The United States reports laboratory results in mg/dL and most of the world in mmol/L. There is no factor between them — there is a different factor for every test.
Calculator
Try:
Glucose — 100 mg/dL
5.551 mmol/L
The factor is the molar mass over ten: Glucose is 180.156 g/mol, so mg/dL ÷ 18.0156 gives mmol/L. That number is different for every test on this list.
The same 100 mg/dL, read as each test
Why there is no single mg/dL to mmol/L factor
Test
Molar mass
Divisor
Would be
Glucose
180.156
18.0156
5.551 mmol/L
Cholesterol (total, HDL, LDL)
386.654
38.6654
2.586 mmol/L
Triglycerides
885.432
88.5432
1.129 mmol/L
Creatinine
113.118
11.3118
8,840.3 µmol/L
Urea
60.056
6.0056
16.651 mmol/L
Urea nitrogen (BUN)
28.014
2.8014
35.696 mmol/L
Uric acid
168.112
16.8112
5,948.4 µmol/L
Bilirubin
584.673
58.4673
1,710.4 µmol/L
Calcium
40.078
4.0078
24.951 mmol/L
Magnesium
24.305
2.4305
41.144 mmol/L
The divisors span a factor of 36. Using the glucose divisor of 18 on a cholesterol result gives an answer more than twice too large, and both numbers look entirely reasonable on a report. Cholesterol and triglycerides sit on the same lipid panel and their divisors differ by 2.3.
Urea and BUN are not the same measurement
The same sample as urea30.01 mg/dL× 2.1438
In SI5.00 mmol/L of ureareading the BUN as urea instead would give 2.33
BUN weighs only the two nitrogen atoms; urea weighs the whole molecule, which is 2.14 times heavier. Both are reported in mg/dL and the label is the only thing that distinguishes them, so a renal result carried across systems without checking which measurand it is comes out low by that factor. US reports say BUN; most other countries report urea in mmol/L.
Haemoglobin has three units
The same haemoglobin, in each unit
Unit
Value
Where it is used
g/dL
14.000
United States, India, much of Asia
g/L
140.0
United Kingdom, Canada, Australia
mmol/L
8.688
Netherlands, Scandinavia
The g/dL to g/L step is exactly ten, which is why a result of 14 and a result of 140 can describe the same sample — and why the difference reads as a misplaced decimal point when it is not one. The mmol/L figure counts haem groups and is a different scale again.
Two more places the prefix matters
Which SI unit each test is reported in
Test
US
SI
Note
Glucose
mg/dL
mmol/L
The famous divisor of 18. Reported in mmol/L almost everywhere except the United States.
Cholesterol (total, HDL, LDL)
mg/dL
mmol/L
Divisor 38.67. All three fractions use the same factor, since they are all cholesterol.
Triglycerides
mg/dL
mmol/L
Divisor 88.57, taking triolein as the representative molecule. Nearly five times the cholesterol divisor, on the same lipid panel.
Creatinine
mg/dL
µmol/L
Reported in µmol/L, not mmol/L — the multiplier is 88.4.
Urea
mg/dL
mmol/L
The whole molecule. Divisor 6.006.
Urea nitrogen (BUN)
mg/dL
mmol/L
Counts only the two nitrogens, so it is a different measurand from urea and 2.14 times lighter. Divisor 2.80.
Uric acid
mg/dL
µmol/L
Reported in µmol/L; the multiplier is 59.48.
Bilirubin
mg/dL
µmol/L
Reported in µmol/L; the multiplier is 17.10.
Calcium
mg/dL
mmol/L
The smallest divisor here, 4.008 — a quarter of glucose and a twentieth of triglycerides.
Magnesium
mg/dL
mmol/L
Divisor 2.431. Also reported in mEq/L, which is twice the mmol/L figure since it is divalent.
