Masaji Nishimura

doi:10.4266/acc.000312

Articles

Page Path: HOME > Acute Crit Care > Forthcoming articles > Article

Review Article Pulmonary Optimal oxygenation in the intensive care unit: what recent trials do—and do not—tell us: DOI: https://doi.org/10.4266/acc.000312
Published online: May 13, 2026

Masaji Nishimura

Aizenbashi Hospital, Osaka, Japan

Corresponding Author: Masaji Nishimura Aizenbashi Hospital, 5-16-15 Nipponbashi, Naniwa-ku, Osaka 565-0005, Japan Tel: +81-6-6633-2801 Email: kumorigarasu@gmail.com

• Received: December 24, 2025 • Revised: February 10, 2026 • Accepted: February 22, 2026

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

1,478 Views
58 Download

Full Article

Download PDF

Abstract
INTRODUCTION
EVIDENCE FROM RETROSPECTIVE STUDIES
EVIDENCE FROM RCTs
DISCUSSION
CONCLUSIONS
KEY MESSAGES
NOTES
REFERENCES

Abstract

Oxygen therapy is one of the most widely used interventions in critical care. Numerous observational studies and randomized controlled trials (RCTs) have evaluated conservative versus liberal oxygenation strategies, yet their results remain inconsistent, a pattern commonly observed in large clinical trials across multiple fields. Recent advances in intensive care unit practices further complicate the interpretation of prior findings, highlighting the need for a clearer framework to interpret oxygenation targets. This narrative review summarizes major observational studies, RCTs, and meta-analyses evaluating oxygenation strategies in critically ill patients. We examine variation in study designs, target and achieved oxygenation levels, and evolving critical care practices. Physiological principles, including oxygen delivery, oxygen utilization, and the oxyhemoglobin dissociation curve, are applied to interpret current evidence and define clinically meaningful safety margins, while acknowledging the limitations of interpreting heterogeneous trial and meta-analytic results. Recent RCTs have found relatively small differences in oxygenation between study groups; both conservative and liberal oxygenation targets are generally within a safe physiological range. However, evidence from the LOCO₂ trial indicates excessively low targets may be harmful, suggesting a safer lower limit of approximately peripheral oxygen saturation (SpO₂) 92%–93%. Retrospective studies indicate acceptable outcomes with arterial oxygen partial pressure (PaO₂) values up to approximately 120 mm Hg. Because the upper portion of the oxyhemoglobin dissociation curve is flat, SpO₂ cannot reliably detect hyperoxemia. Rather than determining the superiority of conservative or liberal oxygenation strategies, current evidence supports defining a practical therapeutic window. For most critically ill patients, an SpO₂ range of 92–97% likely represents a safe and clinically meaningful target.
Key Words: critical care; hyperoxemia; oxygen saturation targets; oxygen therapy

INTRODUCTION

The last universal common ancestor is thought to have arisen approximately 3.8 billion years ago, during a period when molecular oxygen was absent from Earth’s atmosphere [1]. Oxygen began to accumulate around 2.5 billion years ago, and molecular oxygen is inherently reactive and toxic. Approximately 2 billion years ago, an archaeal host cell engulfed an alpha-proteobacterium capable of aerobic respiration. The resulting endosymbiotic relationship gave rise to mitochondria, enabling eukaryotic cells to generate adenosine triphosphate far more efficiently through oxidative phosphorylation. Prior to its utilization in mitochondria, the reactivity of oxygen must be tightly regulated. Over time, oxygen-binding proteins such as hemoglobin and hemocyanin emerged, allowing for efficient and controlled transport of oxygen.

Joseph Priestley and Carl Wilhelm Scheele independently isolated oxygen by heating mercuric oxide [2]. Antoine Lavoisier, a French chemist, was the first to scientifically demonstrate its central role in respiration. More than a century later, medical journals began reporting the therapeutic use of oxygen [3]. Systems for controlled oxygen delivery were subsequently developed [4], and oxygen therapy became a first-line treatment for patients with acute illnesses, saving innumerable lives [5].

Hypoxemia (low oxygen level) is life-threatening, whereas hyperoxemia—although lacking a universally accepted definition—has traditionally been regarded as tolerable and clinically unimportant. Indeed, hyperoxemia remains common in clinical practice [6-9]. However, the potential hazards of oxygen administration have long been recognized. As early as 1917, John Scott Haldane cautioned that oxygen should be administered as sparingly as possible [4]. Furthermore, it is widely accepted that oxygen must be used judiciously in patients with chronic obstructive pulmonary disease [10]. It is not invariably beneficial, even in patients with acute myocardial infarction [11,12] or severe congestive heart failure [13], and hyperoxemia has been shown to be associated with adverse outcomes in stroke [14]. Furthermore, even short-term administration of high-concentration oxygen—such as during general anesthesia—can increase the risk of postoperative pulmonary complications [15,16]. Under 100% oxygen, absorption atelectasis develops rapidly [17,18] and increases the risk of pneumonia [19]. In one study of trauma patients, the high-oxygenation group exhibited a higher incidence of atelectasis than the normo-oxygenation group [20].

Oxygen administration is among the most widely employed medical therapies, and optimizing its use is essential for improving patient outcomes. Numerous clinical investigations have sought to determine the optimal strategy for oxygen therapy, including retrospective and prospective observational studies as well as large-scale randomized controlled trials (RCTs). However, the findings have been inconsistent.

Why have recent RCTs reported conflicting results? Why are oxygen targets still controversial, even after so many large-scale clinical trials? Several factors may contribute, including the inherent limitations of an RCT design in oxygenation studies, rather than flaws in methodological quality. First, the underlying diseases differ across studies. Second, target oxygenation levels vary. Third, the achieved oxygenation levels often deviate from the intended targets. In addition to these issues, real-world practice relies on arterial or peripheral oxygen saturation (SaO₂/SpO₂) and/or arterial oxygen partial pressure (PaO₂) as indicators of oxygenation. However, oxygen utilization is not always linearly related to these measurements. Hemoglobin concentration, cardiac output, and the oxyhemoglobin dissociation curve are also important determinants of oxygenation levels [21,22], yet these factors are often not assessed. Our aim in this manuscript was to examine the optimal approach to oxygen therapy by reviewing the evidence from these clinical trials. Because terminology varies across studies, we refer to the cohort receiving higher oxygenation as the high-oxygenation group, and to the more restrictive cohort as the conservative group.

EVIDENCE FROM RETROSPECTIVE STUDIES

de Jonge et al. [6] conducted a retrospective analysis of PaO₂ within the first 24 hours of intensive care unit (ICU) admission in more than 36,000 mechanically ventilated patients. Their findings indicated that both high fractions of inspired oxygen (FiO₂) and elevated PaO₂ levels were associated with increased mortality. Similarly, following resuscitation from cardiac arrest—whether occurring in-hospital or out-of-hospital—hyperoxemia has been associated with reduced survival and poor neurological outcomes [23].

The adverse effects of oxygen do not appear only beyond a discrete PaO₂ threshold; rather, they seem to increase continuously with the degree of hyperoxemia. As PaO₂ increases, the risks of mortality and functional impairment rise correspondingly [24]. In patients with chronic congestive heart failure, cardiac output tends to decrease as oxygenation levels increase [13].

