Type 1 diabetes in pregnancy is associated with higher rates of maternal and infant complications. The complications are associated with maternal hyperglycaemia. Thus, the main goal of treatment for these women is to optimise glycaemic control and thereby improve clinical outcomes for themselves and for their baby.

This thesis examines glycaemic control in the mothers and infants of pregnancies affected by type 1 diabetes. I present the first home studies of closed-loop insulin delivery in this population. The aim of these studies was to assess the feasibility, efficacy, and utility of overnight and then day-and-night closed-loop insulin delivery in pregnant women with type 1 diabetes. The overnight study, which examined 16 pregnant women (mean age 34.1 years, HbA1c 6.8%, 14.4 weeks gestation), compared overnight use of the closed-loop system with sensor-augmented pump therapy in a 2x4-week randomised crossover design. We found that closed-loop therapy was associated with a 15% improvement in overnight time spent with target glucose concentration (3.5-7.8 mmol/L; 74.7% during closed-loop use vs 59.5% during sensor-augmented pump therapy use). The day-and night study also examined 16 pregnant women (mean age 32.8 years, HbA1c 8.0%, 16.4 weeks’ gestation) using a 2x4-week randomised crossover design to compare continuous day-and-night use of closed-loop insulin delivery with sensor-augmented pump therapy. This study enrolled a more diverse range of participants than the overnight study, but found that closed-loop therapy was associated with comparable glucose control and significantly less hypoglycaemia than sensor-augmented pump therapy. Chapter 4 examines women’s experiences of using the closed-loop system during pregnancy. While the system was generally well-received by participants, individual interactions and perceptions of the system varied markedly, and often did not align with biomedical measures of glycaemic response.

After participation in either crossover study, participants could choose to continue using the technology until delivery (overnight study), or until 6 weeks post-partum (day and night study). Those data are presented in Chapters 2 and 3. The combined data from the women who used the closed-loop system during labour and delivery in either study are presented in Chapter 5. Tight glycaemic control during labour and delivery has traditionally been considered important for reducing rates of neonatal hypoglycaemia. However, despite very tight maternal glycaemic control in the women who used closed-loop insulin delivery, rates of neonatal hypoglycaemia were high.

In order to better characterise the relationship between maternal glucose control in type 1 diabetes pregnancy and neonatal hypoglycaemia, Chapter 6 details an observational study in which continuous glucose monitoring was used to measure maternal and neonatal glycaemic control in 16 mother-infant pairs. The study found that, while neonatal hypoglycaemia was very frequent, it was generally, but not always, detected and treated effectively.

Together, these studies suggest that a novel management tool, closed-loop insulin delivery, can improve overnight glycaemic control, and perhaps reduce hypoglycaemia during type 1 diabetes-affected pregnancies above what is possible with currently available treatments. However, complication rates remain high for these women and their babies. Further research is needed both to further develop treatments that can improve maternal glycaemic control, and to better understand the pathogenesis of diabetes-related pregnancy complications, with the ultimate goal of improving outcomes for women and their children. A definitive trial to assess the clinical efficacy, patient acceptability, and cost effectiveness of closed-loop is now warranted.