Effects of Habitual T'ai Chi Exercise on Adiponectin,Glucose Homeostasis,Lipid Profile,and Atherosclerotic Burden in Individuals with Cardiovascular Risk Factors

Abstract

Objectives:

A single bout of t'ai chi (TC) exercise was previously found to be associated with a significant increase in post-exercise adiponectin levels in individuals with cardiovascular risk factors. The objective of this study was to examine the long-term effects of 24-month habitual TC exercise on adiponectin levels, glucose homeostasis, lipid profile, and atherosclerotic burden in individuals with cardiovascular risk factors.

Design:

This was a prospective observational study.

Settings/location:

The study was conducted at a regional hospital in south Taiwan.

Subjects:

Participants of a TC exercise program held by the clinics of cardiology and cardiovascular surgery for individuals with cardiovascular diseases were recruited to the TC group. Individuals who did not join the program were recruited as controls. All study subjects had at least one cardiovascular risk factor.

Interventions:

Ninety-minute session of Yang's style TC at least once a week.

Outcome measures:

Measurements on adiponectin, glucose homeostasis, lipid profile, and atherosclerotic burden were made at three time points—baseline, 12 months, and 24 months. Two-way repeated-measures general linear model was used to assess the changes over the study period between the TC and control groups.

Results:

Thirty-seven patients in both the TC and control groups completed the study. The TC group showed a greater increase in natural logarithmic transformed (Ln) adiponectin values than the control group over the study period (interaction effect p=0.009). Glucose homeostasis, lipid profile, risk of atherosclerosis, and atherosclerotic burden did not showed significant changes with TC compared with the controls over the 24-month period.

Conclusions:

The results of the present study indicate that for individuals with at least one cardiovascular risk factor, engaging in habitual TC exercise could lead to favorable changes in levels of adiponectin. The association between habitual TC and cardiovascular events and diabetic complications will require further investigations.

Introduction

T 'ai chi (TC) is a form of traditional Chinese martial art that is currently being practiced as a mind–body exercise for health and fitness.¹ Studies have shown that TC is associated with favorable blood lipid profile changes,² reductions in blood pressure,³ and improvements in aerobic endurance.⁴ Systematic reviews have also reported a beneficial association between TC practice and psychological well-being.⁵

In our previous study of 26 patients with at least one cardiovascular risk factor, we observed a significant increase in adiponectin levels in participants after a single bout of TC exercise.⁶ Adiponectin (also called ARCP30, AdipoQ, apM1, and GBP28) is an adipocyte secretory protein that primarily exerts strong anti-diabetic, anti-atherogenic and anti-inflammatory effects.^7,8 Levels of adiponectin are inversely correlated with insulin resistance,⁹ dyslipidemia,¹⁰ diabetes,¹¹ and cardiovascular disease.¹² Adiponectin level is also directly correlated with high-density lipoprotein level and inversely correlated with triglycerides, percent body fat, and serum leptin.¹³ Circulating adiponectin levels have been observed to be inversely associated with markers of endothelial dysfunction and systemic inflammation, such as tumor necrosis factor-α and C-reactive protein (CRP) in diabetic patients and individuals at risk for diabetes.¹⁴ Although TC appeared to have acute beneficial effect on levels of adiponectin and glucose homeostasis, to our knowledge, no studies have investigated whether the effects persist in individuals engaged in habitual TC exercise. Therefore, a follow-up study of 24 months was conducted to investigate the effect of habitual TC exercise on adiponectin and glucose homeostasis in individuals with at least one cardiovascular risk factor.

Methods

Individuals with at least one cardiovascular risk factor were recruited from the clinics of cardiology and cardiovascular surgery in a regional hospital in south Taiwan. A total of 37 individuals were recruited from an on-going TC exercise program held at the hospital for individuals with cardiovascular diseases. These individuals had been regularly practicing Yang's style TC for at least 3 months at the time of recruitment. Thirty-seven controls were recruited from either the family members of the individuals in the TC group or from the nonmedical staff working in the hospital. To minimize the potential confounding effects of age, controls were matched to cases for age (±5 years).

