Serum small RNA profiling identifies prognostic biomarkers for sepsis mortality prediction.

Background: Sepsis remains a leading cause of mortality in critically ill patients, with current prognostic tools showing limited accuracy for outcome prediction. While traditional clinical parameters and inflammatory biomarkers provide some prognostic information, there is an urgent need for novel molecular biomarkers that can accurately predict sepsis outcomes to guide clinical decision-making and therapeutic interventions. Circulating small RNAs, including tRNA-derived small RNAs (tsRNAs) and microRNAs (miRNAs), have emerged as potential biomarkers due to their stability in circulation and regulatory roles in immune responses and inflammatory processes. Methods: This study enrolled 26 sepsis patients admitted to the intensive care unit, who were stratified into recovery (n = 17) and death (n = 9) groups based on clinical outcomes. Comprehensive clinical parameters including demographic characteristics, severity scores, inflammatory markers, organ function indicators, metabolic parameters, and acid-base balance were analyzed. Serum samples underwent optimized small RNA profiling using high-throughput sequencing with de-modification protocols to enhance tsRNA and miRNA detection. Differential expression analysis was performed to identify outcome-associated small RNAs, and receiver operating characteristic curve analysis was conducted to evaluate diagnostic performance of individual biomarkers and combined panels. Results: Traditional clinical parameters showed limited prognostic value, with only specific markers including SOFA scores, procalcitonin, interferon-γ, glucose levels, and acid-base parameters demonstrating significant associations with outcomes. Small RNA profiling revealed 22 differentially expressed tsRNAs (12 downregulated, 10 upregulated) and 5 differentially expressed miRNAs (3 downregulated, 2 upregulated) in the death group compared to the recovery group. Individual biomarkers showed substantial discriminatory power, with top-performing tsRNAs achieving AUCs of 0.827-0.837 and miRNAs reaching AUCs of 0.797-0.850. Notably, combined biomarker panels demonstrated exceptional diagnostic performance, with the tsRNA signature achieving an AUC of 0.967 and the miRNA panel reaching an AUC of 0.902. Conclusion: This study identifies circulating small RNAs as highly promising novel biomarkers for sepsis outcome prediction, substantially outperforming traditional clinical parameters. The exceptional diagnostic accuracy of combined tsRNA and miRNA signatures suggests significant potential for clinical translation to improve sepsis prognosis and patient stratification. These findings provide a foundation for developing molecular-based prognostic tools that could enhance sepsis management and guide therapeutic decision-making in critically ill patients.