Variations in shock indicators in arterial blood gas analysis and prognosis in patients with severe trauma

Shock indicators in arterial blood gas analysis in the context of trauma are essential tools for early diagnosis, when clinical signs are not yet present. The study aimed to analyze the potential prognostic role these indicators may play in the evaluation of trauma patients.

A retrospective study of 294 patients with severe trauma treated between 2016 and 2025 at a tertiary care hospital. Only patients with blood gas analysis at admission and after 24 hours, activation of the massive transfusion protocol, and exploratory laparotomy were included.

Of the 294 patients, 74 (25.2%) died. Lactate variation was significantly higher in patients who died (p < 0.001). The AUC of the ROC curve was 0.6949, with a sensitivity of 72%, specificity of 98.5%, and positive predictive value of 96.4%. The odds ratio for death in patients with an increase in lactate ≤1 mmol/L was 167.14. In contrast, the variation in base excess was not statistically significant (p = 0.054), with an AUC of 0.5748. There was a weak correlation between the variation in base excess and the outcome (r = -0.1124; p = 0.0063), and a weak to moderate correlation for the variation in lactate (r = 0.2931; p < 0.0001). There was also a significant association between outcome and age, Glasgow Coma Scale score, and laboratory parameters at admission and 24 hours.

Lactate variation in the first 24 hours was associated with mortality in patients with severe trauma and demonstrated superior prognostic performance compared to base excess in this sample, with moderate discriminatory power.