SummaryIn order to improve the basis upon which to advise women with diabetic nephropathy about pregnancy, we studied the effect of diabetic nephropathy on the course of pregnancy, perinatal outcome, infant development and long-term outcome of the mothers. All pregnancies of women with diabetic nephropathy (defined as proteinuria > 400 mg/day (n = 26), creatinine clearance < 80 ml/min and hypertension in the first trimester (n = 10)) followed at our centre from 1982 to 1992 were identified (34 White class F and 2 White class T) and the women and their children re-examined in the spring 1993. From the first to the third trimester the percentage of women with proteinuria over 3 g/day increased from 14 to 53 % and those treated with antihypertensive medication from 53 to 97 %. There were no intrauterine or perinatal deaths, but one child died suddenly 4 weeks postpartum. Of 36 newborns (gestational week at birth 36(3), birth weight 2384(834) g)), 11 were born before week 34 and 8 had respiratory distress syndrome. Renal function in the first trimester, diastolic blood pressure in the third trimester and an HbAlc above normal were predictive of gestational age at delivery and low birth weight (stepwise regression analysis). At follow-up of the children (n = 35, age 4.5 (0.4-10) years) the majority (n = 27) were normally developed but seven had psychomotor retardation (four of them major). One child had a severe motor retardation due to a congenital anomaly. At follow up, 21 of the 29 mothers had preserved renal function (creatinine 1.3 (0.8-4.3) mg/dl and 8 had developed end stage renal disease and required dialysis (2 of whom were White class T) within 3 (1-9) years postpartum. Of those, 4 women (3 White F and i White T) had died. Pregnancy did not seem to specifically accelerate the rate of decline of renal function. In women with diabetic nephropathy perinatal mortality can be prevented but perinatal and long-term infant morbidity remains elevated. Women with severely impaired renal function before pregnancy are at risk for serious morbidity when their children are still young. Improvement might be made if all women were to receive specialized care and counselling before, throughout and after pregnancy. [Diabetologia (1995) Today the perinatal outcome of pregnancies of diabetic women without advanced diabetic complications is similar to that in non-diabetic women, if there is normoglycaemic control throughout pregnancy. Therefore, in women with well-controlled diabetes who do not have nephropathy or other advanced vascular complications their decision about whether or not to have children will largely be influenced by factors not associated with diabetes itself. In contrast, for women with diabetic nephropathy (who in most cases also have proliferative retinopa-