Cardiovascular risk factors were defined as cigarette smoking, hypertension (systolic blood pressure ≥140 mm Hg, diastolic blood pressure ≥90 mm Hg), dyslipidemia (serum total cholesterol ≥200 mg/dL, triglyceride ≥150 mg/dL, low density lipoprotein-cholesterol [LDL-C] ≥130 mg/dL, high density lipoprotein-cholesterol [HDL-C]<40 mg/dL in men and<50 mg/dL in women), diabetes mellitus (fasting glucose ≥126 mg/dL, 2 hours plasma glucose after a 75 g oral glucose load ≥200 mg/dL), obesity (body mass index [BMI]>30 kg/m²), and abdominal obesity (waist circumference>90 cm in men and>80 cm in women), family history of premature coronary artery disease (male first-degree relative<55 years old, female first-degree relative<65 years old), and age (men ≥45 years old, women ≥55 years old).^15
–17 All participants gave written informed consent, and the study was approved by the institutional review board of the study hospital.

The intervention of the TC group was a 90-minute session of Yang's style TC at least once a week throughout the 24-month duration of this study. The session was held at the study hospital under the guidance of professional Yang's style TC instructors. The participants were encouraged to practice TC at home at least twice a week. The Yang's style TC exercise consists of 108 forms that are made up of 37 different movements. Each TC session was preceded by a 25-minute warm-up including stretching, and a balance program followed by a 90-minute TC practice with light classic Chinese music playing in the background. Participants in both TC and control groups continued to receive their usual care, which included medication and dietary and exercise advice. There were no changes in the dosage of the medication in any participant during the study period.

Measurements of adiponectin, glucose homeostasis, lipid profile, high-sensitivity CRP (hs-CRP), and atherosclerotic burden were made at three time points—baseline, 12 months, and 24 months. Risk of atherosclerosis was assessed using hs-CRP,¹⁸ and atherosclerotic burden was assessed using carotid intima-media thickness (CIMT).¹⁹ Anthropometric measures and levels of physical activity were also collected at baseline.

Anthropometric measures

Body weight and height were determined with a physician's scale. BMI was calculated by dividing body weight in kilograms by height in meters squared. Circumferences of waists and hips were measured and waist-to-hip ratio was calculated. The characteristics of the study participants are shown in Table 1.

Table 1.

Baseline Characteristics of Participants in T'ai Chi and Control Groups

	Group (mean±standard deviation)
	T'ai chi (n=37)	Control (n=37)	p value
Age (years)	57.6±8.7	59.0±10.7	0.562
Males, n (%)	21 (56.8)	20 (54.1)	1.000
Body height (cm)	161.5±7.4	161.4±7.8	0.957
Body weight (kg)	64.3±10.0	68.7±13.3	0.113
Body mass index (kg/m²)	24.7±3.2	26.3±3.9	0.058
Waist circumference (cm)	83.5±8.8	88.1±8.3	0.024
Waist/hip circumference	0.9±0.1	0.9±0.1	0.150
Physical activity scores (MET/week)^a
Total	2561.8±2214.2	2463.1±2764.8	0.867
High intensity	302.7±1221.7	824.4±2109.7	0.203
Moderate intensity	1877.8±1605.5	781.1±1029.2	0.001
Walking	381.3±461.3	857.5±1040.0	0.015
Smoking, n (%)	2 (5.4)	1 (2.7)	1.000
Hypertension, n (%)	11 (29.7)	17 (45.9)	0.231
Type 2 diabetes, n (%)	9 (24.3)	10 (27)	1.000
Hyperlipidemia, n (%)	26 (70.3)	28 (75.7)	0.794
10-year CHD Framingham risk score (%)^b	5.8±5.0	8.4±8.3	0.106

MET, metabolic equivalent.

CHD, coronary heart disease.

Physical activity

Levels of physical activity of the participants at baseline were assessed using the International Physical Activity Questionnaire Taiwan Version by the Bureau of Health Promotion, Department of Health, Taiwan.²⁰ Responses were converted to metabolic equivalent (MET) per week separately for total, high intensity, and moderate intensity of physical activity.

