Journal of Obesity & Metabolic Syndrome

Search

Article

March, 2024 | Vol.33 No.1

J Obes Metab Syndr 2024; 33(1): 54-63

Published online March 30, 2024 https://doi.org/10.7570/jomes23040

Copyright © Korean Society for the Study of Obesity.

Effects of a 2-Week Kinect-Based Mixed-Reality Exercise Program on Prediabetes: A Pilot Trial during COVID-19

So Young Ahn1, Si Woo Lee2, Hye Jung Shin1,2, Won Jae Lee3, Jun Hyeok Kim2, Hyun-Jun Kim2, Wook Song1,2,4,*

1Institute of Sports Science, Department of Physical Education, Seoul National University, Seoul; 2Research Institute, Dr.EXSol Inc., Seoul; 3Department of Physical Education, Kyungnam University, Changwon; 4Institute on Aging, Seoul National University, Seoul, Korea

Correspondence to:
Wook Song
https://orcid.org/0000-0002-8825-6259
Department of Physical Education, Seoul National University, 1 Gwanak-ro, Gwanak-gu, Seoul 08826, Korea
Tel: +82-2-880-7804
Fax: +82-2-886-7804
E-mail: songw3@snu.ac.kr

Received: June 21, 2023; Reviewed : November 7, 2023; Accepted: November 29, 2023

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

Background: Pre-diabetes can develop into type 2 diabetes mellitus, but can prevented by regular exercise. However, the outcomes when combining unsupervised Kinect-based mixed-reality (KMR) exercise with continuous glucose monitoring (CGM) remain unclear. Therefore, this single-arm pilot trial examined changes in blood glucose (BG) concentrations over 672 hours (4 weeks), including a 2-week period of KMR exercise and CGM in individuals with pre-diabetes.
Methods: This was a pre-and post-treatment case-control study with nine participants. General questionnaires were administered and body composition, fasting BG concentrations, and 2-hour oral glucose tolerance test (2-OGTT) results were measured pre-and post-treatment. Weekly average glucose concentrations, hyperglycemia rate, hypoglycemia rate, average glucose concentration over time, amount of physical activity, amount of food intake, and pre- and postprandial BG (immediately and 30, 60, 90, and 120 minutes after lunch) were measured over 4 weeks (pre-test, exercise, and post -test weeks). Glucose concentrations were measured before exercising, between sets, and 30 and 60 minutes after exercise during the 2 weeks of unsupervised exercise (3 days/week).
Results: In all participants, body mass index (27.16±2.92 kg/m²), fasting BG (108.00±7.19 mg/dL), 2-OGTT (162.56±18.12 mg/dL), hyperglycemia rate (P=0.040), and 90-minute postprandial BG (P=0.035) were significantly reduced during the 2 exercise weeks, and the 2-OGTT result (P=0.044) and diastolic blood pressure (DBP) (P=0.046) were significantly reduced at the post -test as compared with the pre-test.
Conclusion: This study found that 2 weeks of unsupervised KMR exercise reduced 2-OGTT, DBP, hyperglycemia rate, and 90-minute postprandial BG concentration. We believed this effect could be identified more clearly in studies involving a larger number of participants and longer durations of exercise.

Keywords: Blood glucose, Continuous glucose monitoring, Extended reality, Pre-diabetes, Physical exercise

Impairment of blood glucose (BG) control in diabetes leads to serious complications, such as heart disease, chronic kidney disease, oral disease, nerve damage, hearing damage, visual impairment, and mental disease, by increasing reactive oxygen species, inflammation, and advanced glycation end-products.1 Given the increasing prevalence of diabetes, it is expected that by 2030, Asia and Africa will have the highest proportion of patients with type 2 diabetes mellitus.2 Additionally, the prevalence of diabetes in Koreans is increasing,3,4 as is the prevalence of pre-diabetes, which is associated with complications similar to those of diabetes, reaching 26.9% in 2018.5 Furthermore, up to 24.6% of patients with pre-diabetes develop diabetes within 10 years.6 These findings suggest that the number of individuals at high risk of developing diabetes is on the rise, and preventive measures may be essential to limit further increases in the prevalence of diabetes.

Previous studies have identified lifestyle modifications, including exercise, as major strategies for diabetes prevention.7-9 While exercise is an effective strategy to prevent diabetes, low compliance and insufficient intensity can limit its effectiveness.10,11 The use of the Kinect-based mixed-reality (KMR) system to monitor an otherwise unsupervised participants’ movements in real time and provide feedback on exercise postures has been shown to improve body composition and/or enhance physical performance while maintaining high compliance.12-14 The utilization of continuous glucose monitoring (CGM) to measure interstitial glucose concentrations during and after exercise has also been of great interest.15,16 A CGM system (CGMS) is a non-invasive device that automatically detects the glucose concentration of the interstitial fluid (ISF) every minute; the ISF glucose concentration is used to estimate BG concentrations.16 Previous studies have demonstrated the feasibility17 and accuracy of implementing CGM while performing various exercises, and found an approximate 10-minute lag between BG and CGMS measurements.18 However, there has been no prior study that utilized CGMS to monitor the entire course of an unsupervised exercise intervention. Therefore, the objective of this study was to investigate changes in ISF glucose concentrations over a period of 4 weeks, including 2 weeks of unsupervised KMR exercise. This research aimed to assess the effectiveness of a non-face-to-face digital health intervention in individuals with pre-diabetes. We hypothesized that this KMR exercise program could improve glucose control, including effects on hypoglycemia, hyperglycemia, mean glucose, fasting glucose, 2-hour oral glucose tolerance test (2-OGTT), and postprandial glucose. The KMR platform may therefore enhance adherence to exercise programs and promote self-management of health among pre-diabetic individuals.

