Red cell transfusion thresholds in haematopoietic stem cell transplantation

Abstract

Red cell transfusions are part of the routine supportive care following haematopoietic stem cell transplantation. There is increasing evidence from randomized controlled trials demonstrating that restrictive transfusion thresholds are equivalent to liberal thresholds in other patient populations. However, given the lack of patient‐centred outcomes and the longer transfusion support required for patients with haematologic malignancies, the results of these trials may not be generalizable to haematology–oncology patients. In the recently completed multicentre TRIST trial, we randomized 300 patients requiring an autologous or allogeneic haematopoietic stem cell transplantation to either a restrictive (haemoglobin threshold <70 g/l) or liberal (haemoglobin threshold <90 g/l) red cell transfusion strategy. Our primary outcome was health‐related quality of life measured by FACT‐BMT through day 100 post‐transplant. The median number of red cell units transfused was lower in the restrictive strategy group (2 vs. 4). The FACT‐BMT total score was higher at all time points for patients in the restrictive transfusion group including at baseline (109 vs. 103) and at 100 days (116·3 vs. 109·2). After adjustment for baseline, the FACT‐BMT score at day 100 in the restrictive group was statistically non‐inferior to the liberal group (P < 0·0001). There were no significant differences between the two groups for any other clinical outcomes. Based on the results of the TRIST trial, a restrictive haemoglobin threshold of 70 g/l should become standard practice in haematopoietic stem cell transplant patients, and, given the lack of other clinical studies, all other haematology–oncology inpatients.