Sage Journals: Discover world-class research

Abstract

Automated insulin delivery (AID) systems mimic an artificial pancreas via a predictive algorithm integrated with continuous glucose monitoring (CGM) and an insulin pump, thereby providing AID. Outside of pregnancy, AID has led to a paradigm shift in the management of people with type 1 diabetes (T1D), leading to improvements in glycemic control with lower risk for hypoglycemia and improved quality of life. As the use of AID in clinical practice is increasing, the number of women of reproductive age becoming pregnant while using AID is also expected to increase. The requirement for lower glucose targets than outside of pregnancy and for frequent adjustments of insulin doses during pregnancy may impact the effectiveness and safety of AID when using algorithms for non-pregnant populations with T1D. Currently, the CamAPS^® FX is the only AID approved for use in pregnancy. A recent randomized controlled trial (RCT) with CamAPS^® FX demonstrated a 10% increase in time in range in a pregnant population with T1D and a baseline glycated hemoglobin (HbA1c) ≥ 48 mmol/mol (6.5%). Off-label use of AID not approved for pregnancy are currently also being evaluated in ongoing RCTs. More evidence is needed on the impact of AID on maternal and neonatal outcomes. We review the current evidence on the use of AID in pregnancy and provide an overview of the completed and ongoing RCTs evaluating AID in pregnancy. In addition, we discuss the advantages and challenges of the use of current AID in pregnancy and future directions for research.

Keywords

automated insulin delivery type 1 diabetes pregnancy continuous glucose monitoring

Introduction

Pregnant women with type 1 diabetes (T1D) have an increased risk for maternal and neonatal complications, such as congenital malformations, preeclampsia, preterm delivery, and perinatal mortality as well as the need for prolonged hospital admissions after delivery and need for neonatal intensive care compared with the background population.¹ Several population-based cohort studies have shown that the use of technical devices, such as continuous glucose monitoring (CGM) systems and insulin pumps have not significantly improved pregnancy outcomes over time.^2,3 It is often challenging for women with T1D to achieve and maintain strict glycemic control throughout pregnancy due to the metabolic changes, such as the increased risk of hypoglycemia in early pregnancy,^4,5 and the increasing insulin resistance with delayed insulin absorption later in pregnancy.⁶

Continuous glucose monitoring is increasingly being used in women with T1D, especially following the results of the “Continuous Glucose Monitoring in Pregnant Women with Type 1 Diabetes” (CONCEPTT) trial,⁷ and CGM is routinely offered during pregnancy in some countries, including the United Kingdom, Denmark, Belgium, and Sweden.^8-14 International consensus on CGM targets recommends to spend > 70% of time in range in pregnancy (TIRp: 3.5-7.8 mmol/L or 63-140 mg/dL) and < 4% of time below range in pregnancy (TBRp: < 3.5 mmol/L or 63 mg/dL).¹⁵ However, pregnant women with T1D often do not achieve TIRp > 70% until the final weeks of pregnancy, if at all.^8,9,16-18

Automated insulin delivery (AID) systems mimic an artificial pancreas via a predictive algorithm that can be hosted on a smartphone that integrates CGM and insulin pump data and thereby provide automated glucose-responsive insulin delivery between meals and overnight with manually triggered premeal doses. Some AID also provide automated hyperglycemia correction boluses.^19,20 Outside of pregnancy, the use of AID has led to a paradigm shift in the management of T1D, with a 10% higher time in range (TIR; 3.9-10.0 mmol/L or 70-180 mg/dL), with low hypoglycemia risk and improved quality of life compared with stand-alone insulin pumps.^21-23 As the use of AID in clinical practice is increasing, the number of women of reproductive age becoming pregnant while using AID is also expected to increase. The requirement for a lower glucose target than outside of pregnancy and the need for frequent adjustments of insulin doses in women with T1D during pregnancy may impact the effectiveness and safety of AID when using algorithms for non-pregnant populations.²⁴ Clinicians and pregnant women are therefore faced with the dilemma whether to continue with the AID in pregnancy or to switch back to stand-alone insulin pumps, sensor-augmented pump (SAP) therapy, or multiple daily injections (MDIs). However, women are often hesitant to cease the use of AID during pregnancy.²⁵ For this narrative review, we review the current evidence on the use of AID in pregnancy and provide an overview of the completed and ongoing randomized controlled trials (RCTs) evaluating AID in pregnancy. In addition, we discuss the advantages and challenges of current AID for use in pregnancy and future directions for research.

AID Approved for Use in Pregnancy

The first feasibility studies evaluating AID with an algorithm developed for use in T1D pregnancy were performed more than ten years ago, confirming their safety in the second and third trimesters in small studies (n ≤ 12) (Table 1).^26,27

Table 1.

Results of the Smaller AID Studies in Pregnant Women With Type 1 Diabetes.

