Abstract

The aim of this study was to evaluate the treatment, course and outcome of pregnancy in Finland using two cohorts of diabetic women. The clinical cohort consisted of data from all 210 women with Type 1 diabetes and their 296 pregnancies managed between 1986 and 1995 in the two northernmost provinces of Finland. The register-based study population included all 1442 mothers with a singleton birth who had insulin treatment during pregnancy in 1991–1995 according to the Medical Birth Register. Of these mothers, 954 (66%) had pre-existing diabetes.

Insulin-treated diabetes complicated 4.5/1000 births in Finland in 1991–1995, the prevalence of Type 1 diabetes being 2.9/1000 in the whole country and 3.3/1000 in Northern Finland. In the 1990’s the care of these women shifted from tertiary level only to include the secondary level hospitals as well, and was more often carried out on an out-patient basis. This care policy in association with the self-monitoring of blood glucose levels contributed to an obvious improvement in glycaemic control during pregnancy. Despite that, the high proportion (73%) of women entering pregnancy with unsatisfactory glycaemic control did not decrease during the study period.

Retinopathy complicated 134 (45.3%) diabetic pregnancies, while clinical nephropathy was found in 23 (7.8%) cases. Although retinopathy was more often aggravated during the first pregnancy, the occurrence of retinopathy or its severe form was not increased at the beginning of consecutive pregnancies. Of the mothers, 50 (16.9%) had pre-eclampsia during pregnancy, and in 28% of these cases it was classified as superimposed. It was found more often among primiparous than multiparous (25.6% vs. 11.0%, respectively), and its occurrence rose with the severity of diabetes.

In both cohorts, the rates of preterm deliveries, Caesarean sections and large for gestational age (LGA) infants were significantly (p < 0.001) higher in Type 1 diabetic pregnancies than in the background population. The rates of congenital anomalies (CA) were 540–629/10000 in two study populations, both being 2–3-fold as compared to the background population. Cardiac malformations were most common, with anomalies in the genitourinary tract and the musculoskeletal organs being next in frequency. Sixty-three percent of malformed infants were boys.

Though pregnancy itself was not found to worsen the prognosis of diabetes, at least in the short term, pregnancy in diabetic women still remains a high risk state with an increased rate of prematurity, operative deliveries, CAs and peri- and neonatal mortality. In order to decrease the mortality rate in diabetic births, attention should be directed at both the prevention of CA and at identifying the foetuses at risk for intrauterine death. The postneonatal mortality rate is also high, reflecting a shift in the deaths from the early neonatal period to a later age. Therefore, a combined mortality, including induced abortions, stillborns and infant deaths, would give a more realistic idea of the outcomes in diabetic pregnancies.