Introduction: Due to concerns of oxidative stress and injury, most clinicians currently use lower levels of fractional inspired oxygen (FiO2, 0.21–0.3) to initiate respiratory support for moderate to late preterm (MLPT, 32–36 weeks gestation) infants at birth. Whether this practice achieves recommended oxygen saturation (SpO2) targets is unknown. Methods: We aimed to determine SpO2 trajectories of MLPT infants requiring respiratory support at birth. We conducted a prospective, opportunistic, observational study with consent waiver. Preductal SpO2 readings were obtained during the first 10 min of life from infants between 32 and 36 weeks gestation requiring respiratory support in the delivery room. Primary outcome was reaching a minimum SpO2 80% at 5 min of life. The study was prospectively registered (ACTRN12620001252909). Results: A total of 76 eligible infants were recruited between February 2021 and March 2022 from 5 hospitals in Australia. Most (n = 58, 76%) had respiratory support initiated with FiO2 0.21 (range 0.21–1.0) using CPAP (92%). Median SpO2 at 5 min was 81% (interquartile range [IQR] 67–90) and 93% (IQR 86–96) at 10 min. At 5 min, 18/43 (42%) infants had SpO2 below 80% and only 8/43 (19%) reached SpO2 80–85%. Conclusions: Many MLPT infants requiring respiratory support do not achieve recommended SpO2 targets. In very preterm infants, SpO2 <80% at 5 min of life increases risk of death, intraventricular haemorrhage, and neurodevelopmental impairment. The implications on this practice on the health outcomes of MLPT infants are unclear and require further research.

Oxygen is essential for life, but the inappropriate amount of oxygen can be harmful and even fatal. Both too much (hyperoxia) and too little (hypoxia) oxygen rapidly lead to cellular injury and death [1, 2]. Supplemental oxygen is an integral part of the resuscitation process in newborn infants [3], and for decades, pure oxygen (fractional inspired oxygen [FiO2] 1.0) was used for respiratory support of depressed infants. However, the Resair studies in the 1990s [4, 5] showed that it was possible to use air (FiO2 0.21) to initiate respiratory support of hypoxic, full-term infants without increasing risk of neurodevelopmental impairment (NDI) [6]. Using air also reduced oxidative stress and injury to major organs such as the heart and kidneys [7]. Subsequent meta-analyses involving >1,300 infants showed that air significantly reduced the risk of death in hypoxic term or near-term infants compared to using pure oxygen (typical relative risk 0.71, 95% confidence interval [CI] 0.54–0.94) [8] with little impact on neurodevelopmental outcomes [9].

In 2006, an international consensus on science statement suggested that air could be used as an alternative to pure oxygen for newborn respiratory support [10]. Concurrently, oxygen saturations (SpO2) were found to rise gradually and not surpass 90% until at least 7–8 min, even in healthy full-term infants [11, 12]. Administering high levels of FiO2 rapidly increased SpO2 and increased oxidative stress markers of organ injury [13, 14].

In 2010, expert treatment recommendations stated that FiO2 should be titrated to target preductal trajectories of SpO2 of healthy term infants [15] even though the initial studies in full-term infants [4‒6] did not measure SpO2 or titrate FiO2 with blenders. With these recommendations, equipoise was rapidly lost for using higher oxygen levels for infant resuscitation, even for preterm infants with lung disease [16]. Indeed, the largest study to examine the use of air or pure oxygen to resuscitate preterm infants, the Targeted Oxygen in the Resuscitation of Preterm Infants and their Developmental Outcomes (To2rpido), had to be curtailed at 15% recruitment (n = 292) because of clinician reticence to enrol into the high oxygen arm [17].

In 2018, individual patient data were collated from 8 randomised controlled trials (RCTs) that examined lower (FiO2 ≤0.3) and higher (FiO2 ≥0.6) oxygen to initiate resuscitation of infants <32 weeks gestation [18, 19]. Data from 768 infants found that initial FiO2 made no difference to major outcomes such as death or morbidities such as bronchopulmonary dysplasia and NDI. However, not reaching SpO2 80% by 5 min significantly increased risk of intraventricular haemorrhage, death, and NDI [18, 19]. Indeed, 76% of infants did not reach study SpO2 targets despite vigilant trial conditions [18].

The impacts of using lower oxygen strategies on preterm infants between 32 and 36 weeks gestation (moderate to late preterm [MLPT]) are unknown [20]. Considering the known risks of suboptimal oxygenation in very preterm infants even for a few minutes, the deficit of knowledge for this massive group of infants which account for 75% of all preterm births (>10 million per year) [21] is of concern. Even though MLPT infants have less severe lung disease than very preterm infants and are less likely to die, inappropriate oxygenation may potentially increase risk of adverse outcomes, including respiratory and neurodevelopmental sequelae [20, 22, 23].

In this study, we sought to determine if SpO2 trajectories of MLPT infants requiring respiratory support at birth met recommended SpO2 targets. We hypothesised that MLPT infants using currently recommended FiO2 strategies will not meet target SpO2 levels derived from healthy, spontaneously breathing infants.

Study Design

This is a multicentre, observational, and opportunistic cohort study conducted between February 2021 and December 2022. Infants were recruited from five hospitals – two tertiary-level hospitals with neonatal intensive care units (NICU) and three district-level hospitals with special care nurseries managing infants with a minimum gestation of 32 weeks.