Author, year/country/registration number	Type of AID	Study design	Primary outcome	Main results
Murphy, 2011/UK / ISRCTN62568875²⁶	Insulin pump (Deltec CozmoV, Smiths Medical, St Paul, MN) with an MPC algorithm and FreeStyle Navigator CGM sensor (Abbott Diabetes Care, Alameda, CA, USA)	Phase Ia exploratory safety study, T1D, n = 10, HbA1c ≤ 86 mmol/mol (≤ 10%), in which AID was evaluated over 24 hours in early (12-16 weeks) and late pregnancy (28-32 weeks) in a supervised hospital clinical research facility setting	Mean glucose, TIRp, 3.5 to 7.8 mmol/L (63-140 mg/dL), TBRp < 3.5 mmol/L (63 mg/dL), TARp > 7.8 mmol/L (140 mg/dL)	AID was associated with near normoglycemia overnight, both in early (84%) and in late (100%) pregnancies (overnight TARp of 7% and 0% with TBRp of 0% and 0% in early and late pregnancies, respectively)
Murphy, 2011/UK / ISRCTN50385583²⁷	FreeStyle Navigator CGM sensor (Abbott Diabetes Care, Alameda, CA, USA) with an IV sampling catheter and study pump (Animas 2020; Johnson & Johnson, New Brunswick, NJ, USA), AID: Cambridge MPC algorithm	Phase Ib randomized crossover trial with 12 pregnant women with T1D, median HbA1c 46 mmol/mol (6.4%), comparing AID with conventional pump at two 24-hour visits, separated by a one- to six-week interval (at 19 and 23 weeks)	Plasma glucose TIRp, 3.5 to 7.8 mmol/L (63-140 mg/dL) from 14 h 00 on day 1 to 12 h 30 on day 2	TIRp was comparable for AID and conventional pump (81% vs 81%, P = .75) with less time spent extreme hypoglycemic 2.5 mmol/L (45 mg/dL) (0.0% vs 0.3%, P = .04)
Stewart, 2016/UK / ISRCTN71510001²⁸	DANA Diabecare R Insulin Pump^® (SOOIL) and the FreeStyle Navigator II (Abbott Diabetes Care), AID: Cambridge MPC algorithm	Phase II open-label, multicenter, randomized, crossover study comparing overnight AID therapy with SAP therapy each used during four weeks in 16 pregnant women with T1D, followed by a continuation phase in which the AID system was used day and night until delivery in 14 pregnant women with T1D	Overnight TIRp, 3.5 to 7.8 mmol/L (63-140 mg/dL) during each four-week study phase	Overnight AID therapy resulted in higher TIRp than SAP therapy: 74.7% vs 59.5%, P = .002. During the continuation phase, glucose levels were in the target range 68.7% of the time throughout pregnancy
Stewart, 2018/UK/ISRCTN83316328²⁹	CGM (FreeStyle Navigator II^®; Abbott Diabetes Care, Alameda, CA, USA) and insulin pump (DANA Diabecare R^®; Sooil, Seoul, Republic of Korea) with control Cambridge MPC algorithm (Florence D2A, version 0.3.41p; University of Cambridge, UK).	Phase II open-label, randomized, two-period crossover study in 16 pregnant women with T1D (GA 8-24 weeks, with HbA1c ≥ 48 to ≤ 86 mmol/mol [≥ 6.5% to≤ 10%]), 28 days of day and night AID and SAP insulin delivery. Continuation phase possible up to six weeks postpartum	TIRp, 3.5 to 7.8 mmol/L (63-140 mg/dL) during the 4-week intervention periods	AID was associated with comparable glucose levels (TIRp: 62.3% vs 60.1%, P = .47) and significantly less hypoglycemia than SAP therapy (< 63 mg/dL: 1.6% vs 2.7%, P = .02), less nocturnal hypoglycemia (2300-0700 hours) (1.1% vs 2.7%; P = .008)
Stewart, 2018 /UK/ISRCTN71510001 and ISRCTN83316328³⁰	CGM (FreeStyle Navigator 2^®; Abbott Diabetes Care, Alameda, CA, USA), insulin delivery was through a DANA-R pump^® (Diabecare, Sooil, Seoul, South Korea) and the AID systems used were the Florence D2W and Florence D2A (both University of Cambridge, Cambridge, UK). The control algorithm was on a tablet PC for the former and on an Android phone for the latter	Secondary analyses of all women included in the two randomized crossover trials who chose to continue using AID intrapartum and immediate postpartum period (27 [84.4%] of the potential 32 trial participants)	TIRp, 3.5 to 7.8 mmol/L (63-140 mg/dL) during pregnancy and 3.9 to 10 mmol/L (70-180 mg/dL) after delivery	Use of AID was associated with TIRp 82.0% (IQR = 49.3, 93.0) during labor and delivery. Postpartum TIR was 83.3% (IQR = 75.2, 94.6). There was no difference in maternal intrapartum glucose concentration between mothers of infants with and without neonatal hypoglycemia (6.9 mmol/L or 124 mg/dL and 6.8 mmol/L or 122 mg/dL, respectively, P = .84)
Ozaslan, 2022/USA/NCT04492566³¹	CGM, Dexcom G6 (Dexcom, Inc., San Diego, CA, USA) and a research insulin pump (Tandem Diabetes Care, San Diego, CA, USA) with zone-MPC algorithm. The devices are connected wirelessly to the iAPS installed on an unlocked study phone (Google Pixel 2 or 3a)	Multicenter observational pilot study evaluating the feasibility of a pregnancy-specific zone-MPC CL system in 11 pregnant women with T1D from GA 14 to 32 weeks already using SAP therapy, consisting of up to two weeks run-in at home, followed by a two-day AID insulin delivery period in a supervised outpatient setting	TIRp, 3.5 to 7.8 mmol/L (63-140 mg/dL) compared with participants’ one-week run-in period	There was an increased TIRp during supervised AID (81.5% vs 64%, P = .007) with less TARp (16.5% vs 30.8%, P = .029) and TBRp (2.0% vs 5.2%, P = .039)
Levy, 2023/USA/NCT04492566³²	CGM, Dexcom G6 (Dexcom, Inc., San Diego, CA, USA) and a research insulin pump (Tandem Diabetes Care, San Diego, CA, USA) with zone-MPC algorithm. The devices are connected wirelessly to the iAPS installed on an unlocked study phone (Google Pixel 2 or 3a)	Multicenter observational pilot study evaluating at-home use of a pregnancy-specific zone-MPC AID in ten pregnant women with T1D already using SAP therapy from GA 14 to 32 weeks until delivery	TIRp, 3.5 to 7.8 mmol/L (63-140 mg/dL) compared with run-in period	TIRp increased by 14.1% (run-in: 64.5% vs AID: 78.6%, P = .002), overnight TIRp increased by 23.5% (run-in: 61.3% vs AID: 84.8%, P = .005), lower TBRp < 3.5 mmol/L (63 mg/dL) (run-in: 3.7% vs AID: 1.57%, P = .037)

Abbreviations: AID, automated insulin delivery; MPC, model predictive controller; T1D, type 1 diabetes; TARp, time above range in pregnancy; TBRp, time below range in pregnancy; CGM, continuous glucose monitoring; TIRp, time in range in pregnancy; HbA1c, glycated hemoglobin; GA, gestational age; SAP, sensor-augmented pump; PC, personal computer; IQR, interquartile range; C-section, cesarean section; TIR, time in range; iAPS, interoperable artificial pancreas system.

This table was updated based on Benhalima et al.²⁴

Two crossover RCTs with this algorithm, confirmed proof of concept of the CamAPS^® FX AID (CamDiab Ltd. Cambridge, UK) in pregnancy (Table 1) with improvement of TIRp overnight and less hypoglycemia when using the AID compared with SAP therapy. These RCTs had, however, only small sample sizes (16 women) and limited duration of AID use of four weeks.^28,29 A secondary analysis of these two crossover RCTs (including the continuation phase) reported a mean TIR of 82.0% during delivery and 83.3% postpartum.³⁰ The AID performed well, both in vaginal and cesarean section deliveries without hypoglycemia.³⁰ In general, women expressed satisfaction with improved glucose control and increased peace of mind, and they also reported concerns about CGM accuracy and burden of maintenance requirements.³³ Most women became more positive with the technology throughout pregnancy.³⁴ However, the insulin pump used in these crossover RCTs was less sophisticated than the current commercially available insulin pumps and the algorithm was housed on a tablet, which might have led to more device burden.²⁴

Based on these pilot RCTs,^28,29 the CamAPS^® FX AID has been approved for use in pregnancy in the United Kingdom, Europe, and Australia, and to date, this is the only AID approved for use in pregnancy (Table 2). The CamAPS^® has been available since 2020, and is the first interoperable AID android app. It is compatible with different insulin pumps: YpsoPump, with mylife loop (Ypsomed AG, Burgdorf, Switzerland), DANA Diabecare RS and DANA-IIs insulin pumps (SOOIL Development Gyeonggi-do, Korea) with the Dexcom G6/G7 (Dexcom, San Diego, CA, USA) and Free Style Libre 3 sensors (Abbott Inc., Abbott Park, IL, USA).^35,36 Both rapid and ultra-rapid insulin analogues can be used.³⁷

Table 2.

Overview of the Different Commercially Available AID Systems.

