[Sources of error when using haemoglobin A1c]. / Feilkilder ved bruk av hemoglobin A1c.

BACKGROUND: Measurement of glycated haemoglobin A1 in whole blood (b-HbA1c) can be used in both diagnosing and following up patients with diabetes. Correct interpretation of analytical results is contingent on agreement between average plasma glucose (p-glucose) and b-HbA1c. This article provides an overview of factors that may result in a discrepancy between average glucose concentration and b-HbA1c. METHOD: Literature search in PubMed to identify scientific articles that describe strengths and weaknesses of b-HbA1c. RESULTS: The b-HbA1c reading usually provides a good picture of average p-glucose for the preceding two to three months. Patients who are being treated with iron/vitamin B12 supplements, have liver failure, haemolytic anaemia or bleeding usually have a lower b-HbA1c than their p-glucose level would suggest. With increasing patient age, B12 deficiency or iron deficiency anaemia, higher values of b-HbA1c are seen for the same p-glucose level. Some ethnic groups have a higher b-HbA1c than their average p-glucose would suggest, but the risk of long-term complications appears generally to be more closely associated with b-HbA1c than with the glucose level. Pregnancy, renal failure or haemoglobinopathies may make the b-HbA1c value unreliable as an expression of average p-glucose. INTERPRETATION: Correct interpretation of b-HbA1c is conditional on the requisitioner being aware of possible sources of error. If the patient is suspected to have a condition that leads to lack of consistency between b-HbA1c and average p-glucose, glucose-based criteria must be used in diagnosing diabetes.