A high fraction of inspired oxygen may increase mortality in intubated trauma patients - A retrospective cohort study

Background: Mechanical ventilation of trauma patients is common, and many will require a higher than normal fraction of inspired oxygen (FiO(2)) to avoid hypoxaemia. The primary objective of this study was to assess the association between FiO(2) and all-cause, one-year mortality in intubated trauma patients. Methods: Adult trauma patients intubated in the initial phase post-trauma between 2015 and 2017 were retrospectively identified. Information on FiO(2) during the first 24 hours of hospitalisation and mortality was registered. For each patient the number of hours of the first 24 hours exposed to an FiO(2) >= 80%, >= 60%, and >= 40%, respectively, were determined and categorised into exposure durations. The associations of these FiO(2) exposures with mortality were evaluated using Cox regression adjusting for age, sex, body mass index (BMI), Injury Severity Score (ISS), prehospital Glasgow Coma Scale (GCS) score, and presence of thoracic injuries. Results: We included 218 intubated trauma patients. The median prehospital GCS score was 6 and the median ISS was 25. One-year mortality was significantly increased when patients had received an FiO(2) above 80% for 3-4 hours compared to <2 hours (hazard ratio (95% CI) 2.7 (1.3-6.0), p= 0.011). When an FiO(2) above 80% had been administered for more than 4 hours, there was a trend towards a higher mortality as well, but this was not statistically significant. There was a significant, time-dependent increase in mortality for patients who had received an FiO(2) >= 60%. There was no significant relationship observed between mortality and the duration of FiO(2) >= 40%. Conclusion: A fraction of inspired oxygen above 60% for more than 2 hours during the first 24 hours of admission was associated with increased mortality in intubated trauma patients in a duration-dependent manner. However, given the limitations of this retrospective study, the findings need to be confirmed in a larger, randomized set-up. (C) 2021 Elsevier Ltd. All rights reserved.