Comparing the ratio of mean red blood cell transfusion episode rate of 1 unit versus 2 units in hematopoietic stem cell transplant patients

Abstract

BACKGROUND: Two units of red blood cells (RBCs) were typically transfused with each transfusion among hematopoietic stem cell transplant (HSCT) patients. Concerns regarding this practice are increased morbidity, mortality, hospital-acquired infections, length of stay (LOS), and transfusion-related complications. This study compared outcomes of transfusing 1 unit of RBCs per transfusion episode to 2 units of RBCs per episode among HSCT patients. STUDY DESIGN AND METHODS: A retrospective record review was used to evaluate a practice change of transfusing 1 RBC unit per episode among autologous and allogeneic HSCT patients. Primary endpoints included: 1) mean number of RBC transfusion episodes during the hospital stay, 2) mean number of RBC units transfused adjusted by LOS, and 3) mean LOS. RESULTS: Among autologous patients, the ratio of mean rate of transfusion episodes for transfusing 1 unit versus 2 units per transfusion was 1.24, with a one-tailed 95% upper limit of 1.42. With a noninferiority upper bound of 1.50, using 1 unit per transfusion episode was noninferior to 2 units per transfusion episode (p = 0.011). Among allogeneic HSCT patients, the ratio of mean transfusion episode rate was 1.26 with a one-tailed 95% upper limit of 1.52, which was slightly above the 1.50 noninferiority bound (p = 0.061). CONCLUSION: A single-unit transfusion policy was not inferior to the 2-unit policy for autologous HSCT patients and trended toward noninferiority for allogeneic transplant patients. The mean volume of blood per LOS was lower for the 1-unit practice for both groups. The gains from the practice change may outweigh the risks of not changing.