Beyond the trial data: why conservative oxygenation targets require nuanced implementation in pediatric critical care

The optimal oxygenation target for critically ill children has long been uncertain. In most pediatric intensive care units (PICUs), clinicians traditionally pursue liberal oxygenation [oxygen saturation (SpO₂) >94%]. Although intended to ensure safety, this practice potentially exposes patients to harmful oxygen toxicity, including oxidative stress and hyperoxic vasoconstriction. Severe hyperoxemia is consistently associated with increased mortality in critically ill children (1-3). Pediatric studies in bronchiolitis have demonstrated that lowering oxygenation targets to 88% is safe (4) in hospitalized patients and significantly reduces hospital stay (5). An additional trial evaluating the safety and conservative oxygenation in respiratory distress is underway in the pediatric population (6). Adult critical care trials exploring liberal versus conservative oxygenation targets have shown no mortality benefit in larger studies (7,8). Against this backdrop, the Oxy-PICU (oxygen in PICU) trial was launched as the first large, pragmatic randomized controlled trial (RCT) in emergency PICU admissions to evaluate whether conservative oxygenation targets could reduce the duration of organ support or death compared with usual liberal practice.

Enrolling more than 2,000 critically ill children across 15 UK PICUs, Oxy-PICU (9) demonstrated a small but statistically significant benefit of conservative oxygenation (SpO₂: 88–92%) in reducing the composite outcome of duration of organ support or death at 30 days. Importantly, this was achieved without evidence of increased harm, providing the first trial-based support for a conservative oxygen strategy in this setting. Notably, there was no statistical difference for the secondary outcome of mortality at 30 days between the two groups, underscoring that conservative oxygenation does not seem to confer any specific mortality benefit to patients.

A pre-specified follow-up analysis (10) examined whether benefits persisted beyond the acute phase in mechanically ventilated children, assessing survival at 90 days and one year, health-related quality of life, and cost-effectiveness. One-year mortality and average healthcare costs were both lower in the conservative oxygenation group, though the differences were not statistically significant, while quality of life measures remained similar between groups. Combined with the 30-day findings, these results indicate a possible benefit from conservative oxygenation targets in mechanically ventilated PICU patients, even in those with severe lung injury (11).

While Oxy-PICU demonstrates that conservative oxygen targets can be safe and effective at a population level, translating these findings into routine practice requires scrutiny of how SpO₂ is measured, interpreted, and acted upon. Three critical factors shape the practical application of conservative oxygenation strategies: pulse oximetry limitations, occult hypoxemia, and the clinical context of relative hypoxemia.

Pulse oximetry estimates arterial SpO₂ by exploiting differential absorption of red and infrared light by oxy- and deoxy-hemoglobin in pulsatile tissue. This is accomplished by calculating the ratio of the two signals against their non-pulsatile components and against each other and calibrating that ratio-of-ratios against blood gas derived SaO₂ data from human subjects. The signal-to-noise ratio (also known as accuracy) as well as the ratio-of-ratios is affected by tissue molecules that may absorb and scatter light, most notably melanin, as in silico simulations have demonstrated (12). Developers of medical oxygen sensors have not historically accounted for this property (13). As a result, pulse oximetry has been shown to overestimate arterial SpO₂ in children with darker skin (14,15), resulting in a higher risk of occult hypoxemia.

This is especially relevant to the case of patients hospitalized with sickle cell disease, who are at higher risk of developing acute chest syndrome (ACS) with hypoxemia (16). Practice guidelines (17) strongly recommend SpO₂ targets >95% in hospitalized sickle cell patients, making conservative oxygenation targets difficult to justify in mechanically ventilated sickle cell patients, given the high mortality risk associated with ACS. Notably, Oxy-PICU investigators found no significant variance in outcomes based on ethnicity (a poor but potentially useful surrogate for skin pigmentation) in a post-hoc analysis (18).

