Chia-Chi Wangd, Chen-Hua Liud, Chih-Lin Lin, Pin-Chao Wang, Tai-Chung Tseng, Hans Hsienhong Lin, Jia-Horng Kao
Abstract
Background/Purpose
Liver biopsy is the gold standard to determine the severity of hepatic fibrosis despite its risk and invasiveness. The aspartate aminotransferase/platelet ratio index (APRI) could noninvasively predict the severity of hepatic fibrosis in chronic hepatitis C (CHC) patients. Whether fibrosis index based on four factors (FIB-4) could better predict the severity of hepatic fibrosis than APRI in CHC patients remains inconclusive.
Liver biopsy is the gold standard to determine the severity of hepatic fibrosis despite its risk and invasiveness. The aspartate aminotransferase/platelet ratio index (APRI) could noninvasively predict the severity of hepatic fibrosis in chronic hepatitis C (CHC) patients. Whether fibrosis index based on four factors (FIB-4) could better predict the severity of hepatic fibrosis than APRI in CHC patients remains inconclusive.
Methods
This retrospective study enrolled 1473 CHC patients (784 men and 689 women) with liver biopsy and clinical data including age, aspartate aminotransferase, alanine aminotransferase, and platelet count. FIB-4 and APRI were calculated with a formula using the four clinical parameters. Hepatic fibrosis was staged using the Metavir classification system.
Results
The areas under the receiver operating characteristics of FIB-4 for the diagnosis of significant fibrosis (≥ F2), advanced fibrosis (≥ F3), and cirrhosis (F4) were 0.816 [95% confidence interval (CI), 0.795–0.836], 0.827 (95% CI, 0.806–0.849), and 0.849 (95% CI, 0.830–0.867), respectively, compared with those of APRI—0.799 (95% CI, 0.778–0.819), 0.791 (95% CI, 0.770–0.812), and 0.802 (95% CI, 0.781–0.922). In addition, the areas under the receiver operating characteristics of FIB-4 were significantly greater than those of APRI for patients with advanced fibrosis and cirrhosis, respectively (p < 0.0001).
The areas under the receiver operating characteristics of FIB-4 for the diagnosis of significant fibrosis (≥ F2), advanced fibrosis (≥ F3), and cirrhosis (F4) were 0.816 [95% confidence interval (CI), 0.795–0.836], 0.827 (95% CI, 0.806–0.849), and 0.849 (95% CI, 0.830–0.867), respectively, compared with those of APRI—0.799 (95% CI, 0.778–0.819), 0.791 (95% CI, 0.770–0.812), and 0.802 (95% CI, 0.781–0.922). In addition, the areas under the receiver operating characteristics of FIB-4 were significantly greater than those of APRI for patients with advanced fibrosis and cirrhosis, respectively (p < 0.0001).
Conclusion
FIB-4 could predict hepatic fibrosis in CHC patients. By adding two parameters (age and alanine aminotransferase), FIB-4 better predicts advanced fibrosis and cirrhosis than APRI in CHC patients.
FIB-4 could predict hepatic fibrosis in CHC patients. By adding two parameters (age and alanine aminotransferase), FIB-4 better predicts advanced fibrosis and cirrhosis than APRI in CHC patients.
Discussion Only
Full Text Article
Article Outline
Introduction
Methods
Study overview
Biochemical and virological tests
Histological evaluations of liver biopsy specimens
Ethical considerations
Statistical analysis
Results
Patients
Comparison of diagnostic accuracy between APRI and FIB-4
Comparison of diagnostic accuracy between APRI and FIB-4 categorized by HCV genotype 1 or 2
Selective cutoff values of FIB-4 for different stages of hepatic fibrosis
Discussion
In this study, 1473 CHC patients with mainly genotype 1 and/or 2 infection were enrolled. By using AUROC analysis, our data show that both FIB-4 and APRI can predict the severity of hepatic fibrosis in CHC patients. In addition, FIB-4 can better predict advanced fibrosis and cirrhosis than APRI (p < 0.0001). These findings hold true in either genotype 1 or 2 CHC patients. In clinical practice, if FIB-4 was < 1.45, the NPV for advanced hepatic fibrosis was 92.7%. If the FIB-4 index was > 2.15, the PPV of significant fibrosis (≥ F2) was 91%.
