Ultrasound biomicroscopy for the assessment of early-stage nonalcoholic fatty liver disease induced in rats by a high-fat diet
Article information
Abstract
Purpose
The aim of this study was to assess the ability of ultrasound biomicroscopy (UBM) to diagnose the initial stages of nonalcoholic fatty liver disease (NAFLD) in a rat model.
Methods
Eighteen male Wistar rats were allocated to control or experimental groups. A high-fat diet (HFD) with 20% fructose and 2% cholesterol, resembling a common Western diet, was fed to animals in the experimental groups for up to 16 weeks; those in the control group received a regular diet. A 21 MHz UBM system was used to acquire B-mode images at specific times: baseline (T0), 10 weeks (T10), and 16 weeks (T16). The sonographic hepatorenal index (SHRI), based on the average ultrasound image gray-level intensities from the liver parenchyma and right renal cortex, was determined at T0, T10, and T16. The liver specimen histology was classified using the modified Nonalcoholic Steatohepatitis Clinical Research Network NAFLD activity scoring system.
Results
The livers in the animals in the experimental groups progressed from sinusoidal congestion and moderate macro- and micro-vesicular steatosis to moderate steatosis and frequent hepatocyte ballooning. The SHRI obtained in the experimental group animals at T10 and T16 was significantly different from the SHRI of pooled control group. No significant difference existed between the SHRI in animals receiving HFD between T10 and T16.
Conclusion
SHRI measurement using UBM may be a promising noninvasive tool to characterize early-stage NAFLD in rat models.
Introduction
Obesity, a widespread disease throughout the world, has reached epidemic proportions; its age-standardized prevalence among adults (18 years and older) has increased globally by 1.5 times since 2000, and its crude prevalence in children (5-19 years) has more than doubled between 2000 and 2016 [1]. Obesity promotes an increased risk of heart disease, diabetes, depression, and many cancers, as well as premature death [2]. The global increase in obesity has been accompanied by an increase in nonalcoholic fatty liver disease (NAFLD), which refers to fat accumulation in the liver without excessive alcohol consumption and represents the hepatic manifestation of metabolic syndrome based on the consistent association between NAFLD and hypertension, dyslipidemia, insulin resistance, type 2 diabetes, and visceral adiposity [3]. Moreover, obesity and type 2 diabetes are strong risk factors for NAFLD [4].
The global pooled NAFLD prevalence is 25.24%, with wide geographical variation [5], and this prevalence ranges between 42.6% and 69.5% in patients affected by type 2 diabetes mellitus [6]. These epidemiological indicators show that NAFLD is a growing public health problem. Nonalcoholic steatohepatitis (NASH) represents its most severe form, resulting in chronic liver disease, liver cirrhosis, and hepatocellular carcinoma [7]. According to Schattenberg et al. [8], the prompt diagnosis and care of NASH patients will likely reduce future healthcare costs.
As discussed by Perumpail et al. [9], the increased prevalence of NAFLD poses challenges to healthcare systems regarding how to screen for the progressing epidemic of liver disease. Efforts to address these challenges should focus on understanding NAFLD pathophysiology and its treatment, which requires new diagnostic tools and methods to complement the diagnostic tests already employed, including liver biopsy as the gold standard, serum biomarkers and biomarker panels [10] and noninvasive imaging techniques such as hepatic ultrasonography [11], computed tomography, and magnetic resonance imaging [10]. NAFLD and NASH experimental models are also important elements of translational science in order to identify targets for diagnosis, monitoring, therapy, and histopathological outcomes in advance of costly clinical research [12]. Animal models have been used to investigate NAFLD and NASH [12-15], including C57BL/6 mice. Wistar and Sprague-Dawley rats are usually preferred due to their predilection to develop obesity, type 2 diabetes, NAFLD, and NASH [16,17].
In translational studies, animal models based on diet-induced NAFLD paralleling human dietary habits are predominant; these diets include a high-fat diet (HFD) and fructose [14,18], a monosaccharide primarily metabolized in the liver [18] and excessively consumed in Western societies. An HFD has between 45% and 75% of the kilocalories allocated to lipids and promotes a rat phenotype that is similar to human obesity, insulin resistance, glucose intolerance, and dyslipidemia. When a diet contains high levels of fat, cholesterol, and fructose, NAFLD progresses to NASH, and the three stages of NAFLD (steatosis, steatohepatitis with fibrosis, and cirrhosis) develop [19].