AID system	Algorithm and glucose targets	Approval for pregnancy	Lowest possible glucose settings for pregnancy
Medtronic MiniMed 670/770G^®	Treat to target proportional integral derivative technology with insulin feedback with Guardian 3 CGM), non-customizable glucose target 6.7 mmol/L (120 mg/dL), temporary target of 8.3 mmol/L (150 mg/dL)	Not approved	Non-customizable glucose target 6.7 mmol/L (120 mg/dL), active insulin time of two hours
Medtronic MiniMed 780G^®	Proportional-integral-derivative technology with insulin feedback with the most advanced SmartGuard™ technology, with Guardian 3 or 4 CGM, customizable glucose target 5.5 to 6.7 mmol/L (100-120 mg/dL), in 0.5 mmol/L (10 mg/dL) increments, temporary target of 8.3 mmol/L (150 mg/dL)	Not approved	Glucose target of 5.5 mmol/L (100 mg/dL), active insulin time of two hours, Adapt CI ratios as needed. If safe meal bolus^a occurs when lowering meal CI ratios, increase CI ratios and add extra (“fake”) carbs with meals
Tandem t:slimX2 with Control IQ^® technology	Treat to range predictive algorithm, currently with Dexcom G6 CGM Non-customizable target range options responsive to user basal rate and sensitivity adjustments. Sleep activity range, 6.2 to 6.7 mmol/L (112.5-120 mg/dL), which can be used 24 hours a day with optional exits to regular control IQ activity target range of 6.2 to 8.8 mmol/L (112.5-160 mg/dL), or exercise activity range of 7.7 to 8.8 mmol/L (140-160 mg/dL)	Not approved	Use sleep mode day and night, adapt basal rate, and CI ratios as needed, lower insulin sensitivity as needed
CamAPS FX ^®	Cambridge treat-to-target adaptive MPC with mylife YpsoPump^®, DANA Diabecare RS^®, and DANA-i^® insulin pumps (with Dexcom G6/G7 or Freestyle Libre 3 CGM) offer fully customizable glucose targets between 4.4 and 11.1 mmol/L (80-200 mg/dL). Ease-off function to temporarily increase glucose target with 2.5 mmol/L (45 mg/dL) and stop insulin delivery when glucose < 7.7 mmol/L (< 139 mg/dL); Boost function to temporarily increase insulin delivery with 35% (until target is reached)	Approved in > 13 countries (UK, across Europe, and Australia)	Personalized glucose target recommended setting of 5.5 mmol/L (100 mg/dL) before 16-week gestation and 4.4 to 5.0 mmol/L (80-90 mg/dL) from 16 weeks until delivery, adapt CI ratios as needed
Diabeloop^®	DBLG1 (a machine learning-based algorithm Diabeloop Generation 1 with Dexcom G6 CGM), customizable glucose targets between 5.5 and 7.2 mmol/L (100-130 mg/dL). Following features can be adapted: aggressiveness in hyperglycemia (conditions the administered boluses when glucose > 10 mmol/L [180 mg/dL]), aggressiveness in normoglycemia (conditions the basal rate provided by the system when glucose values are between 3.8 and 10 mmol/L [70-180 mg/dL]), and aggressiveness at breakfast, lunch, and dinner (conditions the amount of bolus in each of the intakes). CI ratios and insulin sensitivity cannot be adapted	Not approved	Lowest glucose target 5.5 mmol/L (100 mg/dL). Increase aggressiveness in hyperglycemia, normoglycemia and at meals as needed
Omnipod 5^® (SmartAdjust, with Dexcom G6 CGM)	Model predictive controller algorithm with customizable glucose targets between 6.1 and 8.3 mmol/L (110-150 mg/dL) in 0.5 mmol/L (10 mg/dL increments)	Not approved	Lowest glucose target of 6.1 mmol/L (110 mg/dL), adapt active insulin time, CI ratios, and insulin sensitivity as needed, turn off reverse correction^b

Abbreviations: AID, automated insulin delivery; CGM, continuous glucose monitoring; CI, carbohydrate-to-insulin ratio; MPC, model predictive controller; T1D, type 1 diabetes.

Safe meal bolus refers to a meal bolus amount that is adjusted downward due to various factors the algorithm considers, such as hypoglycemia risk prediction within the next four hours, active insulin and total daily insulin dose.

Reverse correction refers to a calculated dose of insulin lower than that strictly derived from the carbohydrate-to-insulin ratio when the glucose level is below target range at the time of the carbohydrate announcement.

The CamAPS^® FX AID contains a treat-to-target adaptive model predictive controller (MPC) algorithm. The algorithm is initialized using total daily insulin dose and bodyweight. Insulin sensitivity and active insulin time are automatically calculated and adjusted over time by the algorithm. Adaptive learning is incorporated with regards to total daily insulin requirements, diurnal variations, and meal patterns. It also contains an optional exercise mode (Ease-off function), which temporarily raises the glucose target and suspends algorithm-directed insulin delivery if sensor glucose is lower than 7·0 mmol/L or 140 mg/dL. In addition, this AID contains a Boost function to intensify algorithm-driven insulin delivery by approximately 35% when glucose is above target level.³⁸

Based on the experiences with the pilot RCTs,^28,29 this AID was updated since 2018. Glucose measurements from CGM can now be used to inform user-initiated premeal boluses.³⁹ In addition, the glucose target can be intensified or relaxed. In contrast to other available commercial AID, the CamAPS^® FX offers fully customizable glucose targets between 4.4 mmol/L or 80 mg/dL and 11.1 mmol/L or 200 mg/dL. During pregnancy, targets are typically set at 5.5 mmol/L or 100 mg/dL during the first trimester and 4.4 to 5.0 mmol/L or 80 to 90 mg/dL thereafter (Table 2).³⁹

Other Pilot Studies in Pregnancy

A consortium from the United States designed a zone-MPC algorithm specifically customized for use in pregnancy, allowing for lower glycemic targets during the day (4.4-6.0 mmol/L or 80-110 mg/dL) and during the night (4.4-5.5 mmol/L or 80-100 mg/dL)⁴⁰ compared with most commercially available AID. The zone-MPC algorithm runs on an interoperable artificial pancreas system, which consists of a CGM, an insulin pump (Tandem research pump), and an unlocked smartphone.⁴¹ The optimal insulin delivery plan is updated every five minutes based on the glucose data from the CGM. Meals need to be announced to the system by the user, and meal boluses are calculated based on the participants’ prescribed bolus settings (ie, carbohydrate ratio and insulin sensitivity factor) and user-estimated carbohydrate intake. Two small observational studies in women using this algorithm showed a significant increase in TIRp of 14% to 17% (Table 1).^31,32 The question remains whether AID with a customized pregnancy algorithm should preferably be used, as illustrated by the promising results of these small pilot studies.^31,32 Since this algorithm allows for lower glycemic targets compared with most other commercially available AID, this might be particularly useful for the stricter glycemic targets needed during pregnancy. However, larger and longer-duration trials are warranted.

