Current evidence-based clinical review on screening for glucose intolerance in pregnancy, standardization of diagnostic equations, glycemic targets, pharmacotherapy steps, fetal monitoring and birth timing.
⚕️ Target audience: This clinical review is intended for physicians and healthcare professionals; It is a summary of current literature and guidelines for educational purposes. Clinical decisions should be made based on patient-specific evaluation and applicable institutional protocols.
Gestational diabetes mellitus (GDM) is carbohydrate intolerance that occurs when pancreatic beta-cell function cannot accommodate the insulin resistance that accompanies pregnancy. Pregnancy is a physiological state in which both beta-cell function increases and insulin resistance increases under the influence of placental-derived diabetogenic hormones (especially human placental lactogen, growth hormone, corticotropin-releasing hormone, prolactin and progesterone); These changes are most evident in the third trimester (UpToDate, "GDM: Screening, diagnosis, and risk reduction", current).
There is an important distinction in terminology: The American Diabetes Association (ADA) defines GDM as diabetes diagnosed after the 15th week of pregnancy and not clearly present before conception; Hyperglycemia detected in early pregnancy (first visit/first trimester) is most likely considered previously undiagnosed type 2 diabetes (“overt diabetes”) and managed as pregestational diabetes (ADA Standards of Care 2026, §15). This distinction is clinically critical because the risk of congenital anomalies is not increased in GDM that develops after organogenesis is completed, whereas the risk of miscarriage and malformations is increased in early pregnancy hyperglycemia.
GDM affects approximately 7.8% of births in the United States by traditional Carpenter-Coustan criteria; Global prevalence is estimated at approximately 17% (North America 10%, Southeast Asia 25%) by IADPSG (one-step) criteria. Prevalence is increasing over time with increasing maternal age, rising obesity rate, and more sensitive screening methods (HAPO-based IADPSG data; Metzger et al., NEJM 2008, PMID 18463375).
From the middle of the second trimester (approximately week 24) insulin resistance increases markedly; Hyperglycemia occurs in pregnant women who do not have sufficient insulin secretory reserves. The risks of preeclampsia, large baby (LGA), shoulder dystocia, cesarean section, and neonatal hypoglycemia increase with GDM; The association between hyperglycemia and adverse pregnancy outcomes is continuous, not threshold (HAPO study, NEJM 2008). This continuity also explains why diagnostic thresholds remain controversial.
The group at increased risk of GDM includes: GDM in a previous pregnancy (approximately 40% risk of recurrence), diabetes in a first-degree relative, prepregnancy BMI ≥30 kg/m², advanced maternal age (≥35), PCOS, previous baby ≥4000 g, A1C ≥5.7%, and high-risk ethnicity. Because approximately 90% of pregnant women in the United States have at least one risk factor and approximately 20% of pregnant women with GDM have no risk factor, universal screening is recommended rather than selective screening (UpToDate, current).
There is no single universally accepted standard for GDM screening; Practitioners generally follow their national organisation's guidelines. Screening for overt diabetes in early pregnancy (e.g., A1C or targeted fasting glucose/75g test at first visit) is recommended due to the increasing prevalence of type 2 diabetes: A1C ≥6.5% (≥48 mmol/mol) is in favor of overt diabetes and the patient is managed as for pregestational diabetes. ADA 2026 stated that in cases with fasting plasma glucose predominantly ≥110 mg/dL in early pregnancy (≤15 weeks), the treatment option can be evaluated within the framework of shared decision-making (ADA Standards of Care 2026, §15).
GDM screening is performed between 24–28 weeks of pregnancy; This is the period when insulin resistance increases significantly. Two basic approaches are used:
This two-step approach is the most widely used approach in the United States and recommended by ACOG (ACOG Practice Bulletin No. 190, 2018). The first step is a 50g 1-hour oral glucose challenge test (screening test) that does not require fasting. Pregnant women who exceed the threshold are referred to the second step, 100g 3-hour OGTT. The UpToDate author prefers a threshold of ≥135 mg/dL for the high-risk population; Thresholds of ≥130 mg/dL (more sensitive) and ≥140 mg/dL are also commonly used. A preliminary diagnosis of GDM can be made directly in cases with a 50g result of ≥200 mg/dL. Diagnosis is made when two or more values meet/exceed the threshold in the 100g test.
This single-step approach is the widely adopted approach internationally, preferred by the IADPSG; The ADA and ACOG consider the one-step method an acceptable alternative. A 75g 2-hour OGTT is performed during fasting and a single value meeting/exceeding the threshold is sufficient for the diagnosis of GDM. Thresholds are based directly on outcome data from the HAPO study (odds ratio of 1.75).
In a 2022 meta-analysis comparing the two approaches, the one-step method approximately doubled the diagnosis of GDM and the need for pharmacotherapy (16.3% versus 8.3%; and 7.1% versus 3.8%, respectively) but did not provide significant benefit in outcomes such as LGA, preeclampsia, cesarean section, and shoulder dystocia; neonatal hypoglycemia was more common in the one-step group. The optimal approach therefore remains controversial (UpToDate, current; conflicting evidence — disclose ).
| Test | Hunger | 1st hour | 2nd hour | 3rd hour | Diagnostic rule |
|---|---|---|---|---|---|
| 50g scan (1 hour) | — | ≥130 / 135 / 140 | — | — | If positive, switch to 100g (≥200: GDM preliminary diagnosis) |
| 75g (IADPSG) | ≥92 | ≥180 | ≥153 | — | ≥1 value is diagnostic |
| 100g (Carpenter-Coustan) | ≥95 | ≥180 | ≥155 | ≥140 | ≥2 values are diagnostic |
| 100g (NDDG) | ≥105 | ≥190 | ≥165 | ≥145 | ≥2 values are diagnostic |
Carpenter-Coustan detects approximately 22% more cases because its thresholds are lower than NDDG; Both sets continue to be used in the United States. The UpToDate writer prefers Carpenter-Coustan because it is more stringent/lower and captures the highest risk cases. The OGTT Evaluation tool (50g/75g/100g, ACOG PB 190 compliant) on our site can be used to quickly interpret your results. (OGTT Assessment tool)
The first step after diagnosis is medical nutrition therapy (MNT). Up to 85% of cases diagnosed with Carpenter-Coustan can reach target glucose levels with lifestyle changes alone. Calorie distribution is typically planned as three main meals and two to four snacks per day; The UpToDate approach is roughly a 40% carbohydrate, 20% protein, 40% fat (saturated fat <7%) distribution. The minimum daily carbohydrate intake for all pregnant women should be 175 g and fiber 28 g; A high-protein snack before bed is recommended to prevent night fasting ketosis (UpToDate, "GDM: Glucose management", current).
As with adults with diabetes, 30-60 minutes of moderate-intensity aerobic exercise (at least 150 minutes per week) is recommended most days of the week unless there are medical/obstetric contraindications. Exercise reduces both fasting and postprandial glucose by increasing tissue insulin sensitivity and may eliminate the need for insulin in some cases.
The patient should measure his or her glucose at least four times a day: before breakfast (fasting) and 1 or 2 hours after the start of each meal. UpToDate prefers 1-hour postprandial measurement as it better corresponds to the peak effect of rapid-acting analogue insulin.
The ADA and ACOG recommend initiating pharmacotherapy when the following upper limits are exceeded (these targets are considered extrapolated from pregestational diabetes recommendations):
| Time | Target upper limit |
|---|---|
| Fasting and preprandial | <95 mg/dL (5.3 mmol/L) |
| 1st hour postprandial | <140 mg/dL (7.8 mmol/L) |
| 2nd hour postprandial | <120 mg/dL (6.7 mmol/L) |
Overly tight control (mean glucose ≤86 mg/dL) provides no additional benefit and increases the risk of iatrogenic growth restriction; Therefore, the goal is to keep most of the values below the upper limit without creating hypoglycemia.
If more than 30% of glucose values are above target within a week despite MNT and exercise, pharmacotherapy is started. Additionally, even if maternal glucose is normal, insulin initiation is reasonable in the presence of evidence of fetal overgrowth (abdominal circumference >75th percentile or estimated fetal weight ≥90th percentile on early third trimester ultrasound) as indirect evidence of fetal hyperinsulinism (UpToDate Grade 2C).
Insulin is the first-line treatment. UpToDate prefers insulin because it is effective, easy to titrate to glucose, safe for the fetus, and does not cross the placenta (Grade 2C). Oral antihyperglycemic medications (metformin and glyburide) are a reasonable alternative for patients who refuse or are noncompliant with insulin; however, both cross the placenta and there is insufficient data on the long-term consequences of in utero exposure.