Concurrent with the reported animal diet-based models for NAFLD and NASH, ultrasound imaging methods have been used to diagnose liver disease in animals. In addition to providing a noninvasive method, ultrasound imaging is an adequate tool for detecting liver steatosis, longitudinally analyzing liver fibrosis progression, and diagnosing hepatic cirrhosis. Ultrasonographic methods consider the echographic relationship between the liver and the renal cortex [20,21], tissue characterization based on the statistical properties of the gray-level image intensity [22], elastography [23], or the liver parenchyma echotexture with [24] or without [25] the use of an ultrasound contrast agent. These ultrasound methods have been performed in vivo on rat models employing distinct approaches that did not mimic human dietary habits for diet-induced NAFLD. Furthermore, many of the aforementioned investigations based on rat models have evaluated hepatic impairment in its final stages.
The current work was conducted to diagnose the initial stages of NAFLD based on ultrasound biomicroscopy (UBM) images acquired in vivo from rats fed with HFD. The rationale for using UBM is that this technique uses ultrasonic frequencies higher than those typically employed in clinical ultrasonography to provide adequate spatial resolution to inspect small animals, with typical image resolutions of 30 µm (axial) and 72 µm (lateral) obtained using a 40 MHz UBM system [26]. The spatial resolution provided by high-frequency ultrasound provides more detailed images of tissue microstructure, at the expense of lower penetration in comparison with typically used frequencies in clinical ultrasonography. The backscattering decreases if the frequency is reduced and increases if the dimensions of the tissue microstructure are higher; and these compensating effects, in association with the spatial resolution, justify using high-frequency ultrasound in experiments with small animals and low-frequency ultrasound with human patients, considering the organs and microstructure dimensions in each situation.
The UBM images obtained in the current work were processed to generate the sonographic hepatorenal index (SHRI) [27], which was used for clinical quantification of liver steatosis and was calculated as the ratio between the average gray-level intensities of two regions of interest (ROI): one over the UBM image from liver parenchyma and another from the renal cortex [27]. The SHRI is a sensitive, reproducible, and operator-independent parameter used to quantify steatosis, even for small amounts of liver fat that would likely be missed using conventional ultrasonography. In human patients, the index cutoff points to predict steatosis greater than 25% and 60% are 1.86 and 2.23, respectively [27]. With rats receiving diethylnitrosamine orally over the course of 12 weeks to induce hepatic fibrosis, the SHRI ranged from 0.25 (baseline) to 0.46 and 0.53 when determined at 10 and 13 weeks, respectively [28].
Materials and Methods
Compliance with Ethical Standards
The experiments were conducted in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institute of Health. The local Ethics Committee for Animal Use in Research (CEUA) of the Biological Science Institute/Federal University of Rio de Janeiro approved the research protocol (061/19), which was prepared before the study and registered at CEUA. The Animal Research: Reporting of In Vivo Experiments (ARRIVE) guidelines were followed for the replacement, refinement, and reduction of animals. All the study data are available and maintained under the responsibility of the last author.
Study Design and Animals
Eighteen 2-month-old healthy male Wistar rats, Rattus norvegicus, from the Center of Experimental Surgery (Medical School from the Federal University of Rio de Janeiro, RJ, Brazil) and weighing between 250 and 300 g were selected from the vivarium, with female rats excluded because male sex is a predisposing factor for the development of NAFLD [16]. Each animal was anesthetized with 1.5% isoflurane in 1.5 L/min oxygen using a laboratory animal compact anesthesia system (Vevo Anesthesia System, VisualSonics, Toronto, Canada). The animals were marked for identification, and the identified animals were randomly assigned into four groups: control groups at 10 and 16 weeks, GCT10 (n=3) and GCT16 (n=3), respectively, and experimental groups at 10 and 16 weeks, GEP10 (n=6) and GEP16 (n=6), respectively. Animals in the same group were housed three per cage and maintained in a 12-hour light/dark cycle at 23°C. Food was offered ad libitum (25 g per animal per day). The animals in the GCT group (animals from both GCT10 and GCT16) received standard vivarium pelleted formula and water ad libitum, whereas those in GEP10 and GEP16 received an HFD rich in cholesterol and fructose for 10 and 16 weeks, respectively. The HFD (Table 1) was accompanied by water and a cola-based carbonated drink ad libitum. Adverse events for animals in GEP10 and GEP16 included diarrhea, which was more intense until the fifth week, when the signs started a gradual regression until complete disappearance at 8 weeks. The animals were weighed once a week in order to exclude any animal with significant weight loss.
