Comparable Glucose Control with Fast-Acting Insulin Aspart Versus Insulin Aspart Using a Second-Generation Hybrid Closed-Loop System During Exercise

Abstract

Background:

This study compared glucose control with fast-acting insulin aspart (FiAsp) versus insulin aspart following moderate-intensity exercise (MIE) and high-intensity exercise (HIE) using a second-generation closed-loop (CL) system in people with type 1 diabetes.

Materials and Methods:

This randomized crossover study compared FiAsp versus insulin aspart over four sessions during MIE and HIE with CL insulin delivery by the MiniMed™ Advanced hybrid CL system. Participants were randomly assigned FiAsp and insulin aspart each for 6 weeks and within each period performed, in random order, 40 min MIE (∼50% VO₂max) and HIE (6 × 2 min ∼80% VO₂max; 5 min recovery). The primary outcome was continuous glucose monitoring (CGM) time in range (TIR, 3.9–10.0 mM) for 24 h following exercise.

Results:

Sixteen adults (9 male; age 48 [37, 57] years; hemoglobin A1c (HbA1c) 7.0 [6.4, 7.2] %; duration diabetes 30 [17, 41] years) were recruited. In the 24 h postexercise, median TIR was >81%, time in hypoglycemia (<3.9 mM) was <4%, and time in hyperglycemia (>10 mM) was <17% for both exercise conditions and insulin formations, with no significant differences between insulins (P > 0.05). In the 2 h postexercise and overnight, the TIR approached 100% for all conditions.

Conclusions:

There were no differences in TIR during and 24 h after MIE or HIE when comparing insulin aspart with FiAsp delivered by a second-generation CL system. Insulin formulations with an offset in action greater than FiAsp are needed to provide a meaningful improvement in CL glucose control with exercise.

Clinical Trial Registration number: ACTRN12619000469112.

Introduction

There is a plethora of research demonstrating the positive health benefits of exercise, including improvements in dyslipidemia, hypertension, visceral obesity, chronic inflammation, insulin resistance, and a sense of subjective well-being in people without diabetes,^1,2 all of which are particularly relevant to the person living with type 1 diabetes. In support of this, there is a significant body of evidence validating that exercise improves these measures of physical and psychological well-being in people with type 1 diabetes.¹ The American Diabetes Association (ADA) recommends 150 min of exercise per week, with 75 min of vigorous intensity exercise and resistance exercise at least 2–3 times per week, and avoidance of two consecutive days without exercise.³ However, despite the associated health benefits and the ADA recommendations, the additional challenges with glucose control (such as fear of hypoglycemia) act as a significant disincentive to exercise for many people with type 1 diabetes.⁴

Automated insulin delivery through closed loop (CL) has been shown to improve glycemia in people with type 1 diabetes.^5

–9 However, even CL systems are challenged by exercise as insulin requirements can change rapidly and are often dependent upon the intensity and duration of the exercise.^10
–12 For individuals with type 1 diabetes undertaking moderate-intensity exercise (MIE), plasma insulin concentration cannot be decreased rapidly at the start of exercise and might even rise in the systemic circulation^13,14 due to increased subcutaneous blood flow, generally causing blood glucose levels to fall. Conversely, with increasing exercise intensity, there is a corresponding increase in counter-regulatory hormone-mediated insulin resistance and hepatic glucose production,^11,12 which may cause blood glucose levels to rise.

The pharmacodynamic limitations associated with the subcutaneous delivery of rapid-acting insulins, characterized by a delayed onset and offset in action,^15
–17 limit the ability of CL systems to address the rapidly changing insulin requirements associated with exercise. A sufficiently responsive insulin could address these limitations in glucose control. Fast-acting insulin aspart (FiAsp) has a more rapid onset and a shorter duration of insulin action^18,19 compared with insulin aspart, although there are limited data available regarding the use of FiAsp in insulin pumps.²⁰ Several recent studies have shown that the use of FiAsp in a CL system provides similar clinical outcomes to insulin aspart.^21

–25 However, only one study has specifically examined exercise as an intervention when using FiAsp in a first-generation CL device.²³

The MiniMed™ advanced hybrid CL (AHCL) system represents a second-generation CL system. Insulin delivery parameters are broadened with the aim of remaining in CL for longer periods of time without compromising safety. An auto-bolus function was incorporated to enhance the glucose time in range (TIR).²⁴ The latter function may be advantageous in leveraging the pharmacokinetic characteristics of FiAsp. It is still unclear whether the faster action of FiAsp used in an AHCL device will result in better glycemic control during and after exercise.

Thus, the primary study aim was to collect pilot data in adults with type 1 diabetes comparing glucose levels and performance of the MiniMed AHCL system delivering FiAsp versus insulin aspart during and post-MIE and high-intensity exercise (HIE).

