by mean kidney values. The heterogeneity index for the liver was calculated by dividing the liver standard deviation by mean liver values. Hepatorenal ratio and heterogeneity index values of the 3 groups were correlated with pathologic results and compared by a 1-way analysis of variance. A receiver operating characteristic curve analysis was performed, and cutoff values were determined.

RESULTS

The hepatorenal ratio of the hepatic steatosis group was significantly greater than those of the control and nonsteatotic liver disease groups (P < .001). The heterogeneity index of the hepatic steatosis group was significantly greater than that of the control group (P = .046). For a hepatorenal ratio cutoff value of 1.5, 88.2% sensitivity, 91.4% specificity, 88.3% positive predictive value, and 94.1% negative predictive value were obtained for predicting hepatic steatosis.

CONCLUSIONS

PACS-based quantitative grayscale US quantification is a safe, accurate, and easily applicable objective method for the diagnosis of hepatic steatosis in children. A hepatorenal ratio of greater than 1.5 can be used as a conservative parameter, permitting increased confidence in discriminating hepatic steatosis from other conditions.

Quantification of liver fat: A comprehensive review

Fat accumulation in the liver causes metabolic diseases such as obesity, hypertension, diabetes or dyslipidemia by affecting insulin resistance, and increasing the risk of cardiac complications and cardiovascular disease mortality. Fatty liver diseases are often reversible in their early stage; therefore, there is a recognized need to detect their presence and to assess its severity to recognize fat-related functional abnormalities in the liver. This is crucial in evaluating living liver donors prior to transplantation because fat content in the liver can change liver regeneration in the recipient and donor. There are several methods to diagnose fatty liver, measure the amount of fat, and to classify and stage liver diseases (e.g. hepatic steatosis, steatohepatitis, fibrosis and cirrhosis): biopsy (the gold-standard procedure), clinical (medical physics based) and image analysis (semi or fully automated approaches). Liver biopsy has many drawbacks: it is invasive, inappropriate for monitoring (i.e., repeated evaluation), and assessment of steatosis is somewhat subjective. Qualitative biomarkers are mostly insufficient for accurate detection since fat has to be quantified by a varying threshold to measure disease severity. Therefore, a quantitative biomarker is required for detection of steatosis, accurate measurement of severity of diseases, clinical decision-making, prognosis and longitudinal monitoring of therapy. This study presents a comprehensive review of both clinical and automated image analysis based approaches to quantify liver fat and evaluate fatty liver diseases from different medical imaging modalities.

Systematic review to identify and appraise outcome measures used to evaluate childhood obesity treatment interventions (CoOR): evidence of purpose, application, validity, reliability and sensitivity

BACKGROUND

Lack of uniformity in outcome measures used in evaluations of childhood obesity treatment interventions can impede the ability to assess effectiveness and limits comparisons across trials.

OBJECTIVE

To identify and appraise outcome measures to produce a framework of recommended measures for use in evaluations of childhood obesity treatment interventions.

DATA SOURCES

Eleven electronic databases were searched between August and December 2011, including MEDLINE; MEDLINE In-Process and Other Non-Indexed Citations; EMBASE; PsycINFO; Health Management Information Consortium (HMIC); Allied and Complementary Medicine Database (AMED); Global Health, Maternity and Infant Care (all Ovid); Cumulative Index to Nursing and Allied Health Literature (CINAHL) (EBSCOhost); Science Citation Index (SCI) [Web of Science (WoS)]; and The Cochrane Library (Wiley) - from the date of inception, with no language restrictions. This was supported by review of relevant grey literature and trial databases.

REVIEW METHODS

Two searches were conducted to identify (1) outcome measures and corresponding citations used in published childhood obesity treatment evaluations and (2) manuscripts describing the development and/or evaluation of the outcome measures used in the childhood intervention obesity evaluations. Search 1 search strategy (review of trials) was modelled on elements of a review by Luttikhuis et al. (Oude Luttikhuis H, Baur L, Jansen H, Shrewsbury VA, O'Malley C, Stolk RP, et al. Interventions for treating obesity in children. Cochrane Database Syst Rev 2009;1:CD001872). Search 2 strategy (methodology papers) was built on Terwee et al.'s search filter (Terwee CB, Jansma EP, Riphagen II, de Vet HCW. Development of a methodological PubMed search filter for finding studies on measurement properties of measurement instruments. Qual Life Res 2009;18:1115-23). Eligible papers were appraised for quality initially by the internal project team. This was followed by an external appraisal by expert collaborators in order to agree which outcome measures should be recommended for the Childhood obesity Outcomes Review (CoOR) outcome measures framework.

