Share this article on:

Monitoring Exercise-Induced Changes in Glycemic Control in Type 2 Diabetes


Medicine & Science in Sports & Exercise: February 2006 - Volume 38 - Issue 2 - pp 201-207
doi: 10.1249/01.mss.0000183852.31164.5a
CLINICAL SCIENCES: Clinical Case Studies

Purpose: The present study determined the efficacy of the Continuous Glucose Monitoring System (CGMS) during moderate exercise and monitored the changes in whole-day glucose profiles using the CGMS in individuals with and without type 2 diabetes.

Methods: Six, obese, diet-treated individuals with and four age-matched individuals without type 2 diabetes were monitored using the CGMS for 3 d. Subjects cycled at 90% of a predetermined lactate threshold for 1 h at approximately 09:00 h on day 2, during which venous blood was sampled at 10-min intervals and immediately analyzed for glucose concentrations.

Results: Venous blood glucose and CGMS values declined during exercise in the diabetes (P < 0.001) but not the control group (P = 0.085). The CGMS overestimated blood glucose in the control (P = 0.003) and the diabetes (P = 0.045) groups during exercise. The number of data points outside of the 95% confidence intervals was <5% in both groups, showing that there is a statistically acceptable level of agreement between venous blood glucose and CGMS values during exercise. Moderate exercise improved whole-day average glucose concentrations (P = 0.007) and whole-day area under the glucose curve (P = 0.016) values (AUCglu), and the time spent within ±10% of fasting venous glucose (FVG) increased in the diabetes group (P = 0.021). No such effects were seen in the control group.

Conclusion: Using continuous glucose monitoring we were able to demonstrate that a period of moderate exercise improved whole-day glycemic control in obese individuals with type 2 diabetes. The CGMS should only be used as an adjunct and not as an alternative to frequent blood glucose sampling when examining the changes in glucose values during exercise in individuals with and without type 2 diabetes.

1School of Pharmacy and Biomolecular Sciences, University of Brighton, Brighton, UNITED KINGDOM; and 2Chelsea School, Welkin Performance Laboratories, University of Brighton, Eastbourne, UNITED KINGDOM

Address for correspondence: Adam L. Macdonald, School of Pharmacy and Biomolecular Sciences and Chelsea School, Welkin Laboratories, 30 Carlisle Road, Eastbourne, BN20 7SP, United Kingdom; E-mail:

Submitted for publication February 2005.

Accepted for publication August 2005.

Exercise has become a cornerstone treatment for many chronic diseases and is often recommended in combination with dietary alterations as the initial treatment modality in newly diagnosed individuals with type 2 diabetes (1,2). Exercise is recommended in the treatment of glycemic profiles in individuals with type 2 diabetes, principally because of its beneficial effect on blood glucose profiles (2). Changes in glycemic profiles during, immediately post (∼3 h), and in the day(s) following exercise are often studied in individuals with type 2 diabetes and would be of particular interest to clinicians prescribing exercise to patients with the disorder.

In contrast to healthy subjects, blood glucose declines during exercise in most (16,23-25) but not all (6) individuals with type 2 diabetes. This fall in blood glucose during exercise in individuals with type 2 diabetes is likely due to a disproportionate increase in glucose utilization (17,24) or a blunted (or delayed) hepatic glucose output response to exercise (23).

It has been established that a single period of exercise can improve glycemic control in individuals with type 2 diabetes (14). A single bout of exercise in overweight individuals with type 2 diabetes has been shown to reduce the glycemic excursions and to elevate plasma insulin concentrations in the 4-h period following a standard breakfast meal (20). Molecular evidence suggests that the improvements in glycemic profiles are regulated largely by changes in skeletal muscle glucose uptake through the glucose transporter, GLUT4 (7,14). Acute exercise seems to change the glycemic profiles of individuals with type 2 diabetes by increasing skeletal muscle glucose transport activity to an extent similar to that observed in nondiabetic subjects (16). Enhanced glucose transport activity continues into the postexercise period and is associated with improvements in insulin action (23). This effect remains evident at 20 h (9) but not 24 h (8) after a period of moderate exercise.

