Wang Qian-Qian, Zhang Ning, Xu Xiang, Lv Si-Ang, Huang Zhuo-Deng, Long Xi-Dai, Wu Jun
Department of Laboratory Medicine, Jiading Branch of Shanghai General Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, 201803, China.
Department of Pathology, the Affiliated Hospital of Youjiang Medical University for Nationalities, Baise, 533000, China.
BMC Gastroenterol. 2025 Apr 4;25(1):220. doi: 10.1186/s12876-025-03786-2.
BACKGROUND/AIMS: The incidence of metabolic dysfunction-associated steatotic liver disease (MASLD) among individuals with hyperuricemia is significantly high. The aim of this study was to identify effective biomarkers for the detection of MASLD among patients with hyperuricemia.
We conducted an analysis involving 3424 participants with hyperuricemia from the National Health and Nutrition Examination Survey (1999-2020). To identify potential significant variables, we employed Boruta's algorithm, SHapley Additive exPlanations (SHAP) and random forests. Multivariable logistic regression models were utilized to assess the odds ratio (OR) of developing MASLD. To evaluate the accuracy and clinical value of our prediction model, we employed receiver operating characteristic (ROC) curves and decision curve analysis (DCA) curves.
Among the study population of 3424 participants (mean [SD] age, 54 [20] years, 1788 [52.22%] males) with hyperuricemia, 1670 participants had MASLD. Using Boruta's algorithm, SHAP and random forests, our analysis suggested that Triglyceride Glucose-Waist Circumference (TyG_WC) was one of the most significant variables in predicting MASLD risk, with an area under the receiver operating characteristic (AUROC) of 0.865. The restricted curve spline (RCS) revealed a positive association between the odds ratio of TyG_WC and MASLD, when compared with lowest quantile of TyG_WC, the risk of MASLD for highest quantile was 137.96 times higher. The predictive strategy incorporating TyG_WC notably enhanced the clinical model, with threshold probabilities spanning from approximately 0% to 100%, resulting in a significant improvement of the net benefit.
Our analysis found that TyG_WC was one of the most significant variables in predicting MASLD risk among individuals with hyperuricemia.
背景/目的：高尿酸血症患者中代谢功能障碍相关脂肪性肝病（MASLD）的发病率显著较高。本研究的目的是确定用于检测高尿酸血症患者中MASLD的有效生物标志物。
我们对来自国家健康与营养检查调查（1999 - 2020年）的3424名高尿酸血症参与者进行了分析。为了确定潜在的显著变量，我们采用了博鲁塔算法、夏普利加性解释（SHAP）和随机森林。多变量逻辑回归模型用于评估发生MASLD的优势比（OR）。为了评估我们预测模型的准确性和临床价值，我们采用了受试者工作特征（ROC）曲线和决策曲线分析（DCA）曲线。
在3424名高尿酸血症参与者（平均[标准差]年龄，54[20]岁，1788[52.22%]为男性）的研究人群中，1670名参与者患有MASLD。通过博鲁塔算法、SHAP和随机森林分析，我们发现甘油三酯血糖 - 腰围（TyG_WC）是预测MASLD风险的最重要变量之一，受试者工作特征曲线下面积（AUROC）为0.865。受限曲线样条（RCS）显示，与TyG_WC最低分位数相比，TyG_WC的优势比与MASLD呈正相关，TyG_WC最高分位数发生MASLD的风险高137.96倍。纳入TyG_WC的预测策略显著增强了临床模型，阈值概率范围从约0%到100%，从而显著提高了净效益。
我们的分析发现，TyG_WC是预测高尿酸血症个体中MASLD风险的最重要变量之一。