Eligibility Criteria

Infants were eligible if they were born between 32+0 and 36+6 weeks gestation from best estimate including maternal dates and early antenatal ultrasound assessments, and required respiratory support, with continuous positive airway pressure (CPAP), intermittent positive pressure ventilation by mask, or intubation and mechanical ventilation, after birth.

Exclusion Criteria

Infants were excluded if they (1) did not require respiratory support at birth; (2) had congenital conditions that could impact on oxygenation (e.g., cardiorespiratory abnormalities); (3) were born when there was no personnel or equipment to obtain data; (4) were of unknown gestation.

Study Procedure

The attending clinical team determined all interventions at birth including respiratory support and oxygen administration. A Masimo Root® platform with the Pathway™ system installed (Masimo Corporation, Irvine, CA, USA) was used to collect heart rate (HR) and SpO2 data every 2 s with maximum sensitivity settings for 10 min. Oximetry data were also recorded manually every 60 s in addition to the Root® recordings. Manual recordings were used for analysis when the Root® was unavailable. Following cord clamping and after the resuscitation team received the infant, an oximeter probe was placed on the infant’s right wrist for a preductal reading. Oximetry data, FiO2 at each minute, and standard demographic and outcome data were collected. Definitions and measurement tools for the collected variables are provided in online supplementary Appendix 1 (for all online suppl. material, see https://doi.org/10.1159/000539221). Observations were recorded by a person who was not involved in the clinical care of the infant, and completed data sheets were transferred to the coordinating principal investigator using encrypted file transfer with password protection. Details about other delivery room interventions including type of respiratory support, delayed cord clamping, and surfactant administration were also collected. The data were stored securely on a REDCap database [24] at the University of New South Wales, Sydney.

Outcome Measures

The primary outcome was SpO2 at 5 min. Secondary outcomes were reaching SpO2 80% and 80–85% at 5 min, HR, and need for respiratory support at 24 h of age.

Ethics Considerations

Since this was an observational study using interventions and procedures that were standard practice (including oximetry measurements), the study was approved by the Hunter New England Local Health District HREC (2020/ETH01038) for consent waiver. Site-specific approval was also provided by each local institution.

Sample Size Calculation

Power calculations for one sample test for proportions with continuity correction were conducted using the MKpower package for R [25]. A total of 78 infants were needed to detect a 15% absolute difference in the proportion of infants to reach SpO2 80% at 5 min compared to population norms (75% vs. 60%, alpha 0.05, power 0.8).

Statistical Analyses and Data Transformations

Statistical analyses were conducted using the open-source software R [26] with the gtsummary package [27]. Participants were stratified by initial FiO2 (room air or supplementary oxygen). Descriptive statistics were used to present the results, and χ2 tests, or Fisher’s exact tests for cell counts fewer than five, were used to evaluate proportions between categorical variables/groups where applicable. Continuous variables were conservatively assumed non-normally distributed. Statistically significant differences between continuous variables were assessed with the Wilcoxon rank-sum test. We set two-tailed significance at the 5% level. Automatically recorded SpO2 and HR data were downloaded from the Masimo Root® platform with the Pathway™ system installed and transformed into minute values by taking the mean of valid recordings 30 s either side of the time point. Centile curves for SpO2 were generated with quantile regression. Post hoc analyses were conducted to examine the relationship of SpO2 and respiratory support at 24 h to explore potential relationships between initial hypoxia and early respiratory morbidity [28]. These are reported as exploratory where appropriate.

Reporting

The results of this study are reported in line with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement (online suppl. Appendix 2) [29].

A total of 102 infants were enrolled. Data from 76 infants were included in the final analyses. The most common reasons for exclusion were (1) not needing respiratory support (n = 22) and (2) birth determined to be outside target gestation range (32–36 weeks) after delivery room stabilisation; see Figure 1 for participant flow diagram. No infant died in the delivery room or during the hospitalisation. A summary of the number of participants with missing data for each variable is reported in online supplementary Appendix 3. These data were not suitable for imputation [30].

Fig. 1.

Flow diagram of study participant enrolment.

Fig. 1.

Flow diagram of study participant enrolment.

Close modal

Patient Demographics

Median (interquartile range [IQR]) gestational age of the infants was 34 (33–35) weeks. Most (n = 54/74, 71%) were born by caesarean section and had delayed cord clamping ≥30 s (n = 54/68, 79%). Other participant characteristics are shown in Table 1. All infants were admitted to either a special care nursery or NICU.

Table 1.

