University of Iowa Hospitals and Clinics, Iowa, United States
Background/Case Studies: To reduce septic transfusion reactions due to bacterial contamination of platelets, several mitigation strategies have been implemented that impact the shelf-life of the unit. Large volume delayed sampling ≥ 48 hours (LVDS48) for platelet bacterial testing allows for 7-day storage and has been widely adopted. A previous study demonstrated an increase in transfusion reaction rates from day 3 to day 5, however, it is unknown whether this trend will continue with day 6-7 platelets. If older platelets are associated with increased reactions, this could lead to an increase in product wastage, testing, and labor costs for sites using LVDS48 platelets.
Study
Design/Methods: Transfusion reaction rates to platelets aged 4-7 days were determined over a 3-year period (2018-2021) at two independent academic institutions. After exclusion of platelets given to patients < 1yo, our final dataset contained 19,276 platelet transfusions (5,813 + 13,463) in 4,376 unique patients (1,700 + 2,676). Differences in practice between sites include universal product irradiation at one site and non-mandatory reporting of mild allergic reactions at the other site. Multivariable Bayesian hierarchical regression models accounting for sex, patient age, blood type, compatibility, date, site, patient and hospital were used to determine the odds ratio (OR) and statistical significance for differences in reaction rates by storage age. A p-value less than 0.05 denotes statistical significance.
Results/Findings: Overall transfusion reaction rates differed between institutions (2.3% and 0.7%), likely due to differences in reporting practices. However, there was no statistical difference in transfusion reaction rates when comparing day 4-5 platelets to day 6-7 platelets (1.1% vs. 1.4% respectively; OR=1.11, p=0.51). When the data was subdivided by transfusion reaction type, this finding remained the same for febrile nonhemolytic transfusion reactions at both institutions separately and in the combined data set (OR=1.29, p=0.30), and when stratified by ABO compatibility (Identical: OR=1.28, p=0.46; Not Identical: OR=1.27, p=0.48). Allergic transfusion reactions demonstrated an increased rate of reaction in non-ABO identical transfusions (OR=3.32, p=0.0005) only at the institution that required reporting of all allergic reactions. Conclusions: This collaborative multicenter retrospective review of platelet transfusion reaction rates did not overall demonstrate a significant increase in transfusion reaction rates with increased storage age. This study provides reassurance that the implementation of LVDS48 for platelet bacterial testing is unlikely to significantly increase the burden of transfusion reactions due to the increased use of older platelet units.
Importance of research: As the use of 6–7-day old platelets increase, it is important to evaluate the risk of transfusion reactions (as a function of platelet age) to maximize the safety and utility of blood products.
