Comparing advanced hybrid closed loop therapy and standard insulin therapy in pregnant women with type 1 diabetes (CRISTAL): a parallel-group, open-label, randomised controlled trial

Background Advanced hybrid closed loop (AHCL) therapy can improve glycaemic control in pregnant women with type 1 diabetes. However, data are needed on the efficacy and safety of AHCL systems as these systems, such as the MiniMed 780G, are not currently approved for use in pregnant women. We aimed to investigate whether the MiniMed 780G can improve glycaemic control with less hypoglycaemia in pregnant women with type 1 diabetes. Methods CRISTAL was a double -arm, parallel -group, open -label, randomised controlled trial conducted in secondary and tertiary care specialist endocrinology centres at 12 hospitals (11 in Belgium and one in the Netherlands). Pregnant women aged 18-45 years with type 1 diabetes were randomly assigned (1:1) to AHCL therapy (MiniMed 780G) or standard insulin therapy (standard of care) at a median of 10<middle dot>1 (IQR 8<middle dot>6-11<middle dot>6) weeks of gestation. Randomisation was done centrally with minimisation dependent on baseline HbA (1c) , insulin administration method, and centre. Participants and study teams were not masked to group allocation. The primary outcome was proportion of time spent in the pregnancy -specific target glucose range (3<middle dot>5-7<middle dot>8 mmol/L), measured by continuous glucose monitoring (CGM) at 14-17 weeks, 20-23 weeks, 26-29 weeks, and 33-36 weeks. Key secondary outcomes were overnight time in target range, and time below glucose range (<3<middle dot>5 mmol/L) overall and overnight. Analyses were conducted on an intention -to -treat basis. This trial is registered with ClinicalTrials.gov (NCT04520971). Findings Between Jan 15, 2021 and Sept 30, 2022, 101 participants were screened, and 95 were randomly assigned to AHCL therapy (n=46) or standard insulin therapy (n=49). 43 patients assigned to AHCL therapy and 46 assigned to standard insulin therapy completed the study. At baseline, 91 (95<middle dot>8%) participants used insulin pumps, and the mean HbA 1c was 6<middle dot>5% (SD 0<middle dot>6). The mean proportion of time spent in the target range (averaged over four time periods) was 66<middle dot>5% (SD 10<middle dot>0) in the AHCL therapy group compared with 63<middle dot>2% (12<middle dot>4) in the standard insulin therapy group (adjusted mean difference 1<middle dot>88 percentage points [95% CI -0<middle dot>82 to 4<middle dot>58], p=0<middle dot>17). Overnight time in the target range was higher (adjusted mean difference 6<middle dot>58 percentage points [95% CI 2<middle dot>31 to 10<middle dot>85], p=0<middle dot>0026), and time below range overall (adjusted mean difference -1<middle dot>34 percentage points [95% CI, -2<middle dot>19 to -0<middle dot>49], p=0<middle dot>0020) and overnight (adjusted mean difference -1<middle dot>86 percentage points [95% CI -2<middle dot>90 to -0<middle dot>81], p=0<middle dot>0005) were lower with AHCL therapy than with standard insulin therapy. Participants assigned to AHCL therapy reported higher treatment satisfaction. No unanticipated safety events occurred with AHCL therapy. Interpretation In pregnant women starting with tighter glycaemic control, AHCL therapy did not improve overall time in target range but improved overnight time in target range, reduced time below range, and improved treatment satisfaction. These data suggest that the MiniMed 780G can be safely used in pregnancy and provides some additional benefits compared with standard insulin therapy; however, it will be important to refine the algorithm to better align with pregnancy requirements.