Research Article

Diagnostic Performance of Non-invasive Fibrosis Indexes in Hepatitis B Related Fibrosis

10.4274/vhd.galenos.2020.2019.0038

  • Süleyman Sayar
  • Roni Atalay
  • Şüheda Çakmak
  • Gülçiçek Ayrancı
  • Kemal Kürbüz
  • Resul Kahraman
  • Zuhal Çalışkan
  • Oğuzhan Öztürk
  • Hakan Demirdağ
  • Gupse Adalı
  • Kamil Özdil
  • Hamdi Levent Doğanay

Received Date: 24.10.2019 Accepted Date: 02.05.2020 Viral Hepat J 2020;26(2):78-84

Objectives:

The aim of this study was to evaluate the diagnostic performance of non-invasive fibrosis markers [AST to platelet ratio (APRI), Fibrosis Index based on four factors (FIB-4) İndex, AST/platelet/GGT/Alpha-fetoprotein (AFP) Index (APGA), FI, Fibro-quotient (FibroQ), AST/ALT ratio (AAR), GGT/Platelet ratio (GPR), Platelet-age-phosphatase- AFP-AST (PAPAS) and S-Index] in chronic hepatitis b (CHB) patients.

Materials and Methods:

Treatment naive CHB patients who underwent liver biopsy were screened. Four hundred seventeen patients were included in the study. Fibrosis stage was reevaluated according to ISHAK score. The diagnostic efficacy of non-invasive fibrosis indicators for significant fibrosis (≥F3) and cirrhosis (≥F5) was evaluated. The diagnostic performance of the non-invasive markers was defined as the AUROC value of ≥0.9 as excellent, 0.9> AUROC ≥0.8 as good, 0.8> AUROC ≥0.7 as moderate and AUROC <0.7 as poor.

Results:

AUROC values of S-index, GPR, APRI, FIB-4 index, FibroQ and PAPAS for diagnosing significant fibrosis were 0.683, 0.667, 0.679, 0.679, 0.585, 0.606 respectively. AUROC values of S-Index, GPR, APGA and FIB-4 index, APRI, FibroQ, PAPAS, FI for diagnosing cirrhosis were 0.841, 0.833, 0.819, 0.802, 0.767, 0.700, 0.697, 0.620 respectively.

Conclusion:

Diagnostic performance of S-Index for diagnosing cirrhosis and significant fibrosis was found superior to other indexes, but diagnostic performance of all these indexes was poor in predicting significant fibrosis. Diagnostic performance of S-Index, APGA, GPR, and FIB-4 index were good in determining cirrhosis.

Keywords: Hepatitis B, Liver fibrosis, non-invasive fibrosis indexes

Introduction

Hepatitis B virus (HBV) infection is one of the major causes of morbidity and mortality around the world. About 240 million people worldwide are known to have chronic HBV infection (1). The prevalence of chronic HBV infection in adult population in Turkey is 4%, also 40-45% of all patients with chronic hepatitis and cirrhosis have HBV infection (2).

To reduce the mortality associated with cirrhosis and hepatocellular carcinoma caused by HBV infection, it is important to start treatment in optimal time. Liver fibrosis grade is the most important indicator for timing of treatment. Also, it is a predictor of treatment response and prognosis (3,4,5). Liver biopsy is the gold standard for evaluating fibrosis. However, liver biopsy is an invasive procedure and it is not always accepted by patients and requires expert histopathological interpretation. There are also limitations of biopsy such as interobserver variability and sample variability (6). These drawbacks have led to conduction of studies on the evaluation of liver fibrosis by non-invasive methods. According to research in this field, it may be possible to diagnose fibrosis grade with using fibrosis biomarkers.

