Introduction: Patient Blood Management (PBM) is an evidence-based, multidisciplinary approach designed to avoid unnecessary transfusions, optimize hemostasis, and minimize blood loss to improve patient outcomes. Our Blood and Marrow Transplant (BMT) program, in collaboration with our PBM program, changed the Hb guidelines for RBC transfusions to 7g/dL in the fall of 2015 for all patients without acute bleeding or cardiac history on our inpatient transplant floor. This study aimed to assess the compliance with the guideline and the impact of the change on allogeneic patient outcomes (readmission within 30 days and mortality).

Methods: Data was extracted from our Electronic Medical Record and our transplant outcomes REDCAP database for an 18 month period before and after we changed the guidelines, 1/2014-6/2015 (pre) and 1/2016-6/2017 (post). Only RBC transfusions on the stem cell transplant unit were included. The main outcomes analyzed were a) all-cause mortality (at 100 days, 180 days and 365 days post-transplant) and b) hospital readmission (at days 1-7 and days 8-30 post discharge). Statistical analysis included Fisher's exact test and Wilcoxon rank-sum for non-parametric data. Univariate followed by multivariable stepwise logistic regression was performed for generating adjusted odds ratios (adjOR) for various outcomes. Testing for interaction and multi-collinearity was performed at each step of the model building. A two-sided p-value of <0.05 was the threshold for significance. Statistics were performed using Stata 14 and SAS 9.4.2.

Results: During the time periods studied, 293 patients received an allogeneic stem cell transplant (pre=153, post=140): females n=134, 46%; median (IQR) age 57 years (46-64); Karnofsky performance status 90 (80-90); transplant hospital length of stay 18 days (15-25). The most common reasons for transplant were AML (n=144, 49%), MDS/Myeloproliferative Disease (n=44, 15%), NHL (n=39, 13%), and ALL (n=30, 10%). The most common types of transplant were haplo/cord (n=113, 39%), MUD (n=94, 32%) and MRD (n=80, 27%). In the pre and post periods, 883/1147 (77%) and 186/775 (24%) RBC transfusions were at a Hb>=7g/dL, respectively, (p<0.0001) (Figure 1A). In the pre-period 141 (92%) and the post-period 117 (84%) patients were transfused with RBCs (p=0.03) (Figure 1B). The all-cause mortality at 100, 180 and 365 days w n=31 (10.6%), n=58 (19.8%) and n=102 (34.8%) respectively. Among the ever-transfused subjects, all-cause re-admission rates at 1-7 days post discharge were n=69 (26.2%) and at day 8-30 were n=77 (29.3%). After adjusting for age, gender, ASBMT risk classification, Karnofsky score, relapse status, underlying cardiac disease and acute GVHD; there was no difference in adjOR (+/- 95% confidence intervals) for all-cause post-transplant mortality and re-admission at specified time points between the two time periods compared (Figure 1C).

Conclusion: A RBC transfusion Hb threshold of 7 g/dL can be effectively implemented on an inpatient stem cell transplant unit, avoids RBC exposure in a significant percent of patients, and does not lead to increased all-cause post-transplant mortality or readmissions within one month.


Rossi: Celgene: Consultancy; Thrassos: Consultancy. Lee: Clinipace: Consultancy; Baxalta: Consultancy; Amgen: Consultancy; Alexion Pharmaceuticals: Consultancy.

Author notes


Asterisk with author names denotes non-ASH members.