Correlation between viral load, histopathologic necroinflammatory grade and fibrotic stage in viral hepatitis


 Introduction: Liver needle biopsy and histopathology of liver parenchyma are one of the most accurate methods for assessing the severity of inflammation and fibrosis in these patients, which is based on Knodell’s standard system. Objectives: The aim of this study was to investigate the relationship between viral load of hepatitis B and C in polymerase chain reaction (PCR) with pathologic necroinflammatory grade and fibrotic stage of hepatitis in liver biopsy. Patients and Methods: Pathologic evaluation performed to determine the grade and stage of viral hepatitis by using 5μ sections stained with hematoxylin-eosin, Masson trichrome and reticulin in unique center. Data were analyzed using SPSS 16 software. Results: There was a significant relationship between viral load and grade of hepatitis B and C. No association was found between viral load of hepatitis B and C and fibrotic stage of hepatitis. Additionally, relationship between spotty necrosis and portal inflammation with viral load of hepatitis B and relationship between periportal interface hepatitis and viral load of hepatitis C were statistically significant. Conclusion: Despite the association between viral load in PCR and necroinflammatory grade in the histopathology of liver needle biopsy, the relationship between viral load and fibrotic stage is rejected, which can be affected by the duration of the disease.


Introduction
Chronic hepatitis is referred to a set of liver complications with different causes and severities in which inflammation and liver necrosis last at least six months. The chronic hepatitis is classified based on the causing agent into viral hepatitis-especially the types B and C, autoimmune hepatitis, drug-related hepatitis, and uncertain hepatitis (1).
Hepatitis B virus (HBV) is one of the most common causes of chronic hepatitis. It is estimated that one third of the world's population has serologic reasons in favor of a previous involvement with HBV infection. More than 400 million people worldwide have the disease, which 15%-40% of them finally develop long-term complications including liver failure, cirrhosis, and even hepatocellular carcinoma (HCC), while one million people die of the disease every year. HBV belongs to the Hepadnaviridae

