Background and Rationale: Gestational diabetes mellitus (GDM) complicates 3-7% of pregnancies in the US and increases affected women’s risk of developing type 2 diabetes. The prevalence of GDM is increasing in tandem to rising rates of type 2 diabetes and obesity. Since unintended pregnancy in women with pre-diabetes and diabetes imposes substantial maternal and fetal risks, these women have a critical need for safe, effective contraception. Data are lacking regarding the best contraceptive methods for women with recent GDM, and no studies address intrauterine contraceptive use in these patients. The proposed study evaluates the effect of the levonorgestrel intrauterine device (IUD) on carbohydrate metabolism in women with recent GDM. Study Aims and Hypothesis: The primary aim of this randomized controlled trial is to compare glucose tolerance in the first 12 months postpartum in two groups of women with GDM in the last pregnancy: 1) those using a levonorgestrel IUD, and 2) those using a copper IUD. The primary hypothesis is that glucose tolerance 12 months postpartum in the group of women using the levonorgestrel IUD will be similar to that of copper IUD users. Secondary aims include comparison of lipid metabolism, blood pressure, body mass index, continuation rates, pregnancy rates, contraceptive satisfaction, and adverse events in the two groups. Methods: Patients will be recruited during the third trimester of pregnancy. Women will undergo 2-hour oral glucose tolerance testing 6 weeks postpartum; those with normal glucose tolerance will be eligible for randomization and receive either a levonorgestrel or copper IUD. Seventy-two subjects will be enrolled and followed for 12 months after delivery. Twelve months postpartum, subjects will repeat glucose tolerance testing. The primary outcome measure is glucose tolerance at 12 months postpartum. Fasting glucose, lipids, and insulin will be compared between groups at 6 weeks, 6 months, and 12 months postpartum. Information on pregnancy rates, continuation rates, and adverse events will be collected. Contraceptive satisfaction will be measured using a visual analog scale.