A practical physician-led guide to the gestational diabetes test: what you drink, when blood is checked, which numbers matter, and what happens next.
- 1-hour screening test uses 50 g of glucose, is usually done at 24–28 weeks, and usually does not require fasting.
- Abnormal 1-hour result is commonly ≥130, ≥135, or ≥140 mg/dL, depending on the clinic’s chosen cutoff.
- 3-hour glucose tolerance test uses 100 g of glucose after an 8–14 hour fast and checks fasting, 1-hour, 2-hour, and 3-hour glucose.
- Carpenter-Coustan diagnostic cutoffs are fasting ≥95, 1-hour ≥180, 2-hour ≥155, and 3-hour ≥140 mg/dL.
- Two abnormal values on the 3-hour test usually diagnose gestational diabetes in the two-step U.S. approach.
- One-step 75 g testing diagnoses gestational diabetes if fasting glucose is ≥92, 1-hour ≥180, or 2-hour ≥153 mg/dL.
- After diagnosis most clinics target fasting glucose <95 mg/dL, 1-hour post-meal <140 mg/dL, or 2-hour post-meal <120 mg/dL.
- Postpartum follow-up should include a 75 g glucose tolerance test at 4–12 weeks after delivery, then diabetes screening every 1–3 years.
What the glucose tolerance test in pregnancy actually checks
Glucose tolerance test pregnancy usually means a 1-hour, 50 g glucose screening test at 24–28 weeks; if it is high, you usually take a fasting 3-hour, 100 g diagnostic test. A 1-hour value at or above 130–140 mg/dL is an abnormal screen, while the 3-hour Carpenter-Coustan diagnostic cutoffs are fasting 95, 1-hour 180, 2-hour 155, and 3-hour 140 mg/dL.
The test asks a simple question: can your body move a measured glucose load from the bloodstream into cells quickly enough during pregnancy? Pregnancy hormones naturally increase insulin resistance, usually most strongly after 20 weeks, which is why a person with normal early pregnancy glucose can still develop gestational diabetes later.
I’m Thomas Klein, MD, and in clinical review I see the same surprise often: the 1-hour gestational diabetes test is not a diagnosis. It is a screen designed to catch more people than it ultimately labels, much like other prenatal lab checks are built to flag risk early rather than give a final answer.
Kantesti is an AI blood test interpretation platform that reads glucose results in context, including pregnancy timing, units, reference ranges, and whether a value came from a screening or diagnostic test. That distinction matters because a 142 mg/dL 1-hour screen and a 142 mg/dL 3-hour value do not mean the same thing clinically.
When pregnancy glucose testing is usually done
Most pregnant patients are screened for gestational diabetes between 24 and 28 weeks, because insulin resistance rises sharply in the late second trimester. Earlier testing is often used if someone has prior gestational diabetes, obesity, polycystic ovary syndrome, a strong family history of diabetes, or a previous baby over 4,000 g.
The 24–28 week window is not random. Placental hormones such as human placental lactogen and progesterone increase insulin resistance, and by around week 26 the pancreas must often produce 2–3 times the usual insulin response to keep glucose normal.
The American College of Obstetricians and Gynecologists supports screening all pregnant patients at 24–28 weeks, with earlier risk-based evaluation for people more likely to have undiagnosed type 2 diabetes or early gestational diabetes (ACOG Practice Bulletin No. 190, 2018). If you already had a high fasting glucose, HbA1c, or random glucose early in pregnancy, your clinician may not wait for the routine screening window.
Patients sometimes ask whether a normal 10-week glucose result means they can skip the later test. Usually, no. Early tests look for pre-existing diabetes; later screening looks for pregnancy-driven insulin resistance, so it sits alongside other time-sensitive pregnancy blood test red flags.
How to prepare for the 1 hour glucose test pregnancy visit
The 1 hour glucose test pregnancy visit usually does not require fasting, but your clinic’s instruction wins because local protocols differ. You drink 50 g of glucose, finish it within about 5 minutes, and have plasma glucose measured exactly 1 hour later.
Most patients can eat normally before the 1-hour screen, though I suggest avoiding a giant sweet drink or dessert immediately beforehand because it can push a borderline result over the cutoff. A balanced meal with protein, fibre, and slower carbohydrates 2–3 hours before the appointment is less likely to create noise.
