Objectives: Acute kidney injury (AKI) after percutaneous coronary intervention (PCI) is a common and serious complication of the procedure and is associated with adverse outcomes. Early detection and management of AKI is crucial to minimize its impact and to improve prognosis. This study is to investigate serum metabolites associated with AKI after PCI Methods: We performed proton nuclear magnetic resonance-based metabolomics analyses of serum from patients undergoing PCI to identify metabolic profiles underling AKI. Total 31 samples from non-AKI group and 18 samples for AKI group were analyzed. Results: Multivariate analysis identified four metabolites that allowed the diagnosis of AKI with an area under the receiver operating characteristic (ROC) curve of 0.8. Among those, histidine (P<0.01) and threonine (P=0.046) were significantly associated with AKI in multivariate logistic regression. In multivariate ROC analysis, adding histidine and threonine to clinical variables showed the higher discriminatory ability to predict AKI (AUC 0.929 vs 0.845, P =0.063). Conclusions: The present study identified histidine and threonine as potential candidate biomarkers of AKI after PCI.