Blood Transfusions May Be Harmful To Some Patients

The team examined almost 24,000 records of patients in 27 countries who suffered a certain type of heart attack, and found that non-U.S. patients were 80 percent less likely to get a transfusion when undergoing non-invasive treatments, 70 percent less likely to get blood when having an invasive procedure and 60 percent less likely to undergo transfusion as a result of coronary bypass surgery – a difficult and bloody procedure where transfusion rates might be expected to be similar.

"This is interesting because the data also show that patients do pretty much the same, whether they get a transfusion or not," says Dr. Sunil Rao, a cardiologist at Duke and the lead author of the study. "We have to conclude that some of us are doing too many transfusions or others are doing too few." Rao says clinical guidelines aren't clear enough to help them figure out which approach is best.

The study, published in the January 1 issue of the American Journal of Cardiology, comes at a time when increasing evidence suggests transfusions may not only be unnecessary but may actually be harmful to some patients. In earlier studies, Duke scientists found that heart attack patients with hematocrit above 25 (hematocrit is a measure of the supply of oxygen-carrying red blood cells) were more likely to have a second heart attack and were four times more likely to die within a month if they got transfusions.

Rao says most American physicians are trained to prescribe a transfusion when a cardiac patient's hematocrit falls below 30.

"But that's not based on good science," says Rao. "The first successful blood transfusion was done decades ago, and yet we still haven't conducted the randomized, prospective clinical trials we need to do in order to find out which cardiac patients should get transfusions, and when they should get them."

Rao says there's no doubt some transfusions are necessary. In extreme cases, for example, where patients undergo massive blood loss or become severely anemic, transfusion can save lives. But he feels physicians often rush to prescribe the procedure when it may not be needed. "We believe the body can automatically respond to lower hematocrit levels by manufacturing more red blood cells. We need to allow time for that to happen."

Researchers aren't sure why transfusions might hurt some patients. Recent research by Duke's Jonathan Stamler found that banked blood quickly loses nitric oxide, a chemical important in the transfer of oxygen from red blood cells to the tissues that need it.

"It's not surprising that outcomes are not better for heart attack patients who get transfusions," says Stamler. "But they should be. The problem lies with the quality of banked blood. We need to correct that, and then do more studies."

"There is too much confusion and controversy over blood transfusions today," says Rao. "It is amazing to me that in 2007, we don't know how to appropriately prescribe transfusion. Blood is a national resource donated by the public. We need to be accountable to the public and to our patients as well."

The study was funded by the Duke Clinical Research Institute.

Colleagues who contributed to the study include senior author Robert Harrington, director of DCRI; Christopher Granger, Kristin Newby and Jie-Lena Sun, also of DCRI; Robert Califf, director of the Duke Translational Medicine Institute, Karen Chiswell, North Carolina State University; Frans Van de Werf, Universitaire Ziekenhuizen Leuven; Harvey White, Auckland City Hospital; and Paul Armstrong, University of Alberta.