Abstract
BACKGROUND AND AIMS: Accurate assessment of liver inflammation and fibrosis is of vital importance in the clinical management of patients with liver diseases. Our aim is to conduct a meta-analysis to evaluate the diagnostic accuracy of serum cytokeratin 18 (CK18) for staging of liver inflammation and fibrosis against a liver biopsy in adults. METHODS: We systematically searched articles from eight electronic databases. Two authors independently selected included studies, extracted data, and assessed quality. In our meta-analysis, we used the random-effects meta-analysis model. Publication bias, sensitivity analysis, heterogeneity analysis, and post-test probability were used in this meta study. RESULTS: A total of 20 studies with 2235 patients were initially found by the search strategies. The pooled sensitivity, specificity, and area under the curve (AUC) of the summary receiver operating characteristic curve were 0.56, 0.81, and 0.810 for significant fibrosis; 0.64, 0.76, and 0.785 for advanced fibrosis; 0.53, 0.76, and 0.830 for cirrhosis; and 0.68, 0.73, and 0.786 for significant inflammation, respectively. High heterogeneity was observed in our meta-analysis because of factors such as the proportion of males, total number, and antigens of CK-18. CONCLUSION: Serum CK18 had moderate diagnostic value (AUC > 0.7) in different stages of liver fibrosis and significant inflammation, offering a complementary approach to other non-invasive indicators such as serological biomarkers and imaging techniques. Future research should focus on elucidating the role of CK18 in the occurrence and progression of hepatitis and liver fibrosis, particularly in liver diseases with diverse etiologies.