Direct fibrosis biomarkers (enzymatic indicators, collagen markers, glycoproteins and matrix-metalloproteinase indicators and glycosaminoglycans) reflect fibrogenic changes and extracellular matrix cycle at the cellular level in the liver. However, these indicators are not liver-specific, but also have disadvantages such as cost and availability difficulties in routine clinical practice (7). Indirect markers include gamma glutamyl transferase (GGT), alanine aminotransferase (ALT), aspartate aminotransferase (AST), prothrombin time, albumin and bilirubin levels, reflecting alteration in hepatic function. These markers are also useful in diagnosing, evaluating severity and assessing the prognosis of liver diseases (7).

Combination of different indirect fibrosis markers such as AST to platelet ratio (APRI), Fibrosis Index based on four factors (FIB-4), AST/platelet/GGT/Alpha-fetoprotein (AFP) Index (APGA), Fibrosis Index, Fibro-quotient (FibroQ), AST/ALT ratio (AAR), GGT/Platelet ratio (GPR), Platelet-age-phosphatase-AFP-AST (PAPAS) and S-index can improve sensitivity and specificity of these tests (7,8,9,10,11,12,13,14,15,16).

However, most of these scores have not been validated in independent data sets, therefore they cannot be used routinely in clinical practice (17). The aim of this study was to evaluate the diagnostic performance of APRI, FIB-4 Index, APGA, FI, FibroQ, AAR, GPR, PAPAS and S-index in chronic hepatitis B (CHB) patients.


Materials and Methods

A total of 466 consecutive treatment naive CHB patients who underwent liver biopsy between 2012 and 2017 were screened. Demographic, serologic and biochemical data performed within one month before the biopsy were recorded from file and computer database of patients. CHB defined as hepatitis B surfage antigen positivity for more than six months.

Patients who have hepatitis C, delta virus, human immunodeficiency virus (HIV) infection, with a history of alcohol intake higher than 20 gr/day, accompanying autoimmune hepatitis, fewer than 6 portal areas on liver biopsy, and lack of any biochemical parameters used to calculate non-invasive markers were excluded.

Non-invasive fibrosis scores (APRI, FIB-4, APGA, FibroQ, FI, AAR, GPR, PAPAS, S-index) of patients were calculated. Methods for calculating non-invasive markers are shown in Table 1.

Liver biopsies of all patients were reevaluated by an experienced pathologist who was blinded to the clinical and laboratory findings. Fibrosis stage and histological activity were recorded according to ISHAK score. The diagnostic efficacy of non-invasive fibrosis indicators for significant fibrosis (≥F3) and cirrhosis (≥F5) was evaluated.

The diagnostic performance of the non-invasive markers was defined as the AUROC value of ≥0.9 as excellent, 0.9> AUROC ≥0.8 as good, 0.8> AUROC ≥0.7 as moderate and AUROC <0.7 as poor (18). This study was approved by the Local Ethical Committee of Ümraniye Training and Research Hospital (approval number: B10.1TKH.4.34.H.GP.0.01/34). İnformed consent of patients couldn’t obtained due to retrospective design of study.


Statistical Analysis

Descriptive (mean, standard deviation, minimum, median, maximum) statistics were used to define continuous variables. The relationship between independent two categorical variables was evaluated by Fisher’s exact test. The comparison of two continuous variables who distributed not normally was evaluated by Mann-Whitney U test. Logistic regression analysis was performed to identify independent risk factors for  significant fibrosis and cirrhosis. Diagnostic performance of non-invasive fibrozis markers were evaluated by receiver operating curve (ROC) analysis. Significance was set at a p-value of <0.05. The analyzes were performed using the MedCalc Statistical Software version 12.7.7 (MedCalc Software bvba, Ostend, Belgium; http://www.medcalc.org; 2013).


Results

Forty-nine patients who have at least one of the exclusion criteria were excluded from the study. Four hundred seventeen patients were included in the study. Flow chart the of study is shown in Figure 1.


1. Demographic Characteristics of Patients

One hundred and sixty-one (38.6%) of the patients were female and 256 (61.4%) were male. The mean age was 42.26±11.88 years. Two hundred and twenty-one (52.7%) of the patients had significant fibrosis, 80 (19.1%) had advanced fibrosis and 29 (6.9%) had cirrhosis. Demographic characteristics of the patients are shown in Table 2.


