Blood Sugar Converter

Two units, one number, and a world that never agreed

Your blood sugar is one physical fact. There are two ways of writing it down, and which one you use depends almost entirely on where you happen to be standing.

In the United States, and in a few other places, glucose is reported in mg/dL, milligrams per decilitre. That is a measure of mass: how many milligrams of glucose are floating in each tenth of a litre of blood. Most of the rest of the world, including the UK, Australia, Canada, New Zealand, China and most of Europe, uses mmol/L, millimoles per litre. That is a measure of molecules: how many of them there are, rather than what they weigh. International guidance from the World Health Organization and the International Diabetes Federation uses mmol/L.

Neither is more correct. mmol/L is the SI unit and the one chemists would pick, since counting molecules tells you more about what is chemically going on. mg/dL got there first in American labs and stayed. So a fasting reading of 5.5 and a fasting reading of 99 describe exactly the same blood, and this tool moves you between them.

Where the 18 comes from

Not from a committee. From chemistry, and it is worth thirty seconds because it makes the number stop feeling arbitrary.

A molecule of glucose has a molar mass of about 180.16 g/mol, which means one mole of glucose weighs 180.16 grams, and one millimole weighs 180.16 milligrams. So a concentration of 1 mmol/L is 180.16 mg per litre.

But mg/dL is per decilitre, and there are ten decilitres in a litre. So divide by ten:

180.16 ÷ 10 = 18.0 mg/dL per 1 mmol/L

That is the whole derivation. The factor is glucose's own molecular weight, scaled for the fact that one unit uses litres and the other uses tenths of a litre. The tool rounds 18.0182 to 18, which is what clinicians do too, and the difference is negligible: across the entire normal-to-diabetic range it never shifts the answer by more than about 0.01 mmol/L.

The conversion that matters most

Of all the numbers you could put through this converter, one deserves to be memorised:

126 mg/dL ÷ 18 = 7.0 mmol/L

That is the fasting glucose threshold at which diabetes is diagnosed, and it is the same threshold in both systems. When you see "126" in an American article and "7.0" in a British one, they are not different standards. They are the same line, written in two alphabets.

It also happens to be a small gift from the arithmetic: 126 divided by 18 is exactly 7, with no rounding at all. The other direction works just as tidily. 7.0 × 18 = 126.

The table worth bookmarking

Every number that matters, in both units, so you never have to convert the important ones again. These are fasting values unless stated:

One thing to hold in mind: a single reading does not diagnose anything. Unless someone has unmistakable symptoms, a diagnosis needs an abnormal result confirmed, either by a second test or by a different test pointing the same way. One high number on one morning is a reason to get tested properly, not a conclusion.

Mental shortcuts that actually work

For when you are reading a lab report on your phone and cannot be bothered to open a converter.

mmol/L to mg/dL: multiply by 20, then knock off 10 percent. Because 18 is 20 minus a tenth of 20. So 5.5 mmol/L: 5.5 × 20 = 110, minus 11 = 99 mg/dL. Which is exactly right.

mg/dL to mmol/L: divide by 20, then add 10 percent. So 126: 126 ÷ 20 = 6.3, plus 0.63 = 6.93, near enough 7.0.

Or just remember the anchors. Most people only need four: 100 mg/dL is about 5.5, 126 is 7.0, 140 is 7.8, and 180 is 10.0. Learn those and you can eyeball almost any reading you will meet.

The 180 to 10.0 pairing is a nice one, incidentally, because 180 mg/dL is the common post-meal target for people with diabetes and it lands on a perfectly round 10.

Why mixing up the units is genuinely dangerous

The two scales differ by a factor of eighteen, which means a number that is perfectly healthy in one unit is either impossible or catastrophic in the other. There is no overlap where a mistake would be harmless.

So the unit is not a formatting detail, it is half the information. This matters most in a few specific situations: reading a lab report issued in another country, using a meter bought abroad or gifted by a relative, reading research papers, or being advised by family overseas who learned their numbers on the other scale. If someone tells you their sugar was "110" and they live in London, ask, because they either mean a normal American reading or something that would have them in resuscitation.

Meters can usually be set to either unit, and a meter quietly displaying the wrong one is a real hazard. If a reading ever looks wildly out of character, check the unit on the screen before you panic or act on it.

The 18 works for glucose and nothing else

Worth stating plainly, because it is a natural and consequential mistake.

The factor 18 is derived from glucose's molar mass. Every other substance in your blood has its own molar mass, so every other substance has its own conversion factor. Cholesterol converts at about 38.67, triglycerides at 88.57, creatinine at 88.4. Using 18 on any of them gives you a confidently wrong answer.

So this converter is a glucose converter, and only that. If your lab report has cholesterol in mmol/L and you want mg/dL, this is the wrong tool and the right answer is a different factor.

