The use of blood components prior to bedside procedures

Abstract

A transition from liberal use of transfusions prior to invasive procedures to a thoughtful, restrictive approach to transfusion is underway. This shift is being driven by the publication of very large observational studies showing a very low incidence of bleeding complication from most common procedures (even in the presence of severe thrombocytopenia and abnormal tests of coagulation) in conjunction with an evidence-based 2019 guideline from the Society for Interventional Radiology recommending restrictive use of pre-procedure transfusion. Many common invasive procedures have a major bleeding risk well less than 1% with image-guided techniques. This is excellent for patient care, however prospective randomized trials of transfusion vs. no transfusion before invasive procedures are unattainable, given the studies would require an impracticable sample size due to low event rates and would expose the transfusion group to the harms of transfusion. Indeed, a recent pilot randomized trial not only found challenges with recruitment but high rates of transfusion complications suggesting that transfusion risks currently exceed bleeding risks. Utilization studies find approximately 25% of plasma and 10% of platelets are transfused to patients as prophylaxis for bleeding prevention prior to procedures. This suggests that adherence to restrictive practices could substantially reduce adverse reactions from transfusion, minimize blood product shortages, and minimize delays in procedures for transfusion. In addition to unnecessary transfusions, the unselected use of preprocedure laboratory testing is unwarranted for all procedures. This testing is expensive, has a low positive predictive value for bleeding complications, and delays procedures unnecessarily. Numerous studies have also shown that the infusion of plasma for mildly elevated international normalized ratio (INR) test results (INR of 1.5–1.9) does not alter the INR and therefore is very unlikely to reduce the bleeding risk. Lastly, the INR does not predict the risk of bleeding and the coagulation status of patients with liver cirrhosis. Many large centers have successfully transitioned to a restrictive use of blood before procedures and published the safety of this approach. This review will provide the evidence to convince others to follow suit.