Insulin µIU/mL to pmol/L Converter

Insulin in two units: the short answer

To convert insulin from µIU/mL (also written µU/mL or mU/L) to pmol/L, multiply by 6. To convert pmol/L back to µIU/mL, divide by 6. A fasting insulin of 10 µIU/mL is therefore 60 pmol/L, and 90 pmol/L is 15 µIU/mL. The converter above works in both directions and compares your value with a typical fasting laboratory range.

Insulin is measured far less often than glucose, but fasting insulin has become popular for assessing insulin resistance, and it is an essential input to scores such as HOMA-IR. Those scores are defined in µIU/mL, so a result in pmol/L must be converted before use – one of the most common sources of wildly wrong HOMA-IR values online. This guide covers the conversion and its history, what fasting insulin can and cannot tell you, typical values, and the tests in which insulin plays a key role.

What insulin does

Insulin is a peptide hormone produced by the beta cells of the pancreatic islets. It is the body's main signal of the fed state. After a meal, rising glucose triggers insulin release, which:

• moves glucose into muscle and fat cells
• tells the liver to stop releasing glucose and to store glucose as glycogen
• promotes fat storage and inhibits fat breakdown
• promotes protein synthesis

Between meals and overnight, insulin falls to a low baseline, allowing the liver to release glucose and fat stores to be used.

Insulin resistance describes a state in which tissues respond less to insulin, so the pancreas must produce more to keep glucose normal. For years, high insulin can keep glucose normal; type 2 diabetes develops when beta cells can no longer keep up. Fasting insulin is therefore often raised in insulin resistance, prediabetes and early type 2 diabetes, and low in type 1 diabetes and advanced type 2 diabetes.

µIU/mL and pmol/L explained

• µIU/mL (micro-international units per milliliter) – the usual unit in the US and in many clinical calculators. It is numerically identical to µU/mL and mU/L (milli-units per liter).
• pmol/L (picomoles per liter) – the SI unit used in the UK, Europe and many research papers.

Why the factor is 6

International units for insulin were originally defined by biological activity and are now tied to the WHO International Standard for human insulin, in which 1 IU corresponds to about 0.0347 mg of insulin. Human insulin has a molar mass of about 5,808 g/mol. One µIU therefore weighs about 34.7 pg, which is about 6.0 femtomoles, so 1 µIU/mL ≈ 6.0 pmol/L. A 1993 analysis by Vølund recommended 6.00 as the conversion factor, and it is the factor most laboratories and clinical calculators use. Some older sources quote 6.945, which came from earlier potency estimates. If your laboratory states a different factor, use theirs.

How to convert insulin step by step

1. Check the unit: µIU/mL, µU/mL, mU/L or pmol/L.
2. For µIU/mL (or µU/mL, mU/L) to pmol/L, multiply by 6.
3. For pmol/L to µIU/mL, divide by 6.
4. Round to whole numbers for pmol/L and one decimal place for µIU/mL.

Worked examples

• 2 µIU/mL × 6 = 12 pmol/L – a low fasting value, typical of a very insulin-sensitive person.
• 5 µIU/mL × 6 = 30 pmol/L.
• 10 µIU/mL × 6 = 60 pmol/L.
• 20 µIU/mL × 6 = 120 pmol/L – the upper end of a typical fasting range.
• 25 µIU/mL × 6 = 150 pmol/L – a high fasting value suggesting insulin resistance.
• 48 pmol/L ÷ 6 = 8 µIU/mL.
• 174 pmol/L ÷ 6 = 29 µIU/mL.

Insulin conversion chart

Normal fasting insulin

A typical laboratory fasting insulin range is about 2–20 µIU/mL (12–120 pmol/L), but ranges vary widely, partly because insulin assays have historically differed substantially from one another. The reference population also matters: populations with high rates of obesity and insulin resistance produce higher upper limits.

There is no universally agreed cut-off for "insulin resistance" based on fasting insulin alone. In research, fasting insulin above roughly 12–15 µIU/mL (72–90 pmol/L) is often associated with insulin resistance in adults, but the threshold depends on the assay and population, and many guidelines do not recommend fasting insulin for routine clinical screening.

