Glycemia Around Exercise in Adults with Type 1 Diabetes Using Automated and Nonautomated Insulin Delivery Pumps: A Switch Pilot Trial

Abstract

In an in-patient switch study, 10 adults with type 1 diabetes (T1D) performed 45 min of moderate-intensity exercise on 2 occasions: (1) when using their usual insulin pump (UP) and (2) after transitioning to automated insulin delivery (AID) treatment (MiniMed™ 780G). Consensus glucose management guidelines for performing exercise were applied. Plasma glucose concentrations measured over a 3-h monitoring period were stratified into time below range (TBR, <3.9 mmol/L), time in range (TIR, 3.9–10.0 mmol/L), and time above range (TAR, >10.0 mmol/L).

Overall, TBR (UP: 11 ± 21 vs. AID: 3% ± 10%, P = 0.413), TIR (UP: 53 ± 27 vs. AID: 66% ± 39%, P = 0.320), and TAR (UP: 37 ± 34 vs. AID: 31% ± 41%, P = 0.604) were similar between arms. A proportionately low number of people experienced exercise-induced hypoglycemia (UP: n = 2 vs. AID: n = 1, P = 1.00).

In conclusion, switching to AID therapy did not alter patterns of glycemia around sustained moderate-intensity exercise in adults with T1D.

Clinical Trial Registration number: NCT05133765.

Introduction

In the past decade, the treatment landscape of type 1 diabetes (T1D) has changed considerably with new pharmacological and technological therapies penetrating the market. In the case of the latter, recent advances have led to the development of automated insulin delivery (AID) systems that combine an infusion pump with a control algorithm that regulates subcutaneous insulin delivery based on continuous glucose monitoring (CGM) data.

The safety, efficacy, and feasibility of transitioning to AID therapy in T1D have been documented in various inpatient and real-world studies with associated improvements in glycemic control.^1

–5 The increased automaticity of AID therapy lessens the constraints of constant self-management, thereby liberating users to focus on other life factors.⁶ Nevertheless, some user involvement is still needed in circumstances wherein an adjustment in insulin infusion rate may be required to temper anticipated glycemic excursions (i.e., meals and exercise).

However, there is limited knowledge about the integration of AID use around exercise. This is somewhat surprising as exercise can cause considerable and sometimes clinically concerning glyemic disturbance in those with T1D. This is particularly evident when exercise is performed postprandially,^7
–9 at a time when circulating insulin levels are raised due to the pharmacokinetics of the concomitant bolus insulin dose.¹⁰ Although consensus glucose management guidelines for exercise exist,¹¹ it is unclear whether they are applicable to the newest generation of diabetes technologies. Hence, as it stands, exercise advice for AID use is largely based on expert opinion and/or experience.¹²

With the commercial introduction of AID systems for the glycemic management of people with T1D, there is a pressing need to test their safety and efficacy around exercise. As such, we sought to detail the glycemic responses to a controlled bout of moderate-intensity exercise in adults with T1D using automated and nonautomated insulin delivery pumps.

Materials and Methods

Study design and ethical approval

This was a two-period, in-patient, switch study involving adults with T1D. The study was carried out in accordance with the Helsinki Declaration, EU Directive on good clinical practice, and ICH-GCP guidelines after approval by the Regional Scientific Ethical Committee and the Capital Region's Videnscenter for Dataanmeldelser (P-2021-169). All participants were provided with a full written and verbal description of the study and gave informed consent before taking part. The study was registered as a clinical trial.

Screening procedures

Participants in this study were recruited from a separate, but simultaneously conducted and ongoing, randomized crossover trial exploring the efficacy of the MiniMed 780G in people with elevated HbA_1c during which the following inclusion and exclusion criteria applied.

Main inclusion criteria were: aged 18–75 years; T1D ≥2 years; HbA_1c; ≥7.5% (58 mmol/mol); use of insulin pump treatment for ≥12 months; use of a continuous, or an intermittently scanned, glucose monitoring system for ≥6 months; and use of insulin Aspart (Novo Nordisk A/S, Bagsværd, Denmark) for ≥1 week. Main exclusion criteria were: females who were pregnant or breastfeeding; use of glucose-lowering medications (other than insulin), corticosteroids, and/or other drugs affecting glucose metabolism during the study period or within 30 days before study start; use of an AID system; daily use of acetaminophen; alcohol or drug abuse; and conditions contraindicating HbA_1c <7% (53 mmol/mol).

