Dose-response relationship between transfusion and the incidence of infection in critically ill patients: a systematic review and dose-response meta-analysis.

Purpose: To estimate the association between red blood cell (RBC) transfusion volume and hospital-acquired infections (HAI) in critically ill patients, with a particular focus on identifying the potential threshold volumes at which infection risk changes.

Methods: The MEDLINE, CENTRAL, Embase, and Transfusion Evidence Library databases were searched for studies published from database inception to November 2024. Citation searches and reference checks of the relevant guidelines were combined. Studies that evaluated transfusion and anemia pharmacotherapy in critically ill patients were included. Outcome of interest was the incidence of HAI. We conducted a dose-response meta-analysis (DRMA) using a one-stage random-effects model.

Results: We identified 39,453 records after searching the databases. After combining the results of citation searches and reference checks of the guidelines, 45 studies were eligible. For the DRMA, we eliminated 14 studies without results and 15 with a critical risk of bias. We included 9587 patients from 16 studies. Our DRMA showed a non-linear risk curve, with odds ratio for HAI decreasing and reaching a trough at three units of RBC transfusion. Three units of RBC may not increase the risk of HAI in critically ill patients. However, the clinical implications of higher RBC transfusion volumes remain unclear.

Conclusions: Our findings suggest a non-linear relationship between RBC transfusion volume and HAI risk in critically ill patients, highlighting the need for further research to inform individualized transfusion strategies. Clinical Trial Registration http://osf.io/a9cwd.