Characteristics and outcomes of study participants

Overall, N = 76Oxygen, N = 18Room air, N = 58p value2
Characteristics1 
Maternal age 32.0 (29.5–36.0) 34.0 (30.2–36.0) 31.0 (29.0–36.0) 0.3 
Received antenatal care 72/74 (97) 18/18 (100) 54/56 (96) >0.9 
Mode of delivery 54/74 (73) 14/18 (78) 40/56 (71) 0.9 
PPROM 14/76 (18) 2/18 (11) 12/58 (21) 0.5 
Complete antenatal steroid cover 46/75 (61) 12/18 (67) 34/57 (60) 0.6 
Gestational age (completed weeks) 34 (33–35) 33 (32–34) 34 (33–35) 0.092 
Sex (male) 37/72 (51) 11/18 (61) 26/54 (48) 0.3 
Sex (female) 35/72 (49) 7/18 (39) 28/54 (52)  
Delayed cord clamping ≥30 s 54/68 (79) 12/14 (86) 42/54 (78) 0.7 
Apgar (1 min) 8 (6–9) 7 (6–8) 8 (6–9) 0.10 
Apgar (5 min) 9 (8–9) 8 (7–9) 9 (8–9) 0.065 
Outcomes 
Surfactant administration before 24 h 8/75 (11) 4/17 (24) 4/58 (6.9) 0.072 
Cardiopulmonary pathology at 24 h 27/74 (36) 5/17 (29) 22/57 (39) 0.6 
Disposition at 24 h    >0.9 
 Discharged from nursery/NICU 4/76 (5.3) 1/18 (5.6) 3/58 (5.2) >0.9 
 Remains in nursery/NICU 72/76 (95) 17/18 (94) 55/58 (95)  
Ongoing respiratory support at 24 h 52/76 (68) 15/18 (83) 37/58 (64) 0.12 
Overall, N = 76Oxygen, N = 18Room air, N = 58p value2
Characteristics1 
Maternal age 32.0 (29.5–36.0) 34.0 (30.2–36.0) 31.0 (29.0–36.0) 0.3 
Received antenatal care 72/74 (97) 18/18 (100) 54/56 (96) >0.9 
Mode of delivery 54/74 (73) 14/18 (78) 40/56 (71) 0.9 
PPROM 14/76 (18) 2/18 (11) 12/58 (21) 0.5 
Complete antenatal steroid cover 46/75 (61) 12/18 (67) 34/57 (60) 0.6 
Gestational age (completed weeks) 34 (33–35) 33 (32–34) 34 (33–35) 0.092 
Sex (male) 37/72 (51) 11/18 (61) 26/54 (48) 0.3 
Sex (female) 35/72 (49) 7/18 (39) 28/54 (52)  
Delayed cord clamping ≥30 s 54/68 (79) 12/14 (86) 42/54 (78) 0.7 
Apgar (1 min) 8 (6–9) 7 (6–8) 8 (6–9) 0.10 
Apgar (5 min) 9 (8–9) 8 (7–9) 9 (8–9) 0.065 
Outcomes 
Surfactant administration before 24 h 8/75 (11) 4/17 (24) 4/58 (6.9) 0.072 
Cardiopulmonary pathology at 24 h 27/74 (36) 5/17 (29) 22/57 (39) 0.6 
Disposition at 24 h    >0.9 
 Discharged from nursery/NICU 4/76 (5.3) 1/18 (5.6) 3/58 (5.2) >0.9 
 Remains in nursery/NICU 72/76 (95) 17/18 (94) 55/58 (95)  
Ongoing respiratory support at 24 h 52/76 (68) 15/18 (83) 37/58 (64) 0.12 

1n/N (%) for categorical variables, median (IQR) for continuous variables.

2Wilcoxon rank-sum test; Fisher’s exact test; Pearson’s χ2 test.

Oxygen Supplementation

Most infants (n = 58/76, 76%) had respiratory support commenced with FiO2 0.21 (room air). Median initial FiO2 was 0.35 (IQR 0.28–0.4). Five infants were initially administered FiO2 0.4, and two infants were initially administered FiO2 1.0 before being weaned to FiO2 0.21–0.3 after 3–4 min. At 5 minutes, median FiO2 was 0.4 (IQR 0.3–0.4) in those initially given supplemental oxygen and 0.3 (IQR 0.21–0.40) for those initially given air (p = 0.05). At 10 minutes, median FiO2 was higher in those initially given supplementary oxygen (0.38 vs. 0.23, p = 0.01). Median minute-to-minute FiO2 adjustment was 0.1 (IQR 0.09–0.22) for both groups. Median FiO2 over 10 min was not different between those given oxygen (0.3 [IQR 0.24–0.35]) or air (0.22 [IQR 0.21–0.35], p = 0.07). Adjustments in FiO2 over the ten-minute observation period are shown graphically in online supplementary Appendix 4.

Respiratory Support

Most infants received mask CPAP (n = 70/76, 92%) and/or intermittent positive pressure ventilation (n = 16/76 21%). Five infants were intubated. Eight infants received surfactant in the NICU. Infants who received surfactant were more premature than those who were not treated with surfactant (median gestation 32 vs. 34 weeks p = 0.01). Surfactant treatment was not different between air and oxygen infants (p = 0.07; see Table 1). At 24 h, 52/76 (68%) infants required respiratory support and 6 were intubated. No infant received inhaled nitric oxide or high-frequency ventilation. In post hoc exploratory analyses, infants requiring respiratory support at 24 h were less likely to have reached SpO2 80% at 5 min compared to those not needing respiratory support (53 vs. 73%, p = 0.03).