These retrospective studies highlighted the potential harms of hyperoxemia, and motivated the transition to RCTs. However, many of these trials did not define an upper limit of oxygenation, and very high PaO₂ levels—often exceeding 300 mm Hg—were permitted. Additionally, illness severity was not controlled for. Because sicker patients typically receive more oxygen for a longer period of time, this time-dependent exposure represents a form of time-varying confounding and may have contributed to the observed association between higher oxygen exposure and increased mortality. These findings suggest an association between hyperoxemia and poor outcomes, but are limited by confounding factors related to disease severity and treatment indications.

EVIDENCE FROM RCTs

Acute Myocardial Infarction

Rawles et al. administered 6 L/min of gas to patients with acute myocardial infarction (AMI) using visually identical cylinders containing either air or oxygen to ensure blinding of both patients and clinical staff [12]. Serum aspartate aminotransferase concentrations were higher, and tachycardia occurred more frequently, in the high-oxygenation group. Mean PaO₂ was 178 mm Hg, which is considerably higher than in subsequent trials.

To eliminate potential confounding from pre-hospital oxygen exposure in patients with ST-elevation myocardial infarction (STEMI), Stub et al. initiated their intervention upon paramedic arrival [25]. In the conservative group, oxygen was withheld unless SpO₂ fell below 94%, whereas the high-oxygenation group received 8 L/min of oxygen. The high-oxygenation group exhibited higher rates of recurrent myocardial infarction and larger infarct volumes. The authors concluded that routine oxygen administration is not warranted in STEMI patients without hypoxemia. Median SpO₂ in the high-oxygenation group was 100%.

A subsequent large RCT followed patients for 1 year after randomization and found no significant differences between high-oxygenation and conservative oxygenation groups in all-cause mortality, cardiovascular mortality [26], inflammatory markers [27], or quality of life [28]. At around 12 hours after randomization, the average SpO₂ in the high-oxygenation group was 99% versus 97% in the conservative group.

Across these studies, oxygen was administered regardless of the presence or absence of hypoxemia, and oxygenation levels in the high-oxygenation groups were consistently elevated. Two studies demonstrated differences in morbidity between groups, whereas one study found no significant differences. Although all trials enrolled patients with AMI and used similar methods of oxygen delivery, the duration of oxygen administration, the timing of randomization (pre-hospital vs in-hospital), the definition of hypoxemia, and the timing for initiating oxygen therapy (90% or 94%) varied among studies. These factors may have contributed to the inconsistent results. Overall, oxygen administration does not appear to be beneficial in the absence of hypoxemia.

Stroke

Oxygen administration is common in patients with stroke. Roffe et al. administered oxygen at rates of 2–3 L/min, either continuously or only at night, for the first 3 days after stroke [29]. Because desaturation is usually detected at night and daytime oxygen administration may interfere with rehabilitation, the study compared these two approaches. The primary outcome was the Modified Rankin Scale score (mRS) at 90 days post-stroke. No differences were observed between the high-oxygenation groups—whether they received continuous oxygen or nighttime-only oxygen—compared to the group that did not receive any supplementary oxygen administration. There was also no difference in mRS between the two oxygen regimens. Similar to the findings in AMI trials, routine oxygen administration does not appear to be beneficial in the absence of hypoxemia.

The AMI/stroke findings, however, may have limited generalizability as the myocardium and brain are sensitive to the vasoconstrictive effects of hyperoxia and the pathophysiology of patients with AMI/stroke differs from that of general ICU patients.

Resuscitation after Cardiopulmonary Arrest

Although animal studies have reported that high FiO₂ levels and hyperoxemia after cardiopulmonary arrest are associated with poor outcomes [30,31], patients are often ventilated with 100% oxygen in clinical practice. However, clear evidence in humans is lacking [32,33]. In comatose patients who achieved return of spontaneous circulation after out-of-hospital cardiac arrest, two PaO₂ targets—9–10 kPa (68–75 mm Hg) and 13–14 kPa (98–105 mm Hg)—were randomly assigned [34]. The primary outcome was death or severe neurological impairment (Cerebral Performance Category 3 or 4) at 90 days or at hospital discharge. No significant difference was observed between these two groups.

Kilgannon et al. [23] examined the relationship between oxygenation levels and outcomes in observational studies. They defined hyperoxia as PaO₂ ≥300 mm Hg and found that hyperoxia was significantly associated with poor outcomes. In contrast, the RCTs compared relatively narrow oxygenation targets, both of which fell within the normoxemic range. This likely explains why the RCTs did not demonstrate significant differences in clinical outcomes.

Septic Shock And Severe Sepsis

In septic shock and severe sepsis, oxygen delivery to peripheral tissues is impaired, and a low PaO₂ level is considered problematic [35]. Oxygen administration constricts peripheral arteries, reduces the need for vasoactive agents [36], and is expected to enhance oxygen delivery [37]. In a study of patients with septic shock, the high-oxygenation group inhaled 100% oxygen for 24 hours [38]. Mortality at 28 and 90 days did not differ between the high-oxygenation and conservative groups. However, the incidence of atelectasis and ICU-acquired weakness was significantly higher in the high-oxygenation group than the conservative group. Because this trial used a markedly different approach—specifically, 100% oxygen for 24 hours—it is not appropriate to directly compare its findings with those of other clinical trials.

ICU Patients

Girardis et al. [39] assigned ICU patients to two groups. In the high-oxygenation group, PaO₂ was permitted to reach up to 150 mm Hg, SpO₂ was maintained at 97%–100%, and FiO₂ was kept above 0.4. In the conservative group, PaO₂ was maintained between 70–100 mm Hg, SpO₂ between 94%–98%, and FiO₂ was titrated to the lowest possible level. ICU mortality was significantly lower in the conservative group than in the high-oxygenation group (11.6% [25/216] vs. 20.2% [44/218], respectively).

The ICU-ROX trial enrolled nearly 1,000 patients [40]. In the conservative group, SpO₂ was maintained between 90%–97%. In contrast, the high-oxygenation group had no upper limit for SpO₂ or FiO₂, although FiO₂ was kept above 0.3. The number of ventilator-free days (VFD) at day 28 and mortality at day 180 did not differ significantly between the two groups.

The UK-ROX trial recruited more than 16,000 mechanically ventilated patients and assigned them to a conservative or high-oxygenation group [41]. In the conservative group, the target SpO₂ was 90% (range,88%–92%), whereas in the high-oxygenation group, oxygen levels were managed at the discretion of the attending physicians. Mortality at day 90 did not differ between the two groups. Other secondary outcomes—including ICU length of stay, hospital readmission, and life-support–free days at day 30—also showed no significant differences. As in previous studies, the difference in oxygenation levels between groups was minimal. Because clinicians are now more aware of the potential risks associated with high oxygen concentrations, large differences in oxygenation levels in contemporary trials are rare.

In the ICONIC trial, van der Wal et al. [42] noted that the difference in target oxygenation between groups was small, with the high-oxygenation group assigned a PaO₂ target of 110–150 mm Hg. During mechanical ventilation, the median PaO₂ in the high-oxygenation group was 115 mm Hg—slightly higher than in other recent trials—but after extubation it decreased to 85.5 mm Hg. Consequently, no significant differences were observed in VFD or mortality between the conservative and high-oxygenation groups. Another RCT also reported no significant differences in VFD or mortality between conservative and high-oxygenation groups [43]. In many of these studies, the separation in oxygenation levels between groups may not have been large enough to detect a meaningful effect.

