Platelet Transfusions and Adverse Outcomes

In the August issue of Transfusion, Dr. Spiess and co-authors reported a study investigating the hypothesis that platelet (PLT) transfusion is associated with adverse outcomes in patients undergoing coronary artery bypass graft (CABG) surgery (Spiess BD et al., Transfusion 2004;44:1143-8). The study was a re-analysis of data collected during six randomized, double-blind, placebo-controlled trials performed between January 1990 and May 1995 at 27 medical centers, aiming to evaluate aprotinin in CABG surgery. The authors retrospectively investigated adverse outcomes data on the 1,720 patients enrolled in these trials and compared the incidence of perioperative adverse events between patients that received and those who did not receive platelet transfusions.

Adverse events were found to be significantly more frequent in patients administered one or more PLT transfusions. “Logistic regression analysis,” the authors wrote, “showed that PLT transfusion was associated with infection, vasopressor use, respiratory medication use, stroke, and death.” The authors acknowledged that their study was not the randomized controlled trial required to prove the existence of a cause-and-effect relationship between PLT transfusion and adverse outcomes, and that PLT transfusions may simply be a surrogate marker for sicker patients. In addition, they recognized that aprotinin administration may have been a confounding variable, for instance through an independent effect on stroke. They also noted that the randomized trials that formed the basis of their analysis had been performed before the widespread use of leukoreduction. Nevertheless, they concluded that their findings supported “the conservative and targeted use of PLT transfusions.”

In the accompanying editorial (Dzik W, Transfusion 2004;44:1132-4), Dr. Dzik, Associate Editor of Transfusion, pointed out that “retrospective association studies, although provocative and stimulating, are not capable of getting to the truth regarding adverse effects of transfusion and should not be used as the basis of health policy affecting patients. The matter can be resolved through the use of randomized controlled clinical trials (RCTs).” “Transfusion medicine”, he added, “has had too few RCTs and we will continue to suffer as a medical discipline until we remedy this deficiency.” In order to remedy this deficiency, the National Heart Lung and Blood Institute recently created the Clinical Trials Network in Transfusion Medicine and Hemostasis (www.tmhnetwork.org).

With regard to the study by Spiess et al., in a comment published in the November 13 issue of The Lancet (Vamvakas EC, Lancet 2004;364:1736-8), Dr. Vamvakas from the Canadian Blood Services notes that the possible adverse effect of PLT transfusion might be attributed to “an ill-defined effect of allogeneic white cells in the platelet components,” to supernatant allogeneic plasma or to the PLTs themselves. A possible explanation for the association between PLT transfusions and stroke observed by Dr. Spiess and colleagues has been reported recently: Dr. Akbiyik and colleagues have demonstrated that human PLTs express peroxisome proliferator-activated receptor gamma (PPARgamma), a ligand-activated transcription factor that plays an important role in lipid metabolism, diabetes, and inflammation. (Akbiyik F et al., Blood 2004;104:1361-8).

Randomized trials, Dr. Vamvakas concludes, must be done to test the hypothesis that PLT transfusions are associated with adverse outcomes and, should the association be confirmed, to identify the specific culprit within the PLT product. “Clinical practice,” the author states, “should not be modified” in response to observational studies. However, “conservative use of platelet transfusions in coronary-artery-bypass grafting is warranted because of the various infectious and non-infectious complications of allogeneic blood transfusion.”