Accuracy of prehospital point-of-care lactate in early in-hospital mortality

Background Emergency medical services (EMS) routinely face complex scenarios where decisions should be taken with limited clinical information. The development of fast, reliable and easy to perform warning biomarkers could help in such decision-making processes. The present study aims at characterizing the validity of point-of-care lactate (pLA) during prehospital tasks for predicting in-hospital mortality within two days after the EMS assistance. Materials and methods Prospective, multicentric, ambulance-based and controlled observational study without intervention, including six advanced life support and five hospitals. The pLA levels were recorded during EMS assistance of adult patients. The validity of pLA to determine the in-hospital mortality was assessed by the area under the curve (AUC) of the receiver operating curve (ROC). Results A total of 2997 patients were considered in the study, with a median of 69 years (IQR 54-81) and 41.4% of women. The median pLA value was 2.7 mmol/L (1.9-3.8) in survivors and 5.7 mmol/L (4.4-7.6) in nonsurvivors. The global discrimination level of pLA reached an AUC of 0.867, being 1.9 mmol/L and 4 mmol/L the cut-off point for low and high mortality. The discrimination value of pLA was not affected by sex, age or pathology. Conclusions Our results highlight the clinical importance of prehospital pLA to determine the in-hospital risk of mortality. The incorporation of pLA into the EMS protocols could improve the early identification of risky patients, leading to a better care of such patients.