This article is more than 5 years old. Information may no longer be current.
Lower oxygen-saturation targets increase risk for death among preterm infants
Click Here to Manage Email Alerts
In a combined analysis of two trials, reduced oxygen-saturation targets resulted in significantly higher rates of death and disability at age 2 years in preterm infants, when compared with higher targets, according to research in the New England Journal of Medicine.
“These trials, known collectively as the Neonatal Oxygen Prospective Meta-analysis Collaboration, were designed to compare the effects of a lower oxygen-saturation target range vs. a higher target range on a primary outcome of death or major disability at 18 to 24 months, with age corrected for prematurity,” William Tarnow-Mordi, MB, ChB, professor of neonatal medicine at Sydney Medical School, University of Sydney, and colleagues wrote.
The researchers conducted two clinical trials with 2,108 preterm infants in Australia and the United Kingdom between 2006 and 2010. In each trial, neonates born at 28 weeks’ gestation were assigned to either a lower oxygen-saturation target of 85% to 89%, or a higher oxygen-saturation target of 91% to 95%. The researchers evaluated death and disability at age 2 years as the primary outcome.
When each trial was independently evaluated, death rates were insignificantly higher among children in the lower saturation target group. However, when both trials were combined for post hoc analyses, death as well as the combined outcome of death and disability at age 2 years were significantly higher in the lower saturation target groups.
In the Australian trial, death or disability occurred in 45% of infants in the lower target group, compared with 39.8% in the higher target group. Likewise, in the U.K. trial, death or disability occurred in 50.5% of infants in the lower target group, vs. 45.9% in the higher target group. These statistically insignificant results gained significance when the data was combined into a single analysis, with death and disability in 48.1% of all lower target group participants, compared with 43.1% of higher target group infants (P = .02).
“Targeting an intermediate oxygen-saturation range, such as 87% to 93%, vs. a higher range is an untested practice that may increase mortality, because current oximeters permit increasingly disproportionate exposure to hypoxemia as oxygen saturation decreases to below 93%,” Tarnow-Mordi and colleagues wrote. “At present, the most rigorously evaluated evidence for policy is that targeting an oxygen saturation of 91% to 95% is safer than targeting an oxygen saturation of 85% to 89%.” – by David Costill
Disclosure: Tarnow-Mordi reports no relevant financial disclosures. Please see the full study for a list of all other authors’ relevant financial disclosures.
Perspective
Back to Top
David A. Kaufman, MD
Neonatologists and parents of preterm infants want to know what is the safest oxygen saturation range for infants born before 28 weeks gestation.
In their most recent publication, BOOST II reported their primary outcome of morality or 2-year developmental impairment comparing two different oxygen saturation ranges. Two large studies performed in Australia and the UK (BOOST II) is part of 5 large studies with similar design. The goal of these studies is to see if assigning oxygen saturations to either 85-90% or 91-95% effects morbidity and mortality when used until 36 weeks’ postmenstrual age (PMA). These five studies deserve applause in doing a good job collaborating together.
First, one has to know that initially the oximeters had an error in what was displayed, so the revised oximeter data gives a better prediction of how oximeters are programmed in the NICU currently. Without the revision, the lower target group’s oximeters displayed approximately 30% of the time outside their actual range. Specifically values above 87% were shifted up by 1 to 2 percentage points. The company has provided revised software to correct this inaccuracy.
From the revised oximeters, the primary outcome of death OR disability at 2 years was increased in the lower-target group (85-89%) compared with the higher-target group (91-95%). If you break down the outcomes, there was a difference between groups regarding mortality, but there was no difference in disability between the groups. When examining mortality data of the revised oximeters, more infants (24.5%) died in the lower-target group (85-89%) compared to 16.9% in the higher-target group (91-95%). Of note, the UK BOOST II study had the highest overall mortality (23%) compared to 15-17% in three of the other studies and 20% in the SUPPORT study. The Canadian Oxygen Trial (COT) which had an overall mortality rate of 15%, did not find a difference in mortality between groups. This raises the question if the higher oxygen range is only beneficial for centers with higher mortality rates?