Key point
There are several methods for assessing the extent of damage to liver tissue due to viral hepatitis. In this study, several methods of histological examination of the liver following viral hepatitis have been compared with each other. The sensitivity and specificity of each method were discussed and compared.
family. It is a partially double-stranded DNA virus, which its genome is surrounded by a nucleocapsid or core antigen (HBcAg), which itself is also surrounded by a spherical shape, lipid envelope containing hepatitis B surface antigen (HBsAg). The HBV genome also produces a peptide called e-antigen (HBeAg) released into the bloodstream that is an indicator of active viral replication. To determine the patient's status, the serum level of several antigens and antibodies, including HBsAg, HBc Ab, HBeAg, and HBe Ab, as well as liver enzymes (AST, ALT) and, if necessary, HBV-PCR are required to measure serum viral load.
About 130-170 million people are infected with hepatitis C virus (HCV) worldwide, which develops into chronic disease in 85% of cases and can progress to cirrhosis and sometimes HCC. The prevalence of cirrhosis in HCV infection is over 50%, and 10%-30% of people who have HCV infection more than 30 years, may develop cirrhosis. The HCV is a small, enveloped, single-stranded RNA virus of the Flaviviridae family. HCV has seven major genotypes and genotype 1 is the most common cause of the disease in the United States and Europe. The basis for diagnosis is the detection of HCV Ab in serum, followed by HCV-PCR to measure serum viral load; nevertheless, histopathology of liver is probably the best prognostic indicator of chronic HCV infection (7)(8)(9).
In various studies, such as those of Guido et al in Italy and Theise et al in the United States, the importance of liver biopsy has been evaluated to determine the prognosis of viral hepatitis and the relationship between the severity of the disease and the findings of the biopsy. Using a scoring system causes an improvement in the accuracy of biopsy to assess the severity of inflammation and necrosis. Scoring system can be applied to detect the degree of fibrosis and to reconcile the reports and make the results repeatable too (10). Several scoring systems are designed based on the histopathological examination of tissue samples obtained by core needle biopsy of the liver to evaluate necroinflammatory activity and the progression toward cirrhosis that the most popular of which is the Knodell system with a histological activity index (HAI), which classifies chronic hepatitis into mild, moderate, and severe degrees.
In this histological scoring, grade is an indicator of necroinflammatory activity, which includes the sum of the scores of the four following main items ranging from 0 to 18; 1. The degree of periportal inflammatory activity depends on the severity and the extent of necrosis around the portal areas and destruction of limiting plate of hepatocytes adjacent to the portal tract by inflammatory cells called piecemeal necrosis or periseptal interface hepatitis scored 0 to 4. 2. The extent of confluent necrosis in the centrilobular areas, which sometimes causes connection between the portal tract and the central vein called portalcentral bridging necrosis to pan-lobular necrosis scored 0 to 6. 3. The degree of focal degradation and necrosis of hepatocytes in the liver lobule based on the number of foci in the low power field of microscope called spotty necrosis scored 0 to 4. 4. The inflammation degree of the portal is based on the severity of inflammation and the extent of involved portal tracts scored 0 to 4. In this assessment, stage represents the classification of the disease progression toward cirrhosis based on the severity and the extent of fibrosis in the liver parenchyma by forming various forms of fibrosis as follows; fibrous expansion of portal areas with or without short fibrous septa, portal-portal bridging fibrosis, portal-central bridging fibrosis, and cirrhotic nodule scored 0 to 6. If the intensity of fibrosis is increased to such an extent that can change the normal structure of the liver lobule, it is referred to as cirrhosis (1)(2)(3)(4)(5)(6).
Several studies were published regarding the relationship between viral load and histopathological findings in recent years with various results. In a study by Shafaei et al, 72 patients with chronic HBV infection were evaluated from 2006 to 2011 in Babol University, Iran. The patients were divided into three groups of <10 3 , 10 3 -10 5 , and >10 5 copy/ mL based on viral load, and the groups were compared in terms of total score of grading and score of piecemeal necrosis. The results indicated a significant relationship between these two obtained scores and viral load; however there was no significant relationship between viral load and the stage of disease or the degree of fibrosis. In addition, the presence of confluent necrosis was indicative of severe disease (8). In a study by Rodriguez-Inigo et al, in the United States, 27 patients with chronic HCV infection were examined for viral load in peripheral blood by polymerase chain reaction (PCR); the percentage of involved cells was also investigated in a liver biopsy by in situ hybridization. Then their association with pathologic grading was evaluated based on biopsy findings. In their study, a direct correlation was seen between viral load in the blood and the rate of infected hepatocytes, while no relationship was observed between viral load and rate of infected cells with the degree of pathologic grade (11). In the study by Pie et al, in China, the biochemical findings (alanine aminotransferase, ALT), virology data (viral load), and histopathology score (grading and staging) were investigated in 132 patients with HCV infection. In their study, despite the relationship between ALT and necroinflammatory grade, the association between ALT and fibrotic stage was absent. It was also suggested that patients with advanced hepatitis can have normal or near normal ALT level. In addition, no relationship was observed between the viral load and histopathological grade and stage of the disease (12). In the study by Xu et al, in China, 233 patients with chronic HBV infection were evaluated; the study showed, viral load has no correlation with the grade or stage of the disease in a pathologic exam. They also showed that the ALT level can indicate the grade of the disease; however it has no correlation with the stage of the disease (13).

Objectives
According to the presence of standard grading system that was agreed by pathologists to increase repeatability of the results of grading and staging in viral hepatitis and the availability of a quantitative PCR to determine the viral load before onset of treatment, the current study aimed at determining the relationship between these two findings as well as the relationship between viral load and the details of histopathologic grading and staging in liver biopsies.

Study design
The current descriptive-analytical study was conducted at Sina hospital in Hamadan, Northwest of Iran in 2017. A total of 122 new patients who had positive serology including HBS-Ag or HCV-Ab by electrochemiluminescence (ECL) method and were willing to conduct liver biopsy and give blood samples, were enrolled in the study after obtaining written consent. A 5-mL blood sample was taken from each subject to test the biochemical parameters of AST and ALT, and another 5-mL blood sample was also obtained for quantitative PCR in the same laboratory. The AST and ALT tests were performed by Mindray-480 autoanalyzer using Pars-Azmoon kit (Iran). Quantitative-PCR was performed to detect HBV and HCV by Cobas TaqMan-48 analyzer using Roche viral N/A purification real time kit (USA) with 25 IU/ml sensitivity, 95% specificity and detection range of 25-11×10 7 IU/mL.
The liver core needle biopsy was transferred to the laboratory in 10% buffered formalin. After performing fixation and 24-hours tissue processing, the paraffin block was prepared and accordingly, three 5-μm sections were cut for hematoxylin and eosin, Masson trichrome and reticulin staining. Histopathological assessments were performed by an expert pathologist who was blind to the study and revealed the grade and stage of chronic hepatitis based on the Knodell histology activity index (HAI) ( Tables 1 and 2). In addition, owing to the wide range of histopathological grading scale in the Knodell system, from 0 to 18, another variable was defined as modified grade with a smaller range as follows; total grade 1 to 4 as the 1 st rank, total grade 5 to 8 as the 2 nd rank, total grade 9 to 12 as the 3 rd rank, and total grade 13 to 18 as the 4 th rank.
The census method was employed in the current study to set the sample size and all patients with HBV or HCV infection referring to Sina hospital willing to participate in the study were selected within a five-year interval from 2012 to 2017 after obtaining written consent.