Materials and Methods

Study design

This was a two-stage, unblinded, randomized crossover study comparing FiAsp versus insulin aspart over four sessions during MIE and HIE with CL insulin delivery by AHCL (Supplementary Figure S1). The exercise study was conducted at St Vincent's Hospital Melbourne, from June 2019 to November 2020 as part of larger study examining glucose control during AHCL with FiAsp versus insulin aspart, each infused over 6 weeks, with an emphasis on postprandial glucose control. Participants were commenced on the AHCL automated insulin delivery system (Medtronic, Northridge, CA) and randomly assigned FiAsp and insulin aspart, each for 6 weeks. Each stage was preceded by 2 weeks in open loop. During each stage, participants performed, in random order and at least 3 weeks post-CL activation, 40 min MIE (∼50% maximum pulmonary oxygen consumption [VO₂max]) and HIE (6 × 2 min at ∼80% VO₂max with 5 min recovery) with at least 72 h between exercise bouts. Randomization for assignment to insulin and exercise interventions was implemented using a computerized random number generated sequence. All study procedures were approved by the St. Vincent's Hospital Human Research Ethics Committee and all subjects provided written informed consent to participate in the study. The study was registered at www.ANZCTR.org.au. The institutional ethics reference number is HREC005/19 if this is needed.

Participants

Main inclusion criteria included the following: age ≥18 years; clinical diagnosis of type 1 diabetes for at least 1 year; insulin pump therapy use for >3 months; continuous glucose monitoring (CGM) experience; HbA1c <86 mmol/mol (<10.0%); and ability to carbohydrate count. Main exclusion criteria included the following: current or planned pregnancy; estimated glomerular filtration rate <40 mL/min/1.73 m²; history of diabetic ketoacidosis or severe hypoglycemia in the prior 3 months; or major medical or psychiatric illness.

CL insulin delivery system

The study device consisted of a Medtronic 600 series insulin pump with the AHCL algorithm; a Guardian™ 3 sensor; Guardian Link 3 transmitter; and a CONTOUR™ NEXT LINK 2.4 blood glucose meter (Ascensia Diabetes Care, Parsippany, NJ). The MiniMed AHCL system incorporates enhanced features compared to the MiniMed 670G system, including (1) option of two fixed basal targets of 5.6 or 6.7 mM; (2) automated correction boluses based on sensor glucose levels delivered up to every 5 min to achieve a target of 6.7 mM; and (3) improved safety features to enable increased time spent in CL. For the purposes of this study, basal targets were set at 5.6 mM and active insulin time was set at 4 h for all participants throughout the study. FiAsp and insulin aspart formulations (Novo Nordisk, Bagsvaerd, Denmark) were used in the insulin pump.

Preliminary testing

During the initial visit, anthropometric data were collected, and participants were provided with, and educated on the use of, the study devices. Preliminary exercise testing involved determination of VO₂max through graded exercise test to volitional exhaustion on a cycle ergometer (LODE, Groningen, Netherlands). VO₂max was determined using previously described criteria²⁶ and was used to prescribe the experimental exercise intensities.

Experimental exercise testing

Participants inserted a new glucose sensor and infusion set ∼48 h pre-exercise. Participants were in free-living conditions before experimental testing but were asked to consume the same meal of their choice before each exercise bout and take their usual bolus insulin for the meal at least 4 h before arriving to the laboratory to ensure little or no active bolus insulin at exercise onset. Participants were also asked to refrain from additional food consumption following their last meal before exercise, unless hypoglycemia developed (glucose level <3.9 mM). If hypoglycemia developed before exercise, participants were instructed to treat with 15 g of fast-acting carbohydrate and continue to monitor their own glucose level. Supplemental carbohydrate (15 g), without an insulin bolus, was administered 15 min pre-exercise if the glucose level was <7.0 mM.

Participants were contacted approximately 2 h before exercise as a reminder to set the elevated temporary glucose target of 8.3 mM. The exercise interventions were completed in the afternoon between 14:30 and 17:00 to reflect patterns of behavior in the type 1 diabetes community. MIE was performed on the same upright ergometer as used for the cardiopulmonary exercise test and prescribed as 40 min of continuous exercise at an intensity of ∼50% of the predetermined VO₂max. This intensity was chosen to be below the anaerobic threshold for all participants. HIE was performed on the upright ergometer for ∼40 min comprising 6 × 2-min intervals at ∼80% of VO₂max and 5-min active recovery between intervals at ∼40% of VO₂max. This regimen was calculated as having an equivalent workload to the continuous MIE regimen.

For all exercise interventions, the temporary glucose target was ceased immediately postexercise with reversion to the usual target of 5.6 mM. All participants left the clinical trial center and returned to their usual daily activities immediately following the completion of exercise.

Statistical analysis

The primary outcome was the CGM TIR (3.9–10.0 mM) for the 24 h following exercise commencement with comparisons made between insulin types separately for each exercise protocol. Results are reported as median (interquartile range [IQR]) or frequency (percentage) unless otherwise specified. Percentage of VO₂max and heart rate for exercise performed with FiAsp and insulin aspart was compared using a paired t-test. CGM outcomes from exercise commencement to 120 min after exercise completion (0–160 min) and in the 8 and 24 h after exercise commencement were compared between FiAsp and insulin aspart using Wilcoxon signed rank test. All statistical analyses were conducted using Stata 15.1 (StataCorp. 2017. Stata Statistical Software: Release 15. College Station, TX: StataCorp LLC.). The level of significance was set at P < 0.05.

Although this study is exploratory in nature, the study was powered for the primary outcome of the larger study, where sample size calculations used data from an antecedent study.²⁷ Assuming a conservative standard deviation of 7%, to detect ≥5% improvement in TIR with 80% power and 5% significance level, 21 participants were required. To allow for 15% dropout rate, 25 participants were recruited. All 25 participants of the main study were invited to take part in the exercise substudy; however, 9 refused due to lack of interest or time.