However, monitoring exercise-induced changes in daily glucose profiles is time consuming and often invasive, typically requiring multiple venous and/or fingertip blood samples. The recent development of continuous glucose monitoring systems has provided an opportunity to overcome these practical difficulties and so serve as a powerful tool to monitor the glycemic state of individuals with type 2 diabetes during and up to 72 h after exercise. Continuous glucose monitoring has the benefit of enabling patients to return to their normal daily activities, which could provide "real life" information on the effects of exercise (26). This information may then be used to identify exercise- or diet-induced changes in glucose tolerance and provide a useful source of additional information for healthcare professionals for making better exercise recommendations to treat the glycemic profiles of individuals with type 2 diabetes (26).

One such continuous glucose monitor device is the Medtronic-MiniMed Continuous Glucose Monitoring System (CGMS; MiniMed, Sylmar, CA). It is a commercially available holster-style system that was the first indwelling glucose sensor to be approved by the United States Food and Drug Administration, in 1999. The device is intended to supplement, not replace blood glucose information by providing glucose pattern and trend information for approximately 72 h (21,22). The CGMS is attached to an electrochemical sensor, which is inserted into the anterior abdominal wall, and provides a continuous measure of interstitial glucose up to 288 times per day for three consecutive days (4,13,21). The CGMS measures interstitial glucose by converting glucose at a glucose oxidase interface to hydrogen peroxide, which is oxidized to produce an amperometric signal (21). This signal is proportional to the interstitial glucose concentration and is stored in a pager-style monitor. The stored amperometric data was transferred and converted to glucose concentrations after data collection via an infrared link to a personal computer (21) and analyzed using the CGMS system solutions software (version 3.0B).

When compared with conventional, intermittent blood sampling defined as three to four blood glucose measurements per day, continuous glucose monitoring provides a greater insight into the direction, magnitude, duration, and frequency of fluctuations in glucose levels (19). The mean absolute difference between sensor and blood glucose meter values has been shown to be between 1.3 and 2.6 mmol·L−1, likely reflecting the biological time delay between interstitial and blood glucose concentrations (27). The sensitivity and specificity of the CGMS to detect hypoglycemic episodes, defined as a glucose value of 55 mg·dL−1 (∼3 mmol·L−1), has been examined in individuals with type 1 diabetes treated with continuous subcutaneous insulin infusion (12). In 276 paired values of blood glucose and CGMS values, the CGMS correctly identified 33% of the hypoglycemic events, whereas 96% of the CGMS values correctly identified values in the absence of hypoglycemia (12).

A previous study that used the International Organization for Standardization (ISO) standards determined the accuracy of the CGMS over a range of glucose excursions (10). The investigators demonstrated that the performance of the CGMS was poor, with only 41% of the sensor readings meeting the ISO criteria when blood glucose concentrations were ≤ 4.1 mmol·L−1. The performance of the CGMS markedly increased when blood glucose concentrations were > 4.1 mmol·L−1, with between 45 and 60% of readings meeting the ISO performance criteria (10). It seems that the performance of CGMS is greatest in the euglycemic and/or hyperglycemic range. The CGMS's ability to accurately reflect blood glucose concentrations is insufficient, and its use seems limited to that of an adjunct for measurements rather than an alternative to intermittent blood glucose measurements by direct blood measurement.

Recently, the device has been successfully used to monitor the benefits of an 18-d calorie-restriction diet on whole-day changes in glycemic control in obese individuals with type 2 diabetes (7). In a separate study the CGMS was effectively used as an education tool to provide information for improving glycemic control in children with type 1 diabetes (30). The authors reported a significantly lower whole-day area under the curve 2.5 months after an initial assessment (30). Although the CGMS has been used to detect significant hyper- and hypoglycemic events in individuals with type 1 diabetes during intense exercise (5), its ability to measure and provide information in glycemic control after exercise in individuals with type 2 diabetes has yet to be determined.