More generally, global tissue hypoperfusion results in a decrease in the ratio of the pulsatile to non-pulsatile component of the pulse oximetry signal, which can be quantified as a perfusion index (19), and utilized to classify shock states and assess responsiveness to ICU interventions. Importantly, a decreased perfusion index increases the discrepancy between SpO₂ and SaO₂, likely resulting in an increased likelihood of occult hypoxemia in these patients (20). From a clinical perspective, occult hypoxemia is associated with higher organ dysfunction scores and in-hospital mortality in critically ill adult patients (21,22) and overestimation of SaO₂ by SpO₂ in adult patients with coronavirus disease 2019 (COVID-19) led to delayed delivery of COVID-19-directed therapy and a high probability of hospital readmission (23). There is rising concern for occult hypoxemia similarly being associated with worse outcomes in critically ill pediatric patients (24).

When pulse oximetry is reliable and occult hypoxemia is unlikely, clinicians must understand what “relative hypoxemia” (SpO₂: 88–92%) means physiologically and when it might be safe. The most benign scenario occurs when relative hypoxemia results from ventilation-perfusion mismatch or intrapulmonary shunting, if overall alveolar ventilation remains adequate. In mechanically ventilated children with this pattern, we would expect to see reassuring signs: normal lung mechanics (reasonable compliance and resistance), relatively low ventilator support requirements, and absence of respiratory distress. Under these conditions, the lower SpO₂ likely reflects localized lung pathology rather than global respiratory failure, making conservative oxygen targets potentially safe.

However, when relative hypoxemia is new or accompanied by increased work of breathing, tachycardia, hypertension, ventilator dyssynchrony, or elevated CO₂ levels, it may herald worsening lung disease from various conditions—including acute respiratory distress syndrome (ARDS) progression, mucous plugging, atelectasis, or pulmonary edema. Liberal oxygenation targets provide both early warning of deteriorating lung physiology and buffer against significant hypoxemia, whereas conservative targets may provide false reassurance to inexperienced providers and lead to profound hypoxemia if underlying physiology is addressed in delayed fashion.

This concern is amplified by the sigmoid relationship between SpO₂ and partial pressure of arterial oxygen (PaO₂): SpO₂ falling below 88% is much more likely to cause profound PaO₂ decreases compared to SpO₂ falling below 92% or 94% before clinical intervention. This is especially applicable to patients with low functional reserve capacity—young infants with bronchiolitis or pediatric patients with chronic respiratory failure.

Further support for relative hypoxemia serving as an important indicator of respiratory pathophysiology can be inferred by evaluating respiratory care management around intubation and extubation. A recent adult trial evaluating pre-oxygenation with non-invasive ventilation versus oxygen mask during emergency intubation (25) found that non-invasive ventilation significantly reduced hypoxemia incidence (SpO₂ <85%) during intubation, with more oxygen-mask patients experiencing SpO₂ less than 95% during pre-oxygenation. Secondary analysis of a pediatric ARDS trial found that mechanically ventilated children likely to fail spontaneous breathing trials (26) had significant (though small) SpO₂ decreases alongside significantly decreased tidal volumes and increased work of breathing. These examples illustrate a key principle: SpO₂ is just one vital sign among many, and targeting 88–92% may be reasonable when considered alongside the complete clinical picture including ventilatory mechanics, hemodynamics, and overall trajectory of illness.

Overall, conservative oxygenation targets are likely reasonable for patients without occult hypoxemia, significant hemodynamic impairment, or substantial respiratory pathophysiology, thereby resulting in relative hypoxemia that is likely benign. This aligns with Oxy-PICU’s population-level findings, which showed modest benefit in duration of organ support but equivocal effects on mortality and quality of life. One might infer that the reduced duration of organ support in the conservative arm was possibly driven by earlier liberation from respiratory support—shorter courses of invasive and non-invasive mechanical ventilation rather than large shifts in much more critical outcomes like death or need for cardiac, renal, or neurologic support. This suggests any benefit may be mediated more through ICU processes than through altering the trajectory of critical illness itself. However, important questions remain unanswered. These findings do not account for subtler forms of harm including mixed hypoxemic events, neurocognitive sequelae, or the cumulative burden of recurrent desaturations that could disproportionately affect vulnerable patients. Hence, these targets are difficult to implement universally in pediatric ICU medicine, especially for patients in their sickest moments, when they may cause real harm, even if difficult to directly measure.