Total dose varies individually; Most studies report a range of 0.7-2 units/kg based on current gestational weight. UpToDate's practical approach is to start with the simplest regimen and complicate it as needed: typically 10-20 units of intermediate-acting basal insulin (NPH) and 6-10 units of rapid-acting insulin (lispro or aspart) in the morning just before breakfast. The peak of NPH 4-6 hours after the dose may also cover the noon postprandial rise. Only if fasting glucose is high, 0.2 units/kg of basal insulin is added at bedtime. Those with proven safety and efficacy in pregnancy are human NPH and the lispro and aspart analogues; glargine is also considered safe with available data.
Metformin: Extended-release (XR) 500 mg once daily (with dinner); If tolerated, increase by 500-1000 mg per week to reach an effective dose of 1500-2000 mg per day (divided into two doses; maximum 2500 mg). Fetal metformin level reaches approximately 200% of the maternal level. Metformin does not cause hypoglycemia; The most common side effects are gastrointestinal. While the ADA recommends avoiding metformin in patients with hypertension or preeclampsia who are at risk for growth restriction based on placental insufficiency, ACOG and SMFM do not share this warning (difference between guidelines—disclose).
Glyburide: Start at 2.5-5 mg once daily, increasing as needed to a maximum of 20 mg daily (two doses are usually required). The fetal glyburide level is approximately 70% of the maternal level. In meta-analyses, glyburide was associated with higher birth weight, more macrosomia (OR 1.38), and more frequent neonatal hypoglycemia compared to insulin; Maternal hypoglycemia is also significantly higher with glyburide than with insulin (28.8% vs 3.5%). The combined use of two oral agents (metformin + glyburide) is not recommended during pregnancy. If the target is not achieved with the oral agent, supplemental insulin is added (Cochrane systematic review overview, PMID 30103263; UpToDate, current).
The intensity of management and monitoring is determined according to the White classification: A1 GDM is euglycemic with diet alone, A2 GDM is a case that requires pharmacotherapy. A single growth ultrasound at 36–39 weeks is recommended to screen for macrosomia in all GDM patients (UpToDate, "GDM: Obstetric issues and management", current).
| Category | Antenatal follow-up | Time of birth |
|---|---|---|
| A1 GDM (diet euglycemic) | No increased risk of stillbirth; routine NST/BPP not required (≥week 36 if selected) | 39+0 to 41+0 weeks |
| A2 GDM or A1 suboptimal control | NST + amniotic fluid index twice a week starting from the 32nd week | 39+0 weeks (earlier with clinical indication in comorbidity/suboptimal control) |
ACOG states that planned delivery should not occur before 39+0 weeks if there is no indication, and expectant management until 40+6 weeks with antepartum testing is generally appropriate. Early term delivery (37+0–38+6) is not indicated in well-controlled, uncomplicated A2 GDM; However, if there is an additional condition such as hypertension or suboptimal control, delivery is planned earlier according to clinical indication. Antenatal corticosteroids are generally not recommended in pregnant women with diabetes at ≥34 weeks (the ALPS study excluded those with diabetes).
The hyperglycemic effect of placental hormones disappears rapidly after birth; Most patients return to prepregnancy glycemic status in a short time, and most of those using oral agents/insulin antepartum do not require them postpartum. However, people with GDM are at high risk for type 2 diabetes: in a meta-analysis, the risk of type 2 diabetes in people with GDM was found to be approximately 10-fold increased (RR 9.51).
Therefore, all GDM patients should be re-screened with a 75g OGTT 4-12 weeks after birth, and if the result is normal, they should be monitored at least every three years (ADA recommendation). Pre-discharge testing at the birth hospital is a reasonable alternative as it increases compliance to nearly 100% (ACOG considers this acceptable). It should be emphasized in counseling that breastfeeding may reduce the long-term risk of type 2 diabetes; Lifestyle intervention and/or metformin can reduce the incidence of diabetes by approximately 50%.
The notes below cover only the topics that have changed or become clear in the 2024-2026 period, in addition to the principles of screening/diagnostic thresholds, glycemic targets, pharmacotherapy and postpartum follow-up in the body of the guideline.
ACOG, 2018 Practice Bulletin No. Rather than replacing 190 with the new number, it issued a focused update that reformulated screening and postpartum monitoring: ACOG Clinical Practice Update — Screening for Gestational and Pregestational Diabetes in Pregnancy and Postpartum (Obstet Gynecol 2024;144:e20-e23, July 2024). PB 190 is in force as the basic document; However, this update is now guiding for early pregnancy and postpartum screening and is more aligned with the ADA criteria.
