Reducing Target Oxygen Saturation Levels to Prevent Retinopathy of Prematurity

In recent years, several studies have found an association between lower target oxygen saturation levels in the first few weeks of life and a reduction in the incidence of severe retinopathy of prematurity (ROP). As a result, many neonatal nurseries now routinely aim for oxygen saturation levels in the upper 80s or lower 90s in lieu of the previous practice of keeping saturations in the upper 90s. It is plausible that these lower oxygen saturations are more physiologic because they more closely mimic the relatively hypoxic intrauterine environment. Although observational data suggest that policies to reduce oxygen exposure are beneficial, high-level evidence from clinical trials is needed to weigh the benefits versus the potential systemic risks of oxygen curtailment.

Brief History of Oxygen and ROP

In the 1950s, it was discovered that the incidence of ROP was very high because premature infants were exposed to unmonitored, high concentrations of oxygen. Over the ensuing years, strategies to restrict oxygen dramatically reduced incidence of ROP, but had the tragic effect of increasing the incidence of severe brain damage and death. At that time, continuous monitoring of oxygen saturation levels was not available as it is today. Gradually, many neonatologists have reduced target oxygen saturation levels in the first few weeks of life to the upper 80s or lower 90s in an attempt to reduce ROP and lung disease. It is known that oxygen contributes to severe ROP because it is a potent vasoconstrictor, and prolonged vasoconstriction can lead to vaso-obliteration and peripheral retinal ischemia, initiating a cascade of events culminating in neovascularization. Relative hyperoxia probably has its most detrimental effect on retinal vascular development in the first few weeks of life, which is before an ophthalmologist even begins looking for ROP. By the time severe ROP develops (at a mean age of 35-36 weeks postmenstrual age), it has been hypothesized that supplemental oxygen may actually be beneficial because it decreases peripheral retinal hypoxia. This premise led to the supplemental oxygen for prethreshold ROP (STOP-ROP) clinical trial, which found no statistically significant difference between standard and supplemental oxygen groups in their incidence of progression from prethreshold to threshold ROP.¹

Studies of Oxygen and ROP

Several studies have suggested that lowering target oxygen saturation levels in the first few weeks of life can reduce the incidence of severe ROP. One of the first such studies was done by Chow et al., who implemented a strict oxygen management policy with saturation limits of 85% to 93% for infants younger than 32 weeks gestation. They found a reduction in incidence of stages 3 and 4 ROP from 12.5% in 1997 to 2.5% in 2001 and a decrease in laser treatment from 4.5% to 0%.² Tin et al. compared the risks of severe ROP in infants with a target oxygen saturation range of 88-98% to those with a target range of 70-90%. The risk of severe ROP was 27.7% in the higher oxygen group versus 6.2% in the lower oxygen group.³ VanderVeen et al. reported a dramatic reduction in the incidence of prethreshold in at least one eye from 17.5% (44/251 infants) to 5.6% (4/72 infants) after lowering oximetry alarm limits from 87-97% in 2000-2003 to 85-93% after June 1, 2003.⁴ Britt and Sandoval reported a reduction in the incidence of severe ROP after institution of a policy to keep oxygen saturation between 85 and 93% for infants younger than 30 weeks gestation. The incidence of stage 3 or more advanced ROP decreased from 49% (18/37) to 15% (5/33), and the need for laser decreased from 38% (14/37) to 15% (5/33).⁵ Our group studied the effect of a more modest reduction in target oxygen saturation levels from the upper 90s to 90-96%, and we reported a small but statistically insignificant effect after adjusting for several potentially confounding variables.⁶

Observational versus Clinical Trial Data

The aforementioned observational studies certainly suggest that lower target oxygen saturation levels will be beneficial with regard to ROP. However, we do not yet know if oxygen policy changes will reduce severe ROP without increasing neurodevelopmental morbidity and death. There are several examples in the clinical sciences of observational data that strongly suggested benefits of treatment, but when a randomized clinical trial was performed, the intervention was later found to be ineffective or even harmful. One well publicized example is the detrimental effect of estrogen plus progestin on coronary heart disease and breast cancer in postmenopausal women. Observational studies lack randomization, therefore biases are introduced that can invalidate comparisons between groups receiving different treatments.

On the Horizon

Cynthia Cole, MD, MPH, has led international efforts to organize parallel, international, prospective, randomized clinical trials to study the effect of reducing target oxygen saturation levels on ROP and other morbidities, and funding for such studies has already been secured in Australia, the United Kingdom, and Canada. These trials are needed to control for potential confounding variables and to assess adverse events and long-term outcomes such as learning disabilities, cerebral palsy, and death. Until these studies are completed, we are limited to knowing that observational data strongly suggest that lowering oxygen saturation limits reduces severe ROP, but there remains the possibility of an unacceptable risk to the overall development of the premature infant.

Author Disclosure

The author states that he has no financial relationship with the manufacturer or provider of any product or service discussed in this article or with the manufacturer or provider of any competing product or service.