Open AccessA retrospective analysis of venous thromboembolism trends in chemotherapy‐induced anemia: Red blood cell transfusion versus erythrocyte stimulating agent administration
Author(s) -
Bryer Emily J.,
Kallan Michael J.,
Chiu TingShan,
Scheuba Katharina M.,
Henry David H.
Publication year - 2020
Publication title -
ejhaem
Language(s) - English
Resource type - Journals
ISSN - 2688-6146
DOI - 10.1002/jha2.18
Subject(s) - medicine , hazard ratio , chemotherapy , anemia , retrospective cohort study , proportional hazards model , blood transfusion , red blood cell , malignancy , surgery , confidence interval
Abstract Background Patients receiving a variety of chemotherapy regimens often develop chemotherapy‐induced anemia (CIA), which contributes to poor outcomes including increased mortality. Prompt and effective treatment of CIA is essential to prevent fewer chemotherapy dose delays and reductions. Optimal therapy of CIA is controversial and involves the solitary and combined use of intravenous iron, red blood cell (RBC) transfusions, and erythropoietin stimulating agents (ESAs). Despite the baseline coagulopathies present in patients with malignancy, administration of both RBC transfusions and ESAs is associated with venous thromboembolism (VTE). It remains unknown whether the risk of VTE in patients with CIA is greater among patients who receive RBC transfusions or ESAs. Methods A retrospective study analyzed 10,269 University of Pennsylvania Health System patients with malignancies of various type, stage, and histopathology who developed CIA between 2008 and 2017. Using multivariate Cox regression, we determined adjusted hazard ratios (and corresponding 95% confidence intervals) of VTE development after adjusting for RBC and ESA intervention (all during the 90 days following CIA diagnosis). Results Among the 10,269 patients with CIA, 2,642 (25.7%) developed a VTE within the 90‐day period. VTE risk following RBC transfusion (HR = 1.37, 95% CI 1.24‐1.50, P < .001) was more than twice as common as VTE risk following ESA administration (HR = 0.53, 95% CI 0.40‐0.69, P < .001). Conclusion While both RBC transfusion and ESA are independently associated with VTE, our data suggest a greater risk of VTE development with RBC transfusion as compared with ESA.