Background: Acute kidney injury (AKI) is common in patients with community acquired pneumonia (CAP), and associated with poor prognosis. Cystatin C had been demonstrated to be a biomarker for the early detection AKI in in several other clinical settings. Therefore, in this study we evaluated whether AKI in patients with CAP could be well predicted by serum Cystatin C within 24 hours after admission.Methods: Univariate and multivariate logistic regression analyses were used to investigate independent factors of AKI in patients with CAP. Prediction performance of all independent factors for AKI was measured using area under the receiver operating characteristic (ROC) curves (AUCs).Results: 2716 patients with CAP were eligible in this study, and 766 (28%) patients developed AKI. After multivariate logistic regression analysis, Cystatin C, albumin, uric acid, platelet count, white blood cell count, CURB-65 score, and acute respiratory failure were independent factors for AKI in patients with CAP. The maximum AUC was reported for serum Cystatin C within 24 hours after admission. Cystatin C had an AUC of 0.81 (95% CI: 0.79–0.83, P < 0.001) for predicting AKI, with an optimal cut-off value of 1.37 mg/L, computing 68% sensitivity, 80% specificity, 57% positive predictive value and 86% negative predictive value. In the stratification analyses, Cystatin C still had a good performance for predicting AKI in all the subgroups (AUCs: 0.69-0.84).Conclusions: The level of serum Cystatin C within 24 hours after admission appears to be a good biomarker for predicting of AKI in patients with CAP.