Diagnostic Accuracy Of Apparent Diffusion Coecient Values Combined With Γ-Glutamyl Transpeptidase-To-Platelet Ratio Parameters For Predicting Hepatitis B-Related Fibrosis

Background: The accuracy of non-invasive liver brosis diagnosis based on the apparent diffusion coecient value combined with the γ-glutamyl transpeptidase-to-platelet ratio model to predict the degree of hepatitis B-related brosis has not been reported. This study aimed to evaluated the diagnostic ecacy of the apparent diffusion coecient value combined with the γ-glutamyl transpeptidase-to-platelet ratio for liver brosis grading. Methods: The data of 180 patients with chronic hepatitis B diagnosed by liver biopsy were analyzed.The apparent diffusion coecient value, γ-glutamyl transpeptidase-to-platelet ratio, as well as their combination were assessed in different cirrhosis stages using receiver operating characteristic curve analysis to evaluate their value in diagnosing hepatic brosis in chronic hepatitis B patients. Results: We observed that liver brosis stages were inversely associated with apparent diffusion coecient values (r=−0.691, P<0.001), and positively associated with the γ-glutamyl transpeptidase-to-platelet ratio parameters (r=0.502, P<0.001). The area under the curve for diagnostic ecacy of apparent diffusion coecient values, the γ-glutamyl transpeptidase-to-platelet ratio parameters, and their combination for F ≥ 2 liver brosis was 0.831, 0.749, and 0.858, respectively, and for F ≥ 3 was 0.872, 0.771, and 0.903, respectively. Conclusions: Using apparent diffusion coecient values combined with γ-glutamyl transpeptidase-to-platelet ratio parameters was better than using apparent diffusion coecient values or γ-glutamyl transpeptidase-to-platelet ratio parameters alone (all P<0.01). The combined diagnostic tool may improve the accuracy of chronic hepatitis B-related liver brosis diagnosis, especially for F ≥ 3. to evaluate the degree of liver brosis has not been reported. Here, a pathological biopsy of the liver was used as the standard to explore the diagnostic ecacy of the ADC value combined with the GPR parametric model for CHB liver brosis, aiming to provide valuable diagnostic indicators for the diagnosis, treatment, and prognosis of liver brosis. H test and the Mann–Whitney U test. The Spearman correlation coecient was used for correlation analysis. Taking liver histopathological results as the standard, we used the receiver operating characteristic (ROC) curve to evaluate the accuracy of GPR parameters, ADC values, and their combination for diagnosing liver brosis in patients with CHB. changes in γ-glutamyl transferase and platelet levels [32]. Hence, although the ADC values and GPR parameters have good clinical values for diagnosing signicant and progressive hepatic brosis, they may miss the opportunity for early antiviral treatment due to their low sensitivity. Therefore, they still cannot completely replace liver puncture biopsy. Nevertheless, our results demonstrate that the diagnostic ecacy of combining ADC values and GPR parameters for diagnosing hepatic brosis was superior to that of a single ADC value or GPR parameter. This compensated for the ADC values’ lack of ability to diagnose light and moderate liver brosis, although the combined tool having a low sensitivity is acceptable. Its high sensitivity should not miss diagnosis, when ADC values were used for independent diagnosis. Moreover, ADC values combined with GPR parameters were the most advantageous in diagnosing liver brosis with F ≥ 3 (AUC is 0.903), which suggested that when diagnosing mild, moderate or severe liver brosis, joint diagnosis can be made. This can complement each other's advantages and greatly improve the diagnostic capabilities of both.


Background
China is one of the countries with a high prevalence of hepatitis B. Without effective treatment, chronic hepatitis B (CHB) can further develop into liver brosis and cirrhosis, and the latter can also lead to primary hepatocellular carcinoma. Liver brosis can be cured with effective etiological treatment [1]. Therefore, its early detection and precise staging are crucial [2,3]. Needle biopsy of the liver is traditionally regarded as the "gold-standard" for diagnosing hepatic brosis and cirrhosis. However, as an invasive operation, liver biopsy has numerous drawbacks (e.g. high risk, small sampling, poor reproducibility, bleeding, infection, sampling error, inability to repeat the examination in a short time, and poor patient compliance, etc.) [4][5][6][7]. Therefore, liver biopsy is di cult to implement widely, and using it as a general screening diagnostic method is challenging.
In 2015, Lemoine and his colleges constructed a new mathematical model, γ-glutamyl transpeptidase-to-platelet ratio (GPR), to predict CHB related brosis [8]. The parametric model is a combination of different biochemical serological indicators to obtain laboratory indicators for evaluating liver brosis. Some studies have concluded that the feasibility of GPR prediction was better than the aspartate aminotransferase-to-platelet ratio index (APRI) or brosis-4 index [8 -10]. However, the accuracy of the GPR prediction needs improvement, and its performance requires further veri cation.
Magnetic resonance diffusion weighted imaging (DWI) re ects the microscopic water molecule dispersion phenomenon. In liver brosis, the apparent diffusion coe cient (ADC) value is lower than normal liver parenchyma due to collagen deposition, increased ber, low water content, and limited diffusion of water molecules. Because DWI technology does not require injection of a contrast agent, the imaging speed is faster, and the diseased tissue can be quantitatively evaluated. DWI has exhibited great potential in diagnosing liver brosis as the degree of liver brosis can be quantitatively re ected by measuring the change in the ADC value.
At present, most joint researches on non-invasive diagnosis of hepatic brosis mainly focus on simple CHB or nonalcoholic fatty liver disease. However, the diagnostic accuracy of the ADC value in magnetic resonance imaging (MRI) combined with the serological GPR parametric model to evaluate the degree of liver brosis has not been reported. Here, a pathological biopsy of the liver was used as the standard to explore the diagnostic e cacy of the ADC value combined with the GPR parametric model for CHB liver brosis, aiming to provide valuable diagnostic indicators for the diagnosis, treatment, and prognosis of liver brosis.

