ABSTRACT Serum Mac‐2 binding protein glycosylation isomer (M2BPGi) formation requires sugar chain glycosylation on M2BP. Whether glycated hemoglobin (HbA1c) levels affect M2BPGi in diagnosing hepatitis C virus (HCV)‐related hepatic fibrosis remains unclear. We enrolled 2064 patients with available M2BPGi, HbA1c, and valid vibration‐controlled transient elastography (VCTE) for fibrosis staging. Associations between M2BPGi and HbA1c levels across fibrosis stages were analyzed using Kruskal–Wallis test and Spearman's correlation. Multivariable linear regression with Akaike information criterion (AIC) identified factors associated with M2BPGi levels. The diagnostic accuracy of M2BPGi in fibrosis staging was evaluated using the areas under receiver operating characteristics (AUROCs) with 95% confidence intervals (CIs). M2BPGi levels showed no correlation with HbA1c categories (< 5.7%, 5.7%–6.4%, 6.5%–6.9%, 7.0%–7.9%, ≥ 8.0%) in fibrosis stages F2 ( p = 0.48), F3 ( p = 0.63), or F4 ( p = 0.22), except in F0–F1 ( p < 0.001). Spearman's ρ for F0–F1, F2, F3, and F4 were 0.086 ( p = 0.0088), 0.087 ( p = 0.057), 0.056 ( p = 0.41), and −0.023 ( p = 0.62), respectively. Multivariable regression identified fibrosis stage ( β for F4, F3, F2 vs. F0–F1 = 4.2724, 1.7623, 0.7223; p < 0.001), age ( β = 0.0159; p < 0.001), ALT quotient ( β = 0.0094; p < 0.001), steatosis ( β = 0.2460; p < 0.001), but not HbA1c, as independent predictors of M2BPGi. The diagnostic accuracy of M2BPGi for fibrosis staging was comparable across HbA1c categories. In conclusion, HbA1c levels do not influence the diagnostic performance of M2BPGi for HCV hepatic fibrosis staging.
Lin et al. (Thu,) studied this question.