The findings will be a relief to blood collection agencies, which have faced calls to shorten the length of time blood can be stored before it is transfused. In fact, Canadian Blood Services said they were very pleased with the outcome.
"The study supports our current inventory management practices for helping patients receiving transfusions in the intensive care setting," said Dr. Dana Devine, the agency's chief medical and scientific officer.
Currently blood can be stored for up to 42 days, though most transfusions involve blood that is about three weeks old. But there has been a growing belief that fresher is better.
"When you look at all that evidence, over time it was building pressure on the blood system that fresh was better, that we need to perhaps change policy," said one of the lead authors, Dean Fergusson, a scientist at the Ottawa Hospital Research Institute.
Why? For one thing, blood changes as it ages; those changes can be seen under a microscope. The assumption has been that those changes would have an impact. Some animal studies and even observational studies in people have suggested that is likely true.
In fact, a large observational study published in 2008 in the New England Journal of Medicine suggested cardiac surgery patients who were given blood that had been stored for more than two weeks had a 30 per cent increased risk of dying after surgery compared to those who received blood that had been stored for less than two weeks.
But observational studies, which try to tease out patterns by watching what happens when people do or take certain things, can't prove one thing causes another. That takes a randomized controlled trial, which is what Fergusson and his colleagues conducted.
Their work involved nearly 2,500 patients in intensive care units in Canada, Britain, France, the Netherlands and Belgium. The study, funded by the Canadian Institutes of Health Research, was published online Tuesday by the New England Journal of Medicine.
ICU patients who were expected to need transfusions were randomly assigned to get blood that had been stored for fewer than eight days or the blood that would normally be sent from the blood bank. To minimize waste, blood banks will grab the oldest usable blood in the fridge when a transfusion order comes in.
The scientists hypothesized that those who received the fresher blood would do better. But when they monitored the patients they found no real differences between the two groups.
The number of deaths was essentially the same. There were no differences in terms of the length of their hospital stays, the rates of major secondary illnesses they suffered, or other important health measures.
Fergusson had previously conducted a similarly designed trial in premature infants. It too found fresh blood was not better. And a yet-to-be published randomized trial in cardiac surgery patients also supports these findings.
Dr. Bill Flegel said the latest study convincingly shows that there are not huge benefits from transfusing fresher blood.
"And that's very good news," said Flegel, who is chief of the laboratory services section in the department of transfusion medicine at the U.S. National Institutes of Health's Clinical Center in Bethesda, Md.
But Flegel noted that because of the way the trial was designed, it could not detect if there was a survival benefit that was less than five per cent. So if older blood was associated with a four per cent increased risk of dying, the study would not capture that effect. A four per cent difference would be clinically important, he noted.
Flegel said several other trials, including a larger Canadian-led one, are still in the works and one may answer the question of whether a smaller benefit exists.