Isbister JP, Shander A, Spahn DR, Erhard J, Farmer SL, Hofmann A. Adverse blood transfusion outcomes: establishing causation.
Transfus Med Rev 2011;
25:89-101. [PMID:
21345639 DOI:
10.1016/j.tmrv.2010.11.001]
[Citation(s) in RCA: 127] [Impact Index Per Article: 9.8] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 01/01/2023]
Abstract
The transfusion of allogeneic red blood cells (RBCs) and other blood components is ingrained in modern medical practice. The rationale for administering transfusions is based on key assumptions that efficacy is established and risks are acceptable and minimized. Despite the cliché that, "the blood supply is safer than ever," data about risks and lack of efficacy of RBC transfusions in several clinical settings have steadily accumulated. Frequentist statisticians and clinicians demand evidence from randomized clinical trials (RCTs); however, causation for the recognized serious hazards of allogeneic transfusion has never been established in this manner. On the other hand, the preponderance of evidence implicating RBC transfusions in adverse clinical outcomes related to immunomodulation and the storage lesion comes from observational studies, and a broad and critical analysis to evaluate causation is overdue. It is suggested in several circumstances that this cannot wait for the design, execution, and conduct of rigorous RCTs. We begin by examining the nature and definition of causation with relevant examples from transfusion medicine. Deductive deterministic methods may be applied to most of the well-accepted and understood serious hazards of transfusion, with modified Koch's postulates being fulfilled in most circumstances. On the other hand, when several possible interacting risk factors exist and RBC transfusions are associated with adverse clinical outcomes, establishing causation requires inferential probabilistic methodology. In the latter circumstances, the case for RBC transfusions being causal for adverse clinical outcomes can be strengthened by applying modified Bradford Hill criteria to the plethora of existing observational studies. This being the case, a greater precautionary approach to RBC transfusion is necessary and equipoise that justifying RCTs may become problematic.
Collapse