To evaluate and compare the outcome of pregnancies in women with type 1 diabetes treated with continuous subcutaneous insulin pump or multiple insulin injections.Thirty-three patients treated with pump, 23 with multiple injections. Were mainly investigated: glycemic control (HbA1C), fetal outcome (congenital malformations, preterm delivery, perinatal mortality), rates of gestational hypertension and cesarean section.HbA1C (1st, 2nd and 3rd trimester) with pump was not significantly different from that obtained with multiple injections (respectively, 7.5 +/- 1.24 vs 7.6 +/- 1.4; 6.34 +/- 0.6 vs 6.6 +/- 0.8; 6.6 +/- 0.7 vs 6.4 +/- 0.7). Fetal prognosis was also similar. The occurrence of gestational hypertension was linked to diabetic complications [OR 5,7, IC95% 1.24-25, p = 0.0025]. The rate of cesarean section (70% in women treated with pump) was influenced by diabetes duration. In planned pregnancy, independently of therapeutic device, HbA1C during organogenesis was better (6.9% +/- 0.78 vs 8.51 +/- 1.3, p0.0001) with no perinatal death and no major congenital malformation.Fetal prognosis is not overall significantly different with insulin pump compared with intensified conventional therapy. The initiation of insulin pump therapy should not be systematic. Instead, the benefits risks ratio must be assessed resulting in a tailored prescription according to individual needs. This therapeutic choice should be discussed before conception, as planned pregnancy is a main prognostic factor. Diabetes duration and complications remain key factors for the prognosis.