Early pregnancy (<15 weeks) abnormal glucose metabolism—threshold expanded: The ADA does not define abnormal glucose metabolism in early pregnancy by fasting glucose alone; A1C 5.9-6.4% (41-47 mmol/mol) or fasting plasma glucose 110-125 mg/dL (6.1-6.9 mmol/L) indicates this group and is at high risk for adverse pregnancy/neonatal outcome (ADA Standards of Care 2026, Chapter 15). The A1C 5.9-6.4% arm should be added to the guideline's only "FPG ≥110" threshold.
The statement "Early treatment before 20 weeks did not provide significant benefit" in the "controversial points" section of the guideline is outdated. In the TOBOGM multicenter randomized controlled trial (802 women, with risk factors, diagnosis at <20 weeks by WHO-2013 criteria), early/immediate treatment reduced the composite adverse neonatal outcome from 30.5% to 24.9% (adjusted risk difference −5.6 points; 95% CI −10.1 to −1.2; approximately NNT 18). No difference was seen in pregnancy-associated hypertension and neonatal fat-free mass, with the benefit being particularly evident in subgroups with a higher glycemic range (fasting 95-109 mg/dL) and diagnosed at <14 weeks (TOBOGM; N Engl J Med 2023;388:2132-2144; PMID 37144983, According to PubMed — DOI ). In secondary analyses, early treatment approximately halved the incidence of neonatal respiratory distress (aOR 0.50; 95% CI 0.31-0.79; BJOG 2024;132:1087-1095; PMID 39157877, DOI), while in economic evaluation early diagnosis-treatment dominated—more effective and tended to be cheaper (−$1373; high glycemic net savings in the sub-group) were found (EClinicalMedicine 2024;71:102610; PMID 38813447, DOI). (DOI) (DOI) (DOI)
ADA 2026 reiterated the recommendation against first-line use of metformin and glyburide (Recommendation 15.21, Grade A) and cited increased lean mass/abdominal fat monitoring (MiG TOFU) at 5-9 years of age in metformin-exposed children; Metformin used for ovulation induction in PCOS should be discontinued at the end of the first trimester (Recommendation 15.22, Grade A). In contrast, in children aged 5-11 years, 2024 systematic reviews found no significant differences in height/weight/BMI z-scores at age 5 and DXA/MRI fat measurements at age 9; so the long-term effect is still unclear.
This is where the guidelines diverge. ACOG and the ADA consider insulin first-line and position metformin only as an alternative in patients who refuse insulin, cannot safely administer it, or cannot afford it. The SMFM describes metformin as “a reasonable and safe first-line alternative to insulin — with the recognition that approximately half of women will still require insulin” and reaffirmed this stance in 2024 (SMFM Statement, 2018; reaffirmed 2024). A 2024 network meta-analysis (38 RCTs, 6046 women) provides current evidence supporting the SMFM stance, associating metformin with less maternal weight gain (WMD −2.39 kg), less maternal hypoglycemia (OR 0.34), and less LGA (OR 0.61) compared to insulin.
ADA 2026 strongly recommends CGM in type 1 diabetes (Grade A); In other types of gestational diabetes, including GDM, it remains at the "maybe useful" level (Grade E) because GDM-specific evidence is insufficient and one RCT showed no difference in adverse perinatal outcomes between CGM and finger-stick measurement (ADA 2026, Recommendation 15.10). The guideline's statement that "CGM is unproven" is correct; however, this distinction of ADA (type 1 ≠ GDM) should be added.
USPSTF maintains screening at ≥24 weeks (Grade B) and insufficient evidence at <24 weeks (I statement) and no new GDM recommendations have been issued in 2024-2026 (USPSTF; JAMA 2021;326:531; effective as of January 2025).
Items to update in the guideline: (1) add the July 2024 Clinical Practice Update alongside "ACOG PB 190 (2018)" in the references; (2) revise "early treatment provided no benefit" using TOBOGM data (RD −5.6%, NNT≈18, pronounced at <14 weeks / in the higher glycemic range, no benefit for gestational hypertension); (3) add an A1C 5.9–6.4% category to the early screening threshold; (4) separately note SMFM's "reasonable first-line alternative" position on metformin alongside ACOG/ADA; (5) specify the GDM vs. type 1 diabetes distinction for CGM.
Last updated: . Guide numbers and publication years have been verified against current catalogues.