The Effect of Water-Based Exercise on Glucose and Insulin Response in Overweight Women: A Pilot Study

Abstract

Objective:

The aim of this study was to investigate the effectiveness of a 12-week moderate intensity, water-based circuit-type training intervention on glucose and insulin responses in overweight women with normal or impaired glucose tolerance.

Methods:

Fifteen overweight women (body mass index [BMI] > 25 kg/m²) with normal glucose tolerance (NGT; n = 7) or impaired glucose tolerance (IGT; n = 8) were recruited for this study. All women completed a 12-week training intervention utilizing a combination of aerobic and resistance exercises in an aquatic environment, 3 days per week and 60 min per session at 70–75% mode-specific maximum heart rate. A standard 75-g oral glucose tolerance test (OGTT) was administered pre- and post-intervention, from which fasting and post-load plasma insulin and glucose levels were assessed.

Results:

Following the 12-week period, fasting insulin levels had decreased by 44% and 2-h glucose by 30.4% in the group with IGT. Waist circumference (WC) had decreased by 5.3% in this group at the end of the intervention. Only WC and waist-to-hip ratio (WHR) decreased (6.0% and 5.5%, respectively) following the intervention in the NGT group.

Conclusions:

Moderate intensity, water-based circuit-type exercises appear to be an effective exercise modality to improve glucose and insulin response to a glucose challenge in overweight women with IGT.

Introduction

Impaired glucose tolerance (IGT), characterized by post-challenge hyperglycemia (2-h glucose 7.8–11.1 mmol/l), is a recognized pre-diabetic state, and is also closely associated with insulin resistance, obesity (particularly abdominal obesity), and cardiovascular disease (CVD).^1,2 Obesity and insulin resistance are, in turn, independent risk factors for the development of IGT, type 2 diabetes, and CVD.³ Isolated impaired fasting glucose (IFG), defined as a fasting blood glucose level of 5.6–6.9 mmol/l,¹ is also considered a pre-diabetic state, which occurs as a result of defective insulin production and beta-cell dysfunction.⁴

Physical activity is a key component in the prevention and management of obesity and diabetes,^5,6 and in reducing the risk of progression from IGT to diabetes.^7
–9 Both resistance and aerobic training are effective in increasing insulin action and glucose tolerance in those with IGT or diabetes.^8,10,11 Moreover, circuit training, which combines resistance training and aerobic exercise, is also effective in improving glycemic control in IGT and diabetic populations, and may be more effective than either modality alone.^11

–14 Circuit training is normally undertaken in a weight-bearing environment and, while effective, may not be practical for overweight individuals with limited mobility and/or joint problems. Recently, water-based exercise has been considered an attractive alternative exercising environment, as it allows overweight individuals to exercise without putting undue stress on joints.¹⁵ There is substantial evidence to support the effectiveness of various forms of water-based exercise in improving and/or maintaining cardiovascular fitness and strength in many sectors of the population.^15
–17 Swim training lowered 2-h glucose in adolescent humans with Type 1 diabetes;¹⁸ however, this form of water-based exercise has not demonstrated the same glucose-lowering effects in older obese men and women.¹⁹ Furthermore, there is a conundrum as to the appropriate exercise intensity to effect changes in glucose and insulin metabolism. Improvements in insulin sensitivity have been observed with low-intensity,²⁰ moderate-intensity,²¹ and vigorous-intensity²¹ exercise. However, lower intensity exercise may have the added benefits of improving adherence and enjoyment,²² and improving metabolic fitness.^20,21 There are no studies, to our knowledge, that have investigated the effects of moderate-intensity, water-based circuit-type exercise on glucose and insulin responses in humans. Therefore, the purpose of this study was to investigate the effects of a 12-week moderate-intensity, water-based circuit exercise program on glucose tolerance and insulin responses in overweight women identified as either normal or impaired glucose tolerant from a 75-g oral glucose tolerance test (OGTT).

Methods

Previously inactive, overweight women (BMI > 25 kg/m²) were recruited via advertisement to participate in this study, which was approved by the University of Otago Ethics Committee. Exclusion criteria included women on hormone replacement therapy and those with any known medical condition that would prevent them from participating in an exercise intervention. Prior to acceptance into the study, all participants were medically screened using the modified PAR-Q and a glucose challenge undertaken (OGTT) to assess glucose tolerance. Recruitment was dependent on 2-h post-challenge glucose, IGT characterized by plasma glucose of 7.8–11.1 mmol/l and normal glucose tolerance (NGT) of <7.8 mol/l. Individuals with isolated IFG were not excluded. Fifteen participants with a BMI of >25 kg/m² were recruited for the study: eight identified as IGT and seven with NGT. Any medications were noted, and participants were accepted into the study if they had been regularly taking the medications for more than 6 months without alterations to drug and/or dosage. Four participants in the NGT group reported using medications on a regular basis: three taking a thyroid medication and one using a diuretic. Of those women identified as having IGT, four individuals were regularly taking a combination of oral hypoglycemics, statins, diuretics, ACE inhibitors, fibrates, α1-adrenoreceptor blockers, and/or anti-depressants; a β-blocker was taken by three of these four women. One individual in the IGT group was taking an anti-depressant only. All participants completed the intervention without alteration to type or dosage of their regular medication. Participants were also asked not to change their dietary habits over the course of the exercise intervention and were instructed to replicate their diet in the 24-h period immediately prior to OGT testing. Informed written consent was obtained from all participants prior to testing.

Oral glucose tolerance test

A standard glucose tolerance test was administered to determine insulin levels and glucose tolerance. Baseline measures were taken after an overnight fast (10–12 h), after which subjects ingested a 300-ml glucose solution containing 75 g of a glucose polymer (Glucaid, Histolabs, Riverstone, Australia). Thereafter, approximately 10 ml of blood was drawn at 30, 60, 90, and 120 min. Plasma glucose was analyzed using the hexokinase method on a Roche Cobas Mira automated analyzer (Roche Diagnostics Corp., Indianapolis, IN), while insulin was analyzed by radioimmunoassay (RIA) using the Coat-A-Count assay (Diagnostic Products Corp., Indianapolis, IN) auto analyzer. To reduce the acute effect of exercise on glucose tolerance and insulin action, subjects underwent their post-intervention blood tests 72 h after completing their final exercise session. Area under the curve (AUC) was calculated for glucose and insulin responses using the trapezoidal rule.

Anthropometry

Height, weight, and waist and hip circumferences were taken immediately prior to the initial deep water run test for aerobic fitness and again at 3–5 days after the final training session. Height was determined without shoes using a free-standing stadiometer, while weight was taken using electronic scales (Digi, D1-10; Teraoka Ltd., Tokyo, Japan) with participants wearing their swimsuit. A standard fibreglass anthropometric tape was used to measure waist circumference, defined as the narrowest part between the last rib and the anterior superior iliac spine (ASIS), and hip circumference, defined as the widest part between the ASIS and the greater trochanter. All measurements were duplicated, and a tolerance limit was set at 1 kg for weight and 1 cm for height and circumference measures. Weight was measured to the nearest 0.1 kg; if the two measures differed by more than 1 kg, a third measure was taken. Similarly, if height and circumference differed by more than 1 cm on the first two measures, a third was taken. The average of the two closest measurements was used in the analyses. Body mass index (BMI: weight [kg]/height [m²]) was used to estimate adiposity, while waist/hip ratios (WHR) were calculated to describe body fat distribution.

Maximal oxygen uptake testing and exercise intervention

A mode-specific deep water running test was undertaken to determine maximal oxygen consumption. Maximal oxygen consumption was defined as V̇O_2peak, rather than V̇O_2max. V̇O_2max is usually identified as the point where a plateau of oxygen uptake occurs despite an increase in workload and/or in healthy subjects,²³ and/or attainment of a certain concentration of blood lactate, a specific respiratory exchange ratio (RER) and some percentage of age-adjusted maximal heart rate (HRmax).²⁴ Participants in this study were older, sedentary, and overweight, potentially affecting the criteria generally used to determine V̇O_2max; thus, the term V̇O_2peak was considered more appropriate.

