The use of oxygen therapy for preterm infants was introduced in the 1940s, only to cause some degree of heartbreak when thousands of the so-treated children went blind. The cause was attributed to overly liberal use of oxygen therapy, leading to retinopathy of prematurity (ROP). The health of preterm infants in following decades has continued to teeter between too little oxygen and too much – but setting the balance to even the scale has proved a tremendous challenge.
Exactly a decade ago Lisa Askie, a researcher at the University of Sydney, Australia, observed, “Although oxygen is probably the most common therapy delivered to small or sick newborns in the past 75 years, what constitutes appropriate oxygenation for these infants remains highly controversial . The major unanswered question that requires urgent resolution is the important long-term effects of targeting lower versus higher oxygen saturation ranges during the first days/weeks of a preterm infant’s life. Such an important clinical question can only be answered rigorously and definitively using the randomised trial methodology.”1
While the answer has continued to be shrouded in disagreement and some degree of controversy, new understanding is emerging by way of precisely what the researcher requested – randomized trials.
“There were three clinical trials published in the last couple of years all addressing the same question: What is the best oxygenation range for extreme premature infants,” explained Eduardo Bancalari, MD, director of the division of neonatology at the Miller School of Medicine at the University of Miami, and chief of newborn service at Jackson Memorial Hospital, Miami. “The three studies used similar protocols and evaluated similar populations of preterm infants. The Surfactant, Positive Pressure and Pulse Oximetry Randomized Trial (SUPPORT) conducted in the United States by the NICHD Neonatal Research Network and the Benefits of Oxygen Saturation Targeting (BOOST) II trial conducted in Australia, New Zealand and the United Kingdom showed that targeting oxygen saturations between 85% and 89% compared to 91% to 95% reduced the rate of severe ROP.
“Unexpectedly, these trials also showed lower survival in the group of infants randomized to the lower saturation range,” said Bancalari. “The third trial, the Canadian Oxygen Trial (COT) conducted in Canada, Europe, United States and Latin America, showed that rates of survival, ROP and bronchopulmonary dysplasia (BPD) did not differ significantly between the lower and higher SpO2 target ranges but survival, ROP and BPD had the same general trend observed in SUPPORT and BOOST II trials.”
In May 2013, when findings from the BOOST II Collaborative Groups were published in the New England Journal of Medicine (Oxygen Saturation and Outcomes in Preterm Infants, lead author Ben J. Stenson, MD, Royal Infirmary of Edinburgh) the conclusion was somewhat startling. In three trials the BOOST researchers had evaluated the effects of targeting oxygen saturation of 85% to 89% compared to 91% to 95% on disability-free survival at two years in 2448 infants born before 28 weeks gestation. Researchers found the rate of death “was significantly higher in the lower-target group than the higher target group,” however, “those in the lower-target group for oxygen saturation has a reduced rate of retinopathy prematurity . and an increased rate of necrotizing enterocoloitis.”
They concluded, “Targeting an oxygen saturation below 90% with the use of current oximeters in extremely preterm infants was associated with an increased risk of death . Our findings strongly favor the avoidance of targeting an oxygen saturation of less than 90% among such infants, according to readings on current oximeters.”
A Different Finding
Barbara Schmidt, MD, MSc, of the Children’s Hospital of Philadelphia and the Hospital of the University of Pennsylvania, Philadelphia, was first author of the COT study, Effects of Targeting Higher vs Lower Arterial Oxygen Saturations on Death or Disability in Extremely Preterm Infants, also published May 2013, by JAMA. These findings were somewhat different.
In the trial, 1201 infants were enrolled within 24 hours of birth between 2006 and 2010, and were re-assessed .between 2008 and 2012. The infants were assigned to treatment groups and their caregivers were given pulse oximeters that displayed saturations of either 3% above or below the true values but returned to true values below 84% and above 96%. Caregivers were instructed to adjust the concentration of oxygen to maintain saturation levels between 88% and 92%, which produced two treatment groups with true target saturations of 85% to 89% or 91% to 95%. Alarms sounded if the displayed saturation level went to, or below, 86% or increased to 94%. Oxygen saturation data was stored every 10 seconds.
“In 1201 COT babies, targeting oxygen saturations of 85% to 89% as compared with 91% to 95% had no significant effect on the rate of death or disability at 18 months, death before 18 months, and the rate of severe ROP,” Schmidt told ADVANCE. “We performed very frequent audits during the COT trial. We observed less hyperoxia and less hypoxia in COT than the investigators of other similar trials. RTs and nurses are not walking around the neonatal ICU with target ranges in their mind. They respond to alarms. The alarm settings are very important and need to be audited frequently. This may explain why we had neither excess mortality in the lower saturation target group nor excess retinopathy in the higher target group.”
Schmidt said findings have been reflected in practice at participating facilities. “We polled our 25 international study centers at the end of November 2013 and asked for their current alarm settings for extremely preterm infants: While in supplemental oxygen, the vast majority of these centers are currently using upper alarm settings between 93% and 95%. 20 of the 25 centers are currently using lower alarm settings, between 85% and 88%. The COT results support those practice choices.”
When Askie revisited the topic in a paper, Optimal oxygen saturations in preterm infants: a moving target, published April 2013 in Current Opinions in Pediatrics, she remained unconvinced of the precision of the various trials’ findings. She noted, “A recently published systematic review summarized the existing evidence for currently used oxygen saturation targets in preterm infants and highlighted the paucity of randomized trials addressing this topic. It appears that higher oxygen saturation levels increase the risk of severe retinopathy of prematurity and pulmonary morbidities. However, data regarding the effects of various target ranges on early mortality and long-term neurodevelopmental outcomes are lacking. A collaborative group of investigators from five independent randomized trials was established to answer this question definitively. Although the final analysis will not be available until 2014, interim results . revealed an increase in early mortality when the lower oxygen saturation range is targeted. At present, it may be prudent not to target oxygen saturation levels below 90%. Whatever the optimal range, consistently maintaining the newborn’s oxygen saturation levels within target proves an additional challenge for providers. However, ongoing investigative collaborations may soon provide guidance.”
Bancalari further detailed the continuing challenge. “The findings of these randomized trials have placed neonatal clinicians in a very difficult conundrum. Use of lower oxygen targets to reduce complications such as ROP and BPD that are associated with hyperoxemia in preterm infants can affect their survival and result in higher death rate. While there are still many unanswered questions regarding optimal oxygen targets in these infants, most clinicians have adopted a safe position and are recommending avoiding both extremes in oxygenation by targeting saturations between 90% and 95%,” he noted. “This may sound easy but in reality targeting a relatively narrow range in these infants is extremely challenging because premature infants have frequent fluctuations that require constant attention and frequent adjustments in FiO2. It is also not known if different targets should be used in infants with certain conditions such as ROP or pulmonary hypertension or in infants of different gestational or postnatal ages.”
While vital research continues to blossom, there are still questions to be answered and optimal saturation targets to be defined. Understanding is growing, but it is not yet fully ripe. Bancalari reminded that while oxygen in excess is toxic, especially in the more immature infants, supplemental oxygen is also essential for the intact survival in many of these babies. “Finding the right balance on oxygen therapy requires learning from past experience and adopting therapeutic strategies that are based on evidence from good prospective randomized trials like the studies mentioned here.”
Valerie Neff Newitt is on staff at ADVANCE. Contact: email@example.com.