Haemoglobin A1c (revision 23)
Old revision·19:12, 20 Mar 2026·EpitalonEoin
| Haemoglobin A1cLaboratory medicine | |
|---|---|
| Abbreviation | HbA1c |
| Reflects | Average glucose over ~8–12 weeks |
| Units | mmol/mol (IFCC); % (DCCT/NGSP) |
| Confounded by | Anything altering red-cell lifespan |
| Topic infobox · conventions | |
Haemoglobin A1c is the fraction of haemoglobin that has been non-enzymatically glycated. Because glycation is irreversible and proceeds at a rate proportional to glucose concentration, the fraction integrates glucose exposure over the lifespan of the circulating red cells — roughly the preceding eight to twelve weeks, weighted towards the more recent.[1]
It is reported in two unit systems: mmol/mol under the IFCC reference method, and percentage under the older DCCT-aligned convention. Both appear in the literature and conversion between them is defined.[2]
Its principal limitation follows from its mechanism: anything that alters red-cell lifespan alters the result independently of glucose. Haemolysis and blood loss lower it; iron deficiency and reduced turnover raise it.[1]
What it measures
[edit]Glycation is a slow non-enzymatic reaction between glucose and the N-terminal valine of the haemoglobin β-chain. Its extent depends on both glucose concentration and exposure time, and because red cells are replaced continuously the population of cells sampled at any moment carries a weighted history.[2]
The weighting is not uniform: roughly half the value reflects the preceding month, and the remainder the two months before. A change made three weeks ago is therefore only partly expressed.[1]
Estimated average glucose can be derived from HbA1c by a published regression, and the derivation carries substantial individual variation — two people with identical HbA1c can have appreciably different mean glucose.[2] The measure is a laboratory determination with its own method dependence, as any determination is.[3]
Confounders
[edit]| Condition | Direction |
|---|---|
| Haemolysis, blood loss, transfusion | Lower |
| Iron or B12 deficiency | Higher |
| Chronic kidney disease | Variable |
| Haemoglobin variants | Assay-dependent interference |
| Pregnancy | Lower, from increased turnover |
Haemoglobin variants interfere with some assay methods and not others, so a discrepancy between HbA1c and other glycaemic measures should prompt a question about the method rather than an assumption about glucose.[1]
Where HbA1c is unreliable, fructosamine or continuous glucose monitoring metrics provide alternatives measuring different windows.[2]
Use in this field
[edit]HbA1c is the primary glycaemic endpoint in essentially every type 2 diabetes trial discussed on this wiki, including the SUSTAIN and SURPASS programmes. Reductions of 1–2.5 percentage points are the range these agents produce.[4]
Because the measure integrates over months, a trial cannot demonstrate a glycaemic effect faster than the analyte can move, which is why glycaemic trials run at least six months.[1]
It appears on most of the panels discussed in this field; see Baseline laboratory panel. Nothing on this wiki is medical advice.[1]
See also
References
- ^ a b c d e f American Diabetes Association. "Standards of Care in Diabetes." Diabetes Care 47(Suppl 1) (2024).
- ^ a b c d Nathan DM, Kuenen J, Borg R, et al. "Translating the A1C assay into estimated average glucose values." Diabetes Care 31(8):1473–1478 (2008). PMID 18540046.
- ^ United States Pharmacopeia, General Chapter <1503>, Quality Attributes of Synthetic Peptide Drug Substances.
- ^ Drucker DJ. "Mechanisms of action and therapeutic application of glucagon-like peptide-1." Cell Metabolism 27(4):740–756 (2018). PMID 29617641.