The role of lactate-to-albumin ratio to predict 30-day risk of death in patients with sepsis in the emergency department: a decision tree analysis

Abstract

Background Accurately estimating the prognosis of septic patients on arrival in the emergency department (ED) is clinically challenging. The lactate-to-albumin ratio (LAR) has recently been proposed to improve the predictive performance of septic patients admitted to the ICU. Objectives This study aims to assess whether the LAR could be used as an early prognostic marker of 30-day mortality in patients with sepsis in the ED. Methods A prospective observational study was conducted in the ED of the Hospital of Merano. All patients with a diagnosis of sepsis were considered. The LAR was recorded on arrival in the ED. The primary outcome measure was mortality at 30 days. The predictive role of the LAR for mortality was evaluated with the area under the ROC curve, logistic regression adjusted for the Charlson Comorbidity Index value, National Early Warning Score, and Sequential Organ Failure score, and with decision tree analysis. Results 459 patients were enrolled, of whom 17% (78/459) died at 30 days. The median LAR of the patients who died at 30 days (0.78 [0.45–1.19]) was significantly higher than the median LAR of survivors (0.42 [0.27–0.65]) (p < 0.001). The discriminatory ability of the LAR for death at 30 days was 0.738, higher than that of lactate alone (0.692), and slightly lower than that of albumin alone (0.753). The decision trees confirmed the role of the LAR as an independent risk factor for mortality. Conclusion The LAR can be used as an index to better predict the 30-day risk of death in septic patients.