The feeding periods of 10 and 16 weeks were selected based on the literatures [14,16], considering findings in mice fed an HFD. Ten weeks after receiving the HFD, the animals presented obesity, hyperlipidemia and hyperinsulinemia, and glucose intolerance after 12 weeks. Together, these signs represent a phenotype similar to the human disease [16]. Moreover, male C57BL/6 mice fed an HFD up to 16 weeks presented hepatic steatosis, hepatocyte ballooning, Mallory bodies, higher fasting serum glucose levels, and decreased adiponectin levels, suggesting hyperglycemia and insulin resistance [14]. Therefore, 16 weeks of feeding the animals with HFD seemed sufficient to promote lesions suggestive of NASH.
UBM Image Acquisition
Prior to UBM image acquisition using the experimental facilities at CENABIO (National Center of Structural Biology and Bioimaging), the animals were anesthetized with 2% isoflurane in 1.5 L/min oxygen in the supine position over a heated platform for rats (AS-11550, VisualSonics) and underwent abdominal trichotomy. The images were acquired using a UBM machine (Vevo 2100, VisualSonics) operating with a 21 MHz probe (MS-250, VisualSonics) oriented to acquire B-mode images in the sagittal plane containing parts of the liver and the right kidney cortex. Standard ultrasound gel was used to couple the probe and the animal skin. Fig. 1 depicts a picture of the imaging setup.
A cineloop with 100 frames of B-mode images was acquired during each animal examination and stored. The UBM image acquisition timeline comprised three events, beginning with T0, when all 18 animals had their images acquired and the animals in the GEP10 and GEP16 groups began the HFD. Ten weeks later (T10), the animals in the GCT10 (n=3) and GEP10 (n=6) groups had their images acquired. Finally, the animals in the GCT16 (n=3) and GEP16 (n=6) groups had images acquired at T16 (16 weeks after T0).
UBM Image Processing
The cineloop from each animal was inspected, supervised by an expert in ultrasound imaging of small animals, and three B-mode images (256 grayscale values) in the JPEG format were selected that contained views of the liver parenchyma and right renal cortex, with minor artifacts caused by respiratory or bowel motions. These three images were analyzed using the public domain Java image processing computational program ImageJ (version 1.47, NIH, Bethesda, MD, USA). The processing started with one ROI over the liver parenchyma and another over the renal cortex defined for each selected image. Each ROI was drawn based on the mouse pointer motion over the UBM selected image and using the LiveWire 2D tool, from ImageJ, to create the ROI closed contour. Subsequently, the mean gray-level intensities AIL,i and AIK,i, (1≤i≤3), for each ROI over the liver parenchyma and renal cortex, respectively, were computed using the Measure tool from ImageJ. The SHRI was then calculated as
Histological Analysis
Following UBM image acquisition at T10 or T16, the anesthetized animal was placed in a closed chamber saturated with isoflurane to induce euthanasia, and total hepatectomy was performed. The liver was weighed, fixed in a 10% neutral-buffered formalin solution, and sent en bloc to a single-blinded experienced hepatic pathologist that performed the cleavage in transverse sections encompassing all hepatic lobes, and produced 5-μm-thick sliced sections that were prepared for routine hematoxylin and eosin and picrosirius red staining. The hepatic pathologist evaluated all histological samples and graded them based on the modified version of the Nonalcoholic Steatohepatitis Clinical Research Network NAFLD activity scoring system [29], as shown in Table 2.
Statistical Analysis
Statistical tests were used to determine the sample size, the correlation coefficient between SHRI and animal weights, and significant differences among animal groups and for liver weights and the SHRI. The sample size, calculated based on the comparison of two means, used one-way analysis of variance (ANOVA) with pairwise 2-sided tests and the following statistical coefficients: 0.8, 1.0, and 1.2 for the SHRI means of the GCT, GEP10, and GEP16 groups, respectively; 0.15 for the standard deviation; 3 for the number of comparisons between pairs; 0.80 for statistical power; and 0.05 as the level of significance. Further statistical analysis was performed using Jamovi software (version 1.1.9), a free and open-source statistical platform built on top of the R statistical language. Initially, the Shapiro-Wilk test for normality was employed, and if any group of experimental data was not normally distributed, then the nonparametric Kruskal-Wallis test followed by Dwass-Steel-Critchlow-Fligner pairwise comparisons was applied to determine whether the medians of the groups were significantly different. However, for all groups with normally distributed experimental data, the equality of the variances was determined, and one-way ANOVA was applied followed by the post hoc Tukey (equal variances) or Games-Howell (unequal variances) test to determine the significance of differences between the means of the groups. Values of P<0.05 (default) were considered to indicate statistical significance.