Off-Label Use During Pregnancy

The MiniMed 670G^® and the next version 770G^® (Medtronic Inc., Northridge, CA, USA) were the first commercially available AID. The limitation is that their algorithms target an average glucose of 6.7 mmol/L or 120 mg/dL, which is not customizable to lower glucose levels and is higher than the recommended target for use in pregnancy⁴² (Table 2). A few case reports and case series reported improvement in TIR with no pregnancy complications nor other safety issues, in pregnant women with T1D switching to this AID during pregnancy (often starting in auto mode later on in pregnancy).^43-45 Although glycemic control improved, the TIRp was often not reached.^43-45

Several next generation AID, such as the MiniMed 780G^®, Tandem t:slim X2 Control IQ^®, Diabeloop^® and Omnipod 5^® systems received approval for use in non-pregnant people in 2020, but are not approved for use in pregnancy.^46-48 These AIDs allow for a lower glucose target, which might potentially lead to improved glycemia compared with the MiniMed 670G/770G^® systems (Table 2). The MiniMed 780G^® uses an algorithm that automatically adapts the basal rate and also provides automated hyperglycemia correction boluses. This type of AID might be particularly useful for women who prefer an AID without the need for much intervention. In contrast, with the Tandem t:slim X2 with control IQ^®, the basal rate and insulin sensitivity can be adapted, which might be a good option for women who prefer an AID that is less automated and allows for more personal intervention.²⁴ A case report was presented on the use of the Diabeloop^® (Grenoble, France) in a pregnant T1D patient with highly unstable glycemic control and severe hypoglycemia unawareness.⁴⁹ Both daytime and nighttime periods in target range were steadily improved, with a mean TIRp of 82.1% for the daytime and 77.1% for the nighttime from 13 weeks until delivery. The Omnipod 5^® is the first patch pump that is available as an AID. However, there are currently no data on the use of the Omnipod 5^® in pregnancy. Recent case series on the use of the MiniMed 780G^® and Tandem t:slim X2 control IQ^® in pregnancy show, respectively, that women spent more time in range and had lower glucose variability compared with their previous pregnancy using SAP therapy,⁵⁰ and that control IQ technology could be safely used in pregnancy with increased TIRp and less diabetes management burden.⁵¹ In contrast, a recent report of six pregnant women with a baseline HbA1c of 6.6% using the MiniMed 780G^® throughout pregnancy, indicated challenges in achieving the desired glycemic targets for pregnancy in most women.⁵² Differences between these case series, could be related to differences in baseline characteristics, such as glycemic control at the off-set of pregnancy. It could be expected that AID can more easily increase time spent in target glucose range in women starting with a higher baseline HbA1c in early pregnancy, compared with women with already tight glycemic control on SAP therapy. Data from RCTs are therefore needed to evaluate whether AID can provide additional benefit for glycemic control compared with SAP therapy.

Data on do-it-yourself artificial pancreas systems are limited to case reports in selected women with T1D and strong diabetes self-management skills, showing that these systems might be used during pregnancy with improvements in quality of life.^53-56 Insulin pump settings should be adapted similarly to that of conventional insulin pumps to ensure safety in case of loss of AID. However, since these systems are unregulated with no formal technical support, their use is not recommended during pregnancy.

RCTs on AID During Pregnancy or Postpartum

Three open-label RCTs have recently been completed^39,57,58 and one is still ongoing,⁵⁹ to determine the clinical efficacy and safety of AID in T1D pregnancy. These studies are powered for maternal glycemic outcomes, with the primary outcome TIRp of 3.5 to 7.8 mmol/L or 63 to 140 mg/dL. Secondary outcomes include several pregnancy and neonatal outcomes, adverse events (including severe hypoglycemia and diabetic ketoacidosis), and patient-reported outcomes (including treatment satisfaction and quality of life).²⁴ In addition, one small RCT evaluates AID in early postpartum⁶⁰ (Table 3).

Table 3.

The Larger (Ongoing) AID Studies in Pregnant Women With Type 1 Diabetes.

Name of the study, registration number	Intervention vs comparator	Countries, centers, study end	Study design	Primary outcome(s)	(Main) secondary outcomes	Main results
AiDAPT (automated insulin delivery among pregnant women with type 1 diabetes) study, ISRCTN56898625 ^39,61	CamAPS^® FX AiD (Dana Diabecare RS + Dexcom G6 vs CSII or MDI) with Dexcom G6	UK, 9 centers, November 2022	Multicenter, open-label, RCT N = 124 1:1 randomization with stratification by site Eligibility: GA < 14 weeks, HbA1c ≥ 48 mmol/mol (≥ 6.5%) and ≤ 86 mmol/mol (≤ 10%), 18 to 45 years, T1D diagnosis ≥ 1 year, intensive insulin therapy Run-in phase of masked Dexcom G6 for baseline data collection	TIRp, 3.5 to 7.8 mmol/L (63-140 mg/dL) between 16 weeks until delivery	HbA1c; DKA; severe hypoglycemia episodes; mode of delivery; GA at delivery; infant birth weight; LGA; SGA; neonatal morbidity; NICU admission; pregnancy loss < 24 weeks; stillbirth; neonatal death; fear of hypoglycemia, health status, diabetes distress; sleep quality; cost-effectiveness	AID led to a significant increase in TIRp of 10.5% (mean TIR 68.2% ± 10.5% in the AID group vs 55.6% ± 12.5% in the standard of care group), less TARp (difference of 10.2%), higher TIRp overnight (difference of 12.3%), without increase in TBRp and without unanticipated safety problems. Pregnancy outcomes were similar except for a significantly lower rate of hypertensive disorders and less gestational weight gain in the intervention group vs comparator group
CRISTAL(closed-loop insulin delivery in pregnant women with type 1 diabetes: a randomized controlled trial) study, NCT0452097⁵⁸	MiniMed 780G^® system with Guardian 3 or 4 vs continue SoC (MDI or open-loop pump) with any type of sensor)	Belgium, 11 centers and the Netherlands, 1 center, May 2023	Multicenter, open-label RCT N = 95 1:1 randomization with stratification by center, HbA1c and treatment (pump or MDI), GA < 14 weeks Eligibility: GA < 12 weeks, HbA1c ≤ 86 mmol/mol (≤ 10%), 18 to 45 years, T1D diagnosis ≥ 1 year, intensive insulin therapy Run-in phase of ten days with masked Guardian 3 CGM as needed	TIRp, 3.5 to 7.8 mmol/L (63-140 mg/dL) at GA 14 to 17, 20 to 23, 26 to 29, 33 to 36 weeks	Prespecified: time < 3.5 mmol/L (63 mg/dL); overnight time < 3.5 mmol/L (63 mg/dL); overnight TIRp, 3.5 to 7.8 mmol/L (63-140 mg/dL) Exploratory: device deficiencies; time > 7.8 mmol/L (140 mg/dL); > 10 mmol/L (180 mg/dL); time < 2.8 mmol/L (50 mg/dL); < 3.0 mmol/L (54 mg/dL); severe hypo’s; DKA; miscarriage; stillbirth; neonatal death; LGA; respiratory distress; birth trauma; shoulder dystocia; neonatal hypoglycemia; NICU admission; fear of hypoglycemia, health status, diabetes-related emotional distress, treatment satisfaction, cost-effectiveness	Expected in 2024
CIRCUIT(closed-loop insulin delivery by glucose responsive computer algorithms in type 1 diabetes pregnancies) study, NCT04902378⁵⁹	Tandem t:slim X2 pump with control IQ^® with Dexcom G6 vs SoC (MDI or pump) with Dexcom G6	Canada, nine centers + Australia, five centers, 2026	Multicentric open-label RCT N = 90 Eligibility: GA < 14 weeks, HbA1c ≥ 48 mmol/mol-< 86 mmol/mol (≥ 6.5 to < 10%), 18 to 45 years, T1D diagnosis ≥ 1 year, intensive insulin therapy Run-in phase of one week with Dexcom G6 CGM	TIRp, 3.5 to 7.8 mmol/L (63-140) mg/dL, GA, 16 to 34 weeks of gestation	TARp; TBRp; mean blood glucose; Severe hypoglycemia; DKA; fear of hypoglycemia, fear of hyperglycemia,; sleep quality; work productivity, preeclampsia; gestational hypertension; worsening chronic hypertension events; C-sections; preterm delivery; LGA; SGA; mean neonatal birth weight; neonatal hypoglycemia; NICU admission; pregnancy loss	Expected in 2026
PICLS (pregnancy intervention with a closed-loop system) study, NCT03774186⁵⁷	MiniMed 670G^® vs SAP therapy	USA, two centers March 2022	Open-label, pilot RCT 1:1 randomization 14 to 18 weeks N = 24, aim for 20 completers Eligibility: women GA ≤ 11 weeks, HbA1c 37 to 75 mmol/mol (5.5%-9%), 18 to 45 years, T1D diagnosis > 1 year, intensive insulin therapy (MDI or CSII) Run-in phase of seven days with CGM	Severe hypoglycemic episodes Time < 3.0 mmol/L (54 mg/dL), time < 3.5 mmol/L (63 mg/dL), 3.5 to 7.8 mmol/L (63-140 mg/dL), > 7.8 mmol/L (140 mg/dL) and > 10 mmol/L (180 mg/dL)	DKA; duration of hypoglycemic episodes; health status, device satisfaction, preeclampsia; C-sections; average gestational weight gain; miscarriage or stillbirth; LGA; neonatal hypoglycemia	Expected end 2023/early 2024
CLIMB(closed-loop insulin delivery postpartum in mothers with type 1 diabetes and their babies’ feeding practices) study⁶⁰	MiniMed 670G/770G^® with early (6-10 days) postpartum vs delayed (12 weeks postpartum) auto mode	Canada, four centers, March 2023	Pilot open-label RCT N = 18 Randomization after given birth Eligibility: GA 12 to 32 weeks, HbA1c < 85 mmol/mol (< 9.9%), 18 to 45 years, T1D diagnosis ≥ 2 years, intensive insulin therapy	Effect of auto mode vs manual mode of MiniMed 670G/770G on TIR 3.9 to 10 mmol/L (70-180 mg/dL) 12 weeks postpartum	TBR; TAR; glycemic variability; diabetes-related distress, fear of hypoglycemia, sleep quality, feeding practices	AID compared with SAP: no difference in TIR (non-pregnancy) 79.2 ± 8.7% vs 78.2% ± 6.0%); less time < 3.9 mmol/L (70 mg/dL) 1.7 ± 0.8% vs 5.5 ± 3.3%, P < .001, less time < 3.0 mmol/L (54 mg/dL) 0.3 ± 0.2% vs 1.1 ± 0.9%, P = .008, no episodes of severe hypoglycemia

