Abstract
Hybrid closed-loop systems (HCLS) are the recommended therapy for children and adults with type 1 diabetes mellitus (T1DM), offering evident advantages in glycemic control and quality of life. Nevertheless, official recommendations for its use during pregnancy remain limited. We present three cases of pregnant women with T1DM treated off-label with MiniMedTM 780G HCLS in auto mode throughout pregnancy, labor, delivery and postpartum. System settings were maintained during labor and delivery, except for insulin-to-carbohydrate ratios, which were reduced by 15%–90%. The mean time-in-range (TIR) during labor and delivery was 87.7%, with no hypoglycemia. In postpartum, mean TIR was 90.0%, with 1.0% time-above-range. These cases showed that MiniMed™ 780G was effective and safe in labor, delivery and postpartum, adapting to rapid insulin requirements changes while reducing the burden of T1DM management for women and obstetric staff.
Introduction
Type 1 diabetes mellitus (T1DM) during pregnancy is associated with adverse maternal and fetal outcomes. Achieving tight glycemic control can improve these outcomes. 1 International guidelines recommend pregnancy glycemic targets, including a pregnancy-specific time-in-range (TIRp) (3.5–7.8 mmol/L [63–140 mg/dL]) ≥70%, and hemoglobin A1c (HbA1c) below 42 mmol/mol (6.0%) without significant hypoglycemia. 2 However, reaching these targets is challenging for most pregnant women with T1DM. Glycemic control during labor and delivery is critical to minimize neonatal hypoglycemia and potential long-term neurological sequels. 3 This period is marked by glycemic instability and increased hypoglycemia risk due to factors such as changes in cortisol secretion; increased glucose consumption by uterine and skeletal muscles; and a rapid decline in insulin resistance after the feto-placental unit is expelled.3,4 Consequently, achieving optimal glycemic control is challenging.
Hybrid closed-loop systems (HCLS) use advanced algorithms to adjust basal insulin rates, deliver automated insulin boluses for hyperglycemia, and suspend insulin delivery when hypoglycemia is predicted. 5 HCLS have become a standard approach for managing T1DM, significantly improving glycemic control and quality of life. 6 Currently, international guidelines recommend HCLS for T1DM management to both youth and adults who can safely use these devices. 6 A recent systematic review showed that the MiniMedTM 780G HCLS (Medtronic®) increased TIR (3.9–10.0 mmol/L [70–180 mg/dL]) to 75%–80%, reduced time-above-range (TAR), time-below-range (TBR), HbA1c, mean glucose, and coefficient of variation. This study also noted reductions in psychological distress and disease burden, with only exceptional cases of diabetic ketoacidosis and severe hypoglycemia observed. 7
The American Diabetic Association, in Standards of Care in Diabetes-2025, acknowledges that none of the current HCLS have algorithms specifically designed for pregnancy goals. However, it may be appropriate for carefully selected pregnant individuals with T1DM, in the setting of using assistive techniques with expert guidance. 6 In Europe, the CamAPS FX system is currently the only HCLS licensed for use in pregnancy. 8
We describe three cases of pregnant women with T1DM who were treated off-label with MiniMedTM 780G HCLS in auto mode (AM) throughout pregnancy, labor, delivery, and postpartum, focusing on glycemic control during peripartum period.
Case reports
Case 1
A 38-year-old woman with a 31-year history of T1DM and proliferative diabetic retinopathy began using MiniMed™ 780G HCLS with Guardian 4 continuous glucose monitor (CGM) at 36th week of gestation. Her medical history included autoimmune hypothyroidism, polycystic ovary syndrome, and overweight, and her treatment included Fiasp® (aspart) insulin and metformin (3 g/day). Before switching to HCLS, she used an Accu-Chek Spirit Combo® (Roche®, Switzerland) insulin pump, with HbA1c ranging from 50 to 57 mmol/mol (6.7%–7.4%) before pregnancy. During pregnancy, TIRp was 60% in the first two trimesters and 61% in the third, with TBR decreasing from 13% to 7%.
HCLS settings included a glucose target of 5.5 mmol/L (100 mg/dL) and an active insulin time (AIT) of 2 h. After starting HCLS, time in AM was 97%, achieving a TIRp of 82% without hypoglycemia events.