Data analysis
The data were then transferred into SPSS software (version 16) for statistical analysis. To express the quantitative data, the mean and standard deviation (±SD) were used, Tukey post hoc test was also used. To determine the correlation between viral load and other quantitative variables, Spearman's correlation coefficient was used.

Results
Of 122 patients, 71 had HCV and 51 HBV infections; 85.2% of the subjects were male. The mean age of patients in the HCV and HBV groups was 40.59 ± 12.54 and 35.16 ± 11.15 years, respectively. The mean viral load in the HBV and HCV groups was 13 076 151.66 ± 57 587 802.53 and 27 793 360.14 ± 87 925 501.63 IU/mL, respectively.
Based on the results of one-way ANOVA, the relationship between viral load and different necroinflammatory grades was evaluated in the HBV and HCV groups as well as the total population and the results indicated a significant relationship between, the severity of grades -based on histopathological findings -and viral load -based on PCR results -in the HBV group (P = 0.004). In addition, using the Tukey post hoc test, a significant difference was observed in the HBV group in terms of viral load between grade 4 and the rest of grades, except grades 8 and 9 (P = 0.007). However, the same relationship was not statistically significant in the HCV group (Table 3).
Based on the results of the one-way ANOVA, no significant relationship between viral load and different fibrotic stages in the HBV and HCV groups as well as the total population was seen ( Table 4).
The relationship between viral load and different scores of spotty necrosis in the HBV and HCV groups was evaluated using one-way ANOVA and the results indicated a statistically significant correlation between the severity of spotty necrosis and viral load in the HBV group (P = 0.013). In addition, a significant difference was observed in viral load between spotty necrosis grades 0 and 1 in the HBV group using Tukey post hoc test (P = 0.009). In the HCV group, the same relationship was not statistically significant (Table 5).
In the HCV group, the Spearman's Rho test showed a statistically significant correlation between viral load and the severity of periportal interface hepatitis, in one direction (P = 0.028). Additionally, the relationship between viral load and the modified grade was consistently significant (P = 0.020). In addition, the relationship between grade and viral load was near the significant level.
In the HBV group, there was a consistently significant correlation between viral load with the severity of portal inflammation and serum ALT levels, according to the Spearman's Rho test (P = 0.004 and 0.034 respectively).
In the total study population, a statistically significant correlation was observed between the severity of portal inflammation-based on histopathological findings, and viral load-based on PCR results, in the same direction (P = 0.002). Furthermore, this relationship was nearly significant in terms of periportal interface hepatitis. However, the same correlation was statistically insignificant in terms of the other variables (Table 6).

Discussion
The current study aimed at determining the association between serum viral load of HBV and HCV infections and the grade and stage of hepatitis in histopathologic evaluation of the liver biopsy that was performed on 122 patients.
The results of our study showed a significant relationship between viral load and grade of hepatitis in the HBV group. There was also a statistically significant correlation between viral load and modified grade in the HCV group; in other words, by increasing the viral load, the necroinflammatory grade also increased.
In the study by Shafaei et al, (8) the relationship between three different levels of viral load and total necroinflammatory grade was statistically significant in patients with HBV infection, which was consistent with the results of the current study; however in the studies by Xu et al (13) and Martinot-Peignoux et al (14) on patients with HBV infection, the association between viral load and grade of the disease was statistically insignificant,

