The Usefulness of MRI R2* Value In Diagnosing And Staging of Rat Liver Fibrosis

Title: The usefulness of MRI R2* value in diagnosing and staging of rat liver fibrosis


Background
Liver fibrosis is a consequence of chronic injury from a variety of causes and represents a common feature of almost all chronic liver diseases, which can be the development of cirrhosis, portal hypertension and hepatocellular carcinoma [1][2].
Progression of early fibrosis can be reversed by the effective intervention, such as specific antifibrotic therapy or elimination of the cause [3][4].Therefore, the early identification of liver fibrosis has crucial clinical implications in the determination of treatment options and prognosis.Liver biopsy is a good standard for detecting and staging liver fibrosis.However, it is an invasive procedure with inherent risks [5], and is subject to sampling error [6].Serum markers are noninvasive alternatives to biopsy for staging liver fibrosis, which include measurement of the doses of specific markers of fibrosis, such as N-terminal collagen III propeptide and hyaluronic acid.However, because fibrosis is not specific to the liver, the role of these markers is limited [2,3,5].
Iron-induced oxidant stress is involved in fibrogenesis, and causes hepatocytes necrosis and activates hepatic stellate cells (HSCs) and Kupffer cells [7,8], which ultimately cause liver fibrosis and other diseases.
Magnetic resonance imaging (MRI) is considered as a noninvasive and reliable method for detecting LIC [9][10][11].The local field inhomogeneity caused by the paramagnetic effect of tissue iron is the base of detecting iron with MRI [12].This causes more rapid signal decay resulting in increased the transverse relaxation rates of R2 and R2* (the reciprocal of T2 and T2* transverse relaxation times, respectively).
In particular, multi-echo gradient (mGRE) sequences are used to measure R2* of liver tissue [9][10][11].Liver biopsy-proven LIC and measured R2* have a good correlation [11,12].Because liver fibrosis involves the increase of iron deposition [7], we hypothesize that R2* measurement may be used as a noninvasive method to characterize processes of fibrogenesis with iron deposition.The aim of our study was to investigate the correlation between R2* measurements and the degree of liver fibrosis in a rat model.

Animal model
Our study was approved by our Animal Experimentation Ethics Committee (approval number: K2015122) and was carried out in accordance with the guidelines of north sichuan medical college and ARRIVE guidelines.Thirty-five male Sprague-Dawley rats (200 -250 g, Laboratory Animal Center of North Sichuan Medical College, Nanchong, China) were randomly divided into fibrosis and control groups.Thirty rats in the fibrosis group were injected intraperitoneally with a mixture of 1.0 ml/kg of carbon tetrachloride (CCl4) and oil (1:1 v/v) twice a week for 12 weeks.The other five healthy rats were used as the control animals.The animals were fed on a standard diet and subjected to a 12-hour light/dark cycle in an air-conditioned room at 25 °C.

Image processing
The software of R2Star (Function tool 4.4, GE Healthcare, Wisconsin, USA) was employed for calculating the R2* values on R2* map.For R2* values measurement, two radiologists with five years' experience in abdominal MRI were blinded to histopathologic results and the same two consecutive R2* sections at mid liver were analyzed.Three regions of interest (ROIs) of approximately 2-3mm 2 were placed on each slice avoiding major blood vessels and artifacts (Fig. 1a, b).R2* values for all six ROIs were recorded and the mean values for all six ROIs were calculated in each liver.

Histopathological examinations and detection of liver iron content
After the MR examinations, the animals were sacrificed by intraperitoneal injection of 3% pentobarbital sodium (3.0 ml/kg).The livers, which were harvested from the rats were divided into two parts for histopathological examinations and detection of liver iron content, respectively.A small amount of liver tissue (approximately 10×10×10mm 3 ) was used for masson trichrome staining.The residual liver tissue was used for detecting LIC by an atomic absorption spectrophotometer (Hitachi Z-5000, Tokyo, Japan) as previously described [13].

Statistical Analysis
All the data were expressed as mean ± standard deviation (SD).SPSS software (SPSS Inc., Chicago, IL, USA) was used for statistical analyses.A one-way ANOVA test was used to determine the significance of the differences in R2*values between different stages of fibrosis.Spearman correlation coefficients were used to evaluate the correlations between the R2* value and the degree of fibrosis and between the R2* value and LIC.The receiver operating characteristic (ROC) curve analysis was used to determine the optimal cutoff values of the R2* for fibrosis stage.A P＜0.05 was considered to be significant.