Therefore, the initial aim of this study was to observe the effects of moderate exercise on glycemic control for 72 h and to measure the scale of changes in whole-day glucose profiles using the CGMS. The study also determined the level of agreement between venous blood glucose concentrations and CGMS values during exercise in individuals with and without type 2 diabetes.

Back to Top | Article Outline



Six (5 male, 1 female) sedentary individuals with recently diagnosed (2.0 ± 0.5 yr), diet-treated type 2 diabetes (diabetes group) and four male, sedentary, age-matched subjects without type 2 diabetes (control group) were recruited for this investigation. Diet-treated individuals with type 2 diabetes were recruited because this cohort is typically encouraged to increase their physical activity levels together with dietary alterations.

Body fat percentage was assessed from the sum of the skinfold thickness of the biceps, triceps, subscapular, and suprailliac sites, and by determining whole-body fat percentage using the equations of Durnin and Womersley (11). Standard clinical assessments demonstrated no evidence of preexisting diabetic complications or abnormalities during a 12-lead electrocardiogram. No subject was taking any antihyperglycemic drug at the time of data collection, ensuring that the glycemic profiles were not influenced by the action of such pharmacological agents. Three subjects in the diabetes group were being treated for hypertension and were treated with low doses of calcium channel blockers (5-10 mg, twice daily). Experimental procedures were approved by the East Sussex local research ethics committee. This study was conducted in accordance with the World Medical Association's revised Declaration of Helsinki for research into human volunteers. Subjects gave written informed consent before participating in the study.

Back to Top | Article Outline

Oral glucose tolerance test.

A 75-g, 2-h oral glucose tolerance test (OGTT) was performed before the experimental procedure to assess each subject's glucose tolerance to a bolus glucose load (28). Results of this 3-h OGTT ensured that patients in the diabetes group and control group met the classification criteria of the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus (28).

Back to Top | Article Outline

Experimental Procedure

Subjects reported to the laboratory following a 12-h overnight fast on two consecutive days, having abstained from alcohol and caffeine intake for 24 h and from exhaustive exercise for 48 h. Subjects were asked to continue with their normal daily life activities during the 3-d monitoring period but to refrain from exhaustive exercise.

Back to Top | Article Outline


The glucose sensor was inserted subcutaneously, at approximately 09:00 h on admission to the study (day 1), into the anterior abdominal wall of the subjects using a 22-gauge needle. Each subject wore the CGMS for a period of 3 d and was asked to calibrate the device by entering fingertip blood glucose readings from a portable blood glucose monitor (TheraSense Inc.) before breakfast, lunch, and dinner and before going to bed. Whole-day interstitial glucose profiles were collected continuously for 72 h. Glucose profiles were collected the day before (day 1), the day of (day 2), and the day following (day 3) a single 1-h bout of moderate exercise. At the end of the study period, CGMS sensors were removed and the stored data in the monitor was transferred by infrared through a serial port to an external computer. Subjects were asked to closely match their daily nutritional intake and kept detailed food diaries across the 3-d period.

At approximately 08:30 h on day 2 subjects reported to the laboratory, at which time an 18-gauge cannula was inserted into a dorsal forearm vein to enable frequent sampling of venous blood. At approximately 09:00 h, a baseline venous blood sample (∼10 mL) was drawn and dispensed into an EDTA coated sample tube. Subjects exercised on a cycle ergometer (Jaeger, ER 800, Germany) for 1 h at a power output corresponding to 90% of a predetermined LT. Venous blood samples were drawn every 10 min during exercise and were immediately analyzed for glucose using the glucose oxidase method, using a automated glucose analyser (YSI stat 2300 PLUS, Yellows Springs, OH).

Lactate threshold tests were performed 3 d before the main experimental procedures on an electronically braked cycle ergometer (Jaeger ER 800, Germany) using an incremental protocol. During the test, subjects were asked to maintain a pedal cadence of approximately 60 rpm. Tests began at a power output of 40 W, and the power output was increased by 10 W once every 3 min. At the end of each 3-min stage a fingertip capillary blood sample was drawn and immediately analyzed for blood lactate concentrations using an automated lactate analyser (YSI stat 2300 PLUS, Yellows Springs, OH). The lactate threshold was determined by a sudden and sustained increase in blood lactate concentrations above resting values (15). The power output corresponding to this lactate threshold was identified at a later stage by two independent, experienced exercise physiologists.

