Data driven analysis reveals shared transcriptome response and immune cell composition across aetiologies of critical illness.

bioRxiv(2019)

引用 0|浏览20
暂无评分
摘要
Sepsis and trauma are frequent and challenging health problems in critical care. The diversity of patient response to these conditions complicates both disease management and outcome prediction. Whole blood transcriptomics allows the analysis of response in the critically ill at a molecular level. Prior results in this field demonstrate robust and diverse genomic response in the acute phase and others have shown shared biological mechanisms across wide disease aetiologies. We hypothesize that specific biological mechanisms, particularly those related to immune processes, are shared between sepsis and trauma cohorts. These may serve as universal markers for patients vulnerable to a complicated clinical course and/or mortality. We present a systems level analysis of gene expression for a total of 317 patients with abdominal sepsis (51), pulmonary sepsis (108) or trauma (158) and compare them to healthy controls (68). Our results confirm that immune processes are shared across disease aetiologies in critical illnesses. We identify two consistent and distinct subgroups of critical illness: 1) increased dendritic cell and CD4 T helper fractions but suppressed neutrophils and 2) high neutrophils and otherwise suppressed leukocyte fractions. These subgroups validate in an independent cohort of 181 paediatric patients suffering from septic shock of diverse aetiologies. Furthermore, we found immune and inflammatory processes derived from gene co-expression networks were downregulated in subgroup 1. This paralleled prior results which find similar leukocyte configuration by deconvolving whole blood transcriptomics, and this leukocyte configuration associates with greater susceptibility to multi organ failure. We are the first to identify a patient subgroup with a preserved leukocyte configuration across aetiologies of critical illness, which may serve as a universal predictor of complicated clinical course/poor outcome.
更多
查看译文
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要