• Universal screening for Gestational Diabetes Mellitus (GDM) between 24–28 weeks of gestation.
• Early screening in the first trimester for high-risk patients (BMI > 30, history of GDM, strong family history) to identify overt pre-existing T2DM.

• Administer a 75g Oral Glucose Tolerance Test (OGTT) while fasting.
• Diagnosis is made if AT LEAST ONE value is elevated:
• Fasting ≥ 92 mg/dL (5.1 mmol/L)
• 1-Hour ≥ 180 mg/dL (10.0 mmol/L)
• 2-Hour ≥ 153 mg/dL (8.5 mmol/L)

• Step 1: Non-fasting 50g Glucose Challenge Test (GLT). If 1-hour value is ≥ 130, 135, or 140 mg/dL (institution dependent), proceed to Step 2.
• Step 2: Fasting 100g 3-Hour OGTT (Carpenter-Coustan criteria). Diagnosis requires ≥ 2 elevated values:
• Fasting ≥ 95 mg/dL
• 1-Hour ≥ 180 mg/dL
• 2-Hour ≥ 155 mg/dL
• 3-Hour ≥ 140 mg/dL

• The 75g One-Step criteria drastically increases GDM prevalence to ~18% of all pregnancies.
• ACOG argues this heavily medicalizes borderline pregnancies, increasing anxiety and healthcare costs WITHOUT strongly proven reductions in clear morbidity (e.g., shoulder dystocia, NICU admissions).
• If a patient vomits the 100g glucose load during the 3-hour test (very common), the test is invalid. Alternative strategies like fasting blood sugar + random postprandial tracking or immediate progression to a 75g test may be necessary depending on institutional policy.

• Initiate Medical Nutrition Therapy (MNT): Caloric restriction tailored to BMI, substituting high glycemic index carbs with complex carbs/protein.
• Prescribe glucometer: Track fasting (goal < 95 mg/dL) and either 1-hr postprandial (< 140 mg/dL) or 2-hr postprandial (< 120 mg/dL).
• If >20% of values over 1-2 weeks cross thresholds: Initiate pharmacotherapy.
• Insulin is the absolute gold standard because it does not cross the placenta. Metformin is used extensively but DOES cross the placenta (long-term metabolic effects on offspring are under intense study).

The original 100g 3-hour test thresholds were established in 1964 by O'Sullivan and Mahan, based solely on predicting which mothers would develop Type 2 Diabetes later in life, entirely ignoring fetal outcomes. Carpenter and Coustan later mathematically lowered these thresholds in 1982 to match modern plasma-based glucose testing.