Oxygen: Breath of Life or Kiss of Death*
Prior to the extensive use of pulse oximetry, hypoxemia and hyperoxemia could not be easily detected clinically and oxygen was used liberally in any potentially hypoxemic patient. However, in today’s ICUs, there is no reason to administer oxygen to patients with a PaO2 greater than 100 mm Hg or a pulse oximeter oxygen saturation (SpO2) greater than 97%. Oxygen therapy can be deleterious: it triggers formation of reactive oxygen species (4), induces hemodynamic and inflammatory changes (3, 5), contributes to tissue injury, and influences the progression or development of multiple system organ dysfunction (6). Hyperoxemia is associated with increased risk of death compared with normoxemia or hypoxemia following cardiopulmonary resuscitation (7). In a recent randomized controlled trial (RCT) in patients with acute myocardial infarction, routine oxygen administration was not associated with a reduction of symptoms; instead, oxygen supplementation was accompanied by a significant increase in biomarkers of myocardial damage and a larger infarct size (8). Furthermore, hyperoxia and hyperoxemia may promote lung injury during mechanical ventilation and have been linked to poor outcome in various subgroups (9). Despite these findings, no large RCTs have investigated the effects of different oxygenation targets in patients with acute respiratory failure managed with mechanical ventilation and supplemental oxygen.
Current guidelines recommend PaO2 levels (independent of the applied FIO2) of approximately 60–90 mm Hg and arterial oxygen saturation (SaO2) of 90–97%, but these target ranges are based on expert consensus more than on evidence from clinical studies. As a result, attitudes regarding the management of oxygen therapy vary considerably, and clinicians often consider hyperoxemia acceptable as long as the FIO2 is low. Evidence clearly indicates that hypoxemia, defined as a SaO2 less than 90% or a PaO2 less than 60 mm Hg, may predispose patients to cardiac arrest and death (10). However, we do not know whether there is a safe upper limit for PaO2, once a critically ill patient is resuscitated or during disease recovery.
Panwar et al (11) performed a pilot RCT in 103 critically ill patients mechanically ventilated for more than or equal to 24 hours to determine whether a conservative oxygenation strategy (target SpO2, 88–92%) was a feasible alternative to a liberal oxygenation strategy (target SpO2, ≥ 96%) and found that there was no evidence of any harm associated with a conservative target, as assessed by organ dysfunction and mortality. This pilot study was intended to inform the design of any subsequent larger RCTs on the use of conservative versus liberal oxygen therapy in mechanically ventilated patients.