2. Risk Factors Associated with Fibrosis


2.1 Factors associated with significant fibrosis

Risk factors for significant fibrosis were determined as AST [p=0.014, odds ratio (OR): 1.026, 95% confidence interval (CI) Lower: 1.005-95% CI Upper: 1.048], GGT (p=0.001, OR: 1.022, 95% CI Lower: 1.008-95% CI Upper: 1.035), Albumin (p=0.009, OR: 0.456, 95% CI Lower: 0.252-95% CI Upper: 0.825) and PLT levels (p=0.001, OR: 0.994, 95% CI Lower: 0.990-95% CI Upper: 0.997). These findings are shown in Table 3.


2.2 Factors Associated with Cirrhosis

Risk factors for cirrhosis (F≥5) was determined as male gender (p=0.020, OR: 4.078, 95% CI Lower: 1.246-95% CI Upper: 13.348), GGT (p=0.031, OR: 1.013, 95% CI Lower: 1.001-95% CI Upper: 1.025) and AFP level (p=0.006, OR: 1.062, 95% CI Lower: 1.017-95% CI Upper: 1.109). These findings are shown in Table 3.


3. Diagnostic Performance of Non-invasive Markers


3.1 Significant fibrosis

Statistically significant difference was found between F≥3 and F <3 groups in terms of APGA, FIB-4 Index, FibroQ, S-index, APRI, PAPAS, GPR Index distributions (Mann-Whitney U, p<0.05). The mean values ​​of these markers were higher in the patients with significant fibrosis. In the ROC analysis, S-index, GPR, APRI, FIB-4 Index, FibroQ and PAPAS scores showed poor diagnostic performance (AUROC <0.7). AUROC value S-index, GPR, APRI, FIB-4 index, FibroQ and PAPAS for diagnosing significant fibrosis were 0.683, 0.667, 0.679, 0.679, 0.585, 0.606 respectively. Cut off points, sensitivity, specificity, positive predictive values (PPV), negative predictive values (NPV), positive and negative likely ratios (LR) of these markers in diagnosing significant fibrosis are shown in Table 4. ROC analysis of non-invasive markers was shown in Figure 2. FI, APGA Index and AAR were not useful in the diagnosis of significant fibrosis.


3.2 Cirrhosis

Statistically significant difference was found between the groups F≥5 and F<5 (cirrhosis vs non-cirrhosis) in terms of APGA, FIB-4, FI, FibroQ, S-index, APRI, PAPAS, GPR distributions (Mann-Whitney U, p<0.05). The mean values of these indicators were significantly higher in the cirrhosis group. The diagnostic performance of these indicators was evaluated by ROC analysis. Diagnostic performance of S-index (AUROC: 0.841), GPR (AUROC: 0.833), APGA (AUROC: 0.819) and FIB-4 Index (AUROC: 0.802) were good, APRI (AUROC: 0.767), FibroQ (AUROC: 0.700) were moderate and PAPAS (AUROC: 0.697), FI (AUROC: 0.620) were poor. Cut off points, sensitivity, specificity, PPV, NPV, positive and negative LR of these markers in diagnosing cirrhosis are shown in Table 5. ROC analysis of non-invasive markers are shown in Figure 3A, B.


Discussion

In the management of CHB, the grade of liver fibrosis is an important determinant of prognosis and timing of the treatment decision. Liver biopsy is the gold standard for detecting fibrosis, however the procedure is invasive, costly and not always repeatable. Furthermore, biopsy may not accurately reflect the stage of fibrosis due to heterogeneous distribution and a small sampling size (19). In addition, biopsy material should be evaluated by experienced pathologists (11). For this reason, studies are carried out to determine fibrosis grade by the non-invasive methods. In this study, the diagnostic performance of simple non-invasive fibrosis markers (APGA, FI, FIB-4, FibroQ, S-index, APRI, AAR and GPR) were evaluated.