Results

Sixteen adults (9 male; age 48 [37, 57] years; HbA1c 7.0 [6.4, 7.2] %; and duration of diabetes 30 [17, 41] years) completed the study. Participant characteristics are summarized in Table 1. Fifteen (94%) participants were using sensor-augmented pump therapy before study entry. There were no significant differences between the intensity of exercise sessions performed between FiAsp and insulin aspart stages (metrics for the intensity of the prescribed exercise sessions are summarized in Supplementary Table S1). All exercise sessions were performed between 21 and 44 days after initiation of CL insulin delivery (median [IQR]: 28 [21, 35] days). In the 24 h after exercise commencement, the number of carbohydrate events entered into the insulin pump, total amount of carbohydrates in grams, and the time from exercise commencement until the first meal postexercise were not different between conditions (Supplementary Table S2).

Table 1.

Participant Characteristics

Participant characteristics	n = 16
Sex, Male \| Female	9 \| 7
Age, years	48 (37, 57)
Height, cm	175 (165, 178)
Weight, kg	81.9 (75.7, 91.8)
BMI, kg/m²	28.0 (25.1, 29.7)
HbA1c, mmol/mol \| %	53 (46, 55) \| 7.0 (6.4, 7.2)
Duration of diabetes, years	30 (17, 41)
Duration of insulin pump therapy, years	9 (6, 15)
VO₂max, ml.kg⁻¹·min⁻¹	30.00 (21.33, 33.83)
Wmax, watts	205 (172, 262)

Values are median (IQR). Wmax, maximal power output.

BMI, body mass index; HbA1c, hemoglobin A1c; IQR, interquartile range.

The sensor glucose immediately before the beginning of exercise was not different between FiAsp and insulin aspart for MIE (8.33 [5.83, 9.83] vs. 8.27 [7.16, 8.91] mM, P = 0.860, median [IQR]) or HIE (7.69 [4.72, 9.96] vs. 6.80 [5.88, 7.72] mM, P = 0.562, median [IQR]).

Hypoglycemic events and supplemental carbohydrates

In the 15 min before exercise, supplemental carbohydrate (15 g) was administered for blood glucose levels below 7.0 mM in six participants before MIE (5 aspart and 1 FiAsp) and seven participants before HIE (4 aspart and 3 FiAsp) as per published guidelines.²⁸ In addition, there were two episodes of hypoglycemia (≤3.9 mM) in the 15 min before MIE (0 aspart, 2 FiAsp) and two episodes of hypoglycemia before HIE (0 aspart, 2 FiAsp) treated with 15 g carbohydrate. None met the criteria for major hypoglycemia; the nadir for all was >3.5 mM and exercise was initiated after 15 min where blood glucose returned to >4.0 mM.

During exercise, one participant experienced hypoglycemia during HIE (0 aspart and 1 FiAsp) and four participants experienced hypoglycemia during MIE (1 aspart and 3 FiAsp). In none of the events did the nadir glucose drop below 3.5 mM and all lasted <20 min. In all cases of hypoglycemia during exercise, the exercise protocol was able to be completed once glucose levels returned to a level >3.9 mM.

Glycemic outcomes

Median TIR for the primary outcome 24 h after exercise commencement was >80% for all exercise conditions (Fig 1. TIR (median [IQR]) with MIE was 85.7 [79.2, 92.5]% vs. 81.1 [71.0, 92.7]%; P = 0.737 for FiAsp and insulin aspart, respectively, and with HIE was 83.0 [78.3, 93.6]% vs. 87.3 [81.1, 90.8]%; P = 0.587). Time in hyperglycemia was less than 17%, while time in hypoglycemia was 0% for MIE and less than 4% for HIE. There were no significant differences between FiAsp and insulin aspart for any glycemic or insulin delivery outcome (Table 2). Supplementary Figure S2 shows the frequency distribution of TIR for the 24 h after exercise commencement.

FIG. 1.

CGM glucose for the 24 h after exercise commencement (A: HIE, B: MIE). Lines represent median values for FiAsp (red) and insulin aspart (blue), while shaded areas represent IQRs. Dashed horizontal lines indicate the target glucose range between 3.9 and 10.0 mM. Gray shaded bars indicate the 40-min exercise period. CGM, continuous glucose monitoring; FiAsp, fast-acting insulin aspart; HIE, high-intensity exercise; IQRs, interquartile ranges; MIE, moderate-intensity exercise.

Table 2.

Glucose Control And Insulin Delivery Through CL in the 24 h Post-MIE and Post-HIE Commencement with FiAsp and Insulin Aspart