After confirmation of suitability for the main study, participants were asked whether they would be interested in participating in the present exercise substudy. Interested individuals then performed a graded exercise test to volitional exhaustion on a workload-controlled cycle ergometer (Corival; Lode^©, Groningen, The Netherlands). The results were used to determine the individualized workload (watts) required to complete the moderate-intensity (∼60% V̇O_2peak) exercise bout incorporated in each of the exercise trial experimental visits.

Insulin pump therapy switch

Participants were switched from their usual insulin pump (UP) to the MiniMed™ 780G system (Medtronic, Northridge, CA). The technology automatically adjusts basal insulin every 5 min based on CGM input, houses adjustable glucose targets of 100 (5.5), 110 (6.1), and 120 (6.7) mg/dL (mmol/L), and includes an automatic correction bolus feature. A raised temporary glucose target of 150 (8.3) mg/dL (mmol/L) can be set for scenarios such as exercise. By doing so, the autocorrection feature is suspended, and the automatic basal insulin delivery rate is adjusted in an attempt to attain the temporary target glucose.

User-initiated meal announcements are required for optimal glycemic results.^12,13 In this study, each participant had been using the MiniMed 780G system >4 weeks before the AID trial day. Participants used Guardian 3 link or Guardian 4 transmitters connected to the MiniMed 780G system and were advised to change their sensor 24 h before the trial visit.

Experimental trial day procedures

In a switch manner, participants attended the laboratory and performed a 45-min bout of exercise 90 min after consuming a carbohydrate-based drink on two separate occasions: first, when using their UP and second, after transitioning to the AID system (AID). Participants arrived at the research facility after an overnight fast (≥10 h) from food with water ad libitum. Upon arrival, participants adopted a bed-rest position and were fitted with an indwelling cannula ahead of the interventional period.

After the first sample draw (baseline i.e., t = −90 min), participants consumed a standardized low-glycemic index, carbohydrate-based drink ([Isomaltulose; BENEO GmbH, Mannheim, Germany] equating to 0.75 g of carbohydrates per kg body mass) with a 25% bolus insulin dose reduction.¹¹ At the same time, participants reduced their basal insulin rate by 20% in the UP arm or applied the temporary target in the AID system arm¹² as per consensus guidelines. In both arms, these settings were maintained until 15 min after exercise (t = +60 min).

Fifteen minutes before the anticipated exercise start time, plasma glucose (PG) concentrations were checked to ensure safe starting concentrations.¹¹ If PG was <5.0 mmol/L (<90 mg/dL), participants were given 15 g of oral glucose (Dextro Energy GmbH & Co. KG, Krefeld, Germany). If PG was ≥15.0 mmol/L (≥270 mg/dL) and blood ketones levels were <0.6 mmol/L, exercise went ahead but only at the discretion of the participant with frequent monitoring for ketone body formation. If ketone levels were ≥1.5 mmol/L, the visit was canceled and rescheduled.¹¹

After 90-min of bed rest (t = 0 min), participants commenced a bout of moderate-intensity (∼60% VȮ_2peak) exercise on a workload-controlled cycle ergometer (Corival; Lode^©). The exercise session lasted for 45-min (t = +45 min), or until hypoglycemia (PG <3.9 mmol/L [<70 mg/dL]). In the case of the latter, exercise was stopped immediately, and a standardized hypoglycemia treatment protocol was initiated, i.e., provided 15 g of oral carbohydrates (Dextro Energy GmbH & Co. KG), waited 15 min, repeated if necessary.^14
–16 After exercise, participants remained within the laboratory for a further 60-min of observational bed rest.

Blood sampling procedures

Venous-derived whole blood samples were obtained in 15-min intervals from −90 to −15 min, 5-min intervals from −15 to +60 min, and 15-min intervals from +60 to +105 min (Fig. 1) for PG determination (YSI, Inc., Ohio).

FIG. 1.