Participants

After considering the likely dropout rate for out study, we initially recruited a group of 12 obese participants with pre-diabetes with a 1:1 ratio of males and females. The study protocol was approved by the Institutional Review Board of Kyungnam University (IRB no. 1040460-A-2022-029). The eligibility criteria were (1) a body mass index (BMI) ≥23 kg/m2;19 (2) age 29 to 59 years; (3) fasting BG concentration 100 to 125 mg/dL; and (4) 2-OGTT result 140 to 199 mg/dL. The exclusion criteria were (1) acute hepatitis or a history of malignant tumors in the previous month; (2) cerebral infarction or myocardial infarction in the previous 6 months; (3) peripheral vascular disease or dementia; (4) participation in regular exercise in the previous 3 months; and (5) engagement in moderate- to high-intensity exercise.

Study procedure

This study was conducted for 4 weeks using a pre-and post-treatment design (Fig. 1). Before participating in the study, all participants were informed of its purpose and design and voluntarily signed a consent form.

Screening/pre-test day

The first week was designated as the “pre-test” week (week 1), with screening and pre-test procedures conducted on the first day of this week. Before the first visit, the participants were instructed to maintain a fasting state for >9 hours, not consume alcohol or caffeine, and avoid extreme exercise within 24 hours. The researchers measured the participants’ blood pressure, body composition, and waist circumference and completed a physical activity questionnaire (The Global Physical Activity Questionnaire was used for screening only [e.g., occupation, leisure activity, etc.]).20 The participants’ fasting BG concentration and 2-OGTT results were also measured. If the fasting BG concentration was between 100 and 125 mg/dL, a 2-OGTT was performed. Participants with 2-OGTT results between 140 and 199 mg/dL were selected as having pre-diabetes. Additionally, the participants were fitted with a CGMS (Adela Healthcare) and provided a smartwatch (Galaxy Fit2; Samsung) to collect weekly glucose and physical activity (steps) data.

Pre-test week (study week 1)

After the pre-test (week 1) was conducted for a week (Monday to Sunday), the screening and pre-tests were performed on Monday. At the end of week 1, participants were asked to report their weekly average glucose concentration, hyperglycemia rate, hypoglycemia rate, and average glucose level by time and amount of physical activity, food intake, and postprandial glucose on Sunday.

Exercise weeks 1 and 2 (study weeks 2 and 3)

During study weeks 2 and 3, the participants were asked to perform the exercise program for 6 days (3 days per week). Each exercise session was conducted using a KMR device (Virtual Mate; My Benefit) at the same place and time, between 7:00 AM and 9:00 AM. The first day of week 2 was designed as an adaptation phase, during which participants were taught how to use the KMR device and perform proper exercise movements. For the five remaining exercise sessions, the participants were instructed to exercise independently without supervision (Supplementary Fig. 1). One hour before each exercise session, the participants were asked to consume a nutritional drink (New Care; Daesang) containing 30 g of carbohydrates, 7 g of protein, and 6 g of fat in order to maintain the same state and prevent hypoglycemia. Additionally, the same data as in week 1 were collected every last day of the week for weeks 2 and 3.

Post-test week (study week 4)

A post-test was conducted on the first day of week 4. For the post-test and week 4, all measurements, except for the height and additional exercise intention survey, were executed in the same manner as in week 1.

Exercise program

The exercise program was inserted into the KMR device. The device and its use were described in our previous study.12 All participants were taught how to use the device and perform the exercises independently (Supplementary Table 1). The program included approximately 30 minutes of circuit-based exercises: the circuit comprised aerobic, resistance, and high-intensity interval training as recommended by the American College of Sports Medicine21,22 and the design was adapted from Shabani et al.23 The exercise circuit consisted of 5 minutes of warm-up (dynamic stretching, hip rotation, etc.), followed by three 8-minute sets of the eight main exercises, which included exercises using all the major muscle groups of the body (e.g., quadriceps, hamstrings, deltoids, etc.), and ended with a cool-down (static stretching, groin stretching, etc.), as suggested by the American Diabetes Association.24 Each main exercise was performed for 45 seconds while the KMR device monitored the participants’ movements and provided feedback, followed by a 15 seconds rest period. A pair of 5 kg dumbbells for male and 3 kg dumbbells for female participants were used for upper-body exercises. As shown in Supplementary Fig. 2,25 the exercise program provided a recommended number of repetitions; however, the number of repetitions was not fixed, and the participants were allowed to perform more repetitions in the 45 seconds exercise period if they could do so with correct form. The participants were asked to achieve at least 80% of the suggested number of repetitions to avoid rough participation.

Pre- and post-tests

Blood pressure (systolic blood pressure [SBP], diastolic blood pressure [DBP]), and resting heart rate (beats/min) were measured using an automatic blood pressure meter (BPBIO 320S; InBody Co.). All participants were instructed to rest for at least 10 minutes before measurement to ensure physiological stability. To assess body composition, the participants’ height was measured using a passive extensometer, and body weight (kg), skeletal muscle mass (kg), fat-free mass (kg), and BMI (kg/m2) were measured using bioelectrical impedance analysis (InBody720). Waist circumference (cm) was measured in the laboratory when the participants were exhaling using an elastic measuring tape (SECA201; SECA) at the waistline, which is the midpoint between the ribs and the iliac ridge, for accurate measurement.26 All data were recorded to the first decimal point.

For BG measurements, fasting BG measurement was performed first, followed by the 2-OGTT (200 mL of water mixed with 75 g of glucose).27 During the 2-OGTT, participants were advised to rest under supervision until the end of the measurement. Before conducting the test, finger disinfection and thorough alcohol evaporation procedures were meticulously carried out. The measurements were performed using an Accu-Chek GuideMe device (Roche Diabetes Care GmbH): this device is an U.S. Food and Drug Administration-approved and highly validated tool, as evidenced by the data available on the official website (data on file: https://www.accu-chek.com/blog/glucose-meter-accuracy).28 To ensure accuracy, both fasting BG and 2-OGTT measurements were performed twice, and the average value was utilized for the analyses.

For the exercise intention survey, three questions (e.g., “Do you have an intention for planning on doing this workout for another month?”) were asked, and the responses reported on a 7-point Likert scale ranging from extremely unlikely (0) to extremely likely (7). All measurements were by performed by the same sports science professional, and the time allocation for the pre-and post-treatment tests was fixed simultaneously using the same devices.

