Examining the Effects of Morning and Evening Exercise on Glucose Regulation in Adults With Prediabetes

Participant Recruitment, Screening, and Randomization. Participants (n=20, females=10) will be recruited from the University of Colorado Denver Anschutz Medical Campus (CU AMC) and the surrounding Denver Metro area using campus emails and flyers. Volunteers who meet initial eligibility criteria will be scheduled for an in-person screening visit including screening labs (thyroid stimulating hormone (TSH), complete blood count (CBC), comprehensive metabolic panel (CMP), lipid panel, glycated hemoglobin (A1c), urine pregnancy test in women (beta human chorionic gonadotropin), 12-lead electrocardiogram (ECG), medical history review, and physical exam with a physician or advanced practice provider. Potential participants will complete the physical activity readiness questionnaire (PAR-Q) to further screen for contraindications to exercise. 38 Potential participants will also be screened for obstructive sleep apnea using the WatchPat (Itamar Medical, Israel) with an apnea hypopnea index score > 5 being exclusionary as it significantly affects systemic metabolism. The WatchPat uses a combination of peripheral arterial tone, pulse oximetry, heart rate, accelerometers, and a snore detector to measure respiratory effort-related arousals and has been validated. 39,40 Participants who meet study eligibility criteria below will be block randomized in a 1:1 ratio (stratified by sex) to complete either AM-Ex or PM-Ex first in the crossover study.

Study Protocol and Assessments. The study will be performed at the CU AMC Anschutz Health and Wellness Center (AHWC) and the CU AMC Clinical Translational Research Center (CTRC). See Facilities and Other Resources for details of the AHWC and CTRC. Participants will undergo assessments at baseline and again after completing each of the exercise conditions.

Supervised AM and PM Exercise Conditions. Participants will complete a 3-day aerobic exercise condition (Note: 1 bout of exercise has been shown to shift melatonin). Exercise will be performed at 70% of age-predicted MHR for 45 minutes per day. Individual participant exercise time windows will be defined by the individual's habitual wake time (measured by actigraphy), with AM-Ex occurring at 0-3 hours and PM-Ex occurring at 9-12 hours post habitual wake time. These time windows were chosen to define morning and evening for several reasons. Firstly, during the traditional workday, most adults engage in exercise either prior to their work day or after their work day before bed. Secondly, these time windows have been demonstrated to be when circadian patterns of SI and melatonin are most evident and consistent. 13,35 Exercise sessions will be performed on the treadmill at the AHWC and supervised by study staff. HR monitors (Polar H10) will be worn to ensure achievement of target intensity. Session time (time of day), duration, and mean intensity will be recorded for each session. Exercise sessions will be separated by a 3-4 week washout period to both minimize training effects on SI and to account for phases of the menstrual cycle, as different menstrual phases have been shown to exhibit differing SI.

Dietary Prescription. During each condition, participants will be asked to eat a balanced diet consisting of approximately 50% carbohydrate, 30% fat, 20% protein. A controlled diet was chosen to reduce the effects of different meal compositions on the following day insulin clamp trial and so a standardized free-living day could be examined via CGM.

Assessments. All measures will be performed at baseline and after each exercise condition.

Insulin sensitivity (hyperinsulinemic-euglycemic clamp): Protocol derived from original glucose clamp validation study. 45 After fasting overnight for 12 hours, participants will arrive at the CTRC 1 hour after habitual wake time and baseline vital signs and capillary blood glucose will be obtained. Two IVs will be placed by nursing staff on contralateral arms, one in the antecubital vein and one in the wrist. Insulin (42 units/700mL NS concentration) will be started at 40mU/m2/min and dextrose 20% solution will also be started and titrated to maintain a capillary blood glucose of 90ng/dL. Potassium phosphate (5mmol/100mL NS) will also be run at 4mL/hr to prevent hypokalemia. Following 100 minutes of euglycemia, capillary blood glucose and serum insulin will be obtained every 10 minutes x3. At 120 minutes, a serum glucose, insulin, triglycerides, FFA, glycerol, and lactate will be drawn, insulin infusion will be stopped, and the patient will be fed a meal. Thirty minutes post-meal the dextrose will be stopped, and the participant will be monitored for hypoglycemia until discharged. SI will be calculated using glucose infusion rate, urinary losses of glucose, and space correction via the formula described previously.

Free living average glucose, time in range (TIR), and glucose variability (continuous glucose monitor): A Continuous Glucose Monitoring System (CGM; FreeStyle Libre Pro; Abbott Industries, Chicago, IL, USA) will be placed on the first day of baseline and exercise conditions. The study participants will be trained in how to maintain the device. Participants will wear the monitor for 5 days and return the device at the post condition visit (device in place from days 1 to 5). Briefly, a sensor is inserted at a 45° angle just under the skin on the participant's arm using a removable introducer needle. A small, flexible electrode is left behind that will remain embedded in the skin for 5 days. The insertion is quick, and very little discomfort is experienced. The participant's interstitial fluid glucose reacts with the sensor's glucose oxidase layer, producing an electrical signal that is recorded by the digital receiver. The system stores an average glucose value every 15 minutes and participants will be blinded to all glucose data. All glucose data are retrieved via the scanning reader, which will be utilized prior to device removal.

