Evaluation and comparison of the diagnostic performance of routine blood tests in predicting liver fibrosis in chronic hepatitis B infection

ABSTRACT

Background & aims: Biopsy is the gold standard for staging liver fibrosis, but it may be accompanied by complications. As an alternative, non-invasive markers such as transient elastography (for liver fibrosis) and certain combinations of routine blood markers (liver function tests, full blood count) have been developed although their clinical significance remains controversial. Here, we compare the diagnostic values of non-invasive markers for liver fibrosis in patients with chronic hepatitis B infection.

Methods: Transient elastography and routine laboratory tests were performed in 196 patients. Diagnostic performances were compared and were assessed based on the area under the curve (AUC) of a receiver operating characteristic (ROC) analysis.

Results: Elevated GGT to platelet ratio (GPR), the fibrosis index FIB-4 [based on age, AST, platelets and ALT], platelet to lymphocyte ratio (PLR) and total bilirubin were independent predictors of liver stiffness defined by transient elastography (all P < 0.001). The AUCs of GPR in predicting both advanced fibrosis and cirrhosis were significantly larger than that of FIB-4 (P = 0.037 and P = 0.008, respectively) and AST-to-platelet ratio index (APRI) (P = 0.008 and P = 0.005). FIB-4, APRI and red cell volume distribution width-to-platelet ratio (RPR) had similar diagnostic values in discriminating different levels of liver fibrosis.

Conclusions: GPR showed the best diagnostic value and RPR and PLR are easily available and inexpensive markers in evaluating fibrosis and cirrhosis. The diagnostic values of these laboratory markers are useful in diagnosing advanced fibrosis or cirrhosis, and in confirming the different levels of liver fibrosis.

KEYWORDS:

• Chronic hepatitis B infection
• liver fibrosis
• transient elastography
• non-invasive marker

Summary table

Table

Acknowledgements

Lu W and Zhang YP made the equal contributions to this paper.

Disclosure statement

No potential conflict of interest was reported by the authors.