Comparison of FIB‐4 index, NAFLD fibrosis score and BARD score for prediction of advanced fibrosis in adult patients with non‐alcoholic fatty liver disease: A meta‐analysis study

Aim Non-alcoholic fatty liver disease (NAFLD)-related advanced hepatic fibrosis is associated with liver and cardiovascular morbidity and mortality. This study aims to compare the FIB-4 index, NAFLD fibrosis score (NFS) and BARD score for prediction of advanced liver fibrosis. Methods Pooled sensitivity, specificity, diagnostic odds ratio (DOR), summary receiver–operator curves (SROC) and Spearman's rank correlation coefficient were used to examine the accuracy of each non-invasive scoring system for predicting NAFLD-related advanced fibrosis. Results Four studies with 1038 adult patients were included in this meta-analysis. A total of 135 patients (13.0%) had advanced fibrosis. In the FIB-4 index group, pooled sensitivity and specificity with 95% confidence interval (CI), and the area under the ROC (AUROC) were 0.844 (0.772–0.901), 0.685 (0.654–0.716) and 0.8496 ± 0.0680, respectively, at a cut-off of 1.30. At a threshold of 3.25, the same parameters were 0.38 (0.30–0.47), 0.96 (0.95–0.98) and 0.8445 ± 0.0981. At a cut-off of −1.455, values were 0.77 (0.69–0.84), 0.70 (0.67–0.73) and 0.8355 ± 0.0667, respectively. At a 0.676 cut-off, pooled sensitivity and specificity with 95% CI were 0.27 (0.19–0.35) and 0.98 (0.96–0.98), respectively; and the AUROC was 0.647 ± 0.2208. In the BARD score group, pooled sensitivity and specificity with 95% CI were 0.74 (0.66–0.81) and 0.66 (0.63–0.69), respectively; and the AUROC was 0.7625 ± 0.0285. Conclusion FIB-4 index with a 1.30 cut-off has better diagnostic accuracy than the FIB-4 index with a 3.25 cut-off, NFS and BARD score, despite showing its limited value for predicting NAFLD-related advanced fibrosis.