Supplementary Material for: Mac-2 Binding Protein Glycosylation Isomer to Albumin Ratio Predicts Bacterial Infections in Cirrhotic Patients

Introduction
Mac-2-binding protein glycosylation isomer (M2BPGi) is a novel biomarker for liver fibrosis, but little is known about its role in cirrhosis-associated clinical outcomes. This study aimed to investigate the predictive role of M2BPGi in cirrhosis-associated complications.
Methods
One hundred and forty-nine cirrhotic patients were retrospectively enrolled. Patients were followed up for one year and cirrhosis-associated clinical events were recorded. Receiver operating characteristic curve (ROC) analysis was used to establish the values of the predictive models for cirrhotic outcomes. and Cox proportional hazards regression models were used to identify predictors of clinical outcomes,
Results
Sixty (40.3%) patients experienced cirrhosis-associated clinical events and had higher M2BPGi levels compared to those without events (8.7 vs. 5.1 cut-off index, p < 0.001). The most common cirrhosis-associated complications were bacterial infections (24.2%). On ROC analysis, M2BPGi to albumin ratio (M2BPGi/albumin) had comparable discriminant abilities for all cirrhosis-associated events [area under the ROC curve (AUC) = 0.74] compared with M2BPGi, Child-Pugh, Model for End-Stage Liver Disease, Albumin-Bilirubin scores, and neutrophil to lymphocyte ratio and was superior to M2BPGi alone for all bacterial infectious events (AUC = 0.80). Cox regression analysis revealed that the M2BPGi/albumin, but not M2BPGi alone, independently predicted all cirrhosis-associated events [Hazar ratio (HR) = 1.34, p = 0.038] and all bacterial infectious events (HR = 1.51, p = 0.011) within one year. However, M2BPGi/albumin did not predict other cirrhotic complications and transplant free survival.
Discussion/Conclusion
M2BPGi/albumin might serve as a potential prognostic indicator for patients with cirrhosis, particularly for predicting bacterial infections.

引言
Mac-2结合蛋白糖基化异构体（Mac-2-binding protein glycosylation isomer, M2BPGi）是一种新型肝纤维化生物标志物，但目前对其在肝硬化相关临床结局中的作用尚缺乏深入了解。本研究旨在探讨M2BPGi在肝硬化相关并发症中的预测价值。

方法
本研究回顾性纳入149例肝硬化患者，对其进行为期1年的随访并记录肝硬化相关临床事件。采用受试者工作特征曲线（Receiver Operating Characteristic curve, ROC）分析构建肝硬化结局预测模型的相关参数，并使用Cox比例风险回归模型识别临床结局的预测因子。

结果
60例（40.3%）患者发生肝硬化相关临床事件，其M2BPGi水平高于未发生事件的患者（截断指数分别为8.7与5.1，p<0.001）。最常见的肝硬化相关并发症为细菌感染（24.2%）。经ROC分析，M2BPGi与白蛋白比值（M2BPGi/albumin）对所有肝硬化相关事件的判别能力与M2BPGi、Child-Pugh评分、终末期肝病模型（Model for End-Stage Liver Disease, MELD）、白蛋白-胆红素评分以及中性粒细胞与淋巴细胞比值相当[受试者工作特征曲线下面积（area under the ROC curve, AUC）=0.74]；相较于单独使用M2BPGi，其对所有细菌感染性事件的判别能力更优（AUC=0.80）。Cox回归分析显示，M2BPGi/白蛋白比值而非单独的M2BPGi，可独立预测1年内所有肝硬化相关事件[风险比（Hazard ratio, HR）=1.34，p=0.038]以及所有细菌感染性事件（HR=1.51，p=0.011）。然而，M2BPGi/白蛋白比值无法预测其他肝硬化并发症以及无移植生存率。

讨论与结论
M2BPGi/白蛋白比值或可作为肝硬化患者潜在的预后指标，尤其在预测细菌感染方面具有应用价值。