The Lactate Dehydrogenase-to-Albumin Ratio (LAR) Predicts Mortality in Sepsis-Induced Myocardial Injury

Background Sepsis-induced myocardial injury (SIMI), a fatal complication seen in 13%-65% of septic patients with mortality exceeding 50% in severe cases, is characterized by profound metabolic and inflammatory dysregulation.While the Sequential Organ Failure Assessment (SOFA) score is commonly used to assess disease severity in these patients, there remains a need for biomarkers that more directly reflect the metabolic and inflammatory components of SIMI.The lactate dehydrogenase-to-albumin ratio (LAR), which integrates markers of cellular damage and systemic inflammation, has shown promise in sepsis but its prognostic utility specifically in SIMI remains unexplored. Method This retrospective cohort study analyzed data from the MIMIC-IV database involving 4,692 ICU patients with sepsis-induced myocardial injury (SIMI). The primary endpoint was 28-day all-cause mortality. The association was assessed using Cox regression, with restricted cubic splines examining nonlinearity. Subgroup analyses were performed by age, sex, and diabetes. Machine learning models (random forest, XGBoost, decision tree) were developed to validate LAR's predictive value, with performance evaluated by receiver operating characteristic (ROC) curves and decision curve analysis. Results The Random Survival Forest (RSF) model incorporating LAR achieved an AUC of 0.775 (95% CI: 0.752–0.803) for predicting 28-day ICU mortality, outperforming the SOFA score alone (AUC = 0.698, P < 0.001).The prognostic value of LAR remained consistent across key subgroups, including age, sex, and diabetic status (all interaction P > 0.05), underscoring its robustness as a risk stratifier. Furthermore, decision curve analysis confirmed the clinical utility of the model, demonstrating a superior net benefit across a wide range of risk thresholds. Conclusion The lactate dehydrogenase-to-albumin ratio (LAR) is an independent predictor of 28-day mortality in sepsis-induced myocardial injury. Incorporating LAR into clinical risk stratification could improve early identification of high-risk patients.