Creatinine, uric acid and bilirubin are reported in µmol/L rather than mmol/L, because their concentrations are a thousandth of the others — so a creatinine quoted in the wrong prefix is out by a thousand. And calcium and magnesium are sometimes given in mEq/L, which for a divalent ion is twice the mmol/L figure: 2.4 mmol/L of calcium is 4.8 mEq/L. Sodium and potassium are monovalent, so their two numbers coincide — which is precisely why the divalent ones catch people out.
What this converter covers
Ten common analytes, each with the molar mass its factor comes from — plus urea against BUN, and haemoglobin’s three units.
Free, no signup — exact by definition, not an estimate.
Updated 8 September 2026
At a glance
Formula shown
mmol/L = 10 × (mg/dL) ÷ molar mass · urea mg/dL = BUN mg/dL × 2.1438
Scenario support
Glucose 100 mg/dL is 5.55 mmol/L · cholesterol 200 is 5.17 · BUN 14 is urea 5.00
Educational estimate
Planning support from the values you enter — not professional advice.
There is no single factor
“Divide by 18” is the answer everyone remembers, and it is right for exactly one test. The conversion is ten divided by the molar mass of whatever is being measured, so it changes with the substance.
The divisor for each test, and where it comes from
Test
Molar mass
mg/dL ÷
Gives
Magnesium
24.31
2.431
mmol/L
Urea nitrogen (BUN)
28.01
2.801
mmol/L
Calcium
40.08
4.008
mmol/L
Urea
60.06
6.006
mmol/L
Creatinine
113.12
11.312
mmol/L (÷1000 for µmol)
Uric acid
168.11
16.811
mmol/L
Glucose
180.16
18.016
mmol/L
Cholesterol
386.65
38.665
mmol/L
Bilirubin
584.67
58.467
mmol/L
Triglycerides
885.43
88.543
mmol/L
A spread of 36 times from top to bottom. Applying the glucose divisor to a cholesterol result gives an answer more than twice too high, and nothing about it looks wrong: 200 mg/dL of cholesterol becomes 11.1 instead of 5.17, and both are numbers a reader could believe.
The two lipids on a single panel already differ by 2.3 — cholesterol divides by 38.67 and triglycerides by 88.57, because a triglyceride molecule is more than twice as heavy. So even converting one report consistently means using two different numbers on adjacent lines.
The reason is straightforward once stated. mg/dL is a mass per volume and mmol/L is an amount of substance per volume, and the bridge between mass and amount is the molar mass — the same relationship that makes grams-to-moles impossible without knowing the substance.
Urea and BUN are different measurements
This one is worse than a unit mismatch, because both figures are reported in mg/dL and only the label distinguishes them.
Blood urea nitrogen counts only the nitrogen in the urea — two atoms, 28.01 g/mol. Urea counts the whole molecule, 60.06 g/mol. The same specimen therefore gives two numbers 2.14 times apart, and US reports use the first while most of the world uses the second.
One specimen, described four ways
As
Value
Where reported
BUN
14 mg/dL
United States
Urea
30.0 mg/dL
Some conventional-unit reports
Urea
5.00 mmol/L
UK, EU, India, Australia
BUN
5.00 mmol/L
Rare, and ambiguous without the label
Reading a BUN of 14 as though it were urea, and dividing by 6.006, gives 2.33 mmol/L instead of 5.00 — under half. It is the commonest error in carrying a renal panel between systems, and it fails quietly because both the input and the output are plausible figures.
The habit that avoids it: read the label rather than the number. “BUN” or “urea nitrogen” is one measurand and “urea” is another, and a report that says only “urea” with a mg/dL figure needs confirming before it is converted.
The prefix belongs to the test
Even within the SI convention, not every result is in mmol/L. Creatinine, uric acid and bilirubin are reported in µmol/L, because their concentrations are around a thousandth of glucose or cholesterol and a mmol/L figure would be an awkward string of zeros.
