Insufficient rate of recruitment

Despite almost universal usage of supplemental oxygen therapy in patients presenting in the emergency department with traumatic brain injury (TBI), optimal oxygen levels are unclear.

The investigators propose a pilot multi-center randomized controlled trial to test the hypothesis that maintaining intermediate normal as opposed to high normal oxygen levels in patients presenting in the emergency department with TBI is feasible, and to obtain preliminary data on the efficacy of the two approaches to oxygen therapy. The aim is that the investigators produce pilot data, which could inform the design of potential subsequent larger clinical trials.

Despite the worldwide burden of traumatic brain injury (TBI), medical research on the field as opposed to other health problems is underrepresented. Consequently, there are few data to support commonly used interventions for the management of TBI, especially in the setting of the emergency department. For example, despite almost universal usage of supplemental oxygen therapy, the effects of different oxygenation levels under normobaric conditions on outcomes of patients presenting in the emergency department with TBI are unknown.

On the one hand, liberal oxygenation may provide a margin of safety against hypoxemia and may be needed to meet the high oxygen demands of an acutely altered brain physiology. On the other hand, there are increasing concerns that excessive oxygen supplementation may have harmful effects, such as central nervous system toxicity, cerebral vasoconstriction, impaired immunity leading to predisposition to infections (including pneumonia) and acute lung injury/acute respiratory distress syndrome. Such effects could be avoided by intermediate normal oxygen levels.

Taken together, the relative merits and risks of the abovementioned two approaches to oxygen therapy (namely, intermediate normal versus high normal oxygen levels) of patients with TBI in terms of important clinical outcomes (namely, development of nosocomial pneumonia, acute respiratory distress syndrome, disability and mortality) remain undefined. This suggests the need for randomized controlled trials. However, randomized controlled trials focusing on patient-centered outcomes should be preceded by pilot randomized controlled trials, which demonstrate a separation in treatment and protocol compliance (feasibility) associated with the studied interventions.

Therefore, the investigators propose a pilot multi-center randomized controlled trial to test the hypothesis that maintaining intermediate normal as opposed to high normal oxygen levels in patients presenting in the emergency department with TBI is feasible, and to obtain preliminary data on the efficacy of the two approaches to oxygen therapy.

Inclusion Criteria:

• Adult patient ≥18 years
• Glasgow Coma Scale ≤ 8
• Non-penetrating traumatic brain injury
• Intubated patient

Exclusion Criteria:

• Age <18 years
• Lack of intention to admit to the intensive care unit
• Moribund patient expected to die within 24 hours
• Expected need for mechanical ventilation < 24 hours
• Time interval from intubation to group allocation more than 60 minutes
• Penetrating traumatic brain injury
• Pregnancy
• Lack of equipoise of the treating clinician
• Lack of informed consent

To avoid selection bias, the allocation sequence will be blinded from researchers involved in patient enrolment. Although study subjects will be unaware of the assigned group, blinding of treating clinicians is not considered feasible.