Weekly measurement

Weekly step and food intake data were collected to monitor the participants’ lifestyle maintenance over the 4-week study period. Step data was collected through the Galaxy Fit2 smartwatch and the Samsung application, while food intake was measured using the “3-day recall” method. Participants were asked to submit their food records for 3 days (2 weekdays and 1 weekend day) on a diet-recording worksheet. The total calories of all foods consumed were quantified using Can-Pro 5.0 (Computer Aided Nutritional Analysis Program; The Korean Nutrition Society).29

Each participant’s glucose concentrations were monitored by the CGMS every minute, and the mean ISF glucose30 and frequency of hypoglycemia and hyperglycemia were recorded and presented in the application. Postprandial glucose concentration data were collected before, immediately after, and 30, 60, 90, and 120 minutes after lunch on the “3-day recall” date, and additional CGMS glucose concentration measurements were recorded before and after each set of exercises, and also 30 and 60 minutes after exercise. The participants were not allowed to consume food during the assessment period.

The CGMS device was placed on the participants’ arms on the first days of weeks 1 and 3 for measurement. At least two CGMS were used for expiration and detachment. Additionally, an anonymous chat (KakaoTalk; Kakao) was created to deliver notices and collect the necessary data (every Sunday), such as by tagging the smartphone to the device three times a day. Personal questions or questions unrelated to the study were not included.

Statistical analyses

All data were analyzed using SPSS version 25.0 (IBM Co.) and were presented as mean±standard deviation. Nutrient intake was analyzed using a Can-Pro 5.29 Descriptive statistical analyses were used for the demographic variables. The Shapiro-Wilk test was used to evaluate normality at baseline (P>0.05). Paired-sample Wilcoxon tests were used to analyze the pre- and post-treatment test results, and the Friedman’s test with multiple comparison was applied to the repeated measurements. Statistical significance was set at P<0.05 and presented as effect size and Cohen’s d.31

Participants

After considering the likely dropout rate for out study, we initially recruited a group of 12 obese participants with pre-diabetes with a 1:1 ratio of males and females. The study protocol was approved by the Institutional Review Board of Kyungnam University (IRB no. 1040460-A-2022-029). The eligibility criteria were (1) a body mass index (BMI) ≥23 kg/m2;19 (2) age 29 to 59 years; (3) fasting BG concentration 100 to 125 mg/dL; and (4) 2-OGTT result 140 to 199 mg/dL. The exclusion criteria were (1) acute hepatitis or a history of malignant tumors in the previous month; (2) cerebral infarction or myocardial infarction in the previous 6 months; (3) peripheral vascular disease or dementia; (4) participation in regular exercise in the previous 3 months; and (5) engagement in moderate- to high-intensity exercise.

Study procedure

This study was conducted for 4 weeks using a pre-and post-treatment design (Fig. 1). Before participating in the study, all participants were informed of its purpose and design and voluntarily signed a consent form.

Screening/pre-test day

The first week was designated as the “pre-test” week (week 1), with screening and pre-test procedures conducted on the first day of this week. Before the first visit, the participants were instructed to maintain a fasting state for >9 hours, not consume alcohol or caffeine, and avoid extreme exercise within 24 hours. The researchers measured the participants’ blood pressure, body composition, and waist circumference and completed a physical activity questionnaire (The Global Physical Activity Questionnaire was used for screening only [e.g., occupation, leisure activity, etc.]).20 The participants’ fasting BG concentration and 2-OGTT results were also measured. If the fasting BG concentration was between 100 and 125 mg/dL, a 2-OGTT was performed. Participants with 2-OGTT results between 140 and 199 mg/dL were selected as having pre-diabetes. Additionally, the participants were fitted with a CGMS (Adela Healthcare) and provided a smartwatch (Galaxy Fit2; Samsung) to collect weekly glucose and physical activity (steps) data.

Pre-test week (study week 1)

After the pre-test (week 1) was conducted for a week (Monday to Sunday), the screening and pre-tests were performed on Monday. At the end of week 1, participants were asked to report their weekly average glucose concentration, hyperglycemia rate, hypoglycemia rate, and average glucose level by time and amount of physical activity, food intake, and postprandial glucose on Sunday.

Exercise weeks 1 and 2 (study weeks 2 and 3)

During study weeks 2 and 3, the participants were asked to perform the exercise program for 6 days (3 days per week). Each exercise session was conducted using a KMR device (Virtual Mate; My Benefit) at the same place and time, between 7:00 AM and 9:00 AM. The first day of week 2 was designed as an adaptation phase, during which participants were taught how to use the KMR device and perform proper exercise movements. For the five remaining exercise sessions, the participants were instructed to exercise independently without supervision (Supplementary Fig. 1). One hour before each exercise session, the participants were asked to consume a nutritional drink (New Care; Daesang) containing 30 g of carbohydrates, 7 g of protein, and 6 g of fat in order to maintain the same state and prevent hypoglycemia. Additionally, the same data as in week 1 were collected every last day of the week for weeks 2 and 3.

Post-test week (study week 4)

A post-test was conducted on the first day of week 4. For the post-test and week 4, all measurements, except for the height and additional exercise intention survey, were executed in the same manner as in week 1.

Exercise program

The exercise program was inserted into the KMR device. The device and its use were described in our previous study.12 All participants were taught how to use the device and perform the exercises independently (Supplementary Table 1). The program included approximately 30 minutes of circuit-based exercises: the circuit comprised aerobic, resistance, and high-intensity interval training as recommended by the American College of Sports Medicine21,22 and the design was adapted from Shabani et al.23 The exercise circuit consisted of 5 minutes of warm-up (dynamic stretching, hip rotation, etc.), followed by three 8-minute sets of the eight main exercises, which included exercises using all the major muscle groups of the body (e.g., quadriceps, hamstrings, deltoids, etc.), and ended with a cool-down (static stretching, groin stretching, etc.), as suggested by the American Diabetes Association.24 Each main exercise was performed for 45 seconds while the KMR device monitored the participants’ movements and provided feedback, followed by a 15 seconds rest period. A pair of 5 kg dumbbells for male and 3 kg dumbbells for female participants were used for upper-body exercises. As shown in Supplementary Fig. 2,25 the exercise program provided a recommended number of repetitions; however, the number of repetitions was not fixed, and the participants were allowed to perform more repetitions in the 45 seconds exercise period if they could do so with correct form. The participants were asked to achieve at least 80% of the suggested number of repetitions to avoid rough participation.