CGM parameter	FiAsp (n = 16)	Aspart (n = 16)	P
24 h after exercise commencement
MIE
CGM% time 3.9–10 mM, median (IQR)	85.73 (79.17, 92.53)	81.08 (70.99, 92.71)	0.737
CGM% time 3.9–7.8 mM, median (IQR)	59.19 (49.65, 70.31)	58.95 (51.74, 66.32)	0.717
CGM% time >10 mM, median (IQR)	11.18 (4.72, 18.92)	16.32 (6.94, 23.98)	0.552
CGM% time >13.9 mM, median (IQR)	0.00 (0.00, 0.00)	0.00 (0.00, 1.57)	0.415
CGM% time <3.9 mM, median (IQR)	0.00 (0.00, 0.87)	0.00 (0.00, 2.98)	0.833
CGM% time <3.0 mM, median (IQR)	0.00 (0.00, 0.00)	0.00 (0.00, 0.00)	0.617
Mean glucose, median (IQR)	7.37 (6.66, 8.10)	7.56 (7.04, 8.08)	1
Coefficient of variation, median (IQR)	26.64 (22.28, 31.20)	27.59 (19.58, 34.16)	1
Total insulin delivery (U), median (IQR)	25.12 (18.53, 34.22)	24.12 (16.02, 30.24)	0.255
HIE
CGM% time 3.9–10 mM, median (IQR)	83.03 (78.30, 93.58)	87.33 (81.15, 90.80)	0.587
CGM% time 3.9–7.8 mM, median (IQR)	58.01 (49.92, 71.18)	61.99 (50.00, 76.91)	0.918
CGM% time >10 mM, median (IQR)	12.12 (4.51, 19.62)	9.03 (4.17, 15.62)	0.313
CGM% time >13.9 mM, median (IQR)	0.00 (0.00, 1.38)	0.00 (0.00, 1.07)	0.262
CGM% time <3.9 mM, median (IQR)	1.22 (0.35, 5.05)	3.34 (1.04, 5.73)	0.313
CGM% time <3.0 mM, median (IQR)	0.00 (0.00, 0.00)	0.00 (0.00, 0.00)	0.617
Mean glucose, median (IQR)	7.42 (6.65, 7.92)	7.34 (6.56, 7.73)	0.469
Coefficient of variation, median (IQR)	30.32 (24.05, 33.84)	27.15 (22.28, 30.71)	0.234
Total insulin delivery (U), median (IQR)	28.18 (19.59, 35.38)	23.35 (17.07, 32.99)	0.234

Data are median (IQR).

CGM, continuous glucose monitoring; CL, closed-loop; FiAsp, fast-acting insulin aspart; HIE, high-intensity exercise; MIE, moderate-intensity exercise.

In the 40 min from the commencement until the completion of exercise, the change in sensor glucose levels was not significantly different for insulin formations: HIE (median [IQR], −0.45 [−2.44, −0.16] vs. −1.00 [−1.46, −0.22] mM, P = 0.669) or MIE (−1.25 [−3.14, −0.08] vs. −0.93 [−2.92, −0.19] mM, P = 0.84) for FiAsp versus insulin aspart, respectively.

In the 160 min from exercise commencement until 2 h after exercise completion, median TIR (3.9–10.0 mM) was 89.6% for the FiAsp-HIE condition and 100% for all other exercise conditions. Median time in hyperglycemia (>10.0 mM) and hypoglycemia (<3.9 mM) was 0% for all exercise conditions. There were no significant differences between FiAsp and insulin aspart for any glycemic or insulin delivery outcomes (Table 3).

Table 3.

Glucose Control And Insulin Delivery Through CL from the Beginning Of Exercise Until 2 H Post-MIE and Post-HIE Completion with FiAsp and Insulin Aspart

CGM parameter	FiAsp (n = 16)	Aspart (n = 16)	P
2 h after exercise (0–160 min)
MIE
CGM% time 3.9–10 mM, median (IQR)	100.0 (79.17, 100.0)	100.0 (95.83, 100.0)	0.423
CGM% time 3.9–7.8 mM, median (IQR)	66.67 (45.83, 77.08)	64.58 (50.00, 91.67)	0.146
CGM% time >10 mM, median (IQR)	0.00 (0.00, 4.17)	0.00 (0.00, 0.00)	0.418
CGM% time >13.9 mM, median (IQR)	0.00 (0.00, 0.00)	0.00 (0.00, 0.00)	0.964
CGM% time <3.9 mM, median (IQR)	0.00 (0.00, 0.00)	0.00 (0.00, 0.00)	0.799
CGM% time <3.0 mM, median (IQR)	0.00 (0.00, 0.00)	0.00 (0.00, 0.00)	0.964
Mean glucose, median (IQR)	6.89 (6.23, 7.88)	6.64 (5.95, 7.45)	0.326
Coefficient of variation, median (IQR)	15.72 (10.75, 23.09)	16.25 (10.89, 26.35)	0.756
Total insulin delivery (U), median (IQR)	4.61 (2.72, 8.56)	2.99 (1.03, 6.30)	0.234
HIE
CGM% time 3.9–10 mM, median (IQR)	89.58 (77.68, 100.0)	100.0 (77.08, 100.0)	0.938
CGM% time 3.9–7.8 mM, median (IQR)	62.50 (37.50, 79.17)	75.00 (52.08, 89.58)	0.234
CGM% time >10 mM, median (IQR)	0.00 (0.00, 18.75)	0.00 (0.00, 0.00)	0.093
CGM% time >13.9 mM, median (IQR)	0.00 (0.00, 0.00)	0.00 (0.00, 0.00)	n/a
CGM% time <3.9 mM, median (IQR)	0.00 (0.00, 2.08)	0.00 (0.00, 18.75)	0.344
CGM% time <3.0 mM, median (IQR)	0.00 (0.00, 0.00)	0.00 (0.00, 0.00)	0.317
Mean glucose, median (IQR)	7.00 (5.49, 8.13)	6.42 (6.04, 7.13)	0.326
Coefficient of variation, median (IQR)	20.33 (12.37, 25.93)	17.46 (11.03, 26.99)	1
Total insulin delivery (U), median (IQR)	4.36 (2.91, 6.87)	2.37 (1.05, 6.18)	0.121

Data are median (IQR).

