Background: Transfusion-related acute lung injury (TRALI) is a leading cause of transfusion-related mortality. A concern with passive surveillance to detect transfusion reactions is underreporting. Our aim was to obtain evidence-based estimates of TRALI incidence using meta-analysis of active surveillance studies and to compare these estimates with passive surveillance.
Study design and methods: We performed a systematic review and meta-analysis of studies reporting TRALI rates. A search of Medline and Embase by a research librarian identified studies published between January 1, 1991 and January 20, 2023. Prospective and retrospective observational studies reporting TRALI by blood component (red blood cells [RBCs], platelets, or plasma) were identified and all inpatient and outpatient settings were eligible. Adult and pediatric, as well as general and specific clinical populations, were included. Platelets and plasma must have used at least one modern TRALI donor risk mitigation strategy. A random effects model estimated TRALI incidence by blood component for active and passive surveillance studies and heterogeneity was examined using meta-regression.
Results: Eighty studies were included with approximately 176-million blood components transfused. RBCs had the highest number of studies (n = 66) included, followed by platelets (n = 35) and plasma (n = 34). Pooled TRALI estimates for active surveillance studies were 0.17/10,000 (95% confidence intervals [CI]: 0.03-0.43; I2 = 79%) for RBCs, 0.31/10,000 (95% CI: 0.22-0.42; I2 = <1%) for platelets, and 3.19/10,000 (95% CI: 0.09-10.66; I2 = 86%) for plasma. Studies using passive surveillance ranged from 0.02 to 0.10/10,000 among the various blood components.
Discussion: In summary, these estimates may improve a quantitative understanding of TRALI risk, which is important for clinical decision-making weighing the risks and benefits of transfusion.
Keywords: non-infectious; transfusion complications.
© 2023 AABB.