Abbreviations: AID, automated insulin delivery; CSII, continuous subcutaneous insulin infusion; DKA: diabetic ketoacidosis; MDI, multiple daily injections; RCT, randomized controlled trial; GA, gestational age; T1D, type 1 diabetes; HbA1c, glycated hemoglobin; LGA, large-for-gestational age; SGA, small-for-gestational age; NICU, neonatal intensive care unit; SoC: standatd of care threatment; TIRp, time in pregnancy range; TARp, time above range in pregnancy; TBRp, time below range in pregnancy; C-sections, cesarean sections; CGM, continuous glucose monitoring; SAP, sensor-augmented pump.

This table was updated based on Benhalima et al.²⁴

The AiDAPT trial investigated the updated CamAPS^® with Dexcom G6 compared with standard of care in 124 women with an HbA1c between 48 and 86 mmol/mol (6.5%-10%).⁶¹ The AID led to a significant increase in TIRp of 10.5% without increased TBRp.³⁹ Pregnancy outcomes were similar except for a significantly lower rate of hypertensive disorders and less gestational weight gain in the AID group.³⁹ The study was not powered for pregnancy outcomes and women were randomized at about 11 weeks, thus missing the crucial period of organogenesis. Interviews with women using the AID in pregnancy, highlight that it can facilitate positive pregnancy experiences, with less fear of complications and better sleep quality. Women also indicated the need for intensive support during early pregnancy to develop confidence in the AID technology.⁶²

The CRISTAL study, including 95 women, compared the MiniMed 780G^® system and Guardian 3 or 4 sensor with routine care. The study allows for the use of all types of CGM in line with routine care and with no lower HbA1c limit for inclusion, thus enabling the evaluation of women with tight glycemic control in early pregnancy. The study is completed and results are expected in 2024.⁵⁸ The CIRCUIT study, including 90 women, compares the Tandem t:slim X2 pump with control IQ^® and Dexcom G6 vs routine care with Dexcom G6 sensor⁵⁹ (Table 3). Results are expected in 2026.

The PICLS (pregnancy intervention with a closed-loop system) study is a pilot RCT on 24 women with T1D investigating the MiniMed 670G^® compared with SAP therapy in pregnancy.⁵⁷ The study was completed in 2023 but is not yet published.

The CLIMB (closed-loop insulin delivery postpartum in mothers with type 1 diabetes and their babies’ feeding practices) study was a pilot RCT followed by an observational cohort study in 18 women, of the MiniMed 670G/770G^® system with early (6-10 days postpartum) versus delayed (12 weeks postpartum) use of the auto mode.⁶⁰ Participants in the AID group had less hypoglycemia compared with the SAP group (Table 3).⁶⁰ These findings are reassuring for the use of AID postpartum. However, the AID was used after the first week postpartum. Studies exploring the safety of AID in the first week after delivery are needed.

The RCTs on AID in pregnant women with T1D will report on whether AID can improve glycemic management. Several studies will also investigate the cost-effectiveness of AID in women with T1D during pregnancy.^58,61

Advantages and Challenges of Current AID in Pregnancy

Although many AID are not specifically designed for use in pregnancy, these systems may also offer several advantages, such as less glycemic variability, better TIRp overnight, less risk for hypoglycemia, and improved treatment satisfaction.²⁴ Due to the magnitude in increase in TIRp, it can be anticipated that if the CamAPS^® AID is started at preconception or very early in pregnancy, this will probably also lead to reduced rates of other complications, such as congenital anomalies and large-for-gestational age (LGA) infants in women with higher baseline HbA1c.

