Skip to main content
Patient Care

Journal Watch: End-Tidal Carbon Dioxide and the Need for Transfusion

Reviewed This Month

Prehospital End-Tidal CO2 as an Early Marker for Transfusion Requirement in Trauma Patients 

Authors: Wilson BR, Bruno J, Duckwitz M, et al. 

Published in: Amer J Emerg Med,  2020 Aug 22 [epub online]

Most of you probably know trauma is a leading cause of death in the U.S. Unintentional injury is the top cause of death for those younger than 45. An estimated 40% of trauma deaths are due to uncontrolled hemorrhage. And delays in transfusions for trauma patients have been shown to increase mortality. 

Studies have also shown many trauma patients have better outcomes when they receive care at trauma centers. Yet we currently do not have an adequate measure to predict the need for transfusion or the need for transport to a trauma center. Even the American College of Surgeons Committee on Trauma guidance and CDC field triage scheme have been found less than ideal in identifying major trauma patients. However, low EtCO2 measurements have been shown to be associated with hemorrhagic shock and worse outcomes for those injured. 

The authors of this month’s study sought to determine if low EtCO2 measurements among prehospital trauma patients could be used as criteria for transport to a trauma center. Specifically they evaluated whether low prehospital EtCO2 was associated with the need for transfusion, or “operative intervention,” in the first 24 hours after admission to a suburban academic Level 1 trauma center. 

The authors evaluated all trauma alerts transported to this hospital via EMS from December 1, 2017 to November 30, 2018. They also then evaluated only patients less than 65 years old. They indicated this separate analysis was necessary because geriatric patients are more likely to have underlying pulmonary disease that might affect EtCO2 readings. 

Data obtained from the prehospital care reports included mechanism of injury, vital signs, EtCO2, and intubation status, as well as patient demographics. Hospital records were used to obtain in-hospital intubations, vital signs, GCS in the trauma bay, Injury Severity Score, and patient outcomes including need for transfusion, need for surgery to control hemorrhage, and death within 24 hours. 

Low EtCO2 was defined as less than or equal to 30 mmHg. Sensitivity and specificity analyses were also performed to find the optimal EtCO2 cut point to predict the need for transfusions. Generally, high-sensitivity tests have low specificity, and you can find the ideal cut point by identifying the point where sensitivity and specificity are highest. 


There were 235 trauma patients included in this analysis. Of those 105 had EtCO2 documented. The ages of these patients ranged from 17 to 97 years with a median of 61.5. The interquartile range (or middle 50% of data, 25%–75%) was 38–82. More than half the patients with documented EtCO2 were male. Blunt trauma was the most common mechanism of injury, most often as the result of a fall, followed by motor vehicle crashes. Roughly two-thirds (68%) of those with EtCO2 of 30 mmHg or below were fall patients, and 18% were victims of motor vehicle crashes. 

A greater percentage of patients with EtCO2 of 30 or below had injury severity scores greater than 15 (22% vs. 2%). Additionally, more patients with EtCO2 of 30 or less were intubated by EMS (three with EtCO2 of 30 or less vs. none with EtCO2 greater than 30). When evaluating in-hospital intubations, three patients with EtCO2 of 30 or less were intubated in the hospital, compared to four with EtCO2 greater than 30. 

There were two patients who required operative hemorrhage control, eight who required transfusion, and one who died within 24 hours. There was no difference in the age of patients who required the outcome interventions and those who did not (p = 0.47). However, those who required transfusion, operative hemorrhage control, or who died had significantly lower median prehospital EtCO2 measurements (25.5 vs. 31.9, p ≤ 0.05). When excluding geriatric patients, the median EtCO2 for those who required the outcome interventions was 25.0, compared to 34.3 for those who did not (p = 0.01). 

Systolic blood pressure was also associated with the negative outcomes assessed in this study, with those who experienced the outcome of interest having a median SBP of 115.5 mmHg compared to 146 mmHg for those who did not (p = 0.03). Median shock index scores (0.82 vs. 0.61, p = 0.02) and average hemoglobin (9.8 vs. 13.1, p < 0.001) were also significantly associated with the need for transfusion, surgery to control hemorrhage, or death within 24 hours. 

The authors identified the optimal EtCO2 cut point as 27, which resulted in a sensitivity of 73% and a specificity of 72%. 


This is a great study not only because it is well written and the analysis is appropriate, but because it is a wonderful example that shows that you don’t need millions of patients in your study to get it published and add to the literature. This study only had 11 patients that experienced their outcome of interest. Yes, this is clearly a limitation; it also highlights the difficulty of collecting and evaluating data for low-frequency, high-criticality events. 

This study is a preliminary step toward determining if capnography readings, a noninvasive and relatively easily collected vital sign, can help direct the right patient to the right care at the right time. For that reason this is an important addition to the existing literature. The results of this study should lead to future studies that include more patients.   

Antonio R. Fernandez, PhD, NRP, FAHA, is a research scientist at ESO and an assistant professor in the department of emergency medicine at the University of North Carolina–Chapel Hill. He is on the board of advisors of the Prehospital Care Research Forum at UCLA. 


Back to Top