Pre- and post-tests

Blood pressure (systolic blood pressure [SBP], diastolic blood pressure [DBP]), and resting heart rate (beats/min) were measured using an automatic blood pressure meter (BPBIO 320S; InBody Co.). All participants were instructed to rest for at least 10 minutes before measurement to ensure physiological stability. To assess body composition, the participants’ height was measured using a passive extensometer, and body weight (kg), skeletal muscle mass (kg), fat-free mass (kg), and BMI (kg/m2) were measured using bioelectrical impedance analysis (InBody720). Waist circumference (cm) was measured in the laboratory when the participants were exhaling using an elastic measuring tape (SECA201; SECA) at the waistline, which is the midpoint between the ribs and the iliac ridge, for accurate measurement.26 All data were recorded to the first decimal point.

For BG measurements, fasting BG measurement was performed first, followed by the 2-OGTT (200 mL of water mixed with 75 g of glucose).27 During the 2-OGTT, participants were advised to rest under supervision until the end of the measurement. Before conducting the test, finger disinfection and thorough alcohol evaporation procedures were meticulously carried out. The measurements were performed using an Accu-Chek GuideMe device (Roche Diabetes Care GmbH): this device is an U.S. Food and Drug Administration-approved and highly validated tool, as evidenced by the data available on the official website (data on file: https://www.accu-chek.com/blog/glucose-meter-accuracy).28 To ensure accuracy, both fasting BG and 2-OGTT measurements were performed twice, and the average value was utilized for the analyses.

For the exercise intention survey, three questions (e.g., “Do you have an intention for planning on doing this workout for another month?”) were asked, and the responses reported on a 7-point Likert scale ranging from extremely unlikely (0) to extremely likely (7). All measurements were by performed by the same sports science professional, and the time allocation for the pre-and post-treatment tests was fixed simultaneously using the same devices.

Weekly measurement

Weekly step and food intake data were collected to monitor the participants’ lifestyle maintenance over the 4-week study period. Step data was collected through the Galaxy Fit2 smartwatch and the Samsung application, while food intake was measured using the “3-day recall” method. Participants were asked to submit their food records for 3 days (2 weekdays and 1 weekend day) on a diet-recording worksheet. The total calories of all foods consumed were quantified using Can-Pro 5.0 (Computer Aided Nutritional Analysis Program; The Korean Nutrition Society).29

Each participant’s glucose concentrations were monitored by the CGMS every minute, and the mean ISF glucose30 and frequency of hypoglycemia and hyperglycemia were recorded and presented in the application. Postprandial glucose concentration data were collected before, immediately after, and 30, 60, 90, and 120 minutes after lunch on the “3-day recall” date, and additional CGMS glucose concentration measurements were recorded before and after each set of exercises, and also 30 and 60 minutes after exercise. The participants were not allowed to consume food during the assessment period.

The CGMS device was placed on the participants’ arms on the first days of weeks 1 and 3 for measurement. At least two CGMS were used for expiration and detachment. Additionally, an anonymous chat (KakaoTalk; Kakao) was created to deliver notices and collect the necessary data (every Sunday), such as by tagging the smartphone to the device three times a day. Personal questions or questions unrelated to the study were not included.

Statistical analyses

All data were analyzed using SPSS version 25.0 (IBM Co.) and were presented as mean±standard deviation. Nutrient intake was analyzed using a Can-Pro 5.29 Descriptive statistical analyses were used for the demographic variables. The Shapiro-Wilk test was used to evaluate normality at baseline (P>0.05). Paired-sample Wilcoxon tests were used to analyze the pre- and post-treatment test results, and the Friedman’s test with multiple comparison was applied to the repeated measurements. Statistical significance was set at P<0.05 and presented as effect size and Cohen’s d.31

To the best of our knowledge, this is the first study to continuously measure changes in ISF glucose concentrations during unsupervised exercise in obese individuals with pre-diabetes (as identified based on BG concentrations). The study tracked ISF glucose concentrations for approximately 672 hours per participant and found that a 2-week unsupervised exercise program can reduce 2-OGTT results and DBP and may change postprandial glucose concentrations. All nine participants wore smartwatches to detect physical activity and a CGMS to detect ISF glucose concentrations over 4 weeks. All participants showed high compliance and intention to participate in the unsupervised KMR exercise, and our findings suggest that this program may enhance BG control.

The major positive changes observed in this study were in 2-OGTT and DBP. The 2-OGTT is the gold standard for diagnosing and monitoring diabetes32 and DBP is also one of the major factors influenced by BG. The effect of exercise on BG and blood pressure are well known: the mechanism is considered to be improvement of endothelial function leading to increased nitric oxide production, induction of pro-angiogenic pathway and increased insulin sensitivity.33

A previous study by Babraj et al.34 showed similar results to this study: 2 weeks of high-intensity interval training improved the 2-OGTT results, but there was no change in fasting BG concentrations in healthy young male. In the meta-analysis study by MacLeod et al.,35 eight short-term studies showed no change in fasting glucose concentrations in people with diabetes but mean glucose concentrations and hyperglycemia rates were reduced. In contrast, another study showed no change in 2-OGTT results after completing a daily exercise program for 7 consecutive days.36 There have been few studies of CGM and exercise. Further research is needed, but it is noteworthy that the 2-OGTT results in this study were reduced after only 2 weeks of unsupervised exercise.