Results
Animal Weights
Plots of the mean (±standard deviation) of WA/W0, the animal weight WA normalized by W0 (weight at T0), are presented for weeks 1 to 10 for the GCT10 and GEP10 groups and for weeks 1 to 16 for the GCT16 and GEP16 groups (Fig. 2). No significant weight loss occurred, and therefore no animal was excluded.
The WA/W0 results for the GCT10, GCT16, GEP10, and GEP16 groups were not normally distributed; therefore, a nonparametric method was employed to compare the group weight medians. During weeks 1-10, there were no significant differences between the median values of WA/W0 for the GCT10 and GCT16 groups. However, the median values of WA/W0 of the GCT10 and GCT16 groups were significantly different from those of the GEP16 group. Furthermore, there were significant differences between the WA/W0 medians of the GCT10 and GCT16 groups and those of the GEP10 group after week 5. Finally, there were no significant differences between the median WA/W0 values of the GCT10 and GCT16 groups, except at weeks 5, 7, and 8. There was a significant difference between the WA/W0 median values between the GCT16 and GEP16 groups during weeks 11-16.
SHRI
A typical UBM image containing sections of the rat liver and right kidney is depicted in Fig. 3, including the ROIs used to calculate AIL and AIK. Both ROIs had similar depths to provide similar ultrasound attenuation effects of the wave incidence on the liver and kidney ROIs.
The Kruskal-Wallis test revealed no significant difference between the SHRI medians of the four groups determined on T0 and the SHRI medians of the GCT10 group at T10 and the GCT16 group at T16. Therefore, a group GCB (n=24) was formed, pooling the SHRI results of the four groups determined at T0 (n=18) with those for the GCT10 (n=3) and GCT16 (n=3) groups determined at T10 and T16, respectively The Kruskal-Wallis test, followed by Dwass-Steel-Critchlow-Fligner pairwise comparisons, revealed a significant difference between the median SHRI values of the GCB and GEP10 groups and between the GCB and GEP16 groups (P<0.05). No significant difference existed between GEP10 and GEP16 (P=0.183). A boxplot (Fig. 4) summarizes these results.
Liver Weight
The comparative macroscopic HFD effects on the liver are shown (Fig. 5) for one animal from GCT, one from GEP10 and one from GEP16, euthanized at T16, T10, and T16, respectively. The hepatic parenchyma of the animal in the GCT group presented preserved macroscopic appearance, lobes with regular contours and a homogeneous brownish color. In contrast, the macroscopic changes in the liver of the animals from GEP10 and GEP16 presented heterogeneous coloration due to steatosis and vascular congestion. In addition, the liver of the rat from the GEP16 group had lobes with blunt edges.
The liver weights from animals in the GCT, GEP10, and GEP16 groups, either normalized to the corresponding animal weight measured just before euthanasia or in absolute values, were normally distributed. One-way ANOVA was then applied, followed by the Tukey post hoc test, and it was confirmed that the mean values of normalized liver weight showed significant (P<0.001) differences for GCT×GEP10 and GEP10×GEP16, as well as for GCT×GEP16 (P<0.01). The same post hoc test confirmed the absolute liver weights with significant (P<0.01) differences only for GCT×GEP10 and GEP10×GEP16. The liver weights for all animals in each group are depicted in column plots in Fig. 6A and B for normalized and absolute values, respectively, and in boxplots in Fig. 6C and D for normalized and absolute values, respectively.
Histopathological Findings
Typical histological images of the livers of animals from the GCT10, GEP10, and GEP16 groups are presented (Fig. 7). The hepatic pathologist described the liver parenchyma of the animal in the GCT10 group as having a preserved architectural pattern, normal portal tract dimensions, and no inflammation. The same liver also showed an absence of steatosis and other significant histopathological changes in adjacent hepatocytes. Typical histological images of the liver from one animal of each group GCT10, GEP10, and GEP16 are presented (Fig. 7). The hepatic pathologist described the liver parenchyma of the particular animal in the GCT10 group having a preserved architectural pattern, normal portal tract dimensions, and no inflammation. The same liver also showed an absence of steatosis and other significant histopathological changes in adjacent hepatocytes. On the other hand, the specific animal in GEP10 group presented its liver parenchyma showing sinusoidal congestion and moderate macro-and micro-vesicular steatosis. Finally, the liver of the particular animal from the GEP16 group had sinusoidal congestion, moderate steatosis, and frequent hepatocyte ballooning.
The same hepatic pathologist graded the histological specimens of all 18 livers according to the activity scoring system defined in Table 2. The corresponding scores are shown in Table 3. None of the livers of animals in the GCT10 and GCT16 groups exhibited portal or lobular inflammation, ballooning, or fibrosis. Only one animal in the GCT10 group had a liver exhibiting macro-steatosis, which was graded with score 1. In contrast, five animals in the GEP10 group and three in the GEP16 group had liver macro-steatosis. In addition, five out of six (83.3%) of the animals in the GEP10 group had significant inflammatory activity characterized by occasional or frequent hepatocellular ballooning, which includes the presence of swollen hepatocytes and rarefied cytoplasm, which may contain fat droplets and Mallory-Denk corpuscles, which usually represent a significant hepatocellular lesion [30]. Significant inflammatory activity occurred in 50% of animals in the GEP16 group. Lobular inflammation was present in one out of six (16.7%) of the animals in the GEP10 and GEP16 groups. Finally, no livers from animals in the GEP10 and GEP16 groups had portal inflammation or fibrosis.
Discussion
Hepatic fibrosis and cirrhosis represent the end stages of NAFLD, and the annual deaths from cirrhosis in the United States increased by 65% between 1999 and 2016 [31]. Consequently, diagnosing NAFLD/NASH in its early stages is important, as at these stages, there are still possible interventions that can dramatically modify the disease outcomes. Diagnostic and follow-up approaches that are noninvasive, effective, and precise and have low-cost impacts on the general public health are necessary. In this context, Webb et al. [27] used ultrasonography as a promising tool, determining the SHRI to quantify liver steatosis in a clinical trial performed between 2005 and 2006. Subsequently, B-mode ultrasonography was used by Lessa et al. [20] to evaluate hepatic cirrhosis, and more recently, Tang et al. [23] combined quantitative ultrasonography and elastography with machine learning analysis to assess steatohepatitis, liver steatosis, inflammation, and fibrosis. Finally, D’Souza et al. [28] reported an assessment of hepatic fibrosis based on the SHRI. The aforementioned investigations, based on rat models, all evaluated hepatic impairment in its final stages and chemically induced hepatic diseases using a 20% solution of carbon tetrachloride [20], a methionine- and choline-deficient diet [23], or oral administration of diethylnitrosamine [28]. The methods reported previously have the limitation of inducing hepatic diseases that may not reflect concurrent liver diseases of varying etiologies. In contrast, the present work was conducted with liver disease induced by a diet typical of Western societies, consisting of the Amylin liver NASH model [19] modified by adding 5% more lipid calories and maintaining the concentrations of fructose (20%) and cholesterol (2%).
In the present work, the SHRI of the animals in the GEP10 and GEP16 groups increased and were significantly different from the SHRI of the animals in the GCB group. The median SHRI for the GEP16 group was lower than the median SHRI for the GEP10 group, but there was no significant difference between them. It is worth noting (Fig. 2) that the mean normalized animal weight for the animal groups followed a trend similar to the one for the SHRI, with the mean normalized weight of the GEP16 group at T16 being lower than that of the GEP10 group at T10. In this context, a positive Pearson correlation coefficient of 0.54 (P<0.05) was determined, indicating a moderate relationship between the normally distributed SHRI and the normalized weights for animals in the GEP10 group at T10, and in the GEP16 group at T16.
The work conducted by D’Souza et al. [28] and the present study employed similar methods and the same ultrasound frequency (21 MHz) to determine the SHRI, the values of which were similar between the control and experimental animal groups. Furthermore, the method employed to generate the SHRI is noninvasive and immune to variations in instrumentation settings used to optimize animal examinations, which is a relevant advantage of the method. The novelty of the present work in relation to the work reported by D’Souza et al. [28] relates, in essence, to the stage of the analyzed hepatic disease and to the animal model. While the current work imaged liver disease in its early stages and relied on an animal model based on diet-induced NAFLD paralleling human dietary habits, the work conducted by D’Souza et al. [28] inspected hepatic disease at an advanced stage, including liver fibrosis, and relied upon an animal model based on oral administration of diethylnitrosamine.