The challenges of most commercially available AID are that the glucose target cannot be lowered < 5.5 mmol/L or 100 mg/dL (Table 2). The glycemic target might therefore not be strict enough for pregnancy (Figure 1). In addition, the algorithms designed for non-pregnant populations, might not be sufficiently flexible to adapt fast enough to the increased insulin need later in pregnancy. When using AID off-label in pregnancy, it is important to provide counseling on the potential risks and benefits to women, and offer an assisted AID approach.²⁵ The lowest target glucose available for the AID should be used (Table 2). Until now, studies have introduced women to AID in early pregnancy ^39,57-59 or postpartum,⁶⁰ and none initiated AID before pregnancy. Some women may be reluctant to initiate new diabetes technology in pregnancy, suggesting that study participants might be highly selected.

Figure 1.

Challenges and advantages of available automated insulin delivery (AID) systems for use in pregnancy. This Figure was updated based on Benhalima et al.²⁴ RCT, randomized controlled trial.

To facilitate the real-world adoption of AID in pregnant women with T1D, presenting AID to pregnant women (or preferably before pregnancy) and healthcare professionals as a pillar of a three-party collaboration may help to promote optimal use.⁶³ It is also important to develop mentorship services to upskill and support inexperienced staff attending pregnant women using AID.⁶⁴ However, for broad implementation of AID in pregnancy, improved equity in access to AID worldwide is needed. Based on the AiDAPT trial, the “National Institute for Health and Care Excellence” guidance (under review) recommends that AID should be offered to all pregnant persons with T1D.³⁹

Future Research

To have strong evidence on the impact of AID on pregnancy outcomes, a large RCT would be needed powered for maternal and neonatal outcomes with initiation of the AID before pregnancy throughout pregnancy and delivery, and with follow-up postpartum to also evaluate the impact of breastfeeding. However, it would be very challenging to perform such a large study as the number of women of reproductive age already using AID before pregnancy is increasing and due to the advantages associated with AID, this will potentially limit the willingness for recruitment.

As none of the current RCTs are powered for pregnancy outcomes (Table 3), a meta-analysis is needed to evaluate the impact of AID use on obstetric and neonatal outcomes. In addition, this meta-analysis should also include participant-reported outcomes to determine the psychosocial impact of AID use in T1D during pregnancy.²⁴ However, existing standardized questionnaires do not adequately capture important participant-reported outcomes, such as sleep quality, fear of hypoglycemia and hyperglycemia during pregnancy and fear of hypoglycemia postpartum.⁶⁵

More research is also needed to determine whether the current proposed consensus glucose targets for use in pregnancy are adequate or if the glucose targets should be adjusted as AID become more broadly available for use in T1D during pregnancy.¹⁵ When using AID in pregnancy, TIRp might be more easily achieved, allowing for a lower upper target of 6.6 mmol/L or 120 mg/dL instead of 7.8 mmol/L or 140 mg/dL. Until now, there is no consensus on an appropriate target for mean glucose in pregnancy, even though this metric might help to further improve pregnancy outcomes as recent studies in normal glucose tolerant women have demonstrated that mean glycemia is around 5.4 mmol/L or 98 mg/dL throughout pregnancy.^66,67

More research is also needed on the timing of the bolus when using AID in pregnancy. Due to the automatic adaptation of the basal rate, bolus should be given a minimum of 10 to 15 minutes before meals, and there may be less of a need to administer the bolus up to 30 to 45 minutes before meals in late pregnancy, as recommended when using MDI or insulin pumps.⁶ The response should be individualized based on each woman and system used.²⁴

Conclusion

Glycemic management in T1D during pregnancy remains challenging with risk of pregnancy complications. The CamAPS^® FX is the first AID approved for use in pregnancy, demonstrating a 10% increase in TIRp in a population with a baseline HbA1c ≥ 48 mmol/mol (6.5%). Off-label use of AID not approved for pregnancy, are currently also being evaluated in RCTs. More research is needed on the impact of AID on maternal and neonatal outcomes as well as during delivery and postpartum. This will help to guide clinicians and women with T1D on the use of AID during pregnancy.

Footnotes

Abbreviations

AID, automated insulin delivery; CL, closed-loop; MDI, multiple daily injections; T1D, type 1 diabetes; RCT, randomized controlled trial; SAP therapy, sensor-augmented pump therapy; MPC, model predictive controller; CGM, continuous glucose monitoring; HbA1c, glycated hemoglobin; TIRp, time in range in pregnancy; TBRp, time below range in pregnancy; TARp, time above range in pregnancy

Declaration of Conflicting Interests

The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: KB received an unrestricted grant and study devices from Medtronic for the investigator-initiated CRISTAL study. KB received study devices from Dexcom for the investigator-initiated GLORIA study. KB received study medication from Novo Nordisk A/S for the investigator-initiated SERENA study. KB received speaker fee from Novo Nordisk, AstraZeneca and Medtronic. KaatB has no disclosures. LR received a grant from Novo Nordisk A/S for an investigator sponsored study. Grant number: U1111-1209-6358 and participated in the European Association for the Study of Diabetes annual meeting 2023 as an invite by Novo Nordisk A/S. JJ has been a lecturer/member of the scientific advisory board in the following companies: Abbott, AstraZeneca, Boehringer Ingelheim, Eli Lilly, Medtronic, Nordic InfuCare, NovoNordisk A/S, and Sanofi. KB received a fundamental clinical investigatorship from FWO Flanders (1800220N).

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Katrien Benhalima

References

Colstrup

Mathiesen

Damm

Jensen

Ringholm

. Pregnancy in women with type 1 diabetes: have the goals of St. Vincent declaration been met concerning foetal and neonatal complications? J Matern Fetal Neonatal Med. 2013;26(17):1682-1686.

Morrens

Verhaeghe

Vanhole

Devlieger

Mathieu

Benhalima

. Risk factors for large-for-gestational age infants in pregnant women with type 1 diabetes. BMC Pregnancy Childbirth. 2016;16(1):162.

Murphy

Howgate

O’Keefe

, et al. Characteristics and outcomes of pregnant women with type 1 or type 2 diabetes: a 5-year national population-based cohort study. Lancet Diabetes Endocrinol. 2021;9(3):153-164.

Evers

ter Braak

de Valk

van Der Schoot

Janssen

Visser

. Risk indicators predictive for severe hypoglycemia during the first trimester of type 1 diabetic pregnancy. Diabetes Care. 2002;25(3):554-559.

Nielsen

Pedersen-Bjergaard

Thorsteinsson

Johansen

Damm

Mathiesen

. Hypoglycemia in pregnant women with type 1 diabetes: predictors and role of metabolic control. Diabetes Care. 2008;31(1):9-14.

Murphy

Elleri

Allen

, et al. Pathophysiology of postprandial hyperglycaemia in women with type 1 diabetes during pregnancy. Diabetologia. 2012;55(2):282-293.

Feig

Asztalos

Corcoy

, et al. CONCEPTT: continuous glucose monitoring in women with type 1 diabetes in pregnancy trial: a multi-center, multi-national, randomized controlled trial–Study protocol. BMC Pregnancy Childbirth. 2016;16(1):167.

Feig

Donovan

Corcoy

, et al. Continuous glucose monitoring in pregnant women with type 1 diabetes (CONCEPTT): a multicentre international randomised controlled trial. Lancet. 2017;390(10110):2347-2359.