Oxygen Saturation

SpO2 in all infants increased gradually and plateaued to ∼90% by 7 min. Centiles for SpO2 are shown in Figure 2. Five-minute SpO2 was 81% (IQR 67–90, n = 43) and not different between those given air (84%, IQR 66–90) or oxygen (80% IQR 68–89, p > 0.9, Table 2). Most infants (81%, n = 35/43) were outside SpO2 target range of 80–85% [3, 12] (Fig. 3). At 5 min, almost half (42%, n = 18/43) of the infants had SpO2 <80% regardless of initial resuscitation with air or oxygen (46 vs. 40%, respectively, p = 0.7, Fig. 4) and 40% (n = 17/43) had 5-min SpO2 above 85%. One of the two infants given FiO2 1.0 had SpO2 <80% at 5 min, and the other had SpO2 >95%. The proportion of infants in SpO2 ranges are shown in online supplementary Appendix 4. In post hoc univariate linear regression, infants requiring respiratory support with CPAP at 24 h had lower SpO2 at 5 min (mean difference 10% in SpO2, 95% CI: 19–25%, p = 0.04).

Fig. 2.

SpO2 centiles of MLPT infants from zero to 10 min. The solid line represents the 50th centile, interrupted solid lines represent the 75th or 25th centiles, and the dotted lines represent the 90th or 10th centiles.

Fig. 2.

SpO2 centiles of MLPT infants from zero to 10 min. The solid line represents the 50th centile, interrupted solid lines represent the 75th or 25th centiles, and the dotted lines represent the 90th or 10th centiles.

Close modal
Table 2.

Median (IQR) SpO2 of MLPT infants resuscitated with room air or supplementary oxygen from 1 to 10 min

TimeOverall, N = 761Oxygen, N = 181Room air, N = 581p value2
1 min (n = 16, O2 = 5, RA = 11) 72 (60–80) 70 (50–79) 73 (64–80) 0.7 
2 min (n = 28, O2 = 8, RA = 20) 70 (64–79) 66 (62–80) 72 (65–79) 0.8 
3 min (n = 36, O2 = 9, RA = 27) 74 (62–83) 65 (60–79) 77 (64–83) 0.7 
4 min (n = 39, O2 = 11, RA = 28) 79 (66–88) 73 (66–94) 80 (69–86) >0.9 
5 min (n = 43, O2 = 13, RA = 30) 81 (67–90) 80 (68–89) 84 (66–90) >0.9 
6 min (n = 39, O2 = 11, RA = 28) 83 (70–90) 79 (67–86) 84 (72–91) 0.3 
7 min (n = 43, O2 = 13, RA = 30) 85 (74–92) 79 (72–92) 88 (80–94) 0.4 
8 min (n = 41, O2 = 13, RA = 28) 87 (80–94) 87 (79–91) 87 (81–94) >0.9 
9 min (n = 42, O2 = 13, RA = 29) 90 (81–95) 94 (88–95) 88 (81–95) 0.3 
10 min (n = 34, O2 = 12, RA = 30) 93 (86–96) 96 (90–97) 90 (83–95) 0.1 
TimeOverall, N = 761Oxygen, N = 181Room air, N = 581p value2
1 min (n = 16, O2 = 5, RA = 11) 72 (60–80) 70 (50–79) 73 (64–80) 0.7 
2 min (n = 28, O2 = 8, RA = 20) 70 (64–79) 66 (62–80) 72 (65–79) 0.8 
3 min (n = 36, O2 = 9, RA = 27) 74 (62–83) 65 (60–79) 77 (64–83) 0.7 
4 min (n = 39, O2 = 11, RA = 28) 79 (66–88) 73 (66–94) 80 (69–86) >0.9 
5 min (n = 43, O2 = 13, RA = 30) 81 (67–90) 80 (68–89) 84 (66–90) >0.9 
6 min (n = 39, O2 = 11, RA = 28) 83 (70–90) 79 (67–86) 84 (72–91) 0.3 
7 min (n = 43, O2 = 13, RA = 30) 85 (74–92) 79 (72–92) 88 (80–94) 0.4 
8 min (n = 41, O2 = 13, RA = 28) 87 (80–94) 87 (79–91) 87 (81–94) >0.9 
9 min (n = 42, O2 = 13, RA = 29) 90 (81–95) 94 (88–95) 88 (81–95) 0.3 
10 min (n = 34, O2 = 12, RA = 30) 93 (86–96) 96 (90–97) 90 (83–95) 0.1 

n, number of participants with SpO2 at each timepoint; RA, number of participants with SpO2 recording in room air group; O2, number of participants with SpO2 recording in oxygen group.

1Median (IQR).

2Wilcoxon rank-sum test.

Fig. 3.

Median SpO2 values from Dawson nomograms [12] are displayed in red.

Fig. 3.

Median SpO2 values from Dawson nomograms [12] are displayed in red.

Close modal
Fig. 4.

Violin plot of infants’ SpO2 from 1 to 10 min of life. Shaded area represents SpO2 targets at each minute.

Fig. 4.

Violin plot of infants’ SpO2 from 1 to 10 min of life. Shaded area represents SpO2 targets at each minute.

Close modal

Heart Rate

HRs did not significantly differ between the air and oxygen groups at all time points. Median HR (IQR) increased from 127 bpm (80–150) at 1 min (n = 17) to 150 bpm (137–160) at 10 min (n = 54). The 25th, 50th, and 75th centiles of five-minute HR were 133 bpm, 144 bpm, and 158 bpm, respectively (online suppl. Appendix 5). HRs did not differ significantly between infants with and without DCC or cardiopulmonary pathology.