Water is fine before the 1-hour test unless your clinic gives unusual instructions. If you are also having fasting labs the same morning, follow the fasting instruction for those labs; our water-before-fasting guide explains why water rarely disrupts glucose but dehydration can make some blood results look worse.
Do not walk laps around the clinic to “burn off” the drink. Muscle contraction can lower glucose uptake independently of insulin, and even 10–15 minutes of brisk walking during the waiting hour can make the test less reflective of your usual physiology.
How 1-hour screening results are interpreted
A 1-hour glucose screening result is usually considered abnormal at ≥130, ≥135, or ≥140 mg/dL, depending on the clinic. Lower cutoffs catch more gestational diabetes but create more false positives, while 140 mg/dL is more specific but can miss some cases.
A 1-hour screen of 141 mg/dL is not “mild diabetes”; it means your clinic will likely order the 3-hour diagnostic test. I have seen anxious patients change their diet overnight after a 138 mg/dL result, only to have a completely normal diagnostic test 5 days later.
Some practices use 130 mg/dL because it detects roughly 90% of gestational diabetes cases, while 140 mg/dL detects fewer cases but reduces the number of people sent for the longer test. The trade-off is deliberate: screening tests are built to be sensitive, not perfectly specific.
If your 1-hour result is very high, such as ≥200 mg/dL, many clinicians treat it as strongly suggestive of gestational diabetes, although policies vary. For context on isolated glucose elevations outside pregnancy, see our guide to high glucose without diabetes.
What happens during the 3 hour glucose tolerance test
The 3 hour glucose tolerance test is a fasting diagnostic test using 100 g of glucose and four timed blood measurements. You fast for 8–14 hours, have a fasting glucose drawn, drink the glucose solution, then have glucose checked at 1, 2, and 3 hours.
Timing starts after you finish the drink, which is usually expected within 5 minutes. If the 2-hour or 3-hour sample is drawn late by 15–20 minutes, the result may be lower than it would have been on time, especially as insulin response catches up.
Bring something quiet to do and plan to stay seated. Vomiting the glucose drink usually invalidates the test; many clinics reschedule rather than interpret a partial curve, and a chilled drink or straw sometimes helps nausea.
Results may come back the same day if the lab runs plasma glucose onsite, but some clinics batch samples. Our guide to same-day lab results explains why glucose is usually fast while specialised pregnancy tests may take longer.
Which 3-hour glucose numbers diagnose gestational diabetes
In the common U.S. two-step approach, gestational diabetes is usually diagnosed when two or more values meet or exceed diagnostic thresholds on the 100 g 3-hour test. Carpenter-Coustan cutoffs are lower than older National Diabetes Data Group cutoffs, so they diagnose more cases.
ACOG notes that either Carpenter-Coustan or National Diabetes Data Group criteria may be used, but many U.S. practices now favour Carpenter-Coustan because it identifies milder hyperglycaemia linked with pregnancy risk (ACOG Practice Bulletin No. 190, 2018). If your friend was diagnosed at a different number, the lab may simply be using a different standard.
One abnormal value is a grey zone. Many clinicians do not formally diagnose gestational diabetes after one abnormal value, but I have seen practices increase monitoring or repeat testing when the abnormal value is high, such as a fasting glucose of 104 mg/dL or a 1-hour value above 190 mg/dL.
The diagnostic labels overlap with broader diabetes testing, but pregnancy thresholds are deliberately lower because fetal glucose exposure matters even below non-pregnant diabetes cutoffs. For non-pregnancy criteria, our diabetes blood test guide separates diagnostic fasting glucose, HbA1c, and OGTT thresholds.
Why some countries use a 75 g two-hour pregnancy test
Some clinics skip the 1-hour screen and use a 75 g 2-hour oral glucose tolerance test as a one-step diagnostic test. In the IADPSG/ADA approach, gestational diabetes is diagnosed if fasting glucose is ≥92 mg/dL, 1-hour glucose is ≥180 mg/dL, or 2-hour glucose is ≥153 mg/dL.
This is where international advice gets messy. The one-step approach diagnoses more gestational diabetes than the two-step approach because just one abnormal value is enough, and the fasting threshold of 92 mg/dL is lower than the 95 mg/dL Carpenter-Coustan fasting threshold.
NICE in the UK uses a 75 g test too, but its diagnostic cutoffs differ: fasting plasma glucose ≥5.6 mmol/L or 2-hour glucose ≥7.8 mmol/L. A patient moving countries mid-pregnancy can therefore receive different labels from the same biology.
Unit conversion adds another layer: mg/dL divided by 18 gives mmol/L for glucose. If your report mixes units or looks different after a move, our guide to lab values in different units is often the fastest way to prevent a false alarm.
What happens after an abnormal pregnancy glucose result
After an abnormal 1-hour screen, the next step is usually a fasting 3-hour diagnostic test within 1–2 weeks. After a diagnostic gestational diabetes result, care usually shifts to home glucose monitoring, nutrition changes, walking after meals, and medication if targets are not met.
Most clinics ask for four daily checks at first: fasting and either 1 hour or 2 hours after each main meal. Common targets are fasting <95 mg/dL, 1-hour post-meal <140 mg/dL, or 2-hour post-meal <120 mg/dL, although individual plans vary.
Diet treatment is not “no carbs.” It is usually consistent carbohydrates spread across meals and snacks, often paired with protein and fibre; our high blood sugar food swaps explain why a small carb serving at breakfast may behave differently from the same grams at dinner.
If more than about 20–30% of values remain above target after 1–2 weeks, medication is often discussed. Insulin is commonly used because it does not cross the placenta in meaningful amounts, while metformin may be used in selected cases after shared decision-making.
When glucose tolerance results may not fit the clinical picture
Glucose tolerance results can be distorted by illness, vomiting, steroid medication, recent bariatric surgery, unusual carbohydrate restriction, or sample timing errors. A single number should be interpreted with the test conditions, gestational age, symptoms, and whether the laboratory processed glucose quickly.
Acute illness can raise glucose through cortisol and adrenaline, even in someone who usually has normal readings. A steroid injection for asthma or severe nausea treatment can push glucose up for 24–72 hours, so tell the clinic about recent medications before the test.
Prior bariatric surgery is a special case because a glucose drink can cause dumping symptoms and unusual glucose swings. Some obstetric teams use home glucose monitoring instead of a standard OGTT, especially after gastric bypass, because the curve can be hard to interpret safely.
HbA1c is not a reliable substitute for diagnosing gestational diabetes at 24–28 weeks because pregnancy changes red cell turnover and HbA1c reflects the previous 8–12 weeks rather than the meal spikes that affect fetal growth. Our HbA1c accuracy guide explains why a reassuring A1c can still miss post-meal hyperglycaemia.
Why mild pregnancy hyperglycaemia still matters
Mildly high pregnancy glucose matters because risk rises continuously, not only after a neat diagnostic line. The HAPO study found graded associations between maternal glucose and birth weight above the 90th percentile, cord C-peptide above the 90th percentile, and newborn body fat (Metzger et al., 2008).
The fetus does not “get diabetes,” but glucose crosses the placenta and stimulates fetal insulin production. Fetal insulin acts like a growth signal, which is one reason higher maternal glucose is associated with larger birth size and shoulder dystocia risk.
Gestational diabetes also clusters with pregnancy blood pressure risk. I pay close attention when a patient has rising fasting glucose plus blood pressure drifting toward 140/90 mmHg, because the combination can change obstetric monitoring and delivery planning; our pregnancy blood pressure guide covers those call thresholds.
None of this is about blame. In my experience, many patients who eat carefully and exercise still develop gestational diabetes because placental insulin resistance can overwhelm pancreatic reserve by the third trimester.
Food and activity rules before glucose tolerance testing
Before the 3-hour diagnostic test, most clinics advise at least 3 days of normal carbohydrate intake, often around 150 g or more per day, followed by an 8–14 hour overnight fast. Low-carb eating before the test can make the glucose load look worse than your usual metabolism.
A practical 150 g carbohydrate day might include oats or whole-grain toast at breakfast, fruit or yogurt, a rice or potato portion at lunch, and beans or whole grains at dinner. This is not a recommendation to overload sugar; it is a way to avoid testing a pancreas that has been temporarily downshifted by carbohydrate restriction.
Do normal activity the 3 days before the test, but avoid unusually intense exercise the day before if it is not part of your routine. A hard workout can alter muscle glucose uptake for 24–48 hours, and pregnancy testing is not the time for a metabolic experiment.
If you usually follow low-carb eating, tell your clinician rather than quietly changing everything. Our low-carb diet lab guide explains why glucose, ketones, triglycerides, and electrolytes can shift together when carbohydrate intake changes.
Postpartum testing after gestational diabetes
After gestational diabetes, a 75 g 2-hour glucose tolerance test at 4–12 weeks postpartum is the preferred follow-up test. Fasting glucose alone misses some impaired glucose tolerance, and HbA1c can be less reliable soon after delivery because blood loss and iron changes affect red cell turnover.
The ADA Standards of Care recommend postpartum testing at 4–12 weeks and lifelong screening at least every 1–3 years after gestational diabetes (American Diabetes Association Professional Practice Committee, 2026). Diabetes is diagnosed postpartum at fasting glucose ≥126 mg/dL or 2-hour glucose ≥200 mg/dL on a 75 g OGTT, using non-pregnant criteria.
Prediabetes postpartum includes fasting glucose 100–125 mg/dL or 2-hour glucose 140–199 mg/dL. In clinic, I treat those numbers as a long runway for prevention rather than a failure; lactation, sleep, weight trajectory, and family support all change what is realistic.
The long-term risk is substantial: roughly 30–70% of people with gestational diabetes have recurrence in a later pregnancy, and up to half develop type 2 diabetes within 10–20 years in some cohorts. Our guide to testing after gestational diabetes lays out which labs to track after the newborn phase.
How Kantesti helps organise pregnancy glucose results
Kantesti helps by separating screening results from diagnostic results, converting glucose units, and showing trends across pregnancy and postpartum labs. Kantesti is an AI blood test analyzer that can read uploaded PDF or photo lab reports and return structured interpretation in about 60 seconds.
Kantesti AI is not a replacement for your obstetric team, and it should not decide medication doses. Its value is pattern recognition: fasting glucose, timed OGTT values, HbA1c, ferritin, kidney markers, thyroid tests, and postpartum follow-up can be reviewed together rather than as disconnected screenshots.
Kantesti is an AI-powered blood test analysis tool used by 2M+ people across 127 countries, with privacy-focused, GDPR-aligned handling and support for 75+ languages. The underlying approach is described in our AI technology guide, and the clinical standards behind interpretation are outlined in our medical validation material.
When I review pregnancy labs as Thomas Klein, MD, I want the same thing our users want: fewer unexplained flags and clearer next questions for the clinician. Kantesti’s neural network maps glucose against thousands of related biomarkers, and our biomarker guide shows why context often beats a single red marker.
Research notes, safety limits, and when to call
Call your maternity unit urgently if you have repeated glucose readings above your care plan’s safety limits, reduced fetal movement, persistent vomiting, dehydration, severe headache, visual symptoms, or blood pressure at or above 140/90 mmHg. A glucose tolerance test result should guide care, not delay urgent assessment when symptoms are concerning.
As of June 9, 2026, the most clinically useful references for gestational diabetes remain guideline-based thresholds plus outcome data, especially ACOG’s two-step approach, ADA postpartum follow-up recommendations, and the HAPO study’s continuous-risk findings. Our broader women’s health guide keeps related pregnancy and hormone testing topics in one place.
Kantesti is an AI lab test interpretation service, but our medical team still treats pregnancy glucose as a clinician-managed condition because fetal growth scans, medication choices, and delivery timing require obstetric judgement. You can see the physician oversight model through the Medical Advisory Board.
Kantesti Ltd. (2026). Nipah Virus Blood Test: Early Detection & Diagnosis Guide 2026. Zenodo. DOI: 10.5281/zenodo.18487418. ResearchGate: publication search. Academia.edu: publication search. Kantesti Ltd. (2026). B Negative Blood Type, LDH Blood Test & Reticulocyte Count Guide. Figshare. DOI: 10.6084/m9.figshare.31333819. ResearchGate: publication search. Academia.edu: publication search.
Frequently Asked Questions
Do I need to fast for the 1 hour glucose test in pregnancy?
Most clinics do not require fasting for the 1-hour glucose test in pregnancy because it is a 50 g screening test, not a fasting diagnostic OGTT. You drink the glucose solution and have glucose measured 1 hour later. If your clinic tells you to fast, follow that local instruction because some practices combine the test with other fasting labs. Water is usually allowed unless your maternity team says otherwise.
What is a normal result for the 1 hour glucose test pregnancy screen?
A normal 1-hour pregnancy glucose screen is usually below the clinic’s cutoff, commonly <130, <135, or <140 mg/dL. The cutoff varies because lower thresholds detect more cases but create more false positives. A result of 140 mg/dL or higher usually leads to a 3-hour diagnostic glucose tolerance test. A very high result, such as ≥200 mg/dL, may be managed differently depending on local policy.
What are the 3 hour glucose tolerance test cutoffs?
The common Carpenter-Coustan cutoffs for the 100 g 3-hour glucose tolerance test are fasting ≥95 mg/dL, 1-hour ≥180 mg/dL, 2-hour ≥155 mg/dL, and 3-hour ≥140 mg/dL. Gestational diabetes is usually diagnosed when two or more values meet or exceed those thresholds. Some laboratories use older NDDG cutoffs: fasting ≥105, 1-hour ≥190, 2-hour ≥165, and 3-hour ≥145 mg/dL. Always compare your result with the criteria printed on your lab report.
Can I drink water during the 3 hour glucose tolerance test?
Most clinics allow plain water during the 3-hour glucose tolerance test, and staying mildly hydrated can make the visit easier. You should not eat, drink coffee, chew sugary gum, smoke, or exercise during the test because those can alter glucose handling. The usual fast before the test is 8–14 hours. If you vomit the glucose drink, the clinic usually reschedules or changes the testing plan.
Does one abnormal value on the 3 hour test mean gestational diabetes?
In the standard two-step U.S. approach, one abnormal value on the 3-hour 100 g test usually does not formally diagnose gestational diabetes; two or more abnormal values usually do. That said, one abnormal value still signals higher metabolic stress, especially if it is fasting glucose ≥95 mg/dL or a very high 1-hour value. Some clinicians recommend dietary counselling, repeat testing, or home glucose checks after one abnormal value. The best next step depends on the exact number, gestational age, and fetal growth pattern.
What happens if I fail the glucose tolerance test during pregnancy?
If you fail the diagnostic glucose tolerance test, your care team usually starts home glucose monitoring, nutrition counselling, and activity guidance. Common targets are fasting glucose <95 mg/dL, 1-hour after meals <140 mg/dL, or 2-hour after meals <120 mg/dL. If readings remain above target after about 1–2 weeks, medication such as insulin may be discussed. Many people achieve targets with food timing, carbohydrate distribution, and walking after meals.
Do I need diabetes testing after the baby is born?
Yes, after gestational diabetes you should usually have a 75 g 2-hour glucose tolerance test at 4–12 weeks postpartum. Fasting glucose alone can miss impaired glucose tolerance, and HbA1c can be less reliable soon after delivery. Postpartum diabetes is diagnosed at fasting glucose ≥126 mg/dL or 2-hour glucose ≥200 mg/dL. Long-term screening every 1–3 years is recommended because type 2 diabetes risk remains higher for years.
Get AI-Powered Blood Test Analysis Today
Join over 2 million users worldwide who trust Kantesti for instant, accurate lab test analysis. Upload your blood test results and receive comprehensive interpretation of 15,000+ biomarkers in seconds.
📚 Referenced Research Publications
Klein, T., Mitchell, S., & Weber, H. (2026). Nipah Virus Blood Test: Early Detection & Diagnosis Guide 2026. Kantesti AI Medical Research.
Klein, T., Mitchell, S., & Weber, H. (2026). B Negative Blood Type, LDH Blood Test & Reticulocyte Count Guide. Kantesti AI Medical Research.
📖 External Medical References
American Diabetes Association Professional Practice Committee (2026). Management of Diabetes in Pregnancy: Standards of Care in Diabetes—2026. Diabetes Care.
⚕️ Medical Disclaimer
This article is for educational purposes only and does not constitute medical advice. Always consult a qualified healthcare provider for diagnosis and treatment decisions.
E-E-A-T Trust Signals
Experience
Physician-led clinical review of lab interpretation workflows.
Expertise
Laboratory medicine focus on how biomarkers behave in clinical context.
Authoritativeness
Written by Dr. Thomas Klein with review by Dr. Sarah Mitchell and Prof. Dr. Hans Weber.
Trustworthiness
Evidence-based interpretation with clear follow-up pathways to reduce alarm.