The optimal oxygenation of the premature, especially the immature, newborn infant has been discussed and searched for the last decades [1, 2]. The Neonatal Oxygenation Prospective Meta-analysis (NeOProM) Collaboration represented a landmark in the study of oxygen therapy in immature infants beyond the delivery room [3]. Higher (91–95%) versus lower (85–89%) functional oxygen saturation (SpO2) target ranges from birth, or soon thereafter, were compared. This prospective meta-analysis comprised five trials, Surfactant, Positive Pressure, and Pulse Oximetry Randomized Trial (SUPPORT), the three Benefits of Oxygen Saturation Targeting (BOOST II) studies, and the Canadian Oxygen Trial (COT), including a total of 4,911 infants, similar concerning design, participants, and intervention [4‒6]. The studies also had identical primary outcome, a composite outcome of death or major disability by 18–24 months corrected age. The large number of enrolled subjects secured the power to draw conclusions on mortality and other outcomes such as injury to the lungs: bronchopulmonary dysplasia (BPD), eye: retinopathy of prematurity (ROP), brain: intraventricular hemorrhage or intestine: necrotizing enterocolitis (NEC).
After the individual studies were published, several investigators examined the results [7‒9]. Mean birthweights in the groups were between 825 and 837 g and gestational age 25.6–26.0 weeks. No significant difference between the 2 groups was found regarding the primary outcome. However, for all studies, combined mortality was 19.3% in the low and 16.2% in the high saturation groups, respectively (RR = 1.18, 95% CI: 1.04–1.34). Combined severe ROP was detected in 10.7% in the low saturation group versus 14.5% in the high saturation group, summary RR 0.74 (0.59–0.92). By contrast, NEC was significantly more common in the low saturation group (RR = 1.25, 95% CI: 0.05–1.49) [7‒9].
Thus, a high SpO2 target reduced both mortality and NEC but increased the risk for severe ROP. Most international guidelines therefore suggested a balance between high and low SpO2 targets, aiming at lower mortality risk at the expense of potentially more severe ROP. For that reason, most international guidelines recommend targeting SpO2 at 91–95% in immature newborn infants [10, 11].
However, after these results were published it has been suggested that perhaps the optimal SpO2 target for these immature infants should be in between the high and low saturation target groups investigated in the NeOProM studies. Taylor and co-workers [12] from the Canadian Neonatal Network present in this issue of Neonatology results from a retrospective cohort analysis of preterm infants <29 weeks GA born between 2011 and 2018 comparing intermediate SpO2 targets, that is, SpO2 88–93%, with high targets, 90–95%. Gestational age and birth weight are comparable with what was found in the NeOProM studies.
Survival without morbidity was higher in the intermediate target group compared to the high SpO2 target group (aOR 1.59, 95% CI: 1.04–2.45). There was, however, no difference in mortality (aOR 0.81, 95% CI: 0.59–1.11), NEC (aOR 0.90, 95% CI: 0.69–1.18), or severe ROP (aOR 1.26, 95% CI: 0.80–1.97). Further, there were no differences between the two groups regarding severe brain injury. BPD was almost significantly reduced in the intermediary compared to the high group (aOR 0.63, 95% CI: 0.38–1.03).
The authors also examined the effects of alarm limits. Alarm limits between 83–95% versus 88–97% were linked to higher survival (aOR 1.67, 95% CI: 1.05–2.65).
Although a low SpO2 target group was not included in this study, these results indicate that an intermediate SpO2 target leads to similar mortality as in the high group without increasing the risk of severe ROP. The study therefore suggests that a SpO2 target of 88–93% is preferable to 90–95%. It also indicates that alarm limits should not be higher than 95%.
Another study investigating the oxygenation of immature newborn infants is published in this issue of Neonatology. The objective of the SafeBoosC-III trial was to investigate the benefits or harms of treatment based on cerebral oximetry monitoring compared to standard (usual) care [13]. Now, Alsina-Casanova et al. [14] investigated whether cerebral oximeter-guided treatment during the first 72 h after birth reduced brain injury by better control of brain oxygenation than those undergoing usual care. A total of 210 patients from 8 centers in 5 countries (Belgium, Greece, Italy, Spain, USA) with a mean (SD) gestational age of 26.3 (1.2) weeks in both groups and birth weight of 856 (197) g and 847 (238) g in the 2 groups, respectively, were included. Careful monitoring of brain oxygenation did not affect outcomes such as severe brain injury assessed by cerebral ultrasound, with 18% versus 22% in the oximetry groups versus usual care. None of the other outcome measures such as BPD, ROP, or NEC differed between the 2 groups. The conclusion was that the use of cerebral oximetry-guided treatment in the delivery room and the first 72 h after birth did not show a benefit as measured by brain MRI at term equivalent age [14].
These 2 studies are filling a gap in our knowledge regarding the oxygenation of immature newborn infants in the delivery room and beyond. There has long been a need to test out during the first weeks of life an intermediate SpO2 target in these infants. However, it is not likely that any large randomized study will be carried out studying this so shortly following the publication of the NeOProM results. A retrospective cohort study as carried out by the authors from the Canadian networks is therefore at present the most realistic way to assess this issue in the near future.
Although the presented data from both studies are interesting, they may not be sufficient to change clinical practice. The 2022 European Consensus Guidelines on respiratory distress syndrome [15] will be revised in 2025. We guess that the results from Taylor and coworkers’ [12] study are not sufficient to change practice. However, they may justify testing an intermediate SpO2 target range of 88–93% in selected cases. It also seems justified to keep the upper alarm limits <97% to avoid the highest SpO2 peaks.
Near-infrared spectroscopy has for several years been tested out in newborn babies both in the delivery room and beyond, and no improved outcome has been found [13, 16]. The present study by SafeBoosC-III investigators reveals that near-infrared spectroscopy does not add clinically important information to justify the introduction of this technique routinely in the DR and NICU.
These 2 studies, however, demonstrate there are still gaps to fill regarding our knowledge of how to best oxygenate immature newborn infants in the delivery room and beyond. The oxygenation safety ranges of the most immature babies <25 weeks gestational age are hardly investigated. These babies are at higher risk of needing oxygen supplementation in a stage of rapid vessel development, which may enhance oxygen-derived chronic diseases such as BPD, ROP, and NEC. A subset study of epidemiological changes in this vulnerable population could guide our future studies. We welcome studies dealing with the delicate oxygen balance in these most vulnerable newborn infants.
Conflict of Interest Statement
The authors have no conflicts of interest to declare.
Funding Sources
No funding was received for writing this editorial.
Author Contributions
O.D.S. wrote the first draft. C.P.S. and M.V. contributed significantly to the writing of this manuscript. All authors have accepted the final version.