This is the first study to examine SpO2 within the first 10 min of life in MLPT infants requiring respiratory support. We showed that most MLPT infants who were resuscitated with oxygen and ventilation strategies recommended by expert committees did not achieve recommended SpO2 targets. The implications of these findings on clinical outcomes are unknown and require evaluation, especially in light of emerging evidence suggesting concerns about hypoxaemia in very preterm (<32 weeks gestation) infants and the potential impacts of oxidative stress in infants exposed to hyperoxaemia.

Recommendations from international guidelines to initiate resuscitation with lower initial FiO2 were extrapolated from RCTs conducted in term, hypoxic infants who were resuscitated with either FiO2 0.21 or 1.0 [31]. Recommendations to adjust FiO2 to target SpO2 readings were derived from observations of spontaneously breathing, full-term healthy infants [12]. Whether these trajectories could be met, and their clinical relevance in preterm infants at risk of oxidative stress, was unknown until the first meta-analyses from individual patient data of infants <32 weeks gestation enrolled in trials of lower (≤0.3) and higher (≥0.65) initial FiO2 for resuscitation. Data from 706 infants <32 weeks gestation found that infants were less likely to reach SpO2 80% by 5 min if resuscitation was initiated with FiO2 <0.3 (OR 2.63, 95% CI: 1.21–5.74). Furthermore, failure to reach SpO2 80% by 5 min was associated with an increased risk of death (OR 4.57, 95% CI: 1.62–13.98) [18] and in survivors, with lower cognitive scores and NDI [19, 32].

The association between SpO2 targets and major clinical outcomes in MLPT infants requiring respiratory support at birth is unknown. MLPT infants are less likely to die or sustain major morbidities such as intraventricular haemorrhage and respiratory complications, compared to very preterm infants. However, their impact on health care resources is concerning because of their large numbers (>10 million MLPT infants are born worldwide each year [33]) and little is known about the impact of current resuscitation strategies on either their short- or longer term outcomes. In a review of the Vermont Oxford Network Database from 2011 to 2020, including 616, 110 infants between 30 and 36 weeks gestation, median 34 weeks, Handley et al. [34] found that 55.3% required delivery room intervention. In our survey of 695 clinicians from 21 countries, we found that most would initiate respiratory support for MLPT infants with FiO2 0.21 (43%) or 0.3 (36%) and only 45% would titrate FiO2 to meet recommended SpO2 targets [34].

We found in a post hoc analysis that infants were more likely to need respiratory support at 24 h if they had lower SpO2 at 5 min (mean difference 10% in SpO2, 95% CI: 19–25%, p = 0.04). The cause for this needs to be examined in larger studies. Lower SpO2 at 5 min may have reflected worse initial respiratory disease, or infants required more prolonged postnatal respiratory support because of insufficient oxygenation at birth. Further evaluation of this relationship is required. In animal foetuses, hypoxia stops breathing movements through active inhibition of respiratory centres [35]. In very preterm infants (median 27 weeks gestation), respiratory drive increased through transient administration of FiO2 1.0 [28]. One RCT of infants <30 weeks gestation also found higher respiratory effort, improved SpO2, and shorter duration of mask ventilation when resuscitation was initiated with FiO2 1.0 compared to 0.3 [36]. Certainly, very preterm infants (23–32 weeks gestation) who require respiratory support take longer to achieve target SpO2 and HR than those not requiring support [37].

Limitations

This study has its limitations. It was opportunistic, and recruitment may have been impacted by selection bias. For example, data were not collected if personnel were unable to record resuscitation information. In some instances, there were difficulties obtaining reliable SpO2 traces, especially before 5 min, which we aimed to account for with parallel manual recordings. Consequently, proportions of missing data were high for some outcomes and time points (e.g., only 57% of infants had SpO2 recordings at 5 min), and data were possibly missing not at random. These data were not suitable for multiple imputation, and a complete case analysis was performed [30]. Treatment discrepancies because of the study’s observational design may have accounted for observed differences and may have biased results. Sicker infants, for example, may have preferentially been given supplemental oxygen or started on higher levels of oxygen, thus potentially introducing allocation bias to the treatment groups. The speed and nature of FiO2 change may have varied across treating clinicians or centres, which would also have influenced SpO2 results. Future studies exploring various titration strategies may add further insights. Timing of delayed cord clamping, and therefore timing of the resuscitation team receiving the infant, varied within and across sites, and this may have led to inconsistencies in the recording of SpO2 and led to fewer infants with valid SpO2 recordings in the first few minutes of observation. The characteristics of the study population, including the high caesarean section rate, significantly limit generalisability to those born via vaginal delivery, other gestational age ranges, and those born at very low birth weight. Large numbers of comparisons were analysed increasing the risk of type 1 error; hence, outcomes were prespecified and analyses were reported as exploratory where relevant [38].

