Quantification of Immune Dysregulation by Next-generation Polymerase Chain Reaction to Improve Sepsis Diagnosis in Surgical Patients

    loading  Checking for direct PDF access through Ovid


Objectives:To quantify immunological dysfunction in surgical patients with presence/absence of sepsis using a droplet digital polymerase chain reaction (ddPCR) transcriptomic analysis. The study also aims to evaluate this approach for improving identification of sepsis in these patients.Background:Immune dysregulation is a central event in sepsis. Quantification of the expression of immunological genes participating in the pathogenesis of sepsis could represent a new avenue to improve its diagnosis.Methods:Expression of 6 neutrophil protease genes (MMP8, OLFM4, LCN2/NGAL, LTF, PRTN3, MPO) and also of 5 genes involved in the immunological synapse (HLA-DRA, CD40LG, CD3E, CD28, ICOS) was quantified in blood from 101 surgical patients with sepsis, 53 uninfected surgical patients, and 16 blood donors by using ddPCR. Areas under receiver operating characteristic curves (AUROC) and multivariate regression analysis were employed to test individual genes and gene ratios to identify sepsis, in comparison with procalcitonin.Results:Sepsis-induced overexpression of neutrophil protease genes and depressed expression of immunological synapse genes. MMP8/HLA-DRA, LCN2/HLA-DRA outperformed procalcitonin in differentiating between patients with sepsis and surgical controls in the AUROC analysis: LCN2/HLA-DRA: 0.90 (0.85–0.96), MMP8/HLA-DRA: 0.89 (0.84–0.95), procalcitonin: 0.80 (0.73–0.88) (AUROC, confidence interval 95%), and also in the multivariate analysis: LCN2/HLA-DRA: 8.57 (2.25–32.62); MMP8/HLA-DRA: 8.03 (2.10–30.76), procalcitonin: 4.20 (1.15–15.43) [odds ratio (confidence interval 95%)]. Gene expression levels of HLA-DRA were an independent marker of hospital mortality.Conclusions:Quantifying the transcriptomic ratios MMP8/HLA-DRA, LCN2/HLA-DRA by ddPCR is a promising approach to improve sepsis diagnosis in surgical patients.

    loading  Loading Related Articles