Testing was performed in the swimming flume at the School of Physical Education, University of Otago. Due to the acute effect of β-blockade on exercise heart rate, patients on β-blockers were tested while medicated, as the derivation of a training heart rate range must be based on an exercise test conducted under conditions as similar as possible, with respect to the timing of medication, to those under which the individual will be exercising.²⁵ Immediately prior to the test, each participant was instructed in correct deep water running technique over a period of 10 min to ensure familiarity with the testing conditions. Participants then rested for a minimum of 10 min before beginning the actual test. As heart rate and oxygen kinetics are affected by water temperature, the flume was maintained at a thermoneutral level of 29°C and the depth constant, at 1.80 m. All participants were required to refrain from alcohol use and strenuous exercise for 24 h before the test and to avoid caffeine intake for 4 h before the test. V̇O_2peak was determined using the progressive deep water protocol of Mercer and Jensen²⁶ to voluntary exhaustion. This protocol is continuous and consists of 1-min stages. Test participants were submerged to neck level while wearing a flotation belt, with a tether attached to the back and run through a series of pulleys. The other end of the tether is attached to a bucket suspended in front of the participant and above a wooden plank that rested across the flume deck. To provide a graded response, a weight was added to the bucket at the beginning of each 1-min stage. If the participant was not able to meet the intensity at a given stage, the bucket dropped onto the wooden plank; this was accepted as the endpoint of the test. This protocol has been found to be valid and reliable for assessing maximal aerobic uptake while deep water running.^26,27 Respiratory gas analysis was carried out during the protocol, with oxygen uptake (V̇O₂) determined using a Beckman metabolic cart (Sensormedics, Yorba Linda, CA). Heart rate was monitored continuously (Polar Favor, Kempele, Finland). The highest oxygen uptake (V̇O₂) and corresponding HR in the final 60 sec of the exercise test were taken as V̇O_2peak and HR_peak, respectively. The corresponding respiratory exchange ratio (RER) to V̇O_2peak was determined from gas analysis.

Participants performed the 12-week water-based circuit program, using a group-based format, under supervision of a physical activity educator 3 days per week, for a total of 60 min per session. Exercise intensity was set at 70–75% maximum heart rate obtained during pre-test maximal water-based exercise testing and at a rating of perceived exertion of 11–14.²⁸ By reducing the speed of movements in the water, all participants were able to meet their target heart rate from week 1 of the intervention. Speed was increased as participants adapted to the exercise movements to ensure target heart rate was met throughout the 12 weeks. Heart rate monitors (Polar Favor, Kempele, Finland) were worn by each participant at all training sessions. The aquatic circuit comprised a 5-min warm-up, 50 min of deep water running interspersed with 60–90 sec of water-based resistance exercises, and a 5-min cool down. The resistance exercises utilized movements designed to work all areas of the body using either the resistive properties of the medium, or foam dumbbells. Each training session consisted of one set of five primary exercises for the upper body: chest press, chest flys (with a slight bend in the elbows and arms stretched out to the sides, hands were brought together at shoulder height against the resistance of the water, using a wide “hugging action”), tricep extension, a combination movement of pushing down and pulling up in a vertical plane, and bicep curls. Four primary exercises for the lower body were as follows: star jumps (hip abduction/adduction), scissor kicks (hip extension and flexion), leg press, and knee extension and flexion. Two exercises were utilized for the abdominals: trunk curls and oblique trunk curls. All exercises were performed in an upright position with participants submerged to neck level whilst wearing a flotation belt.

Statistical analysis

Shapiro-Wilk testing was undertaken to assess normality of data distribution. The effects of exercise training on changes in anthropometric measures, fasting, and post-load glucose and insulin responses were assessed using independent t-tests for between group differences at baseline and on completion of the 12-week intervention. Paired t-tests were undertaken pre- and post-intervention to assess within-group differences for all measures.

Results

Fasting insulin was found to violate normality of data distribution. Values were log transformed (base₁₀) for analysis; however, all data are presented in Table 1 as mean ± SEM for ease of understanding. All eligible participants completed the intervention, with 100% compliance for three sessions per week over the 12-week intervention period. Pre- and post-intervention data are presented in Table 1. Groups did not differ significantly for age, weight, BMI, WC, WHR, V̇O_2peak, HR_peak, or RER before or after 12 weeks of circuit water-based training (Table 1).

Table 1.

