Abstract
BACKGROUND: Sepsis is a prevalent and severe condition. However, research investigating the relationship between the immune microenvironment in sepsis-associated acute kidney injury (SA-AKI) through diagnostic models using RNA biomarkers remains limited. Therefore, this study developed a diagnostic model using gene expression data from the Gene Expression Omnibus (GEO) database, leveraging a sufficient sample size. METHODS: We proposed a computational method to identify RNAs Rela and Stat3 constructing a diagnostic model using Least Absolute Shrinkage and Selection Operator regression algorithms. Gene expression data from the GEO, comprising five samples each of SA-AKI and sepsis, were analyzed. RESULTS: Diagnostic models were developed for the datasets, followed by immune cell infiltration and correlation analyses. Experiments were conducted to test and confirm the high expression of Stat3 via Rela in AKI cells post-sepsis, leading to a worse prognosis. CONCLUSION: This study identified the significant roles of RNAs Rela and Stat3 in SA-AKI. The developed diagnostic model demonstrated improved accuracy in identifying SA-AKI, suggesting that these RNA markers may provide valuable insights into the pathophysiology of SA-AKI and enhance early diagnosis. These findings contribute to a better understanding of immune-related mechanisms underlying SA-AKI and may inform future therapeutic strategies.