Association between the length of storage of transfused leukoreduced red blood cell units and hospital-acquired infections in critically ill children: A secondary analysis of the TRIPICU study

Objective Evaluate the association between leukoreduced red blood cell (RBC) storage length and hospital-acquired infection (HAI) incidence rate in critically ill children. Background RBC transfusions are common in critically ill children. Despite their benefits, observational studies suggest an association between them and HAIs. One possible mechanism for increased HAI is transfusion-related immunomodulation due to bioactive substances' release as transfused blood ages. Methods In this secondary analysis of the 'Transfusion Requirement in Paediatric Intensive Care Units' (TRIPICU) study, we analysed a subset of 257 participants that received only one pre-storage leukoreduced RBC transfusion. RBC storage length was classified as 1) transfusion of 'fresh' RBCs (<= 10 days), 2) transfusion of 'stored' RBCs (21-34 days), and 3) transfusion of 'long-stored' RBCs (>= 35 days). All were compared to a 'golden' period (11-20 days), representing the time between 'fresh' and 'stored'. We used quasi-Poisson multivariable regression models to estimate the HAI incidence rate ratio (IRR) and corresponding 95% confidence interval (CI). Results We found that the association between the length of storage time of leukoreduced RBCs and HAIs was not significant in the 'fresh' group (IRR 1.23; 95% CI 0.55, 2.78) and the 'stored' group (IRR 1.61; 95% CI 0.63, 4.13) when compared to the 'golden' period. However, we observed a statistically significant association between the 'long-stored' group and an increase in the HAI incidence rate (IRR 3.66; 95% CI 1.22, 10.98). Conclusion Transfusion of leukoreduced RBC units stored for >= 35 days is associated with increased HAI incidence rate in haemodynamically stable, critically ill children.