Fibrosis stages and R2* value
Twenty-three rats in the fibrosis group survived and seven died after 12 weeks.The mortality was 23.3%.Five rats in the control group were at stage F0; four rats in the treatment group were at stage F1; eight rats, at stage F2; seven rats, at stage F3 (Fig. 1c); and four rats, at stage F4.

R2* value and LIC
LIC gradually increased during the progression of the liver fibrosis (Table 1).

Discussion
Our initial findings indicate that R2* values are a noninvasive method of detecting and staging liver fibrosis.
Liver fibrosis is initiated by the accumulation of collagen and other materials within the extracellular matrix, including type I collagen which are formed by HSCs and hepatic myofibroblasts (MFB) [15,16].Assessment of liver fibrosis in chronic liver disease is beneficial for determining disease progression and assessing complications, such as esophageal varices and hepatocellular carcinoma.The imaging [15,17,18].The development of larger gradients, higher field strengths, improved surface coils, and parallel imaging techniques have considerably improved the speed and quality of MRI examinations [19].R2* has been proposed to be sensitive to tissue iron.R2* is used for noninvasive quantitative evaluation of LIC and plays an important role in the management of patients undergoing chelation therapy [12].Measured liver R2* does not require the intravenous injection of potentially nephrotoxic contrast agents or the use of an additional transducer hardware, such as in the case of MR elastography.Due to the paramagnetic nature of iron, R2* increases with iron deposits in the liver.Measured liver R2* has been shown to have a strong correlation with biopsy-proven LIC [11,12].Excess accumulation of Fe 2+ generates a radical OH, which leads to apoptosis of hepatocytes, activates HSCs, and promotes the process of fibrosis [16].The R2* is also affected by some confounding factors, including collagen contents, fat and necroinflammation.Results from this study suggest that liver fibrosis is associated with a R2* value increase, and this R2* value increase is correlated with LIC.There were significant differences in R2* values among the stages of liver fibrosis (F=30.84，P<0.001).With the use of ROC analysis, the most discriminating cutoff values of R2* were 46.84 Hz for ≥ F1, 55.30 Hz for ≥ F2, 68.06 Hz for ≥ F3, and 78.79 Hz for F4.In our study, the R2* measurements can clearly separate the different fibrosis stages.For example, the cutoff value of 46.84 Hz for stage 1 or greater fibrosis had a sensitivity was 91.3 % and a specificity of 100%.This high accuracy for the diagnosis of intermediate fibrosis stages is clinically important, because patients with hepatitis C genotype 1 infection should be treated only when substantial fibrosis is observed. 19Furthermore, with an optimized cutoff value of 68.06 Hz for stage 3 or greater fibrosis, the sensitivity of R2* value was 91.2% at a specificity of 94.1%.This high accuracy for the diagnosis of advanced fibrosis is also important because portal hypertension and hepatocellular carcinoma should be alert in patients with advanced fibrosis [20].We believe a clear association between increasing R2* values and fibrosis stage has been shown (r=0.902,P<0.001), R2* values can potentially be used noninvasively for detecting and staging liver fibrosis.
The variability of biopsy LIC depends on the size of the specimen.Biopsy LIC measurement also varies widely among laboratories because of differences in the actual ratios of wet-to-dry weights of tissue samples [11].In our study, the vast majority of liver tissue was used for detecting of LIC by atomic absorption spectrophotometer.Thus, this method can better reflect the LIC compared with biopsy.
There was one limitation to our study.Because this is a report of our initial experience, the results were limited by the sample size; more specifically the small number of rats with stage F1 and stage F4 of liver fibrosis.

Conclusions
In conclusion, this study shows that measured liver R2* is a noninvasive and development of noninvasive markers of liver fibrosis would reduce biopsy-related complications and facilitate early diagnosis and improve monitoring of progression of chronic liver disease.Medical imaging is of great significance for the diagnosis and staging of liver fibrosis.Many imaging based techniques are used including ultrasonography-based elastography, such as 2D-Shear wave elastography, transient elastography, and diverse MRI-based techniques, such as MR elastography, diffusion-weighted imaging (DWI), T1ρ MR imaging, MR perfusion imaging and dynamic contrast-enhanced MR

Fig 4 .
Fig 4. Scatterplots of R2* values and LIC.The spearman correlation coefficient between R2* values and LIC was 0.984.