In the 8 h after exercise commencement, median TIR was >77% for all exercise conditions, and time in hyperglycemia was <19%. Time in hypoglycemia was significantly higher with insulin aspart compared to FiAsp following HIE (median [IQR]: 6.77 [0.00, 10.42]% vs. 0.00 [0.00, 4.76]%), P = 0.043, respectively), while no significant difference for glycemic or insulin delivery outcomes was observed with MIE (Supplementary Table S3).

Overnight postexercise (00:00–06:00), median TIR was 97.2% for the FiAsp-MIE condition and 100% for all other conditions. Time in hyperglycemia and hypoglycemia was 0% for all exercise conditions. There were no significant differences between FiAsp and insulin aspart for any glycemic or insulin delivery outcomes (Table 4).

Table 4.

Glucose Control And Insulin Delivery Through CL in the Overnight Period (00:00–05:59) Following MIE and HIE with Fast-Acting Insulin (FiAsp) and Insulin Aspart

CGM parameter	FiAsp (n = 16)	Aspart (n = 16)	P
Overnight after exercise (00:00–05:59)
MIE
CGM% time 3.9–10 mM, median (IQR)	97.22 (75.00, 100.0)	100.0 (81.25, 100.0)	0.321
CGM% time 3.9–7.8 mM, median (IQR)	69.44 (39.58, 90.97)	75.69 (42.36, 100.0)	0.623
CGM% time >10 mM, median (IQR)	0.00 (0.00, 25.00)	0.00 (0.00, 18.75)	0.385
CGM% time >13.9 mM, median (IQR)	0.00 (0.00, 0.00)	0.00 (0.00, 0.00)	0.317
CGM% time <3.9 mM, median (IQR)	0.00 (0.00, 0.00)	0.00 (0.00, 0.00)	0.302
CGM% time <3.0 mM, median (IQR)	0.00 (0.00, 0.00)	0.00 (0.00, 0.00)	0.964
Mean glucose, median (IQR)	6.86 (5.99, 8.01)	6.78 (5.74, 8.07)	0.918
Coefficient of variation, median (IQR)	15.41 (12.74, 22.66)	13.84 (8.12, 21.01)	0.408
Total insulin delivery (U), median (IQR)	6.56 (5.21, 11.72)	6.16 (5.12, 8.55)	0.063
HIE
CGM% time 3.9–10 mM, median (IQR)	100.0 (88.19, 100.0)	100.0 (88.89, 100.0)	0.957
CGM% time 3.9–7.8 mM, median (IQR)	87.50 (47.22, 100.0)	69.24 (43.75, 100.0)	0.407
CGM% time >10 mM, median (IQR)	0.00 (0.00, 0.00)	0.00 (0.00, 11.11)	0.738
CGM% time >13.9 mM, median (IQR)	0.00 (0.00, 0.00)	0.00 (0.00, 0.00)	0.158
CGM% time <3.9 mM, median (IQR)	0.00 (0.00, 0.00)	0.00 (0.00, 0.00)	0.084
CGM% time <3.0 mM, median (IQR)	0.00 (0.00, 0.00)	0.00 (0.00, 0.00)	n/a
Mean glucose, median (IQR)	6.47 (5.77, 7.79)	7.16 (6.35, 8.34)	0.379
Coefficient of variation, median (IQR)	15.83 (7.32, 20.19)	14.81 (10.92, 18.79)	0.959
Total insulin delivery (U), median (IQR)	7.16 (4.81, 10.86)	6.95 (5.39, 10.81)	0.959

Data are median (IQR).

In the overall AHCL period (6 weeks per stage), TIR was significantly higher for FiAsp versus insulin aspart, with median TIR >79% for both FiAsp and insulin aspart (82.3 [78.8, 83.6]% vs. 79.3 [77.5, 82.6]% for FiAsp and insulin aspart, respectively [median [IQR], P = 0.018). Complete CGM outcomes for the overall AHCL period can be seen in Supplementary Table S4.

System performance

Median % CGM time in CL for each 6-week stage was 99.8% in the insulin aspart group and 99.9% in the FiAsp group, while median % of total time in CL for the study duration was 94.7% in the insulin aspart group and 95.1% in the FiAsp group (n = 16). Sensor MARD of 7.8% was identical in both faster aspart and insulin aspart groups, using self-monitoring blood glucose values as a reference.

Safety outcomes

No episode of severe hypoglycemia, diabetic ketoacidosis, or other serious adverse events was reported during the study. The number of pump alarms relating to no insulin delivery across the 12-week study period, presumed due to line occlusion, was not different between insulin aspart and FiAsp (38 alarms in 9 [36%] participants vs. 20 alarms in 6 [24%] participants, respectively, P = 0.22).

Discussion

This study is the first to compare the use of FiAsp versus insulin aspart in conjunction with a second-generation CL system in adults with type 1 diabetes undertaking MIE and HIE. Regardless of the insulin formulation, we demonstrated excellent overall TIR (3.9–10.0 mM) with a median exceeding 80% in the 24 h following exercise commencement and TIR approaching 100% in the 2 h postexercise, as well as overnight. Overall, these findings demonstrate that FiAsp does not convey a clinical advantage in glycemic control over insulin aspart for those people living with type 1 diabetes, using a second-generation CL system, who wish to exercise.

