Oxygen therapy has always been a critical part of resuscitation to prevent the well-documented harm caused by tissue hypoxia. The concept of more-is-better naturally became the standard practice for most emergency physicians when it came to oxygen delivery, especially in our newly intubated patients. Rarely do we return to the ventilator to turn down FiO2. This practice pattern developed despite the documented negative effects of hyperoxia via free radical production. (Chest 1972;62:162; Chest 2001;120:467.)
Recent research focusing on the potential harm caused by hyperoxia cannot be ignored. Kilgannon, et al., studied a large, multicenter cohort to look at the initial blood gas obtained in the ICU on resuscitated cardiac arrest patients and found a significantly higher in-hospital mortality for patients with a PaO2 of 300 mm Hg or greater. (JAMA 2010;303:2165.) A subsequent study performed by the same group showed a linear correlation between increasing oxygen tension and in-hospital mortality and an inverse relationship with functionally intact survival. (Circulation 2011;123:2717.) Similar findings have been demonstrated in patients undergoing therapeutic hypothermia after cardiac arrest. (Crit Care Med 2012;40:3135; Crit Care Resusc 2013;15:186; Am J Emerg Med 2014;32:55.)
The adverse effects of hyperoxia have been shown in a multitude of neurologic pathologies as well. Brenner, et al., looked at more than 1,500 patients with traumatic brain injury and found that high PaO2 (greater than 200 mm Hg) within the first 24 hours of hospital admission significantly correlated with higher mortality and lower discharge GCS scores compared with patients with normal oxygen levels. (Arch Surg 2012;147:1042.) The probability of being exposed to hyperoxia was independently associated with higher in-hospital mortality in another retrospective, multicenter study of more than 1,200 TBI patients. (J Neurol Neurosurg Psychiatry 2014;85:799.)
These results do not seem to be limited to TBI, either. Jeon, et al., looked at patients with spontaneous aneurysmal subarachnoid hemorrhage and found that exposure to high oxygen levels was associated with an increased risk for delayed cerebral ischemia and poor outcome. (J Neurol Neurosurg Psychiatry 2014;85:1301.)
Authors in the recent CLOSE trial — the first randomized, controlled trial of conservative versus liberal oxygenation strategies — demonstrated the need for more focus on hyperoxia. (Am J Respir Crit Care Med 2016;193:43.) Researchers randomized 103 primarily medical patients who were newly intubated and thought to require extended ventilation to a conservative oxygenation strategy (target SpO2 of 88-92%) or a liberal oxygenation strategy (target SpO2 ≥ 96%). The protocol was successful in producing a clear separation between conservative and liberal oxygenation strategies. The authors found no difference in a multitude of secondary safety outcomes, including ventilator-free days and ICU days as well as 90-day mortality. The data obtained from the study demonstrate not only the safety but the need for further randomized, controlled trials directed at outcomes.
What does this all mean for our day-to-day practice in the ED and ICU? For now, our focus during airway management should remain the avoidance of hypoxemia with the very liberal use of supplemental oxygen — the more the better — until further research becomes available. Once the airway is established, however, we should be vigilant and active in aggressive titration of FiO2, especially in patients with neurological disease or post-cardiac arrest. At supra-physiologic levels, oxygen may have toxic effects. The specific goals for saturation and PaO2 are not yet well-defined, but we believe it is reasonable to titrate the FiO2 down to the lowest level that will maintain a saturation greater than 96 percent.
Share this article on Twitter and Facebook.
Access the links in EMN by reading this on our website or in our free iPad app, both available at www.EM-News.com.
Comments? Write to us at email@example.com.