This study was not designed to examine major clinical outcomes, such as death, neurodevelopment, or need for respiratory support after birth. The AIROPLANE study (ACTRN12621001267842), currently recruiting, is a cluster-randomised cross-over trial of FiO2 0.21 versus 0.3 for delivery room resuscitation of MLPT infants and aims to determine the impact of initial FiO2 on the primary outcome of need for respiratory support after birth. Further studies will be needed to assess if initial oxygenation affects major clinical outcomes, including death and neurodevelopment in this population. The intersection of delivery room interventions, including the interaction of supplemental oxygen, response to ventilatory manoeuvres, and cord management practices, could also be explored in future studies to inform potential synergistic/antagonistic treatment effects. Due to the massive numbers of patients involved to answer these questions, agile and adaptive trial designs [39] conducted under a consent waiver process [40] and utilising collaborative evidence synthesis approaches [41‒43] are likely to be needed.

We found that most MLPT infants who were resuscitated with oxygen supplementation strategies currently recommended by expert committees did not achieve recommended SpO2 at 5 min of life, falling below or above the recommended range of 80–85%. Further research is required to determine the most appropriate oxygen supplementation strategies to understand the oxygen needs of MLPT infants and the impact of these strategies on important clinical outcomes including death and NDI.

We thank Prof Maximo Vento, Dr. Celia Padilla-Sánchez, and Dr. Antonio Cañada-Martínez for sharing R code to assist with data visualisation.

This study protocol was reviewed and approved by Hunter New England Local Health District HREC, New South Wales, Australia (2020/ETH01038). The study was eligible for waiver of consent given the entirely observational design. Site-specific approval was also provided by each local institution.

The authors have no conflicts of interest to declare.

Equipment (Masimo Root® platform with the Pathway™ system installed) and consumables (oxygen saturation probes) were provided in-kind by Masimo Corporation, Irvine, CA, USA. This study was not supported by any sponsor or funder.

J.X.S. and J.L.O. were responsible for the study design and ethical approval process, and oversaw data collection, which was coordinated by site leads (S.B., K.Y., C.L.A., A.K., and M.T.) at their respective hospitals. J.X. and T.A.N.P. were responsible for data entry and data cleaning. J.X.S. was responsible for data analysis, data visualisation, writing – initial review, and statistical analyses under the supervision of J.L.O. All authors (J.X.S., S.B., T.A.N.P., K.Y., C.L.A., A.K., M.T., J.S., J.L.O.) contributed to revising and editing the manuscript, were involved in the decision to submit the manuscript, and met ICMJE authorship requirements.

The data that support the findings of this study are not publicly available due to their containing information that could compromise the privacy of research participants but are available at request for research purposes with appropriate ethical approval. Data requests can be directed to J.X.S. (james.sotiropoulos@sydney.edu.au) and J.L.O. (j.oei@unsw.edu.au).

