Background Early definitive airway protection and normoventilation are key principles in the treatment of severe traumatic brain injury. These are currently guided by end tidal CO2 as a proxy for PaCO2. We assessed whether the difference between end tidal CO2 and PaCO2 at hospital admission is associated with in-hospital mortality.Method We conducted a retrospective observational cohort study of consecutive patients with traumatic brain injury who were intubated and transported by Helicopter Emergency Medical Services to a Level 1 trauma center between January 2014 and December 2019. We assessed the association between the CO2 gap—defined as the difference between end tidal CO2 and PaCO2—and in-hospital mortality using multivariate logistic regression models.Results 105 patients were included in this study. The mean ±SD CO2 gap at admission was 1.64 (± 1.09) kPa and significantly greater in non-survivors than survivors (2.26 ±1.30 kPa vs. 1.42 ±0.92 kPa, p<.001). The correlation between EtCO2 and PaCO2 at admission was low (Pearson's r=.287). The mean CO2 gap after 24 hours was only 0.64 ±0.82 kPa, and no longer significantly different between non-survivors and survivors. The multivariate logistic regression model showed that the CO2 gap was independently associated with increased mortality in this cohort and associated with a 2.7-fold increased mortality for every 1 kPa increase in the CO2 gap (OR 2.692, 95% CI 1.293 to 5.646, p=.009).Conclusions This study demonstrates that the difference between EtCO2 and PaCO2 is significantly associated with in-hospital mortality in patients with traumatic brain injury. EtCO2 was significantly lower than PaCO2, making it an unreliable proxy for PaCO2 when aiming for normocapnic ventilation. The higher-than-expected CO2 gap will lead to iatrogenic hypoventilation when normocapnic ventilation is aimed at, and might thereby increase in-hospital mortality.