In this study, weekly glucose concentrations measured by the CGMS did not show any significant reduction in daily time but increased. It is difficult to explain the mechanism of this result, but it may have been due to the Hawthorne effect,37 in which most participants did not realize their glucose level (which was higher than expected) before this study. In addition, wearing the CGMS sensor caused participants to manage their glucose levels during study week 1 more than they would normally. Furthermore, previous studies have shown that high cognitive demand can decrease glucose levels,38 and our results may therefore have been influenced by the participants having had different weekly or daily work schedules and intensities, making comparisons of glucose concentrations by week and time less meaningful.

Despite these possible limitations, the weekly glucose hyperglycemia rate and postprandial glucose levels after 90 minutes were significantly reduced after the 2-week exercise intervention. The reduction in hyperglycemia rates following the exercise program suggests that regular physical activity may be an effective strategy to manage BG concentrations in individuals with poor glycemic control.39 In particular, lunch-time postprandial glucose concentrations are an indicator of glucose spikes (metabolic functions) that act in the body due to food intake, which could represent metabolic ability,40 such as the 2-OGTT. It has been reported as one of the gold standards for controlling BG.41 This critical indicator changed within 3 days of implementing the exercise program. The reduction in postprandial glucose concentrations after completing the 2-week exercise program may indicate improved insulin sensitivity and glucose uptake in skeletal muscle.42 Additionally, there was a slight reduction after the second week of exercise (week 3) compared to the first exercise week (week 2), which was not significant. We anticipate a significant change after a longer-term exercise intervention.

Another positive change was in DBP. Blood pressure is higher in individuals with poor glycemic control and is indicative of vascular health, which is also related to poor glycemic control.43 A study by Böhm et al.44 showed that blood pressure is an important indicator of cardiovascular disease in people with poor glycemic control; therefore, lowering BG concentrations through exercise can positively affect vascular health. Similarly, in our study DBP was significantly reduced after 2 weeks of exercise, and both mean SBP and DBP approached the normal range; however, the change in SBP was not significant, possibly due to the small sample size, which resulted in a large standard deviation. The observed reduction in DBP in our study after completing the exercise program may be attributed to the beneficial effects of regular physical activity on vascular health, which is often compromised in individuals with poor glycemic control.45

This study had some limitations. The first is the short duration of the intervention (2 weeks): our study was a pilot trial that aimed to validate the assessment protocols and exercise system before conducting a longer-duration study, as no previous study has evaluated ISF glucose in real time in conjunction with multi-component unsupervised exercise. Second, the sample size was small. Although the ratio of male to females was close to equal and showed normality, and the minimum number of participants was determined through G*power, applying the research results to a larger sample size would have been better. Third, there was no control group, and it was therefore difficult to determine whether the observed changes in the outcome measures were solely due to the exercise program. Finally, BG concentrations are influenced by various activities, and efforts were therefore made to regulate the participants’ daily life activities, they could not be completely restricted or controlled. In addition, the amount of physical activity (smartwatch to detect steps) and food intake may differ depending on whether the participants are detached and sincere and may not be as reliable as CGMS measurements. Therefore, generalizing these results may pose a challenge. Additional research is warranted to assess the long-term effectiveness of this exercise program and to validate the reliability of these findings in a study that also includes a control group. Large-scale randomized trials are imperative for assess whether this exercise program can effectively prevent diabetes and replicate the results observed in this study.

Since non-face-to-face interventions, including exercise programs, have become increasingly important due to the COVID-19 pandemic, our study provides evidence that unsupervised KMR exercise programs may effectively manage BG concentrations and improve vascular health in obese individuals with pre-diabetes. Further studies are needed to confirm our findings and determine the optimal duration and intensity of unsupervised exercise programs for pre-diabetes management.

The authors declare no conflict of interest. Dr. Exsol Inc. also declares no conflict of interest.

This research was supported by a grant from the Korean Institute of Marine Science & Technology Promotion (20220027). The study sponsor/funder was not involved in the design of the study, nor in the collection, analysis, and interpretation of data or writing of the report, and did not impose any restrictions regarding the publication of the report. The authors have no other potential conflicts of interest to disclose.

We would like to express our sincere appreciation to our colleagues at the Health and Exercise Science Laboratory at Seoul National University for their support and assistance in this research.

Study concept and design: SYA; acquisition of data: SYA, SWL, HJS, WJL, and JHK; analysis and interpretation of data: SYA; drafting of the manuscript: SYA, SWL, HJS, and WJL; critical revision of the manuscript: SYA, HJK, and WS; statistical analysis: SYA; obtained funding: HJK and WS; administrative, technical, or material support: SYA, SWL, HJS, WJL, JHK, HJK, and WS; and study supervision: SYA, HJK, and WS.

Fig. 1. Design of the 4-week study of the effects of a 2-week period of Kinect-based mixed-reality exercise, as assessed using continuous glucose monitoring. The study was conducted over a period of 4 weeks, with pre- and post-treatment assessments on the first day of the pre- (week 1) and post-treatment (week 4) weeks, respectively. The exercise program was completed during weeks 2 and 3, and consisted 3 non-consecutive exercise days per week. Throughout the 4-week period, interstitial glucose concentrations were tracked by a continuous glucose monitoring system.
Fig. 2. Percent participation in Kinect-based mixed-reality exercise program over 5 exercise days with among nine study participants with pre-diabetes. Data for the first exercise day were not included to allow for participant adaptation and supervision. The X-axis presents exercise days and the Y-axis presents the percentage of participants who completed the study exercise program [(participant’s reps/suggested reps)× 100] of a total of 24 exercises for each participant. As a result, all the participants intended to do more exercise by themselves without any motivation to do so. *P<0.01; P<0.001.
Fig. 3. Interstitial fluid glucose concentrations during Kinect-based mixed-reality exercise: day 2 vs. day 6. This graph shows the mean interstitial fluid glucose concentration data with standard error bars for each time period during exercise days 2 and 6 of 6 days of exercise participation. Data for the first exercise day were not included to allow for participant adaptation and supervision, and the other exercise days were unsupervised. The X-axis presents the various measurement points: before exercise, first rest period between exercises, second rest period between exercises, immediately after exercise, 30 minutes after exercise (minAE), and 60 minAE. The Y-axis presents the interstitial fluid glucose concentration: the left side Y-axis scale is mg/dL and the right side is mmol/L. The mean represents the mean of interstitial fluid glucose concentrations from before exercise to 60 minAE. *P<0.1 (tendency); P<0.05.

Pre- and post-treatment measurements of body composition, blood pressure, and blood glucose level

Variable Pre-treatment Post-treatment P (Cohen’s d) Mean rank (range) Z
Weight (kg) 75.53 ± 4.68 75.71 ± 4.61 0.656 (0.038) –6.00 to 4.50 –0.533
Fasting glucose (mg/dL) 108.00 ± 7.20 109.67 ± 11.50 0.695 (0.174) –5.75 to 4.40 –0.059
Oral glucose tolerance test result (mg/dL) 162.67 ± 18.30 142.56 ± 37.93 0.044* (0.675) –5.06 to 4.50 –2.134
Waist circumference (cm) 90.17 ± 3.86 89.50 ± 4.60 0.477 (0.158) –4.33 to 2.67 –0.526
Skeletal muscle mass (kg) 28.09 ± 3.43 28.49 ± 3.77 0.254 (0.111) –3.50 to 6.20 –1.009
Body fat (%) 29.54 ± 9.34 28.81 ± 9.30 0.166 (0.078) –6.25 to 2.75 –0.980
Fat-free mass (kg) 50.40 ± 5.47 51.13 ± 6.13 0.228 (0.126) –3.50 to 6.20 –1.007
Body mass index (kg/m2) 27.16 ± 2.92 24.03 ± 8.23 0.341 (0.507) –4.80 to 4.00 –0.842
Systolic blood pressure (mmHg) 131.00 ± 9.35 126.11 ± 12.10 0.176 (0.452) –5.50 to 4.00 –1.245
Diastolic blood pressure (mmHg) 82.89 ± 8.45 78.11 ± 12.37 0.046* (0.451) –5.64 to 2.75 –2.018

Values are presented as mean± standard deviation.

*P<0.05; Present for minimum and maximum of mean rank.

Weekly continuous blood glucose monitoring data for 4 weeks

Variable Pre-treatment (week 1) Exercise week 1 (week 2) P * (Cohen’s d) Exercise week 2 (week 3) P (Cohen’s d) Post-treatment (week 4) P (Cohen’s d)
Hypoglycemia (%) 2.11 ± 5.26 0.11 ± 0.33 0.290 (0.537) 2.00 ± 2.55 0.961 (0.027) 1.22 ± 1.72 0.652 (0.023)
Hyperglycemia (%) 9.22 ± 10.95 8.78 ± 10.96 0.695 (0.040) 5.89 ± 8.84 0.040§ (0.335) 11.44 ± 12.49 0.267 (0.189)
Mean BG (mg/dL) 105.44 ± 16.31 109.11 ± 12.19 0.122 (0.255) 102.89 ± 12.74 0.302 (0.174) 108.22 ± 13.82 0.463 (0.184)
Weekly average BG (mg/dL)
12:00 AM–3:00 AM 110.22 ± 41.89 103.56 ± 16.00 0.505 (0.210) 97.67 ± 17.59 0.182 (0.390) 102.56 ± 22.22 0.311 (0.228)
3:00 AM–6:00 AM 92.33 ± 14.07 94.89 ± 7.25 0.482 (0.228) 90.33 ± 11.60 0.508 (0.115) 91.22 ± 6.61 0.839 (0.101)
6:00 AM–9:00 AM 93.78 ± 12.79 104.78 ± 15.39 0.008 (0.777) 98.78 ± 14.95 0.172 (0.359) 99.11 ± 8.89 0.264 (0.484)
9:00 AM–12:00 PM 105.56 ± 16.86 110.22 ± 12.06 0.145 (0.318) 106.11 ± 13.57 0.891 (0.036) 109.22 ± 17.30 0.425 (0.214)
12:00 PM–3:00 PM 121.67 ± 24.41 121.56 ± 21.92 0.968 (0.005) 114.78 ± 27.64 0.075 (0.264) 122.33 ± 18.77 0.908 (0.030)
3:00 PM–6:00 PM 103.44 ± 13.78 110.67 ± 14.84 0.006 (0.505) 106.22 ± 13.56 0.501 (0.203) 109.33 ± 13.46 0.184 (0.432)
6:00 PM–9:00 PM 107.56 ± 11.27 113.33 ± 10.12 0.104 (0.539) 102.89 ± 7.94 0.152 (0.479) 120.22 ± 21.14 0.036§ (0.747)
9:00 PM–12:00 AM 116.22 ± 21.56 116.78 ± 24.59 0.912 (0.024) 113.22 ± 21.52 0.522 (0.139) 114.56 ± 22.96 0.660 (0.075)

Values are presented as mean± standard deviation. Hypoglycemia, percentage participants with BG concentration < 70 mg/dL per week; Hyperglycemia, percentage of participants with BG concentration > 140 mg/dL per week; Mean BG, mean BG concentration over a week-long period; Physical activity, average weekly step counts as recorded using a Samsung Galaxy Fit wearable activity tracker.

*P-value for comparison between pre-treatment week (week 1 of 4) and exercise week 1 (week 2 of 4); P-value for comparison between pre-treatment week (week 1 of 4) and exercise week 2 (week 3 of 4); P-value for comparison between pre-treatment week (week 1 of 4) and post-test week (week 4 of 4); P-values are presented as Cohen’s d: §P<0.05, P<0.01.

BG, blood glucose.

