Implementation of a transfusion bundle reduces inappropriate red blood cell transfusions in intensive care - a before and after study

Restrictive red blood cell (RBC) transfusion has been widely described in transfusion guidelines. However, compliance with these guidelines is often poor. Therefore, we developed a care bundle for the transfusion of RBCs in intensive care. We investigated the effect of the application of the transfusion bundle on transfusion practice, hypothesising that the implementation of the transfusion bundle would lead to a reduction of inappropriate RBC transfusions. We conducted a before and after study between January and December 2014 in a medical-surgical intensive care unit (ICU) of a university hospital in Amsterdam, the Netherlands. The primary outcome was the percentage of appropriate transfusions, referring to those transfusions that were in accordance to the patients' individual preset haemoglobin threshold. The mean pre-transfusion haemoglobin level was 7·3 g dL-1 [standard deviation (SD) = 1·15] during baseline and significantly decreased to 7·1 g dL-1 (SD = 1·04) after transfusion bundle implementation; 95% confidence interval (CI): 0·009-0·308, P-value = 0·04. The number of inappropriate transfusions significantly decreased from 25% (111/439) during baseline to 15% (42/280) during implementation, a difference of 10%; 95% CI: -0·164 to -0·0416, P-value 0·001. This further decreased to 12% (45/370) in the post-implementation phase. A logistic regression analysis showed that the chance to find an appropriate transfusion is approximately twice as high after transfusion bundle implementation. Introduction of a transfusion bundle results in a significant reduction of the number of inappropriate RBC transfusions in the medical-surgical ICU. Our results show that the introduction of a transfusion care bundle helps to improve compliance with transfusion guidelines in daily practice