Modified Histological Activity Index-staging (Architectural Changes, Fibrosis, and Cirrhosis Score) Score
No fibrosis 0 Fibrous expansion of some portal areas, with or without short fibrous septa 1 Fibrous expansion of most portal areas, with or without short fibrous septa 2 Fibrous expansion of most portal areas with occasional portal to portal (P-P) bridging 3 Fibrous expansion of portal areas with marked bridging (P-P) as well as portal-central (P-C) 4 Marked bridging (P-P and/or P-C) with occasional nodules (incomplete cirrhosis) 5 Cirrhosis, probable or definite 6 Total Maximum possible score 6 which may be due to the high percentage of mild grade and lower viral load in their populations. In addition, in the study by Rodriguez-Inigo et al (11) and Pie et al (12), the same relationship was not statistically significant in patients with HCV infection. Moreover, in our study the association between viral load and total grade was statistically insignificant in the HCV group.
In the current study, no relationship between the viral load and the fibrotic stage of disease both in HBV and HCV groups was seen. It should be noted that in the studies by Shafaei et al (8), Xu et al (13), Martinot-Peignoux et al (14), and Chi-Jen Chu et al (15) on patients with HBV as well as the study by Pie et al (12) on patients with HCV infection, the same relationship was statistically insignificant, which was in agreement with the results of the current study. Perhaps, the stage of liver fibrosis is more affected by the duration of the disease than the level of viral load; however further comprehensive studies are required to obtain more conclusive results.
In the current study, in relationship between the components of grade -based on histopathological findings -and viral load -based on PCR results -the association of spotty necrosis and portal inflammation with HBV viral load, and the relationship between periportal interface hepatitis and HCV viral load were statistically significant. In addition, a significant association between viral load and the degree of portal inflammation in all patients with viral hepatitis was seen, which indicates the increase in portal inflammation following the increase in viral load. However, the association with confluent necrosis was insignificant, which can be attributed to the very small number of subjects with confluent necrosis in the current study. In the study by Shafaei et al (8), a significant relationship was observed between HBV viral load and periportal interface hepatitis, but this relationship was insignificant in terms of other components of grade.
The results of his study, despite the similarity in the relationship between total grade and HBV viral load, were inconsistent with the current study findings regarding the grade components in the HBV group. In addition, in the study by Shafaei et al, the presence of confluent necrosis indicated a more severe disease, which was inconsistent with the results of the current study due to the low number of patients with confluent necrosis. Moreover, in the study by Rodriguez et al (11), a direct relationship was observed between the viral load of HCV and the percentage of virus-infected hepatocytes, however no association was observed with the grade in histopathology. In the recent studies, the relationship between grade components and viral load was less considered, especially in HCV infection, since further extensive studies are required in this regard.
In the current study, the relationship between viral load and ALT in the HBV group was significant, but the same relationship was not observed in the HCV group. In the studies by Martinot-Peignoux et al (14) and Chi-Jen Chu et al (15), this relationship was not significant, which was inconsistent with our findings. This difference can be attributed to the higher number of patients with HBV infection in the two former studies compared with that of the current study. In addition, in the study by Martinot-Peignoux et al, most of the subjects had a viral load of <10 5 and a mild histopathological grade that might affect the statistical findings. Nevertheless, in the study by Pie et al (12) on patients with HCV as well as the study by Xu et al (13) on subjects with HBV infection, the relationship between ALT and grade of viral hepatitis was statistically significant. In other words, ALT increases with more activation of liver inflammation, but its association with fibrotic stage was rejected in both studies and their results showed that in patients with advanced hepatitis and a high stage, ALT can even be normal or close to normal.

Conclusion
Despite the association between the viral load -based on PCR results -and necroinflammatory grade in the histopathologic evaluation of liver core needle biopsy, shown in various studies, the association between viral load and fibrotic stage was rejected in all studies, which can indicate more influence of other factors such as the duration of the disease, compared with viral load, in the development of liver fibrosis, which is practically unattainable since the onset of the disease is unclear in most cases. The controversy in the relationship between necroinflammatory grade components and the viral load that was observed in the various studies, indicate the existence of other influential factors such as immunesystem parameters of the infected individuals in the degree of necroinflammatory changes of liver, which apart from viral load, can influence the severity of necrosis and inflammation of the liver parenchyma and contribute as a confounding factor to the relationship between viral load and the grade components.
Overall, due to the significant relationship between viral load and histopathologic grade of hepatitis, the quantitative-PCR can be used to estimate the necroinflammatory grade; however to estimate the fibrotic stage, viral load is not reliable and other non-invasive methods such as liver FibroScan can be employed. Besides, for estimation of the necroinflammatory grade based on viral load, larger studies with adequate sample size for each necroinflammatory grade are needed to calculate cut-off value with ROC curve analysis.

Limitations of the study
Our study requires further investigation by larger samples in multi-centric studies.