At 38w6d, she went into labor, but a caesarean section was required due to labor dystocia. HCLS remained in AM with the previous settings throughout labor and delivery. After delivery, the insulin-to-carbohydrate ratios (ICRs) were reduced by 50%. Her average sensor glucose (SG) was 9.0 ± 1.7 mmol/L (163 ± 30 mg/dL), with 78% TIR (3.9–10.0 mmol/L [70–180 mg/dL]) and no hypoglycemia (Fig. 1A). The newborn, a female weighing 3430 g (z-score 0.51; 70th percentile), had Apgar scores of 9, 10, and 10 at 1, 5, and 10 min, respectively, and required phototherapy for neonatal jaundice.

Continuous glucose monitoring data and insulin events during labor and delivery day, and first day of the puerperium. Graphics were obtained from CarelinkTM Clinic software. A—Case 1; B—Case 2; C—Case 3.
In the first 24 h postpartum, TIR was 80%, with one mild hypoglycemic event (3.2 mmol/L [58 mg/dL]) (Fig. 1A). Her total daily insulin dose (TDID) was 35.9 units (30% basal, 70% bolus), which is similar to the TDID at the end of pregnancy (34.0 units; 49% basal, 51% bolus). For the remainder 72-h hospital stay, TIR was 92% with no hypoglycemia. Over the first 4 weeks postpartum, while breastfeeding, the average SG was 7.2 ± 2.6 mmol/L (129 ± 47 mg/dL), with 81% TIR and 5% TBR.
Case 2
A 29-year-old woman with a 20-year history of T1DM and proliferative diabetic retinopathy started using MiniMed™ 780G HCLS with Guardian 4 CGM and Humalog® (lispro) insulin at the 9th week of gestation. Comorbidities included autoimmune hypothyroidism. She had previously used an Accu-Chek Spirit Combo® insulin pump and reported a preconception HbA1c of 49–54 mmol/mol (6.6–7.1%). During early pregnancy, TIRp was 36%, and pregnancy-specific below-range (TBRp) was 1% before switching to HCLS.
In the first trimester, the glucose target was set at 6.1 mmol/L (110 mg/dL) with an AIT of 3 h to address the patient’s fear of hypoglycemia. She spent 99% time in AM with a TIRp of 44% and no hypoglycemic events. To optimize glycemic control, the target glucose was adjusted to 5.5 mmol/L (100 mg/dL) and AIT to 2 h, although the patient made self-adjustments to settings without medical guidance throughout the pregnancy. By the end of the second trimester, AIT was restored to 3 h. In the third trimester, the glucose target was increased to 6.7 mmol/L (120 mg/dL) with an AIT of 4 h. The ICR became more aggressive as pregnancy progressed, then reduced by the end. She spent 100% time in AM during the second and third trimesters, with TIRp of 40% and 55%, respectively without hypoglycemia. She was hospitalized twice: once for severe hyperemesis gravidarum at 13w1d, and once for a urinary tract infection at 22w6d.
At 37w3d, the patient developed preeclampsia with severe features, leading to a caesarean due to fetopelvic disproportion. HCLS remained in AM, with previous settings. The average SG was 7.0 ± 2.1 mmol/L (126 ± 38 mg/dL), with a TIR of 87% and no hypoglycemic events (Fig. 1B). The male neonate had macrosomia (weight: 4880 g; z-score 3.58; 100th percentile), Apgar scores of 9, 10, and 10 at 1, 5, and 10 min, respectively, and presented clavicle fracture and neonatal jaundice, treated with phototherapy.
In the first 24 h postpartum, TIR was 97%, with no hypoglycemic events, and TDID was 31.0 units (59% basal, 41% bolus) (Fig. 1B), which corresponds to 75.4% of TDID at the end of pregnancy (41.1 units; 64% basal, 36% bolus). On the second day of puerperium, ICR was reduced by 15%. During the next 72 h, TIR was 83%, with no hypoglycemia. Over the first 4 weeks postpartum, while breastfeeding, the average SG was 8.4 ± 2.3 mmol/L (152 ± 42 mg/dL), with 76% TIR and 1% TBR. For optimized glycemic control during this period, the glucose target was set at 5.5 mmol/L (100 mg/dL), AIT at 2 h, and ICR was increased by 20% at breakfast.