Pre- and Post-Intervention Demographic and Oral Glucose Tolerance Data (Mean ± SEM)

	NGT (n = 7)		IGT (n = 8)
	Pre	Post	Pre	Post
Age (years)	57 ± 3.7 (37,65)	57 ± 3.7	58 ± 3.1 (39,64)	58 ± 3.1
Weight (kg)	86.8 ± 7.6 (58.8,123.9)	86.2 ± 7.5 (57.0,121.5)	86.5 ± 5.8 (63.8,119.0)	85.3 ± 6.1 (63.0,119.8)
BMI (kg/m²)	33.6 ± 2.1 (26.9,43.2)	33.2 ± 2.1 (26.8,42.3)	33.3 ± 1.9 (26.2,42.4)	33.1 ± 1.9 (24.9,42.5)
WC (cm)	104.1 ± 5.1 (78–123)	97.9 ± 5.1^a (74–119)	105.1 ± 3.4 (90–122)	99.5 ± 3.6^a (83–118)
WHR	0.91 ± 0.02 (0.84–0.98)	0.86 ± 0.01^a (0.82–0.89)	0.92 ± 0.01 (0.89–0.99)	0.89 ± 0.01 (0.84–0.94)
V̇O_2peak (l/min)	1.58 ± 0.17 (1.05–2.50)	1.63 ± 0.17 (1.01–2.49)	1.29 ± 0.13 (0.80–1.83)	1.50 ± 0.09 (1.08–1.76)
HR_peak (bpm)	141 ± 8 (101–158)	144 ± 7 (115–153)	130 ± 4 (110–166)	143 ± 6 (118–168)
RER	1.06 ± 0.04 (0.87,1.11)	1.02 ± 0.04 (0.79,1.10)	0.97 ± 0.02 (0.85,1.03)	1.00 ± 0.02 (0.92,1.05)
FPG (mmol/l)	5.3 ± 0.4 (4.1–6.9)	5.3 ± 0.3 (4.6–6.9)	6.1 ± 0.5 (3.4–7.7)	6.3 ± 0.3 (5.2–7.5)
2-h glucose (mmol/l)	5.1 ± 0.5 (2.8–6.3)	4.5 ± 0.3 (3.4–6.1)	10.2 ± 0.6^b (8.1–13.4)	7.1 ± 0.6^a,b (4.3–9.2)
AUC_g (mmol/l/min)	793.3 ± 106.6 (474.7,1234.4)	734.7 ± 68.2 (506.9,1007.5)	1177.7 ± 92.3^d (814.1,1633.1)	1163.0 ± 68.8^b (882.5-1460.6)
FPI (pmol/l)	96.5 ± 49.5 (20.3–388.8)	74.4 ± 32.3 (20.1–261.4)	75.4 ± 6.6 (54.9,109.7)	42.0 ± 10.7^c (16.3, 98.1)
2-h insulin (pmol/l)	371.1 ± 137.5 (41.3,1016.0)	334.7 ± 111.2 (45.6,906.5)	620.1 ± 102.2 (63.3,1027.6)	505.3 ± 93.6 (100.2,834.1)
AUC_i ( × 10³) (pmol/l/min)	63.9 ± 20.5 (11.9,175.3)	59.9 ± 15.8 (16.4,137.0)	68.1 ± 11.2 (19.6,118.6)	63.6 ± 10.5 (23.6,109.2)

Within-group difference, p < 0.01.

Between-group difference at equivalent time point, p < 0.01.

Within-group difference for equivalent time point, p < 0.05.

Between-group difference, p < 0.05.

Data are presented as mean ± SEM. Minimum and maximum values for each variable are presented in parentheses.

WC, waist circumference; WHR, waist-hip ratio; V̇O_2peak, peak oxygen consumption; HR_peak, peak heart rate; RER, respiratory exchange ratio; FPG, fasting plasma glucose; AUC_g, area under the glucose curve; FPI, fasting plasma insulin; AUC_i, area under the insulin curve.

Training effects

In the group with IGT, WC, 2-h glucose levels, and fasting plasma insulin decreased significantly by 5.3% (95% confidence interval [CI] of the difference in means 3.5–7.7 cm), 30.4% (95% CI 1.4–4.9 mmol/l), and 44.2% (95% CI 0.05–0.62 pmol/l), respectively, on completion of the training intervention compared to their pre-intervention measures (Table 1). In the group with NGT, only the anthropometric variables of WC 6.0% (95% CI 3.2–9.2 cm) and WHR 5.5% (95% CI 0.02–0.09) showed significant pre- to post-intervention reductions (Table 1).