Acute Hypoxemic Respiratory Failure/Acute Respiratory Distress Syndrome

The HOT-ICU trial similarly enrolled a large cohort of patients with acute hypoxemic respiratory failure [44]. The target PaO₂ was 90 mm Hg in the high-oxygenation group and 60 mm Hg in the conservative group. Mortality at day 90 did not differ between the groups. Secondary analyses focusing on patients with chronic obstructive pulmonary disease and hematologic malignancies likewise showed no significant differences.

The HOT-COVID trial, which followed the same protocol as the HOT-ICU trial, focused on patients with coronavirus disease 2019 (COVID-19) [45]. Unlike the other trials, its primary outcome was the number of life-support–free days at day 90, which was longer in the conservative group. Secondary outcomes—mortality at 90 days, the proportion of patients with serious adverse events or number of days alive and out of hospital—were also similar between the two groups.

In studies conducted by the ARDS Network on acute respiratory distress syndrome (ARDS), the target PaO₂ was set at 55–80 mm Hg. Reports have indicated that deviating from this protocol and permitting higher oxygenation levels may influence prognosis [46-48]. However, long-term complications such as central nervous system disorders are also a concern, and because hypoxemia is a known risk factor [49,50], indiscriminately aiming for low oxygenation is not beneficial. Barrot et al. compared target PaO₂ levels of 55–70 mm Hg (SpO₂ 88–92%) with 90–105 mm Hg (SpO₂ >96%). The 90-day mortality rate was significantly higher in the conservative group, leading to early termination of the study at the interim analysis [51]. Notably, intestinal ischemia occurred in 5% of patients in the conservative group, whereas no cases were reported in the high-oxygenation group.

In contrast to other trials, the LOCO₂ trial provides evidence of potential harm associated with oxygenation targets near the lower limit of the normoxemic range. Even with current standard care, including lung-protective strategies and prone ventilation, oxygenation targets at the lower boundary of normoxemia may result in inadequate oxygen delivery to peripheral tissues.

In the HOT-ICU trial, about 40% of patients were treated with open system or non-invasive positive pressure ventilation. Although the patients were hypoxemic, the results cannot be directly compared with those of the ARDS trials. Similar to other ICU studies, the separation of target and achieved oxygenation levels between the groups was not large enough to produce a clear difference in outcomes. However, the absence of significant differences in most of these trials should not be interpreted as evidence of no biological effect of oxygen administration.

Pediatric Critically Ill Patients

In pediatric patients, risks associated with hyperoxemia have also been reported [52]. In children with bronchiolitis, targets of SpO₂ 94% and 90% were compared [53]. A pulse oximeter with a special algorithm displayed an SpO₂ of 94% when the actual value was 90%. In both groups, oxygen was administered when the oximeter reading fell below 94%. However, in one group, oxygen was withheld unless the actual SpO₂ dropped below 90%. Although the number of children requiring oxygen was reduced in the conservative group, adverse events and the need for escalation of therapy did not differ between the groups.

In patients admitted to the pediatric ICU (PICU), target SpO₂ ranges of 88%–92% and >94% were compared [54]. The median achieved SpO₂ values were 94% (interquartile range [IQR], 93%–96%) and 97% (IQR, 96%–98%), respectively. Neither ICU mortality nor 30-day mortality differed between the groups. However, the number of life-support–free days at day 30 was lower in the high-oxygenation group.

In pediatric patients, long-term outcomes are particularly important. One study reported that neither mortality nor hospital readmission at 1 year differed between high-oxygenation and conservative oxygen groups, and quality of life was also similar [55]. Longer follow-up remains essential as children continue to grow, but this will need to be addressed in future studies.

DISCUSSION

Numerous clinical studies, reviews, and meta-analyses of various patient populations comparing the effects of different levels of oxygenation have been performed to date [31,56-77]. However, the results of RCTs have been inconsistent, and the conclusions of meta-analyses have also varied [52,56,58,60,61,75-77]. How these heterogeneous findings should be interpreted and applied to everyday clinical practice remains unclear. This challenge is not unique to oxygenation research.

Major clinical trials have varied substantially in their enrolled populations, primary and secondary outcomes, target oxygenation ranges, achieved oxygenation levels, and other factors (Tables 1 and 2). Mortality was the most common primary outcome, but the time point at which mortality was assessed has varied widely—28 days, 30 days, 90 days, 180 days, or one year. Several trials instead used life-support-free days as the primary outcome [40,45,54]. This composite measure typically includes ventilator support, cardiovascular support, and renal replacement therapy, with ventilator dependence being the dominant contributor [45,54]. Patients in the high-oxygenation group were maintained on mechanical ventilation at oxygenation levels at which, in the conservative group, they would have been extubated. This difference in extubation thresholds is unlikely to be the sole explanation for outcome differences between groups, but it may have introduced protocol-driven effects that obscured true differences in clinical benefit.

Another challenge is that recent large-scale studies have shown relatively small differences in oxygenation levels between high-oxygenation and conservative oxygenation groups. Moreover, in the high-oxygenation groups, both the target and the achieved oxygenation levels remained within the normoxemic range rather than reaching true hyperoxemia [41,43,51]. This may explain why studies were unable to demonstrate the adverse effects of hyperoxemia; instead, these studies have offered insights into the safe upper limit of oxygenation.

Oxygen utilization depends not only on oxygen delivery—which is determined by hemoglobin concentration, the amount of hemoglobin saturated with oxygen (SaO₂), PaO₂, and cardiac output—but also on the position of the oxyhemoglobin dissociation curve (Figures 1 and 2). In particular, oxygen release from hemoglobin is significantly dependent on the position of the oxyhemoglobin curve. When the curve is shifted to the right, more oxygen is released in the peripheral tissue. Its effect is demonstrated in inherited enzyme defects of the red cell [21,22]. In most clinical trials, however, measurements are typically limited to SpO₂ and/or PaO₂, neither of which directly reflects tissue-level oxygen utilization. Furthermore, oxygen utilization does not change linearly with SpO₂ and PaO₂. Although an SpO₂ target of 90% is commonly used in conservative oxygenation strategies, findings from the LOCO₂ trial—where mortality was higher in the conservative cohort [51]—suggest that an SpO₂ of approximately 92%–93% may represent a safer lower limit, particularly in patients with ARDS.

For the upper safety margin of oxygenation, both pulmonary and peripheral tissue oxygen toxicity must be considered. If pulmonary oxygen toxicity is prioritized, FiO₂ should be kept as low as possible, provided that oxygen utilization remains adequate. Pulmonary oxygen toxicity has been thoroughly investigated in laboratory animals [78-80]. Exposure to high FiO₂ (>0.7) for several days causes death from respiratory failure [81,82]. Human studies have suggested that 50% oxygen does not cause severe pulmonary damage in healthy volunteers for up to 24 hours [83,84]. Although evidence remains incomplete, human lungs may be reasonably tolerant to oxygen toxicity when FiO₂ is kept below approximately 0.5. In recent RCTs involving mechanically ventilated patients, FiO₂ levels were generally maintained below 0.5 in both the high-oxygenation group and the conservative group. Pulmonary oxygen toxicity was therefore not a primary concern in these clinical trials.

Conversely, if peripheral tissue toxicity is considered more important, peripheral hyperoxia may cause tissue injury through the generation of reactive oxygen species. As described above, oxygen release from hemoglobin is significantly dependent on the position of the oxyhemoglobin curve, which may influence not only the lower safety margin but higher safety margin. However, given that tissue oxygen tension cannot be measured in routine practice, SaO₂ and PaO₂ serve as indirect surrogates. Retrospective studies have suggested that the lowest mortality occurs around a PaO₂ values of 70 – 120 mm Hg, with detrimental effects observed both at lower and at higher levels [6]. Girardis et al. [39] allowed PaO₂ values up to 150 mm Hg in the higher-oxygenation group, and found that ICU mortality was significantly increased in this group. Taken together, available evidence suggests that PaO₂ values up to approximately 120 mm Hg are acceptable.

