No Substitute for Real Blood

28 April 2008 (All day)

Tim Boyle/Getty Images

Use with caution.
Blood substitutes, like this one from Northfield Laboratories in Evanston, Illinois, are consistently risky, according to a new analysis.

A team of clinical trial specialists and consumer advocates has concluded that blood substitutes increase death rates by 30% and nearly triple the risk of heart attacks. Not everyone buys the findings, however, and clinical trials in the field continue.

Donated blood has a short shelf life, and the supply is often tight. For decades, researchers have tried to make blood from scratch. Although the products differ, all contain hemoglobin molecules that help deliver oxygen to the body's tissues. A number of products have been tested in trauma patients and various surgical settings, although none has yet been approved for general use in the United States or Europe. Several studies have suggested that blood substitutes may be more dangerous than real blood, but the differences haven't always been statistically significant.

Charles Natanson, a critical care specialist at the U.S. National Institutes of Health Clinical Center in Bethesda, Maryland, wondered whether combining trial data would make safety issues clearer. He teamed up with health advocates from Public Citizen, a public interest group in Washington, D.C., that has expressed grave concerns about the blood substitutes. They scanned publication databases, press releases, and material from U.S. Food and Drug Administration (FDA) meetings and came up with 16 clinical trials, some of them unpublished, of five different blood substitutes dating back to 1998.

Pooling the data, the team gauged two commonly reported risks of blood substitutes: death and heart attacks. Across the trials, 164 people out of 1927 who received fake blood died, compared to 123 out of 1784 in control groups, a risk increase of about 30%. In the treatment groups, 59 had heart attacks, compared to 16 in control cohorts, a risk increase of 2.7-fold.

No one knows what's behind the difference. But one possibility is that hemoglobin molecules in substitute blood capture nitric oxide molecules that normally dilate blood vessels and keep platelets from getting sticky. Losing nitric oxide could, in theory, cause vessels to constrict and clots to form.

Natanson is not in favor of shutting down blood substitute research. "But we need to move from humans back to animals, until we find a formulation that has less toxicity," he says, noting that FDA has continued to approve human trials of blood substitutes despite apparent safety problems. The paper, published today in the Journal of the American Medical Association, notes that five blood-substitute trials are ongoing.

The findings are cause for concern, says Dean Fergusson, an epidemiologist at the Ottawa Health Research Institute in Ontario, Canada. What struck him, he says, was that all the blood-substitute products caused harm. "That tells me we need to step back and say, 'What is going on here?' "

But others aren't convinced. "The conclusions they come up with are far overreaching," says Jonathan Jahr, an anesthesiologist at the University of California, Los Angeles, who has led several blood-substitute trials. Jahr argues that the researchers erred by bringing together blood substitutes with different track records, including one that was abandoned some time ago for safety reasons. He also notes that a 700-person orthopedic surgery trial he expects to publish in June will assuage concerns about one product, Hemopure. There were 10 deaths in patients who got substitute blood and six in those given real blood, a difference that was not statistically significant, says Jahr.

Posted In: