Critical Care Commentary

Hyperoxia in the ICU: Is less more?


“All things are poison and nothing is without poison, only the dose permits something not to be poisonous.” Paracelsus once said.

A bit of history

Oxygen was discovered in 1775 and was since noted to be both vital and poisonous. It was much later in 1899 that it was demonstrated that partial pressures of oxygen up to 75% led to both severe lung injury and death as compared with levels of 40% to 50%. While the administration of oxygen in hypoxic patients is beneficial, this intervention in healthy subjects leads to a reduction in heart rate, cardiac index, and an increase in mean arterial pressure, systemic vascular resistance, and large artery stiffness.

Dr. Said Chaaban, University of Kentucky, Lexington

Dr. Said Chaaban

While oxygen itself is not toxic, the reactive oxygen species that form as a result of oxygen metabolism are. A study showed that supplementation of oxygen in patients with COPD, or in women undergoing C-section with the use of spinal anesthesia, leads to an increase in reactive oxygen species (Winslow RM. Transfusion. 2013;53[2]:424).

Hyperoxia has multiple clinical effects on lung physiology and gas exchange that include worsening hypoxemia secondary to absorptive atelectasis and damage to the airways and lung parenchyma (Sackner MA, et al. Ann Intern Med. 1975;82[1]:40).

High levels of inspired oxygen could also lead to accentuation of hypercapnia as explained by the Haldane effect; a reduction of the affinity for carbon dioxide leading to an increase in PaC02. High oxygen levels can also decrease the hypoxic drive for ventilation leading to worsening hypercapnia.

Hyperoxia is a situation routinely encountered in clinical practice, as well, often resulting from an overzealous attempt to prevent or reverse hypoxia. ICU physicians, though aware of potential threats of hyperoxia, often fail to translate such concerns in their clinical practice (Helmerhorst HJ, et al. Ann Intensive Care. 2014;4:23).

Effects of hyperoxia in CNS and cardiovascular disease

The last 2 decades have seen several studies looking into the effects of hyperoxia in specific clinical scenarios. Arterial hyperoxia was found to be independently associated with in-hospital death in ventilated stroke patients in the ICU, as compared with either arterial normoxia or hypoxia (Rincon F, et al. Crit Care Med. 2014;42[2]:387). The AVOID trial showed that supplemental oxygen therapy in patients with ST-elevation myocardial infarction, but without hypoxia, increased early myocardial injury with risk of larger myocardial infarct size at 6 months. (Stub D, et al. Circulation. 2015;131[24]:2143).

Hyperoxia in the ICU

Although the potential risks of hyperoxia in conditions such as stroke and cardiac arrest had been observed, the jury was still out on its effects on a critically ill, mixed population, as routinely encountered in the ICU. Oxygen-ICU, a single center trial published in 2016, was one of the first looking at a mixed ICU population, while assessing the effects of a conservative oxygen delivery strategy against a conventional one (Girardis M, et al. JAMA. 2016;316[15]:1583). The researchers noted a significant mortality difference favoring conservative oxygen therapy, particularly in intubated patients. The IOTA group’s systematic review and meta-analysis of 16,000 patients, showed an increased relative risk of death in-hospital with hyperoxia, that persisted over a prolonged period while conferring no obvious advantages (Chu DK, et al. Lancet. 2018;391[10131]:1693).


Next Article: