For over 50 years, the diagnosis of gestational diabetes mellitus (GDM) has been based upon an oral glucose tolerance test at 24-28 weeks' gestation. This is the time during pregnancy when insulin resistance is increasing and hyperglycemia develops among those with insufficient insulin secretory capacity to maintain euglycemia. The Hyperglycemia and Adverse Pregnancy Outcomes (HAPO) study and the two major randomized controlled trials of treating GDM are based upon recruitment of women at this time during pregnancy. Meanwhile, the increasing prevalence of type 2 diabetes in pregnancy, with its significant risk of adverse pregnancy outcomes, has led to a need to identify undiagnosed diabetes as near to conception as possible. Screening for undiagnosed diabetes early in pregnancy also identifies women with hyperglycemia less than overt diabetes, yet at increased risk of adverse pregnancy outcomes. Such women are more insulin resistant-with higher blood pressure, triglycerides, perinatal mortality, and neonatal hypoglycemia with a greater need for insulin treatment-than those with GDM diagnosed at 24-28 weeks' gestation. Currently, there is uncertainty over how to diagnose GDM early in pregnancy and the benefits and harms from using the current management regimen. Randomized controlled trials testing the criteria for, and treatment of, GDM early in pregnancy are urgently needed to address this existing equipoise. In the meantime, the importance of early or "prevalent GDM" (i.e., mild hyperglycemia present from early [before] pregnancy) warrants interim criteria and thresholds for medication, which may differ from those in use for GDM diagnosed at 24-28 weeks' gestation.