PG responses before, during, and after a 45-min bout of sustained moderate-intensity exercise when participants were using their UP and after transitioning over to an AID system. PG concentrations on experimental visits when data are expressed as (A) the absolute concentrations at each timepoint and (B) the change from fasted starting values at each timepoint. Unfilled markers indicate no change in the point concentration of PG from fasted levels. Shaded gray rectangle represents the designated moderate-intensity continuous exercise period. Data are presented as mean ± SEM. AID, automated insulin delivery; PG, plasma glucose; SEM, standard error of the mean.

Statistics and computation of glycemic parameters

All statistical analyses were performed via SPSS (IBM^®; SPSS, Inc., Chicago). Unless otherwise stated, data are presented as mean ± SD. Time spent within a specific glucose zone was calculated as the number of PG readings that fell within that zone divided by the total number of glucose readings from the participant represented as a percentage, that is, time below range ([TBR] <3.9 mmol/L [<70 mg/dL]), time in range ([TIR] 3.9–10.0 mmol/L [70–180 mg/dL]), and time above range ([TAR] >10.0 mmol/L [>180 mg/dL]).¹⁷ To account for the confounding effect of rescue carbohydrate provision on subsequent PG concentrations, the first point at which a hypoglycemic event occurred was carried forward for the remainder of the experimental trial day.

Differences in glycemic parameters between experimental arms were assessed through the paired samples t-test or, failing the assumption of normality, the nonparametric equivalent (Wilcoxon Signed Ranks test). The McNemar test was used to identify differences in the prevalence of hypoglycemia between arms. Alpha was set at 0.05 and significance was accepted when P-values were ≤alpha.

Results

Participant characteristics

Data from 10 adults with T1D [7 females, age: 49 ± 15 years, HbA_1c: 67 ± 8 mmol/mol (8.3% ± 0.8%) diabetes duration 28 ± 15 years, body mass index: 27.5 ± 3.5 kg/m², V̇O_2peak: 24.0 ± 7.7 mL/(min·kg)] were included in this study. Before switching to AID, participants used the following insulin pumps: MiniMed 640G, Medtronic (n = 7), Veo Paradigm, Medtronic (n = 2), Accu-Chek Insight, Roche diagnostics, (n = 1); whereof six used CGM and four used intermittently scanned CGM. After transition to the AID system, participants were provided with adequate time for therapy familiarization before conducting the second experimental visit (130 ± 30 days).

PG responses

Pre-exercise period

Fasted PG levels were comparable between arms (UP: 8.3 ± 2.5 vs. AID: 7.9 ± 2.2 mmol/L, P = 0.634) and remained as such at each timepoint throughout the 90-min pre-exercise period (Fig. 1). The meal-induced rise in PG was similar between arms (UP: Δ + 1.6 ± 2.4 vs. AID: +1.9 ± 1.4 mmol/L, P = 0.787) as was the individualized dose of meal-time bolus insulin (UP: 5.3 ± 2.7 vs. AID: 5.2 ± 3.1 IU, P = 0.834). All glycemic parameters were similar between pumps (Table 1 and Fig. 1).

Table 1.