It seems most reasonable, then, that in convalescing patients—those identified by experienced senior clinicians and possibly augmented by future artificial intelligence risk stratification—implementing conservative oxygenation targets may prove quite beneficial in weaning and discontinuing ICU therapies toward recovery and hospital discharge. The key lies not in universal application, but in thoughtful, context-sensitive implementation that accounts for the technological limitations of our monitoring tools, the physiological complexity of critical illness, and the clinical expertise required to distinguish safe relative hypoxemia from impending or evolving cardiorespiratory failure.

Acknowledgments

None.

Footnote

Provenance and Peer Review: This article was commissioned by the Editorial Office, Translational Pediatrics. The article has undergone external peer review.

Peer Review File: Available at https://tp.amegroups.com/article/view/10.21037/tp-2025-616/prf

Funding: This work was supported by NIH National Institute of Child Health and Human Development training grant (#T32HD040686).

Conflicts of Interest: Both authors have completed the ICMJE uniform disclosure form (available at https://tp.amegroups.com/article/view/10.21037/tp-2025-616/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Lilien TA, Groeneveld NS, van Etten-Jamaludin F, et al. Association of Arterial Hyperoxia With Outcomes in Critically Ill Children: A Systematic Review and Meta-analysis. JAMA Netw Open 2022;5:e2142105. [Crossref] [PubMed]
2. Pelletier JH, Ramgopal S, Horvat CM. Hyperoxemia Is Associated With Mortality in Critically Ill Children. Front Med (Lausanne) 2021;8:675293. [Crossref] [PubMed]
3. Ramgopal S, Dezfulian C, Hickey RW, et al. Association of Severe Hyperoxemia Events and Mortality Among Patients Admitted to a Pediatric Intensive Care Unit. JAMA Netw Open 2019;2:e199812. [Crossref] [PubMed]
4. Louman S, van Stralen KJ, Pijnenburg MWH, et al. Oxygen saturation targets for children with respiratory distress: a systematic review. ERJ Open Res 2023;9:00256-2023. [Crossref] [PubMed]
5. Briggs S, Gupta V, Thakkar N, et al. Decreasing Goal Oxygen Saturations in Bronchiolitis Is Associated With Decreased Length of Stay. Hosp Pediatr 2023;13:688-94. [Crossref] [PubMed]
6. Louman S, van Stralen KJ, Koppelman G, et al. Safe, effective and cost-effective oxygen saturation targets for children and adolescents with respiratory distress: protocol for a randomised controlled trial (OxyKids study). BMJ Open 2024;14:e087891. [Crossref] [PubMed]
7. Semler MW, Casey JD, Lloyd BD, et al. Oxygen-Saturation Targets for Critically Ill Adults Receiving Mechanical Ventilation. N Engl J Med 2022;387:1759-69. [Crossref] [PubMed]
8. ICU-ROX Investigators and the Australian and New Zealand Intensive Care Society Clinical Trials Group. Conservative Oxygen Therapy during Mechanical Ventilation in the ICU. N Engl J Med 2020;382:989-98. [Crossref] [PubMed]
9. Peters MJ, Gould DW, Ray S, 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. [Crossref] [PubMed]
10. Moler-Zapata S, Peters MJ, Gould DW, 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. [Crossref] [PubMed]
