A total of 137 consecutive patients with CHB who underwent percutaneous liver biopsy at Shanghai Public Health Clinical Center (China) from 2005 to April 2008 were included in this study. Real-time PCR showed that all patients were positive for HBV DNA and had no chronic liver disease confirmed by standard clinical, serological, biochemical, and radiological criteria. Additional exclusion criteria were antiviral treatment before liver biopsy, alcohol consumption in excess of 40 g/d. Liver biopsies were obtained by either blind or ultrasound-guided techniques using a 16-gauge Klatskin needle. The length of biopsy samples was longer than 1.5 cm. All biopsies were read by a pathologist who had no clinical information on the CHB patients. Formalin-fixed and formalin-embedded liver tissues were cut into 4-μm thick sections with a microtome. One section was stained with hematoxylin and eosin for assessment of hepatic inflammatory activity and the other sections were stained with Gomori stain for evaluation of hepatic fibrosis. Biopsy specimens with at least 4 portal fields were considered representative and scored by a pathologist unaware of the laboratory results. Fibrosis was staged as no (0), mild (1), moderate (2), severe (3), and cirrhosis (4), using the METAVIR score[7]. Hepatic inflammatory activity was also scored.

Serum aspartate aminotransferase (AST), ALT, HA and platelet count in all patients within 2 wk after liver biopsy were routinely determined. The ULN for ALT was 50 U/L, and transformed into ULN for further analysis. The reference range for platelet count was 100 × 10⁹-300 × 10⁹/L. HA was measured using the RIA and the reference range was 9-119 ng/mL. According to the ULN of ALT, we divided the patients into two groups with their ALT ≥ 2ULN and < 2ULN, respectively. APRI was calculated as previously described[8,9].

Baseline demographic data were evaluated for comparability of the two groups using Fisher’s exact test for categorical variables and Student’s t test for continuous variables. Student’s t test or analysis of variance was used to compare the means of different stage groups when appropriate. Correlation was evaluated by the Spearman correlation coefficient. The sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and likelihood ratio (LR) of APRI ≥ 1.5 and APRI < 1.5 in combination with different HA cut-off points were detected for the presence of moderate to severe fibrosis/cirrhosis (fibrosis stages 2 and 4) and no to mild fibrosis/cirrhosis (fibrosis stages 0 and 1). P < 0.05 was considered statistically significant.

A total of 137 patients with CHB who underwent percutaneous liver biopsy were included in this study. In order to adequately estimate the predictive model, the patients were divided into 2 groups with their ALT ≥ 2ULN and ALT < 2ULN, respectively. There was no significant correlation between age, gender, HA, APRI and disease stage. The characteristics of patients in the two groups are shown in Table 1.

The distribution of fibrosis stages in the two groups is shown in Table 1. Of the 137 patients, 57 (41.6%) had no or only mild fibrosis (stages 0 and 1), 46 (33.6%) had moderate fibrosis (stage 2), and 34 (24.8%) had severe fibrosis or cirrhosis (stages 3 and 4). As expected, the APRI and HA cut-off points increased with the stage of fibrosis, but there was no significant difference between the two groups. The mean APRI was ≥ 1.5 in patients with moderate to severe fibrosis or cirrhosis (stages 2-4) and < 1.5 in patients with no or mild fibrosis (stages 0 and 1).

Among the patients with their APRI ≥ 1.5, the Spearman correlation coefficient was r = 0.312 and 0.344 between APRI and fibrosis stage (P < 0.05, Figure 1), and r = -0.717 and -0.812 between HA and fibrosis stage (P < 0.05, Figure 2). However, there was no significant difference in patients with their APRI < 1.5 between APRI or HA count and fibrosis stage.

APRI ≥ 1.5 in combination with different HA cut-off points could detect moderate to severe fibrosis (stages 2-4) in CHB patients (Table 2). The sensitivity, specificity, PPV, NPV, +LR and -LR were determined. APRI alone had a PPV of 41.3% and a specificity of 84.7%. When HA was added, the PPV and specificity increased significantly, indicating that APRI ≥ 1.5 in combination with a HA cut-off point > 300 ng/mL can detect moderate to severe fibrosis or cirrhosis (stages 2-4) in CHB patients. In the present study, the PPV was 93.7% and the specificity was 98.9%. On the other hand, not all patients with moderate to severe fibrosis or cirrhosis could be correctly identified. The identification methods for patients with no or mild fibrosis (stages 0 and 1) using APRI < 1.5 in combination with different HA cut-off points are shown in Table 3. Since the sensitivity, specificity, PPV and NPV were low, mild fibrosis in CHB patients could not be detected using these laboratory parameters.

Liver biopsy is the gold standard for hepatic fibrosis in chronic hepatitis B (CHB) patients, especially in those whose alanine aminotransferase (ALT) is under 2ULN or normal, but its value is questioned because of its potential risk and sampling error. Noninvasive markers of liver fibrosis have been recently proposed as substitutes for liver biopsy, but their reported accuracy is around 80%. They have been mostly validated in hepatitis C but not in hepatitis B. They applied their method in a cohort of patients with chronic hepatitis B.

Since sampling error and inter-observer variability are known limitations of liver biopsy, some noninvasive methods for liver fibrosis have been proposed, but international guidelines still do not recommend a routine use of the markers due to lack of reproducibility and a misdiagnosis rate of 20%. Thus, a trusted method that avoids a number of liver biopsies by maintaining an excellent accuracy is urgently needed.

In this study, they established a noninvasive assessment model consisting of APRI and HA for the detection of liver fibrosis and cirrhosis in CHB patients. APRI ≥ 1.5 in combination with a HA cut-off point > 300 ng/mL could detect moderate to severe fibrosis (stages 2-4) in CHB patients. The PPV and specificity of this model were 93.7% and 98.9%, respectively, showing that it can be used as a non-invasive marker for the detection of liver fibrosis in CHB patients.

The predictive model can be used as a first line assessment of significant hepatic fibrosis in CHB patients, limiting liver biopsy to those who are unclassified or show a low predictive value. In the future, priority should be given to large scale validation studies and the most promising non-invasive markers in patients with all major etiologies of chronic liver disease and most frequent cofactors affecting the diagnostic performance of fibrosis markers.

APRI: a simple test combining aspartate aminotransferase (AST) and platelet count for non-invasive prediction of significant fibrosis and cirrhosis in hepatitis C patients. It is a very simple and economic tool, but it is somehow less accurate than fibrotest.

This is an interesting paper addressing a clinical problem in the management of chronic hepatitis B. The usefulness of this combination of non-invasive markers of fibrosis and its place in clinical practice needs to be further studied.