Association between normalized lactate load and in-hospital mortality in patients with acute myocardial infarction

Background: Lactate was a prognostic indicator for acute myocardial infarction (AMI) patients. However, the association between normalized lactate load, representing hypoxic burden over time, and in -hospital mortality remained uncertain. Methods: The data for this study was obtained from the Medical Information Mart for Intensive Care IV (MIMICIV, version 2.1) database. The normalized lactate load, describing the average intensity of hyperlactatemia, was calculated as the area under the curve (AUC) of lactate divided by time. 5882 AMI patients enrolled in this study were divided into survivor (n = 5015), and non -survivor group (n = 867). The primary endpoint was in -hospital mortality. Receiver operating characteristic (ROC) curves were generated to assess the predictive efficacy of normalized lactate load for in -hospital mortality, and areas under the curves of different parameters were compared using DeLong test. Multivariate binary logistic regression analysis was employed to explore the association between normalized lactate load and in -hospital mortality. The adjusting variables included age, gender, ethnicity, heart rate, systolic blood pressure, congestive heart failure, shock, dyslipidemia, cardiac arrest, cerebrovascular disease, neutrophil, lymphocyte, creatinine, blood nitrogen urea, clopidogrel, beta-blockers, angiotensin-converting enzyme inhibitor (ACEI)/angiotensin receptor blocker (ARB), statins, dialysis, extracorporeal membrane oxygenation (ECMO), the Sequential Organ Failure Assessment (SOFA) score and Simplified Acute Physiology Score II (SAPS II). Restricted cubic spline (RCS) was conducted to evaluate nonlinear associations of normalized lactate load with in -hospital mortality. Results: The overall in -hospital mortality rate was 14.7%. After adjusting for confounding variables, normalized lactate load was independently associated with increased risk of in -hospital mortality (Normalized lactate load >= 2.6 vs Normalized lactate load<2.6: OR, 95% CI: 1.56, 1.27-1.93). The RCS demonstrated a positive linear relationship between normalized lactate load and in -hospital mortality (non-linear p = 0.725). ROC curves showed that normalized lactate load was better than first lactate, maximum lactate, and mean lactate in predicting in -hospital mortality, but lower than SOFA and SAPS II. Among participants with at least nine lactate measures, normalized lactate load showed predictive performance comparable to SOFA and SAPS II. Conclusion: Normalized lactate load can be used to predict the prognosis of in -hospital mortality in AMI patients, and its prediction performance increases with the increase of lactate measurement.