Managing oxygen levels in preterm infants

"Until more is known on morbidity and mortality with low oxygen saturation (SaO₂) targets during the first few weeks of life, most neonatologists say it is probably best to aim for a saturation of at least 85% to 88%, and probably lower than 95%, to avoid wide swings in oxy­gen levels," said Dr Hartnett, professor of ophthalmology, University of Utah, Salt Lake City, Utah, USA.

"Research using suitable models will hopefully help us understand the molecular pathomechanisms of ROP and allow the identification of safer and more effective strategies that perhaps can be used in combination with oxygen to maintain overall health in developing preterm infants," she added.

The Extremely Low Gestational Age Newborn study evaluated associations between ROP and oxygen within the first 3 days of life to test the hypothesis that blood gas derangements during this period would increase preterm infants' risk for severe ROP, Dr Hartnett explained.

Its results showed that being in the highest quartile of PaO₂ or PCO₂ as well as being in the lowest quartile of pH increased the risk of zone I ROP and severe ROP.

"This study provided clinical evidence corroborating clinical observations and evidence from animal models reported years earlier by Ashton, Michaelson and Patz," Dr Hartnett continued. "Based on findings from several studies, it is now believed that oxygen fluctuations during an infant's course in the neonatal nursery are associated with increased risk of ROP.

"We know there are many factors that can cause tissue oxygen fluctuation in preterm infants," she said.

Investigating the role of supplemental oxygen, the Supplemental Therapy with Oxygen to Prevent ROP (STOP-ROP) compared delivery of supplemental oxygen to maintain SaO₂ of 96% to 99% versus conventional levels of 89% to 94% in infants diagnosed with prethreshold ROP.

While the working hypothesis was that higher oxygen would reduce the risk of progression to threshold disease, there was no difference between study groups for that outcome.

Post hoc analysis did show that higher oxygen reduced the risk of threshold disease in a subgroup of infants with no plus disease.

Oxygen saturation ranges

"More recently, smaller clinical studies and trials showed that lower oxygen saturation was associated with reduced risk of severe ROP," Dr Hartnett reported.

"The conflict between findings from STOP-ROP and these other studies help to lay the framework for several recently reported clinical trials to test two different oxygen saturation ranges on risk of ROP in preterm infants," she said.

The latter studies include the Surfactant, Positive Airway Pressure, Pulse Oximetry Randomized Trial (SUPPORT) and Benefits of Oxygen Saturation Targeting Study II (BOOST II).

Their findings suggested that higher oxygen is associated with better infant survival.

SUPPORT compared SaO₂ targets of 85% to 89% versus 91% to 95% in infants born from 24 to 28 weeks gestational age who were intubated and given surfactant within 1 hour of birth or treated with continuous positive airway pressure. Its results showed the lower SaO₂ group had a higher risk of death before discharge.

However, among the surviving infants within that group, the risk of developing severe ROP was lower in the low SaO₂ group than among those targeted for 91% to 95%.

BOOST II, which was conducted in the UK and Australia, compared the same oxygen targets in infants born under 28 weeks gestational age and also found poorer survival in the lower oxygen group.

More recently, the Canadian Oxygen Trial (COT) found no difference in ROP development comparing the same oxygen ranges.

Dr Hartnett pointed out that SUPPORT, BOOST II and COT had important methodological differences.

"The studies varied with respect to how alarms were set and the locations from which infants were enrolled," Dr Hartnett concluded. "Site of enrolment is especially important to consider as there are phenotypic differences in ROP throughout the world based on the availability of resources to provide optimal prenatal and neonatal care, and possible genetic influences."