Clinical Implications

In routine clinical practice, oxygenation is monitored almost exclusively with pulse oximetry. Recent large RCTs have indicated that both conservative oxygenation targets (SpO₂ 88%–92%) and more liberal targets (SpO₂ 96%–98%) are generally safe. Modern pulse oximeters are accurate and widely available, although they are unreliable in specific circumstances—such as carbon monoxide poisoning, methemoglobinemia, severe vasoconstriction, shivering, or interference from nail polish. Awareness of these limitations is essential. Based on physiological considerations and the available clinical evidence, an SpO₂ of approximately 92%–93% appears to be a reasonable lower safety threshold. For the upper margin, pulmonary oxygen toxicity is unlikely when FiO₂ remains below 0.5, and retrospective analyses suggest that PaO₂ levels up to 120 mm Hg are acceptable. However, because the upper portion of the oxyhemoglobin dissociation curve is flat, SpO₂ cannot reliably distinguish between moderate and marked hyperoxemia. Thus, maintaining SpO₂ at or below approximately 97% is a practical and safe upper limit in daily practice.

CONCLUSIONS

Rather than asking whether conservative or liberal oxygenation is superior, it may be more appropriate to recognize that both strategies—as implemented in recent trials—fall within a relatively safe physiological range. The repeated inability of conservative oxygenation strategies to demonstrate superiority likely reflects this therapeutic window. While recent trials have not been able to demonstrate the superiority of conservative or liberal oxygen therapy, their findings suggest that for the vast majority of critically ill patients, an SpO₂ range of 92%–97% likely represents a safe and clinically meaningful target. Determining the optimal oxygenation range for specific underlying pathologies remains an important area for future research.

KEY MESSAGES

▪ Recent randomized controlled trials of oxygenation strategies have yielded inconsistent results.

▪ Excessively low oxygenation targets may be harmful, whereas marked hyperoxemia also carries potential risks.

▪ Rather than determining the superiority of conservative or liberal strategies, current evidence supports a practical therapeutic window for oxygenation corresponding to a peripheral oxygen saturation (SpO₂) in the range of 92%–97% for most critically ill patients.

NOTES

CONFLICT OF INTEREST

No potential conflict of interest relevant to this article was reported.

FUNDING

None.

ACKNOWLEDGMENTS

None.

AUTHOR CONTRIBUTIONS

All the work was done by Masaji Nishimura.

Figure 1.

Oxygen utility. Oxygen delivery depends on cardiac output and arterial oxygen content (CaO₂). CaO₂ is determined by hemoglobin concentration, arterial oxygen saturation (SaO₂), and the partial pressure of arterial oxygen (PaO₂). Although PaO₂ contributes to CaO₂, SaO₂ is the major determinant under physiological conditions. Oxygen utility is dependent on hemoglobin affinity determined by the environment of peripheral tissue. In clinical practice, we measure SaO₂(or SpO₂)/PaO₂, but these values do not directly reflect tissue-level oxygen utilization. Numerous physiological factors influence peripheral oxygen extraction, and understanding these relationships is essential for interpreting oxygenation parameters in real-world settings.

Figure 2.

Oxyhemoglobin dissociation curve. The oxyhemoglobin dissociation curve illustrates the nonlinear relationship between partial pressure of arterial oxygen (PaO₂) and arterial oxygen saturation (SaO₂). Rightward and leftward shifts reflect changes in hemoglobin affinity due to factors such as pH, temperature, and PCO₂. In anemia, the curve typically shifts to the right, maintaining oxygen utility without increasing cardiac output. The position of the curve is influenced by numerous physiological variables. Although PaO₂ and SO₂ are the primary determinants of peripheral oxygen availability, the actual amount of molecular oxygen released to tissues depends heavily on the position of the oxyhemoglobin dissociation curve. In clinical practice, modern blood gas analyzers can provide the P50 value (the PaO₂ at which hemoglobin is 50% saturated); however, neither this value nor hemoglobin levels and cardiac output have been widely evaluated in randomized controlled trials.

Table 1.

Major randomized controlled trials comparing conservative and liberal oxygenation strategies Enrolled patients and primary outcomes of each study

Study	Trial name	Enrolled patients	Primary outcome
Stub et al. [25]	AVOID	STEMI	Size of myocardial infarction
Hofmann et al. [26]	DETO2X-AMI	AMI	365-Day mortality
Roffe et al. [29]	-	Stroke	Neurological outcome at day 90
Schmidt et al. [34]	BOX	Out-of-hospital CPA	Mortality & severe CNS disorders at day 90
Asfar et al. [38]	HYPERS2S	Septic shock/severe sepsis	28-Day mortality
Arleth et al. [20]	TRAUMOX2	Trauma	Death and/or major respiratory complications within 30 days
Girardis et al. [39]	-	ICU patients	ICU mortality
Mackle et al. [40]	ICU-ROX	MV patients in ICU	VFD until day 28
van der Wal et al. [42]	ICONIC	MV patients in ICU	28-Day mortality
Schjørring et al. [44]	HOT-ICU	AHRF	90-Day mortality
Barrot et al. [51]	LOCO₂	ARDS	28-Day mortality
Peters et al. [54]	Oxy-PICU	PICU patients	Duration of oxygen support at day 30
Martin et al. [41]	UK-ROX	MV patients in ICU	90-Day mortality

STEMI: ST-elevated myocardial infarction; AMI: acute myocardial infarction; CPA: cardio-pulmonary arrest; CNS: central nervous system; ICU: intensive care unit; MV: mechanically ventilated; VFD: ventilator-free day; AHRF: acute hypoxemic respiratory failure; ARDS: acute respiratory distress syndrome; PICU: pediatric intensive care unit.

Table 2.

Summary of oxygenation targets and achieved oxygenation levels in key clinical trials

Study	Lower target	Higher target	Lower achievement	Higher achievement
Stub et al. [25]	SpO₂ ≥94%	O₂ mask 8L/min	98.0% (IQR, 96.0%–99.0%)	100.0% (IQR, 99.0%–100.0%)
Hofmann [26]	SpO₂ ≥90%	O₂ mask 6L/min	97% (IQR, 95%–98%)	99% (IQR, 97%–100%)
Roffe et al. [29]	No oxygen	n-cannula 2 or 3 L/min	-	-
Schmidt et al. [34]	PaO₂ 68–75 mm Hg	PaO₂ 98–105 mm Hg	-	-
Asfar et al. [38]	SaO₂ 88%–95%	F_IO₂ 1.0	-	-
Arleth et al. [20]	SpO₂ 94%	15 L/min or FiO₂ 1.0	97% (IQR, 96%–99%)	100% (IQR, 99–100)
Girardis et al. [39]	PaO₂ 70–100 mm Hg	PaO₂ up to 150 mm Hg	87 mm Hg (IQR, 79–97)	102 mm Hg (IQR, 88–116)
Mackle et al. [40]	91%≤ SpO₂<97%	91% ≤SpO₂	F_iO₂ 0.21, 1 hr (IQR, 0–17)	F_IO₂ 0.21, 27 hr (IQR, 5–78)
van der Wal et al. [42]	PaO₂ 55–80 mm Hg	PaO₂ 110–150 mm Hg	75 mm Hg (IQR, 69.8–83.5)	115 mm Hg (IQR, 100.3–129.0)
Schjørring et al. [44]	PaO₂ 60 mm Hg	PaO₂ 90 mm Hg	70.8 mm Hg (IQR, 66.6–76.5)	93.3 mm Hg (IQR, 87.1–98.7)
Barrot et al. [51]	PaO₂ 55–70 mm Hg	PaO₂ 90–105 mm Hg	PaO₂<55 mm Hg: 58 patients, >70 mm Hg: 97 patients	PaO₂<90 mm Hg: all patients, >105 mm Hg: 98 patients
Peters et al. [54]	SpO₂ 88%–92%	SpO₂ ≥94%	94% (IQR, 93–96)	97% (IQR, 96–98)
Martin et al. [41]	SpO₂ 88%–92%	Up to attending physician	93.3% (2.8%), 71.5 mm Hg (13.9)	95.1% (2.4%), 79.5 mm Hg (17.9)

