Oxygen saturation in immature babies: updated recommendations

Although 60 years ago hyperoxia was shown to be a risk factor for retinopathy of prematurity (ROP), for the last 10 years there has been a search for the optimal oxygen saturation (SpO2) target for immature and extremely low birth weight (ELBW) infants, as these babies are vulnerable to even slight hyperoxia.
Recent guidelines recommend SpO2 targets of 85-93% because available evidence indicates that low oxygen saturation is beneficial with respect to minimizing both eye and lung problems in ELBW infants. For example, a recent meta-analysis of randomized trials concluded that low saturation reduced the relative risk of severe ROP and bronchopulmonary dysplasia (BPD)/lung problems. When both randomized and non-randomized studies were assessed together, severe ROP was reduced from 20.9% to 9.5% and BPD/lung problems from 40.8% to 29.7%. This is in agreement with the results of the SUPPORT trial where babies of 24–28 weeks' gestation were randomized to low (85–89%) or high (91–95%) oxygen saturation targets. In this trial, low oxygen saturation was associated to reductions of severe ROP from 17.9% to 8.6% and BPD from 41.7% to 38.0%.
The SUPPORT trial was the only randomized trial reporting mortality data. It was therefore of concern when this study reported a significantly increased mortality in the low compared with the high saturation group (19.9% vs 16.2%). Despite these findings, many neonatologists did not change their clinical practice but waited for the results of other ongoing randomized studies using similar oxygen target limits. Very recently, the BOOST II trials were closed on recommendation of the data monitoring committees after detecting a significant increase in mortality in the lower oxygen saturation group. However, this increase in mortality was only discovered after revision of the calibration algorithm in the Masimo oximeter, which was used in all these trials.
Data from another randomized study from Canada, the COT study, are pending. Until these results and follow-up data of SUPPORT and BOOST II trials become available, SpO2 targets of 85–89% should be avoided. This recommendation may be controversial knowing that, even if mortality is slightly reduced, a higher target may lead to considerably higher rates of severe ROP and BPD in these infants. In the future, a more dynamic approach may be adopted keeping the SpO2 lower in the first weeks of life and increasing the target range after, for instance, 32 weeks' postmenstrual age during the second phase of ROP. There are experimental data indicating that hyperoxia increases oxygenation and oxidative stress in the brain, and that oxidative stress contributes to regulation of the circulation. A more comprehensive understanding of pathogenetic mechanisms leading to oxidative damage in immature infants may also lead to improved outcomes in the future. In conclusion, until further evidence becomes available, SpO2 in immature babies should not be targeted between 85– 89%.

Top