Anemia is commonly observed in patients with cancer, and is caused by blood loss during surgery, chemotherapy, and radiotherapy. Although erythropoiesis-stimulating agents (ESAs) are often used for treating anemic cancer patients, the risk of thromboembolic events and decreased survival with ESA use has prompted the Food and Drug Administration to issue a restrictive advisory. As an alternative treatment, red blood cell (RBC) transfusions are often recommended, but no studies have demonstrated improved outcomes or increased quality of life in this patient population. Thus, a recent study sought to determine the relationship between transfusions and outcomes in cancer patients.

Using the discharge database of the University HealthSystem Consortium, the study examined discharge data for 504,208 patients with cancer who were admitted to United States’ medical centers between 1995 and 2003. Patients were identified based on disease classification codes, followed by statistical analyses to determine if transfusions of (A) only platelets, or (B) only RBC, or (C) both platelets and RBC, are associated with the following three outcomes: (1) venous thromboembolism (VTE), (2) arterial thromboembolism (ATE), and (3) in-hospital mortality.

Overall, 14.7% of the study population received either RBC or platelet transfusions. Of the total population, VTE was observed in 7.2% of patients receiving only RBC transfusions, 6.6% of patients receiving both RBC and platelet transfusions, and 6.4% of patients receiving only platelet transfusions, rates which were significantly higher than cancer patients not receiving transfusions. Likewise, significant rates were observed in cases of ATE. Also, patients receiving blood transfusions showed significantly higher rates of in-hospital mortality than patients not receiving transfusions of any kind.

The study’s results suggest caution in using transfusions as an alternative treatment to ESAs, and a number of possible mechanisms might explain the findings. For instance, the large amount of redox-iron delivered by transfusions has been linked to cardiovascular disease. Also, stored RBCs—resulting from transfusions increasing the circulating RBC mass—are depleted in nitric oxide, which may cause vasoconstriction and thrombosis. This study did not take into account patients who may have concomitantly been receiving ESA treatment, nor did it factor in the timing of transfusion in relation to thromboembolic events. Therefore, further studies are needed to elucidate these underlying mechanisms, as well to improve clinical decisions regarding transfusion therapy.

Khorana AA, Francis CW, Blumberg N, Culakova E, Refaai MA, Lyman GH. Blood transfusions, thrombosis, and mortality in hospitalized patients with cancer. Arch Intern Med. 2008 Nov 24;168(21):2377-81.

NAAC Expert Commentary:
Anemia is the most common cancer complication and one that is associated with an adverse prognosis, regardless of tumor type. Plasma erythropoietin is inappropriately low for the degree of anemia observed in cancer patients, owing to suppression of erythropoietin production by inflammatory cytokines, provoked by the neoplastic process. With the development of recombinant erythropoietin, it was widely assumed that the recombinant hormone would alleviate anemia, reduce the need for blood transfusions, and possibly improve survival as well as quality of life in cancer patients.

However, a number of recent clinical trials in patients with a variety of malignancies have indicated that recombinant erythropoietin may actually increase the incidence of thrombosis and reduce the survival of anemic cancer patients. Most of these trials involved the increase of hematocrit to normal and some involved anemic cancer patients not receiving concomitant chemotherapy. The most recent meta-analysis from a detailed evaluation of 13,933 anemic cancer patients in 53 clinical trials, indicated that recombinant erythropoietin increased the mortality rate and reduced survival only in patients not receiving concomitant chemotherapy.

Also, a black box warning and restrictions on the use of recombinant erythropoietin in cancer-associated anemia have been instituted by the FDA, whose position is that blood transfusion is safer than ever. But how safe is blood transfusion, leaving aside the issue of infectivity and acute lung injury?

The study by Khorana et al directly addresses this question. This group, which has previously identified venous thromboembolism as a frequent complication of cancer chemotherapy, now reports that red cell and platelet transfusions are associated with an increased risk of venous and arterial thrombosis and increased mortality in hospitalized cancer patients. Although this type of study cannot establish causality, and we do not know to what level transfusion increased the hematocrit, or the extent to which prophylactic anticoagulation was employed in the study population, it does establish a baseline from which to operate. Moreover, it reminds us that red cell transfusions are not a benign alternative to recombinant erythropoietin.

The reduction in plasma volume associated with red cell transfusions is also a potential mechanism for toxicity overlooked by the authors. Since recombinant erythropoietin can do the same thing, the use of both together needs to be monitored carefully. It may also be the case that when prophylactic anticoagulation is appropriately employed, the safety of recombinant erythropoietin will improve sufficiently to alleviate the need for red cell transfusions in anemic cancer patients.

Last Updated: February 4, 2009