We found that S-index, APRI, FIB-4 Index, GPR, PAPAS and FibroQ indicators can detect accurately significant fibrosis. In diagnosis of significant fibrosis, the AUROC value of the S-index was higher than the other non-invasive indicators, but the diagnostic performance of the S-index and others were poor (AUROC <0.700). PPV, AUROC and positive likelihood ratio values of S-index, APRI, FIB-4, GPR Index were found close to each other. If these tests are used at optimal cut-off points, 26-28% of patients can be diagnosed to have significant fibrosis as false positive.

In addition, we found that the AUROC value of the S-index in the diagnosis of cirrhosis was high (AUROC: 0.841), and the diagnostic performance was better than the other non-invasive indicators. Along with S-index, we also found that GPR (AUROC: 0.833), APGA (AUROC: 0.819) and FIB-4 (AUROC: 0.802) indexes had good diagnostic performance in detecting cirrhosis. If these tests are used for a value at or below the optimal cut-off point to exclusion of cirrhosis, %98-99 of patients will be determined correctly.

The APRI and FIB-4 Indexes are non-invasive fibrosis indicators used firstly in patients with HCV or HCV/HIV co-infection in the Western population. In 2015, a meta-analysis evaluated the diagnostic performance of APRI, FIB-4 Index in CHB, reported that AUROC values were found to be 0.740, 0.784 for significant fibrosis and 0.726, 0.844 for cirrhosis respectively (20). In 2015, the WHO recommended the use of the APRI score (APRI score >2 in adults) to assessing the presence of cirrhosis where source limited settings in CHB patients (21). However, in our study, diagnostic performance of APRI was not as good as S-index, GPR, APGA and FIB-4 for determining cirrhosis.

Zhou et al. (13) reported that S-index has good diagnostic performance in detecting significant fibrosis (AUROC: 0.812) and cirrhosis (AUROC: 0.890) (14). Also, Tag-Adeen et al. (22) confirmed this result (AUROC: 0.810) in diagnosing significant fibrosis and reported that the S-index was excellent in the diagnosis of cirrhosis (AUROC: 0.960), superior to the APRI, FIB-4 Index. In our study, the diagnostic performance of the S-index was not as good as the results reported by Zhou et al. (13) and Tag-Adeen et al. (22) These inconsistencies may be related to different demographic, viral characteristics of the study groups and ethnic differences. Also, in our study, liver fibrosis was evaluated with Ishak score but Zhou et al. (13) and Tag-Adeen et al. (22) used Scheuer’s and Metavir scores, respectively. ISHAK score documents the minimal changes in fibrosis stage better than other classifications and evaluates fibrosis as 7 stages (23,24).

Lemoine et al. (16) reported that the diagnostic performance of GPR was superior to the APRI and FIB-4 Index in the diagnosis of significant fibrosis (AUROC: 0.720 and 0.730 in different cohorts) and cirrhosis (AUROC: 0.830 and 0.870) in CHB patients (17). However, in a recent meta- analysis it has been shown that, GPR has moderate diagnostic accuracy for predicting HBV-related significant fibrosis, severe fibrosis, and cirrhosis (AUROC values 0.733, 0.777, and 0.796, respectively) (25). Our study demonstrated that diagnostic performance of GPR was poor in significant fibrosis and good in cirrhosis. Different from our study, the use of elastography as a reference in the study of Lemoine et al. (16) may explain the inconsistency of results between studies.

Our study demonstrated that APGA, FI and AAR Idexes could’t determinated the significant fibrosis, also AAR could’t diagnosed the cirrhosis. In addition, FibroQ, PAPAS and FI scores were found to be weaker than other indexes to diagnosis of cirrhosis.


Study Limitations

Retrospective and single centre design are limitation of our study. İn adition, the distribution of most patients between fibrosis stage 2,3 might have negatively affect the diagnostic performance results of non-invasive tests.