Significant differences between the groups for 2-h glucose were observed before (95% CI −6.8 to −3.6 mmol/l) and after (95% CI −4.1 to −1.0 mmol/l) the 12-week exercise program; however, the mean post-load glucose value fell within the normal range (<7.8 mmol/l)¹ for the IGT group at the end of the intervention period (Table 1). There were no significant differences within or between the groups for fasting blood glucose values pre- or post-intervention; however, the IGT group value continued to demonstrate IFG levels at both time points. The integrated AUC for glucose was significantly different between the IGT and NGT groups at both pre-intervention (95% CI −694.0 to −75.0 mmol/l/min) and post-intervention (95% CI −638.7 to −217.9 mmol/l/min) time points (Table 1).

Discussion

The main findings in this study were that following 12 weeks of moderate intensity, water-based circuit training, waist circumference, fasting insulin, and 2-h glucose responses to an oral glucose load were reduced in overweight women with IGT. These changes occurred without any significant alteration in weight or cardiovascular fitness levels. The significance of this study is twofold: it is the first study to investigate the effects of water-based training using a combination of aerobic and resistance training modalities on glucose and insulin responses; and second, moderate intensity exercise can improve markers of metabolic fitness, specifically plasma glucose and insulin levels, and anthropometric variables, in overweight and obese women, despite little change in cardiovascular fitness.

Circuit-based training combines elements of aerobic exercise and resistance training, and has been shown to be more effective than either modality alone.^11

–14 Much of the evidence for the beneficial effects of exercise has been derived from land-based studies;^13,14 however, water provides a non-weight-bearing environment, which can reduce joint stress, making exercise more manageable for sectors of the population such as obese or elderly individuals, or those with joint problems. In the present study, mean BMI for both groups reflected Grade 1 obesity; thus, the non-weight-bearing nature of the exercise program was appropriate. We had 100% attendance, with all women completing 36 sessions of 60 min of moderate intensity training. There were no injuries reported as a result of this intervention. All sessions were supervised, and this may have contributed to the high attendance rate and safety of the exercise program. However, this form of exercise could be undertaken independently following an introductory period to ensure individuals understand monitoring of heart rate and familiarize themselves with the resistance exercises.

The water-based circuit intervention utilized in the present study was successful in reducing fasting insulin and lowering post-load glucose levels to within the normal range (<7.8 mmol/l) in the IGT group. While land-based circuit training interventions have consistently improved glucose control in individuals with Type 2 diabetes or IGT,^11

–14 this is the first study to find positive effects on glucose control using a water-based equivalent. Water-based exercise in the form of swim training has been effective in improving glucose metabolism in adolescent Type 1 diabetics,¹⁸ although it has not had the same effects in older obese individuals.¹⁹ Regular exercise enhances glucose uptake by increasing the efficacy of the molecular components of signaling pathways involved in glucose metabolism and insulin sensitivity in glucose-sensitive tissues.²⁹ It has been suggested that aerobic and resistance exercise affects different components of the metabolic pathways and underpins the additive effects of the two modalities.¹⁴

While our training intervention was successful in reducing fasting insulin and 2-h glucose, the IGT group still demonstrated a mean post-intervention fasting plasma glucose level that placed them within the diagnostic criteria for impaired fasting glucose (IFG).¹ The mechanism responsible for the improvement in 2-h plasma glucose after training is likely due to the different metabolic determinants of IFG and IGT. Normal control of fasting glucose depends on the ability to maintain appropriate basal insulin secretion, and on appropriate levels of insulin sensitivity in the liver to control hepatic glucose output; thus, IFG is indicative of defects in β-cell function.³⁰ However, IGT is associated with peripheral insulin resistance, most importantly at the level of the skeletal muscle, which is the main depot of glucose disposal postprandially.³¹ In those with both IFG and IGT, peripheral insulin sensitivity is significantly reduced compared with individuals with NGT.⁴ Accordingly, physical activity interventions have the greatest effect on reducing 2-h plasma glucose levels in those with IGT,³¹ and are less likely to be effective in reversing isolated IFG or those with combined IFG/IGT.

