Research Roundup
Question: Do blood transfusions in hospitalized cancer patients with anemia or thrombocytopenia affect thrombotic event and in-hospital mortality rates?
Background: Erythropoiesis-stimulating agents have recently been shown to increase thrombotic risk and decrease survival in cancer patients. Blood transfusions are a common alternative for anemic patients. However, there are no randomized trials demonstrating improved outcomes in cancer patients receiving transfusions. Furthermore, the safety of transfusions has not been clearly defined.
Study design: Retrospective, cohort study.
Setting: 60 U.S. medical centers.
Synopsis: Using discharge data from the University Health System Consortium, 504,208 hospitalizations of cancer patients revealed that 14% of patients received at least one RBC transfusion and 3% of patients received a platelet transfusion. RBC and platelet transfusions were associated with increased risk of arterial thrombosis (RBCs: OR 1.53; 95% CI 1.46-1.61; platelets: OR 1.55; 1.4-1.71, P<0.001) and venous thrombotic events (RBCs: OR 1.34; 1.29-1.38; platelets: OR 1.2; 1.11-1.29, P<0.001). Additionally, transfusions were associated with increased in-hospital mortality (RBCs: OR 1.34; 1.29-1.38; platelets: OR 2.4; 2.27-2.52, P<0.001). Study results are limited by several factors, including the observational nature and the use of administrative coding data. Information on venous thromboembolism prophylaxis was not available, and the timing of transfusions in relation to the diagnosis of thrombotic events is unknown. Finally, anemia-necessitating transfusions may be a surrogate for “sicker” patients, explaining the increased in-hospital mortality. Blood transfusions in hospitalized patients require further study to determine whether there is a causal relationship between transfusions and increased thrombotic events and mortality.
Bottom line: Blood transfusions in hospitalized cancer patients should be used cautiously, as they are associated with increased thrombotic events and in-hospital mortality.
Citation: Arch Int Med. 2008;168:2377-2381
—Reviewed for the eWire by Kerry Will, MD, Jayne Barr, MD, Kim Tartaglia, MD, Aaron Wenger, MD, Jonathan Wynbrandt, MD, Nathan J. O’Dorisio, MD, The Ohio State University Medical Center, Columbus, OH.