Case 3
A 35-year-old woman with an 11-year long history of T1DM began using MiniMed™ 780G HCLS with Guardian 4 CGM and Humalog® (lispro) insulin 4 months before conception. She also had autoimmune hypothyroidism. HbA1c was 61 mmol/mol (7.7%) before starting HCLS, which subsequently decreased to 49 mmol/mol (6.6%).
Throughout pregnancy, the glucose target was set to 5.5 mmol/L (100 mg/dL) with an AIT of 2 h. The ICR were intensified, and she was advised to increase the time between bolus and breakfast as pregnancy advanced. In the first, second, and third trimesters, she spent 86%, 94%, and 97% in AM, achieving TIRp of 55%, 59%, and 55%, respectively, with TBRp 1% in the first two trimesters, and no hypoglycemic events in the third.
At 38w4d, the patient underwent a caesarean section due to a prior caesarean section and fetopelvic disproportion. HCLS settings remained unchanged. Following delivery, the ICR was reduced up to 90%, which correspond to70% of prepregnancy ICR. The average SG was 6.8 ± 1.3 mmol/L (123 ± 23 mg/dL), with 98% TIR and no hypoglycemia (Fig. 1C). The male neonate weighted 3700 g (z-score 0.92; 88th percentile), had Apgar scores of 9, 10, and 10 at 1, 5, and 10-min, respectively, and developed neonatal jaundice that did not require phototherapy.
In the first 24 h postpartum, TIR was 100%, and TDID was 35.1 units (51% basal, 49% bolus) (Fig. 1C), decreasing 44.0% from TDID at the end of pregnancy (79.7 units; 26% basal, 74% bolus). Over the next 72 h, TIR was 91% with occasional mild hypoglycemia, primarily occurring after lunch (TBR 3%). During the first 4 weeks postpartum, while breastfeeding, the average SG was 7.8 ± 2.9 mmol/L (141 ± 53 mg/dL), with 80% TIR and 2% TBR.
Table 1 and Supplementary Table S1 provide detailed CGM metrics, insulin doses, and carbohydrate intake during pregnancy, labor, and puerperium, as well as other maternal outcomes.
Maternal and Fetal Outcomes and Continuous Glucose Monitoring Metrics, Insulin Doses, and Carbohydrate Intake During Pregnancy, Labor, Delivery, and Puerperium
Time-above-range: high (10.0–13.8 mmol/L) + very high (≥13.9 mmol/L). Time-below-range: low (3.1–3.8 mmol/L) + very low (≤3.0 mmol/L).
DM, diabetes mellitus; BMI, body mass index; HCLS, hybrid closed-loop system; SG, sensor glucose; SD, standard deviation; GMI, glucose management indicator; SBP, systolic blood pressure; DBP, diastolic blood pressure.
Discussion
To our knowledge, apart from the clinical trial, this is the first real-world case series reporting MiniMed™ 780G HCLS use during labor, delivery, and postpartum. HLCS performed effectively, achieving over 78% TIR with minimal hypoglycemia. In Case 3, hypoglycemia was likely associated with inadequate ICR.