Kristensen

Ögge

Sengpiel

, et al. Continuous glucose monitoring in pregnant women with type 1 diabetes: an observational cohort study of 186 pregnancies. Diabetologia. 2019;62(7):1143-1153.

10.

Flash glucose monitoring: national arrangements for funding of relevant diabetes patients, https://www.england.nhs.uk/wp-content/uploads/2019/03/National-arrangements-for-funding-of-relevant-diabetes-patients-June-2020-Updated-final.pdf. Accessed November 6, 2022.

11.

Health Improvement Scotland/SHTG Advice on Health Technologies. What is the clinical and cost effectiveness of FreeStyle Libre® Flash glucose monitoring for patients with diabetes mellitus treated with intensive insulin therapy? https://shtg.scot/media/1799/freestyle-libre-flash-glucose-monitoring-shtg-evidence-note-81-endocrinology-and-diab.pdf. Accessed November 6, 2022.

12.

Nørgaard

Mathiesen

Nørgaard

Ringholm

. Comparison of glycemic metrics measured simultaneously by intermittently scanned continuous glucose monitoring and real-time continuous glucose monitoring in pregnant women with type 1 diabetes. Diabetes Technol Ther. 2021;23(10):665-672.

13.

Rijksinstituut voor Ziekte-En Invaliditeitsverzekering, https://www.riziv.fgov.be/nl/themas/kost-terugbetaling/ziekten/endocriene-metabole/Paginas/diabetes-tegemoetkoming-kosten-begeleiding-volwassenen-gespecialiseerd-centrum.aspx. Accessed October 19, 2022.

14.

Goldberg

Ursing

Ekéus

Wiberg-Itzel

. Swedish guidelines for type 1 diabetes and pregnancy outcomes: a nationwide descriptive study of consensus and adherence. Prim Care Diabetes. 2021;15(6):1040-1051.

15.

Battelino

Danne

Bergenstal

, et al. Clinical targets for continuous glucose monitoring data interpretation: recommendations from the international consensus on time in range. Diabetes Care. 2019;42(8):1593-1603.

16.

O’Malley

Ozaslan

Levy

, et al. Longitudinal observation of insulin use and glucose sensor metrics in pregnant women with type 1 diabetes using continuous glucose monitors and insulin pumps: the LOIS-P study. Diabetes Technol Ther. 2021;23(12):807-817.

17.

Tundidor

Meek

Yamamoto

, et al. Continuous glucose monitoring time-in-range and HbA1c targets in pregnant women with type 1 diabetes. Diabetes Technol Ther. 2021;23(10):710-714.

18.

Scott

Murphy

Kristensen

, et al. Continuous glucose monitoring metrics and birth weight: informing management of type 1 diabetes throughout pregnancy. Diabetes care. 2022;45:1724-1734.

19.

Phillip

Nimri

Bergenstal

, et al. Consensus recommendations for the use of automated insulin delivery (AID) technologies in clinical practice. Endocr Rev. 2022;44:254-280.

20.

Phillip

Nimri

Bergenstal

, et al. Consensus recommendations for the use of automated insulin delivery technologies in clinical practice. Endocr Rev. 2023;44(2):254-280.

21.

Weisman

Bai

Cardinez

Kramer

Perkins

. Effect of artificial pancreas systems on glycaemic control in patients with type 1 diabetes: a systematic review and meta-analysis of outpatient randomised controlled trials. Lancet Diabetes Endocrinol. 2017;5(7):501-512.

22.

Leelarathna

Choudhary

Wilmot

, et al. Hybrid closed-loop therapy: where are we in 2021. Diabetes Obes Metab. 2021;23(3):655-660.

23.

Boughton

Hovorka

. New closed-loop insulin systems. Diabetologia. 2021;64(5):1007-1015.

24.

Benhalima

Beunen

Siegelaar

, et al. Management of type 1 diabetes in pregnancy: update on lifestyle, pharmacological treatment, and novel technologies for achieving glycaemic targets. Lancet Diabetes Endocrinol. 2023;11(7):490-508.

25.

Szmuilowicz

Levy

Buschur

Polsky

. Expert guidance on off-label use of hybrid closed-loop therapy in pregnancies complicated by diabetes. Diabetes Technol Ther. 2023;25(5):363-373.

26.

Murphy

Elleri

Allen

, et al. Closed-loop insulin delivery during pregnancy complicated by type 1 diabetes. Diabetes Care. 2011;34(2):406-411.

27.

Murphy

Kumareswaran

Elleri

, et al. Safety and efficacy of 24-h closed-loop insulin delivery in well-controlled pregnant women with type 1 diabetes: a randomized crossover case series. Diabetes Care. 2011;34(12):2527-2529.

28.

Stewart

Wilinska

Hartnell

, et al. Closed-loop insulin delivery during pregnancy in women with type 1 diabetes. N Engl J Med. 2016;375(7):644-654.

29.

Stewart

Wilinska

Hartnell

, et al. Day-and-night closed-loop insulin delivery in a broad population of pregnant women with type 1 diabetes: a randomized controlled crossover trial. Diabetes Care. 2018;41(7):1391-1399.

30.

Stewart

Yamamoto

Wilinska

, et al. Adaptability of closed loop during labor, delivery, and postpartum: a secondary analysis of data from two randomized crossover trials in type 1 diabetes pregnancy. Diabetes Technol Ther. 2018;20(7):501-505.

31.

Ozaslan

Levy

Kudva

, et al. Feasibility of closed-loop insulin delivery with a pregnancy-specific zone model predictive control algorithm. Diabetes Technol Ther. 2022;24(7):471-480.

32.

Levy

Kudva

Ozaslan

, et al. At-home use of a pregnancy-specific zone-MPC closed-loop system for pregnancies complicated by type 1 diabetes: a single-arm, observational multicenter study. Diabetes Care. 2023;46(7):1425-1431.

33.

Farrington

Stewart

Hovorka

Murphy

. Women’s experiences of day-and-night closed-loop insulin delivery during type 1 diabetes pregnancy. J Diabetes Sci Technol. 2018;12(6):1125-1131.

34.

Farrington

Stewart

Barnard

Hovorka

Murphy

. Experiences of closed-loop insulin delivery among pregnant women with type 1 diabetes. Diabet Med. 2017;34(10):1461-1469.

35.

Thabit

Tauschmann

Allen

, et al. Home use of an artificial beta cell in type 1 diabetes. N Engl J Med. 2015;373(22):2129-2140.

36.

Bally

Thabit

Kojzar

, et al. Day-and-night glycaemic control with closed-loop insulin delivery versus conventional insulin pump therapy in free-living adults with well controlled type 1 diabetes: an open-label, randomised, crossover study. Lancet Diabetes Endocrinol. 2017;5(4):261-270.

37.

Boughton

Hartnell

Thabit

, et al. Hybrid closed-loop glucose control with faster insulin aspart compared with standard insulin aspart in adults with type 1 diabetes: a double-blind, multicentre, multinational, randomized, crossover study. Diabetes Obes Metab. 2021;23(6):1389-1396.

38.