So a creatinine of 88 is µmol/L and a glucose of 5.5 is mmol/L, and the two units appear on the same page without either being labelled unusual. Converting a creatinine into mmol/L and comparing it with a European reference figure in µmol/L is a thousandfold error — which at least announces itself, unlike the two-fold ones above.
A third convention adds a further trap for the ions. Calcium and magnesium are sometimes reported in mEq/L, which for a divalent ion is twice the mmol/L figure: 2.4 mmol/L of calcium is 4.8 mEq/L. Sodium and potassium are monovalent, so their mmol/L and mEq/L numbers coincide — which is exactly why people carry the habit of treating the two units as interchangeable into the cases where they are not.
None of these is a mistake by anyone. Each convention makes a particular set of results readable. They only collide when a figure travels between systems without its unit.
Haemoglobin, and the factor of ten
Haemoglobin is quoted three ways, and the difference between two of them is exactly ten — which reads as a misplaced decimal point and is not one.
The same haemoglobin, in each convention
Unit
Value
Where
g/dL
14.0
United States, India, much of Asia
g/L
140
United Kingdom, Canada, Australia
mmol/L
8.69
Netherlands, Scandinavia
A decilitre is a tenth of a litre, so the first two are the same measurement with the decimal point moved. The third counts haem groups rather than grams and is a different scale entirely — a figure of 8.69 alongside one of 14.0 is not a disagreement about the sample.
The reason this one is worth watching is that both 14 and 140 are readable numbers in their own convention, so a value copied without its unit is not obviously wrong in either direction. Unlike a thousandfold error, a tenfold one can survive a glance.
What this page will not do
It converts units. It does not carry reference ranges, and that omission is deliberate rather than an oversight.
A reference range is a property of the laboratory, not of the analyte. It depends on the assay and the instrument, on whether the sample was serum or plasma or whole blood, and on the population the range was established in — so two laboratories can report the same specimen against different limits, and both are correct. A range printed next to a converted number would read as a verdict the page is in no position to give.
Several results also depend on how the sample was taken, in ways no conversion can carry across. A fasting glucose and a random one are different measurements of the same analyte. LDL cholesterol is often calculated from the other lipids rather than measured, and the calculation is less reliable when triglycerides are high. Creatinine-based estimates of kidney function use equations that have been revised, so an older result and a newer one are not directly comparable even in the same units.
What the page is good for: reading a report issued in the other convention, comparing a result taken abroad with one taken at home, and checking that a figure quoted in an article or a guideline is in the units you think it is. For anything that bears on a decision, the laboratory that produced the result and the clinician who ordered it have the method, the range and the context that a converter does not.
Related calculators
Other concentration and laboratory tools:
Moles to GramsGrams, moles, molarity, molality and normality — asking for the molar mass and the density, because neither conversion exists without them.
PPMppm to percent, mg/L and µg/m³ — asking which liquid or which gas, because without that the conversion has no answer.
Water Hardnessppm, gpg, °dH, °fH and °e — every factor derived from molar masses, plus the calcium and magnesium a lab actually reports.
WeightKilograms, pounds, ounces, stone and tonnes, listing the short, long and metric ton separately — three masses, one word.
VolumeLitres, gallons, pints and cubic units — with US and imperial named apart, because a UK gallon is 20% larger than a US one.
Densitykg/m³, lb/ft³, lb/in³ and lb/gal, with specific gravity against any water reference and the API scale that runs backwards.
Every factor here is computed from the IUPAC atomic weights and then cross-checked against the published clinical conversion tables — deriving without the check would miss a wrong molar mass, and transcribing without the derivation would miss a wrong formula. The convention that each analyte carries its own SI prefix is the IFCC one, which is why creatinine appears in µmol/L rather than in a very small mmol/L figure.
