But the earlier suggestion that the treatment might protect against cognitive disabilities in the children — an observation made when they were about 18 months old — no longer seems to be true when the children reach the age of five, the researchers reported Tuesday in the Journal of the American Medical Association.
By five, children in both arms of the study — those who had received caffeine when they were preemies and those who had not — scored at about the same range on intelligence tests. And both were near the norm for their age group, a heartening finding, the lead author and others suggested.
While interpreting the results is a bit complex, Dr. Barbara Schmidt said they support continued use of the treatment.
"I think clinicians should continue to use caffeine to reduce apnea, to improve regular breathing and to wean babies off artificial ventilator support," said Schmidt, director of clinical research in the department of pediatrics at the Hospital of the University of Pennsylvania in Philadelphia.
The study is a follow-up to a landmark Canadian effort to assess the safety of a made-in-Montreal treatment that has long been used, but for which safety studies had not been previously done.
The study is funded by the Canadian Institutes of Health Research, though it enrolled infants from academic hospitals in Australia, Europe and Israel, as well as Canada. Schmidt, the principle investigator, got the funding when she was at McMaster University in Hamilton. She moved to the University of Pennsylvania in 2007.
In the late 1970s, Dr. Jack Aranda, who was then at McGill University in Montreal, started using caffeine in preemies, figuring the stimulant would help combat the apnea — sporadic arrests in breathing — commonly seen in babies born before their lungs are fully developed.
The practice became commonplace and remains so. To prevent apnea, preemies are given a starting dose of caffeine equivalent to the amount that would be found in between four and six cups of coffee. That level of caffeine is topped up daily.
But by the 1990s, questions were being asked about the safety of the treatment. Studies in animals suggested caffeine might be toxic for developing brains, explained Dr. Boriana Parvez, medical director of the neonatal intensive care unit at Toronto's Hospital for Sick Children.
In essence, people worried that caffeine, especially in such high doses, might solve a short-term problem but create longer-term ones.
The findings to date dispel those worries. Aranda, now with the Children's Hospital of SUNY (State University of New York) Downstate, in Brooklyn, follows developments of the study with interest. He's not surprised by the positive findings, but is delighted.
"I'm so proud she did it," he said of Schmidt's work.
In the early days, when the babies were still in neonatal intensive care units, those on caffeine had lower rates of chronic lung disease and of serious eye disease related to prematurity. They also spent less time on a ventilator (breathing machine) and less time on supplemental oxygen.
At 18 months of age, fewer babies in the caffeine group had cerebral palsy and other motor function disabilities. In fact, it looked like caffeine was neuro-protective — it appeared the babies who had received it had fewer cognitive delays than the children who hadn't received the therapy.
Those findings were published in 2006 in the New England Journal of Medicine and they changed medical practice, Parvez said. Hospitals that hadn't been using caffeine switched over to it.
In this latest report, the researchers no longer see a difference in intelligence test scores between the two groups as the children age. But that doesn't negate the benefits of the therapy in the earlier stages, or the longer-term motor function benefits, Schmidt said.
"There is no other drug that we give in the neonatal unit that has ever been shown to have any benefits on any aspect of neurodevelopment up to five years," she said.
Parvez, who was not involved in the study, said she still sees evidence of neuro-protection in the caffeine group, noting the better motor function and better visual perception observed in children who had the treatment are signs of neurological gains.
As well, Parvez suggested it will be telling to follow the children over time — which the researchers are doing — saying the tasks asked of older children on standardized tests may show that the better motor and visual perception skills lead to a divergence of scores as the two groups of children get older.
Even if they don't, the fact that the work shows at age five that these children are close to the norm for peers who were born full term is welcome news.
"It's a message of hope," Schmidt said. "It shows that even though very pre-term birth increases the risk of impairment — that's a fact — that some children overcome, especially the cognitive impairment that may be diagnosed early in life."
She said the group is now studying the children at ages 11 and 12.
Note to readers: This is a corrected story. The funding agency is the Canadian Institutes of Health Research, not "for" Health Research.