RESULTS

Three hundred and seventy-nine manuscripts describing 180 outcome measures met eligibility criteria. Appraisal of these resulted in the recommendation of 36 measures for the CoOR outcome measures framework. Recommended primary outcome measures were body mass index (BMI) and dual-energy X-ray absorptiometry (DXA). Experts did not advocate any self-reported measures where objective measurement was possible (e.g. physical activity). Physiological outcomes hold potential to be primary outcomes, as they are indicators of cardiovascular health, but without evidence of what constitutes a minimally importance difference they have remained as secondary outcomes (although the corresponding lack of evidence for BMI and DXA is acknowledged). No preference-based quality-of-life measures were identified that would enable economic evaluation via calculation of quality-adjusted life-years. Few measures reported evaluating responsiveness.

LIMITATIONS

Proposed recommended measures are fit for use as outcome measures within studies that evaluate childhood obesity treatment evaluations specifically. These may or may not be suitable for other study designs, and some excluded measures may be more suitable in other study designs.

CONCLUSIONS

The CoOR outcome measures framework provides clear guidance of recommended primary and secondary outcome measures. This will enhance comparability between treatment evaluations and ensure that appropriate measures are being used. Where possible, future work should focus on modification and evaluation of existing measures rather than development of tools de nova. In addition, it is recommended that a similar outcome measures framework is produced to support evaluation of adult obesity programmes.

FUNDING

The National Institute for Health Research Health Technology Assessment programme.

Evidence and recommendations for imaging liver fat in children, based on systematic review

BACKGROUND & AIMS

Fatty liver is a common problem in children and increases their risk for cirrhosis, diabetes, and cardiovascular disease. Liver biopsy is the clinical standard for diagnosing and grading fatty liver. However, noninvasive imaging modalities are needed to assess liver fat in children. We performed a systematic review of studies that evaluated imaging liver fat in children.

METHODS

We searched PubMed for original research articles in peer-reviewed journals from January 1, 1982, through December 31, 2012, using the key words "imaging liver fat." Studies included those in English, and those performed in children from birth to 18 years of age. To be eligible for inclusion, studies were required to measure hepatic steatosis via an imaging modality and a quantitative comparator as the reference standard.

RESULTS

We analyzed 9 studies comprising 610 children; 4 studies assessed ultrasonography and 5 studies assessed magnetic resonance imaging (MRI). Ultrasonography was used in the diagnosis of fatty liver with positive predictive values of 47% to 62%. There was not a consistent relationship between ultrasound steatosis score and the reference measurement of hepatic steatosis. Liver fat as measurements by MRI or by spectroscopy varied with the methodologies used. Liver fat measurements by MRI correlated with results from histologic analyses, but sample size did not allow for an assessment of diagnostic accuracy.

CONCLUSIONS

Available evidence does not support the use of ultrasonography for the diagnosis or grading of fatty liver in children. Although MRI is a promising approach, the data are insufficient to make evidence-based recommendations regarding its use in children for the assessment of hepatic steatosis.

Relationships of body composition and liver fat content with insulin resistance in obesity-matched adolescents and adults

OBJECTIVE

While in adults not total body- or visceral fat mass, but liver fat content was found to independently determine insulin resistance, it is unclear whether these relationships are already present in obese adolescents.

METHODS

Thirty-nine overweight/obese adolescents were matched for sex and BMI with 39 adults. To compare the age- and sex-specific BMI values of adolescents and adults, the percentile value of each adolescent was projected to the age of 18. Body fat depots were quantified by whole-body magnetic resonance (MR) imaging. Liver fat content was measured with (1)H-MR spectroscopy. Insulin resistance was estimated from the homeostasis model assessment of insulin resistance (HOMA-IR).