SI Unit Conversion Table for common laboratory analytesNational Center for Biotechnology Information · verified 2026-09-08 · The published per-analyte conversion factors this page derives independently and cross-checks against — 18.0 for glucose, 38.67 for cholesterol, 88.57 for triglycerides, 88.4 µmol per mg/dL for creatinine
IUPAC standard atomic weightsInternational Union of Pure and Applied Chemistry · verified 2026-09-08 · The atomic weights each molar mass is summed from, which is why every factor here is derived rather than transcribed and why urea and its nitrogen content come out as separate figures
The International System of Units in clinical laboratory scienceInternational Federation of Clinical Chemistry and Laboratory Medicine · verified 2026-09-08 · The convention that each analyte is reported in the SI prefix appropriate to its concentration — mmol/L for glucose and lipids, µmol/L for creatinine, bilirubin and uric acid — and the distinction between amount-of-substance and mass concentration
Conversion note
This converts units. It does not interpret results, and it deliberately carries no reference ranges — a reference range belongs to the laboratory that produced the result, because it depends on the assay, the instrument, the sample type and the population, and two laboratories can report the same specimen against different ranges. A converted number is the same result in different units and is not a diagnosis, a trend or a reason to change anything. Some results additionally depend on how the sample was taken: a fasting and a random glucose are not comparable, LDL cholesterol is often calculated rather than measured, and creatinine-based kidney estimates use equations that have changed. If a result matters, take it to the clinician or laboratory that ordered it — they have the method, the range and the context this page does not. Nothing here is medical advice.
Published the mg/dL to mmol/L Converter for ten common laboratory analytes: glucose, cholesterol, triglycerides, creatinine, urea, urea nitrogen, uric acid, bilirubin, calcium and magnesium.
Establishes that there is no single mg/dL to mmol/L factor: it is ten divided by the molar mass, so the divisors run from 2.43 for magnesium to 88.54 for triglycerides, a spread of 36 times.
Shows the concrete cost: applying the familiar glucose divisor of 18 to a cholesterol result of 200 gives 11.10 instead of 5.17, and both figures look entirely reasonable on a report.
Notes that cholesterol and triglycerides sit on the same lipid panel and their divisors differ by 2.3, so converting one report consistently already requires two different numbers on adjacent lines.
Separates urea from blood urea nitrogen, which are different measurands both reported in mg/dL: BUN weighs only the two nitrogen atoms and urea the whole molecule, 2.1438 times heavier, so a BUN of 14 is a urea of 30.0 mg/dL or 5.00 mmol/L while reading it as urea gives 2.33.
Records that the SI prefix belongs to the analyte rather than the country -- creatinine, uric acid and bilirubin are reported in micromoles per litre because their concentrations are a thousandth of the others.
Covers the milliequivalent convention for the ions, where a divalent calcium or magnesium figure is twice its millimolar one while monovalent sodium and potassium coincide, which is why the habit of treating the units as interchangeable travels into the cases where it fails.
Handles haemoglobin's three units, where g/dL to g/L is exactly ten and therefore reads as a misplaced decimal point without being one, while the millimolar figure counts haem groups on a different scale again.
Derives every factor from the IUPAC atomic weights AND cross-checks it against the published clinical table, since deriving alone would miss a wrong molar mass and transcribing alone would miss a wrong formula.
Carries no reference ranges by design, and says why: a range belongs to the laboratory that issued the result because it depends on the assay, the instrument, the sample type and the population, so a range printed beside a converted number would read as a verdict this page cannot give.
States what a conversion cannot carry across -- fasting versus random glucose, LDL that is calculated rather than measured, and creatinine equations that have been revised -- and directs anything that bears on a decision to the laboratory and clinician who have the context.
Verified by 45 automated cases, including that the divisors span 36 times, that urea and BUN are stored as separate analytes rather than one with a factor, that each analyte routes to its own SI prefix, and that no reference range has leaked into the data.
Add this calculator to your site
Responsive embed — and private: nothing your visitors type leaves their browser.