Thesis Floris Kranenburg

Summary

Anemia, also referred to as low hemoglobin level, is a common finding in critically ill patients. Red blood cell (RBC) transfusion is used to increase patients’ hemoglobin level. Approximately 26 - 40% of critically ill patients receive RBC transfusions during their stay in the Intensive Care Unit (ICU), making transfusion a substantial element of critical care practice. Current available evidence supports utilising a restrictive transfusion trigger of 7 g/dL in non-bleeding ICU patients. A burning question is whether 7 g/dL is the optimal trigger for all non-bleeding ICU patients. This research focused on increasing our understanding of clinicians’ red cell transfusion decisions, of possible consequences of red cell transfusion in individual non-bleeding critically ill patients, and on the methodology to achieve the latter goal.We showed that there is considerable variation in transfusion decisions amongst critical care physicians, suggesting that clinicians have varying ideas about benefits of transfusion. We identified clinical determinants of transfusion decisions, because these determinants might be associated with observed effects of RBC transfusion. The data were derived from the electronic medical health records of ICU patients. Our results confirmed that besides hemoglobin concentration, inherently the most important predictor of RBC transfusion, a number of other clinical parameters were associated with transfusion decisions. Using the same data, we developed a model for the prediction of the effect of RBC transfusion on subsequent organ functioning in non-bleeding critically ill patients.