Carotid intima media thickness

Left carotid artery was scanned in the long axes using a Philips iE33 xMATRIX ultrasound system fitted with a 7.5-MHz linear probe. The carotid bulb was identified and longitudinal two-dimensional ultrasonographic images of the common carotid artery 1 to 2 cm proximal to the carotid bulb were obtained. The optimal longitudinal image was acquired on the R-wave of the ECG and continuously recorded as digital files for 3 seconds. Measurements of CIMT of the posterior wall of the carotid artery were made from stored digital images with automatic edge detection using QLAB Advanced Quantification Software version 5.0 (Koninklijke Philips Electronics, Netherlands). The three maximum measures from the left common carotid artery in three different frames were averaged. Reading and analysis of images were done at the end of the study by a single physician who was blinded to the identity of the time point and study group. The intraobserver coefficient of variance was 2.1%.

Laboratory data

Serum total adiponectin concentrations were measured by radioimmunosorbent assay (LINCO Research, St. Charles, MO). The intra-assay variability was 3.59%. The adiponectin radioimmunosorbent assay utilizes ¹²⁵I-labeled murine adiponectin and a multispecies adiponectin rabbit antiserum to determine the level of adiponectin in serum by the double antibody/polyethyleneglycol technique.

Serum glucose concentrations were analyzed based on the hexokinase/glucose-6-phosphate dehydrogenase method (Shino-Test Corporation Ltd., Co., Tokyo, Japan). The intra-assay variability of serum glucose concentrations was 1.6%. Serum levels of insulin were analyzed by a two-site sandwich immunoassay using direct chemiluminescence technology (Kyowa Medex Co., Ltd for Bayer HealthCare LLC, Tokyo, Japan) with an intra-assay variability of 3.2%–4.6%. Insulin sensitivity was assessed utilizing the homeostasis model assessment (HOMA) method. The HOMA-estimated insulin resistance (HOMA-IR) index was calculated using the following formula: fasting plasma glucose (mg/dL)×fasting plasma insulin (μU/mL)/405.²¹ Hemoglobin A_1c (HbA_1C) was analyzed by an automated high-performance liquid chromatography instrument (Tosoh HLC-723G8, Bioscience Division, Tokyo, Japan). The intra-assay variability of HbA₁C ratio was 0.2%–0.6%.

Cholesterol was analyzed by cholesterol oxidase-N-(3-sulfopropyl)-3-methoxy-5-methylaniline (HMMPS) method (Wako Pure Chemical Industries, Ltd., Osaka, Japan). The intra-assay variability of cholesterol was <5%, measurable range 1–800 mg/dL. Triglyceride was analyzed by GPO-HMMPS, glycerol blanking method (Wako Pure Chemical Industries, Ltd.). The intra-assay variability of triglyceride was <5%, with a measurable range of 1.5–2000 mg/dL. HDL-C was analyzed by chemically modified enzyme method (Kyowa Medex Co., Ltd., Tokyo, Japan). The intra-assay variability of HDL-C was <5%, with a measurable range of up to 100 mg/dL. LDL-C was analyzed by selective solubilization method (Kyowa Medex Co., Ltd., Tokyo, Japan). The intra-assay variability of LDL-C was <5%, with a measurable range of 1–700 mg/dL.

The hs-CRP levels were analyzed by Latex immunoassay (Wako Pure Chemical Industries, Ltd.). The intra-assay variability of hs-CRP was <10%, with a measurable range of 0.01–35 mg/dL.

Statistical analyses

Data were coded and entered into SPSS for Windows version 12.0 (SPSS, Chicago, IL) for statistical analysis. Participants in TC and control groups at baseline were compared using unpaired t-test for continuous variables and Fisher exact test for categorical parameters. Adiponectin levels did not show Gaussian distribution and therefore, Ln values of adiponectin were also calculated. Changes in anthropometric measurements, intima media thickness of common carotid artery, and laboratory data between the TC and control groups over the three time points were assessed using two-way repeated-measures general linear model. Baseline measurements of BMI and waist circumference were included in the model as covariates to adjust for their differences between the TC and control groups. The group–time interaction effect between baseline, 12 months, and 24 months was the main statistical outcome of interest. A p<0.05 was considered statistically significant.

Results

A total of 74 individuals (TC=37, control=37) were recruited in the present study, and 72 (97%) of them completed the 24-month follow-up. One individual died of sudden death while golfing, and one individual withdrew from the study due to personal reasons. Both of them belonged to the control group. Participants in the TC group had practiced Yang's TC for 4.8±5.2 years (mean±SD) with a self-reported mean practice frequency of 3.8±2.6 times per week.

