When we hear of an 11-pound baby being delivered, maternal diabetes comes to mind. This link between the mother's glucose levels and macrosomia provides the main rationale for treating gestational diabetes. Recently proposed criteria for gestational diabetes will result in the condition being diagnosed in nearly a fifth of all pregnancies.1 In a related article in CMAJ, Retnakaran and colleagues report that, among women who did not have gestational diabetes according to current diagnostic criteria, im paired glucose tolerance was not a significant independent pre dictor of having a large-for- gestational-age infant.2 Given that many of these women would be la belled as having gestational diabetes using the proposed diagnostic criteria,1 this should give us pause before accepting such changes to the criteria. Many studies have shown an association be - tween macrosomia and maternal diabetes. The Hyperglycemia and Adverse Pregnancy Outcome (HAPO) Study (n = 23 316) showed a continuous association between increased infant birth weight and maternal glucose levels below those diagnostic of diabetes.3 In 1952, Pedersen hypothesized that excess maternal glucose crosses the placenta and leads to fetal hyperglycemia, which in turn leads to hyperinsulinemia and excessive fetal growth.4 In large-for-gestational-age infants of women with diabetes, the excess weight is not simply from the glucose being turned into fat, but many internal organs are enlarged as well. Currently, the best explanation may be that an elevated fetal insulin level downregulates insulin growth factor (IGF)- binding protein-1, which allows more unbound IGF-1 to promote excess growth. The large HAPO study showed that the association between maternal glucose levels and birth weight is a continuum with no clear threshold.3 The International Association of Diabetes and Pregnancy Study Groups came to a consensus in choosing diagnostic thresholds of 5.1, 10.0 and 8.5 mmol/L for fasting, one-hour and two-hour plasma glucose concentrations, respectively, on a single 75-g oral glucose tolerance test.1 These thresholds are based on the glucose levels associated with a 1.75-fold increased risk of having a large-forgestational- age infant based on data from the HAPO study. The Canadian Diabetes Association's current guidelines for diagnosing gestational diabetes use higher thresholds for glucose levels that are virtually identical to those associated with a 2-fold increased risk of a large-for-gestationalage infant based on the HAPO data.6