Postprandial interstitial fluid glucose concentration as measured using continuous glucose monitoring for 4 weeks

Blood glucose concentration measurement time Pre-treatment week (week 1) (mg/dL) Exercise weeks 1 and 2 (weeks 2−3) (mg/dL) P * (Cohen’s d) Post-treatment week (week 4) (mg/dL) P (Cohen’s d)
Before lunch 100.00 ± 18.85 106.83 ± 14.19 0.149 (0.409) 102.19 ± 20.64 0.632 (0.111)
Immediately after lunch 124.48 ± 31.44 123.20 ± 22.48 0.884 (0.049) 121.85 ± 24.78 0.755 (0.093)
30 min after lunch 148.82 ± 38.44 136.44 ± 27.07 0.176 (0.372) 138.41 ± 27.01 0.354 (0.313)
60 min after lunch 143.07 ± 32.67 142.78 ± 29.25 0.977 (0.009) 142.96 ± 27.98 0.991 (0.004)
90 min after lunch 129.19 ± 23.91 119.68 ± 17.08 0.035 (0.458) 124.33 ± 19.80 0.542 (0.221)
120 min after lunch 115.00 ± 22.41 115.07 ± 16.40 0.990 (0.004) 110.85 ± 19.32 0.527 (0.198)

Values are presented as mean± standard deviation.

*For comparison of pre-treatment week (week 1) and exercise weeks 1 and 2 (weeks 2 and 3) values; For comparison of pre-treatment week (week 1) and post-treatment week (week 4) values; P<0.05.