Glycemic Parameters During Each Designated Time Period

Glycemic parameter	UP	AID	P
Pre-exercise period_{(−90 min to −5 min)}
Mean (mmol/L)	9.7 ± 3.0 (11.1 [4.7])	9.2 ± 2.3 (8.3 [3.9])	0.476
Minimum (mmol/L)	8.1 ± 2.6 (8.8 [4.0])	7.8 ± 2.2 (7.9 [3.4])	0.705
Maximum (mmol/L)	11.0 ± 3.4 (12.5 [5.3])	10.3 ± 2.5 (8.8 [4.3])	0.509
TBR (%)	0.0 ± 0.0 (0.0 [0.0])	0.0 ± 0.0 (0.0 [0.0])	1.00
TIR (%)	47.2 ± 40.4 (25.0 [75.0])	62.5 ± 45.9 (100.0 [87.5])	0.388
TAR (%)	52.8 ± 40.4 (75.0 [75.0])	37.5 ± 45.9 (0.0 [87.5])	0.388
CV (%)	10.4 ± 4.4 (9.6 [6.4])	10.6 ± 5.4 (8.6 [9.6])	0.943
Exercise period_{(0 min to + 45 min)}
Mean (mmol/L)	8.6 ± 3.0 (9.6 [4.3])	8.4 ± 2.7 (7.4 [4.2])	0.801
Minimum (mmol/L)	7.0 ± 2.5 (7.4 [4.7])	6.7 ± 2.7 (5.9 [3.8])	0.705
Maximum (mmol/L)	10.1 ± 3.5 (11.0 [5.1])	9.8 ± 2.8 (8.6 [4.7])	0.756
TBR (%)	12.2 ± 25.4 (0.0 [20.0])	1.2 ± 3.7 (0.0 [0.0])	0.285
TIR (%)	52.2 ± 38.0 (50.0 [75.0])	67.7 ± 41.7 (88.9 [70.0])	0.325
TAR (%)	35.6 ± 42.2 (10.0 [80.0])	31.1 ± 42.6 (0.0 [70.0])	0.588
CV (%)	13.5 ± 7.2 (11.8 [10.4])	14.6 ± 5.2 (16.3 [9.7])	0.729
Postexercise period_{(+50 min to} ₊ _105 min)
Mean (mmol/L)	7.1 ± 2.7 (7.7 [4.9])	7.1 ± 3.2 (6.7 [4.4])	0.992
Minimum (mmol/L)	6.6 ± 2.4 (7.4 [4.6])	6.5 ± 2.8 (6.0 [4.1])	0.942
Maximum (mmol/L)	7.6 ± 3.0 (8.2 [5.6])	7.6 ± 3.5 (7.2 [4.8])	1.00
TBR (%)	22.2 ± 44.1 (0.0 [50.0])	11.1 ± 33.3 (0.0 [0.0])	0.564
TIR (%)	61.1 ± 44.9 (83.3 [91.7])	66.7 ± 50.0 (100.0 [100.0])	0.796
TAR (%)	16.7 ± 30.0 (0.0 [33.4])	22.2 ± 44.1 (0.0 [50.0])	0.854
CV (%)	5.0 ± 4.0 (3.6 [6.6])	5.6 ± 3.3 (5.1 [4.8])	0.741
Exercise+postexercise period_{(0 min to} ₊ _105 min)
Mean (mmol/L)	8.1 ± 2.8 (8.5 [4.6])	7.9 ± 2.9 (7.1 [4.2])	0.852
Minimum (mmol/L)	6.5 ± 2.4 (7.0 [4.6])	6.5 ± 2.8 (5.9 [4.0])	0.967
Maximum (mmol/L)	10.3 ± 3.4 (11.1 [4.4])	9.8 ± 2.8 (8.6 [4.7])	0.620
TBR (%)	16.0 ± 32.1 (0.0 [31.3])	5.2 ± 15.6 (0.0 [0.0])	0.285
TIR (%)	55.6 ± 35.6 (56.3 [71.9])	67.0 ± 41.3 (75.0 [73.4])	0.594
TAR (%)	28.5 ± 35.5 (6.3 [59.4])	27.8 ± 42.3 (0.0 [62.5])	0.917
CV (%)	16.4 ± 9.2 (13.1 [8.6])	16.5 ± 6.2 (20.0 [11.0])	0.985
Overall period_{(−90 min to + 105 min)}
Mean (mmol/L)	8.6 ± 2.8 (9.3 [3.6])	8.3 ± 2.6 (7.2 [4.1])	0.682
Minimum (mmol/L)	6.2 ± 2.0 (6.7 [3.7])	6.3 ± 2.5 (5.9 [3.0])	0.904
Maximum (mmol/L)	11.2 ± 3.3 (12.6 [4.0])	10.4 ± 2.4 (8.8 [4.1])	0.383
TBR (%)	10.7 ± 21.4 (0.0 [20.9])	3.4 ± 10.1 (0.0 [0.0])	0.285
TIR (%)	52.6 ± 27.1 (54.2 [37.0])	65.6 ± 39.0 (69.6 [68.8])	0.320
TAR (%)	36.7 ± 34.0 (29.2 [62.5])	31.0 ± 40.7 (0.0 [68.8])	0.735
CV (%)	18.6 ± 8.3 (15.4 [10.8])	17.9 ± 6.7 (20.4 [11.6])	0.844

Data are presented as mean ± SD (median [IQR]). UP: The visit in which participants were using their usual insulin pump. AID: the visit in which participants had been switched to an automated insulin delivery system. TBR: time spent with PG below the target range (<3.9 mmol/L). TIR: time spent with PG levels within the target range (3.9–10.0 mmol/L). TAR: time spent with PG above the target range (>10.0 mmol/L).