A 2024 randomized controlled trial (RCT) (CRISTAL) includes 46 women who used MiniMed™ 780G HCLS throughout pregnancy. The mean TIRp was 66.5 ± 10%, 8 higher than in our cohort. Guibert A. et al. conducted a retrospective study, reporting TIRp in the first (54%) and second trimester (64%) that were similar to those observed in our patients. 9 In clinical practice, challenges such as missed or delayed prandial insulin boluses, fear of hypoglycemia, and errors in carbohydrate counting can affect glycemic control.10,11 Our cohort showed near absence of hypoglycemia. CRISTAL trial and Guibert A. et al. reported similar results, with a pTBR of 2.5% ± 2.8% 8 and 1.3%–2.0%, 9 respectively. Hypoglycemia is common under standard treatments, particularly in early pregnancy. Previous studies have shown that up to 50% of women experience severe hypoglycemia.12,13 While hypoglycemia itself is not directly linked to adverse maternal or fetal outcomes, it can limit the ability of patients to achieve tight glycemic control and may lead to severe accidents. 13
Women using standard insulin pump or variable-rate intravenous insulin infusion (VRIII) during labor and delivery generally achieved good mean glucose levels.14–16 However, the hypoglycemia rates during this period varies widely across studies (0%–56%). 17 The CRISTAL trial allowed maintaining AM during labor and delivery, advising a higher glucose target of 6.1–6.7 mmol/L (110 or 120 mg/dL) and an ICR reduction by at least 50% between the end of the dilation phase and delivery. 18 However, 60% (n = 15) of participants opted to maintain the 5.5 mmol/L (100 mg/dL) target. Overall, on delivery day, TIR was 86.6% ± 13.7%, and TBR was 2.0% ± 3.0%. 19 In our patients, glycemic metrics were similar while maintaining previous settings, demonstrating that HLCS adapted quickly to the rapid change in insulin requirements after delivery. This highlights HLCS’s ability to effectively adjust to significant variability in insulin needs. MiniMed™ 780G HLCS features an adaptive algorithm that updates daily at midnight, adjusting auto basal insulin delivery, limits, and mathematical model parameters based on insulin delivery, SG data, and carbs entries. The system can also deliver auto-correction boluses and adjust the auto basal rate every 5 min, using real-time data to achieve the set glucose target. 20
A retrospective study evaluating glycemic control without CGM in women with T1DM during early postpartum showed that most patients had unstable glycemic control from delivery until discharge, with an average of 1.8 ± 1.2 episodes of hyperglycemia daily and 0.5 ± 0.6 episodes of hypoglycemia daily. 3 In the CRISTAL trial, TIR was 86.8% ± 6.7%, with 5.2% ± 3.3% TBR. 19 Conversely, our patients spent an average of only 1% TBR. HLCS has the potential to improve glucose variability postpartum and reduce hypoglycemic episodes. Furthermore, it avoids the transition from subcutaneous insulin therapy to VRIII, which is expensive, invasive, and limits mobility; reduces the burden on obstetric staff with limited experience managing T1DM; minimizes the need for frequent glucose monitoring and insulin dose adjustments; and alleviates distress for women. 4
In conclusion, our cases demonstrate that MiniMed™ 780G has the potential to adapt to rapid changes in insulin requirements during the peripartum period. However, algorithm refinement is necessary to achieve glycemic metrics during pregnancy. Future research, including RCTs with larger sample sizes, is essential to validate its superior efficacy and safety. In addition, studies should evaluate treatment satisfaction among both the women using the MiniMed™ 780G and the obstetric staff, comparing it to current management approaches.
Perspective of patients
Case 1
“I was diagnosed with T1DM when I was 7 years old. In 2007, I was fortunate to be chosen to use an insulin pump, which has significantly and positively changed my life. I really wanted to be a mother. In January 2023, I became pregnant. In September 2023, I received the new MiniMed™ 780G HLCS. With its alerts and ability to automatically deliver insulin boluses when my glycemia begins to rise, it offered me greater support for a more peaceful and safer pregnancy. On October 10th, I went into labor, and the next day, with the insulin pump always in AM, my daughter was born, and everything went perfectly. Having the pump with me at all times made everything go even more smoothly. My glucose levels were automatically managed, and no intervention was needed. I felt much safer, it’s easier to keep glycemia under control, which greatly benefits both my health and quality of life.”
Case 3
“Regarding my labor, which was a caesarean section, I remember that absolutely nothing needed to be done. A few days before the delivery, I was instructed to only adjust the ICR after delivery. No other adjustments were necessary, and that’s exactly what happened. After delivery, I adjusted the ICR. During labor, since everything was well controlled, I didn’t have to worry about managing T1DM, and the insulin pump was the least of my concerns. The staff just kept asking if everything was fine and checking my glycemia. In my previous pregnancy, I had a standard insulin pump and the management during labor was very different. The reality is that this system is much less stressful and has really improved my quality of life.”
Footnotes
Ethics Approval Statement
Ethical approval for this publication was obtained from our Institutional Review Board (or Ethics Committee).
Patient Consent Statement
Written informed consent was obtained from the patients for publication of this case report.
Authors’ Contribution
J.G.: Wrote this case report. S.B., C.G., and G.N.: Involved in the patient care and revised the draft. H.U.F.: Helped in the writing of this case report and revised the draft. J.Q.: Revised the draft.
Author Disclosure Statement
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of these case reports.
Funding Information
There is no source of research support to declare.
Abbreviations Used
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
