BACKGROUND: With the change of disease treatment mode, people have realized the importance of traditional Chinese medicine in the treatment of steroidinduced necrosis of the femoral head (SANFH). Therefore, bioinformatics is used to analyze the pathogenesis of SANFH at the molecular level, build a disease risk model, and predict the potential therapeutic effects of traditional Chinese medicine, so as to provide a theoretical basis for the treatment of SANFH by traditional Chinese medicine in the future. OBJECTIVE: To mine the competing endogenous RNA regulatory network of SANFH based on bioinformatics, analyze its molecular regulatory mechanism in SANFH, predict relevant disease targets, build disease risk models, and predict Chinese herbal medicines with potential therapeutic effects. METHODS: The GEO database was searched to download the SANFH matrix file GSE123568 and gene annotation file GPL15207. The differentially expressed long non-coding RNAs and mRNAs were obtained by software analysis such as R language, and the miRNA-mRNAs associated with the differentially expressed long non-coding RNAs were predicted through the public database. Then, predicted and differentially expressed mRNAs were intersected and integrated to obtain the competing endogenous RNA network. STRING database and Cytoscape software were used to screen key genes and R language was used to analyze the functions and related pathways of key genes and mine the key competing endogenous RNA network. Finally, the risk model of SANFH was constructed according to the key genes and the prediction of traditional Chinese medicine was carried out. RESULTS AND CONCLUSION: Compared with healthy controls, a total of 7 long non-coding RNAs and 1763 mRNAs were differentially expressed in SANFH patients. Six key genes including STAT3, KAT2B, AGO4, JAK2, JAK1, and PTGS2 were identified. The enriched functions of key genes include biological processes such as response to peptide hormones, interleukin-6-mediated signaling pathways, and cell responses to interleukin-6, and are involved in signaling pathways such as JAK-STAT, adipocytokines, and prolactin. Four miRNAs (miR- 135a-5p, miR-137, miR-17-5p, miR-20b-5p) and two long non-coding RNAs (SNHG11, C20orf197) may play a key role in the occurrence and development of SANFH. KAT2B is most likely to be a risk factor for SANFH. Turmeric, Epimedium, and Astragalus have the potential to treat SANFH. Through the analysis of the competing endogenous RNA network mediated by SANFH-related long non-coding RNAs, potential disease targets, signaling pathways, and potential therapeutic traditional Chinese medicines can be identified, providing a reference for further clarifying its pathogenesis in subsequent experimental research. [ABSTRACT FROM AUTHOR]