Appropriate triage of the trauma patient is critical. Low end-tidal carbon dioxide (ETCO2) is associated with mortality and hemorrhagic shock in trauma, but the relationship between low ETCO2 and important clinical variables is not known. This study investigates the association of initial in-hospital ETCO2 and patient outcomes, as well as the utility of ETCO2 as a predictive aid for blood transfusion.

Adult patients who presented to a Level One trauma center from 2019 to 2020 were eligible. Trauma bay ETCO2 measured by side-stream capnography was prospectively obtained for all trauma activations at time of initial evaluation. Using the Liu method of cut point estimation, patients were stratified as having low (≤29.5 mmHg) or normal ETCO2 (>29.5 mmHg). Multivariable regression was used to estimate the association of low ETCO2 with patient outcomes.

A total of 955 patients underwent initial in-hospital ETCO2 measurement. Median time from arrival to ETCO2 measurement was 4 min. Among admitted patients (N = 493), 48.9% had low ETCO2. Compared to patients with normal ETCO2, those with low ETCO2 were older (median age 53 vs 46, p = 0.01) and more likely to have the highest trauma activation (27.4% vs 19.8%, p = 0.048). There was no difference in head injury. After adjustment, patients with low ETCO2 had greater odds of blood transfusion (OR 4.65, 95%CI 2.0–10.7), mortality (OR 5.10, 95%CI 1.1–24.9), inferior disposition (OR 1.64, 95%CI 1.1–2.6), and complications (OR 3.35, 95%CI 1.5–7.4). ETCO2 was more predictive of early blood transfusion than Shock Index (area under ROC = 67.6% vs 58.2%).

Low trauma bay ETCO2 remains significantly associated with inferior clinical outcomes after adjustment. In comparison to other triage tools, low ETCO2 values may be more predictive of the need for blood transfusion. Further studies are needed to evaluate the role of ETCO2 as a decision making tool for early trauma management.