临床荟萃 ›› 2025, Vol. 40 ›› Issue (5): 417-422.doi: 10.3969/j.issn.1004-583X.2025.05.006

• 论著 • 上一篇    下一篇

代谢相关脂肪性肝病肝纤维化危险因素分析及构建预测模型

肇雪婷1, 白佳雯1, 孙军2()   

  1. 1.锦州医科大学 研究生院,辽宁 锦州 121000
    2.锦州市中心医院 消化内科,辽宁 锦州 121000
  • 收稿日期:2024-12-16 出版日期:2025-05-20 发布日期:2025-05-23
  • 通讯作者: 孙军 E-mail:50681118@qq.com

Analysis of risk factors for liver fibrosis in metabolic dysfunction-associated fatty liver disease and construction of a nomogram

Zhao Xueting1, Bai Jiawen1, Sun Jun2()   

  1. 1. Graduate School, Jinzhou Medical University, Jinzhou 121000, China
    2. Department of Gastroenterology, Central Hospital of Jinzhou, Jinzhou 121000, China
  • Received:2024-12-16 Online:2025-05-20 Published:2025-05-23
  • Contact: Sun Jun E-mail:50681118@qq.com

摘要:

目的 探讨代谢相关脂肪性肝病(metabolic dysfunction-associated fatty liver disease,MAFLD)肝纤维化发生的相关危险因素,并构建预测肝纤维化的列线图。方法 回顾性收集MAFLD患者的人口统计学资料及血生化指标等。根据FibroScan检测肝脏组织硬度值(liver stiffness measure,LSM), 判断患者是否存在肝纤维化。利用logistic回归分析MAFLD肝纤维化的潜在预测指标,并构建列线图模型;通过C指数评估模型的区分能力,通过校准图分析模型的一致性,并通过决策曲线评价模型的临床实用性。结果 共纳入MAFLD患者282例,其中97例存在肝纤维化(LSM≥8 kPa),185例为非肝纤维化(LSM<8 kPa)。单因素分析显示,血小板、血小板/白细胞比值(platelet to white blood cell ratio,PWR)、碱性磷酸酶、丙氨酸氨基转移酶、天冬氨酸氨基转移酶(aspartate aminotransferase,AST)、γ-氨酸转肽酶、血尿酸(serum uric acid,UA)、同型半胱氨酸(homocysteine,HCY)是发生肝纤维化的影响因素(P<0.05)。多因素logistic回归分析显示PWR(OR=0.921, 95%CI:0.877~0.967,P<0.001)是肝纤维化的保护因素,HCY(OR=1.101, 95%CI:1.032~1.175,P=0.003)、AST(OR=1.021, 95%CI:1.011~1.031,P<0.001)、UA(OR=1.032, 95%CI:1.001~1.066,P=0.048)是肝纤维化的独立危险因素;基于上述因素构建的列线图模型具有良好的区分度(AUC=0.835,95%CI:0.779~0.891),校准曲线提示肝纤维化的预测概率与观测概率一致性较高,临床决策曲线确定列线图在阈值概率内有较高的净获益。结论 基于PWR、HCY、AST、UA构建的列线图模型可较好地预测MAFLD肝纤维化相关风险,从而进行早期识别和干预,具有较高的临床应用价值。

关键词: 非酒精性脂肪性肝病, 代谢相关脂肪性肝病, 肝纤维化, 列线图, 危险因素

Abstract:

Objective To explore the related risk factors for the occurrence of liver fibrosis in metabolic dysfunction-associated fatty liver disease (MAFLD), and to construct a nomogram to predict liver fibrosis. Methods Demographic data and blood biochemical indicators of MAFLD patients were retrospectively collected. Liver fibrosis was determined by the liver stiffness measure (LSM) detected via FibroScan. Logistic regression analysis was conducted to identify potential predictive indicators for liver fibrosis in MAFLD, and a nomogram was then created. The discrimination ability, consistency, and clinical applicability of the nomogram were assessed through C-index, calibration plots, and decision curves, respectively. Results A total of 282 MAFLD patients were included. Among them, 97 patients had liver fibrosis (LSM≥8 kPa), and 185 patients did not have liver fibrosis (LSM<8 kPa). Univariate analysis indicated that platelet count (PLT), platelet-to-white blood cell ratio (PWR), alkaline phosphatase (ALP), alanine aminotransferase (ALT), aspartate aminotransferase (AST), γ-glutamyl transpeptidase (GGT), serum uric acid (UA), and homocysteine (HCY) were influencing factors for liver fibrosis in MAFLD patients (P<0.05). The multivariate Logistic regression analysis revealed that the PWR (OR=0.921, 95%CI: 0.877-0.967, P<0.001) was a protective factor for liver fibrosis, HCY(OR=1.101, 95%CI: 1.032-1.175, P=0.003), AST(OR=1.021, 95%CI: 1.011-1.031, P<0.001), and UA(OR=1.032, 95%CI: 1.001-1.066, P=0.048) were independent risk factors for liver fibrosis in MAFLD patients. The nomogram constructed based on these factors had good discrimination (area under the curve [AUC]=0.835, 95%CI: 0.779-0.891), and the calibration curve indicated that the predicted probability of liver fibrosis was highly consistent with the observed probability. The clinical decision curve determined that the nomogram had a high net benefit within the threshold probability range. Conclusion The nomogram constructed based on PWR, HCY, AST and UA can well predict the risk of liver fibrosis in MAFLD patients, thereby enabling early identification and intervention. It has high clinical application value.

Key words: non-alcoholic fatty liver disease, metabolic dysfunction-associated fatty liver disease, liver fibrosis, nomogram, risk factors

中图分类号: