Skip to main content

The Need for Automated Insulin Delivery Systems During Pregnancy in Women with Type 1 Diabetes

Updated: 4/15/24 2:58 amPublished: 2/25/23 2:58 am
By Susannah ChenApril Hopcroft

Automated Insulin Delivery for Type 1 Diabetes While Pregnant Keeping up with insulin requirements can be especially challenging during pregnancy. Researchers reviewed the latest data and emphasized the need to include specific software in AID systems for pregnant women with type 1 diabetes.

Using automated insulin delivery during pregnancy would be beneficial for women with type 1 diabetes because they could optimize blood sugar control and improve quality of life, according to experts in diabetes care during pregnancy. 

“Could automated insulin delivery help in pregnancy? Yes, by all means,” said Dr. Sarit Polsky, assistant professor and director of the Pregnancy and Women's Health Clinic at the University of Colorado Barbara Davis Center for Diabetes. 

She cited a handful of current studies PICLS, AiDAPT, CRISTAL, and CIRCUIT, among others – that used AID systems in pregnancy.

Overall, studies demonstrate that pregnancy-specific AID calculations can help women with type 1 diabetes optimize glycemic control during pregnancy, improve quality of life, and potentially reduce the overall cost of maternal health on the hospital system. However, experts emphasized the need to develop algorithms that are safe and approved for pregnancy.

One AID system, the CamAPS Fx, has been licensed for use during pregnancy, but it is only approved for use in Europe, the U.K., and Australia. In the United States, every available AID system currently excludes use during pregnancy.

The challenges of managing diabetes in pregnancy

Dr. Helen Murphy, honorary consultant physician at the Cambridge University NHS Foundation Trust, helped pioneer the micromanagement of diabetes in pregnancy through a series of studies of pregnancy-specific algorithms in AID systems. 

Murphy described a wide range of challenges in using AID systems for pregnant women with type 1 diabetes including:

  • Better managing high-carbohydrate meals 

  • Developing an AID system that can deliver insulin quickly enough to avoid post-meal high blood sugars 

  • Knowing when to increase insulin delivery (such as illness and rising insulin resistance during the third trimester of pregnancy) versus knowing when to reduce or stop insulin delivery (like during low blood sugar events, hot weather, or physical activity)

Lois Donovan, a clinical professor at the University of Calgary, emphasized the psychological impact of living with type 1 diabetes through pregnancy.

As principal investigator for CIRCUIT, a trial that studies the use of Tandem’s Control-IQ software in pregnant women with type 1, Donovan said that patients reported concerns such as fear of the baby dying, the pressure to care for a baby, and the increased burden of work for type 1 women in pregnancy compared to other pregnant women.

Assisted techniques can support AID use in pregnancy

Based on the results of her PICLS study, Polsky offered the opinion that clinicians should only medically allow off-label AID system use in individuals who are able to navigate the system using “assisted” techniques to help optimize blood sugars during pregnancy.

Some examples of  Polsky’s “assisted” techniques include: 

  • Using the lowest glucose target level that the system allows 

  • Entering “fake carbohydrate boluses'' based on insulin sensitivity 

  • Limiting correction boluses to less than once every two hours 

  • Pre-bolusing 15 minutes before meals during the first trimester and 45 minutes before meals in the third trimester 

  • “Super bolusing,” which is when basal insulin delivery is curtailed and delivered as an additional carb or correction bolus to create more impact in situations when insulin is needed quickly

New data suggests potential for MiniMed 780G in pregnancy

The latest insight on using AID in pregnancy comes from the CRISTAL RCT, which tested the MiniMed 780G system in 95 pregnant people with type 1 diabetes in Belgium and the Netherlands. 

At the start of the study, participants had an average A1C of 6.5% – indicating strong initial glycemic control. Participants were randomly assigned to use the MiniMed 780G (with a target of 100 mg/dL and active insulin time of two hours) or standard care (insulin pumps or multiple daily injections). 

What were the key findings? 

Participants using the MiniMed 780G did not see improvements in overall pregnancy time in range (63-140 mg/dL) compared to the standard care group. 

However, compared to standard care, the MiniMed 780G showed: 

  • Improved overnight time in range (63-140 mg/dL) 

  • Reduced time below range (less than 63 mg/dL)

  • Reduced weight gain during pregnancy

  • Reduced hypoglycemia unawareness

  • Minimized glycemic variability

  • Improved treatment satisfaction

There were no unexpected safety concerns and similar numbers of hypoglycemia events in both groups: eight in the AID group and seven in the standard care group. 

Many participants – about 61% – relied on Polsky’s technique of entering “fake carbs” to counteract the 780G’s safe meal bolus feature. This feature adjusts the bolus downwards when the algorithm predicts hypoglycemia within the next four hours. 

Dr. Katrien Benhalima, CRISTAL investigator and deputy head of the Clinic of Endocrinology at UZ Leuven in Belgium, added that the system performed well overnight in pregnant people with type 1 diabetes. However, she said the algorithm lacked flexibility with meals and wasn’t able to adapt fast enough to the increased insulin requirements later in pregnancy. 

The bottom line

AID systems can improve glucose control in pregnant women, but plenty of guidance is needed from knowledgeable healthcare providers. 

“It’s still work. It isn’t just plug-and-play. You have to give pregnant women with diabetes support,” said Donovan.

Overall, experts emphasized the potential of automated insulin delivery systems to improve health outcomes for both mom and baby – provided algorithms are adjusted appropriately to account for the unique challenges of insulin delivery in pregnancy. 

“It is important that the algorithms can be further refined to better align with insulin needs in pregnancy because any improvement in time in range also translates to improved pregnancy outcomes,” said Benhalima. 

Learn more about managing diabetes in pregnancy: 

What do you think?

About the authors

Prior to joining The diaTribe Foundation as a managing editor, Susannah Chen was a freelance writer and editorial content strategist focused on food, drink, and travel. A health journey involving... Read the full bio »
April Hopcroft joined diaTribe in 2023 as a Staff Writer after co-leading the Diabetes Therapy team at Close Concerns. She graduated from Smith College in 2021, where she majored in... Read the full bio »