CV, coefficient of variation; PG, plasma glucose.

Exercise period

PG concentrations immediately before exercise onset were similar between arms (UP: 10.0 ± 3.5 vs. AID: 9.8 ± 2.9 mmol/L, P = 0.850). Both the magnitude (UP: Δ−3.0 ± 1.9 vs. AID: Δ−3.1 ± 1.2 mmol/L, P = 0.818) and rate (UP: −0.07 ± 0.04 vs. AID: −0.07 ± 0.03 mmol/L/min, P = 0.893) of change in PG over exercise were comparable between conditions as were the duration (UP: 40.0 ± 10.6 vs. AID: 45 ± 0.0 min, P = 0.195) and intensity (UP: 65 ± 11 vs. AID: 60% ± 14%V̇O_2peak, P = 0.194) of the exercise bout. Two people experienced hypoglycemia during cycling in UP, whereas one person did so in AID (P = 1.00). All in-exercise glycemic parameters were similar between arms (Table 1 and Fig. 1).

Post-exercise period

Point concentrations of PG throughout the 1-h post-exercise period were similar between arms (Fig. 1) as were all other glycemic parameters (Table 1). There was one episode of hypoglycemia during the postexercise period in UP and none in AID (P = 1.00). Participants completed the trial with equivalent PG concentrations on both occasions (UP: 7.2 ± 3.5 vs. AID: 7.2 ± 3.0 mmol/L, P = 0.970).

Overall period

Overall, two people experienced hypoglycemia during UP, whereas one person did so during AID (P = 1.00). The time to hypoglycemia onset was similar between arms (UP: 130 ± 31.2 vs. AID: 135 ± 0.0 min, P = 0.902). Overall TBR (UP: 11 ± 21 vs. AID: 3% ± 10%, P = 0.285), TIR (UP: 53 ± 27 vs. AID: 66% ± 39%, P = 0.320), and TAR (UP: 37 ± 34 vs. AID: 31% ± 41%, P = 0.735) were similar between arms as were all other glycemic parameters (Table 1 and Fig. 1).

Discussion

This study assessed glycemic responses to a bout of aerobic exercise in adults with T1D before and after therapy switch to an AID system. We found similarities in all glycemic parameters between the automated and nonautomated insulin delivery pumps around a sustained bout of moderate-intensity cycling when consensus guidelines for glucose management were applied.^11,12

Our findings indicate that for 45-min of aerobic exercise undertaken 1.5 h after a meal, a strategy involving both a 25% reduction in the concomitant bolus insulin dose and an increase in the individualized sensor glucose target (to 8.3 mmol/L) is associated with minimal risk of exercise-induced hypoglycemia in adults with T1D using AID therapy. Indeed, during cycling, the AID system achieved 1% TBR, which, although not statistically significant, was proportionately less than the 12% observed in the usual pump arm, a theme that continued into the postexercise period, where one individual experienced a recurrent event while using the predecessor pump.

It is interesting to consider whether the prevalence of hypoglycemia would have been more prominent had our exercise stimulus or indeed, observational window, been longer in duration. Even so, given that hypoglycemia represents a major clinical concern that dissuades regular exercise engagement in those with T1D, the poignance of the present findings should perhaps not be easily overlooked.

In light of the synergistic glucose-lowering effects of exercising muscle tissue and active on-board insulin, the need to consider exercise announcement and bolus dose reductions ahead of activities performed soon after a meal in those using closed-loop systems has been highlighted.⁹ Our data provide extension of findings by those of Paldus and colleagues,¹⁸ who highlighted glycemic superiority when using a hybrid-closed loop system (MiniMed 670G system) relative to standard therapy during moderate- and high-intensity exercise in adults with T1D.

Our results align with recent study by Myette-Cote et al,¹⁹ who in their investigation of AID use around exercise in adults with T1D documented the complete avoidance of hypoglycemia during 60-min of moderate-intensity continuous exercise performed 1 and 2 h after a meal when combining a 33% reduction in the meal-time insulin dose with an increased glucose target (6.0 to 9.0 mmol/L).

Hence, although caution must be erred when considering the pilot nature of both our studies, these data provide empirical evidence to support the suitability and safety of current consensus guidelines for AID use around postprandially performed exercise.¹² Nevertheless, how best to utilize the newest generation of insulin pumps in an exercise setting is yet to be established, and more research is needed to identify strategies that mitigate risk under different exercise scenarios.

With the ongoing technological evolution of diabetes care, access to information that represents the latest management tools is critical in providing the greatest scope for maximizing safety. As we continue to design, develop, and disseminate new diabetes technologies, it is important we consider their integration around exercise.

Study strengths, limitations, and future research directives

This study is among the first to characterize the efficacy of an AID system around physical exercise in adults with T1D. Although inherent limitations of this study include the provision of a carbohydrate-only drink and the small sample size with homogeneity in diabetes and physical fitness characteristics, the novelty of this study provides a foundational basis from which others can start to formulate prudent management strategies using AID therapy in different exercising scenarios. Expansion of this study to a larger and more heterogeneous population should be considered for further research.

Conclusion

In this study, switching to AID therapy did not alter patterns of glycemia around sustained moderate-intensity exercise in adults with T1D. The data herein provides evidence-based infromation that may help govern decision making for exercise management using the newest generation automated insulin pumps.

Footnotes

Acknowledgments

The authors thank the participants for their willingness to contribute and commit to the study protocol. We also thank the Diabetesforeningen (Denmark) for their financial contributions to the project as well as BENEO GmbH for supplying the pre-exercise carbohydrate source used in this research (Palatinose™). Finally, we thank the study biomedical scientist, Sandra Tawfik, for her help in performing trial-day activities.

Authors' Contributions

O.M.M., M.B.C., S.S., R.M.B., S.C.B., and K.N. contributed to the conception and design of the study. O.M.M., K.B.K., M.B.C., S.S., K.N., and A.G.R. contributed to the acquisition of data. O.M.M. was responsible for data analyses. All authors were responsible for data interpretation. O.M.M. wrote the original draft of the article. All authors contributed to revising the article. All authors provided final approval of the version to be published.

Author Disclosure Statement

S.S. is an employee of Novo Nordisk A/S as of May 1st, 2022. SS has received speaker's fee from Novo Nordisk. KN received funding through her institution for participating in advisory boards from Medtronic, Novo Nordisk, and Convatec and for lecturing from Sanofi, Novo Nordisk, Medtronic, and Dexcom. Her institution received funding for studies she performed from Zealand Pharma, RSP Systems, Novo Nordisk, Medtronic, and Dexcom. The remaining authors report having no relevant conflicts of interest to disclose.

Funding Information

The ongoing randomized-controlled trial on the MiniMed 780G from which participants in the current exercise study were recruited was an investigator-initiated study funded by Medtronic diabetes.

References

Brown

, Kovatchev

, Raghinaru

, et al. Six-month randomized, multicenter trial of closed-loop control in type 1 diabetes. N Engl J Med, 2019; 381(18):1707–1717; doi: 10.1056/NEJMoa1907863

Silva J

, Lepore

, Battelino

, et al. Real-world performance of the MiniMed™ 780G system: First report of outcomes from 4120 users. Diabetes Technol Ther, 2022; 24(2):113–119; Available from: https://pubmed.ncbi.nlm.nih.gov/34524003/

Breton

, Kovatchev

. One year real-world use of the control-IQ advanced hybrid closed-loop technology. Diabetes Technol Ther, 2021; 23(9):601–608; doi: 10.1089/dia.2021.0097

Pintaudi

, Gironi

, Nicosia

, et al. MiniMed Medtronic 780G optimizes glucose control in patients with type 1 diabetes mellitus. Nutr Metab Cardiovasc Dis, 2022; 32(7):1719–1724; Available from: https://pubmed.ncbi.nlm.nih.gov/35599092/

