[Insulin dependent diabetes and pregnancy: evaluation of the insulin pump]

To evaluate the effectiveness of continuous insulin infusion (insulin pump) on the materno-foetal morbidity during pregnancy in patients with insulin-dependent diabetes mellitus.

A retrospective study from 1980 to 1991. SITE. Gynecology-Obstetrics Unit, University of Caen.

Eighty-one patients with insulin-dependent diabetes mellitus known to be affected before their pregnancy were followed in the unit from 1980 to 1991. This population was divided into two groups: in the first group, an insulin pump was installed before 15 weeks of amenorrhoea (n = 36) and in the second group, conventional treatment was given with three daily injections of insulin or with a pump installed after 15 weeks of amenorrhoea (n = 45).

In the first group with the insulin pump before 15 weeks, there was a higher proportion of severe diabetes, the first consultation occurred earlier, there were half as many cases of neonatal jaundice and the length of hospitalization during the first trimester of pregnancy was longer. There was no difference in Apgar scores, cord pH, birth weight and the proportion of foetal macrosomia, length of the hospitalization in the neonatality ward, rate of malformation, infection, low blood glucose and calcium, transitive respiratory distress and neonatal polycythaemia, length of hospitalization of the mother during the second and third week postpartum, the rate of urinary infection, high blood pressure, hydramnios during pregnancy, delivery route, haemoglobin Alc or fructosamine during pregnancy. There was no perinatal death.

Although there was no significant difference in the results, which may be explained by the higher number of severe cases of diabetes in the first group, the use of the insulin pump did not appear to improve control of blood glucose levels, and thus to improve the materno-foetal prognosis, except by the bias of earlier attentive management of the pregnancy which led to better outcome.