BACKGROUNDEffective treatment and prevention of delirium remain challenging, largely due to the limited efficiency in identifying and validating drug targets. To address this gap, we performed a systematic Mendelian Randomization (MR) study of the druggable genome to identify potential therapeutic targets for delirium.METHODSTo identify therapeutic targets for delirium, we conducted MR and colocalization analyses using cis-eQTL data (31,684 blood samples and 1387 brain samples) and genome-wide association study (GWAS) data for delirium (discovery cohort N = 431,880; validation cohort N = 403,037). This approach enabled us to identify annotated druggable gene targets with significant associations with delirium. Additional analyses included phenome-wide association studies (PheWAS), enrichment analyses, protein network construction, drug prediction, molecular docking, and mediation analyses to assess the therapeutic and clinical relevance of these targets.RESULTSFive genes were identified as significant therapeutic targets for delirium: C4BPA, A2M, GRIK4, and C1R from blood samples, and SUMF1 from both blood and brain tissues. These genes are biologically significant for their strong association with immune function. At the gene level, PheWAS identified associations between C4BPA and other traits. Molecular docking demonstrated strong binding between drugs and proteins with available structural data. Mediation analysis revealed that C4BPA explained 16.7 % of the aging-related risk for delirium.CONCLUSIONOur MR analysis identifies 5 potential druggable targets for the treatment of delirium. Drugs targeting these genes are more likely to succeed in clinical trials and are expected to prioritize the development of delirium treatments while reducing drug development costs.