2019 Re-audit of the medical use of red blood cells

Key objectives

• To reduce the variation in the medical use of blood so that quality becomes more consistent, reducing the risk to patients
• Audit a representative sample of physician decisions to prescribe red blood cells
• Gather information from a sample of patients under the care of a physician, such as reason for transfusion, clinical picture, co-morbidity and testing
• Identify which areas of physician transfusion practice are amenable to change
• Provide a national picture of process performance for comparative and benchmarking purposes

What did we audit?

• Sites were asked to audit consecutive transfusions of red blood cells (RBC) into adult patients during the audit period. They were asked to report one transfusion episode per patient and to focus on the first transfusion episode in the admission, where possible. All transfusions had to be ordered or prescribed by a physician, or under the guidance of a physician.

Who took part?

• 161 sites contributed data
• 5155 patients were analysed

What did we find?

• Full report (PDF 549KB)
• Key points (PDF 549KB)

Key findings and recommendations

• 96% (4930/5129) of audited patients had a pre-transfusion haemoglobin (Hb) taken within 3 days of transfusion.
  
  
• Significant numbers of asymptomatic or only mildly symptomatic patients are being transfused when their Hb levels are above the recommended thresholds.
  
  
• 30% (816/2745) of patients transfused with an Hb greater than 70g/L did not have an apparent adequate clinical reason for their transfusion.
  
  
• 61% (325/534) of patients with acute coronary syndrome or cardio-respiratory disease were transfused above a threshold of 80g/L without an apparent adequate clinical reason.
  Recommendation: Trusts should ensure that there is an appropriate reason for selecting red cell transfusion as the treatment of choice.
  
  
• 90% (4221/4715) of audited patients had a post-transfusion haemoglobin (Hb) taken within 3 days of transfusion (Standard 3).
  
  
• 21% of non-bleeding, multi-unit transfused patients had their haemoglobin level checked between units, and 27% had a documented clinical review (Standard 4).
  Recommendation: Trusts should assure themselves that the NICE and British Society of Haematology recommendations, amongst others, are supported within their medical settings, and that patients are not put at increased risk of unnecessary transfusion.
  
  
• One in five (1034/5155) patients were transfused because of iron-deficiency anaemia. Only one-third of these (333/1034) presented with a bleeding phenotype which may have explained the use of transfusion. 68% of transfused iron-deficient patients were not bleeding.
  Recommendation: Trusts should ensure that iron supplementation is utilised as the primary intervention unless it is clinically urgent to transfuse.
  
  
• Nearly 5% (231/5155) of transfusions were documented as given because of B12 or folate deficiency or both. Deficiency severe enough to require red cell support is unusual.
  Recommendation: Trusts should carefully examine their individual reports to look at these cases and consider deep-dive analysis of any that they contributed, to ensure that these transfusions were necessary.
  
  
• Only 50% (2581/5155) of transfusions had documentation that risks, benefits and alternatives had been explained (for this or a previous transfusion) (Standard 5).
  Recommendation: Trusts should assure themselves that there is a robust system in place to help transfusion authorisers to have access to patient information sources, in keeping with NICE guidelines and quality standards.