What’s New in the Patient Safety World

February 2012

More Evidence of Harm from Oxygen



In several of our prior columns on use of oxygen (see our Patient Safety Tips of the Week April 8, 2008 “Oxygen as a Medication” and January 27, 2009 “Oxygen Therapy: Everything You Wanted to Know and More!”) we have commented that in the past we often routinely gave oxygen to patients with myocardial infarction or stroke. But such use was more reflexive in nature and not evidence-based.


In our July 2010 What’s New in the Patient Safety World column “Cochrane Review: Oxygen in MI” we discussed a Cochrane Review (Cabello 2010) which suggested that not only is oxygen not likely beneficial in acute MI patients, it may even cause harm. They reviewed the literature but could find only 3 randomized trials of oxygen vs. room air in MI patients. Only 387 patients were included in these 3 trials and only 14 patients died. The pooled relative risk for death was almost 3.00 in the oxygen group but the confidence interval was wide. The authors concluded that there is no evidence to support the routine use of oxygen in the acute MI patient. They suggest that the issue of whether oxygen is harmful could only be answered in a more definitive randomized trial.


Around the same time another study (Kilgannon 2010) demonstrated that the use of hi-dose oxygen in post-cardiac arrest patients had a deleterious effect.


In recent months several reviews have again challenged the conventional practice of giving oxygen routinely to all patients with suspected MI, regardless of whether hypoxemia has been demonstrated. Kones (Kones 2011) reviewed the literature and presented the history, physiology, and clinical evidence (or lack thereof!) for the practice of giving oxygen in MI patients. He discusses the assumptions underlying the practice and some weak historical evidence that supported the practice, then discusses modern physiological knowledge of the effects of hyperoxygenation and the clinical evidence suggesting that hyperoxygenation may be harmful. He discusses the status of recommendations about oxygen in current guidelines and notes the importance of doing a large randomized controlled trial (which is apparently in progress). Kones also voices concern about the editorial (Atar 2010) accompanying the Cochrane review. That editorial had argued that the quality of the studies included in the Cochrane review was poor and that the study did not conclusively demonstrate increased mortality from oxygen and that with no strong current evidence of a deleterious effect of oxygen, treatment of MI patients with oxygen is still merited. Kones, on the other hand, argues that oxygen is indicated in MI patients who are hypoxemic and target oxygen saturations should be in the 94-96% range and that hyperoxia should be avoided until such time that more definitive studies are done.


A second recent paper (Cornet 2012) looks at use of supplemental oxygen in a variety of medical emergencies and notes the collective evidence argues against routine use of oxygen in most emergencies. Instead, they recommend a policy of careful, titrated oxygen supplementation. They discuss the previously mentioned MI data and the Kilgannon study on post-resuscitation hyperoxia. Then they discuss the data on oxygen use in stroke and COPD. They conclude that evidence suggests potential detrimental effects of hyperoxygenation and recommend that, when it is indicated, oxygen therapy should be titrated carefully and cautiously.


One other area in which use of hyperoxygenation has been of interest deals with surgical site infections. However, the PROXI study (Staehr 2011) showed that use of 80% vs. 30% O2 in obese patients undergoing abdominal surgery did not reduce the SSI rate. Though it did not show any difference in pulmonary complications or other adverse events between the groups, it failed to demonstrate any positive value of hyperoxygenation.


Hospitals need to look at their existing protocols (and actual practices) for managing a variety of medical conditions where oxygen use may be considered. How many of you have standardized order sets that directly (or indirectly by poor use of checkboxes) encourage inappropriate use of oxygen in MI or stroke patients? Going back to our Patient Safety Tips of the Week April 8, 2008 “Oxygen as a Medication” and January 27, 2009 “Oxygen Therapy: Everything You Wanted to Know and More!” we strongly support facilities doing audits of their oxygen practices. You’ll probably be surprised at the opportunities you uncover to improve practices (and save money at the same time!).


In addition, in many cases high doses of oxygen are administered by the pre-hospital emergency response teams. Making them aware of the potential dangers is also important.







Cabello JB, Burls A, Emparanza JI, Bayliss S, Quinn T. Oxygen therapy for acute myocardial infarction. Cochrane Reviews 2010; Published online June 16, 2010




Kilgannon JH, Jones AE, Shapiro NI et al. Association Between Arterial Hyperoxia Following Resuscitation From Cardiac Arrest and In-Hospital Mortality
JAMA. 2010; 303(21): 2165-2171




Kones R. Oxygen therapy for acute myocardial infarction-then and now. A century of uncertainty. Am J Med. 2011; 124(11): 1000-1005




Atar D. Should oxygen be given in myocardial infarction? BMJ  2010; 340: c3287 (Published 17 June 2010)




Cornet AD, Kooter AJ, Peters MJL, Smulders YM. Supplemental Oxygen Therapy in Medical Emergencies: More Harm Than Benefit? (Research Letter). Arch Intern Med 2012; published online January 9, 2012




Staehr AK, Meyhoff CS, Rasmussen LS,  PROXI Trial Group. Inspiratory Oxygen Fraction and Postoperative Complications in Obese Patients: A Subgroup Analysis of the PROXI Trial. Anesthesiology 2011; 114(6): 1313-1319
















Tip of the Week Archive


What’s New in the Patient Safety World Archive