Current CGM consensus statements recommend that people with type 1 diabetes target a TIR >70%, a time above range (>10.0 mM) of <25% and a time below range (<3.9 mM) of <4%.²⁹ In the 24 h following exercise commencement, each of these recommendations was exceeded following MIE and HIE, for both FiAsp and insulin aspart, with TIR greater than 80% across all exercise conditions. Although the results of this study represent that of a cohort with already well-managed glycemia at baseline, these outcomes should reassure those living with type 1 diabetes wishing to undertake exercise as prior reports indicate that the fear of hypoglycemia not only during and immediately following exercise but also particularly nocturnal hypoglycemia and subsequent glycemic instability the following day has been a significant disincentive.⁴ Indeed, TIR overnight following exercise exceeded the recommendations by an even greater magnitude and approached 100% in all conditions.

The 8 h after exercise period represented the time interval with the greatest glycemic variability, likely reflecting the changes in insulin sensitivity following exercise impacting the insulin dosing for the evening meal. With HIE, the time below range following the postexercise meal was ∼7% with insulin aspart, significantly greater than when infusing FiAsp, which parallels previous research suggesting that FiAsp may provide improved outcomes with mealtime dosing.^22,24

The proactive interventions employed during the study were aimed at reducing residual insulin action and maintaining glucose levels at the higher end of the target range to minimize the risk for hypoglycemia, should glucose levels fall with exercise. These strategies included a standard lunch 4 h before exercise to minimize mealtime insulin on board, a reminder to set a temporary target 2 h pre-exercise, and instructions to treat with 15 g carbohydrate if blood glucose levels dropped below 3.9 mM or were <7.0 mM 10 min pre-exercise, all steps that would be recommended based on current guidelines for exercise and type 1 diabetes.²⁸ The aforementioned measures subsequently proved to be necessary as with both FiAsp and insulin aspart, we observed a sharp decline in glucose level during exercise, most pronounced during MIE, with a fall of ∼3.0 mM during the 40-min exercise period.

We conclude that the pharmacokinetic differences between FiAsp and insulin aspart are not of a sufficient magnitude to meaningfully impact residual insulin action during exercise and there remains a need for forward planning so that strategies can be implemented to address this. This complements previously published data from our group showing that despite the rapid reductions in insulin delivery provided by the HCL algorithm in response to a declining blood glucose with exercise, changes in circulating free insulin levels do not closely mirror changes in insulin infusion rates. The prevention of hypoglycemia during exercise itself is largely facilitated by the activation of endogenous counter-regulatory mechanisms.^30,31 Although FiAsp may have a shorter duration of action when compared to insulin aspart, our data demonstrate that the incremental shift in pharmacokinetics, particularly the offset in insulin action, remains too slow to respond to the rapidly changing insulin requirements with exercise even when used in conjunction with a second-generation CL system.

To the best of our knowledge, only one previous study has assessed FiAsp use with CL during exercise, although this employed a first-generation CL system and a single type of exercise intervention with outcomes focusing upon exercise and the time immediately following its completion. Dovc et al.²³ demonstrated results comparable to this study for moderate-vigorous interval exercise, where no difference was evident for TIR or hypoglycemia between FiAsp and insulin aspart during 2 h postexercise using a first-generation fully CL system with a fuzzy logic-based algorithm. However, this study also demonstrated that FiAsp does not prevent the precipitous drop in blood glucose during continuous MIE and is the first to analyze glycemic control in the 24 h, including the overnight period, following exercise when using FiAsp and CL insulin delivery.

The lack of advantage with FiAsp over insulin aspart with exercise contrasts with findings relating to mealtime glycemia, where data have shown a benefit for FiAsp compared to insulin aspart^21,22,24,25 under free-living conditions and in the postprandial period. Of note, recent data from our group have shown that FiAsp is superior to insulin aspart when infused through the AHCL system for postprandial glucose control when the insulin bolus is either administered late (20 min postmeal) or omitted.²⁴ In addition, data from the entire 6-week CL period demonstrate a significantly higher TIR for FiAsp when compared to insulin aspart. This improvement in TIR was attributable to differences in postprandial glycemic control.²⁴

The onset and offset in action of subcutaneously administered rapid acting insulin are asymmetric with a prolonged offset.³² Thus, while the earlier onset of action of FiAsp, which is more pertinent to mealtime requirements, appears to provide an improvement of postprandial glucose, the difference in the offset in action of FiAsp compared with insulin aspart is insufficient to improve glucose control with exercise.

A strength of this study is that it represents a realistic best-case scenario for pre-exercise conditions for people with type 1 diabetes using CL insulin delivery. Participants attended the clinical trials center only for the 40-min duration of exercise and returned home immediately after cessation of exercise, making the study generalizable to the population with type 1 diabetes. Combined with the 24-h duration of study analysis, this highlights that exercise can be undertaken with a second-generation CL system in people with type 1 diabetes without a high risk of postexercise hypoglycemia or next-day glycemic variability, when appropriate clinical recommendations are followed.

Conversely, we recognize the exploratory nature of this study, and it cannot be completely ruled out that FiAsp provides superior glucose control to insulin aspart in people with type 1 diabetes, who wish to exercise. While this study is in keeping with previous findings,²³ it remains possible that the data may not be adequately statistically powered to detect changes in TIR at the level observed in this study. Lack of blinding with regard to insulin formation for both investigators and participants is a confounding factor in this study, while the population had well-managed glycemia at baseline and were experienced in diabetes technology, which limits the generalizability of findings to the overall population. The use of the second-generation CL system also precludes generalization to other CL systems employing different control algorithms.

