To investigate the differential expression of microRNAs (miRNAs) in the peripheral blood of children with refractory epilepsy. Methods: A total of 46 pediatric patients with epilepsy were enrolled and divided into a refractory epilepsy group (R group, n=20) and a well-controlled epilepsy group (C group, n= 26). Additionally, 26 healthy children undergoing routine health examinations in the pediatric healthcare department during the same period were included as the healthy control group (N group). High-throughput sequencing technology was used to detect the expression of miRNAs in the peripheral whole blood of children in the three groups, and DEseq2 software was employed to screen for differentially expressed miRNAs. Under the same conditions, an additional 30 children were enrolled in each of the R, C, and N groups, and peripheral blood samples were collected. Fluorescent quantitative polymerase chain reaction (qRT-PCR) amplification technology was used to validate the expression levels of target miRNAs in the peripheral blood of children. The miRanda and RNAhybrid online databases were utilized to predict the target genes of the differentially expressed miRNAs, and Gene Ontology (GO) analysis and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis were conducted for the candidate target genes. Results: Compared with the N group, a total of 124 differentially expressed miRNAs were identified in the R group, including 74 upregulated and 50 downregulated miRNAs, while 66 differentially expressed miRNAs were found in the C group, with 21 upregulated and 45 downregulated. When compared with the C group, the R group exhibited 337 differentially expressed miRNAs, of which 197 were upregulated and 140 were downregulated. Among these, miR-15a-5p showed the most significant difference (P=9.00×10－8 , |log2(fold change)|=2.93). qRT-PCR amplification of miR-15a-5p revealed a statistically significant difference in its expression among the three groups of children (P<0.001), consistent with the high-throughput sequencing results. The receiver operating characteristic (ROC) curve demonstrated that miR-15a-5p had a good area under the curve (AUC) value (0.884), sensitivity (90% ), and specificity (83.3% ) in diagnosing refractory epilepsy in children. Target gene prediction for the differentially expressed miRNAs, along with GO and KEGG analysis of the candidate target genes, suggested that miR-15a-5p may be involved in regulating signaling pathways related to refractory epilepsy (RE) in children, including the MAPK signaling pathway, cAMP signaling pathway, platinum drug resistance signaling pathway, p53 signaling pathway, Ras signaling pathway, Fc-γ-R-mediated phagocytosis signaling pathway, and neurodegenerative disease signaling pathway. Conclusion: The expression profile of miRNAs in the peripheral blood of children with refractory epilepsy differs significantly from that of children with well-controlled epilepsy and healthy children. Among these, miR-15a-5p exhibits the most prominent differential expression and may participate in the pathophysiological process of epilepsy by regulating multiple pathways.