Gestational Diabetes OGTT Interpreter
Gestational Diabetes OGTT Interpreter
Classify a pregnancy glucose tolerance test on the WHO 2013 / IADPSG one-step 75 g criteria, on NICE, or on the two-step 100 g Carpenter-Coustan pathway still standard across much of the United States. Two things this page will not do: blend the pathways, or report a fasting glucose of 7.4 mmol/L as gestational diabetes when it is diabetes in pregnancy.
Pregnancy OGTT classification
Pathway + values → classificationFasting 5.2 mmol/L, 1-hour 9.4, 2-hour 7.6, one-step WHO 2013 / IADPSG criteria
Three pathways, three sets of thresholds
One-step 75 g — NICE NG3. Any ONE of: fasting ≥ 5.6 mmol/L · 2-hour ≥ 7.8 mmol/L. No 1-hour sample
Two-step 100 g — Carpenter-Coustan. TWO OR MORE of: fasting 95 mg/dL (5.3) · 1-hour 180 (10.0) · 2-hour 155 (8.6) · 3-hour 140 (7.8)
Above gestational diabetes. Fasting ≥ 7.0 mmol/L (126 mg/dL) or 2-hour ≥ 11.1 (200) after 75 g = diabetes mellitus in pregnancy
- the ranges are ranges
- WHO’s fasting criterion is 5.1 to 6.9 and its 2-hour criterion is 8.5 to 11.0 — both have an upper bound. A value above the upper bound is not “very abnormal gestational diabetes”, it is a different diagnosis, and this interpreter reports it as one
- one value or two
- the one-step pathways diagnose on a single abnormal value; Carpenter-Coustan requires two. This is the largest single reason the pathways disagree about who has the condition, and it is by design rather than by accident
- the 50 g screen
- the two-step pathway begins with a non-fasting 50 g glucose load and a 1-hour measurement. The ADA records thresholds of 130, 135 and 140 mg/dL all in use, so the two-step pathway’s sensitivity varies between units before the 100 g test is even reached
- why the loads cannot be mixed
- a 100 g load produces a higher and later glucose peak than a 75 g load. Carpenter-Coustan’s 2-hour threshold of 8.6 mmol/L and WHO’s of 8.5 look almost identical and are not interchangeable, because they describe different tests
- timing
- 24 to 28 weeks for routine testing. Earlier testing is offered where there is previous gestational diabetes — NICE says as soon as possible after booking — with a repeat at 24 to 28 weeks if the early test is normal
Worked example
Fasting 5.2 mmol/L, 1-hour 9.4, 2-hour 7.6, one-step WHO 2013 / IADPSG criteria
Fasting 5.2 is at or above the WHO/IADPSG fasting threshold of 5.1 mmol/L (92 mg/dL)
1-hour 9.4 is below 10.0 and 2-hour 7.6 is below 8.5, so neither of those is abnormal
One value is enough on the one-step pathway, so this is gestational diabetes mellitus
Now switch to NICE. NICE needs a fasting value of 5.6 or above, and 5.2 is below it
NICE has no 1-hour criterion, and its 2-hour threshold is 7.8, which 7.6 does not reach
The same woman is therefore not diagnosed under NICE — a real disagreement between guidelines, not a rounding error
And note what this result is not: the fasting value is well below 7.0 mmol/L, so this is gestational diabetes and not diabetes mellitus in pregnancy
The three pathways compared
| WHO 2013 / IADPSG (75 g) | NICE NG3 (75 g) | Carpenter-Coustan (100 g) | |
|---|---|---|---|
| Fasting | 5.1 mmol/L (92 mg/dL) | 5.6 mmol/L | 95 mg/dL (5.3 mmol/L) |
| 1 hour | 10.0 mmol/L (180 mg/dL) | not measured | 180 mg/dL (10.0 mmol/L) |
| 2 hours | 8.5 mmol/L (153 mg/dL) | 7.8 mmol/L | 155 mg/dL (8.6 mmol/L) |
| 3 hours | not measured | not measured | 140 mg/dL (7.8 mmol/L) |
| Values needed | Any one | Any one | Two or more |
| Preceded by a screen? | No | No | Yes — 50 g load, threshold 130, 135 or 140 mg/dL |
| Where it is standard | Much of the world; WHO-recommended | United Kingdom | Much of the United States |
Gestational diabetes or diabetes in pregnancy?
