BackgroundCholestatic hepatitis is recognized as a significant contributor to the development of liver fibrosis and cirrhosis. As a well-known classic formula for the treatment of cholestatic hepatitis, Yinchenhao decoction (YCHD) is widely used in countries in Asia, including China, Japan, and Korea. However, in recent years, a risk of liver injury has been reported from Rheum palmatum L. and Gardenia jasmonoides J.Ellis which are the main ingredients of YCHD. Therefore, the question arises whether YCHD is still safe enough for the treatment of cholestatic hepatitis or whether an optimized ratio of ingredients should be applied. These is inevitable questions for the clinical application of YCHD. PurposeTo provide a scientific basis for the clinical application of YCHD through a combination of meta-analysis and network pharmacology and to find the best ratio of components to ensure optimal therapeutic efficacy and safety. At the same time, a deeper understanding of the mechanisms of YCHD was explored. MethodsWe retrieved relevant trials from various databases including PubMed, Web of Science, EMBASE, Cochrane Library, China National Knowledge Infrastructure (CNKI), VIP and Wanfang databases up to August 2023. After screening for inclusion and exclusion criteria, we assessed efficiency, ALT, AST, and TBIL as outcome parameters. The relevant data underwent a network meta-analysis using STATA 16.0 software. Based on network pharmacology, we screened the disease targets, active ingredients, and targets related to YCHD. The targets were visualized using Cytoscape 3.9.1. Then, potential mechanisms were explored based on bioinformatic techniques. ResultsTwenty eligible studies were finally screened and a total of 1,591 patients who fulfilled the inclusion criteria were enrolled in the study. The meta-analysis results indicated that TG-c (treatment group c) [(Artemisia capillaris Thunb. : Gardenia jasminoides J.Ellis : Rheum palmatum L. = 10:5:2–10:5:3) + CT] was the most promising therapeutic approach, demonstrating superior efficacy and notable improvements in both AST and TBIL levels. For ALT, TG-d [(Artemisia capillaris : Gardenia jasminoides : Rheum palmatum = 5:1:1–5:2:1) + CT] exhibited the greatest potential as optimal therapy option. Based on the surface under the cumulative ranking curve (SUCRA) values, TG-c was the best therapy in terms of efficiency and improvement in TBIL levels, while TG-d was the most effective in reducing ALT levels. For AST levels, TG-e [(Artemisia capillaris : Gardenia jasminoides : Rheum palmatum = 5:2:2–5:3:3) + CT] was the most effective therapy. The comprehensive analysis revealed that TG-c exhibited the most pronounced efficacy. Combined network pharmacology, GO enrichment analysis and KEGG pathway enrichment analysis displayed that the key target genes of Artemisia capillaris, Rheum palmatum, and Gardenia jasminoides were closely involved in inflammation response, bile transport, apoptosis, oxidative stress, and regulation of leukocyte migration. Notably, bile secretion dominated the common pathway of the three herbs. On the other hand, Artemisia capillaris exhibited a unique mode of action by regulating the IL-17 signaling pathway, which may play a crucial role in its effectiveness. ConclusionBased on our findings, the optimal TG-C demonstrated the most favorable overall therapeutic efficacy by increasing the dosage of Artemisia capillaris while reducing the dosage of Gardenia jasminoides and Rheum palmatum. This is attributed to the potent ability of Artemisia capillaris. to effectively modulate the IL-17 signaling pathway, thereby exerting a beneficial therapeutic effect. Conversely, Gardenia jasminoides and Rheum palmatum may potentially enhance the activation of the NF-кB signaling pathway, thereby elevating the risk of hepatotoxicity.