Randomized Trial of Automated Insulin Delivery in Pregnant Women with Type 1 Diabetes

Objective: To examine the clinical efficacy of automated insulin delivery (AID) during type 1 diabetes (T1D) pregnancy. Methods and Outcomes: We randomized pregnant women with T1D from nine UK clinical sites to hybrid closed-loop (CamAPS FX) or standard insulin therapy with continuous glucose monitoring. The primary outcome was the between-group difference in percentage time in the pregnancy-specific target glucose range (63-140mg/dl) from 16-weeks’ gestation until delivery. Analyses were performed according to intention-to-treat principles. Key secondary outcomes included overnight time-in-range (TIR), time spent hyperglycaemic (time-above-range, TAR >140mg/dl), HbA1c and safety events. Results: 124 participants (aged 31.1 ± 5.3yrs, HbA1c 7.7 ± 1.2%, T1D duration 2-31yrs, weight 49.0-138.0kg, total daily insulin doses 0.3-1.4units/kg) were randomized. The percentage of time that maternal glucose levels were within target was higher during AID than standard insulin therapy; mean between-group treatment difference 10%; 95% CI 7 to 10%; p < 0.001. Participants randomized to AID had larger reductions in hyperglycemia (AID vs control -11%; 95% CI -14 to - 7%; p<0.001), higher overnight time-in-range; (13%; 95% CI 9 to 17%; p<0.001), and lower HbA1c (-0.34%; 95%CI -0.52 to -0.15%; p<0.001), without additional insulin, weight gain or hypoglycemia. The treatment effect was apparent from early pregnancy, consistent across clinical sites, maternal HbA1c categories and previous insulin pump or injections. A significantly higher percent of AID participants reached CGM targets (TIR >70%: 46% vs 10%; p<0.001 and TAR <25%: 37% vs 11%; p=0.007 [AID vs. control]). One neonatal death that was unrelated to standard insulin therapy occurred. Conclusions: AID significantly improved maternal glycemia throughout T1D pregnancy. Our results support proposed NICE guideline recommendations that hybrid closed-loop therapy should be offered to all pregnant women with T1D. Disclosure H. R. Murphy: Advisory Panel; Medtronic, Research Support; Dexcom, Inc. L. Shepstone: None. M. E. Wilinska: Consultant; CamDiab Ltd. S. Bergford: None. J. Sibayan: None. C. Kollman: Research Support; Insulet Corporation, Dexcom, Inc., Tandem Diabetes Care, Inc. R. Beck: Consultant; Eli Lilly and Company, Novo Nordisk, Diasome, Insulet Corporation, Research Support; Tandem Diabetes Care, Inc., Beta Bionics, Inc., Dexcom, Inc., Bigfoot Biomedical, Inc., Medtronic, Ascensia Diabetes Care, Roche Diabetes Care, Eli Lilly and Company, Novo Nordisk. R. Hovorka: Advisory Panel; Ypsomed AG, Consultant; Abbott Diabetes, B. Braun, Speaker's Bureau; Eli Lilly and Company, Stock/Shareholder; CamDiab Ltd. T. T. M. Lee: None. C. Collett: None. S. Hartnell: Advisory Panel; Dexcom, Inc., Medtronic, Consultant; CamDiab Ltd., Other Relationship; AskDiabetes Ltd, Speaker's Bureau; Abbott Diabetes, Ypsomed AG. E. M. Scott: Research Support; Abbott Diabetes, Speaker's Bureau; Abbott Diabetes. R. S. Lindsay: None. K. F. Hunt: None. D. R. Mccance: None. M. Hammond: None. Funding National Institute for Health Research (16/35/01); JDRF (22-2013-266, 2-RSC-2019-828-M-N); Diabetes Research and Wellness Foundation (SECF/21 to T.T.M.L.)