Blood transfusions are common in clinical practice. In 2014/15 NHS Blood and Transplant issued 1.7 million units of red blood cells, 275,000 units of platelets, 215,000 units of fresh frozen plasma and 165,000 units of cryoprecipitate to hospitals in England and North Wales. An estimated 430,000 patients received a red blood cell transfusion in 2002; a further study has not been conducted, but given the reduction in blood use since 2002 the number of patients who have had a transfusion is likely to be 10% to 20% lower than this figure.

Despite considerable efforts to ensure the safety of blood transfusions, they are associated with significant risks. The Serious Hazards of Transfusion (SHOT) scheme estimated that in 2014 the risk of transfusion‑related death was 5.6 per million blood components issued, and the risk of transfusion‑related major morbidity was 63.5 per million blood components issued, although it was not always certain that transfusion was the direct cause of death or major morbidity. Removing cases where patient harm was caused by delayed transfusion rather than transfusion itself reduces the risk of transfusion‑related death to 4.5 per million blood components issued, and the risk of transfusion‑related major morbidity was 61.9 per million blood components issued. The most common cause of death associated with transfusion was transfusion associated circulatory overload.

There is evidence from the SHOT scheme and national audits of transfusion practice that:

  • some patients are receiving the wrong blood components

  • the choice of blood component is not always based on clinical findings and laboratory test values

  • patients are not always monitored for the adverse effects of transfusion, and these effects are not always managed correctly

  • some patients are transfused unnecessarily, which is wasteful of a scare and costly resource and put patients at unnecessary risk.

Accurate patient identification is a crucial step. Giving a patient the wrong blood transfusion is an avoidable serious hazard, and can result from errors made anywhere in the transfusion process.

There has been an approximate 25% decline in the transfusion of red blood cells in England in the last 15 years. The red blood cell transfusion rate declined from 45.5 to 36 units per 1,000 people between 1999 and 2009, and since then has dropped further to around 31.5 units per 1,000 people. This rate is a little higher than in Northern Ireland, the Netherlands and Canada, but is considerably lower than in the United States. In contrast, the use of platelets and fresh frozen plasma has been increasing. The proportion of red blood cells used between 1999 and 2009 in surgical patients has declined from 41% to 29% of all red cells transfused, and in medical patients has increased from 52% to 64% of all red cells transfused. Use in obstetrics and gynaecology has remained stable at 6%. A national audit of blood transfusion in 2014 showed that the proportion of red cell transfusions used in surgical patients continues to decline and was 27% of all red cells transfused with a corresponding increase in medical patients to 67%.

This guideline contains recommendations about general principles of blood transfusion, and applies to a range of conditions and different settings. It does not include recommendations relating to specific conditions.

The guideline covers:

  • the appropriate use of blood components

  • alternatives to transfusion for surgical patients

  • ensuring patient safety, including monitoring for transfusion reactions

  • providing patients with information about transfusion.

This guideline focuses on the general principles of transfusion. To do this, it was necessary to limit the scope by excluding:

  • patient groups with special transfusion needs, such as fetuses, neonates and children under 1 year old, pregnant women, and patients with haemoglobinopathies.

  • specialist areas already covered by NICE guidelines, for example, anaemia in chronic kidney disease, upper gastrointestinal bleeding and trauma and massive haemorrhage.

  • the use and administration of blood products, such as intravenous immunoglobulin, anti‑D and recombinant activated factor VII.

  • near‑patient testing for haemoglobin concentration and haemostasis.

  • laboratory procedures relating to the safety and quality of blood, including pre‑transfusion compatibility testing.

  • the diagnosis of anaemia.

  • the management of anaemia in medical patients is out of the scope of this guidance, but it is important to note that the correct approach for managing anaemia in medical patients is important for avoiding unnecessary use of blood.

Despite the lack of specific evidence in the paediatric population, a number of the recommendations have been considered applicable to children following extrapolation from evidence in adults. This was considered to be a reasonable approach to provide some guidance for this age group. However, it should be noted that the guidelines do not cover transfusion for neonates and infants less than a year of age due to the difficulties in extrapolating adult evidence to very young children.