| Result after 75 g | Classification | Why it matters |
|---|---|---|
| Fasting 5.4 mmol/L (97 mg/dL) | Gestational diabetes | Usually manageable with diet and monitoring first; commonly resolves after delivery |
| Fasting 6.8 mmol/L (123 mg/dL) | Gestational diabetes — top of the WHO range | Still gestational diabetes, but insulin is likely to be needed early |
| Fasting 7.4 mmol/L (133 mg/dL) | Diabetes mellitus in pregnancy | A different diagnosis. Implies pre-existing glucose intolerance, higher anomaly risk, persistence after delivery, and a need for retinal and renal assessment |
| 2-hour 10.4 mmol/L (187 mg/dL) | Gestational diabetes | Inside the 8.5 to 11.0 range |
| 2-hour 12.2 mmol/L (220 mg/dL) | Diabetes mellitus in pregnancy | Above 11.1 mmol/L — meets the ordinary diabetes criterion |
Glucose targets once the diagnosis is made
| Guideline | Fasting | 1 hour after a meal | 2 hours after a meal |
|---|---|---|---|
| NICE NG3 (capillary plasma) | below 5.3 mmol/L | below 7.8 mmol/L | below 6.4 mmol/L |
Which criteria, and gestational diabetes or diabetes in pregnancy
There is no single international standard for diagnosing gestational diabetes, and pretending otherwise is the main way pregnancy glucose results get misread. Three frameworks are in routine use. The WHO 2013 criteria, which adopt the IADPSG thresholds derived from the HAPO study, diagnose on a single abnormal value after a 75 g load: fasting 5.1 to 6.9 mmol/L, 1-hour 10.0 or above, or 2-hour 8.5 to 11.0. NICE, in the United Kingdom, also uses a 75 g load and a single value, but sets the fasting bar higher at 5.6 mmol/L, sets the 2-hour bar lower at 7.8, and takes no 1-hour sample at all. And across much of the United States the two-step pathway persists: a non-fasting 50 g screen, then a 100 g three-hour test read against the Carpenter-Coustan thresholds, with two or more abnormal values required. The same woman can be diagnosed by one of these and cleared by another, and this page will show you that rather than average the three into a number that belongs to none of them.
The 75 g and 100 g tests are different tests, not two conventions for one test. A 100 g load produces a higher, later glucose peak, which is why Carpenter-Coustan’s 2-hour threshold of 155 mg/dL sits above WHO’s 153 mg/dL despite the test being less sensitive overall. Reading a value from one load against the other’s thresholds is not a conservative approximation; it is a category error. The requirement for two abnormal values rather than one moves more women between diagnosed and undiagnosed than any of the numeric thresholds do, and it is the deliberate design choice that keeps the two-step pathway labelling fewer pregnancies.
The other thing that is regularly lost is that WHO’s gestational diabetes criteria are bounded ranges rather than open-ended thresholds. The fasting criterion is 5.1 to 6.9 mmol/L and the 2-hour criterion is 8.5 to 11.0. A fasting value of 7.4 mmol/L is not severe gestational diabetes — it is diabetes mellitus in pregnancy, which WHO 2013 defines separately by reference to the ordinary 2006 diabetes criteria. The difference is not semantic. Diabetes in pregnancy usually means glucose intolerance that predates conception, so the organogenesis window has already passed under hyperglycaemia and the congenital anomaly risk is higher; it needs retinal and renal assessment, almost always needs insulin, and does not disappear at delivery, so it requires formal reclassification rather than a routine postnatal glucose check. An interpreter that reported it as gestational diabetes would send a woman down the wrong pathway at the point where it matters most.
