BackgroundHyperlactatemia is a prognostic marker among patients with ST-segment elevation acute myocardial infarction (STEMI). However, the predictive value of lactate and the dynamic change associated with acute kidney injury (AKI) among patients with STEMI, remain poorly understood. We aimed to compare single lactate values at admission (Lacadm) and 12 h after admission (Lac12h) with lactate clearance (LC) 12 h after admission for AKI prediction in patients with STEMI.MethodsA total of 1,784 patients with STEMI were included. The study endpoint was AKI occurrence during hospitalization. The predictive value of lactate levels measured at admission and 12 h after admission and LC for AKI prediction was determined using multivariate logistic regression analyses and compared with receiver operator characteristic (ROC) curve analysis.ResultsOverall, AKI was observed in 353 (19.8%) patients. In multivariate logistic regression analyses, Lacadm ≥ 4.3 mmol/L (OR: 1.53; 95% CI: 1.01–2.30), Lac12h ≥ 2.1 mmol/L (OR: 1.81; 95% CI: 1.36–2.42), and LC ≥ −7.5% (OR: 0.40; 95% CI: 0.30–0.53) were the independent predictive factors for AKI after adjusting for confounders. ROC curve analysis results revealed that Lac12h (0.639; 95% CI: 0.616–0.661) exhibited a significantly higher area under the curve (AUC) than those of Lacadm (0.551; 95% CI: 0.527–0.574) and LC (0.593; 95% CI: 0.570–0.616) in the prediction of AKI. LC (△AUC = 0.037, p < 0.001) and Lac12h (△AUC = 0.017, p = 0.029) enhanced the discrimination capacity of Mehran Risk Score (MRS) for AKI among patients undergoing emergency coronary angiography.ConclusionLac12h is more effective for AKI prediction among patients with STEMI than Lacadm and LC. Furthermore, Lac12h and LC enhance the prediction capacity of MRS for AKI among patients after emergency coronary angiography.