Fasted High-Intensity Interval and Moderate-Intensity Exercise Do Not Lead to Detrimental 24-Hour Blood Glucose Profiles

Sam N. Scott; Matt Cocks; Rob C. Andrews; Parth Narendran; Tejpal S. Purewal; Daniel J. Cuthbertson; Anton J. M. Wagenmakers; Sam O. Shepherd

Disclosures

J Clin Endocrinol Metab. 2019;104(1):111-117. 

In This Article

Research Design and Methods

Fourteen sedentary people with T1D (six men, eight women; age, 26 ± 3 years; body mass index, 27.6 ± 1.3 kg/m2; V̇O2peak, 30.8 ± 2.0 m/kg/min; duration of T1D, 8.2 ± 1.4 years) on a basal-bolus insulin regimen completed the study. Exclusion criteria were duration of T1D <6 months, insulin pump therapy, poor diabetes control (HbA1c >86 mmol/mol), poor diabetes control (HbA1c >86 mmol/mol), frequent hypoglycemia (>5 per week), and/or hypoglycemia unawareness [determined from medical history as patients having no symptoms prior to or at the time of a blood sugar ≤70 mg/dL (3.9 mmol/L) within the last 3 months], obesity (body mass index >30 kg/m2), pregnancy or planning pregnancy, uncontrolled hypertension (>180/100 mm Hg), angina, autonomic neuropathy, taking any medication that affects heart rate, major surgery planned within 6 weeks of the study, and severe nonproliferative and unstable proliferative retinopathy. Testing took place in the laboratory of the School of Sport and Exercise Sciences at Liverpool John Moores University. The study was approved by the Black Country NHS Research Ethics Committee (West Midlands, UK), and all participants gave written informed consent to a protocol conforming to the Declaration of Helsinki.

Pre-experimental Procedures

Participants first performed an incremental exercise test to exhaustion on an electromagnetically braked cycle ergometer (Excalibur Sport V2.0; Lode, Groningen, Netherlands) to determine maximal aerobic power output and V̇O2peak using an online gas collection system (MOXUS modular oxygen uptake system; AEI Technologies, Pittsburgh, PA). This information was used to determine appropriate workloads for subsequent exercise trials. The test consisted of 3-minute stages starting at 60 W, and the workload was increased by 35 W at each stage until subjects could not maintain a cadence of >50 rpm. V̇O2peak was taken as the highest value achieved over a 15-second recording period. Participants also completed a food diary over a minimum of 3 days to calculate their habitual caloric and macronutrient intake.

Study Design and Experimental Protocol

Participants completed a randomized, counterbalanced, crossover experiment consisting of three intervention periods: CON, HIT, and MICT (see Figure 1 for protocol overview). Each intervention period lasted 24 hours, during which the effect of a single session of exercise on subsequent 24-hour glycemic control and risk of hypoglycemia were assessed under standardized dietary, but otherwise free-living, conditions. Periods were identical except for the exercise performed. Participants had a continuous glucose monitor system (CGMS) probe (G4 Platinum; Dexcom, San Diego, CA) inserted subcutaneously into the abdomen at least 24 hours before the intervention period to allow time for a "bedding in" period. Participants were trained on how to use the CGMS and were instructed to calibrate the device a minimum of four times daily using capillary blood tests. Participants were not blinded to the CGMS (i.e., they could see their glucose values). Twenty-four hours after the CGMS was inserted, participants completed the control intervention. Participants did not attend the laboratory on the control day but were provided a standardized diet to consume while going about their normal daily activities.

Figure 1.

Study protocol. At 24 h after the CGMS was inserted, participants completed the control day. Participants did not attend the laboratory on the control day but were provided a standardized diet to consume while going about their normal daily activities. After the control day, participants completed the two exercise intervention periods in a randomized order separated by at least 48 h. The exercise intervention periods were identical to the control intervention except that participants attended the laboratory after an overnight fast and having omitted their short-acting insulin to perform a bout of either HIT or MICT. Wmax, maximal aerobic power output.

The standardized diet was matched to each participant's habitual energy intake and consisted of three meals (breakfast, lunch, and dinner; 50% carbohydrate, 30% fat, 20% protein). Participants were instructed to consume these meals at predetermined time points throughout the day. Participants only consumed the food provided by the research team during this period. Additional snacks were permitted only to prevent hypoglycemia. The diet was a 2-day rolling diet and was matched for macronutrient and energy content between days, which ensured that participants consumed the same food on the experimental days. Participants were also instructed to abstain from caffeine, alcohol, and vigorous exercise. Participants completed a food diary to confirm that they had eaten the prescribed food at the correct times.

After the control day, participants completed the two exercise intervention periods in a randomized order separated by at least 48 hours. The exercise intervention periods were identical to the control intervention except participants attended the laboratory after an overnight fast and having omitted their short-acting insulin to perform a bout of either HIT or MICT. After the exercise, participants left the laboratory and returned to their normal daily activities. As on the control day, participants were provided a standardized diet to consume. This diet was identical to the control day, and participants consumed each meal at the same predetermined time points throughout the day. Insulin dosage was not recorded.

Exercise Protocols

Both exercise protocols were conducted on a stationary cycle ergometer (Excalibur Sport V2.0; Lode, Groningen, Netherlands) and were preceded by a standardized 5-minute warm-up at 50 W. MICT consisted of 30 minutes of continuous cycling at a workload equivalent to 65%. V̇O2peak HIT consisted of six 1-minute intervals at a workload equivalent to 100% V̇O2peak interspersed with 1 minute of rest. The total time commitment of the HIT protocol (17 minutes) was approximately half that of the MICT protocol (35 minutes).

Acute Change in Blood Glucose With Exercise

Blood glucose concentration was recorded before and after exercise through capillary fingertip sampling to ensure that blood glucose levels were between 7 and 14 mmol/L, in accordance with the guidelines we developed in the Exercising for Type 1 Diabetes study,[11] meaning participants were safe to commence exercise and also safe to leave after exercise. If blood glucose was <7 mmol/L before exercise, 20 g of glucose was ingested. If blood glucose was >14 mmol/L, a light walk or insulin was advised, blood ketones were checked.[12]

Statistical Analyses

Continuous glucose monitor data were downloaded from the device using Dexcom Studio™ software (12.0.4.6). Data from the CGMS were analyzed in accordance with the International Consensus on Use of Continuous Glucose Monitoring guidelines.[10] A one-way ANOVA with repeated measures was used to assess glycemic control between the three conditions using the following metrics: percentage of time in level 1 hypoglycemia (≤3.9 mmol/L), percentage of time in level 2 hypoglycemia (≤2.9 mmol/L), time in target range (4 to 10 mmol/L), and hyperglycemia (≥10 mmol/L). Mean glucose and glycemic variability using coefficient of variation were compared between conditions. Episodes of level 1 and 2 hypoglycemia and hyperglycemia were compared between conditions. The 24-hour period was defined as 8:00 PM to 8:00 AM, and the nocturnal period was defined as 12:00 AM to 6:00 AM. A two-factor repeated measures ANOVA was used to assess whether there was an acute change in blood glucose concentration after HIT and MICT in the fasted state with the within subject factors "training mode" and "time point." All analyses were performed using SPSS Statistics for Windows, Version 22.0 (IBM Corp., Armonk, NY). Data are presented as mean ± SEM, and significance was set at P ≤ 0.05.

Comments

3090D553-9492-4563-8681-AD288FA52ACE

processing....