SpO₂: peripheral oxygen saturation; IQR: interquartile range; PaO₂: arterial oxygen partial pressure; SaO₂: arterial oxygen saturation; F_IO₂: fraction of inspired oxygen.

REFERENCES

1. Hsia CC, Schmitz A, Lambertz M, Perry SF, Maina JN. Evolution of air breathing: oxygen homeostasis and the transitions from water to land and sky. Compr Physiol 2013;3:849-915.Article PubMed PMC PDF
2. Hancock JT. A brief history of oxygen: 250 years on. Oxygen 2022;2:31-9.Article
3. Blodgett AN. The continuous inhalation of oxygen in cases of pneumonia otherwise fatal, and in other diseases. Med Surg J 1890;123:481-5.Article
4. Haldane JS. The therapeutic administration of oxygen. Br Med J 1917;1:181-3.Article PubMed PMC
5. Wemple ML, Swenson KE, Swenson ER. Oxygen therapy part 2 - indications and toxicity. NEJM Evid 2023;2:EVIDra2300111.Article PubMed
6. de Jonge E, Peelen L, Keijzers PJ, Joore H, de Lange D, van der Voort PH, et al. Association between administered oxygen, arterial partial oxygen pressure and mortality in mechanically ventilated intensive care unit patients. Crit Care 2008;12:R156.Article PubMed PMC
7. Egi M, Kataoka J, Ito T, Nishida O, Yasuda H, Okamaoto H, et al. Oxygen management in mechanically ventilated patients: a multicenter prospective observational study. J Crit Care 2018;46:1-5.Article PubMed
8. Itagaki T, Nakano Y, Okuda N, Izawa M, Onodera M, Imanaka H, et al. Hyperoxemia in mechanically ventilated, critically ill subjects: incidence and related factors. Respir Care 2015;60:335-40.Article PubMed PDF
9. Madotto F, Rezoagli E, Pham T, Schmidt M, McNicholas B, Protti A, et al. Hyperoxemia and excess oxygen use in early acute respiratory distress syndrome: insights from the LUNG SAFE study. Crit Care 2020;24:125.Article PubMed PMC
10. Massaro DJ, Katz S, Luchsinger PC. Effect of various modes of oxygen administration on the arterial gas values in patients with respiratory acidosis. Br Med J 1962;2:627-9.Article PubMed PMC
11. Maroko PR, Radvany P, Braunwald E, Hale SL. Reduction of infarct size by oxygen inhalation following acute coronary occlusion. Circulation 1975;52:360-8.Article PubMed
12. Rawles JM, Kenmure AC. Controlled trial of oxygen in uncomplicated myocardial infarction. Br Med J 1976;1:1121-3.Article PubMed PMC
13. Haque WA, Boehmer J, Clemson BS, Leuenberger UA, Silber DH, Sinoway LI, et al. Hemodynamic effects of supplemental oxygen administration in congestive heart failure. J Am Coll Cardiol 1996;27:353-7.Article PubMed
14. Rincon F, Kang J, Maltenfort M, Vibbert M, Urtecho J, Athar MK, et al. Association between hyperoxia and mortality after stroke: a multicenter cohort study. Crit Care Med 2014;42:387-96.Article PubMed
15. Kotani N, Hashimoto H, Sessler DI, Muraoka M, Hashiba E, Kubota T, et al. Supplemental intraoperative oxygen augments antimicrobial and proinflammatory responses of alveolar macrophages. Anesthesiology 2000;93:15-25.Article PubMed PDF
16. Müller-Wirtz LM, Lilien TA, Patterson WM, Ott S, Francis RCE, Gama de Abreu M, et al. Individual and combined effects of chemical and mechanical power on postoperative pulmonary complications: a secondary analysis of the REPEAT study. Anaesthesia 2025;80:1510-8.Article PubMed PMC
17. Hedenstierna G. Oxygen and anesthesia: what lung do we deliver to the post-operative ward? Acta Anaesthesiol Scand 2012;56:675-85.Article PubMed
18. Rothen HU, Sporre B, Engberg G, Wegenius G, Högman M, Hedenstierna G, et al. Influence of gas composition on recurrence of atelectasis after a reexpansion maneuver during general anesthesia. Anesthesiology 1995;82:832-42.Article PubMed PDF
19. Six S, Jaffal K, Ledoux G, Jaillette E, Wallet F, Nseir S, et al. Hyperoxemia as a risk factor for ventilator-associated pneumonia. Crit Care 2016;20:195.Article PubMed PMC
20. Arleth T, Baekgaard J, Siersma V, Creutzburg A, Dinesen F, Rosenkrantz O, et al. Early restrictive vs liberal oxygen for trauma patients: the TRAUMOX2 randomized clinical trial. JAMA 2025;333:479-89.Article PubMed PMC
21. Delivoria-Papadopoulos M, Oski FA, Gottlieb AJ. Oxygen-hemoglobulin dissociation curves: effect of inherited enzyme defects of the red cell. Science 1969;165:601-2.Article PubMed
22. Oski FA, Marshall BE, Cohen PJ, Sugerman HJ, Miller LD. The role of the left-shifted or right-shifted oxygen-hemoglobin equilibrium curve. Ann Intern Med 1971;74:44-6.Article PubMed PDF
23. Kilgannon JH, Jones AE, Shapiro NI, Angelos MG, Milcarek B, Hunter K, et al. Association between arterial hyperoxia following resuscitation from cardiac arrest and in-hospital mortality. JAMA 2010;303:2165-71.Article PubMed
24. Kilgannon JH, Jones AE, Parrillo JE, Dellinger RP, Milcarek B, Hunter K, et al. Relationship between supranormal oxygen tension and outcome after resuscitation from cardiac arrest. Circulation 2011;123:2717-22.Article PubMed
25. Stub D, Smith K, Bernard S, Nehme Z, Stephenson M, Bray JE, et al. Air versus oxygen in ST-segment-elevation myocardial infarction. Circulation 2015;131:2143-50.Article PubMed
26. Hofmann R, James SK, Jernberg T, Lindahl B, Erlinge D, Witt N, et al. Oxygen therapy in suspected acute myocardial infarction. N Engl J Med 2017;377:1240-9.Article PubMed
27. Hofmann R, Tornvall P, Witt N, Alfredsson J, Svensson L, Jonasson L, et al. Supplemental oxygen therapy does not affect the systemic inflammatory response to acute myocardial infarction. J Intern Med 2018;283:334-45.Article PubMed PDF
28. Hofmann R, Befekadu Abebe T, Herlitz J, James SK, Erlinge D, Yndigegn T, et al. Routine oxygen therapy does not improve health-related quality of life in patients with acute myocardial infarction-insights from the randomized DETO₂X-AMI trial. Front Cardiovasc Med 2021;8:638829.Article PubMed PMC
29. Roffe C, Nevatte T, Sim J, Bishop J, Ives N, Ferdinand P, et al. Effect of routine low-dose oxygen supplementation on death and disability in adults with acute stroke: the stroke oxygen study randomized clinical trial. JAMA 2017;318:1125-35.Article PubMed PMC
30. Balan IS, Fiskum G, Hazelton J, Cotto-Cumba C, Rosenthal RE. Oximetry-guided reoxygenation improves neurological outcome after experimental cardiac arrest. Stroke 2006;37:3008-13.Article PubMed PMC
31. Pilcher J, Weatherall M, Shirtcliffe P, Bellomo R, Young P, Beasley R, et al. The effect of hyperoxia following cardiac arrest: a systematic review and meta-analysis of animal trials. Resuscitation 2012;83:417-22.Article PubMed
