Department of Endocrinology, Changzhou Third People's Hospital, Changzhou Medical Center, Nanjing Medical University, Changzhou, China.
Department of Clinical Nutrition, Changzhou Third People's Hospital, Changzhou Medical Center, Nanjing Medical University, Changzhou, China.
Ann Med. 2024 Dec;56(1):2337739. doi: 10.1080/07853890.2024.2337739. Epub 2024 Apr 4.
This study aims to validate the efficacy of the conventional non-invasive score in predicting significant fibrosis in metabolic-associated fatty liver disease (MAFLD) and to develop a non-invasive prediction model for MAFLD.
This cross-sectional study was conducted among 7701 participants with MAFLD from August 2018 to December 2023. All participants were divided into a training cohort and a validation cohort. The study compared different subgroups' demographic, anthropometric, and laboratory examination indicators and conducted logistic regression analysis to assess the correlation between independent variables and liver fibrosis. Nomograms were created using the logistic regression model. The predictive values of noninvasive models and nomograms were evaluated using receiver operating characteristic (ROC) curve analysis and decision curve analysis (DCA).
Four nomograms were developed for the quantitative analysis of significant liver fibrosis risk based on the multivariate logistic regression analysis results. The nomogram's area under ROC curves (AUC) was 0.710, 0.714, 0.748, and 0.715 in overall MAFLD, OW-MAFLD, Lean-MAFLD, and T2DM-MAFLD, respectively. The nomogram had a higher AUC in all MAFLD participants and OW-MAFLD than the other non-invasive scores. The DCA curve showed that the net benefit of each nomogram was higher than that of APRI and FIB-4. In the validation cohort, the AUCs of the nomograms were 0.722, 0.750, 0.719, and 0.705, respectively.
APRI, FIB-4, and NFS performed poorly predicting significant fibrosis in patients with MAFLD. The new model demonstrated improved diagnostic accuracy and clinical applicability in identifying significant fibrosis in MAFLD.
本研究旨在验证传统非侵入性评分在预测代谢相关脂肪性肝病（MAFLD）中显著纤维化的有效性，并建立 MAFLD 的非侵入性预测模型。
本横断面研究纳入了 2018 年 8 月至 2023 年 12 月间的 7701 例 MAFLD 患者，所有患者被分为训练队列和验证队列。本研究比较了不同亚组的人口统计学、人体测量学和实验室检查指标，并进行了逻辑回归分析以评估独立变量与肝纤维化之间的相关性。使用逻辑回归模型建立了列线图。使用受试者工作特征（ROC）曲线分析和决策曲线分析（DCA）评估了非侵入性模型和列线图的预测值。
基于多变量逻辑回归分析结果，建立了 4 个用于定量分析显著肝纤维化风险的列线图。列线图的 ROC 曲线下面积（AUC）在整体 MAFLD、OW-MAFLD、Lean-MAFLD 和 T2DM-MAFLD 中的值分别为 0.710、0.714、0.748 和 0.715。在所有 MAFLD 患者和 OW-MAFLD 患者中，列线图的 AUC 均高于其他非侵入性评分。DCA 曲线表明，每个列线图的净获益均高于 APRI 和 FIB-4。在验证队列中，列线图的 AUC 值分别为 0.722、0.750、0.719 和 0.705。
APRI、FIB-4 和 NFS 在预测 MAFLD 患者的显著纤维化方面表现不佳。新模型在识别 MAFLD 中的显著纤维化方面显示出了更高的诊断准确性和临床适用性。