Many exercise studies suggest improvements in insulin sensitivity occur as a result of changes within muscle tissue. Exercise training has been shown to produce an up-regulation of glucose transport through an increase in GLUT-4 transporter expression.³² Other improvements predominantly seen within the skeletal muscle include increased capillary density,³³ increased muscle blood flow,³⁴ increased mitochondrial size and number,³⁵ increased glycolytic and oxidative enzymatic activity,³⁶ increased glycogen synthase activity,^35,37,38 decreased intramyocellular lipid (IMCL) content and droplet size,³⁹ and a shift to more insulin-sensitive fiber types.^33,40 Increases in FFM have also been implicated in improved glucose tolerance after resistance training.⁴¹ Although changes in lean body mass were not assessed in the current investigation, the resistive effect of water provides exercise loading during limb movements, which enhances muscular tension and thus provides a training stimulus for muscular development.⁴²

Post-intervention WC decreased significantly in both groups, with WHR decreasing in the NGT group only; however, only the IGT group demonstrated changes in glucose and insulin responses. No change was observed in body weight; thus, it is possible that improvement in glucose and insulin responses in the IGT group was influenced by increases in lean body mass. Alternatively, changes in WC may be reflective of reductions of visceral or subcutaneous fat mass without any change in body weight or BMI, and it is the change in body fat depot that may affect improved insulin action and/or glucose tolerance.⁴³ Waist circumference predicts diabetes and CVD to a greater extent than BMI, blood pressure, lipoproteins, or glucose;⁴⁴ however, others report that risks of IGT and Type 2 diabetes are reduced with higher levels of physical activity, independent of the level of abdominal obesity.^10,45 Thus, it is difficult to identify whether increased physical activity or reduced WC has the greater impact on 2-h glucose. To further compound the issue, the evidence is equivocal on whether it is exercise intensity, energy expenditure, or exercise duration that plays a greater role in improving glucose control. While exercise duration and energy expenditure contribute to the improvements in glucose tolerance and insulin sensitivity, these metabolic benefits can be conferred with low- to moderate-intensity exercise.⁹ Moreover, improvements are independent of changes in cardiorespiratory fitness, or changes in body composition, weight, or other anthropometric measures.⁴⁶

The strengths of this study were that the exercise was prescribed at a percentage of mode-specific maximum (i.e., our participants undertook their maximal testing in water) and that the exercise modality chosen was well received, evidenced by 100% adherence throughout the 12 weeks of the intervention. We also acknowledge that the small sample size in this study did not permit the use of more advanced statistical procedures that may have resolved some of the interactions between exercise duration, WC changes, and 2-h glucose. Significant changes in cardiovascular fitness measures were also limited by the sample size. Three of our participants in the IGT group were taking beta-blockers, and thus a true maximal heart rate and oxygen consumption could not be detected. Additionally, there are large inter- and intra-individual variations in maximal exercise testing. This test can be substantially influenced by the testing environment and by intrinsic factors, including age, participant motivation, lean body mass, habitual activity levels, diurnal rhythms, and stress hormones.⁴⁷ We endeavored to limit variation by standardizing the testing environment and by providing strong verbal encouragement throughout the duration of the test. However, our results provide the initial evidence that moderate intensity water-based circuit exercise can improve glucose and insulin responses in overweight women with IGT. This evidence can form the basis from which further research may be undertaken using larger sample sizes.

Conclusions

Our findings lend support to the use of moderate-intensity exercise to improve metabolic fitness and presents novel findings that water-based circuit exercise is a suitable modality to effect metabolic improvements in overweight women with IGT. Not only is moderate-intensity exercise more likely to be acceptable to previously inactive sectors of the population,²² but the benefits may be similar to those obtained through more vigorous activity.²¹ Water-based exercise is a non-weight-bearing environment that permits individuals with limited mobility or those who prefer not to perform land-based exercise to obtain health benefits. Given that individuals will respond differently to exercise,⁴⁸ equivalent emphasis needs to be placed on improvements in metabolic measures (blood lipids, glucose tolerance, insulin sensitivity, blood pressure), in addition to the more traditional focus on weight loss and/or cardiorespiratory fitness.

Footnotes

Acknowledgment

The authors would like to thank the City of Dunedin for the use of the community aquatic center.

Disclosure Statement

No competing financial interests exist.

References

American Diabetes Association. Diagnosis and classification of diabetes mellitus. Diabetes Care, 2008; 31:S55–S60.

Baron

. Impaired glucose tolerance as a disease. Am J Cardiol, 2001; 88:16H–9H.

Abate

. Obesity and cardiovascular disease: pathogenetic role of the metabolic syndrome and therapeutic implications. J Diabetes Complications, 2000; 14:154–174.

Tripathy

, Almgren

, Tuomi

et al. Contribution of insulin-stimulated glucose uptake and basal hepatic insulin sensitivity to surrogate measures of insulin sensitivity. Diabetes Care, 2004; 27:2204–2210.

