Review Question
A restrictive transfusion policy involves giving a red blood cell transfusion if the oxygen-carrying capacity of blood (haemoglobin) falls below a certain level. A liberal transfusion policy involves giving a red blood cell transfusion at a higher haemoglobin level.
This review aims to assess whether a restrictive or liberal transfusion policy is superior in terms of death (due to any cause), death due to bleeding, infection, transfusion reactions or iron overload, quality of life, frequency and length of hospital admissions, serious bleeding or infections, and number of red blood cell transfusions required.
Background
The bone marrow is where many types of blood cells are produced. Red blood cells are necessary to bring oxygen to all parts of the body, white blood cells fight against infection, and platelets in the blood help to form clots and prevent bleeding. Bone marrow failure can have different causes and can happen at birth or later in life, and may result in too few of any or all of the three types of blood cells in the body. Too few red blood cells causes a low haemoglobin level, called anaemia, which may result in poor delivery of oxygen to the body. This can cause shortness of breath, tiredness and has a significant impact on quality of life.
Regular red blood cell transfusions are used to support patients and improve their quality of life when treatments that might cure the condition are not an option. However, there are certain risks involved with regular use of red blood cell transfusions, for example, transfusion reactions, transfusion-transmitted infections and iron overload.
Currently it is not clear if the best transfusion policy is a restrictive or liberal one.
Study Characteristics
Only one small study including 13 patients with bone marrow failure was included in this review. The study was funded by two government agencies and one charity. We are aware of three ongoing studies which have not yet been completed.
The evidence is current to 26th May 2015.
Key Results
The one included study was too small to demonstrate any difference in all-cause mortality (death due to any cause) or number of red cell transfusions received between a restrictive compared to a liberal red blood cell transfusion policy. At the current time, there is a lack of evidence to recommend a restrictive transfusion strategy over a liberal one. Trials with good methodology are needed to determine the best transfusion policy for patients with long -term bone marrow failure disorders.
Quality of the Evidence
The evidence for the findings was of very low quality. This was because very small numbers of participants were included in the study. Only 13 patients were recruited to the trial rather than the planned 200 participants due to problems with recruitment.
This review indicates that there is currently a lack of evidence for the recommendation of a particular transfusion strategy for bone marrow failure patients undergoing supportive treatment only. The one RCT included in this review was only published as an abstract and contained only 13 participants. Further randomised trials with robust methodology are required to develop the optimal transfusion strategy for such patients, particularly as the incidence of the main group of bone marrow failure disorders, MDS, rises with an ageing population.
Bone marrow failure disorders include a heterogenous group of disorders, of which myelodysplastic syndrome (MDS), forms the largest subgroup. MDS is predominantly a disease of the elderly, with many elderly people managed conservatively with regular allogeneic red blood cell (RBC) transfusions to treat their anaemia. However, RBC transfusions are not without risk. Despite regular transfusions playing a central role in treating such patients, the optimal RBC transfusion strategy (restrictive versus liberal) is currently unclear.
To assess the efficacy and safety of a restrictive versus liberal red blood cell transfusion strategy for patients with myelodysplasia, acquired aplastic anaemia, and other inherited bone marrow failure disorders.
We searched for randomised controlled trials (RCTs) in the Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library 2015, Issue 4), Ovid MEDLINE (from 1946), Ovid EMBASE (from 1974), EBSCO CINAHL (from 1937), the Transfusion Evidence Library (from 1980) and ongoing trial databases to 26th May 2015.
RCTs including patients with long-term bone marrow failure disorders that require allogeneic blood transfusion, who are not being actively treated with a haematopoietic stem cell transplant, or intensive chemotherapy.
We used standard Cochrane review methodology. One author initially screened all references, and excluded any that were clearly irrelevant or duplicates. Two authors then independently screened all abstracts of articles, identified by the review search strategy, for relevancy. Two authors independently assessed the full text of all potentially relevant articles for eligibility, completed the data extraction and assessed the studies for risk of bias using The Cochrane Collaboration’s ’Risk of bias’ tool.
We included one trial (13 participants) and identified three ongoing trials that assess RBC transfusion strategies in people with MDS.
The quality of the evidence was very low across different outcomes according to GRADE methodology.
The one included study randomised participants to a restrictive [haemoglobin (Hb) transfusion trigger < 72 g/L, 8 participants] or liberal [Hb trigger < 96 g/L, 5 participants] transfusion policy. There was insufficient evidence to determine a difference in all-cause mortality (1 RCT; 13 participants; RR 0.13, 95% CI 0.01 to 2.32; very low quality evidence). There was insufficient evidence to determine a difference in the number of red blood cell transfusions (1 RCT; 13 participants; 1.8 units per patient per month in the liberal group, compared to 0.8 in the restrictive arm, no standard deviation was reported; very low quality evidence). There were no anaemia-related complications reported (cardiac failure) and no reported effect on activity levels (no statistics provided). The study did not report: mortality due to bleeding/infection/transfusion reactions or iron overload, quality of life, frequency and length of hospital admissions, serious infections (requiring admission to hospital), or serious bleeding (e.g. WHO/CTCAE grade 3 (or equivalent) or above).