Here is a thorough summary of HbA1c from authoritative medical textbooks:
HbA1c (Glycated Hemoglobin / Haemoglobin A1c)
What Is It?
HbA1c is formed when blood glucose enters red blood cells and forms a covalent, non-enzymatic adduct with the epsilon-amino groups of lysyl residues and the alpha-amino group of the N-terminal valines of hemoglobin beta chains. This process is called glycation (not glycosylation - it requires no enzyme). The higher the blood glucose, the more glycation occurs.
Because red blood cells survive approximately 120 days, the HbA1c level reflects mean blood glucose over the preceding 8-12 weeks - a far more informative window than a single glucose reading. - Harper's Illustrated Biochemistry, 32nd Ed
Reference Range (in Non-Diabetic Adults)
The scatter plot below (from Tietz Textbook of Laboratory Medicine, 7th Ed) shows the population distribution of HbA1c by age and sex, with reference interval bands:
- Normal: < 5.7% (< 39 mmol/mol)
- HbA1c tends to rise slightly with age in population data
Diagnostic Cut-points (ADA Criteria)
| Glycemic Status | HbA1c (%) | HbA1c (mmol/mol) |
|---|
| Normal | < 5.7% | < 39 |
| Prediabetes | 5.7 - 6.4% | 39 - 47 |
| Diabetes | ≥ 6.5% | ≥ 48 |
- To diagnose diabetes, two abnormal results are generally required (either two different tests or two occasions), unless classic symptoms plus a random glucose ≥ 200 mg/dL are present.
- A random plasma glucose ≥ 200 mg/dL with symptoms is sufficient alone.
- Note: The WHO defines impaired fasting glucose at a slightly higher threshold (FPG > 6.1 mmol/L vs. ADA's ≥ 5.6 mmol/L).
Sources: Goldman-Cecil Medicine, Harrison's Principles of Internal Medicine 22E, Henry's Clinical Diagnosis
Treatment Targets
| Population | HbA1c Target |
|---|
| Most non-pregnant adults | < 7.0% (53 mmol/mol) |
| Stringent target (selected patients) | ≤ 6.5% |
| Elderly with intact cognition | < 7.0–7.5% |
| Elderly with cognitive impairment / limited lifespan | < 8.0% (to avoid hypoglycemia) |
| Dialysis patients (KDOQI/KDIGO) | ~7.0% |
The target should always be individualized - considering age, comorbidities, hypoglycemia risk, cognition, and lifestyle. Large trials (DCCT, UKPDS, ACCORD, ADVANCE) confirm that better glycemic control reduces microvascular complications, and early intensive control has lasting cardiovascular benefit. - Harrison's 22E
- The ADA recommends at least biannual HbA1c measurements for diabetes follow-up; quarterly for patients not at goal or after treatment changes. - Rosen's Emergency Medicine
Why HbA1c Can Be Falsely Low or High
Falsely Low (underestimates glucose):
- Hemolytic anemia (shorter RBC lifespan = less time for glycation)
- Hemoglobinopathies (e.g., sickle cell, thalassemia - variant Hbs may not glycate normally)
- Renal impairment (increased RBC turnover, erythropoietin therapy)
- After splenectomy
Falsely High (overestimates glucose):
- Chronic use of opioids, salicylates, or alcohol (assay interference)
- Iron deficiency anemia (older RBCs persist longer, accumulate more glycation)
In cases where HbA1c is unreliable, fructosamine (glycated serum protein) or glycated albumin can be used - both reflect mean glucose over the preceding 2-3 weeks. These are less standardized but clinically useful as alternatives. - Goldman-Cecil Medicine
Key Points in Summary
- Mechanism: Non-enzymatic glycation of Hb beta chains - proportional to mean blood glucose
- Window: Reflects ~8-12 weeks average glycemia (RBC lifespan ~120 days)
- Normal: < 5.7% | Prediabetes: 5.7-6.4% | Diabetes: ≥ 6.5%
- Therapy goal: < 7% for most adults; individualize for elderly/high-risk
- Limitations: Unreliable in hemolytic anemia, hemoglobinopathies, renal failure, iron deficiency, and certain drug use - use fructosamine/glycated albumin as backup
- Monitoring frequency: At least every 6 months; every 3 months if not at goal