Hill

, Wyatt

. Role of physical activity in preventing and treating obesity. J Appl Physiol, 2005; 99:765–770.

Ivy

. Role of exercise training in the prevention and treatment of insulin resistance and non-insulin-dependent diabetes mellitus. Sports Med, 1997; 24:321–336.

Tuomilehto

, Lindstrom

, Eriksson

et al. Prevention of type 2 diabetes mellitus by changes in lifestyle among subjects with impaired glucose tolerance. N Engl J Med, 2001; 344:1343–1350.

Uusitupa

MIJ

, Lindi

, Louheranta

et al. Long-term improvement in insulin sensitivity by changing lifestyles of people with impaired glucose tolerance. Diabetes, 2003; 52:2532–2538.

Yates

, Khunti

, Bull

et al. The role of physical activity in the management of impaired glucose tolerance: a systematic review. Diabetologia, 2007; 50:1116–1126.

10.

Borodulin

, Tuomilehto

, Peltonen

et al. Association of leisure time physical activity and abdominal obesity with fasting serum insulin and 2-h postchallenge plasma glucose levels. Diabet Med, 2006; 23:1025–1028.

11.

Sigal

, Kenny

, Boule

et al. Effects of aerobic training, resistance training, or both on glycemic control in Type 2 diabetes: a randomized trial. Ann Intern Med, 2007; 147:357–369.

12.

Balducci

, Leonetti

, Di Mario

et al.

Is a long-term aerobic plus resistance training program feasible for and effective on metabolic profiles in type 2 diabetic patients?

Diabetes Care, 2004; 27:841–842.

13.

Eriksson

, Tuominen

, Valle

et al.

Aerobic endurance exercise or circuit-type resistance training for individuals with impaired glucose tolerance?

Horm Metab Res, 1998; 30:37–41.

14.

Maiorana

, O'Driscoll

, Goodman

et al. Combined aerobic and resistance exercise improves glycemic control and fitness in type 2 diabetes. Diabetes Res Clin Pract, 2002; 56:115–123.

15.

Takeshima

, Rogers

, Watanabe

et al. Water-based exercise improves health-related aspects of fitness in older women. Med Sci Sports Exerc, 2002; 34:544–551.

16.

Chu

, Eng

, Dawson

et al. Water-based exercise for cardiovascular fitness in people with chronic stroke: a randomised controlled trial. Arch Phys Med Rehabil, 2004; 85:870–874.

17.

Meyer

, Bucking

. Exercise in heart failure: should aqua therapy and swimming be allowed? Med Sci Sports Exerc, 2004; 36:2017–2023.

18.

Sideraviciuite

, Gailiuniene

, Visagurskiene

et al. The effect of long-term swimming program on glycemia control in 14–19-year aged healthy girls and girls with type 1 diabetes mellitus. Medicina (Kaunas), 2006; 42:513–518.

19.

Tanaka

, Bassett

JDR

, Howley

. Effects of swim training on body weight, carbohydrate metabolism, lipid and lipoprotein profile. Clin Physiol, 1997; 17:347–359.

20.

Healy

, Dunstan

, Salmon

et al. Objectively measured light-intensity physical activity is independently associated with 2-h plasma glucose. Diabetes Care, 2007; 30:1384–1389.

21.

O'Donovan

, Kearney

, Nevill

et al. The effects of 24 weeks of moderate- or high-intensity exercise on insulin resistance. Eur J Appl Physiol, 2005; 95:522–528.

22.

Lind

MSE

, Joens-Matre

, Ekkekakis

. What intensity of physical activity do previously sedentary middle-aged women select? Evidence of a coherent pattern from physio-logical, perceptual, and affective markers. Prev Med, 2005; 40:407–419.

23.

Taylor

, Buskirk

, Henschel

. Maximal oxygen uptake as an objective measure of cardio-respiratory performance. J Appl Physiol, 1955; 8:73–80.

24.

Day

, Rossiter

, Coats

et al. The maximally attainable VO₂ during exercise in humans: the peak vs. maximum issue. J Appl Physiol, 2003; 95:1901–1907.

25.

Wilmore

, Freund

, Joyner

et al. Acute responses to submaximal and maximal exercise consequent to beta-adrenergic blockade: implications for the prescription of exercise. Am J Cardiol, 1985; 55:135D–141D.

26.