About those two decimal places

The tool reports its answers to two decimals, which is more precision than the underlying reality supports, and it is worth knowing so you do not read meaning into digits that have none.

In practice, labs report mmol/L to one decimal place, and mg/dL as a whole number. Nobody has ever been handed a result of 99.00 mg/dL. So when this converter tells you 5.56 mmol/L, the honest reading is 5.6, and the second decimal is decoration.

And the deeper point is worth more than the formatting one. Blood glucose is genuinely variable, even within the same person on the same day. The typical biological variation in a fasting glucose test is around 5.7 percent, which means a result of 126 mg/dL could reflect a true value anywhere from roughly 110 to 142. Against that kind of natural wobble, arguing about the second decimal place is like measuring a beach with a micrometer.

Which is the useful takeaway: read your number as a band, not a point. That is how your doctor reads it too.

HbA1c is a different animal entirely

This trips people up constantly, so: HbA1c is not blood sugar, and this converter will not touch it.

A glucose reading is a snapshot of this moment. HbA1c measures how much glucose has stuck to your haemoglobin over the lifespan of your red cells, so it reflects your average over roughly the past two to three months. Different quantity, different question.

And confusingly, HbA1c has its own two-unit problem:

• Percent, the older unit, still used in the US. Diabetes is diagnosed at 6.5% or above.
• mmol/mol, the newer international unit, used in the UK and much of the world. That same threshold is 48 mmol/mol.

So 6.5% and 48 mmol/mol are the same result, and neither converts with the 18 on this page. If someone quotes you an A1c of 48, they have not got a blood sugar reading of 48; they have a different test in a different unit that happens to share the territory.

Who uses which, and why you might need both

Which sounds tidy until your life crosses a border. The people who actually need this converter tend to be: someone who moved country and has a decade of readings in the old unit; someone whose meter came from abroad; anyone reading research, since papers use both; and anyone with family managing diabetes on the other side of the world, where a phone call about numbers needs a translation layer.

If you keep long-term records, pick one unit and convert everything into it rather than keeping a mixed log. A spreadsheet with both units in one column is a trap you will eventually fall into.

Questions people ask

What is 100 mg/dL in mmol/L?

About 5.5 mmol/L (5.56 by the tool's arithmetic, and your lab would print 5.5). That sits at the top of the normal fasting range, just under the 5.6 mmol/L mark where prediabetes begins. Which is a good illustration of why the unit matters: 100 sounds like a round, comfortable number, and it is sitting right on a line.

Why is 126 mg/dL the diabetes number?

Because it converts to exactly 7.0 mmol/L, and 7.0 is where the evidence put the threshold. The mg/dL figure looks arbitrary precisely because it is a translation of a round number in the other unit. Fasting glucose at or above that level, confirmed, is one of the diagnostic criteria for diabetes.

Can I convert my HbA1c with this?

No. HbA1c is a different measurement entirely, an average over two to three months rather than a reading now, and it uses its own units of percent or mmol/mol. The factor of 18 does not apply to it. An A1c of 6.5% is the same as 48 mmol/mol, and neither has anything to do with the conversion on this page.

Can I use this for cholesterol?

No, and please do not. The 18 comes from glucose's molar mass, and every other substance has a different one. Cholesterol uses roughly 38.67, triglycerides 88.57. Applying 18 to them produces a number that looks plausible and is wrong, which is the worst kind of wrong.

Should I use 18 or 18.0182?

18 is fine, and it is what clinical practice uses. Across the whole range you are likely to care about, the two factors never disagree by more than about 0.01 mmol/L, which is far smaller than the natural variation in the test itself. Both give 7.0 at the diagnostic threshold, which is the case that matters.

My converted number looks high. What now?

Do not diagnose yourself from one reading and a converter. Glucose swings with food, illness, stress, sleep and the time of day, and a single high number is common and often means nothing. What it is worth doing is taking the reading, in whichever unit, to a doctor, who can test properly and interpret it alongside everything else about you. This page converts units. It does not know anything about you.

References

Where the figures come from. The diagnostic thresholds quoted above, namely fasting glucose at or above 126 mg/dL (7.0 mmol/L), a 2-hour glucose tolerance value at or above 200 mg/dL (11.1 mmol/L), and an HbA1c at or above 6.5%, together with the requirement that an abnormal result be confirmed, follow the American Diabetes Association's diagnostic criteria. The description of the fasting, tolerance, A1c and random plasma glucose tests, and the figure of roughly 5.7 percent for typical within-person biological variation in a fasting glucose result, come from the National Institute of Diabetes and Digestive and Kidney Diseases.

1. American Diabetes Association. Diagnosis. https://diabetes.org/about-diabetes/diagnosis
2. National Institute of Diabetes and Digestive and Kidney Diseases. Diabetes Tests & Diagnosis. niddk.nih.gov