Hepatic fibrosis is one of the important factors associated with the long-term prognosis of CHC patients. If noninvasive methods could accurately predict the severity of hepatic fibrosis, the majority of liver biopsies could be avoided.20 In recent years, transient elastography has been accepted as an alternative method to assess the severity of hepatic fibrosis. However, several limitations could affect its clinical usefulness. The cost of transient elastography may preclude its widespread availability, and it has a decreased accuracy in specific population groups including obese patients or patients with ascites or hepatitis flare.21, 22, 23 APRI and FIB-4 can predict the severity of hepatic fibrosis in CHC patients. Furthermore, the difference of FIB-4 per year was reported to predict the risk of disease progression to cirrhosis.24, 25 In addition, they are easily calculated in clinical practice, and serial data can be obtained to follow up on disease progression.
Our results provided additional evidence that FIB-4 can predict the severity of hepatic fibrosis, not only in predicting advanced fibrosis or cirrhosis but also in significant fibrosis (AUROC = 0.816; 95% CI, 0.795–0.836). A previous French study validated that FIB-4 could predict advanced fibrosis or cirrhosis in HCV-monoinfected patients with 847 liver biopsies and 780 FIB-4 index values. Although their AUROC seems better than ours in predicting advanced fibrosis or cirrhosis (0.91 and 0.85 vs. 0.85 and 0.83, respectively), the data of viral load and genotype were not provided. In this study, the sample size was larger and all patients were Taiwanese with mainly (92.6%) HCV genotype 1 and/or 2 infection.
To the best of our knowledge, this study is the first one from Asia to validate the superiority of FIB-4 compared with APRI in CHC patients. APRI is known to be associated with the severity of hepatic fibrosis in CHC patients.26 By adding two clinical parameters (age and ALT) to the backbone of APRI, FIB-4 better predicted advanced fibrosis and cirrhosis than APRI in CHC patients. A recent study from Egypt found that FIB-4 had the best performance curve in predicting moderate or severe fibrosis (≥ F2) compared with APRI using AUROC analysis.27 However, PPV and NPV (0.56 and 0.76, respectively) seemed lower than those in previous reports. In addition, the statistical significance was not confirmed. Recently, a large cohort of Western CHC patients showed that the AUROC of FIB-4 was significantly higher than that of APRI in predicting advanced fibrosis.28 Moreover, another two studies found that FIB-4 was superior to APRI in the prediction of severe fibrosis and cirrhosis.29, 30 In clinical practice, because these high-risk patients could be identified using FIB-4 score, they could receive frequent monitoring, and aggressive treatment should be given as early as possible.
Taking these data together, we suggest using FIB-4 instead of APRI to predict hepatic fibrosis in clinical practice, and urge practicing physicians to measure platelet count in addition to routine AST, ALT, and AFP testing as well as ultrasound examination for the regular follow-up of CHC patients. Just like the role of estimated glomerular filtration rate (eGFR) (formula = 186.3 × serum creatinine −1.154 × age −0.203 × (0.742 if female) × (1.212 if black)) in the prediction of renal function,31 if FIB-4 index can be automatically calculated with the aid of a computer, we can easily understand the status of hepatic fibrosis and its progression.
This study has several unique features. First, this is the first validation of FIB-4 in Asian CHC patients. Thus, we can validate the power of FIB-4 for predicting hepatic fibrosis in Asian patients. Second, the sample size is large, and different severity levels of hepatic fibrosis—including significant, advanced fibrosis, or cirrhosis—were analyzed using AUROC between APRI and FIB-4. Thus, we could understand the performance of FIB-4 in different severity levels of hepatic fibrosis compared with APRI. Third, diagnostic accuracy between FIB-4 and APRI was compared not only in the overall population but also in subgroups categorized by HCV genotype 1 or 2. However, several limitations are also noted. First, although a small proportion of patients (7.4%) lacked data on HCV RNA and genotype, the percentage was acceptable. Second, the possible interobserver variability between different pathologists in determining the severity of hepatic fibrosis could not be avoided.
In summary, the FIB-4 index is validated to noninvasively predict the severity of hepatic fibrosis, and compared with APRI it better predicts advanced fibrosis and cirrhosis in Asian CHC patients. It is noninvasive, simple, and widely available in clinical practice. Just like eGFR for predicting renal function, it can help physicians to understand the severity of hepatic fibrosis and follow up on disease progression in CHC patients.