1.
Torres-Cuevas
I
,
Parra-Llorca
A
,
Sánchez-Illana
A
,
Nuñez-Ramiro
A
,
Kuligowski
J
,
Cháfer-Pericás
C
, et al
.
Oxygen and oxidative stress in the perinatal period
.
Redox Biol
.
2017
;
12
:
674
81
.
2.
Sotiropoulos
JX
,
Kapadia
V
,
Vento
M
,
Rabi
Y
,
Saugstad
OD
,
Kumar
RK
, et al
.
Oxygen for the delivery room respiratory support of moderate to late preterm infants. An international survey of clinical practice from 21 countries
.
Acta Paediatr
.
2021
;
110
(
12
):
3261
8
.
3.
Wyckoff
MH
,
Wyllie
J
,
Aziz
K
,
de Almeida
MF
,
Fabres
J
,
Fawke
J
, et al
.
Neonatal life support: 2020 international consensus on cardiopulmonary resuscitation and emergency cardiovascular care science with treatment recommendations
.
Circulation
.
2020
;
142
(
16_Suppl l_1
):
S185
221
.
4.
Ramji
S
,
Ahuja
S
,
Thirupuram
S
,
Rootwelt
T
,
Rooth
G
,
Saugstad
OD
.
Resuscitation of asphyxic newborn infants with room air or 100% oxygen
.
Pediatr Res
.
1993
;
34
(
6
):
809
12
.
5.
Saugstad
OD
,
Rootwelt
T
,
Aalen
O
.
Resuscitation of asphyxiated newborn infants with room air or oxygen: an international controlled trial: the Resair 2 study
.
Pediatrics
.
1998
;
102
(
1
):
e1
.
6.
Saugstad
OD
,
Ramji
S
,
Irani
SF
,
El-Meneza
S
,
Hernandez
EA
,
Vento
M
, et al
.
Resuscitation of newborn infants with 21% or 100% oxygen: follow-up at 18 to 24 months
.
Pediatrics
.
2003
;
112
(
2
):
296
300
.
7.
Vento
M
,
Sastre
J
,
Asensi
MA
,
Viña
J
.
Room-air resuscitation causes less damage to heart and kidney than 100% oxygen
.
Am J Respir Crit Care Med
.
2005
;
172
(
11
):
1393
8
.
8.
Davis
PG
,
Tan
A
,
O’Donnell
CPF
,
Schulze
A
.
Resuscitation of newborn infants with 100% oxygen or air: a systematic review and meta-analysis
.
Lancet
.
2004
;
364
(
9442
):
1329
33
.
9.
Saugstad
OD
,
Vento
M
,
Ramji
S
,
Howard
D
,
Soll
RF
.
Neurodevelopmental outcome of infants resuscitated with air or 100% oxygen: a systematic review and meta-analysis
.
Neonatology
.
2012
;
102
(
2
):
98
103
.
10.
International Liaison Committee on Resuscitation
.
The international liaison committee on resuscitation (ILCOR) consensus on science with treatment recommendations for pediatric and neonatal patients: pediatric basic and advanced life support
.
Pediatrics
.
2006
;
117
(
5
):
e955
77
.
11.
Rabi
Y
,
Yee
W
,
Chen
SY
,
Singhal
N
.
Oxygen saturation trends immediately after birth
.
J Pediatr
.
2006
;
148
(
5
):
590
4
.
12.
Dawson
JA
,
Kamlin
COF
,
Vento
M
,
Wong
C
,
Cole
TJ
,
Donath
SM
, et al
.
Defining the reference range for oxygen saturation for infants after birth
.
Pediatrics
.
2010
;
125
(
6
):
e1340
7
.
13.
Wang
CL
,
Anderson
C
,
Leone
TA
,
Rich
W
,
Govindaswami
B
,
Finer
NN
.
Resuscitation of preterm neonates by using room air or 100% oxygen
.
Pediatrics
.
2008
;
121
(
6
):
1083
9
.
14.
Vento
M
,
Moro
M
,
Escrig
R
,
Arruza
L
,
Villar
G
,
Izquierdo
I
, et al
.
Preterm resuscitation with low oxygen causes less oxidative stress, inflammation, and chronic lung disease
.
Pediatrics
.
2009
;
124
(
3
):
e439
49
.
15.
Perlman
JM
,
Wyllie
J
,
Kattwinkel
J
,
Atkins
DL
,
Chameides
L
,
Goldsmith
JP
, et al
.
Neonatal resuscitation: 2010 international consensus on cardiopulmonary resuscitation and emergency cardiovascular care science with treatment recommendations
.
Pediatrics
.
2010
;
126
(
5
):
e1319
44
.
16.
Bhola
K
,
Lui
K
,
Oei
JL
.
Use of oxygen for delivery room neonatal resuscitation in non-tertiary Australian and New Zealand hospitals: a survey of current practices, opinions and equipment
.
J Paediatr Child Health
.
2012
;
48
(
9
):
828
32
.
17.
Oei
JL
,
Saugstad
OD
,
Lui
K
,
Wright
IM
,
Smyth
JP
,
Craven
P
, et al
.
Targeted oxygen in the resuscitation of preterm infants, a randomized clinical trial
.
Pediatrics
.
2017
;
139
(
1
):
e20161452
.
18.
Oei
JL
,
Finer
NN
,
Saugstad
OD
,
Wright
IM
,
Rabi
Y
,
Tarnow-Mordi
W
, et al
.
Outcomes of oxygen saturation targeting during delivery room stabilisation of preterm infants
.
Arch Dis Child Fetal Neonatal Ed
.
2018
;
103
(
5
):
F446
54
.
19.
Oei
JL
,
Kapadia
V
,
Rabi
Y
,
Saugstad
OD
,
Rook
D
,
Vermeulen
MJ
, et al
.
Neurodevelopmental outcomes of preterm infants after randomisation to initial resuscitation with lower (FiO2<0.3) or higher (FiO2>0.6) initial oxygen levels. An individual patient meta-analysis
.
Arch Dis Child Fetal Neonatal Ed
.
2022
;
107
(
4
):
386
92
.
20.
Shapiro-Mendoza
CK
,
Lackritz
EM
.
Epidemiology of late and moderate preterm birth
.
Semin Fetal Neonatal Med
.
2012
;
17
(
3
):
120
5
.
21.
Chawanpaiboon
S
,
Vogel
JP
,
Moller
A-B
,
Lumbiganon
P
,
Petzold
M
,
Hogan
D
, et al
.
Global, regional, and national estimates of levels of preterm birth in 2014: a systematic review and modelling analysis
.
Lancet Glob Health
.
2019
;
7
(
1
):
e37
46
.
22.
Cheong
JL
,
Doyle
