中国骨质疏松杂志

铁死亡血清学标志物在绝经后骨质疏松症发病中的作用及预测模型构建

Role of ferroptosis-related serological biomarkers in the pathogenesis of postmenopausal osteoporosis and construction of a diagnostic nomogram

DOI：10.3969/j.issn.1006-7108.2023.05.001

中文关键词: 绝经后骨质疏松症谷胱甘肽过氧化酶4 谷胱甘肽丙二醛随机森林列线图

英文关键词:postmenopausal osteoporosis glutathione peroxidase 4 glutathione malondialdehyde random forest nomogram

基金项目:国家自然科学基金项目（81973886，82174395）；广州中医药大学“双一流”与高水平大学学科协同创新团队重点项目（2021XK21）

作者	单位
林适1 王世浩1，2 刘树华1 杨彬彬1 唐子佳1 东智卓玛1 吴建军1 林贤灿1 陈桐莹1 黄宏兴3* 万雷3*	1.广州中医药大学，广东广州 510006 2.湖州市吴兴区人民医院，浙江湖州 313008 3.广州中医药大学第三附属医院，广东广州 510378

摘要点击次数: 696

全文下载次数: 0

中文摘要:

目的探索铁死亡血清生物标志物谷胱甘肽过氧化酶4（GPX4）、谷胱甘肽（GSH）以及丙二醛（MDA）和绝经后骨质疏松症（postmenopausal osteoporosis，PMOP）发病的关系及对其的预测价值。方法以80名绝经后女性作为研究对象，按照骨质疏松症诊断标准分为骨质疏松组（60例）和非骨质疏松组（20例）。比较两组受试者一般资料及血清GPX4、GSH、MDA水平；Spearman相关分析各变量和骨密度及各变量之间的相关性；随机森林算法评估各变量对PMOP发病的重要性；ROC曲线进一步肯定各变量对PMOP的预测效能。最后，构建PMOP预测模型。结果两组受试者体质量指数、体重、身高相比无明显差异（P＞0.05）；非骨质疏松受试者腰椎骨密度、GPX4、GSH水平均比骨质疏松症组高，年龄、MDA值则比骨质疏松组低（P＜0.05）。Spearman相关分析显示，随着GPX4水平（R=0.42，P＜0.05）、GSH水平（R=0.43，P＜0.05）升高，受试者骨质疏松程度越低；随着MDA水平（R=-0.30，P＜0.05）升高，受试者骨质疏松程度升高。随着GPX4水平（R=-0.30，P＜0.05）、GSH水平（R=-0.42，P＜0.05）降低，受试者MDA水平升高；而GPX4和GSH水平呈正向趋势（R=0.43，P＜0.05）。随机森林算法分析显示年龄、GPX4、GSH、MDA水平在PMOP发病中具有重要作用。同时，ROC曲线也证实了这一结果，并且发现联合这4个指标诊断能更好地预测PMOP的发生。基于这4个变量构建的列线图能帮助识别PMOP高危人群。结论 PMOP人群存在以GPX4为核心的氧化还原能力降低，而脂质过氧化水平升高的现象，表明PMOP的发生和铁死亡存在密切关系。同时，年龄联合外周血GSH、GPX4、MDA水平可以很好地预测PMOP的发生。

英文摘要:

Objective To investigate the relationship between ferroptosis-related serum biomarkers GPX4, GSH, MDA and PMOP, and their predictive values in PMOP. Methods A total of 80 postmenopausal women were divided into osteoporosis group (n=60) and non-osteoporosis group (n=20) according to the diagnostic criteria of osteoporosis. The general data and serum levels of GPX4, GSH, and MDA were compared between the two groups. Spearman correlation analysis was used to analyze the correlation between each variable and bone mineral density. Random forest algorithm was used to evaluate the importance of each variable on the incidence of PMOP. ROC curve further confirmed the predictive efficacy of variables for PMOP. Finally, the PMOP prediction model is constructed. Results There were no significant differences in height, weight, and BMI between the two groups (P>0.05). BMD of the lumbar vertebrae, GPX4, and GSH levels in the non-osteoporosis group were higher than those in the osteoporosis group, while the age and MDA levels were lower than those in the osteoporosis group (P<0.05). Spearman correlation analysis showed that with the increase of GPX4 level (R=0.42, P<0.05) and GSH level (R=0.43, P<0.05), the degree of osteoporosis decreased. With the increase of MDA level (R=-0.30, P<0.05), the degree of osteoporosis increased. With the decrease of GPX4 level (R=-0.30, P<0.05) and GSH level (R=-0.42, P<0.05), MDA level increased. The levels of GPX4 and GSH showed a positive trend (R=0.43, P<0.05). Random forest analysis showed that age, GPX4, GSH, and MDA levels played an important role in the pathogenesis of postmenopausal osteoporosis. At the same time, ROC curve also confirmed this result, and found that the combined diagnosis of these four indicators could better predict the occurrence of postmenopausal osteoporosis. Based on these four variables, the nomogram constructed helped to identify high-risk PMOP population. Conclusion The decrease of GPX4-centered REDOX capacity and the increase of lipid peroxidation in postmenopausal osteoporosis population indicate that there is a close relationship between the occurrence of postmenopausal osteoporosis and ferroptosis. At the same time, age combined with peripheral blood GSH, GPX4, and MDA levels may well predict the occurrence of PMOP.

查看全文查看/发表评论下载PDF阅读器

关闭