Back to Top | Article Outline

Data and Statistical Analysis

The software solutions software (Medtronic-MiniMed, Version 3.0B) generated a table of summary statistics, daily glucose plots, and a database of the recorded 5-min values. Statistical analysis was performed using the SPSS (version 10) software and a P value of <0.05 was taken as evidence of statistical significance. A repeated-measures ANOVA was used to determine differences over time, and Tukey's post hoc test was implemented where appropriate. To assess the statistical validity of the CGMS during exercise, regression-based, 95% limits of agreement (3) were calculated with time-matched venous blood glucose data. The clinical accuracy of the CGMS during exercise was assessed by the ISO standards for accuracy of point blood glucose tests. The sensor accuracy will be expressed as the percentage of CGMS values within 0.8 mmol·L−1 of a time-matched venous blood glucose value ≤ 4.1 mmol·L−1 and within 20% of venous blood glucose concentrations ≥4.1 mmol·L−1 (19). All values are reported as mean ± SEM.

Back to Top | Article Outline


The physical and biochemical characteristics of the subjects are detailed in Table 1. Nutritional analysis (Compeat, version 6; Visual Information Systems Ltd, UK) confirmed that there were no differences (P = 0.671) in total carbohydrate intake within subjects on day 1 (198.9 ± 18.1 g·d−1), day 2 (204.1 ± 14.2 g·d−1), or day 3 (199.6 ± 13.7 g·d−1). There was no significant difference in glycemic index classification of the consumed carbohydrate between the 3 d (P = 0.774). Therefore, changes in whole-day glycemia seen across the 3-d period could largely be attributed to the exercise period conducted on day 2.

Fasting venous glucose concentrations (P = 0.02) and CGMS values (P = 0.035) immediately before the start of exercise were higher in the diabetes group compared with the control group (Table 1). The CGMS overestimated the fasting venous glucose concentration by approximately 1.3 (∼22%) and approximately 2.6 mmol·L−1 (∼20%) in the control (P = 0.01) and diabetes (P = 0.006) groups, respectively, at rest. In the diabetes group, venous blood glucose and CGMS values significantly declined during exercise (Fig. 1). In contrast, venous blood glucose or CGMS values did not change during exercise in the control group (Fig. 1). There was no significant difference in the response of venous blood glucose or CGMS values to exercise in the diabetes group (P = 0.108), yet CGMS values were significantly higher throughout exercise in the control group (P = 0.032).

Back to Top | Article Outline

Level of Agreement

During exercise, the mean differences between venous blood glucose concentrations and CGMS values were −1.7 ± 0.6 and −1.5 ± 1.1 mmol·L−1 in the control group (P = 0.003) and in the diabetes group, respectively (P = 0.045). The size of this difference was not different between groups (P = 0.330), confirming that the CGMS overestimated venous blood glucose to a similar degree in both groups. During exercise in the control group, 3.8% of the data points were outside the 95% limits of agreement (Fig. 3). In the diabetes group, the variability of the differences increased as the average of the CGMS and venous blood glucose increased (Fig. 2). Therefore, regression-based 95% limits of agreement (3) were calculated and similar to that of the control group: 4.5% of the data points were outside the 95% confidence intervals. Although the CGMS overestimated venous blood concentrations, statistically, there was an acceptable level of agreement between venous blood glucose and CGMS values in both groups during exercise (Figs. 2 and 3). According to the ISO standards for accuracy of point glucose tests, 50% of the CGMS values (21 of 42 samples) met the requirements in the diabetes group, whereas none of the 28 CGMS values conformed to the criteria in the control group during exercise.

Whole-day average glucose concentrations were approximately 13% lower on day 3 than on day 1 (P = 0.007; Fig. 4) in the diabetes group. There was no difference in whole-day average glucose concentrations between days 2 and 3 or days 1 and 2 in the diabetes group (Fig. 4).