Study patients
From March 2016 to June 2021, 180 patients with chronic hepatitis B were hospitalized and underwent liver biopsy at the Department of Liver Diseases of Huai'an Infectious Disease Hospital. The patients who met the diagnostic criteria for the management of hepatitis B virus (HBV) infection in the 2017 Clinical Practice Guidelines of the European Association for the Study of the Liver were included in the study. The exclusion criteria were: hepatitis C and D virus infections; hepatocellular carcinoma; transaminase level>400 U/L; heavy drinking (>20 g/day); acute heart failure or pregnancy; those with contraindications to MRI examination; and those with more diffusion-weighted imaging artifacts and of poor quality.

Serological examination
The BC-5380 automatic blood cell analyzer was used for the routine blood test. The Au5811 automatic biochemical detector was used for the liver function test. The ELISA method (the American Abbott i2000SR automatic chemiluminescence analyzer, the reagents are from the instrument manufacturer's supporting reagents) was used for HBV marker detection. The HBV DNA detection adopted realtime uorescent quantitative polymerase chain reaction method (reagents were from Shanghai Fosun Long March Medical Science Co., Ltd., Shanghai, China), and 500 IU/L was considered the lower detection limit of HBV DNA. GPR was calculated using the following formula: GPR = actual value of γ-glutamyl transpeptidase (γ-GT)/upper limit of normal value of γ-GT/platelet count ×100.
The upper limit of the normal γ-GT value was 50 U/L.

Magnetic resonance examination
All patients underwent an upper abdominal magnetic resonance scan and a DWI examination with fasting for over 6 h and breathing training before the examination. The GE MR355 1.5 T superconducting magnetic resonance imaging system was used for the examination. The patients took the advanced foot, supine position, and eight-channel phased array body coil. We rst performed a routine MRI scan with the range from the top of the diaphragm to the lower pole of the liver or spleen.
The diffusion imaging adopted a single-shot plane echo imaging sequence, and the scanning parameters were: TE 61.8 ms, TR 9230.8 ms, slice thickness 6.0 mm, interval 1.5 mm, matrix 256×256, number of signal averaged 2 times, and b value 800 s/mm 2 [11,12]. We rst imported the collected raw DWI data into the commercial analysis software package (Functool) in GE AW4.6 workstation, selected some layers with the best image quality in the ADC map, and randomly drew 3 region of interests (ROIs) on different layers. The size of the ROI was approximately 100 mm 2 . The choice of ROI should avoid the hilar and the large blood vessels in the second hilar. The measurement should be performed by two senior imaging physicians simultaneously, and the results should agree. Finally, the measured ADC value was recorded as the average value.

Histological examination of liver puncture biopsy
The location and depth of the puncture were determined by color Doppler ultrasound (the right lobe of the liver and away from the large blood vessels). After routine skin disinfection, application of a local anesthesia, and instructing the patients to hold their breath, we used a 16G liver puncture needle (Braun, Germany) to extract the liver tissue (sample length 15-25 mm) under negative pressure.
Then, the liver tissue was xed in 4% formaldehyde, embedded in para n, serially sectioned, stained with hematoxylin and eosin and Masson, and nally observed under a multi-eld optical microscope.

Statistical analysis
Measurement data conforming to normal distribution were represented by mean±standard deviation; non-normal distribution were represented by median and interquartile range [M(P 25-P 75)]. The linear-by-linear test was used to test the association among variables in a contingency table with ordered categories. Comparisons between multiple groups and further pairwise comparisons were performed using the Kruskal-Wallis H test and the Mann-Whitney U test. The Spearman correlation coe cient was used for correlation analysis. Taking liver histopathological results as the standard, we used the receiver operating characteristic (ROC) curve to evaluate the accuracy of GPR parameters, ADC values, and their combination for diagnosing liver brosis in patients with CHB.
Statistical tests were all two-sided, and P<0.05 was considered signi cant. All analyses were performed using IBM SPSS 25.0 (SPSS, Chicago, Illinois) and MedCalc 12.0 (Mariakerke, Belgium).