32. Bellomo R, Bailey M, Eastwood GM, Nichol A, Pilcher D, Hart GK, et al. Arterial hyperoxia and in-hospital mortality after resuscitation from cardiac arrest. Crit Care 2011;15:R90.Article PubMed PMC PDF
33. Crescioli E, Lass Klitgaard T, Perner A, Lilleholt Schjørring O, Steen Rasmussen B. Lower versus higher oxygenation targets in hypoxaemic ICU patients after cardiac arrest. Resuscitation 2023;188:109838.Article PubMed
34. Schmidt H, Kjaergaard J, Hassager C, Mølstrøm S, Grand J, Borregaard B, et al. Oxygen targets in comatose survivors of cardiac arrest. N Engl J Med 2022;387:1467-76.Article PubMed
35. Hyun DG, Ahn JH, Huh JW, Hong SB, Koh Y, Oh DK, et al. The association of arterial partial oxygen pressure with mortality in critically ill sepsis patients: a nationwide observational cohort study. Crit Care 2024;28:187.Article PubMed PMC PDF
36. Klitgaard TL, Schjørring OL, Lange T, Møller MH, Perner A, Rasmussen BS, et al. Lower versus higher oxygenation targets in critically ill patients with severe hypoxaemia: secondary Bayesian analysis to explore heterogeneous treatment effects in the Handling Oxygenation Targets in the Intensive Care Unit (HOT-ICU) trial. Br J Anaesth 2022;128:55-64.Article PubMed
37. Ghazaly HF, Aly AA, Tammam AS, Hassan MM, Hammad SS, Mahmoud NM, et al. Influence of liberal versus conservative oxygen therapies on the hemodynamic parameters of mechanically ventilated patients with sepsis: a randomized clinical trial. BMC Anesthesiol 2024;24:469.Article PubMed PMC PDF
38. Asfar P, Schortgen F, Boisramé-Helms J, Charpentier J, Guérot E, Megarbane B, et al. Hyperoxia and hypertonic saline in patients with septic shock (HYPERS2S): a two-by-two factorial, multicentre, randomised, clinical trial. Lancet Respir Med 2017;5:180-90.Article PubMed
39. Girardis M, Busani S, Damiani E, Donati A, Rinaldi L, Marudi A, et al. Effect of conservative vs conventional oxygen therapy on mortality among patients in an intensive care unit: the oxygen-ICU randomized clinical trial. JAMA 2016;316:1583-9.Article PubMed
40. Mackle D, Bellomo R, Bailey M, Beasley R, Deane A, Eastwood G, et al. Conservative oxygen therapy during mechanical ventilation in the ICU. N Engl J Med 2020;382:989-98.Article PubMed
41. Martin DS, Gould DW, Shahid T, Doidge JC, Cowden A, Sadique Z, et al. Conservative oxygen therapy in mechanically ventilated critically ill adult patients: The UK-ROX randomized clinical trial. JAMA 2025;334:398-408.Article PubMed PMC
42. van der Wal LI, Grim CC, Del Prado MR, van Westerloo DJ, Boerma EC, Rijnhart-de Jong HG, et al. Conservative versus liberal oxygenation targets in intensive care unit patients (ICONIC): a randomized clinical trial. Am J Respir Crit Care Med 2023;208:770-9.Article PubMed PMC PDF
43. Semler MW, Casey JD, Lloyd BD, Hastings PG, Hays MA, Stollings JL, et al. Oxygen-saturation targets for critically ill adults receiving mechanical ventilation. N Engl J Med 2022;387:1759-69.Article PubMed PMC
44. Schjørring OL, Klitgaard TL, Perner A, Wetterslev J, Lange T, Siegemund M, et al. Lower or higher oxygenation targets for acute hypoxemic respiratory failure. N Engl J Med 2021;384:1301-11.Article PubMed
45. Nielsen FM, Klitgaard TL, Siegemund M, Laake JH, Thormar KM, Cole JM, et al. Lower vs higher oxygenation target and days alive without life support in COVID-19: The HOT-COVID randomized clinical trial. JAMA 2024;331:1185-94.Article PubMed PMC
46. Boyle AJ, Holmes DN, Hackett J, Gilliland S, McCloskey M, O'Kane CM, et al. Hyperoxaemia and hypoxaemia are associated with harm in patients with ARDS. BMC Pulm Med 2021;21:285.Article PubMed PMC PDF
47. Dorado JH, Pérez J, Navarro E, Gogniat E, Torres S, Cagide S, et al. Impact of liberal versus conservative saturation targets on gas exchange indices in COVID-19 related acute respiratory distress syndrome: a physiological study. Rev Bras Ter Intensiva 2021;33:537-43.Article PubMed
48. Aggarwal NR, Brower RG, Hager DN, Thompson BT, Netzer G, Shanholtz C, et al. Oxygen exposure resulting in arterial oxygen tensions above the protocol goal was associated with worse clinical outcomes in acute respiratory distress syndrome. Crit Care Med 2018;46:517-24.Article PubMed PMC
49. Hopkins RO, Weaver LK, Pope D, Orme JF, Bigler ED, Larson-LOHR V, et al. Neuropsychological sequelae and impaired health status in survivors of severe acute respiratory distress syndrome. Am J Respir Crit Care Med 1999;160:50-6.Article PubMed PDF
50. Mikkelsen ME, Christie JD, Lanken PN, Biester RC, Thompson BT, Bellamy SL, et al. The adult respiratory distress syndrome cognitive outcomes study: long-term neuropsychological function in survivors of acute lung injury. Am J Respir Crit Care Med 2012;185:1307-15.Article PubMed PMC
51. Barrot L, Asfar P, Mauny F, Winiszewski H, Montini F, Badie J, et al. Liberal or conservative oxygen therapy for acute respiratory distress syndrome. N Engl J Med 2020;382:999-1008.Article PubMed
52. Lilien TA, Groeneveld NS, van Etten-Jamaludin F, Peters MJ, Buysse CM, Ralston SL, et al. Association of arterial hyperoxia with outcomes in critically ill children: a systematic review and meta-analysis. JAMA Netw Open 2022;5:e2142105. Article PubMed PMC
53. Cunningham S, Rodriguez A, Adams T, Boyd KA, Butcher I, Enderby B, et al. Oxygen saturation targets in infants with bronchiolitis (BIDS): a double-blind, randomised, equivalence trial. Lancet 2015;386:1041-8.Article PubMed PMC
54. Peters MJ, Gould DW, Ray S, Thomas K, Chang I, Orzol M, et al. Conservative versus liberal oxygenation targets in critically ill children (Oxy-PICU): a UK multicentre, open, parallel-group, randomised clinical trial. Lancet 2024;403:355-64.Article PubMed
55. Moler-Zapata S, Peters MJ, Gould DW, Giallongo E, Orzol M, Ray S, et al. Longer-term survival, quality of life, and cost-effectiveness of conservative versus liberal oxygenation targets in critically ill children: a pre-specified analysis from Oxy-PICU, a multicentre, open, parallel-group, randomised controlled trial. Lancet Child Adolesc Health 2025;9:16-24.Article PubMed
56. Barbateskovic M, Schjørring OL, Krauss SR, Meyhoff CS, Jakobsen JC, Rasmussen BS, et al. Higher vs lower oxygenation strategies in acutely ill adults: a systematic review with meta-analysis and trial sequential analysis. Chest 2021;159:154-73.Article PubMed