Whole-day area under the glucose response curve was approximately 27% lower on day 3 in comparison with day 1 (Fig. 5; P = 0.016), and approximately 15% lower on day 3 compared with day 2 (Fig. 5; P = 0.236) in the diabetes group. The total time spent per day within ±10% of fasting venous glucose (FVG) concentrations in the diabetes group increased by approximately 54.9% between days 1 and 3 and by approximately 59.4% between day 2 and 3 (Fig. 7; P = 0.018), yet had no significant effect when comparing day 1 with day 2 (Fig. 7). There were no significant differences in the whole-day average glucose (Fig. 4; P = 0.644), whole-day AUC (Fig. 5; P = 0.305), or the total time spent within ±10% of fasting venous glucose (Fig. 6; P = 0.237) across the 3 d in the control group.

Back to Top | Article Outline


Clinical benefits of acute exercise on glycemic control can be seen during and up to 20 h following exercise. To quantify these changes in glycemia, multiple venous or fingertip blood samples are required. The recent introduction of continuous glucose monitoring offers the opportunity to overcome these impracticalities and generate "real life" information on the glycemic responses to exercise. Therefore, the purpose of the present study was to assess the efficacy of continuous glucose monitoring during 1h of moderate exercise and to quantify exercise-induced changes in whole-day measurement of glycemic control in subjects with and without type 2 diabetes. The data presented in this study demonstrate that there is statistically acceptable agreement between the CGMS and venous blood glucose concentrations during moderate exercise in both groups. Therefore, the CGMS seems to be a useful tool that, in conjunction with measures of blood glucose, could aid in the mapping of changes in glucose metabolism during and in the short and medium term following exercise in individuals with and without type 2 diabetes.

The data collected in this study are in keeping with the majority (16,20,23-25) but not all (6) published data indicating that moderate exercise causes a significant reduction in blood glucose concentration in individuals with type 2 diabetes. The magnitude of the decline in blood glucose during exercise in the diabetes group, in this study, is comparable with other previously reported data (16,18,25) that have utilized similar exercise intensities (∼50-60% V̇O2peak) and durations (40-60 min). This reduction in blood glucose is likely due to a blunted or delayed hepatic glucose production (23) coupled with a normal or enhanced glucose utilization in response to exercise in individuals with type 2 diabetes (24).

The present findings show that a single bout of moderate exercise has a beneficial effect on glycemia for at least 24 h after exercise in obese individuals with type 2 diabetes, an effect that was not observed in the control group using the CGMS. The improvement in glycemic control seen in the diabetes group is in keeping with the findings of others who have shown improvements immediately after (20) and up to 20 h following exercise (9).

In the diabetes group, the total time spent within ±10% of fasting venous glucose concentration increased between days 2 and 3, with no significant effect reported in whole-day average glucose concentrations or AUC. This demonstrates that the exercise-induced, whole-day improvements in glycemic control can be seen immediately following and up to 24 h after exercise. The effect of the exercise must therefore be to reduce the amplitude of the glycemic excursions rather than to simply lower the total daily blood glucose concentration. These changes in glycemic control, as measured by the CGMS, were not seen in the control group, likely because of the inability of the CGMS to detect small changes in glycemic control and the expected good glycemic control that should be present in nondiabetic subjects.

In the present study, at least in the diabetes group, the percentage of CGMS values that conform to the ISO standards is similar to other studies (10). It seems that the accuracy of the CGMS is influenced by the glycemic state, as the accuracy seems to be reduced in conditions of hypoglycemia and greatest in the hyperglycemic range. However in the present study, the CGMS has shown that it can only accurately measure between 0 and 50% of time-matched reference values, demonstrating its inadequacies in measure point glucose values during exercise in individuals with and without type 2 diabetes.

The effectiveness of continuous glucose monitoring in subcutaneous tissue is based on the assumption that interstitial glucose accurately reflects whole-body blood glucose concentrations during glycemic variations (27). The present study used venous blood glucose as a comparative marker to assess the efficacy of the CGMS to reflect changes in glucose concentrations during exercise. Under basal conditions the difference between venous blood and interstitial glucose values has been reported to be approximately 1.3 mmol·L−1, whereas during a hyperinsulinemic clamp this discrepancy increases to approximately 2.6 mmol·L−1 (27). The overestimation of blood glucose concentrations by the CGMS in the present study may therefore reflect the biological time delay between venous blood and interstitial glucose concentrations (27).

