Abstract
OBJECTIVE: This study aimed to assess the accuracy of Mac-2 binding protein glycosylation isomer (M2BPGi) in predicting the stage of liver fibrosis. METHODS: Articles published until October 10, 2023, were searched in the PubMed, Embase, Web of Science, and Cochrane Library databases. 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 M2BPGi in predicting the stage of liver fibrosis. A 95% confidence interval (CI) was provided for each estimate. RESULTS: Twenty-four studies were included in this meta-analysis, including 3,839 patients with liver fibrosis, 409 of whom progressed to stage 4 or above. The pooled sensitivity, specificity, and area under the ROC (AUC) for M2BPGi predicting liver fibrosis ≥F3 were 0.74 (95% CI [0.65-0.82]), 0.84 (95% CI [0.76-0.89]), and 14.99 (95% CI [9.28-24.21]), respectively. The pooled sensitivity, specificity, and AUC for ≥F4 were 0.80 (95% CI [0.70-0.88]), 0.80 (95% CI [0.73-0.86]), and 16.43 (95% CI [0.84-0.90]), respectively. CONCLUSION: Among different sample partitions, M2BPGi has the best diagnostic performance for liver fibrosis stage ≥4. Furthermore, the cutoff of 1-2 is more accurate than that of 0-1 or 2-3 for fibrosis ≥ F3 and ≥ F4. REGISTRATION: CRD42023483260.