Some researchers argue that you need to include death in the combined outcome because these patients are not able to be tested. It falsely makes the assumption, that if those patients survived they would have disability.
Mortality needs to be treated separately for several reasons. Mortality may be unrelated to the study intervention, therefore overestimating negative outcomes in both groups. For infants <28 weeks gestation, several hurdles need to be overcome for survival as death can occur due to not just one, but several complications of prematurity, including, respiratory failure, bacterial infection, necrotizing enterocolitis, focal bowel perforation, severe chronic lung disease and intracranial pathology. Now in these studies it is possible that hypoxemia or targeting a lower saturation range may increase mortality related to several different disease processes.
What is the cause of mortality related to hypoxemia in these studies? The studies were not designed and powered to be able to pinpoint which diseases or complications were responsible for the increased mortality. In BOOST II supplemental tables, the primary causes of death were analyzed. For the BOOST II UK revised criteria, NEC, sepsis and pneumonia were all twice as high in the lower vs. higher saturation target group. Comparatively in all the BOOST II study patients, respiratory distress syndrome and pulmonary hemorrhage was twice as high for the higher saturation target group. The category of “other” was the most common choice and twice as high for the lower saturation target group for all the BOOST II patients. A priori definitions delineating the different cause of mortality are variable and often lacking specific definitions. This is not unique to these studies and neonatology needs studies simply validating definitions or processes to define causes of death.
Finally, now that we have data studying one model of keeping the saturation range the same from birth until 36 weeks PMA, should other models such as changing the oxygen target range every 4 weeks or so be studied?
Perspective
Back to Top
Rangasamy Ramanathan, MD
According to the study, use of lower oxygen saturation targeting 85-89% vs. 91-95% resulted in a nonsignificantly higher rates of death or disability at 2 years of age. However, in the COT study, published in JAMA in 2013, no increase in mortality or disability at 18 months was seen. Study authors recommend targeting oxygen saturation of 91-95%. Though, in all these studies, severe retinopathy of prematurity (ROP) and bronchopulmonary dysplasia were significantly higher in the high oxygen saturation group.
ROP is the leading cause of blindness in children globally. ROP is a biphasic disease, with phase 1 due to hyperoxia-mediated vasoobliteration and phase 2 due to hypoxia of the metabolically active, maturing retina leading to vasoproliferation. Recently, adoption of high oxygen saturation (91-95%) resulted in a significant increase in severe ROP, without a decrease in mortality and the number needed to harm (NNH) for stage III or worse ROP was 25, consistent with SUPPORT (NNH 11 for severe ROP) and BOOST-II (NNH 35 for ROP needing treatment).
Another issue with these studies is that there is no data on hemoglobin levels at different postnatal ages. Typically, clinicians tend to accept lower hemoglobin values with advancing postnatal age. Therefore, oxygen delivery is significantly decreased in the low oxygen saturation target group when ROP enters phase 2. Based on the pathophysiology of ROP, it may not be the best strategy to target low or high oxygen saturation during phase 1 and phase 2 ROP.
Chen ML et al., published a meta-analysis of studies using early-low versus late-high oxygen saturation and concluded that this approach reduced the risk for severe ROP without an increase in mortality. They also recommended a sufficiently powered randomized, controlled trial on optimal oxygen delivery in the early and late stages of ROP and long-term visual, pulmonary and neurodevelopment follow-up.
References:
1. Schmidt B, Whyte RK, Asztalos EV, et al; Canadian Oxygen Trial (COT) Group. Effects of targeting higher vs lower arterial oxygen saturations on death or disability in extremely preterm infants: a randomized clinical trial. JAMA 2013; doi:10.1001/jama.2013.5555.
2. Manley BJ, Kuschel CA, Elder JE, Doyle LW, Davis PG. Higher rates of retinopathy of prematurity after increasing oxygen saturation targets for very preterm infants: experience in a single center. J Pediatr 2015; doi: 10.1016/j.jpeds.2015.10.005.
3. Chen ML, Guo L, Smith LE, Dammann CE, Dammann O. High or low oxygen saturation and severe retinopathy of prematurity: a meta-analysis. Pediatrics 2010; doi: 10.1542/peds.2009-2218.
Click Here to Manage Email Alerts