Effects of red blood cell transfusion on patients undergoing cardiac surgery in Queensland - a retrospective cohort study

Nchafatso G Obonyo,Vikash Dhanapathy,Nicole White,Declan P Sela,Reema H Rachakonda,Matthew Tunbridge,Beatrice Sim,Derek Teo,Zohaib Nadeem,Louise E See Hoe,Gianluigi Li Bassi,Jonathon P Fanning,John-Paul Tung,Jacky Y Suen,John F Fraser
DOI: https://doi.org/10.1186/s13019-024-02974-7
2024-08-01
Abstract:Background: Packed red blood cell (pRBC) transfusion is a relatively safe and mainstay treatment commonly used in cardiac surgical patients. However, there is limited evidence on clinical effects of transfusing blood nearing end-of shelf life that has undergone biochemical changes during storage. Objective: To investigate evidence of associations between morbidity/mortality and transfusion of blood near end of shelf-life (> 35 days) in cardiac surgical patients. Methods: Data from the Queensland Health Admitted Patient Data Collection database 2007-2013 was retrospectively analysed. Coronary artery bypass graft and valvular repair patients were included. Multivariable logistic regression was used to examine the effect of pRBC age (< 35 days vs. ≥ 35 days) on in-hospital mortality and morbidity. As secondary analysis, outcomes associated with the number of pRBC units transfused (≤ 4 units vs. ≥ 5 units) were also assessed. Results: A total of 4514 cardiac surgery patients received pRBC transfusion. Of these, 292 (6.5%) received pRBCs ≥ 35 days. No difference in in-hospital mortality or frequency of complications was observed. Transfusion of ≥ 5 units of pRBCs compared to the ≤ 4 units was associated with higher rates of in-hospital mortality (5.6% vs. 1.3%), acute renal failure (17.6% vs. 8%), infection (10% vs. 3.4%), and acute myocardial infarction (9.2% vs. 4.3%). Infection carried an odds ratio of 1.37 between groups (CI = 0.9-2.09; p = 0.14) and stroke/neurological complications, 1.59 (CI = 0.96-2.63; p = 0.07). Conclusion: In cardiac surgery patients, transfusion of pRBCs closer to end of shelf-life was not shown to be associated with significantly increased mortality or morbidity. Dose-dependent differences in adverse outcomes (particularly where units transfused were > 4) were supported.
What problem does this paper attempt to address?