Objective: To stratify systemic lupus erythematosus (SLE) patients clinically, to analyze the clinical characteristics of patients with and without disease activity, and to explore the application va-lue of key clinical indicators in assessing disease activity, as well as to construct an evaluation model.
Methods: A retrospective analysis was conducted on clinical data of the SLE patients diagnosed at Peking University People' s Hospital from May 1995 to April 2014. Demographic information, clinical manifestations, laboratory tests, and antibody detection results were collected. The patients were divided into active and inactive groups based on systemic lupus erythematosus disease activity index 2000(SLEDAI-2000)scores. t-tests, Mann-Whitney U tests, and χ2 tests were used to compare the differences between the groups. Spearman correlation analysis was used to evaluate the relevant clinical indicators associated with SLE activity in the active disease group. Based on the results of statistical analysis, a Logistic regression model was constructed, and the performance of the model was evaluated.
Results: No significant differences were found in demographic characteristics between the two groups. In the active disease group, positive rates of antinuclear antibodies (ANA) and anti-double-stranded DNA antibodies (anti-dsDNA) were increased; white blood cell count (WBC), red blood cell count (RBC), hemoglobin (HGB), lymphocytes (LY), total protein (TP), albumin (ALB), and complement 3(C3) levels were significantly decreased; while immunoglobulin A and G levels were markedly elevated. The correlation analysis results showed that hemoglobin, albumin, C3, and complement 4(C4) had higher correlation indices compared with other clinical indicators. Among these, C3 exhibited a certain negative correlation with disease activity. The Logistic regression model based on 12 significantly different indicators (P < 0.05) achieved an accuracy of 91.4%, sensitivity of 94.4%, specificity of 81.0%, and the area under curve (AUC) of the receiver operating characteristic (ROC) was 0.944.
Conclusion: This study comprehensively evaluated a range of clinical indicators related to SLE disease activity, providing a thorough understanding of both laboratory and clinical markers. The Logistic regression model, which was primarily constructed using laboratory test indicators, such as inflammatory markers, immune response parameters, and organ involvement metrics, demonstrated a high degree of accuracy in assessing the disease activity in SLE patients. Consequently, this model might provide a new basis for the diagnosis and treatment of SLE patients, offering significant clinical diagnostic value.
目的: 对系统性红斑狼疮(systemic lupus erythematosus, SLE)患者进行临床分层，分析不同活动度患者的临床特征，并探讨关键临床指标在SLE疾病活动度评估中的应用价值及评估模型的构建。
方法: 回顾性分析1995年5月至2014年4月北京大学人民医院确诊的SLE患者临床资料。收集患者的人口学信息、临床表现、实验室检查结果，根据SLE疾病活动度指数2000(systemic lupus erythematosus disease activity index 2000, SLEDAI-2000)将患者分为疾病活动组与无疾病活动组。采用t检验、Mann-Whitney U检验以及χ2检验比较两组间差异，并使用Spearman相关性分析评估疾病活动组中与SLE活动度相关的临床指标。基于统计学分析结果，构建了逻辑回归模型，并对模型性能进行评估。
结果: 两组在基本人口学特征上差异无统计学意义。疾病活动组中，抗核抗体(antinuclear antibodies，ANA)及抗双链DNA抗体(anti-double-stranded DNA antibodies，anti-dsDNA)阳性率升高；白细胞计数(white blood cell, WBC)、红细胞计数(red blood cell, RBC)、血红蛋白(hemoglobin, HGB)、淋巴细胞(lymphocytes, LY)、总蛋白(total protein, TP)、白蛋白(albumin, ALB)及补体C3水平显著降低，免疫球蛋白A(immunoglobulin A，IgA)、免疫球蛋白G(immunoglobulin G, IgG)水平明显升高。相关性分析结果显示，血红蛋白、白蛋白、补体C3和补体C4与其他临床指标相比具有更高的相关性指数，其中补体C3与疾病活动度具有一定负相关性。基于12项差异具有统计学意义的指标构建的逻辑回归模型准确率为91.4%，敏感性为94.4%，特异性为81.0%，受试者工作特征曲线(receiver operating characteristic, ROC)的线下面积(area under curve, AUC)为0.944。
结论: 全面评估与SLE疾病活动度相关的临床指标，以实验室检查指标为主构建的逻辑回归模型能够较准确地评估SLE患者疾病活动度，为SLE患者的个体化治疗提供参考。
Keywords: Clinical indicators; Clinical stratification; Disease activity; Systemic lupus erythematosus.