Comparison of ADC values and GPR parameters
ADC values gradually decreased with the aggravation of liver brosis (P trend <0.001). In contrast, GPR parameters gradually increased with the aggravation of liver brosis (P trend <0.001) (Figure 1, Table 2).   Magnetic resonance DWI can be used to assess the physiological and pathological characteristics of the body tissue through the microscopic situation of water molecules [14]. It is an imaging technique particularly sensitive to the transverse dispersion motion of water molecules, and the size of the ADC value re ects the rapid and slow motion of tissue water molecules [15,16]. Normal hepatocyte morphology and arrangement are orderly, and the hepatic parenchymal cells and hepatocyte matrix are stable. When hepatic brosis occurs in patients with chronic hepatic disease, hepatocyte degeneration, necrosis, abnormal hyperplasia of intrahepatic ber tissue, and collagen ber deposits in the stroma of liver cells occur, and water molecular activity is limited, resulting in the reduction of water molecular activity of the liver tissue, which decreases the ADC value of the pathological tissue.
However, some studies have reported that the accuracy of the MRI ADC value was relatively low in the diagnosis of segmentation point F≥2, which suggested that its ability to distinguish between mild and moderate brosis was relatively weak [17][18][19]. Thus, the cause may be related to the extremely low deposition of extracellular matrix and no hepatic brosis for grade F1, and con ned peripheral sinus, con uence area and lea et brosis, sparse distribution of collagen ber, no ber interval and no lea et structure disorder, and the diffusion restriction of water molecules for grade F2 [20][21][22].
In contrast, the ability of GPR to diagnose hepatic brosis at different stages was relatively stable. Previous studies have reported on the noninvasive diagnosis of hepatic brosis by serology [23]. The GPR model established by Lemoine and his colleagues was better than the APRI and brosis-4 models in the evaluation of hepatic brosis in patients with CHB in West Africa [8]. More recent studies have also demonstrated the advantages of GPR model [24][25][26][27][28][29][30], which were also con rmed in our previous study [31]. However, GPR only re ects matrix changes in the serum, rather than liver parenchymal degradation, lacks speci city for the liver, and is easily affected by changes in γ-glutamyl transferase and platelet levels [32]. Hence, although the ADC values and GPR parameters have good clinical values for diagnosing signi cant and progressive hepatic brosis, they may miss the opportunity for early antiviral treatment due to their low sensitivity. Therefore, they still cannot completely replace liver puncture biopsy.
Nevertheless, our results demonstrate that the diagnostic e cacy of combining ADC values and GPR parameters for diagnosing hepatic brosis was superior to that of a single ADC value or GPR parameter. This compensated for the ADC values' lack of ability to diagnose light and moderate liver brosis, although the combined tool having a low sensitivity is acceptable. Its high sensitivity should not miss diagnosis, when ADC values were used for independent diagnosis. Moreover, ADC values combined with GPR parameters were the most advantageous in diagnosing liver brosis with F≥3 (AUC is 0.903), which suggested that when diagnosing mild, moderate or severe liver brosis, joint diagnosis can be made. This can complement each other's advantages and greatly improve the diagnostic capabilities of both.
No non-invasive testing method or a single laboratory index that completely replaces the examination results of liver biopsy exists, so combining diagnostic models of non-invasive testing and laboratory indicators may overcome the shortcomings of using a separate method or indicator. In assessing the degree of hepatic brosis, the ADC values were more effective than the GPR parameters. The GPR parameters can be used as a prognostic evaluation of hepatic brosis staging. Combining the ADC values and GPR parameters can improve the diagnostic ability for hepatic brosis, and provide a more reliable diagnostic index for its early diagnosis, treatment, and prognosis. A non-invasive detection joint or multiple index model should be explored in future research.
This study has some limitations. First, the patients were enrolled from a single center, and the sample size was small, which make it di cult to accurately evaluate the overall situation of patients with HBV in China. Second, the study was designed to exclude concomitant diseases that could have effects on γ-GT or platelets, such as non-alcoholic fatty hepatic disease, alcoholic hepatic disease, combined with other hepatitis virus infections. Third, MRI examination is more expensive, require longer inspection time, and cannot be available due to contraindications. Finally, fat deposition can also affect the ADC value, so further research needs to be done in the future.
In summary, we found that combining ADC and GPR was better than using the ADC value or GPR parameter alone in diagnosing liver brosis. The ADC value combined with GPR parameter may improve the accuracy of chronic hepatitis B-related liver brosis, especially for F≥3. We believe that our study makes a signi cant contribution to the literature because the accuracy of non-invasive liver brosis diagnosis based on the ADC value combined with the GPR model to predict the degree of hepatitis B-related brosis has not yet been reported.

Declarations
Ethics approval and consent to participate The study has followed the tenets of the Declaration of Helsinki. The Institutional Review Board of Huai'an Infectious Disease Hospital approved this retrospective study (batch number: 2020011) and waived the requirement for written informed consent due to its retrospective nature.

Consent for publication
Not applicable.

Availability of data and materials
The datasets generated and/or analysed during the current study are not publicly available due to participant con dentiality but are available from the corresponding author on reasonable request. Figure 1 Correlation between ADC values, GPR parameters, and liver brosis.