Air or supplemental (extra) oxygen: which works better to help babies breathe at birth?

Key messages

• When full-term or near-term babies need help to start breathing at birth, using normal air instead of extra oxygen may reduce the risk of death before leaving hospital. However, we cannot be certain about this because the current studies are not of high enough quality.

• Using air instead of extra oxygen may make little or no difference to the risk of moderate or severe brain injury (hypoxic ischaemic encephalopathy (HIE)). This happens when a baby’s brain does not get enough oxygen or blood flow around the time of birth. It is diagnosed in the first hours after birth.

• The long-term effects of using air versus extra oxygen are unclear. No studies followed children past one year of age to report on neurodevelopmental outcomes (long-term brain development or disability) according to widely accepted and rigorous definitions. Two studies looked at children up to two years of age. However, they used less reliable assessment methods, meaning we remain very uncertain about whether using room air instead of extra oxygen makes any real difference.

What happens when a baby has difficulty breathing at birth?

Most babies begin to breathe on their own immediately after birth. This allows their lungs to fill with air and their body to switch from getting oxygen through the placenta to getting it from their own lungs. However, about one in 10 babies born at or near their due date (term or late preterm) need help to start breathing. If a baby cannot breathe effectively, they may suffer from asphyxia. This is a condition where the body — and especially the brain — does not get enough oxygen. Asphyxia is a serious issue that can lead to death or long-term brain injury.

How is this condition treated?

Healthcare workers often use a mask to provide "breathing support" (positive pressure ventilation) to babies having difficulties expanding their lungs. During this process, they can use air (which contains 21% oxygen) or supplemental oxygen (extra oxygen that can be up to 100% concentration).

What did we want to find out?

We wanted to find out whether using normal air was better than extra oxygen for helping full-term and near-term babies to start breathing at birth. We specifically looked at whether normal air improved the chances of:

• the baby surviving (reducing deaths before leaving the hospital); and

• avoiding brain injury (specifically moderate to severe HIE).

We also wanted to find out if using air was associated with any unwanted effects, such as long-term problems with a child's development or physical abilities.

What did we do?

We searched for studies that compared air with extra oxygen given to babies born at or near their due date who needed help to start breathing. We compared and summarised the results of these studies. We also rated our confidence in the evidence based on factors such as how the studies were conducted and the number of babies involved.

What did we find?

We found 12 studies that involved 2054 babies born on their due date, or up to five weeks earlier. These studies were conducted in both high-income and low- to middle-income countries. All studies compared air (21%) to 100% oxygen to help babies breathe at birth.

Our findings suggest that using air instead of 100% oxygen may reduce the number of deaths before babies leave the hospital. However, using air likely makes little or no difference to the risk of moderate or severe brain injury (HIE). None of the studies followed children for long enough to report on moderate or severe long-term problems with development or disability after the babies reached one year of age, at least not according to widely accepted and rigorous definitions.

Overall, we are not yet certain about the effects on these important health outcomes because the available evidence is limited and comes from studies that were not always well-conducted.

What are the limitations of the evidence?

We have little confidence in these results for several reasons. Most of the studies were designed or carried out in ways that may have made the results less accurate. For example, there were concerns about how babies were put into treatment groups and how the results were reported. The evidence is also limited because the studies were small and did not involve enough babies to give us a clear answer. While the studies compared air to 100% oxygen, none of them looked at "middle ground" oxygen levels or adjusted the oxygen based on the babies' actual needs. Because of these gaps and the small amount of data available for long-term health and brain development, we cannot be certain if these findings apply to all newborns in modern hospital settings.

How up to date is this evidence?

The evidence is up to date to January 2026.