Mercer

, Jensen

. Reliability and validity of a deep water running graded exercise test. Meas Phys Educ Exerc Sci, 1997; 1:213–222.

27.

Phillips

, Legge

, Jones

. Maximal physiological responses between aquatic and land exercise in overweight women. Med Sci Sports Exerc, 2008; 40:959–964.

28.

Borg

GAV.

Psychophysical basis of perceived exertion. Med Sci Sports Exerc, 1982; 14:377–381.

29.

Hawley

, Lessard

. Exercise training-induced improvements in insulin action. Acta Physiol, 2008; 192:127–135.

30.

Perreault

, Bergman

, Playdon

et al.

Impaired fasting glucose with or without impaired glucose tolerance: progressive or parallel states of prediabetes?

Am J Physiol, 2008; 295:E428–E435.

31.

Unwin

, Zimmet

, Alberti

. Impaired glucose tolerance and impaired fasting glycaemia: the current status on definition and intervention. Diabet Med, 2002; 19:708–723.

32.

Dela

, Ploug

, Handberg

et al. Physical training increases muscle GLUT4 protein and mRNA in patients with NIDDM. Diabetes, 1994; 43:862–865.

33.

Lillioja

, Young

, Cutler

et al. Skeletal muscle capillary density and fiber type are possible determinants of in vivo insulin resistance in man. J Clin Invest, 1987; 80:415–424.

34.

Utriainen

, Holmang

, Bjorntorp

et al. Physical fitness, muscle morphology, and insulin-stimulated limb blood flow in normal subjects. Am J Physiol, 1996; 270:E905–E911.

35.

Turcotte

, Fisher

. Skeletal muscle insulin resistance: roles of fatty acid metabolism and exercise. Phys Ther, 2008; 88:1279–1296.

36.

Simoneau

J-A

, Colberg

, Thaete

et al. Skeletal muscle glycolytic and oxidative enzyme capacities are determinants of insulin sensitivity and muscle composition in obese women. FASEB J, 1995; 9:273–278.

37.

Bruce

, Thrush

, Mertz

et al. Endurance training in obese humans improves glucose tolerance and mitochondrial fatty acid oxidation and alters muscle lipid content. Am J Physiol, 2006; 291:E99–E107.

38.

Simoneau

J-A

, Kelley

. Altered glycolytic and oxidative capacities of skeletal muscle contribute to insulin resistance in NIDDM. J Appl Physiol, 1997; 83:166–171.

39.

, Goodpaster

, Kelley

. Effects of weight loss and physical activity on muscle lipid content and droplet size. Obes Res, 2004; 12:761–769.

40.

Houmard

, Weidner

, Gavigan

et al. Fiber type and citrate synthase activity in the human gastrocnemius and vastus lateralis with aging. J Appl Physiol, 1998; 85:1337–1341.

41.

Poehlman

, Dvorak

, DeNino

et al. Effects of resistance training and endurance training on insulin sensitivity in nonobese, young women: a controlled randomized trial. J Clin Endocrinol Metab, 2000; 85:2463–2468.

42.

Poyhonen

, Sipila

, Keskinen

et al. Effects of aquatic resistance training on neuromuscular performance in healthy women. Med Sci Sports Exerci, 2002; 34:2103–2109.

43.

Ross

, Janssen

, Dawson

et al. Exercise-induced reduction in obesity and insulin resistance in women: a randomized controlled trial. Obes Res, 2004; 12:789–798.

44.

Janiszewski

, Janssen

, Ross

. Does waist circumference predict diabetes and cardiovascular disease beyond commonly evaluated cardiometabolic risk factors? Diabetes Care, 2007; 30:3105–3109.

45.

Lee

, Kuk

, Katzmarzyk

et al. Cardiorespiratory fitness attenuates metabolic risk independent of abdominal subcutaneous and visceral fat in men. Diabetes Care, 2005; 28:895–901.

46.

Ekelund

, Franks

, Sharp

et al. Increase in physical activity energy expenditure is associated with reduced metabolic risk independent of change in fatness and fitness. Diabetes Care, 2007; 30:2101–2106.

47.

Wasserman

, Hansen

, Sue

et al. Principles of exercise testing and interpretation, 4th. Philadelphia: Lippincott, Williams & Wilkins, 2005.

48.

Bouchard

, Rankinen

. Individual differences in response to regular physical activity. Med Sci Sports Exerc, 2001; 33:S446–S451.