Abstract
BACKGROUND: Sepsis is a life-threatening condition that induce tens of million death each year, yet early diagnosis remains a formidable challenge. Many studies have focused on the diagnostic accuracy of microRNAs (miRNAs) for sepsis in recent years, particularly miR-155-5p, miR-21, miR-223-3p, miR-146a, and miR-125a. Thus, we conducted this meta-analysis to explore if miRNAs may be used as a biomarker for sepsis detection. METHODS: We searched PubMed, the Cochrane Central Register of Controlled Trials, EMBASE, and China National Knowledge Infrastructure through May 12, 2022. This meta-analysis was conducted using Meta-disc 1.4 and STATA 15.1 in a fixed/random-effect model. RESULTS: The analysis included a total of 50 relevant studies. The overall performance of total miRNAs detection was: pooled sensitivity, 0.76 (95% confidence interval [CI], 0.75 to 0.77); pooled specificity, 0.77 (95%CI, 0.75 to 0.78); and area under the summary receiver operating characteristic curves value (SROC), 0.86. The subgroup analysis suggested that detection in miR-155-5p group had the highest area under the curve (AUC) of SROC among all miRNAs: pooled sensitivity, 0.71 (95%CI, 0.67 to 0.75); pooled specificity, 0.82 (95%CI, 0.76 to 0.86); and SROC, 0.85. MiR-21, miR-223-3p, miR-146a, and miR-125a had SROC values of 0.67, 0.78, 0.69, and 0.74, respectively. The specimen type was found to be a source of heterogeneity in the meta-regression study. The SROC of serum was higher than that of plasma (0.87 and 0.83, respectively). CONCLUSIONS: Our meta-analysis revealed that miRNAs, specifically miR-155-5p, could be useful biomarkers for detecting sepsis. A clinical serum specimen is also indicated for diagnostic purposes.