"If the parents of these premature babies had been informed of the risk ... many parents would have said, 'We don't want our babies in this research,'" said Michael Carome, deputy director of Public Citizen's Health Research Group. Given how many universities were involved with the study, he said, "clearly there was a systemwide failure."
The U. was one of 23 academic medical center who participated in the SUPPORT study, which tested more than 1,300 babies between 2004 and 2009. The University of Alabama at Birmingham was the lead institution in the study, and it received the government's notice of a violation.
The study began because more hospitals were reducing the amount of oxygen given to premature babies to lower the risk of severe eye damage, or retinopathy, which is a known risk of treating premature babies with oxygen, said Bradley Yoder, one of the Utah authors. Retinopathy increased in the United States starting the 1940s, and it's still a serious issue more than half of the babies weighing less than 2 3/4 pounds at birth will have at least a mild form of the disorder.
But lowering the levels of oxygen to avoid eye damage can also increase the risk of brain damage and death.
As monitoring technology improved, lowering the oxygen levels "was becoming very common," said Yoder. "A lot of people said we should be lowering and lowering without any scientific evidence."
Under standard care, doctors were assessing individual babies and generally keeping their blood oxygen saturation the percentage of hemoglobin with oxygen bound to it between 85 and 95 percent.
In the study, babies were chosen randomly to either be kept in a "low" range, at 85 to 89 percent, or a "high" range, at 91 to 95 percent.
But according to the federal government, the risks of the different levels were already known and the researchers should have told parents that when asking for their consent.
"The researchers had sufficient available information to know, before conducting the study, that participation might lead to differences in whether an infant survived, or developed blindness, in comparison to what might have happened to a child that had not been enrolled in the study," according to the March 7 violation letter.
According to the results, published in the New England Journal of Medicine in 2010:
• Among infants treated with low oxygen, only 41 out of 475 developed severe retinopathy, or 8.6 percent. In the high oxygen group, the rate was 17.9 percent.
• In the low oxygen group, 130 out of the 654 infants died, or 19.9 percent. In the high oxygen group, 16.2 percent died.
Those results indicate that oxygen concentrations shouldn't fall below about 88 percent, Yoder said. Before those conclusions, researchers contend it wasn't clear how the different oxygen levels, both considered within the safe range, would affect children.
The risks "had largely been brought up in the '50s and the '60s ... they did not reflect the more recent results of more robust studies," said Roger Faix, another U. author on the study. He pointed out the subjects were "very sick, very tiny babies," and said the mortality rates of infants in the study were actually lower than infants who weren't part of it.
But those infants weren't a control group in the study, and Carome said it's "disingenuous" to compare the two because the babies who were tested most about 26 weeks old started out healthier than those who weren't part of the study.
The violation notice is one of handful the government issues every year, he said. "The severity of these deficiencies and the implications are significant," Carome said, especially considering the vulnerable nature of the subjects. It's unclear whether the children would have survived unscathed if they weren't part of the study, he said.
The investigation started after the government received a complaint two years ago. The authors of the study have responded to the violation finding, but have not yet received a final response from the government, Faix said.
Meanwhile, the University of Utah's Institutional Review Board is doing its own investigation, and plans to have results in about a month.
"There aren't any hard and fast rules" about consent forms, said Jeffrey Botkin, the associate vice president for research integrity at the U. The consent form in the SUPPORT study was developed by the National Institute of Child Health and Human Development with the input of the nearly two dozen schools that did the research. Each institution could then alter the 2 1/2-page form to suit their individual policies.
"We'll look at our own consent forms that we approved and make a determination whether any mistakes were made," Botkin said, "and, if so, how we want to respond to those."
Premature babies, oxygen and blindness
Retinopathy of prematurity, or ROP, is caused by abnormal blood vessel growth in the eye. Scar tissue that builds up around the retina can cause blindness.
A sudden surge: Doctors treating premature infants with oxygen in the 1940s saw a dramatic increase in the previously rare disorder.
Breakthrough: Controversial research in the 1950s showed infants who received a lower level of oxygen had a much lower incidence of ROP than those receiving the then-standard higher levels. As doctors used less oxygen, the number of blind American children dropped by 60 percent.
New concerns: Research in the 1970s indicated the lower levels may have led to increased deaths. One analysis concluded: "Each sighted baby gained [by limiting the use of oxygen] may have cost some 16 deaths." Doctors began treating some premature infants with higher levels of oxygen.
A continuing challenge: As new technology saves premature infants with ever-lower weights, it appears ROP has grown again.
The impact: Of the 28,000 premature infants born weighing less than 2 ¾ pounds each year in the U.S., more than half have at least a mild form of ROP. More than 1,000 require treatment. About 400 to 600 become legally blind.
Source: Department of Health and Human Services, National Eye Institute