LW
,
Burnett
AC
,
Lee
KJ
,
Walsh
JM
,
Potter
CR
, et al
.
Association between moderate and late preterm birth and neurodevelopment and social-emotional development at age 2 years
.
JAMA Pediatr
.
2017
;
171
(
4
):
e164805
.
23.
Townley Flores
C
,
Gerstein
A
,
Phibbs
CS
,
Sanders
LM
.
Short-term and long-term educational outcomes of infants born moderately and late preterm
.
J Pediatr
.
2021
;
232
:
31
7.e2
.
24.
Harris
PA
,
Taylor
R
,
Thielke
R
,
Payne
J
,
Gonzalez
N
,
Conde
JG
.
Research electronic data capture (REDCap): a metadata-driven methodology and workflow process for providing translational research informatics support
.
J Biomed Inform
.
2009
;
42
(
2
):
377
81
.
25.
Kohl
M
.
MKpower: power analaysis and sample size calculation
. R Package Version 0.72023.
26.
R Core Team
.
R: a language and environment for statistical computing
.
Vienna, Austria
:
R Foundation for Statistical Computing
.
27.
Sjoberg
D
,
Whiting
K
,
Curry
M
,
Lavery
JA
,
Larmarange
J
.
Reproducible summary tables with the gtsummary package
.
R J
.
2021
;
13
(
1
):
570
80
.
28.
van Vonderen
JJ
,
Narayen
NE
,
Walther
FJ
,
Siew
ML
,
Davis
PG
,
Hooper
SB
, et al
.
The administration of 100% oxygen and respiratory drive in very preterm infants at birth
.
PLoS One
.
2013
;
8
(
10
):
e76898
.
29.
von Elm
E
,
Altman
DG
,
Egger
M
,
Pocock
SJ
,
Gøtzsche
PC
,
Vandenbroucke
JP
, et al
.
The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies
.
Lancet
.
2007
;
370
(
9596
):
1453
7
.
30.
Jakobsen
JC
,
Gluud
C
,
Wetterslev
J
,
Winkel
P
.
When and how should multiple imputation be used for handling missing data in randomised clinical trials: a practical guide with flowcharts
.
BMC Med Res Methodol
.
2017
;
17
(
1
):
162
.
31.
Saugstad
OD
,
Ramji
S
,
Soll
RF
,
Vento
M
.
Resuscitation of newborn infants with 21% or 100% oxygen: an updated systematic review and meta-analysis
.
Neonatology
.
2008
;
94
(
3
):
176
82
.
32.
Oei
JL
,
Kapadia
V
,
Rabi
Y
.
Meta-analysis of neurodevelopmental outcomes after randomization of preterm infants to resuscitation with lower (FiO2 < 0.3) or higher (FiO2 > 0.6) oxygem strategies: are boys and girls equal? Proceeds of the Pediatrics Academic Societies Meeting
.
Toronto, Canada
;
2018
.
33.
Ohuma
EO
,
Moller
A-B
,
Bradley
E
,
Chakwera
S
,
Hussain-Alkhateeb
L
,
Lewin
A
, et al
.
National, regional, and global estimates of preterm birth in 2020, with trends from 2010: a systematic analysis
.
Lancet
.
2023
;
402
(
10409
):
1261
71
.
34.
Handley
SC
,
Salazar
EG
,
Greenberg
LT
,
Foglia
EE
,
Lorch
SA
,
Edwards
EM
.
Variation and temporal trends in delivery room management of moderate and late preterm infants
.
Pediatrics
.
2022
;
150
(
2
):
e2021055994
.
35.
Walker
DW
,
Lee
B
,
Nitsos
I
.
Effect of hypoxia on respiratory activity in the foetus
.
Clin Exp Pharmacol Physiol
.
2000
;
27
(
1–2
):
110
3
.
36.
Dekker
J
,
Martherus
T
,
Lopriore
E
,
Giera
M
,
McGillick
EV
,
Hutten
J
, et al
.
The effect of initial high vs. Low FiO2 on breathing effort in preterm infants at birth: a randomized controlled trial
.
Front Pediatr
.
2019
;
7
:
504
.
37.
Phillipos
E
,
Solevåg
AL
,
Aziz
K
,
van Os
S
,
Pichler
G
,
O'Reilly
M
, et al
.
Oxygen saturation and heart rate ranges in very preterm infants requiring respiratory support at birth
.
J Pediatr
.
2017
;
182
:
41
6.e2
.
38.
Proschan
MA
,
Waclawiw
MA
.
Practical guidelines for multiplicity adjustment in clinical trials
.
Control Clin Trials
.
2000
;
21
(
6
):
527
39
.
39.
Pallmann
P
,
Bedding
AW
,
Choodari-Oskooei
B
,
Dimairo
M
,
Flight
L
,
Hampson
LV
, et al
.
Adaptive designs in clinical trials: why use them, and how to run and report them
.
BMC Med
.
2018
;
16
(
1
):
29
.
40.
Katheria
A
,
Schmölzer
GM
,
Janvier
A
,
Kapadia
V
,
Saugstad
OD
,
Vento
M
, et al
.
A narrative review of the rationale for conducting neonatal emergency studies with a waived or deferred consent approach
.
Neonatology
.
2023
;
120
(
3
):
344
52
.
41.
Seidler
AL
,
Hunter
KE
,
Cheyne
S
,
Ghersi
D
,
Berlin
JA
,
Askie
L
.
A guide to prospective meta-analysis
.
BMJ
.
2019
;
367
:
l5342
.
42.
Sotiropoulos
JX
,
Oei
JL
,
Schmölzer
GM
,
Hunter
KE
,
Williams
JG
,
Webster
AC
, et al
.
NETwork meta-analysis of trials of initial oxygen in preterm newborns (NETMOTION): a protocol for systematic review and individual participant data network meta-analysis of preterm infants <32 Weeks’ gestation randomized to initial oxygen concentration for resuscitation
.
Neonatology
.
2022
;
119
(
4
):
517
24
.
43.
Sotiropoulos
JX
.
NETwork Meta-analysis of Trials of Initial Oxygen in preterm Newborns (NETMOTION): a systematic review and individual participant data network meta-analysis
. In: Proceedings of pediatric academic societies (PAS) annual meeting.
Washington DC, USA
;
2023
; p.
1418056
.