In 1997, the American Diabetes Association (ADA) announced a new diagnostic criterion for diabetes and set the definition of gestational diabetes mellitus (GDM). Before 1991, GDM was defined as “a transient abnormality of glucose tolerance during pregnancy” (2–4). However, the 1997 definition of GDM by the ADA includes diabetes diagnosed during pregnancy. This definition ignores the added risks to the mother and to the fetus when the mother has undiagnosed type 2 diabetes. We propose reconsideration of the definition, which would separate diabetes and slight abnormal carbohydrate, so-called GDM, to provide a better model of care for type 2 diabetic pregnant women.
There are three problems concerning an undiagnosed type 2 diabetic woman that are not major issues in pregnant women who are first diagnosed with abnormal glucose tolerance in pregnancy that resolves after pregnancy. First, the entire pregnancy is associated with abnormal carbohydrate metabolism, not just the second half. The second problem is related to the rate of congenital malformations of newborns from these pregnant women. The third is concerned with undiagnosed diabetic retinopathy.
In our Japanese cohort, we observed the results of 75-g oral glucose tolerance tests (OGTTs) (Japan criteria: two or more values above fasting glucose >100 mg/dl, 1-h glucose >180 mg/dl, and 2-h glucose >150 mg/dl) for 1,416 pregnant women who had risk factors for GDM. We found the frequency of GDM in the first trimester is the highest (33/250 [13.2%]), followed by the second (32/417 [7.7%]) and third trimesters (37/749 [4.9%]). Similarly, the frequency of type 2 diabetes is the highest in the first trimester at 6.0%, with 2.6% in the second trimester and 1.3% in the third trimester. Thus, in women with positive OGTT, GDM accounts for 7.2% and type 2 diabetes diagnosed during pregnancy accounts for 2.5% of the total pregnant population. In other words, 35% of women with a positive OGTT have type 2 diabetes diagnosed for the first time in pregnancy.
In this cohort, the congenital malformation rate from GDM patients was 1.9% and was no different from the rate in the general Japanese population. In contrast, the congenital malformation rate in infants of type 2 diabetic mothers diagnosed during pregnancy was higher than that of children from pregestational diabetic mothers treated during pregnancy, 12.7 vs. 4%, respectively.
There were no GDM patients with retinopathy. However, the rate of background retinopathy was 12.7% and proliferative retinopathy was 4.2% in the type 2 diabetic women diagnosed for the first time during pregnancy.
Similar rates and complications were seen in a cohort of pregnant women in Santa Barbara, California, where a total of 49,861 pregnancies occurred in our Mexican-American population from 1997 to 2004. A total of 4,133 (8.3%) had a positive OGTT based on the ADA criteria (1). However, 40% of the GDM women had type 2 diabetes first diagnosed during pregnancy based on our criteria: acanthosis nicgrans, requiring insulin before the 12th week of gestation, because they failed to maintain goals with dietary intervention alone (6). Five percent of the type 2 women had retinopathy, and 7% had significant proteinuria at time of diagnosis.
O’Sullivan (2) defined GDM as “a transient abnormality of glucose tolerance during pregnancy.” We should return to this time-honored definition. If type 2 diabetes is first detected during pregnancy, then it should be named as such. Data presented here underscores that this is a worldwide problem. In preparation for the November 2005 Fifth International Gestational Diabetes Conference, it is timely that we reconsider our definition of GDM.