Whichever criteria are used, the pre-analytical rules are the same and are unforgiving. The sample must be venous plasma collected into a tube that inhibits glycolysis or separated promptly, because glucose in an uncentrifuged tube falls by roughly 5 to 7% an hour — enough, at these thresholds, to convert a diagnosis into a normal result. Timing runs from the first sip of the load and the woman should stay seated. Testing is normally at 24 to 28 weeks, earlier where there has been gestational diabetes before, with a repeat at 24 to 28 weeks if that early test is normal. For the non-pregnant version of the same test, see the OGTT result interpreter, and note that its thresholds are much higher — reading a pregnancy test against them would miss almost every case of gestational diabetes.
Frequently asked questions
What are the WHO criteria for gestational diabetes?
Any one of a fasting plasma glucose of 5.1 to 6.9 mmol/L (92 to 125 mg/dL), a 1-hour value of 10.0 mmol/L (180 mg/dL) or above, or a 2-hour value of 8.5 to 11.0 mmol/L (153 to 199 mg/dL) after a 75 g load. One abnormal value is sufficient; there is no requirement for two.
What is the difference between gestational diabetes and diabetes in pregnancy?
The thresholds. A fasting plasma glucose of 7.0 mmol/L (126 mg/dL) or above, or a 2-hour value of 11.1 mmol/L (200 mg/dL) or above after 75 g, is diabetes mellitus in pregnancy — usually pre-existing glucose intolerance, with higher congenital anomaly risk, a need for retinal and renal assessment, and persistence after delivery. WHO’s gestational diabetes criteria are bounded ranges that stop below those values.
Why do NICE and the international criteria disagree?
NICE sets its fasting threshold at 5.6 mmol/L against the WHO/IADPSG figure of 5.1, and takes no 1-hour sample, but sets its 2-hour threshold lower at 7.8 against 8.5. A woman with a fasting glucose of 5.3 mmol/L and normal post-load values has gestational diabetes internationally and not under NICE.
How many abnormal values does the two-step test need?
Two or more of the four Carpenter-Coustan thresholds on the 100 g load — fasting 95 mg/dL, 1-hour 180, 2-hour 155, 3-hour 140. A single abnormal value is not diagnostic on that pathway, although it is associated with higher birthweight and worse outcomes than a normal test, and the same woman would often be diagnosed on the one-step criteria.
Can a 75 g result be read against the Carpenter-Coustan thresholds?
No. The 100 g load produces a higher and later glucose peak, so the two sets of thresholds describe different tests. Blending them either over- or under-diagnoses depending on which value is used, and the direction of the error is not predictable.
When should the test be done?
Routinely at 24 to 28 weeks. Where there has been gestational diabetes in a previous pregnancy, NICE recommends testing as soon as possible after booking, with a repeat at 24 to 28 weeks if the early result is normal. Other risk factors — BMI above 30, a previous baby of 4.5 kg or more, a first-degree family history, or a high-prevalence ethnic background — prompt testing at 24 to 28 weeks.
Related calculators
References
- World Health Organization. Diagnostic criteria and classification of hyperglycaemia first detected in pregnancy. WHO/NMH/MND/13.2. Geneva: WHO; 2013.
- National Institute for Health and Care Excellence. Diabetes in pregnancy: management from preconception to the postnatal period. NICE guideline NG3.
- American Diabetes Association Professional Practice Committee. 2. Diagnosis and Classification of Diabetes: Standards of Care in Diabetes—2025. Diabetes Care. 2025;48(Suppl 1):S27–S49.
- HAPO Study Cooperative Research Group. Hyperglycemia and adverse pregnancy outcomes. N Engl J Med. 2008;358(19):1991–2002.
Medical Disclaimer: The tools and content provided here are for educational and reference purposes only. They are not intended to substitute for professional medical advice, diagnosis, or treatment. Clinical decisions should always be based on the comprehensive assessment of a qualified healthcare professional.