Conclusion

We found that the diagnostic performance of S-index for diagnosing cirrhosis and significant fibrosis was superior to GPR APRI, FIB-4 Index, FibroQ, PAPAS, AAO, APGA and FI indices in patients with CHB. However, diagnostic performance of S-index, GPR, APRI, FIB-4, FibroQ, PAPAS indices were poor in predicting significant fibrosis (AUROC <0.700). Therefore, we believe that these indirect non-invasive fibrosis indicators have limited value for diagnosing significant fibrosis. The diagnostic performance of S-index, APGA, GPR, and FIB-4 Index were good for excluding cirrhosis. We think that these indexes can be used to excluding CHB related cirrhosis in source limited regions.


Ethics

Ethics Committee Approval: The ethics committee approval for the study was obtained from the Local Ethics Committee of Ümraniye Training and Research Hospital (approval number: B10.1TKH.4.34.H.GP.0.01/34, date: 20/02/2019).

Informed Consent: Couldn’t obtained due to retrospective design.

Peer-review: Externally peer-reviewed.

Authorship Contributions

Surgical and Medical Practices: K.K., R.K., Z.Ç., S.S., H.D., Gü.A., R.A., Concept: S.S., Design: S.S., Data Collection or Processing: S.S., R.A., Ş.Ç., K.K., Gü.A., Analysis or Interpretation: G.A., K.Ö., S.S., Literature Search: R.A., H.L.D., O.Ö., Writing: S.S.

Conflict of Interest: The authors declare no conflict of interest.

Financial Disclosure: The authors declare that this study has not received any financial support.

