基于实验室指标构建银屑病联合诊断模型在临床应用的研究

A study on clinical application of constructing a combined diagnosis model for psoriasis based on laboratory indicators

目的探讨基于实验室指标构建联合诊断模型对银屑病诊断的临床应用。方法2021年6月—2022年6月在本院门诊和住院确诊银屑病患者52例以及体检健康者30例,检测WBC、NC、LC、PLT、ALT、AST、ALB、TBI、IgA、IgM、IgG、C3、C4、SCCA等实验室指标,并计算NLR、PLR比值。单因素logistic回归筛选并用多因素logistic回归拟合得到联合诊断模型,生成受试者工作曲线(ROC),并将所有受检者数据代入,评价模型诊断效果。结果ALB、SCCA、NLR纳入拟合得到新的联合诊断模型,ROC曲线显示该模型的AUC域下面积高于其他单项指标。联合诊断模型的敏感度(82.7%)、特异度(93.3%)、阳性预测值(95.6%)、阴性预测值(75.7%)、诊断符合率(86.6%)均优于ALB、SCCA、NLR单独使用。结论NLR、白蛋白ALB、SCC构建的联合诊断模型在银屑病实验室诊断中具有更好地临床应用价值。

Objective This paper aims to explore the clinical application of constructing a combined diagnostic model based on laboratory indicators for the diagnosis of psoriasis.Methods From June 2021 to May 2022,52 outpatients and inpatients with psoriasis and 30 healthy people were enrolled in this study.WBC,NC,LC,PLT,ALT,AST,ALB,TBI,IgA,IgM,IgG,C3,C4,SCCA and other laboratory indicators were detected,and the NLR/PLR ratio was calculated.Univariate logistic regression screening and multivariate logistic regression fitting were used to construct the combined diagnostic model,generate the subject working curve(ROC),and all the subject data was substituted to evaluate the diagnostic effect of the model.Results ALB,SCCA and NLR were included in the fitting to get a new joint diagnostic model.The ROC curve showed that the area under AUC domain of this model was higher than that of other single indicators.The sensitivity(82.7%),specificity(93.3%),positive predictive value(95.6%),negative predictive value(75.7%)and diagnostic coincidence rate(86.6%)of the combined diagnostic model were observed,higher than that of single ALB,SCCA,and NLR.Conclusion The combined diagnostic model constructed by NLR,ALB and SCC has better clinical value in the laboratory diagnosis of psoriasis.