Objective: The aim of our work was to analyze the profile of patients, the course of pregnancy and the occurrence of maternal complications, fetal and neonatal along 3 levels of treatment of gestational diabetes mellitus : monitoring rules only lifestyle modifications, the use of low-dose insulin therapy, the use of high dose insulin. Materials and Methods: A retrospective study was conducted at the University Hospital of Angers on the period from June 12, 2008 to January 19, 2009. Patients included in the study had a single pregnancy complicated with gestational diabetes. The records of 94 patients were studied: 54 patients received only treatment with diet recommandations (diet group), 19 patients received insulin at low doses (≤ 0.25 IU / kg / d) and 18 patients received insulin at high doses (> 0.25 IU / kg / d). Results: Patients who used insulin (and a fortiori a high insulin dose) had a higher BMI before pregnancy. (Median BMI in the diet group was 25.6 kg/m2 versus 27.1 in the insulin group ≤ 0.25 IU/kg/day and 29.1 in the insulin group > 0.25 IU/kg/day, non- significant difference between groups). Women who had gestational diabetes with insulin treatment were diagnosed and care before (the first diabetologic consultation was 31.5 SA in the diet group, 30.9 SA in the insulin group ≤ 0 , 25 IU/kg/d and 28.4 SA in the insulin group > 0.25 IU/kg/day (p <0.05). We did not observe any difference in the occurrence of maternal and fetal complications during pregnancy. The birth was earlier among women in the group with insulin> 0.25 IU / kg / d (38.6 versus 39.3 SA SA in the other groups).The rate of induction of labor was higher in patients treated with insulin (33.3% in groups with insulin versus 19.3% in diet group). Conclusion: Patients who are going to use insulin at high doses early carry no more risk of developing complications during their pregnancies. The newborns to these mothers have no more complication.