Hepatic fibrosis is one of the important factors associated with the long-term prognosis of CHC patients. If noninvasive methods could accurately predict the severity of hepatic fibrosis, the majority of liver biopsies could be avoided.20 In recent years, transient elastography has been accepted as an alternative method to assess the severity of hepatic fibrosis. However, several limitations could affect its clinical usefulness. The cost of transient elastography may preclude its widespread availability, and it has a decreased accuracy in specific population groups including obese patients or patients with ascites or hepatitis flare.21, 22, 23 APRI and FIB-4 can predict the severity of hepatic fibrosis in CHC patients. Furthermore, the difference of FIB-4 per year was reported to predict the risk of disease progression to cirrhosis.24, 25 In addition, they are easily calculated in clinical practice, and serial data can be obtained to follow up on disease progression.
Our results provided additional evidence that FIB-4 can predict the severity of hepatic fibrosis, not only in predicting advanced fibrosis or cirrhosis but also in significant fibrosis (AUROC = 0.816; 95% CI, 0.795–0.836). A previous French study validated that FIB-4 could predict advanced fibrosis or cirrhosis in HCV-monoinfected patients with 847 liver biopsies and 780 FIB-4 index values. Although their AUROC seems better than ours in predicting advanced fibrosis or cirrhosis (0.91 and 0.85 vs. 0.85 and 0.83, respectively), the data of viral load and genotype were not provided. In this study, the sample size was larger and all patients were Taiwanese with mainly (92.6%) HCV genotype 1 and/or 2 infection.
To the best of our knowledge, this study is the first one from Asia to validate the superiority of FIB-4 compared with APRI in CHC patients. APRI is known to be associated with the severity of hepatic fibrosis in CHC patients.26 By adding two clinical parameters (age and ALT) to the backbone of APRI, FIB-4 better predicted advanced fibrosis and cirrhosis than APRI in CHC patients. A recent study from Egypt found that FIB-4 had the best performance curve in predicting moderate or severe fibrosis (≥ F2) compared with APRI using AUROC analysis.27 However, PPV and NPV (0.56 and 0.76, respectively) seemed lower than those in previous reports. In addition, the statistical significance was not confirmed. Recently, a large cohort of Western CHC patients showed that the AUROC of FIB-4 was significantly higher than that of APRI in predicting advanced fibrosis.28 Moreover, another two studies found that FIB-4 was superior to APRI in the prediction of severe fibrosis and cirrhosis.29, 30 In clinical practice, because these high-risk patients could be identified using FIB-4 score, they could receive frequent monitoring, and aggressive treatment should be given as early as possible.
Taking these data together, we suggest using FIB-4 instead of APRI to predict hepatic fibrosis in clinical practice, and urge practicing physicians to measure platelet count in addition to routine AST, ALT, and AFP testing as well as ultrasound examination for the regular follow-up of CHC patients. Just like the role of estimated glomerular filtration rate (eGFR) (formula = 186.3 × serum creatinine −1.154 × age −0.203 × (0.742 if female) × (1.212 if black)) in the prediction of renal function,31 if FIB-4 index can be automatically calculated with the aid of a computer, we can easily understand the status of hepatic fibrosis and its progression.
This study has several unique features. First, this is the first validation of FIB-4 in Asian CHC patients. Thus, we can validate the power of FIB-4 for predicting hepatic fibrosis in Asian patients. Second, the sample size is large, and different severity levels of hepatic fibrosis—including significant, advanced fibrosis, or cirrhosis—were analyzed using AUROC between APRI and FIB-4. Thus, we could understand the performance of FIB-4 in different severity levels of hepatic fibrosis compared with APRI. Third, diagnostic accuracy between FIB-4 and APRI was compared not only in the overall population but also in subgroups categorized by HCV genotype 1 or 2. However, several limitations are also noted. First, although a small proportion of patients (7.4%) lacked data on HCV RNA and genotype, the percentage was acceptable. Second, the possible interobserver variability between different pathologists in determining the severity of hepatic fibrosis could not be avoided.
In summary, the FIB-4 index is validated to noninvasively predict the severity of hepatic fibrosis, and compared with APRI it better predicts advanced fibrosis and cirrhosis in Asian CHC patients. It is noninvasive, simple, and widely available in clinical practice. Just like eGFR for predicting renal function, it can help physicians to understand the severity of hepatic fibrosis and follow up on disease progression in CHC patients.
Introduction
Methods
Study overview
Biochemical and virological tests
Histological evaluations of liver biopsy specimens
Ethical considerations
Statistical analysis
Results
Patients
Comparison of diagnostic accuracy between APRI and FIB-4
Comparison of diagnostic accuracy between APRI and FIB-4 categorized by HCV genotype 1 or 2
Selective cutoff values of FIB-4 for different stages of hepatic fibrosis
Discussion
No comments:
Post a Comment