Utilizing Point-of-Care Lactate Testing for Rapid Prediction of Clinical Outcomes in Patients with Acute Gastrointestinal Bleeding in the Emergency Department

Abstract

ObjectivesWe aimed to rapidly predict the prognosis of patients who present to the emergency department (ED) with acute gastrointestinal bleeding (AGIB) using point-of-care (POC) lactate testing. MethodsThis single-center retrospective observational study included 327 patients (survival group, 287; non-survival group, 40) who presented to the ED with AGIB between March 2021 and February 2022. We compared POC-measured lactate levels with laboratory-measured lactate levels using Pearson’s correlation. Multivariate logistic regression analysis was used to identify early predictors of in-hospital mortality and correlated clinical outcomes. Receiver operating characteristic (ROC) curves were used to determine the optimal cutoff for POC-measured lactate levels for predicting in-hospital mortality, and the ROC curves for POC-measured lactate levels and AIMS65 scores were compared using the DeLong test. ResultsPOC-measured lactate levels strongly correlated with laboratory-measured lactate levels (R2=0.82). Patients in the non-survival group had higher POC-measured lactate levels than did those in the survival group (2.6 mmol/L vs. 1.4 mmol/L, p<0.001). POC-measured lactate level, age, systolic blood pressure, heart rate, and malignancy were identified as early predictors of in-hospital mortality (adjusted odds ratio [aOR] for POC-measured lactate levels: 1.15; 95% confidence interval [CI] 1.02–1.30). The optimal POC-measured lactate level cutoff was 3.2 mmol/L. Areas under the ROC curves for POC-measured lactate level and the AIMS65 score were 0.70 and 0.73, respectively, showing statistical compatibility. Higher POC-measured lactate levels correlated with ICU admission, blood transfusion, and mechanical ventilation (aOR: 1.16, 95% CI 1.05–1.27; 1.16, 1.04–1.30; and 1.31, 1.13–1.53, respectively]. Further, the hyperlactatemia subgroup (POC-measured lactate level >3.2 mmol/L) exhibited a lower survival probability in the Kaplan–Meier survival analysis (p<0.01). ConclusionsOur study shows that rapidly obtainable POC-measured lactate levels are valuable for predicting critical outcomes in AGIB patients and should be considered an early prognostic indicator for in-hospital mortality in the ED.