预测系统性红斑狼疮患者的焦虑与抑郁:炎症、社会人口学变量与临床因素的作用
本研究为横断面研究,纳入238例系统性红斑狼疮(SLE)患者,采用随机森林分析血清神经炎症相关分析物、社会人口学与临床变量与抑郁(贝克抑郁量表-II)及焦虑(贝克焦虑量表)症状的关系。结果显示,抑郁主要受SLE病程影响,其次为MMP-9浓度;焦虑主要受NGAL和IFN-γ影响;多变量模型预测抑郁、焦虑及两者兼有的AUC分别为0.85、0.77和0.82。
要点
- 抑郁症状主要受SLE病程影响,MMP-9浓度影响较小。
- 焦虑症状主要受NGAL和IFN-γ浓度影响,其他变量影响较小。
- 多变量模型预测抑郁AUC为0.85、焦虑为0.77、两者兼有为0.82。
核心结论
- 作者认为,这项探索性研究的发现提供了预测SLE患者抑郁和焦虑症状的潜在模型,为进一步研究构建解释SLE中这些以患者为中心的重要结局机制的概念模型奠定了基础。
研究设计与样本
- 研究为横断面研究,分析对象为一项SLE认知纵向研究入组参与者的基线数据(N=238)。
主要发现
- 抑郁症状受SLE病程影响最大,受MMP-9浓度影响较小。
- 焦虑症状受NGAL和IFN-γ影响最大,其他变量影响较小。
- 多变量模型预测抑郁的AUC为0.85,预测焦虑的AUC为0.77,预测抑郁与焦虑兼有的AUC为0.82。
关键数值与推荐
- 抑郁采用贝克抑郁量表-II测量,焦虑采用贝克焦虑量表测量。
- 临床显著的抑郁和焦虑症状采用已在SLE中验证的界值定义。
- 分析采用随机森林,并用混淆矩阵生成模型性能统计量。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
原文
OBJECTIVES: Characterizing the contribution of specific pro-inflammatory mediators (analytes) to depression and anxiety in systemic lupus erythematosus (SLE) is a crucial step in illuminating the mechanisms of these disabling symptoms. The aims of this study were to investigate i) the relationship between depression and anxiety symptoms with relevant clinical and sociodemographic variables and neuroinflammation-associated analytes in a cohort of SLE patients, and ii) the ability of models including these sociodemographic, clinical and biological variables to discriminate between SLE patients with and without clinically significant depression or anxiety. METHODS: This is a cross-sectional study of baseline data from participants enrolled in a longitudinal study of cognition in SLE (N = 238). We examined the relationship between serum concentrations of a group of analytes associated with neuroinflammation, relevant sociodemographic and clinical variables and i) depression (measured with the Beck Depression Inventory-II) and anxiety (measured with the Beck Anxiety Inventory) scores, as well as between clinically significant depression and anxiety symptoms (defined using cut-off scores validated in SLE) via random forests. We generated model performance statistics from these models using confusion matrices. RESULTS: Depression is most influenced by duration of SLE and, to a lesser extent, concentrations of the analytes MMP-9. Anxiety is most influenced by the analytes NGAL and IFN- Ɣ, with other variables being less influential. The multivariate models exhibited AUCs of 0.85 (depression), 0.77 (anxiety) and 0.82 (both depression and anxiety). CONCLUSIONS: The findings of this exploratory study provide potential models predicting depression and anxiety symptoms in SLE. This produces a basis for further research building a conceptual model explaining the mechanisms of these important patient-centered outcomes in SLE.
引用
- Sigel M, Muñoz-Grajales C, Barraclough ML, et al. Predicting anxiety and depression in systemic lupus erythematosus: the role of inflammation, sociodemographic variables and clinical factors. Semin Arthritis Rheum. 2025. PMID: 40286553. doi:10.1016/j.semarthrit.2025.152718
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。