Objective: To compare the outcomes of pregnancies in women with pre-existing, type 1 and type 2, diabetes and to examine the influence of ethnicity on these outcomes.
Design: Prospective cohort study.
Setting: Large district hospital in Yorkshire with an ethnically mixed population.
Sample: Case series of all 202 pregnancies in women with pre-existing diabetes, ending in miscarriage, termination of pregnancy or delivery between January 1994 and December 2002.
Methods: Univariate and multivariate logistic regression analysis comparing outcomes in type of diabetes and in ethnic group.
Main outcome measures: Fetal loss, perinatal and infant mortality and congenital anomaly.
Results: All 14 stillbirths and infant deaths and 13 of the 15 congenital malformations were to Asian women. Analysis within this ethnic group showed a very high rate of adverse birth outcome for type 1 diabetic women and for type 2 diabetic women on insulin before the pregnancy. Total pregnancy loss among type 1 diabetic women was 156 per 1000 and among type 2 diabetic women on insulin was 167 per 1000. Congenital abnormality rates were 156 per 1000 for type 1 diabetic women and 261 per 1000 for type 2 diabetic women on insulin. Asian type 2 diabetic women not on insulin prior to pregnancy had significantly better outcomes: Total pregnancy loss was 123 per 1000 and congenital abnormality rate was 32 per 1000. After adjustment for confounders, including type of diabetes, Asian women had significantly worse outcomes (combined perinatal loss and malformation) than Caucasian women [odds ratio (OR) 4.96, 95% confidence interval (CI) 1.16-21.1].
Conclusion: Ethnicity has a significant impact on the outcome of diabetic pregnancies, with worse outcomes for babies born to Asian mothers compared with Caucasian mothers. The use of insulin pre-pregnancy rather than type of diabetes appears to predict adverse outcome.