Prevalence of antibodies to glutamic acid decarboxylase in women who have had gestational diabetes.

Objectives: Our purpose was to determine the prevalence of autoantibodies to glutamic acid decarboxylase in women who had had gestational diabetes, including those in whom insulin-requiring or non-insulin-requiring diabetes mellitus has since developed.

Study Design: The study group comprised 734 women with previous gestational diabetes who were consecutive attendees to a follow-up clinic. These women were tested for autoantibodies to glutamic acid decarboxylase with a radioimmunoprecipitation assay. We similarly tested 104 women in whom permanent diabetes mellitus developed after gestational diabetes, of whom 20 were using insulin and 84 were not. Those using insulin also had fasting C-peptide levels measured.

Results: Thirteen of the 734 (1.8%, 95% confidence interval 0.9% to 3.0%) women with previous gestational diabetes were positive for autoantibodies to glutamic acid decarboxylase. Of the 20 women with diabetes treated with insulin, 12 had insulin deficiency confirmed by low levels of C peptide; all 12 were positive for autoantibodies to glutamic acid decarboxylase. Of the 84 women with diabetes not requiring insulin, 6 (7.1%, 95% confidence interval 2.7% to 14.9%) were positive for autoantibodies to glutamic acid decarboxylase.

Conclusions: The prevalence of autoantibodies to glutamic acid decarboxylase in women with previous gestational diabetes was 1.8%. Our data also showed that insulin-dependent diabetes mellitus will develop in 1.7% of women with gestational diabetes. A positive test for autoantibodies to glutamic acid decarboxylase may help in the early identification of insulin-dependent diabetes mellitus. Adult-onset insulin-dependent diabetes mellitus developed in only 5.2% (12/230) of women with previous gestational diabetes who later had diabetes mellitus.