Salivary Dim Light Melatonin Offset (DLM-Off): Participants will be instructed to avoid over the counter or melatonin products or supplements containing melatonin for the duration of the study. On the day prior to DLM-Off testing, participants will be instructed to avoid caffeine, alcohol, marijuana, nonsteroidal anti-inflammatory medications, and will follow the prescribed diet. In the evening prior to DLM-Off testing they will be instructed to place light blocking goggles, 1 saliva collection tube, sample log, and a pen at bedside and to set their alarm for the next morning at their habitual wake time. On the morning of the DLM-Off collection, participants will remain in bed and immediately don the light blocking goggles. Five minutes after awakening, participants will collect the first saliva sample in the tube at their bedside and record the time. The saliva collection will be completed within 5 minutes. This tube will be placed in the freezer and transported to the CTRC on ice by the participant. Light blocking goggles will be worn until arrival at the CTRC in a light-controlled room. Saliva samples will then be collected by study staff hourly from hours 1-5 after awakening.

Serum AM cortisol: On arrival at the CTRC for the post condition visit, and after IV placement for hyperinsulinemic-euglycemic clamp (see above), a blood sample for AM serum cortisol will be taken prior to the onset of the insulin infusion.

Device-measured Physical Activity (PA) and Sedentary Behavior (SB): With each condition, PA and SB will be measured using the activPAL (PALTechnologies: Glasgow, Scotland) activity monitor. The activPAL will be given to the participant on day 1 of the condition and returned at the post condition visit. The activPAL is a small (53 x 35 x 7 mm) and light (15 grams) device that uses accelerometer-derived information about thigh position to estimate time spent in different body positions (i.e., sitting/lying, standing, and stepping). The device is attached using a non-allergenic adhesive pad and positioned on the midline of the thigh, one-third of the way between the hip and knee and has been validated as an objective measure of PA and SB.

Sleep duration, sleep quality, and sleep timing (wrist actigraphy, questionnaire): Wrist actigraphy (Actiwatch, Philips Respironics) is well validated for assessment of sleep duration and quality and is commonly used for free-living sleep assessment48. The Actiwatch will be worn on the non-dominant wrist for 5 consecutive days. In addition to assessing sleep using the Actiwatch, participants will be asked to keep a sleep diary for these same 5 days. The sleep diary will include bedtime, wake time, daytime naps, and a rating of sleep quality. Daytime sleepiness will be assessed using the Epworth Sleepiness Scale (ESS). The ESS provides a valid measure of daytime sleepiness over the past seven days.

Chronotype: Individual chronotype will be measured at the baseline visit using the Morningness-Eveningness Questionnaire (MEQ).

Menstrual Tracking: For menstruating participants, a menstrual history will be taken at the screening visit and pregnancy will be ruled with a urine pregnancy test prior to the onset of each condition. Sexually active individuals will be required to use contraception throughout the duration of the study protocol. Each measure (baseline, condition 1, and condition 2) will be separated by intervals of 28 days, therefore aligning with menstrual cycle of most menstruating participants. Menstruating participants will be evaluated during the follicular phase with the exercise prescription to begin on day 1 of the menstrual cycle.

Analytic Plan Power Calculation Our sample size and power estimates were based off a similar study investigating the effect of brisk walking on SI in older women. To achieve 80% power with a two-sided paired t-test for this 2x2 crossover design, a sample size of n=20 is necessary. This power estimate assumes a mean difference of 412µmol/min in glucose clearance (Rd), with a standard deviation of the paired differences assumed to be 626.2µmol/min, and the significance level is 0.05. Anticipating a 20% dropout rate, 25 subjects will be enrolled to obtain a final sample size of 20 subjects that have completed all conditions.

Statistical Analyses for Aims 1 and 2 Analyses will be performed using SAS 9.4 software (SAS Institute, Inc (Cary, NC, USA)). Conventional approaches to the interpretation and reporting of results from statistical analyses will be used (2-sided p-values and 95% confidence intervals, and significance levels of 0.05 will be used for all hypothesis tests). The primary comparison will be change in SI (exercise - baseline) with different exercise conditions. Potential changes in SI, glycemic variability, DLM-off timing, serum cortisol peak and offset, light exposure, non-exercise physical activity, sedentary time, sleep duration/timing, and sleep quality/subjective sleepiness will be explored. Within group changes will be analyzed using two-sided paired t-tests for normally distributed data. Continuous variables will be assessed for normality, and transformations will be utilized for non-normally distributed data. Between group differences will be explored using repeated measures ANOVA for normally distributed variables. Summary statistics for feasibility and acceptability outcomes will be calculated.