Ware

Boughton

Allen

, et al. Cambridge hybrid closed-loop algorithm in children and adolescents with type 1 diabetes: a multicentre 6-month randomised controlled trial. Lancet Digit Health. 2022;4(4):e245-e255.

39.

Lee

TTM

Collett

Bergford

, et al. Automated insulin delivery in women with pregnancy complicated by type 1 diabetes. N Engl J Med. 2023;389:1566-1578.

40.

Ozaslan

Deshpande

Doyle

3rd Dassau

. Zone-MPC automated insulin delivery algorithm tuned for pregnancy complicated by type 1 diabetes. Front Endocrinol. 2021;12:768639.

41.

Deshpande

Pinsker

Zavitsanou

, et al. Design and clinical evaluation of the interoperable artificial pancreas system (iAPS) smartphone app: interoperable components with modular design for progressive artificial pancreas research and development. Diabetes Technol Ther. 2019;21(1):35-43.

42.

American Diabetes Association Professional Practice C, Draznin

Aroda

, et al. 6. Glycemic targets: standards of medical care in diabetes-2022. Diabetes Care. 2022;45(suppl 1):S83-S96.

43.

Polsky

Akturk

. Case series of a hybrid closed-loop system used in pregnancies in clinical practice. Diabetes Metab Res Rev. 2020;36(3):e3248.

44.

Moreno-Fernández

García-Seco

. Commercialized hybrid closed-loop system (minimed medtronic 670g) results during pregnancy. AACE Clin Case Rep. 2021;7(3):177-179.

45.

Guzmán Gómez

Viggiano

Silva-De Las Salas

Martínez

Urbano Bonilla

. The closed-loop system improved the control of a pregnant patient with type 1 diabetes mellitus. Case Rep Endocrinol. 2021;2021:7310176.

46.

Forlenza

Ekhlaspour

Breton

, et al. Successful at-home use of the tandem control-IQ artificial pancreas system in young children during a randomized controlled trial. Diabetes Technol Ther. 2019;21(4):159-169.

47.

Amadou

Franc

Benhamou

, et al. Diabeloop DBLG1 closed-loop system enables patients with type 1 diabetes to significantly improve their glycemic control in real-life situations without serious adverse events: 6-month follow-up. Diabetes Care. 2021;44(3):844-846.

48.

Bergenstal

Nimri

Beck

, et al. A comparison of two hybrid closed-loop systems in adolescents and young adults with type 1 diabetes (FLAIR): a multicentre, randomised, crossover trial. Lancet. 2021;397(10270):208-219.

49.

Vambergue

Madleen

Desir

, et al. Management of pregnancy in a patient with highly unstable type 1 diabetes with DBL-hu closed-loop system. Diabetes Technol Ther. 2022;24(9):672-674.

50.

Munda

Kovacic

Pongrac Barlovic

. Carbohydrate intake and closed-loop insulin delivery system during two subsequent pregnancies in type 1 diabetes. Metabolites. 2022;12(11):1137.

51.

Wang

Dunlop

McKeen

Feig

Donovan

. Real-world use of Control-IQ technology automated insulin delivery in pregnancy: a case series with qualitative interviews. Diabet Med. 2023;40(6):e15086.

52.

Albert

Romero

Capel

, et al. Off-label use of an automated insulin delivery system in pregnant women with type 1 diabetes. Diabetes Res Clin Pract. 2023;202:110800.

53.

Lemieux

Yamamoto

Donovan

. Do-it-yourself artificial pancreas system use in pregnant women with type 1 diabetes in a real-world setting: 2 case reports. Can J Diabetes. 2021;45(8):804-808.

54.

Schütz-Fuhrmann

Schütz

Eichner

Mader

. Two subsequent pregnancies in a woman with type 1 diabetes: artificial pancreas was a gamechanger. J Diabetes Sci Technol. 2020;14(5):972-973.

55.

Bukhari

Malek

. Open-source automated insulin delivery systems for the management of type 1 diabetes during pregnancy. BMJ Case Reports. 2021;14(9):e243522.

56.

Schütz

Schütz-Fuhrmann

Stadler

Staudinger

Mader

. Management of type 1 diabetes mellitus using open-source automated insulin delivery during pregnancy: a case series. Diabetes Technol Ther. 2022;24(3):227-230.

57.

ClinicalTrials.gov. PICLS study. https://clinicaltrials.gov/ct2/show/NCT03774186?term=PICLS. Accessed October 19, 2022.

58.

Beunen

Van Wilder

Ballaux

, et al. Closed-loop insulin delivery in pregnant women with type 1 diabetes (CRISTAL): a multicentre randomized controlled trial-study protocol. BMC Pregnancy Childbirth. 2023;23(1):180.

59.

ClinicalTrials.gov. CIRCUIT study, https://clinicaltrials.gov/ct2/show/NCT04902378?term=CIRCUIT. Accessed October 19, 2022.

60.

Donovan

Feig

Lemieux

, et al. A randomized trial of closed-loop insulin delivery postpartum in type 1 diabetes. Diabetes Care. 2023;46:2258-2266.

61.

Lee

TTM

Collett

Man

, et al. AiDAPT: automated insulin delivery amongst pregnant women with type 1 diabetes: a multicentre randomized controlled trial-study protocol. BMC Pregnancy Childbirth. 2022;22(1):282.

62.

Lawton

Kimbell

Closs

, et al. Listening to women: experiences of using closed-loop in type 1 diabetes pregnancy. Diabetes Technol Ther. 2023;25:845-855.

63.

Lawton

Rankin

Hartnell

Lee

, et al. Healthcare professionals’ views about how pregnant women can benefit from using a closed-loop system: qualitative study. Diabet Med. 2023;40(5):e15072.

64.

Rankin

Hart

Kimbell

, et al. Rollout of closed-loop technology to pregnant women with type 1 diabetes: healthcare professionals’ views about potential challenges and solutions. Diabetes Technol Ther. 2023;25(4):260-269.

65.

Chaput

Dunlop

Booth

Feig

Donovan

. Existing standardised questionnaires do not adequately capture quality-of-life outcomes of greatest importance for those living with type 1 diabetes in pregnancy. Diabet Med. 2023;40(4):e15044.

66.

Scott

Murphy

Myers

Saravanan

Poston

Law

. MAGIC (maternal glucose in pregnancy) understanding the glycemic profile of pregnancy, intensive CGM glucose profiling and its relationship to fetal growth: an observational study protocol. BMC Pregnancy Childbirth. 2023;23(1):563.

67.

Durnwald

Beck

, et al. 173-OR: CGM-measured differences in pregnancies with vs. without gestational diabetes (GDM) and adverse perinatal outcomes. Abstract at ADA, San Diego, CA, Section Pregnancy—Clinical/Epidemiology, June 20, 2023.

Automated Insulin Delivery for Pregnant Women With Type 1 Diabetes: Where Do We Stand?

Abstract

Keywords

Introduction

AID Approved for Use in Pregnancy

Other Pilot Studies in Pregnancy

Off-Label Use During Pregnancy

RCTs on AID During Pregnancy or Postpartum

Advantages and Challenges of Current AID in Pregnancy

Future Research

Conclusion

Footnotes

Abbreviations

Declaration of Conflicting Interests

Funding

ORCID iD

References