TORONTO — Clinical trials have shown that many automated insulin delivery (AID) systems are safe and effective for nonpregnant patients with type 1 diabetes (T1D). They are associated with improved glycemic outcomes, reduced nighttime hypoglycemia, and no increased risk for diabetic ketoacidosis.

But T1D during pregnancy presents unique challenges for these systems, said Denice Feig, MD, professor of medicine at the University of Toronto, Toronto, and Lois Donovan, MD, clinical professor at the University of Calgary, Calgary, in a presentation at the Diabetes Canada and Canadian Society of Endocrinology and Metabolism (CSEM) Professional Conference 2025.

Time spent in the pregnancy-specific glucose range needs to be tighter compared with outside of pregnancy, given the risks to the mother and fetus, said Feig. There is also an increased risk for hypoglycemia early in pregnancy and a risk for hyperglycemia later when insulin resistance is increasing.

Closed-Loop AID

Feig reported on her group’s recent randomized clinical trial, which examined the safety and efficacy of a closed-loop AID system during pregnancy for women with T1D who were at less than 14 weeks’ gestation. The system combines a continuous glucose monitor (CGM) and insulin pump with an algorithm that tells the pump how much insulin to deliver based on data from the CGM.

The CIRCUIT trial was carried out at 14 diabetes-in-pregnancy clinics in Canada and Australia. Participants had an average age of 31.7 years; they had an A1c level of 7.4% during early pregnancy. For the primary analysis, researchers included data from 88 women.

Researchers collected 1 week of CGM data at baseline, after which participants were randomized to either continue with standard care (insulin pump or multiple daily injections) or to closed-loop therapy with an AID system. Data were collected throughout the study to 6 weeks post-partum.

All participants used the Dexcom G6 CGM to measure their glucose levels. Women in the closed-loop group used the Tandem t:slim X2 insulin pump with Control-IQ technology, which integrates with the Dexcom G6 CGM.

More Time in Range

The study showed that between 16 and 34 weeks’ gestation, women in the closed-loop group spent 65.4% of the time within the pregnancy-specific glucose range (63-140 mg/dL) compared with 50.3% of the time for women in the standard-care group. This meant that women in the closed-loop group spent 3 more hours per day in pregnancy range; they also spent less time below and above range.

A1c levels were also lower in the closed-loop group than in the standard-care group: 0.49% lower at 24 weeks’ gestation and 0.43% lower at 34 weeks’ gestation.

Exploratory results showed that fewer women had preeclampsia in the closed-loop group than in the standard-care group. The closed-loop group also had lower daily insulin doses. Gestational age at delivery and number of preterm births were similar in both groups.

There was also less weight gain during pregnancy in the closed-loop group. “This may be due to the fact that insulin is being delivered more efficiently or because there is less hypoglycemia, which means there is less need for patients to eat empty carbs to treat hypoglycemia,” Donovan told Medscape News Canada.

There were no safety concerns with the closed-loop group. One of the limitations of the study is that because participants had to interact with the AID system, it was an open-label study. Another is that the study was not large enough to assess pregnancy outcomes.

Better Quality of Life

Melissa-Rosina Pasqua, MD, PhD, assistant professor of endocrinology and metabolism at McGill University in Montreal, said during a separate presentation at the conference that while AID systems have improved, they’re still not perfect.

For example, “there’s the ongoing need to administer insulin for meals,” she said, “and even though we sometimes use harm reduction in the clinical setting, the AID systems are usually built with precise carb counting in mind,” which can be challenging for people to do accurately all the time.

Donovan agreed that while more work is needed on the AID systems, “we are getting closer to having a system that works well most of the time.”

The ability of AID systems to keep blood glucose within range during pregnancy, especially at night, has also improved women’s quality of life. Donovan said, “A 30-year-old patient who had been diagnosed with type 1 diabetes at age 12 said to me a week after starting on this AID system, ‘I haven’t slept that well since I was 12.’”

Donovan added, “I also sleep easier at night seeing the kind of control pregnant patients are getting.”

The study was funded by MSI Foundation and Diabetes Canada End Diabetes 100 award 2021 grant OG-3-21-5570-LD, the University of Calgary Clinical Research Fund, the Buckley Family Trust, the Mount Sinai Department of Medicine Research Fund, and the Medical Research Futures Fund. Feig reported receiving research funding or other support from Dexcom and Tandem. Donovan reported receiving research funding or other support from Dexcom, Diabetes Canada, Interanalytics, Medtronic, MSI, and Tandem Diabetes Care. Pasqua reported receiving speaker honoraria from Abbott, Medtronic, and Sanofi.

Shawn Radcliffe is a freelance health and science journalist based in Ontario, Canada, with more than 15 years writing about general health, medical research, mental health, and other topics for print and online publications. He previously performed laboratory and clinical research and research administration at universities in Philadelphia and Portland, Oregon. He has a master’s degree in science education from Drexel University.