Supplemental oxygen during cardiac arrest resuscitation

by Michael Smith

Administering high concentrations of supplemental oxygen during resuscitation after cardiac arrest is associated with increased inhospital mortality, researchers said.

So-called blood hyperoxia resulting from being given pure oxygen after cardiac arrest was associated with an 80% increase in the risk of death, compared with patients whose blood gases were normal on admission to the intensive care unit (ICU), according to Stephen Trzeciak, MD, of Cooper University Hospital in Camden, N.J., and colleagues.

The finding, from a retrospective cohort study of more than 6,000 patients treated in hospital ICUs over a four year period, adds to data regarding the controversial practice of giving 100% oxygen during resuscitation efforts — but cannot show a causal link, Trzeciak and colleagues said in the June 2 issue of the Journal of the American Medical Association.

In neonatal resuscitation, the use of 100% oxygen has been shown to have detrimental effects and there is animal data suggesting its use leads to worse neurological outcomes. But the “clinical data are lacking” in human adults, the researchers noted.

To help fill this knowledge gap, Trzeciak and colleagues turned to Project IMPACT, a large administrative database containing information from 131 adult intensive care units across the U.S. that included data from more than 400,000 patients.

For this analysis, they decided that hyperoxia after the return of spontaneous circulation would be defined as a partial arterial oxygen pressure of 300 millimeters of mercury seen in the first blood gas results after a patient was admitted to intensive care.

Hypoxia was defined as partial arterial oxygen pressure of less than 60 millimeters of mercury, while normoxia included all other results.

The database included 6,326 patients who met the inclusion criteria: age greater than 17, nontraumatic cardiac arrest, cardiopulmonary resuscitation within 24 hours before admission to intensive care, and an arterial blood gas analysis within 24 hours of ICU admission.

In that group, the researchers found, hyperoxia was common — 18% were hyperoxic, 63% were hypoxic, and 19% were normoxic.

The primary outcome measure was inhospital mortality, the researchers said, noting that death in such circumstances is the rule, not the exception: Even if return of spontaneous circulation is achieved, about 60% of patients do not live to discharge.

In this case, 56% of the patients — some 3,561 — died before discharge, with the mortality highest in the hyperoxia group, at 63%. The mortality rate in the hypoxia group was 57%, and 45% in the normoxia group. The differences between the hyperoxia group and the other groups were both significant at P<0.001.

Even if they survived, however, only 29% of the hyperoxia group were able to live independently after discharge, the researchers found, compared with 38% of the normoxic patients, a difference that was significant at P=0.002.

Limited patient functioning after discharge could be due to brain damage. In background provided in the paper, the researchers noted that reperfusion can lead to oxidative stress, cell damage, and brain injury, which “can be perpetuated in a persistently hyperoxic environment.”

In a multivariate regression analysis, nine factors independently predicted death, the researchers reported, including demographic factors such as age and independent living before the attack, and physiological factors such as hypotension on ICU arrival, tachycardia, and hypoxia.

Hyperoxia was included with a hazard ratio for inhospital death of 1.8 (and a 95% confidence interval from 1.5 to 2.2), which was significant at P<0.001, the researchers said.

The study should be interpreted with caution, they said, because of its observational nature. Other limitations included the sole use of first blood gas results after ICU admission, and the inability of a database designed from an ICU perspective that cannot capture variables such as initial cardiac rhythm and the period without circulation.

Nonetheless, the study “provides an impetus for better defining the use of oxygen in all settings of cerebral resuscitation,” said Patrick Kochanek, MD, and Hülya Bayır, MD, both of the University of Pittsburgh School of Medicine.

Writing in an accompanying editorial, they said the study “underscores the possibility” that outcomes might be improved by more careful titration of oxygen to prevent hyperoxia.

Michael Smith is a MedPage Today North American Correspondent.

Originally published in MedPage Today. Visit for more critical care news.