Metabolic disorders (MDs) are a group of medical conditions that impact the metabolism. These complex processes may have common characteristics among various diseases, thereby suggesting the potential of drug repurposing. Employing Mendelian Randomization (MR), we constructed a causal network between 2,478 targetable drug-gene expressions (eQTLs) and 30 broadly reported metabolic disorders. Our study identified 499 drug target genes significantly associated with 27 metabolic disorders (|MR coefficient| > 0.2). Pathway enrichment analysis of drug target genes indicates that regulation of response stimulus may serve as a common pathway across 14 diseases. Based on 53 commonly used clinical drugs for 18 diseases, we elucidated novel therapeutic mechanisms of some drugs, such as the potential of Valproic Acid to treat Schizophrenia by affecting key genes in Alcoholism, SLC29A1, and HDAC4. Furthermore, we identified potential for drug repurposing in four diseases, with Manic Episode and Type 1 Diabetes sharing four novel drugs: Cannabidiol, Doxorubicin, Genistein, and Propylthiouracil. Additionally, we predicted 189 potential therapeutic drugs affecting the causal genes of diseases. Overall, we established a causal network between metabolic disorders and drug target genes, explored possible pathways for drug action on disease treatment, and proposed drug repurposing strategies for four diseases.