Choudhary

, Kolassa

, Keuthage

, et al. Advanced hybrid closed loop therapy versus conventional treatment in adults with type 1 diabetes (ADAPT): A randomised controlled study. Lancet Diabetes Endocrinol, 2022; 10(10):720–731; Available from: http://www.thelancet.com/article/S2213858722002121/fulltext

Sherr

, Heinemann

, Fleming

, et al. Automated insulin delivery: Benefits, challenges, and recommendations. A Consensus Report of the Joint Diabetes Technology Working Group of the European Association for the Study of Diabetes and the American Diabetes Association. Diabetes Care, 2023; 66(1):3–22; doi: 10.2337/dci22-0018/147674

McCarthy

, Deere

, Churm

, et al. Extent and prevalence of post-exercise and nocturnal hypoglycemia following peri-exercise bolus insulin adjustments in individuals with type 1 diabetes. Nutr Metab Cardiovasc Dis, 2021; 31(1):227–236.

West

, Stephens

, Bain

, et al. A combined insulin reduction and carbohydrate feeding strategy 30 min before running best preserves blood glucose concentration after exercise through improved fuel oxidation in type 1 diabetes mellitus. J Sports Sci, 2011; 29(3):279–289.

Tagougui

, Taleb

, Legault

, et al. A single-blind, randomised, crossover study to reduce hypoglycaemia risk during postprandial exercise with closed-loop insulin delivery in adults with type 1 diabetes: Announced (with or without bolus reduction) vs unannounced exercise strategies. Diabetologia, 2020; 63(11):2282–2291; Available from: https://pubmed.ncbi.nlm.nih.gov/32740723/

10.

Heise

, Hövelmann

, Brøndsted

, et al. Faster-acting insulin aspart: Earlier onset of appearance and greater early pharmacokinetic and pharmacodynamic effects than insulin aspart. Diabetes Obes Metab, 2015; 17(7):682; Available from:pmc/articles/PMC5054830/

11.

Riddell

, Gallen

, Smart

, et al. Exercise management in type 1 diabetes: A consensus statement. Lancet Diabetes Endocrinol, 2017; 8587(17):1–14.

12.

Phillip

, Nimri

, Bergenstal

, et al. Consensus Recommendations for the Use of Automated Insulin Delivery (AID) technologies in clinical practice. Endocr Rev, 2022;bnac022; doi: 10.1210/endrev/bnac022/6692818

13.

Guides and Manuals | Medtronic Diabetes [Internet]. [cited January 4, 2023]. Available from: https://www.medtronic-diabetes.com.au/support/guides-and-manuals/minimed-780g/guides [Last accessed: January 4, 2023].

14.

Destree

, Vercellino

, Armstrong

Interventions to improve adherence to a hypoglycemia protocol. Diabetes Spectr, 2017; 30(3):195; Available from:pmc/articles/PMC5556581

15.

Slama

, Traynard

P-Y

, Desplanque

, et al. The search for an optimized treatment of hypoglycemia: Carbohydrates in tablets, solution, or gel for the correction of insulin reactions. Arch Intern Med, 1990; 150(3):589–593; Available from: https://jamanetwork.com/journals/jamainternalmedicine/fullarticle/612981

16.

American Diabetes Association. Hypoglycemia (Low Blood Glucose) | ADA [Internet]. [cited January 4, 2023]. Available from: https://diabetes.org/healthy-living/medication-treatments/blood-glucose-testing-and-control/hypoglycemia [Last accessed: January 4, 2023].

17.

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.

18.

Paldus

, Lee

, Morrison

, et al. First randomized controlled trial of hybrid closed loop versus multiple daily injections or insulin pump using self-monitoring of blood glucose in free-living adults with type 1 diabetes undertaking exercise. J Diabetes Technol, 2021; 15(6):1399–1401; Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8655292/

19.

Myette-Cote

, Molveau

, Wu

, et al. A randomized crossover pilot study evaluating glucose control during exercise initiated 1 or 2 h after a meal in adults with type 1 diabetes treated with an automated insulin delivery system. Diabetes Technol Ther, 2022. Available from: https://pubmed.ncbi.nlm.nih.gov/36399114/