  1. Ott JJ, Stevens G A, Groeger J, Wiersma ST. Global epidemiology of hepatitis B virus infection: new estimates of age-specific HBsAg seroprevalence and endemicity. Vaccine. 2012:30;2212-2219.
  2. Tozun N, Ozdogan O, Cakaloglu Y, Idilman R, Karasu Z, Akarca U, Kaymakoglu S, Ergonul O. Seroprevalance of hepatitis B and C virus infections and risk factors in Turkey: a fieldwork TURHEP study. Clin Microbiol Infect. 2015:21;1020-1026.
  3. Tsochatzis EA, Bosch J, Burroughs AK. Liver cirrhosis. Lancet. 2014;383:1749-1761.
  4. Weissberg JI, Andres LL, Smith CI, Weick S, Nichols J E, Garcia G, Gregory PB. Survival in chronic hepatitis B: an analysis of 379 patients. Ann Intern Med. 1984:101;613-616.
  5. Kim SU, Kim BK, Park JY, Kim DY, Ahn SH, Song K, Han KH. Transient Elastography is Superior to FIB-4 in Assessing the Risk of Hepatocellular Carcinoma in Patients with Chronic Hepatitis B. Medicine (Baltimore). 2016;95:e3434.
  6. Castera L. “Noninvasive methods to assess liver disease in patients with hepatitis B or C.” Gastroenterology 2012;142:1293-1302.
  7. Nallagangula KS, Nagaraj SK, Venkataswamy L, Chandrappa M. Liver fibrosis: a compilation on the biomarkers status and their significance during disease progression.   Future Sci OA. 2017;4:FSO250.
  8. Chin JL, Pavlides M, Moolla A, Ryan JD. Non-invasive markers of liver fibrosis: adjuncts or alternatives to liver biopsy? Front Pharmacol. 2016;7:159.
  9. Fung J, Lai CL, Fong DYT, Yuen JCH, Wong DKH, Yuen MF. Correlation of liver biochemistry with liver stiffness in chronic hepatitis B and development of a predictive model for liver fibrosis. Liver Int. 2008;28:1408-1416.
  10. Ohta T, Sakaguchi K, Fujiwara A, Fujioka SI, Iwasaki Y, Makino Y, Shiratori Y. Simple surrogate index of the fibrosis stage in chronic hepatitis C patients using platelet count and serum albumin level. Acta Med Okayama. 2006;60:77-84.
  11. Kim BK, Kim DY, Park JY, Ahn SH, Chon CY, Kim JK, Han KH. Validation of FIB-4 and comparison with other simple noninvasive indices for predicting liver fibrosis and cirrhosis in hepatitis B virus-infected patients. Liver Int. 2010;30:546-553.
  12. Hsieh YY, Tung SY, Lee IL, Lee K, Shen CH, Wei KL, Lin YH. FibroQ: an easy and useful noninvasive test for predicting liver fibrosis in patients with chronic viral hepatitis. Chang Gung Med J. 2009;32:614-622.
  13. Zhou K, Gao CF, Zhao Y P, Liu HL, Zheng RD, Xian JC, Lu LG. “Simpler score of routine laboratory tests predicts liver fibrosis in patients with chronic hepatitis B.” J Gastroenterol Hepatol. 2010;25:1569-1577.
  14. Shin WG, Park SH, Jang  MK, Hahn TH, Kim JB, Lee MS, Park CK. Aspartate aminotransferase to platelet ratio index (APRI) can predict liver fibrosis in chronic hepatitis B. Dig Liver Dis. 2008;40:267-274.
  15. Williams AL, Hoofnagle JH. Ratio of serum aspartate to alanine aminotransferase in chronic hepatitis relationship to cirrhosis.” Gastroenterology. 1988;95:734-739.
  16. Lemoine M, Shimakawa Y, Nayagam S, Khalil M, Suso P, Lloyd J, Cooke G. The gamma-glutamyl transpeptidase to platelet ratio (GPR) predicts significant liver fibrosis and cirrhosis in patients with chronic HBV infection in West Africa. Gut. 2016;65:1369-1376.
  17. Parikh P, Ryan JD, Tsochatzis EA. Fibrosis assessment in patients with chronic hepatitis B virus (HBV) infection. Ann Transl Med. 2017;5:40.
  18. Hanley JA, McNeil BJ. “The meaning and use of the area under a receiver operating characteristic (ROC) curve.” Radiology. 1982;143:29-36.
  19. Regev A, Berho M, Jeffers LJ, Milikowski C, Molina EG, Pyrsopoulos NT, Schiff ER. “Sampling error and intraobserver variation in liver biopsy in patients with chronic HCV infection.” Am J Gastroenterol. 2002;97:2614-2618.
  20. Xiao G, Yang J, Yan L. “Comparison of diagnostic accuracy of aspartate aminotransferase to platelet ratio index and fibrosis‐4 index for detecting liver fibrosis in adult patients with chronic hepatitis B virus infection: a systemic review and meta‐analysis.” Hepatology. 2015;61:292-302.
  21. World Health Organization. Guidelines for the Prevention Care and Treatment of Persons with Chronic Hepatitis B Infection: Mar-15. World Health Organization. (2015)
  22. Tag-Adeen M, Omar MZ, Abd-Elsalam FM, Hasaneen A, Mohamed MA, Elfeky HM, Said EM, Abdul-Aziz B, Osman AH, Ahmed ES, Osman GS, Abdul-Samie T. “Assessment of liver fibrosis in Egyptian chronic hepatitis B patients: A comparative study including 5 noninvasive indexes.” Medicine (Baltimore). 2018;97:e9781.
  23. Ishak K, Baptista A, Bianchi L, Callea F, De Groote J, Gudat F, Denk H, Desmet V, Korb G, MacSween RN, et al. Histological grading and staging of chronic hepatitis. J Hepatol. 1995;22:696-699.
  24. Bedossa P, Poynard T. An algorithm for the grading of activity in chronic hepatitis C. The METAVIR Cooperative Study Group. Hepatology. 1996;24:289-293. 
  25. Lian MJ, Zhang JQ, Chen SD, Zhang DD, Yang YY, Hong GL. Diagnostic accuracy of γ-glutamyl transpeptidase-to-platelet ratio for predicting hepatitis B-related fibrosis: a meta-analysis. Eur J Gastroenterol Hepatol. 2019;31:599-606.