A large variation in transfusion policy between countries and even between hospitals within a country can still be seen as common practice:

  • In the first Austrian benchmark study, Gombotz and colleagues found a transfusion rate varying from 16 to 85% for patients undergoing primary total hip replacement (THR) and a 12-87% transfusion rate for patients undergoing primary total knee replacement (TKR) surgery (Gombotz et al. Transfusion 2007). 
  • In the second benchmark study, which was published recently, transfusion rates have decreased in THR (from 41 to 30%) and TKR (41 to 25%) however with still substantial inter-centre variability (Gombotz et al. Transfusion 2014) These studies show, that the implementation of guidelines in daily practice is often difficult. This is not only relevant for implementing transfusion thresholds, but also for the number of transfusions at each event to reach a particular target Haemoglobin (Hb) level. 

Barr and co-workers investigated red blood cell transfusion practice in Northern Ireland in 2005 and still found a two-unit instead of single-unit transfusion practice in medical and surgery patients (n=1474) (Barr et al. Vox Sang 2010).


Scientific evidence to support restrictive transfusion thresholds

Key messages:

  • Use of a restrictive blood transfusion protocol can be regarded as an important blood saving and cost-saving strategy.
  • Restrictive transfusion does not lead to an increased rate in mortality, cardiac events, myocardial infarction, stroke, pneumonia, infections and thromboembolism when compared to liberal transfusion strategies in the majority of patients.
  • Effects of restrictive transfusion triggers in high risk groups such as acute coronary syndrome and in specific subgroups (see below) need to be tested further in large clinical trials.



In 1988 the NIH published consensus guidelines for red blood cell transfusions. Since then, several guidelines have been published, recommending that a range of Hb levels between 6 and 10 g/dL can be used, depending on the presence of serious co-morbidity:

Randomized studies comparing liberal to restrictive transfusion triggers

Nowadays, a large number of high-quality trials (more than 30 randomized controlled trials) on RBC transfusion thresholds in hemodynamically stable hospitalized adult patients have been published, which included over 19,000 patients in total, comparing restrictive transfusion thresholds (transfusing RBCs only when haemoglobin level reaches 7-8 g/dL) with liberal transfusion thresholds (transfusing RBCs only when haemoglobin level reaches 9-10 g/dL). 

A recent Cochrane review concluded that “transfusing at a restrictive haemoglobin concentration of between 7 g/dL to 8 g/dL decreased the proportion of participants exposed to RBC transfusion by 43% across a broad range of clinical specialties (RR 0.97, 95% CI 0.81 to 1.16). There was no evidence that a restrictive transfusion strategy impacts 30-day mortality or morbidity (i.e. mortality at other points, cardiac events, myocardial infarction, stroke, pneumonia, thromboembolism, infection) compared with a liberal transfusion strategy. There were insufficient data to inform the safety of transfusion policies in certain clinical subgroups, including acute coronary syndrome, myocardial infarction, neurological injury/traumatic brain injury, acute neurological disorders, stroke, thrombocytopenia, cancer, haematological malignancies, and bone marrow failure. The findings provide good evidence that transfusions with allogeneic RBCs can be avoided in most patients with haemoglobin thresholds above 7 g/dL to 8 g/dL.” (5) 

Another recent systematic review focused on patients with cardiovascular disease and found that  cardiac surgery patients had a comparable 30 day-mortality between groups (risk ratio 0.99; 95% confidence interval 0.74-1.33). In 2 small trials (n=154) in patients with myocardial infarction, the point estimate for the mortality risk ratio was 3.88 (95% CI, 0.83-18.13) favoring the liberal strategy. (6) 

Another systematic review focussed on the effect of transfusion thresholds on postoperative infection rate and concluded that using restrictive transfusion thresholds lowered infection risk (Holst et al. New Eng J Med 2014).

A study of interest was published by Villanueva and coworkers. These authors concluded that a restrictive transfusion strategy significantly improved outcome in patients with an acute upper gastrointestinal bleed (7).



1. Based on recent literature regarding transfusion thresholds and clinical transfusion guidelines, the following recommendations can be made [5-7]:

  • Restrictive transfusion threshold of haemoglobin level of 7 g/dL should be used for hemodynamically stable hospitalized adult patients, including critically ill patients. 
  • Restrictive transfusion threshold of haemoglobin level of 7.5 g/dL should be used for those undergoing cardiac surgery.
  • Restrictive transfusion threshold of haemoglobin level of 8 g/dL should be used for those undergoing orthopaedic surgery, and those with underlying stable cardiovascular disease. 
  • Patients admitted with an acute upper gastrointestinal bleed may have a mortality benefit from using a restrictive transfusion threshold of haemoglobin level of 7 g/dL.

2. There is insufficient evidence to recommend an optimal transfusion threshold in the following groups of patients:

  • Acute coronary syndrome (small trials showed a trend towards lower mortality when using a liberal transfusion threshold)
  • Severe thrombocytopenia in patients treated for hematological or oncological disorders at risk for bleeding
  • Chronic transfusion-dependent anaemia (cancer, haematological malignancies, bone marrow failure)
  • Acute neurologic disorders such as stroke and traumatic brain injury



Manish S. Patel, MD Associate Professor of Medicine

Jeffrey L. Carson, MD Distinguished Professor of Medicine

Provost, Rutgers Biomedical Health Sciences

Division of General Internal Medicine

Rutgers – Robert Wood Johnson Medical School

New Brunswick, New Jersey, USA


The author

 Cynthia So-Osman

Cynthia So-Osman

ISBT Regional Director Europe, Clinical Consultant in Transfusion Medicine, Sanquin Blood Supply, The Netherlands