Why fasting insulin is not a routine test

Major diabetes organisations, including the American Diabetes Association, do not recommend measuring insulin to diagnose prediabetes or diabetes. Glucose, HbA1c and the oral glucose tolerance test are the diagnostic tools. Reasons include:

• insulin assays are less standardised than glucose and HbA1c assays
• insulin levels fluctuate considerably and are released in pulses
• there are no treatment decisions in the general population that depend on the result
• people with insulin resistance are usually identifiable from weight, waist size, triglycerides, HDL, blood pressure and glucose

Fasting insulin does have a place in research, in some specialist clinics, and in the specific situations described below.

HOMA-IR and other insulin-based indices

The homeostatic model assessment of insulin resistance (HOMA-IR) combines fasting glucose and fasting insulin:

HOMA-IR = fasting insulin (µIU/mL) × fasting glucose (mmol/L) ÷ 22.5

or, with glucose in mg/dL: insulin (µIU/mL) × glucose (mg/dL) ÷ 405.

The formula requires insulin in µIU/mL. If your insulin is reported in pmol/L, divide by 6 first. Entering 60 pmol/L directly instead of 10 µIU/mL would make HOMA-IR six times too high. Our HOMA-IR calculator accepts either unit and converts automatically.

Other indices include the QUICKI (quantitative insulin sensitivity check index) and the Matsuda index, which uses insulin and glucose values from an oral glucose tolerance test. All of these are research tools rather than diagnostic tests, and their cut-offs differ between populations.

When insulin testing is clinically useful

Low blood sugar (hypoglycaemia)

The most important clinical use of insulin measurement is in investigating hypoglycaemia in people without diabetes. When glucose is low – typically below 3.0 mmol/L (54 mg/dL) – insulin should be suppressed. The Endocrine Society guideline describes endogenous hyperinsulinism as an insulin of at least 3 µIU/mL (18 pmol/L) together with C-peptide of at least 0.6 ng/mL (0.2 nmol/L) and proinsulin of at least 5 pmol/L at the time of hypoglycaemia. This pattern suggests an insulinoma (a rare insulin-secreting tumour) or another cause of excess insulin production. These tests are usually performed during a supervised 72-hour fast.

Suspected surreptitious insulin use

If hypoglycaemia is caused by injected insulin, insulin will be high but C-peptide low, because C-peptide is released only from the body's own insulin production. This distinction is important in investigating unexplained hypoglycaemia.

Diabetes classification

C-peptide, rather than insulin, is the preferred test for estimating residual beta-cell function – for example, to help distinguish type 1 from type 2 diabetes several years after diagnosis – because insulin-treated patients have injected insulin in their blood and many insulin assays cross-react with insulin analogues.

Insulin and C-peptide

Insulin and C-peptide are released in equal amounts from proinsulin, but they behave differently:

• Insulin is largely cleared by the liver on its first pass; C-peptide is not, so C-peptide better reflects pancreatic secretion.
• C-peptide has a longer half-life (about 30 minutes versus about 5 minutes for insulin), so it fluctuates less.
• Injected insulin contains no C-peptide.

C-peptide is reported in ng/mL or nmol/L (1 ng/mL = 0.331 nmol/L, or 331 pmol/L) – a different conversion from insulin.

Insulin resistance in everyday terms

Insulin resistance is not a disease in itself but a feature of several common conditions. It is closely linked to:

• excess abdominal (visceral) fat and fat in the liver
• physical inactivity
• metabolic syndrome – a cluster of large waist size, high triglycerides, low HDL, high blood pressure and raised fasting glucose
• polycystic ovary syndrome
• non-alcoholic fatty liver disease (now often called metabolic dysfunction-associated steatotic liver disease, MASLD)
• family history of type 2 diabetes and certain ethnic backgrounds, including South Asian, Black African and Hispanic ancestry

Most of these can be identified without measuring insulin. The practical response is also largely the same regardless of the insulin value: regular physical activity (both aerobic and resistance exercise), weight loss where appropriate, better sleep, and a diet lower in refined carbohydrates and higher in fibre. In the Diabetes Prevention Program, modest weight loss of about 7% and 150 minutes a week of activity reduced progression to type 2 diabetes by 58% in people with prediabetes.