RESULTS

Compared to overweight and obese adults, adolescents had higher HOMA-IR (P < 0.001) and lower lean body mass (P = 0.002). Furthermore, they had higher total body- (P = 0.02), but lower visceral- (P < 0.001) fat mass, while liver fat content was not significantly different between the groups (P = 0.16). In both groups liver fat content (both P ≤ 0.007), but not total body- or visceral fat mass (all P ≥ 0.64) was an independent predictor of insulin resistance.

CONCLUSIONS

Having lower visceral fat mass, overweight and obese adolescents are more insulin resistant than sex- and BMI-matched adults. Liver fat content, but not total body- or visceral fat mass, is an independent determinant of insulin resistance in adolescents.

Serial measurement of hepatic lipids during chemotherapy in patients with colorectal cancer: a 1 H MRS study

Hepatic steatosis is a hallmark of chemotherapy-induced liver injury. We made serial (1) H MRS measurements of hepatic lipids in patients over the time course of a 24-week chemotherapeutic regimen to determine whether (1) H MRS could be used to monitor the progression of chemotherapy-induced steatosis. Thirty-four patients with stage III or IV colorectal cancer receiving 5-fluorouracil, folinic acid and oxaliplatin (n=21) or hepatic arterial infusion of floxuridine with systemic irinotecan (n=13) were studied prospectively. (1) H MRS studies were performed at baseline and after 6 and 24 weeks of treatment. A (1) H MR spectrum was acquired from the liver during a breath hold and the ratio of fat to fat+water (FFW) was calculated to give a measure of hepatic triglycerides (HTGCs). The methodology was histologically validated in 18 patients and the reproducibility was assessed in 16 normal volunteers. Twenty-seven patients completed baseline, 6-week and 24-week (1) H MRS examinations and one was censored. Thirteen of 26 patients (50%) showed an increase in FFW after completion of treatment. Six patients (23%) developed hepatic steatosis and two patients converted from steatosis to nonsteatotic liver. Patients whose 6-week hepatic lipid levels had increased significantly relative to baseline also had a high probability of lipid elevation relative to baseline at the completion of treatment. Serial (1) H MRS is effective for the monitoring of HTGC changes during chemotherapy and for the detection of chemotherapy-associated steatosis. Six of 26 patients developed steatosis during chemotherapy. Lipid changes were observable at 6 weeks.

Two- versus three-dimensional dual gradient-echo MRI of the liver: a technical comparison

OBJECTIVE

To compare 2D spoiled dual gradient-echo (SPGR-DE) and 3D SPGR-DE with fat and water separation for the assessment of focal and diffuse fatty infiltration of the liver.

METHODS

A total of 227 consecutive patients (141 men; 56 ± 14 years) underwent clinically indicated liver MRI at 1.5 T including multiple-breath-hold 2D SPGR-DE and single-breath-hold 3D SPGR-DE with automatic reconstruction of fat-only images. Two readers assessed the image quality and number of fat-containing liver lesions on 2D and 3D in- and opposed-phase (IP/OP) images. Liver fat content (LFC) was quantified in 138 patients without chronic liver disease from 2D, 3D IP/OP, and 3D fat-only images.

RESULTS

Mean durations of 3D and 2D SPGR-DE acquisitions were 23.7 ± 2.9 and 97.2 ± 9.1 s respectively. The quality of all 2D and 3D images was rated diagnostically. Three-dimensional SPGR-DE revealed significantly more breathing artefacts resulting in lower image quality (P < 0.001); 2D and 3D IP/OP showed a similar detection rate of fat-containing lesions (P = 0.334) and similar LFC estimations (mean: +0.4 %; P = 0.048). LFC estimations based on 3D fat-only images showed significantly higher values (mean: 2.7 % + 3.5 %) than those from 2D and 3D IP/OP images (P < 0.001).

CONCLUSION

Three dimensional SPGR-DE performs as well as 2D SPGR-DE for the assessment of focal and diffuse fatty infiltration of liver parenchyma. The 3D SPGR-DE sequence used was quicker but more susceptible to breathing artefacts. Significantly higher LFC values are derived from 3D fat-only images than from 2D or 3D IP/OP images.