57. Barbateskovic M, Schjørring OL, Russo Krauss S, Jakobsen JC, Meyhoff CS, Dahl RM, et al. Higher versus lower fraction of inspired oxygen or targets of arterial oxygenation for adults admitted to the intensive care unit. Cochrane Database Syst Rev 2019;2019:CD012631.Article PubMed PMC
58. Cao L, Chen Q, Xiang YY, Xiao C, Tan YT, Li H, et al. Effects of oxygenation targets on mortality in critically ill patients in intensive care units: a systematic review and meta-analysis. Anesth Analg 2024;139:734-42.Article PubMed
59. Chen XL, Zhang BL, Meng C, Huang HB, Du B. Conservative oxygen therapy for critically ill patients: a meta-analysis of randomized controlled trials. J Intensive Care 2021;9:47.Article PubMed PMC PDF
60. Chu DK, Kim LH, Young PJ, Zamiri N, Almenawer SA, Jaeschke R, et al. Mortality and morbidity in acutely ill adults treated with liberal versus conservative oxygen therapy (IOTA): a systematic review and meta-analysis. Lancet 2018;391:1693-705.Article PubMed
61. Cumpstey AF, Oldman AH, Martin DS, Smith A, Grocott MPW. Oxygen targets during mechanical ventilation in the ICU: a systematic review and meta-analysis. Crit Care Explor 2022;4:e0652. Article PubMed PMC
62. Dong WH, Yan WQ, Chen Z. Effect of liberal or conservative oxygen therapy on the prognosis for mechanically ventilated intensive care unit patients: a meta-analysis. Sao Paulo Med J 2022;140:463-73.Article PubMed PMC
63. Klitgaard TL, Schjørring OL, Nielsen FM, Meyhoff CS, Barbateskovic M, Wetterslev J, et al. Letter the editor: serious methodological concerns about a recently published meta-analysis on oxygen therapy. J Intensive Care 2021;9:72.Article PubMed PMC PDF
64. Klitgaard TL, Schjørring OL, Nielsen FM, Meyhoff CS, Perner A, Wetterslev J, et al. Higher versus lower fractions of inspired oxygen or targets of arterial oxygenation for adults admitted to the intensive care unit. Cochrane Database Syst Rev 2023;9:CD012631.Article PubMed PMC
65. Li L, Zhang Y, Wang P, Chong W, Hai Y, Xu P, et al. Conservative versus liberal oxygen therapy for acutely ill medical patients: a systematic review and meta-analysis. Int J Nurs Stud 2021;118:103924.Article PubMed
66. Li X, Liu D, Liu C, Mao Z, Liu Y, Yi H, et al. Conservative versus liberal oxygen therapy in relation to all-cause mortality among patients in the intensive care unit: a systematic review of randomized controlled trials with meta-analysis and trial sequential analysis. Med Intensiva (Engl Ed) 2023;47:73-83.Article PubMed
67. Li XY, Dai B, Hou HJ, Zhao HW, Wang W, Kang J, et al. Conservative versus liberal oxygen therapy for intensive care unit patients: meta-analysis of randomized controlled trials. Ann Intensive Care 2024;14:68.Article PubMed PMC PDF
68. Liu L, Tian Y. Liberal or conservative oxygen therapy for ventilated patients in the ICU: a meta-analysis of randomized controlled trials. J Cardiothorac Surg 2021;16:261.Article PubMed PMC PDF
69. Martin DS, Mckenna HT, Rowan KM, Gould DW, Mouncey PR, Grocott MP, et al. The effect of conservative oxygen therapy on mortality in adult critically ill patients: a systematic review and meta-analysis of randomised controlled trials. J Intensive Care Soc 2023;24:399-408.Article PubMed PMC PDF
70. Nielsen FM, Klitgaard TL, Bruun NH, Møller MH, Schjørring OL, Rasmussen BS, et al. Lower or higher oxygenation targets for acute Hypoxaemic respiratory failure: Protocol for an individual patient data meta-analysis. Acta Anaesthesiol Scand 2023;67:811-9.Article PubMed PDF
71. Nielsen FM, Klitgaard TL, Bruun NH, Møller MH, Schjørring OL, Rasmussen BS, et al. Lower or higher oxygenation targets in the intensive care unit: an individual patient data meta-analysis. Intensive Care Med 2024;50:1275-86.Article PubMed PMC PDF
72. Patel JK, Kataya A, Parikh PB. Association between intra- and post-arrest hyperoxia on mortality in adults with cardiac arrest: a systematic review and meta-analysis. Resuscitation 2018;127:83-8.Article PubMed
73. Young PJ, Bailey M, Bellomo R, Bernard S, Bray J, Jakkula P, et al. Conservative or liberal oxygen therapy in adults after cardiac arrest: An individual-level patient data meta-analysis of randomised controlled trials. Resuscitation 2020;157:15-22.Article PubMed
74. Zhao X, Xiao H, Dai F, Brodie D, Meng L. Classification and effectiveness of different oxygenation goals in mechanically ventilated critically ill patients: network meta-analysis of randomised controlled trials. Eur Respir J 2021;58:2002928.Article
75. Crescioli E, Krejberg KU, Klitgaard TL, Nielsen FM, Barbateskovic M, Skrubbeltrang C, et al. Long-term effects of lower versus higher oxygenation levels in adult ICU patients-a systematic review. Acta Anaesthesiol Scand 2022;66:910-22.Article PubMed PMC PDF
76. Damiani E, Adrario E, Girardis M, Romano R, Pelaia P, Singer M, et al. Arterial hyperoxia and mortality in critically ill patients: a systematic review and meta-analysis. Crit Care 2014;18:711.Article PubMed PMC PDF
77. van der Wal LI, Grim CC, van Westerloo DJ, Schultz MJ, de Jonge E, Helmerhorst HJ, et al. Higher versus lower oxygenation strategies in the general intensive care unit population: a systematic review, meta-analysis and meta-regression of randomized controlled trials. J Crit Care 2022;72:154151.Article PubMed
78. de los Santos R, Seidenfeld JJ, Anzueto A, Collins JF, Coalson JJ, Johanson WG, et al. One hundred percent oxygen lung injury in adult baboons. Am Rev Respir Dis 1987;136:657-61.Article PubMed PDF
79. Frank L, Bucher JR, Roberts RJ. Oxygen toxicity in neonatal and adult animals of various species. J Appl Physiol Respir Environ Exerc Physiol 1978;45:699-704.Article PubMed
80. Robinson FR, Casey HW, Weibel ER. Animal model: oxygen toxicity in nonhuman primates. Am J Pathol 1974;76:175-8.PubMed PMC
81. Clark JM, Lambertsen CJ. Pulmonary oxygen toxicity: a review. Pharmacol Rev 1971;23:37-133.Article PubMed
82. Kallet RH, Matthay MA. Hyperoxic acute lung injury. Respir Care 2013;58:123-41.Article PubMed PMC
83. Caldwell PR, Lee WL, Schildkraut HS, Archibald ER. Changes in lung volume, diffusing capacity, and blood gases in men breathing oxygen. J Appl Physiol 1966;21:1477-83.Article PubMed
84. Davis WB, Rennard SI, Bitterman PB, Crystal RG. Pulmonary oxygen toxicity. Early reversible changes in human alveolar structures induced by hyperoxia. N Engl J Med 1983;309:878-83.Article PubMed