In conclusion, we demonstrate that a period of moderate exercise has a beneficial effect on short- and medium-terminterstitial glucose profiles, as measured by the CGMS following exercise in obese individuals with type 2 diabetes. The indication is that a single episode of moderate exercise can have a beneficial effect on blood and interstitial glucose concentrations both during and following exercise in individuals with type 2 diabetes. The CGMS does provide statistically significant agreement with venous blood glucose during exercise, yet its ability to predict time-matched venous blood glucose measurements is insufficient to warrant its use in place of blood glucose measurements in clinical settings. However, it is simple to use, and a convenient adjunct to frequent fingertip/venous blood sampling that offers detailed information as to the duration, magnitude, and frequency in glucose levels, which are beyond the scope of intermittent blood sampling. Therefore, the CGMS would aid intermittent blood sampling in the monitoring and recording of whole-day exercise-induced changes in glycemia in individuals with type 2 diabetes.

The authors would like to thank Medtronic-MiniMed (Watford, UK) for their generous financial support of this project. A special thank you for technical support goes to Hannah Dickinson, Ian Tilley, and Morag McLaren (Medtronic-Minimed, Watford, UK) and Ann Attfield and Patrick Smith (Chelsea School, University of Brighton).

Back to Top | Article Outline


1. American College of Sports Medicine. Exercise and type 2 diabetes: position stand. Med. Sci. Sports Exerc. 32:1345-1359, 2000.
2. American Diabetes Association. Physical activity/exercise and diabetes. Diabetes Care 27:S58-S62, 2004.
3. Bland, J. M., and D. G. Altman. Measuring agreement in method comparison studies. Stat. Methods Med. Res. 8:135-160, 1999.
4. Boyne, M. S., D. M. Silver, J. Kaplan, and C. D. Saudek. Timing of changes in interstitial and venous blood glucose measured with a continuous subcutaneous glucose sensor. Diabetes 52:2790-2794, 2003.
5. Carr, J. M., and M. H. Corcoran. Continuous glucose monitoring in marathon runners with type 1 diabetes during the 2001 Chicago marathon. Diabetes 51:988, 2002.
6. Colberg, S. R., J. M. Hagberg, S. D. McCole, J. M. Zmuda, P. D. Thompson, and D. E. Kelley. Utilization of glycogen but not plasma glucose is reduced in individuals with NIDDM during mild-intensity exercise. J. Appl. Physiol. 81:2027-2033, 1996.
7. Colette, C., C. Ginet, C. Boegner, et al. Dichotomous responses of inter and postprandial hyperglycemia to short-term calorie restriction in patients with type 2 diabetes. Eur. J. Clin. Invest. 35:259-264, 2005.
8. Cusi, K., K. Maezono, A. Osman, et al. Insulin resistance differentially affects the PI 3-kinase and MAP kinase-mediated signaling in human muscle. J. Clin. Invest. 105:311-320, 2000.
9. Devlin, J., M. Hirshman, E. Horton, and E. Horton. Enhanced peripheral and splanchnic insulin sensitivity in NIDDM men after a single bout of exercise. Diabetes 36:434-439, 1987.
10. Diabetes Research in Children Network (DirecNet) Study Group. The accuracy of the CGMS in children with type 1 diabetes: results of the diabetes research in children network (DirecNet) accuracy study. Diabetes Technol. Ther. 5:781-789, 2003.
11. Durnin, J. V., and J. Worsley. Body fat assessed from total body density and its estimation from skinfold thickness: measurements on 481 men and women aged 16 to 72 years. Br. J. Nutr. 32:77-91, 1974.