Insulin levels are not used to diagnose diabetes; A1C and glucose are. You can translate an A1C result into average glucose with our A1C calculator.

Insulin in polycystic ovary syndrome

Many women with PCOS have insulin resistance, regardless of body weight, and high insulin levels stimulate the ovaries to produce more androgens. The international PCOS guideline recognises insulin resistance as a key feature but does not recommend routine measurement of insulin or HOMA-IR in clinical practice, because the tests are poorly standardised and do not change management. Instead, women with PCOS are advised to have glucose status assessed (preferably with an oral glucose tolerance test in higher-risk women) and lipids checked. Metformin is used in some women with PCOS, particularly alongside lifestyle measures.

Insulin in pregnancy

Insulin resistance increases naturally during pregnancy, especially after about 20 weeks, driven by placental hormones. This helps ensure a steady glucose supply for the baby. In women whose pancreas cannot increase insulin output enough, gestational diabetes develops. Gestational diabetes is diagnosed with glucose testing, not insulin levels, and insulin treatment, when needed, is adjusted using glucose readings.

Insulin during an oral glucose tolerance test

Some clinics measure insulin at several points during a glucose tolerance test, producing an "insulin curve". Typical peak values after a 75 g glucose drink are commonly between about 30 and 150 µIU/mL (180–900 pmol/L), with wide variation between healthy people. Patterns of delayed or exaggerated insulin response have been proposed as early markers of insulin resistance, but there are no internationally agreed reference values, and this test is not recommended for routine diagnosis by major diabetes organisations.

Comparing insulin results across laboratories

Insulin assays from different manufacturers can differ by 20% or more, a problem highlighted by international standardisation efforts. A fasting insulin of 14 µIU/mL in one laboratory and 72 pmol/L (12 µIU/mL) in another may simply reflect assay differences rather than a real change. For tracking, use the same laboratory and the same fasting conditions, and focus on trends together with glucose, HbA1c, waist size and lipids.

What affects insulin results

• Fasting status – insulin rises within minutes of eating. Fasting insulin must be taken after at least 8–12 hours without food.
• Obesity and physical inactivity raise fasting insulin; weight loss and exercise lower it.
• Medicines – corticosteroids, some antipsychotics and certain HIV medicines increase insulin resistance; metformin, GLP-1 receptor agonists and pioglitazone lower insulin resistance.
• Insulin treatment – injected insulin and analogues appear in the result to varying degrees depending on the assay.
• Haemolysis – red cells contain an enzyme that degrades insulin, so a haemolysed sample can give a falsely low result.
• Insulin antibodies – in people treated with insulin or with rare autoimmune insulin syndrome, antibodies can interfere.
• Pregnancy – insulin resistance rises in the second and third trimesters, increasing fasting insulin.
• Sleep loss and acute illness temporarily increase insulin resistance.

Insulin as a medicine: units are different again

People who inject insulin for diabetes use doses in units (U), such as "10 units of glargine". These dose units are the same international units used in the blood test, but the numbers are not comparable: a dose of 10 units is 10,000,000 µIU spread through the whole body and absorbed over hours. Blood insulin tests are not used to adjust insulin doses; glucose readings, continuous glucose monitoring and HbA1c are used instead. Concentrated insulins (such as U-200, U-300 and U-500) contain more units per milliliter than standard U-100 insulin, which affects how many milliliters are injected but not the meaning of a "unit".

When to see a doctor

See a doctor if you have symptoms of low blood sugar – shakiness, sweating, palpitations, confusion or unusual behaviour that improves with eating – especially if you do not have diabetes. Seek urgent care for severe confusion, seizures or loss of consciousness. If a fasting insulin test suggests insulin resistance, discuss it with a clinician who can assess glucose, HbA1c, lipids, blood pressure and liver health, which are more useful for guiding prevention than insulin alone.