TherapyLog
HomeLab markers › HbA1c and fasting glucose

HbA1c and fasting glucose: the conversion, and why donating blood distorts one of them

HbA1c is an average, and averages have assumptions. The main one is that your red cells live about as long as everyone else's — which is exactly the assumption that routine phlebotomy breaks.

Last reviewed: 4 September 2026

What each of the three measures

Fasting glucose is a single point: what your blood glucose was at the moment of the draw, after a fast. It is sensitive to the previous day, to sleep, to stress and to whether the fast was real.

HbA1c is glycated haemoglobin, and it reflects average glucose exposure over the lifespan of the circulating red cells — conventionally described as two to three months, weighted towards the most recent weeks. It is stable and needs no fast.

Fasting insulin is the one most often left off and often the most informative here. Glucose can sit in range for years while insulin climbs to keep it there. A normal glucose with a high fasting insulin is a different metabolic picture from a normal glucose with a low one, and only one of them is reassuring.

The unit conversion

HbA1c is reported as a percentage (NGSP) in the United States and as mmol/mol (IFCC) in much of the rest of the world, and the two are related by a non-linear master equation rather than a simple factor. The converter above runs the app's own formula, which is exactly why it is worth using: it is one of the conversions people most often do wrong by reaching for a multiplier.

As a sanity check, 48 mmol/mol is about 6.5%, and 42 mmol/mol is about 6.0%. Fasting glucose converts more simply — mmol/L times 18.02 gives mg/dL — and 5.5 mmol/L is about 99 mg/dL.

Why HbA1c can mislead this audience specifically

Established clinical use HbA1c assumes a normal red cell lifespan. Anything that shortens it — haemolysis, some haemoglobinopathies, recent blood loss, transfusion — means the average cell has had less time to glycate, and the result reads falsely low.12 The reverse is true of anything that lengthens it.

That is not an exotic caveat here. Anyone managing testosterone-driven erythrocytosis with regular therapeutic phlebotomy or blood donation is deliberately and repeatedly shortening the average age of their circulating red cells,3 and the expected consequence is an HbA1c that understates true glucose exposure — potentially by enough to make a deteriorating metabolic picture look stable.

Off-label or community practice If you donate or undergo phlebotomy on a schedule, treat HbA1c as the weaker of your two glucose markers and lean on fasting glucose and fasting insulin, which do not depend on red cell lifespan. Where the question is important, fructosamine measures glycated serum protein over a shorter window and is unaffected by red cell turnover.2 Which to use is worth raising with your clinician rather than deciding alone — and the hematocrit page covers the other half of that trade.

What moves these on protocol

GH secretagogue effects

What drives it: MK-677, GHRP-2/6, ipamorelin, CJC — water retention and blood-sugar drift come with the territory.

What people report: Tingling hands or carpal-tunnel feeling, puffiness, vivid dreams, raging hunger (GHRP-6 and MK-677 especially).

What to measure: Fasting glucose and HbA1c on anything run for months — GH antagonizes insulin.

Worth knowing: Rising fasting glucose on a GH secretagogue is the signal to cut the dose or stop — insulin resistance is the long-term cost nobody logs.

Anything here is a reason to talk to the clinician who prescribes for you, with the result and the assay in front of you — not a reason to change a protocol on your own.

Off-label or community practice Growth hormone antagonises insulin, so GH secretagogues run for months are the compounds in this space most likely to move fasting glucose in the wrong direction. The app's guidance is to check fasting glucose and HbA1c on anything run long term, and to treat a rising fasting glucose as the signal to reconsider rather than as noise. The IGF-1 page covers the other side of that.

GLP-1 receptor agonists move these markers in the opposite direction, and substantially. A falling HbA1c on semaglutide or tirzepatide is the expected effect rather than a surprise. What that means for anything else you are taking, and for how often these are worth measuring, is a question for the prescriber.

What the app records for this marker

Marker
HbA1c
Panel group
metabolic
Canonical unit
%
Units accepted
% (canonical)
mmol/mol (converted by formula)
Generic reference range
4–5.6 % — generic reference range, male default. Your lab's printed interval takes precedence over this.
Optimal band
4.5–5.3 % non-diagnostic
LOINC
4548-4, 17856-6 — unverified seed, not checked against a vendor payload
Unit converter

Runs the app's own conversion, which means it refuses the conversions the app refuses instead of approximating them.

Your lab's range wins

The reference interval printed on your own report is the one that counts, and it is the one TherapyLog flags against when your report carries it. That is not deference for its own sake: a reference interval belongs to the assay, the instrument and the population the lab validated it on, and two labs measuring the same blood can legitimately print different intervals. A generic range — like the one in the fact box above — is what the app falls back to when your report did not include one, and it is the weaker of the two.

The same applies to any optimal band. Those are drawn from clinical literature and community practice for reading a trend over time, they are non-diagnostic, and sitting outside one is not by itself a finding. What a value means for you is a question for the clinician who can see your whole chart.

Compounds whose monitoring panel includes this marker

Generated by matching this marker's own aliases against the monitoring note on every compound in the app's reference, so the list is the app's, not an author's.

Questions people actually ask

My HbA1c is great but my fasting glucose is creeping up. Which do I believe?

Look at whether anything is shortening your red cell lifespan — regular donation or phlebotomy is the common one in this population, and it biases HbA1c low. If that applies, the fasting glucose is the more trustworthy of the two. Adding a fasting insulin usually settles the question, and the answer is worth taking to a clinician rather than watching for another year.

Does HbA1c need to be fasted?

No. That is one of its genuine advantages — it is an average over months, so a meal does not move it. Fasting glucose and fasting insulin do require a real fast, and the registry marks fasting status as context those two results cannot be read without.

48 mmol/mol — what is that in percent?

About 6.5%. The converter above uses the app's own formula rather than a multiplier, because the NGSP-to-IFCC relationship is non-linear and approximating it with a single factor is a common source of error.

Sources

Every claim above that is not a definition is either labelled by evidence tier or carries a numbered reference to one of these.

  1. Diagnostic Limitations of Hemoglobin A1c in the Setting of Compound Hemoglobinopathy Worked example of altered red cell turnover producing a falsely low HbA1c and delaying diagnosis.
  2. Unexpectedly Low HbA1c in a Patient With Newly Diagnosed Diabetes Mellitus and Thalassemia Trait The same failure mode, and the use of alternative glycaemic markers where red cell lifespan is altered.
  3. Diagnosis, management, and outcomes of drug-induced erythrocytosis: a systematic review, Blood Advances 9(9):2108 (2025) How commonly phlebotomy is used in this population, which is what makes the HbA1c caveat relevant.

TherapyLog logs this marker with the unit, the reference interval your report printed and the assay method beside it, so a trend cannot silently switch methods on you.

Log your bloodwork

Built by Joel Gonzales, founder of TherapyLog. Not a clinician. Last reviewed 4 September 2026. The calculator on this page runs the same code as the app; how these pages are written, sourced and corrected is set out in the editorial policy.

This calculator does the arithmetic you typed and nothing else. It does not know what is actually in your vial, whether the label is accurate, or anything about you. Confirm the vial strength and the diluent volume on your own label before you draw, and take dosing decisions to a qualified provider.

TherapyLog is an informational tracking tool and is not a substitute for professional medical advice, diagnosis, or treatment. Always consult a qualified healthcare provider before starting, changing, or stopping any medical protocol.
TherapyLog LLC · Floresville, Texas · Privacy Policy · Consumer Health Data · Terms of Use · About · hello@therapylog.app