In 2001, gestational diabetes (GD) remains the subject of much debate, particularly as regards screening and diagnosis, therapeutic management and effects on perinatal outcome. It continues to be defined as a disorder of glucose tolerance occurring for the first time during pregnancy, whatever the outcome during the post-partum period. There is as yet still no consensus concerning screening and diagnosis criteria, partly due to the existence of a continuum between maternal blood glucose levels and perinatal outcome, which means that any threshold is necessarily arbitrary. There is no general agreement concerning therapeutic management. As regards diet, there are insufficient randomised prospective studies that could serve to establish minimum required calorie intakes and various regimen. Intervention studies should provide more accurate information concerning the role of diet in therapeutic strategies. It is also necessary to define the role of other therapeutic alternatives such as insulin analogues, or more recently, the sulphonylureas. Nevertheless, although there is still a large debate, the physiopathological mechanisms are becoming increasingly clear. GD and type 2 diabetes appear to be the same entity, with the former constituting an early sign of the latter. Long-term management of women with GD and treatment modalities also require better definition, since these patients are at risk for diabetes, mainly type 2.