Background
Diabetes in pregnancy is associated with preterm delivery, birthweight extremes, and increased rates of congenital anomaly, stillbirth, and neonatal death. We aimed to identify and compare modifiable risk factors associated with adverse pregnancy outcomes in women with type 1 diabetes and those with type 2 diabetes and to identify effective maternity clinics.
Methods
In this national population-based cohort study, we used data for pregnancies among women with type 1 or type 2 diabetes collected in the first 5 years of the National Pregnancy in Diabetes audit across 172 maternity clinics in England, Wales, and the Isle of Man, UK. Data for obstetric complications (eg, preterm delivery [<37 weeks' gestation], large for gestational age [LGA] birthweight [>90th percentile]) and adverse pregnancy outcomes (congenital anomaly, stillbirth, neonatal death) were obtained for pregnancies completed between Jan 1, 2014, and Dec 31, 2018. We assessed associations between modifiable (eg, HbA1c, BMI, pre-pregnancy care, maternity clinic) and non-modifiable risk factors (eg, age, ethnicity, deprivation, duration of type 1 diabetes) with pregnancy outcomes in women with type 1 diabetes compared with those with type 2 diabetes. We calculated associations between maternal factors and perinatal deaths using a regression model, including diabetes type and duration, maternal age, BMI, deprivation quintile, first trimester HbA1c, preconception folic acid, potentially harmful medications, and third trimester HbA1c.
Findings
Our dataset included 17 375 pregnancy outcomes in 15 290 pregnant women. 8690 (50·0%) of 17 375 pregnancies were in women with type 1 diabetes (median age at delivery 30 years [10–90th percentile 22–37], median duration of diabetes 13 years [3–25]) and 8685 (50·0%) were in women with type 2 diabetes (median age at delivery 34 years [27–41], median duration of diabetes 3 years [0–10]). The rates of preterm delivery (3325 [42·5%] of 7825 pregnancies among women with type 1 diabetes, 1825 [23·4%] of 7815 with type 2 diabetes; p<0·0001), and LGA birthweight (4095 [52·2%] of 7845 with type 1 diabetes, 2065 [26·2%] of 7885 with type 2 diabetes; p<0·0001) were higher in type 1 diabetes. The prevalence of congenital anomaly (among women with type 1 diabetes: 44·8 per 1000 livebirths, terminations, and fetal losses; among women with type 2 diabetes: 40·5 per 1000 livebirths, terminations, and fetal losses; p=0·17) and stillbirth (type 1 diabetes: 10·4 per 1000 livebirths and stillbirths; type 2 diabetes: 13·5 per 1000 livebirths and stillbirths; p=0·072) did not significantly differ between diabetes types, but rates of neonatal death were higher in mothers with type 2 diabetes than in those with type 1 diabetes (type 1 diabetes: 7·4 per 1000 livebirths; type 2 diabetes 11·2 per 1000 livebirths; p=0·013). Across the whole study population, independent risk factors for perinatal death (ie, stillbirth or neonatal death) were third trimester HbA1c of 6·5% (48 mmol/mol) or higher (odds ratio 3·06 [95% CI 2·16–4·33] vs HbA1c <6·5%), being in the highest deprivation quintile (2·29 [1·16–4·52] vs the lowest quintile), and having type 2 diabetes (1·65 [1·18–2·31] vs type 1 diabetes). Variations in HbA1c and LGA birthweight were associated with maternal characteristics (age, diabetes duration, deprivation, BMI) without substantial differences between maternity clinics.
Interpretation
Our data highlight persistent adverse pregnancy outcomes in women with type 1 or type 2 diabetes. Maternal glycaemia and BMI are the key modifiable risk factors. No maternity clinics were had appreciably better outcomes than any others, suggesting that health-care system changes are needed across all clinics.
Funding
None.
CORE (COnnecting REpositories)
CORE (COnnecting REpositories)