12. Guerci, B., M. Floriot, P. Bohme, et al. Clinical performance of CGMS in type 1 diabetic patients treated by continuous subcutaneous insulin infusion using insulin analogs. Diabetes Care 26:582-589, 2003.
13. Gross, T. M., B. W. Bode, D. Einhorn, et al. Performance evaluation of the MiniMed continuous glucose monitoring systemduring patient home use. Diabetes Technol. Ther. 2:49-56, 2000.
14. Henriksen, E. J. Exercise effects of muscle insulin signaling and action: invited review: effects of acute exercise and exercise training on insulin resistance. J. Appl. Physiol. 93:788-796, 2002.
15. Jones, A. M., and J. H. Doust. Limitations to submaximal exercise performance. In: Kinanthropometry and Exercise Physiology Laboratory Manual, R. Eston and T. Reilly (Eds.). London: Routledge, Human Kinetics, 2002, pp. 235-262.
16. Kang, J., D. E. Kelley, R. J. Robertson, et al. Substrate utilization and glucose turnover during exercise of varying intensities in individuals with NIDDM. Med. Sci. Sports Exerc. 31:82-89, 1999.
17. Kennedy, J., M. Hirshman, E. Gervino, et al. Acute exercise induces GLUT4 translocation in skeletal muscle of normal human subjects and subjects with type 2 diabetes. Diabetes 48:1192-1197, 1999.
18. Kim, J., R. S. Solis, E. B. Arias, and G. D. Cartee. Postcontraction insulin sensitivity: relationship with contraction protocol, glycogen concentration, and 5′AMP-activated protein kinase phosphorylation. J. Appl. Physiol. 96:575-583, 2004.
19. Klonoff, D. C. Continuous glucose monitoring. roadmap for 21st century diabetes therapy. Diabetes Care 28:1231-1239, 2005.
20. Larsen, J. J., F. Dela, M. Kjaer, and H. Galbo. The effect of moderate exercise on postprandial glucose homeostasis in NIDDM patients. Diabetologia 40:447-453, 1997.
21. Mastrototaro, J. J. The MiniMed continuous glucose monitoring system (CGMS). J. Pediatr. Endocrinol. Metab. 12:751-758, 1999.
22. Mastrototaro, J. J., and T. M. Gross. Reproducibility of the continuous glucose monitoring system matches previous reports and the intended use of the product. Diabetes Care 26:256, 2003.
23. Minuk, H. L., M. Vranic, E. B. Marliss, A. K. Hanna, A. M. Albisser, and B. Zinman, Glucoregulatory and metabolic response to exercise in obese noninsulin-dependent diabetes. Am. J. Physiol. 240:E458-464, 1981.
24. Musi, N., N. Fujii, M. F. Hirshman, et al. AMP-activated protein kinase (AMPK) is activated in muscle of subjects with type 2 diabetes during exercise. Diabetes 50:921-927, 2001.
25. Poirier, P., S. Mawhinney, L. Grondin, et al. Prior meal enhances the plasma glucose lowering effect of exercise in type 2 diabetes. Med. Sci. Sports Exerc. 33:1259-1264, 2001.
26. Reach, G. Continuous glucose sensing for the treatment of diabetes mellitus. Analysis 39:609-612, 1993.
27. Rebrin, K., G. M. Steil, W. P. Van Antwerp, and J. J. Mastototaro. Subcutaneous glucose predicts plasma glucose independent of insulin: implications for continuous monitoring. Am. J. Physiol. 40:E561-E571, 1999.
28. The Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Report of the expert committee on the diagnosis and classification of diabetes mellitus. Diabetes Care 25:S5-S20, 2002.
29. Wallerberg-Henriksen, H., S. H. Constable, D. A. Young, and J. O. Holloszy. Glucose transport in rat skeletal muscle: interaction between exercise and insulin. J. Appl. Physiol. 65:909-913, 1988.
30. Weintrob, N., A. Schechter, H. Benzxzaquen, et al. Glycemic patterns detected by continuous subcutaneous glucose sensing in children and adolescents with type 1 diabetes mellitus treated by multiple daily injections vs. continuous subcutaneous insulin. Arch. Pediatr. Adolesc. Med. 158:677-684, 2004.


©2006The American College of Sports Medicine