Review of: Kilgannon JH, Jones AE, Shapiro NI, et al. Association between arterial hyperoxia following resuscitation from cardiac arrest and in-hospital mortality. JAMA. 2010;303:2165–2171.
This large, prospective observational study investigated the association between elevated arterial oxygen levels (hyperoxia) and increased hospital mortality post-resuscitation from cardiac arrest. The data was pulled from the Project IMPACT database, which encompasses 120 intensive care units (ICUs) across the U.S. Hyperoxic patients (defined as PaO2 of > 300 mmHg) were compared against those who were hypoxic (defined as PaO2 < 60 mmHg) and normoxic post-cardiac arrest. A total of 6,326 patients were eligible for review: 1,156 were hyperoxic, 3,999 were hypoxic, and 1,171 patients were normoxic with mortality rates of 63%, 57% and 45%, respectively. The investigators were able to show that hyperoxic states are associated with increased risk of death.
Medic Marshall: Reading this study was a real eye-opener for me with regard to the potential lethality of oxygen. How many times do you show up on a scene—any scene, for that matter—and your first responders have a non-rebreather mask on your patient? This patient just so happens to be complaining of stomach cramps, isn’t short of breath and isn’t complaining of chest pain or dizziness … well, you get the picture. Yet this study was able to show that too much oxygen post-resuscitation might actually be killing people.
Here’s a more likely scenario: You roll up on a cardiac arrest to find the patient being ventilated at rate of 40 plus breaths per minute with 100% oxygen. We already know patients are being hyperventilated, but now hyper-oxygenated? I find this a little scary, personally. I also don’t know how to address the issue.
So what does this all really mean? Do we stop administering oxygen in high doses to everyone? Are there only certain conditions under which oxygen should be administered? I don’t know, but it would sure seem that way.
Doc Wesley: Our understanding of cardiac arrest physiology continues to significantly expand with research such as this. There’s a growing body of science to indicate that oxygen is not a benign drug. This concept runs counter to our longstanding prolific use of oxygen in every Department of Transportation and National Highway Traffic Safety Administration curriculum since their inception in the 1970s.
However, the U.S. has lagged behind the recommendations of such notable organizations as the British Thoracic Society. They released guidelines for oxygen administration in 2008 that prescribed that there’s no reason to exceed an oxygen saturation of greater than 96%.
In the study, the data clearly showed a significant increase in mortality when the arterial oxygen content exceeded 300 mmHg and, in fact, was worse than that of patients who were hypoxic. Additionally, the neurologic status of survivors was worse if they experienced post-arrest hyperoxemia.
The theory is that during ischemia there are several protective mechanisms the body uses to inhibit cell death. These anti-inflammatory and antioxidant responses are inhibited by high concentrations of oxygen, particularly immediately after perfusion is restored. In this study, the first measurement of arterial oxygen content was made when the patient arrived in the ICU, which was more than likely an hour or more after return of spontaneous circulation (ROSC).
Although this study may appear to have more impact on emergency department care of the cardiac arrest patient, the fact that animal data indicates that oxygen causes cellular damage immediately following ROSC may force us to consider titrating oxygen delivery in the field. This will be problematic because few EMS providers are equipped with variable oxygen concentration delivery systems.