The PaCO 2 -EtCO 2 gradient in patients with a prehospital inserted arterial catheter for a trauma mechanism by a helicopter emergency medical service in the United Kingdom

Introduction

Tight arterial carbon dioxide (CO₂) control is fundamental to reducing the impact of secondary brain injury in patients with traumatic brain injury. ¹ Unintentional hypercapnia and hypocapnia may worsen outcomes for patients, particularly those with a traumatic brain injury (TBI). ¹ Current guidance recommends using end-tidal carbon dioxide (EtCO₂) as a surrogate for PaCO₂ to guide ventilation strategy when arterial blood gas (ABG) sampling is less readily available. ¹ The EtCO₂ is recommended to be maintained between 4.0 and 4.5 kPa, and the accepted gradient between the PaCO₂–ETCO₂ is considered to be 0.5–1.0 kPa; however, reappraisals have demonstrated heterogeneity in the agreement between the PaCO₂ and EtCO₂, as well as variability in the gradient.1–3

This study aims to describe the gradient and the frequency with which the EtCO₂ misclassifies the PaCO₂ in the prehospital environment.

Methods

This retrospective observational study included adult patients (aged ⩾ 18 years) who underwent prehospital arterial catheterisation following a traumatic mechanism who were conveyed to the regional Major Trauma Centre (MTC), 1st February 2015–17th April 2023. Data were obtained from both the East Anglian Air Ambulance (EAAA) electronic medical record system (HEMSbase, Medic One Systems Ltd, UK) and the hospital electronic medical record system (Epic Hyperspace Production^®, Epic Systems Corporation, Verona, WI, USA).

EtCO₂ measurements were matched within 5 min of the prehospital ABG sample. Hypocapnoea was defined as <4 kPa, normocapnoea as 4–6 kPa, and hypercapnoea as >6 kPa.^1,3 Misclassification was defined as a discrepancy between the EtCO₂ and PaCO₂ categories (e.g. one indicating normocapnoea and the other hypocapnoea or hypercapnoea). Data were described as number (percentage) for categorical variables and median [interquartile range (IQR)] for continuous variables.

Results

During the study period, 53 patients with prehospital arterial catheterisation for a trauma mechanism and who had an ABG recorded prehospital were identified (Figure 1).

OPEN IN VIEWER

Figure 1.

Flow diagram of included and excluded patients.

Demographics and injury characteristics are described in Table 1.

OPEN IN VIEWER

Table 1.

Demographics and injury characteristics.

Data variable	Descriptive statistics
Male sex, n (%)	42 (79.2%)
Age in years, median [IQR]	56 [40–69]
Injury severity score, median [IQR]	29 [25–38]
Pre-hospital shock index, median [IQR]	0.946 [0.636–1.215]
Most injured region
Head and neck, n (%)	46/ 53 (86.7%)
Chest, n (%)	4/53 (7.5%)
Abdomen and pelvis, n (%)	3/53 (5.8%)
Isolated head injury, n (%)	17/53 (32.1%)
Intubated at time of blood prehospital blood gas, n (%)	51/53 (96.2%)
Mortality, n (%)	16/53 (30.2%)
Timings
Injury time to blood gas measurement in minutes, Median [IQR]	86 min [75–114 min]
Injury time to hospital blood gas measurement in minutes, median [IQR]	147 min [131−178 min]
Pre-hospital CO2 values
PaCO2 kPa, median [IQR]	5.8 [5.3–6.9]
EtCO2 kPa, median [IQR]	4.1 [3.8–4.5]
PaCO2–EtCO2 gradient kPa, median [IQR]	1.7 [1.2–2.6]
Pre-hospital PaCO2–ETCO2 gradient differences and misclassification
PaCO2–EtCO2 > 1 kPa difference, n (%)	43/53 (81.1%)
EtCO2 misclassified the PaCO2, n (%)	38/53 (71.7%)

Most patients were intubated at the time of ABG sampling (51/53 (96.2%)). The PaCO₂–EtCO₂ gradient was more than >1 kPa difference in most patients (43/53(81.1%)) and the ETCO₂ misclassified the PaCO₂ in over two-thirds of patients (38/53 (71.7%)).

Discussion

This study observed that PaCO₂–EtCO₂ gradient was frequently more than the accepted gradient of 0.5–1.0 kPa, and the EtCO₂ often misclassified the PaCO₂.

This study supports previous studies demonstrating the wider gradient and its association with mortality.^2–4 However, previous studies have either not focussed on trauma where targeted normocapnia is vital, or been undertaken later in the injury journey in the emergency department or intensive care unit.^2–4 Given the differences in methodology between this paper and smaller, previously published papers utilising the same database, only one patient was found to overlap.^2,3 Moreover, this study is advantageous in closely matching the EtCO₂ and PaCO₂ to within 5 min, which is particularly important given the variability in gradient demonstrated in prior studies.^2–4

Currently, only a small proportion of critical care services in the UK can measure PaCO₂ prehospital, and since the data comes from a single centre, the results may not be widely applicable. ⁵

Overall, EtCO₂ is not a reliable surrogate for PaCO₂ in guiding ventilator settings during trauma. Clinicians should be mindful of this and consider ABG sampling when available.