11. Ray S, Wiegand M, Gould DW, et al. Severity of Impaired Oxygenation and Conservative Oxygenation Targets in Mechanically Ventilated Children: A Post Hoc Subgroup Analysis of the Oxy-PICU Trial of Conservative Oxygenation. Pediatr Crit Care Med 2025;26:e492-7. [Crossref] [PubMed]
12. Al-Halawani R, Qassem M, Kyriacou PA. Monte Carlo Simulation of the Effect of Melanin Concentration on Light-Tissue Interactions in Reflectance Pulse Oximetry. Sensors (Basel) 2025;25:559. [Crossref] [PubMed]
13. Keller MD, Harrison-Smith B, Patil C, et al. Skin colour affects the accuracy of medical oxygen sensors. Nature 2022;610:449-51. [Crossref] [PubMed]
14. Ruppel H, Makeneni S, Faerber JA, et al. Evaluating the Accuracy of Pulse Oximetry in Children According to Race. JAMA Pediatr 2023;177:540-3. [Crossref] [PubMed]
15. Andrist E, Nuppnau M, Barbaro RP, et al. Association of Race With Pulse Oximetry Accuracy in Hospitalized Children. JAMA Netw Open 2022;5:e224584. [Crossref] [PubMed]
16. Wesevich A, Woelkers M, Adegunsoye A, et al. Triaging acute chest syndrome clinical decision-making using bedside SaO2/FiO2 ratio. Blood Adv 2025;9:2970-9. [Crossref] [PubMed]
17. Mekontso Dessap A, Dauger S, Khellaf M, et al. Guidelines for the management of emergencies and critical illness in pediatric and adult patients with sickle cell disease. Ann Intensive Care 2025;15:74. [Crossref] [PubMed]
18. Jones GAL, Wiegand M, Ray S, et al. Ethnicity and Observed Oxygen Saturations, Fraction of Inspired Oxygen, and Clinical Outcomes: A Post-Hoc Analysis of the Oxy-PICU Trial of Conservative Oxygenation. Pediatr Crit Care Med 2024;25:912-7. [Crossref] [PubMed]
19. Elshal MM, Hasanin AM, Mostafa M, et al. Plethysmographic Peripheral Perfusion Index: Could It Be a New Vital Sign? Front Med (Lausanne) 2021;8:651909. [Crossref] [PubMed]
20. Wilson BJ, Cowan HJ, Lord JA, et al. The accuracy of pulse oximetry in emergency department patients with severe sepsis and septic shock: a retrospective cohort study. BMC Emerg Med 2010;10:9. [Crossref] [PubMed]
21. Wong AI, Charpignon M, Kim H, et al. Analysis of Discrepancies Between Pulse Oximetry and Arterial Oxygen Saturation Measurements by Race and Ethnicity and Association With Organ Dysfunction and Mortality. JAMA Netw Open 2021;4:e2131674. [Crossref] [PubMed]
22. Saidy S, Iqbal A, Baig SH. Pulse Oximetry Discrepancies and Occult Hypoxemia in ICU Patients: Predictors and Clinical Outcomes. J Intensive Care Med 2025;40:1269-78. [Crossref] [PubMed]
23. Fawzy A, Wu TD, Wang K, et al. Clinical Outcomes Associated With Overestimation of Oxygen Saturation by Pulse Oximetry in Patients Hospitalized With COVID-19. JAMA Netw Open 2023;6:e2330856. [Crossref] [PubMed]
24. Chen QC, Shen JJ, Huang YL, et al. Discrepancy between arterial oxygen saturation and pulse oximetry measurement in a Chinese pediatric patient cohort. Heliyon 2024;10:e28336. [Crossref] [PubMed]
25. Gibbs KW, Semler MW, Driver BE, et al. Noninvasive Ventilation for Preoxygenation during Emergency Intubation. N Engl J Med 2024;390:2165-77. [Crossref] [PubMed]
26. Knox KE, Hotz JC, Newth CJL, et al. A 30-Minute Spontaneous Breathing Trial Misses Many Children Who Go On to Fail a 120-Minute Spontaneous Breathing Trial. Chest 2023;163:115-27. [Crossref] [PubMed]