Lately, I’ve been doing reading on retinopathy of prematurity for a section of my book. Our Gabriel was diagnosed with stage 2 retinopathy, but thankfully, he grew out of it without needing surgery. It is a very common disease for extremely-low-birthweight babies, with 80 percent of them being diagnosed with some stage of it.

A large study that was published in 2010 was called Surfactant, Positive Pressure and Pulse Oximetry Randomized Trial, or SUPPORT for short, and one of the things researchers believed at the beginning was that providing slightly less oxygen to preemies would result in lower rates of retinopathy of prematurity without an increase in mortality. The results, however, were surprising – the lower-oxygen had a higher mortality rate, at 20 percent, versus 16 percent for the higher-oxygen group. They did also have a small decrease in retinopathy, but when you’re comparing a condition that can be treated with surgery versus death, you have to go for the method that prevents death.

The study involved randomly dividing babies in to two groups, one with a goal of 85 to 89 percent blood-oxygen saturation, the other with a goal of 91 to 95 percent blood-oxygen saturation. At the time the study was planned, the American Academy of Pediatrics considered 85 to 95 percent blood-oxygen saturation the “acceptable range,” so it seemed like a valid comparison to make. The study was randomized by modifying the babies’ pulse oximeters. Parents of preemies will remember the pulse oximeter as the red glowing light on a bracelet around their baby’s foot or arm. Another family we knew called this the “ET phone home light.” The device shines a red light through the baby’s skin and then the sensor looks at the color of the blood, thus giving his or nurse instant information about how much oxygen the baby’s blood has. If the oxygen level is too low, the nurse can then increase the enriched air that the baby gets through a breathing tube or a CPAP mask. When this would happen, a bell would go off that we would call it the “desat alarm,” and the nurse would come over to the isolette to adjust things. Over five weeks of mechanical ventilation, we really, really, came to dislike the desat alarm.

In the SUPPORT test, the pulse oximeters were modified for the babies in the lower-oxygen group so that the nurses would not know which group the babies were in. This meant that if a baby in the lower-oxygen group had a blood-oxygen saturation of 88, the number on the screen would be 92. The idea was to prevent nurse bias, who would be predisposed to higher oxygen levels.

After the study was completed, a federal agency that oversees medical trials and ethics criticized the ethical planning of SUPPORT, specifically the consent forms that the parents signed to have their children enrolled in the study. Jerry Menikoff, director of the Office for Human Research Protections, told the Washington Post in an article published on April 10, 2013, “They went out of their way to tell you how your kid might benefit, but they didn’t give the flip side, which is that there is a chance your kid might end up worse.” An editorial in The New York Times on April 15, 2013 struck a similarly negative note, using the headline “An Ethical Breakdown” and called it a “startling, deplorable failure.”

These negative articles in 2013 were the first mention that SUPPORT had gotten in the national press, which made me feel kind of badly for the researchers. Before these articles, the study had only been in medical journals as the results only applied to a small group of patients, micropreemies.

Reading about the results also highlighted for us what a delicate path Gabriel had walked. The difference between 88 and 92 on the saturation level did not seem that big to us. Most of the time Gabriel was on breathing support, the nurses set up his desat alarm to go off if he got below 80, and even then, it was a minor adjustment that would then follow. If he got down to the 60 or 70 range, his skin would turn slightly blue. All of this was concerning, but we came to accept it as part of the life of a 22-week micropreemie. We, of course, wanted him to get enough oxygen, but we didn’t realize how small the target was.

If you’d like to read more, the authors of the study defended themselves in a letter to the editor in The New York Times, here, and a letter to the editor in the New England Journal of Medicine, here. Additionally, The New England Journal of Medicine published another editorial defending the study, here.