Also, we acknowledge that many people with type 1 diabetes may not adopt all the recommended practices implemented in this study (i.e., minimizing insulin on board and setting temporary targets) and thus, “real-world” glycemic outcomes may be inferior to those seen in this study. Finally, the active insulin time was fixed at 4 h for both FiAsp and insulin aspart during this study for safety purposes. However, AHCL permits an active insulin time as low as 2 h, which tunes the algorithm to become more aggressive. Such tuning might enhance the differences between the two insulin formulations, particularly in the postprandial period, where data have shown FiAsp to be more effective than insulin aspart.^21,22,24,25 With exercise, unlike meals, the offset in insulin action is likely the key pharmacokinetic parameter benefiting glucose control, and as such, a reduction in the active insulin action time enabling more aggressive insulin delivery would be of less relevance.

Conclusion

FiAsp delivery using a second-generation CL system was similar in terms of safety and glucose control both during and 24 h following MIE and HIE when compared to insulin aspart. While FiAsp has a faster onset and offset of action compared to standard insulin aspart, there was no clinically meaningful benefit demonstrated for glucose control during, or in the 24 h following exercise. We recognize that complete restoration of the physiological responsiveness in insulin action may not be feasible when insulin is administered subcutaneously and not directly into the portal circulation. Nevertheless, insulin formulations with an offset in action faster than FiAsp are needed to provide an improvement in CL control during exercise. We also recognize that our findings are limited to a cohort of individuals with already well-managed glycemia at baseline, and specifically to exercise, and that FiAsp may yet have an overall positive effect due to its faster onset of action in reducing glucose excursions during postprandial periods.

Data Availability

The datasets generated during and/or analyzed during this study are available from the corresponding author on reasonable request.

Footnotes

Authors' Contributions

D.M., D.P.Z., M.H.L., B.P., S.V., and D.N.O. contributed to the study design. D.M., D.P.Z., and D.N.O. assisted with implementation of the study. D.M., D.P.Z., S.V., and D.N.O. contributed to data analysis and interpreted the data. B.G., A.R., and N.K. provided technical support for the study and reviewed the final article for technical accuracy. All authors critically reviewed the article. All authors approved the final version of the article. D.N.O. is the guarantor of the study and had full access to the data and accepts full responsibility for the conduct of the study, the integrity of data, and the accuracy of data analysis.

Acknowledgments

The authors are grateful to all the participants for their dedication and commitment to this study. The authors acknowledge Ms. Catriona Sims (University of Melbourne Department of Medicine, Melbourne, Australia) and Ms. Marion Jamieson (Medtronic, Melbourne, Australia) for providing administrative and logistical support.

Author Disclosure Statement

M.H.L. and B.P. have received speaker honoraria from Medtronic. D.P.Z. has received speaker honoraria from Medtronic, Insulet, and Ascensia Diabetes Care. B.G., A.R., and N.K. are employees of Medtronic. D.N.O. has served on advisory boards and received research support and honoraria from Medtronic, Novo Nordisk, Sanofi, and Abbott. No other potential conflicts of interest relevant to this article were reported.

Funding Information

Funds were provided for this investigator-initiated study by Medtronic and Novo Nordisk. Material support was provided from Medtronic and Novo Nordisk. M.H.L. is supported by a National Health and Medical Research Council (NHMRC) postgraduate scholarship, cofounded by Diabetes Australia. B.P. is supported by a University of Melbourne scholarship and research support from JDRF. D.Z. is supported by the Leona M. and Harry B. Helmsley Charitable Trust and the International Society for Pediatric and Adolescent Diabetes (ISPAD) Fellowship. The funders of the study had no role in the study design; the collection, analysis, and interpretation of data; and writing the report; and did not impose any restriction regarding the publication of the report.

Supplementary Material

Supplementary Figure S1

Supplementary Figure S2

Supplementary Table S1

Supplementary Table S2

Supplementary Table S3

Supplementary Table S4

References

Chimen

, Kennedy

, Nirantharakumar

, et al.: What are the health benefits of physical activity in type 1 diabetes mellitus? A literature review. Diabetologia, 2012; 55:542–551.

Warburton

, Bredin

: Health benefits of physical activity: a systematic review of current systematic reviews. Curr Opin Cardiol, 2017; 32:541–556.

Colberg

, Sigal

, Yardley

, et al.: Physical activity/exercise and diabetes: a position statement of the American Diabetes Association. Diabetes Care, 2016; 39:2065–2079.

Lascar

, Kennedy

, Hancock

, et al.: Attitudes and barriers to exercise in adults with type 1 diabetes (T1DM) and how best to address them: a qualitative study. PLoS One, 2014; 9:e108019.

Nimri

, Muller

, Atlas

, et al.: MD-Logic overnight control for 6 weeks of home use in patients with type 1 diabetes: randomized crossover trial. Diabetes Care, 2014; 37:3025–3032.

Sharifi

, De Bock

, Jayawardene

, et al.: Glycemia, treatment satisfaction, cognition, and sleep quality in adults and adolescents with type 1 diabetes when using a closed-loop system overnight versus sensor-augmented pump with low-glucose suspend function: a randomized crossover study. Diabetes Technol Ther, 2016; 18:772–783.