There was a weight increase in the animals in the control groups, whereas those receiving the HFD did not gain significant weight. The overall difference between animal weights normalized by the corresponding animal weight at T0 for the control and experimental groups was significant after the fifth week. The literature presents conflicting results regarding weight changes in NAFLD animal models, as well as variation related to the length of the diet period, animals (strain, sex, and age), and the amount of calories allocated to fat [13]. Although some authors have reported that HFDs promote body weight increase, with physiological and phenotypical characteristics similar to human NAFLD [12,32,33], others did not report body weight differences between HFD and control groups [34-37]. Conversely, some studies reported weight decreases for animals in experimental groups [15,30,38] when employing diets with high-cholesterol concentrations, thus precluding the characteristic phenotype of obesity and metabolic syndrome. Moreover, the supplementation of a high-cholesterol and high-cholate diet with high-fat levels to animals can cause hepatic morphological alterations found in humans with NASH, as well as 9% body weight loss at the same time [18].
Although the animals in the present work receiving an HFD did not gain weight, their liver weight ratios did increase and normalized to the corresponding animal weight, did increase and the mean values were significantly higher (P<0.01) than that of the animals that did not receive an HFD (Fig. 6C). Additionally, the mean value of this ratio for the animals in the GEP10 group was higher (P<0.001) than in the animals in the GEP16 group; this should be explored in future works to determine whether the finding of less ballooning in animals from the GEP16 group is consistent with livers weighing less than those in animals from the GEP10 group. Regarding the liver weight in absolute values, only the animals in the GEP10 group presented a significant (P<0.01) difference in their mean liver weight in comparison with the corresponding results for the animals in the GCT and GEP16 groups (Fig. 6D). As was observed for animal weight, the liver weight (normalized or not) followed a similar trend to that for the SHRI, with a relative increase at T10 in comparison to corresponding evaluations at T0 and T16.
As presented in Table 3, the animals in the GEP10 group exhibited more pronounced effects regarding macro-steatosis and ballooning than the animals in the GEP16 group, as expressed by the quantity of animals with scores greater than zero and with median scores of 2.0 and 0.5 for the GEP10 and GEP16 groups, respectively. For macro-steatosis, the corresponding median scores were 1.0 and 0.5. These findings should be explored in future works to determine whether they can explain why the median SHRI of the GEP16 group was lower than that of the GEP10 group. From a histopathological point of view, the hallmark of NAFLD is hepatocellular steatosis, the presence of which is required in more than 5% of hepatocytes to diagnose this disease [30,39,40]. The histopathological findings in animals from the GEP10 and GEP16 groups are consistent with some of the NAFLD outcomes and include macro-steatosis, lobular inflammation and ballooning, with ballooning showing the highest scores. Despite the absence of the obesity and metabolic syndrome phenotype in the animals in this study fed an HFD, the histopathological findings enable the conclusion that the dietary model used herein effectively induced NAFLD and NASH in rats.
The findings in terms of animal and liver weights, as well as histopathology, provide important insights into the animal response to the HFD and the establishment of NAFLD and NASH. Furthermore, the current work concentrated on the ultrasonographic evaluation of NAFLD and NASH in their early stages in an animal model that relates more closely to translational fundamentals. The use of ultrasound imaging to image animal-induced hepatic disease, as introduced in the current work, establishes a basis for other investigations designed to diagnose NAFLD and NASH in animal models. The SHRI, based on a noninvasive method that is immune to variations in the instrumentation settings used to optimize animal examinations, represents a promising tool to characterize NAFLD in rat models.
Notes
Author Contributions
Conceptualization: Pantaleão Jr. ACS. Data acquisition: Pantaleão Jr. ACS, de Castro MP, da Silva ALA, Manso JEF, Machado JC. Data analysis or interpretation: Pantaleão Jr. ACS, Araujo KSFM, Campos CFF, Machado JC. Drafting of the manuscript: Pantaleão Jr. ACS, Manso JEF. Critical revision of the manuscript: Pantaleão Jr. ACS, de Castro MP, Araujo KSFM, Campos CFF, da Silva ALA, Machado JC. Approval of the final version of the manuscript: all authors.
No potential conflict of interest relevant to this article was reported.
Acknowledgements
The authors thank CENABIO (National Center of Structural Biology and Bioimaging), from Federal University of Rio de Janeiro, for the care and handling of the animals and for allowing the use of the ultrasound image equipment. The authors thank Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) for financial support.
References
Article information Continued
Notes
Key point
An animal model for nonalcoholic fatty liver disease (NAFLD) was developed based on a frequently used high fat Western diet consisting of 20% fructose and 2% cholesterol. NAFLD was diagnosed in the initial stages. The hepatorenal sonographic index was determined at three points along a timeline extending up to 16 weeks.