The diagnostic value of PLS-DA model based on seven serum cytokines for children patients with mycoplasma pneumonia / 实用医学杂志

ABSTRACT

Objective To investigate the diagnostic value of the partial least square discriminant analysis (PLS-DA) model based on seven serum cytokines for children patients with mycoplasma pneumonia,the 7 cytokines includ interleukin-2 (IL-2),IL-4,IL-6,IL-8,IL-10,tumor necrosis factor alpha (TNF-α) and interferonγ(IFN-γ).Methods Serum levels of cytokines were measured by the double antibody sandwich method of ELI-SA in 140 patients with pneumoniae infection and 135 normal healthy controls,and data was analyzed with the receiver operation characteristic (ROC) curve and the PLS-DA.Results The area under the ROC curve (AUC) of serum IL-10 and IFN-γfor the diagnosis of MPP was 0.84 (95％ CL0.79 ～ 0.89).The sensitivity of IL-10 and IFN-γ for the diagnosis of MPP was 91.4％ and 82.1％,with the specificity was 77.0％ and 82.2％,respectively.The diagnostic accuracy of the PLS-DA model based on seven serum cytokines for children MPP and the controls was 90.0％ and 88.15％,and the prediction accuracy was 86.4％ and 87.4％,respectively.Conclusion Seven serum cytokines based on PLS-DA model was helpful for the diagnosis of children patients with MPP.