Figure & Data

References

Citations

Citations to this article as recorded by

PubReader
ePub Link

Cite

CITE: export Copy Format; Close

Download Citation

Download a citation file in RIS format that can be imported by all major citation management software, including EndNote, ProCite, RefWorks, and Reference Manager.

Format:

RIS — For EndNote, ProCite, RefWorks, and most other reference management software
BibTeX — For JabRef, BibDesk, and other BibTeX-specific software

Include:

Citation for the content below
Citation and abstract for the content below

Optimal oxygenation in the intensive care unit: what recent trials do—and do not—tell us

DOI: https://doi.org/10.4266/acc.000312

XML Download

Figure

Optimal oxygenation in the intensive care unit: what recent trials do—and do not—tell us

Figure 1. Oxygen utility. Oxygen delivery depends on cardiac output and arterial oxygen content (CaO2). CaO2 is determined by hemoglobin concentration, arterial oxygen saturation (SaO2), and the partial pressure of arterial oxygen (PaO2). Although PaO2 contributes to CaO2, SaO2 is the major determinant under physiological conditions. Oxygen utility is dependent on hemoglobin affinity determined by the environment of peripheral tissue. In clinical practice, we measure SaO2(or SpO2)/PaO2, but these values do not directly reflect tissue-level oxygen utilization. Numerous physiological factors influence peripheral oxygen extraction, and understanding these relationships is essential for interpreting oxygenation parameters in real-world settings.

Figure 2. Oxyhemoglobin dissociation curve. The oxyhemoglobin dissociation curve illustrates the nonlinear relationship between partial pressure of arterial oxygen (PaO2) and arterial oxygen saturation (SaO2). Rightward and leftward shifts reflect changes in hemoglobin affinity due to factors such as pH, temperature, and PCO2. In anemia, the curve typically shifts to the right, maintaining oxygen utility without increasing cardiac output. The position of the curve is influenced by numerous physiological variables. Although PaO2 and SO2 are the primary determinants of peripheral oxygen availability, the actual amount of molecular oxygen released to tissues depends heavily on the position of the oxyhemoglobin dissociation curve. In clinical practice, modern blood gas analyzers can provide the P50 value (the PaO2 at which hemoglobin is 50% saturated); however, neither this value nor hemoglobin levels and cardiac output have been widely evaluated in randomized controlled trials.

Figure 1.

Figure 2.

Optimal oxygenation in the intensive care unit: what recent trials do—and do not—tell us

Study	Trial name	Enrolled patients	Primary outcome
Stub et al. [25]	AVOID	STEMI	Size of myocardial infarction
Hofmann et al. [26]	DETO2X-AMI	AMI	365-Day mortality
Roffe et al. [29]	-	Stroke	Neurological outcome at day 90
Schmidt et al. [34]	BOX	Out-of-hospital CPA	Mortality & severe CNS disorders at day 90
Asfar et al. [38]	HYPERS2S	Septic shock/severe sepsis	28-Day mortality
Arleth et al. [20]	TRAUMOX2	Trauma	Death and/or major respiratory complications within 30 days
Girardis et al. [39]	-	ICU patients	ICU mortality
Mackle et al. [40]	ICU-ROX	MV patients in ICU	VFD until day 28
van der Wal et al. [42]	ICONIC	MV patients in ICU	28-Day mortality
Schjørring et al. [44]	HOT-ICU	AHRF	90-Day mortality
Barrot et al. [51]	LOCO₂	ARDS	28-Day mortality
Peters et al. [54]	Oxy-PICU	PICU patients	Duration of oxygen support at day 30
Martin et al. [41]	UK-ROX	MV patients in ICU	90-Day mortality

Study	Lower target	Higher target	Lower achievement	Higher achievement
Stub et al. [25]	SpO₂ ≥94%	O₂ mask 8L/min	98.0% (IQR, 96.0%–99.0%)	100.0% (IQR, 99.0%–100.0%)
Hofmann [26]	SpO₂ ≥90%	O₂ mask 6L/min	97% (IQR, 95%–98%)	99% (IQR, 97%–100%)
Roffe et al. [29]	No oxygen	n-cannula 2 or 3 L/min	-	-
Schmidt et al. [34]	PaO₂ 68–75 mm Hg	PaO₂ 98–105 mm Hg	-	-
Asfar et al. [38]	SaO₂ 88%–95%	F_IO₂ 1.0	-	-
Arleth et al. [20]	SpO₂ 94%	15 L/min or FiO₂ 1.0	97% (IQR, 96%–99%)	100% (IQR, 99–100)
Girardis et al. [39]	PaO₂ 70–100 mm Hg	PaO₂ up to 150 mm Hg	87 mm Hg (IQR, 79–97)	102 mm Hg (IQR, 88–116)
Mackle et al. [40]	91%≤ SpO₂<97%	91% ≤SpO₂	F_iO₂ 0.21, 1 hr (IQR, 0–17)	F_IO₂ 0.21, 27 hr (IQR, 5–78)
van der Wal et al. [42]	PaO₂ 55–80 mm Hg	PaO₂ 110–150 mm Hg	75 mm Hg (IQR, 69.8–83.5)	115 mm Hg (IQR, 100.3–129.0)
Schjørring et al. [44]	PaO₂ 60 mm Hg	PaO₂ 90 mm Hg	70.8 mm Hg (IQR, 66.6–76.5)	93.3 mm Hg (IQR, 87.1–98.7)
Barrot et al. [51]	PaO₂ 55–70 mm Hg	PaO₂ 90–105 mm Hg	PaO₂<55 mm Hg: 58 patients, >70 mm Hg: 97 patients	PaO₂<90 mm Hg: all patients, >105 mm Hg: 98 patients
Peters et al. [54]	SpO₂ 88%–92%	SpO₂ ≥94%	94% (IQR, 93–96)	97% (IQR, 96–98)
Martin et al. [41]	SpO₂ 88%–92%	Up to attending physician	93.3% (2.8%), 71.5 mm Hg (13.9)	95.1% (2.4%), 79.5 mm Hg (17.9)

Table 1. Major randomized controlled trials comparing conservative and liberal oxygenation strategies Enrolled patients and primary outcomes of each study

Table 2. Summary of oxygenation targets and achieved oxygenation levels in key clinical trials

SpO₂: peripheral oxygen saturation; IQR: interquartile range; PaO₂: arterial oxygen partial pressure; SaO₂: arterial oxygen saturation; F_IO₂: fraction of inspired oxygen.