  1. Centers for Disease Control and Prevention. Prevent diabetes complications [Internet]. CDC; 2022 [cited 2024 Jan 4]. Available from: https://www.cdc.gov/diabetes/managing/problems.html.
  2. International Diabetes Federation. Global guideline for type 2 diabetes [Internet]. IDF; 2012 [cited 2024 Jan 4]. Available from: https://www.iapb.org/wp-content/uploads/Global-Guideline-for-Type-2-Diabetes-IDF-2012.pdf.
  3. Choi YJ, Kim HC, Kim HM, Park SW, Kim J, Kim DJ. Prevalence and management of diabetes in Korean adults: Korea National Health and Nutrition Examination Surveys 1998-2005. Diabetes Care 2009;32:2016-20.
    Pubmed KoreaMed CrossRef
  4. Choi ES, Rhee EJ, Kim JH, Won JC, Park CY, Lee WY, et al. Insulin sensitivity and insulin secretion determined by homeostasis model assessment and future risk of diabetes mellitus in Korean men. Korean Diabetes J 2008;32:498-505.
    CrossRef
  5. Korean Diabetes Association. Diabetes fact sheet 2020. KDA; 2020.
  6. van Herpt TT, Ligthart S, Leening MJ, van Hoek M, Lieverse AG, Ikram MA, et al. Lifetime risk to progress from pre-diabetes to type 2 diabetes among women and men: comparison between American Diabetes Association and World Health Organization diagnostic criteria. BMJ Open Diabetes Res Care 2020;8:e001529.
    Pubmed KoreaMed CrossRef
  7. Tuomilehto J, Lindström J, Eriksson JG, Valle TT, Hämäläinen H, Ilanne-Parikka P, et al. Prevention of type 2 diabetes mellitus by changes in lifestyle among subjects with impaired glucose tolerance. N Engl J Med 2001;344:1343-50.
    Pubmed CrossRef
  8. Knowler WC, Barrett-Connor E, Fowler SE, Hamman RF, Lachin JM, Walker EA, et al. Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin. N Engl J Med 2002;346:393-403.
    Pubmed KoreaMed CrossRef
  9. Li G, Zhang P, Wang J, Gregg EW, Yang W, Gong Q, et al. The long-term effect of lifestyle interventions to prevent diabetes in the China Da Qing Diabetes Prevention Study: a 20-year follow-up study. Lancet 2008;371:1783-9.
    Pubmed CrossRef
  10. Krousel-Wood MA, Berger L, Jiang X, Blonde L, Myers L, Webber L. Does home-based exercise improve body mass index in patients with type 2 diabetes?: results of a feasibility trial. Diabetes Res Clin Pract 2008;79:230-6.
    Pubmed CrossRef
  11. Hordern MD, Dunstan DW, Prins JB, Baker MK, Singh MA, Coombes JS. Exercise prescription for patients with type 2 diabetes and pre-diabetes: a position statement from Exercise and Sport Science Australia. J Sci Med Sport 2012;15:25-31.
    Pubmed CrossRef
  12. Ahn SY, Sung YH, Bae JH, Lim BG, Li X, Kim DH, et al. Effect of unsupervised Kinect-based mixed reality fitness programs on health-related fitness in men during COVID-19 pandemic: randomized controlled study. J Mens Health 2023;19:104-12.
  13. Cai H, Li G, Jiang S, Yin H, Liu P, Chen L. Effect of low-intensity, KinectTM-based Kaimai-style Qigong exercise in older adults with type 2 diabetes. J Gerontol Nurs 2019;45:42-52.
    Pubmed CrossRef
  14. Ahn SY, Sung YH, Bae JH, Lim BG, Song W. Reliability and validity of the Kinect-based mixed reality device: pilot study. Asian J Kinesiol 2022;24:2-11.
    CrossRef
  15. Yoo JH, Kim JH. Time in range from continuous glucose monitoring: a novel metric for glycemic control. Diabetes Metab J 2020;44:828-39.
    Pubmed KoreaMed CrossRef
  16. Schubert-Olesen O, Kröger J, Siegmund T, Thurm U, Halle M. Continuous glucose monitoring and physical activity. Int J Environ Res Public Health 2022;19:12296.
    Pubmed KoreaMed CrossRef
  17. Bally L, Zueger T, Pasi N, Carlos C, Paganini D, Stettler C. Accuracy of continuous glucose monitoring during differing exercise conditions. Diabetes Res Clin Pract 2016;112:1-5.
    Pubmed CrossRef
  18. Little JP, Jung ME, Wright AE, Wright W, Manders RJ. Effects of high-intensity interval exercise versus continuous moderate-intensity exercise on postprandial glycemic control assessed by continuous glucose monitoring in obese adults. Appl Physiol Nutr Metab 2014;39:835-41.
    Pubmed CrossRef
  19. WHO Expert Consultation. Appropriate body-mass index for Asian populations and its implications for policy and intervention strategies. Lancet 2004;363:157-63.
    Pubmed CrossRef
  20. Lee J, Lee C, Min J, Kang DW, Kim JY, Yang HI, et al. Development of the Korean Global Physical Activity Questionnaire: reliability and validity study. Glob Health Promot 2020;27:44-55.
    Pubmed CrossRef
  21. Kanaley JA, Colberg SR, Corcoran MH, Malin SK, Rodriguez NR, Crespo CJ, et al. Exercise/physical activity in individuals with type 2 diabetes: a consensus statement from the American College of Sports Medicine. Med Sci Sports Exerc 2022;54:353-68.
    Pubmed KoreaMed CrossRef
  22. Muñoz-Martínez FA, Rubio-Arias JÁ, Ramos-Campo DJ, Alcaraz PE. Effectiveness of resistance circuit-based training for maximum oxygen uptake and upper-body one-repetition maximum improvements: a systematic review and meta-analysis. Sports Med 2017;47:2553-68.
    Pubmed CrossRef
  23. Shabani R, Nazari M, Dalili S, Rad AH. Effect of circuit resistance training on glycemic control of females with diabetes type II. Int J Prev Med 2015;6:34.
    Pubmed KoreaMed CrossRef
  24. American Diabetes Association. 2. Classification and diagnosis of diabetes: standards of medical care in diabetes-2021. Diabetes Care 2021;44(Suppl 1):S15-33.
    Pubmed CrossRef
  25. My Benefit. Virtual Mate [Internet]. My Benefit; 2023 [cited 2024 Jan 10]. Available from: https://www.mybenefit.co.kr.
  26. World Health Organization. Waist circumference and waist-hip ratio: report of a WHO expert consultation. WHO; 2011.
  27. World Health Organization. Definition and diagnosis of diabetes mellitus and intermediate hyperglycaemia: report of a WHO/IDF consultation. WHO; 2006.
  28. Roche. Accu-Chek [Internet]. Roche; 2023 [cited 2024 Jan 4]. Available from: https://www.accu-chek.com.
  29. Korean Nutrition Society. Can-Pro 5.0 (Computer Aided Nutritional Analysis Program). Korean Nutrition Society; 2011.
  30. Garg SK, Schwartz S, Edelman SV. Improved glucose excursions using an implantable real-time continuous glucose sensor in adults with type 1 diabetes. Diabetes Care 2004;27:734-8.
    Pubmed CrossRef
  31. Cohen J. Statistical power analysis for the behavioral sciences. 2nd ed. Lawrence Erlbaum Associates; 1988.
  32. DECODE Study Group, the European Diabetes Epidemiology Group. Glucose tolerance and cardiovascular mortality: comparison of fasting and 2-hour diagnostic criteria. Arch Intern Med 2001;161:397-405.
    Pubmed CrossRef
  33. Gambardella J, Morelli MB, Wang XJ, Santulli G. Pathophysiological mechanisms underlying the beneficial effects of physical activity in hypertension. J Clin Hypertens (Greenwich) 2020;22:291-5.
    Pubmed KoreaMed CrossRef
  34. Babraj JA, Vollaard NB, Keast C, Guppy FM, Cottrell G, Timmons JA. Extremely short duration high intensity interval training substantially improves insulin action in young healthy males. BMC Endocr Disord 2009;9:3.
    Pubmed KoreaMed CrossRef
  35. MacLeod SF, Terada T, Chahal BS, Boulé NG. Exercise lowers postprandial glucose but not fasting glucose in type 2 diabetes: a meta-analysis of studies using continuous glucose monitoring. Diabetes Metab Res Rev 2013;29:593-603.
    Pubmed CrossRef
  36. Mikus CR, Oberlin DJ, Libla J, Boyle LJ, Thyfault JP. Glycaemic control is improved by 7 days of aerobic exercise training in patients with type 2 diabetes. Diabetologia 2012;55:1417-23.
    Pubmed KoreaMed CrossRef
  37. Landsberger HA. Hawthorne revisited: management and the worker, its critics, and developments in human relations in industry. Cornell University Press; 1958.
  38. Scholey AB, Harper S, Kennedy DO. Cognitive demand and blood glucose. Physiol Behav 2001;73:585-92.
    Pubmed CrossRef
  39. Colberg SR, Sigal RJ, Yardley JE, Riddell MC, Dunstan DW, Dempsey PC, et al. Physical activity/exercise and diabetes: a position statement of the American Diabetes Association. Diabetes Care 2016;39:2065-79.
    Pubmed KoreaMed CrossRef
  40. Muggeo M, Verlato G, Bonora E, Zoppini G, Corbellini M, de Marco R. Long-term instability of fasting plasma glucose, a novel predictor of cardiovascular mortality in elderly patients with non-insulin-dependent diabetes mellitus: the Verona Diabetes Study. Circulation 1997;96:1750-4.
    Pubmed CrossRef
  41. Cavalot F, Petrelli A, Traversa M, Bonomo K, Fiora E, Conti M, et al. Postprandial blood glucose is a stronger predictor of cardiovascular events than fasting blood glucose in type 2 diabetes mellitus, particularly in women: lessons from the San Luigi Gonzaga Diabetes Study. J Clin Endocrinol Metab 2006;91:813-9.
    Pubmed CrossRef
  42. Devlin JT. Effects of exercise on insulin sensitivity in humans. Diabetes Care 1992;15:1690-3.
    Pubmed CrossRef
  43. Yamada MH, Fujihara K, Kodama S, Sato T, Osawa T, Yaguchi Y, et al. Associations of systolic blood pressure and diastolic blood pressure with the incidence of coronary artery disease or cerebrovascular disease according to glucose status. Diabetes Care 2021;44:2124-31.
    Pubmed CrossRef
  44. Böhm M, Schumacher H, Teo KK, Lonn EM, Mahfoud F, Mann JF, et al. Cardiovascular outcomes and achieved blood pressure in patients with and without diabetes at high cardiovascular risk. Eur Heart J 2019;40:2032-43.
    Pubmed CrossRef
  45. Sparks JR, Kishman EE, Sarzynski MA, Davis JM, Grandjean PW, Durstine JL, et al. Glycemic variability: importance, relationship with physical activity, and the influence of exercise. Sports Med Health Sci 2021;3:183-93.
    Pubmed KoreaMed CrossRef