Thabit

, Tauschmann

, Allen

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

Thabit

, Lubina-Solomon

, Stadlet

, et al.: Home use of closed-loop insulin delivery for overnight glucose control in adults with type 1 diabetes: a 4-week, multicentre, randomised crossover study. Lancet Diabetes Endocrinol, 2014; 2:701–709.

Weisman

, Bai

, Cardinez

, et al.: 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; 5:501–512.

10.

Doyle

, Huyett

, Lee

, et al.: Closed-loop artificial pancreas systems: engineering the algorithms. Diabetes Care, 2014; 37:1191–1197.

11.

Sandoval

, Guy

, Richardson

, et al.: Effects of low and moderate antecedent exercise on counterregulatory responses to subsequent hypoglycemia in type 1 diabetes. Diabetes, 2004; 53:1798–1806.

12.

Yardley

, Kenny

, Perkins

, et al.: Resistance versus aerobic exercise: acute effects on glycemia in type 1 diabetes. Diabetes Care 2013;36 537–542.

13.

Frayn

, Karpe

: Regulation of human subcutaneous adipose tissue blood flow. Int J Obes, 2014; 38:1019–1026.

14.

Mallad

, Hinshaw

, Schiavon

, et al.: Exercise effects on postprandial glucose metabolism in type 1 diabetes: a triple-tracer approach. Am J Physiol Endocrinol Metab, 2015; 308:E1106–E1115.

15.

Guerci

, Sauvanet

: Subcutaneous insulin: pharmacokinetic variability and glycemic variability. Diabetes Metab, 2005; 31:4S7–4S24.

16.

McAuley

, Horsburgh

, Ward

, et al.: Insulin pump basal adjustment for exercise in type 1 diabetes: a randomised crossover study. Diabetologia, 2016; 59:1636–1644.

17.

Mudaliar

, Lindberg

, Joyce

, et al.: Insulin aspart (B28 asp-insulin): a fast-acting analog of human insulin: absorption kinetics and action profile compared with regular human insulin in healthy nondiabetic subjects. Diabetes Care, 1999; 22:1501–1506.

18.

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; 17:682–688.

19.

Heise

, Stender-Petersen

, Hövelmann

, et al.: Pharmacokinetic and pharmacodynamic properties of faster-acting insulin aspart versus insulin aspart across a clinically relevant dose range in subjects with type 1 diabetes mellitus. Clin Pharmacokinet, 2017; 56:649–660.

20.

Zijlstra

, Demissie

, Graungaard

, et al.: Investigation of pump compatibility of fast-acting insulin aspart in subjects with type 1 diabetes. J Diabetes Sci Technol, 2018; 12:145–151.

21.

Hsu

, Buckingham

, Basina

, et al.: Fast-acting insulin aspart use with the MiniMedTM 670G System. Diabetes Technol Ther, 2021; 23:1–7.

22.

Boughton

, Hartnell

, Thabit

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

23.

Dovc

, Piona

, Mutlu

, et al.: Faster compared with standard insulin aspart during day-and-night fully closed-loop insulin therapy in type 1 diabetes: a double-blind randomized crossover trial. Diabetes Care, 2020; 43:29–36.

24.

Lee

, Paldus

, Vogrin

, et al.: Fast-acting insulin aspart versus insulin aspart using a second-generation hybrid closed-loop system in adults with type 1 diabetes: a randomized, open-label, crossover trial. Diabetes Care, 2021; dc210814.

25.

Ozer

, Cooper

, Ahn

, et al.: Fast acting insulin aspart compared to insulin aspart in the Medtronic 670G hybrid closed loop system in type 1 diabetes: an open label crossover study. Diabetes Technol Ther, 2021; 23:286–292.

26.

Mier

, Alexander

, Mageean

: Achievement of VO2max criteria during a continuous graded exercise test and a verification stage performed by college athletes. J Strength Cond Res, 2012; 26:2648–2654.

27.

Lee

, Vogrin

, Paldus

, et al.: Glucose control in adults with type 1 diabetes using a medtronic prototype enhanced-hybrid closed-loop system: a feasibility study. Diabetes Technol Ther, 2019; 21:499–506.

28.

Riddell

, Gallen

, Smart

, et al.: Exercise management in type 1 diabetes: a consensus statement. Lancet Diabetes Endocrinol, 2017; 5:377–390.

29.

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; 42:1593–1603.

30.

Jayawardene

, McAuley

, Horsburgh

, et al.: Closed-loop insulin delivery for adults with type 1 diabetes undertaking high-intensity interval exercise versus moderate-intensity exercise: a randomized, crossover study. Diabetes Technol Ther, 2017; 19:340–348.

31.

Lee

, Vogrin

, Paldus

, et al.: Glucose and counterregulatory responses to exercise in adults with type 1 diabetes and impaired awareness of hypoglycemia using closed-loop insulin delivery: a randomized crossover study. Diabetes Care, 2020; 43:480–483.

32.

Heise

, Zijlstra

, Nosek

, et al.: Pharmacological properties of faster-acting insulin aspart vs insulin aspart in patients with type 1 diabetes receiving continuous subcutaneous insulin infusion: a randomized, double-blind, crossover trial. Diabetes Obes Metab, 2017; 19:208–215.

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.

0.00 MB

0.05 MB

0.34 MB

0.01 MB

0.02 MB

0.01 MB