Baseline characteristics of the participants of the study are shown in Table 1. Age, sex, body weight, body height, BMI, and waist/hip circumference were not significantly different between the TC and control groups at baseline. Waist circumference was significantly higher in control group (p=0.024). Cardiovascular risk factors including hypertension, diabetes mellitus, hyperlipidemia, smoking, and 10-year CHD Framingham risk score were not significantly different between the two groups. Physical activity scores showed that total and high intensity physical activity scores were not significantly different between the two groups but moderate intensity physical activity (p=0.001) was significantly higher in TC group, whereas walking (p=0.015) was significantly higher in control group.

Adiponectin, glucose homeostasis, lipid profile, hs-CRP, and CIMT

A significant interaction between group and time obtained from the repeated measures general linear model indicated that the changes over time were significantly different between TC and control groups. Of the measured variables, only Ln adiponectin exhibited a significant interaction between group and time (p=0.009). The increase in Ln adiponectin over time was significantly higher in TC group compared to the control group. No significant interaction between group and time in untransformed adiponectin was observed.

Over the 24-month study period, HOMA-IR (p=0.004) and insulin (p=0.001) increased significantly, but the increase was not different between the two groups. Cholesterol (p=0.023), LDL-C (p=0.013), and HbA_1C (p=0.002) decreased significantly over time, but the decrease was not different between the two groups. Regarding the comparisons between the two groups, triglyceride (p=0.030) and CIMT (p=0.027) were significantly higher in control group compared to the TC group, but there was no difference in the changes over time between the two groups (Table 2).

Table 2.

Adiponectin, Glucose Homeostasis, Lipid Profiles, hs-CRP, and CIMT Results in T'ai Chi and Control Groups at Baseline, 12 Months, and 24 Months

				Time (month)		p value^a
	Group	n	Baseline	12	24	Group	Time	Interaction
Adiponectin (μg/ml)	T'ai chi	32	8.11±5.03	9.84±5.18	10.56±5.68	0.030	<0.001	0.125
	Control	33	4.68±2.66	7.36±4.80	6.80±4.93
Ln adiponectin	T'ai chi	32	1.90±0.64	2.14±0.58	2.21±0.57	0.018	<0.001	0.009
	Control	33	1.40±0.54	1.83±0.58	1.73±0.61
HOMA-IR	T'ai chi	37	1.80±1.56	2.46±3.34	2.28±2.85	0.812	0.004	0.832
	Control	35	2.43±1.77	3.08±2.54	3.03±2.55
Glucose (mg/dL)	T'ai chi	37	110.97±34.61	110.81±27.64	109.73±29.07	0.692	0.232	0.377
	Control	34	112.97±29.59	118.65±31.78	114.06±25.44
Insulin (μU/ml)	T'ai chi	37	6.23±4.26	7.97±8.06	7.65±7.01	0.348	0.001	0.612
	Control	34	8.61±5.32	10.30±7.14	10.79±8.59
HbA_1C (%)	Tai Ch	37	6.44±1.27	6.11±0.87	6.28±1.01	0.712	0.002	0.200
	Control	35	6.68±1.17	6.50±1.08	6.56±1.19
Cholesterol (mg/dL)	T'ai chi	37	206.16±45.60	194.59±37.93	198.32±44.37	0.181	0.023	0.199
	Control	35	208.89±35.86	206.26±33.65	196.03±33.47
Triglyceride (mg/dL)	T'ai chi	37	108.03±60.07	102.19±60.17	118.27±103.38	0.030	0.151	0.280
	Control	35	165.29±79.54	148.11±79.60	145.37±82.09
HDL-C (mg/dL)	T'ai chi	37	66.70±16.30	63.27±12.66	61.57±13.17	0.322	0.079	0.232
	Control	35	57.14±20.15	56.94±19.87	56.86±19.17
LDL-C (mg/dL)	T'ai chi	37	124.38±37.70	115.97±31.03	115.38±34.17	0.169	0.013	0.141
	Control	35	126.37±33.04	128.06±29.82	115.26±26.27
hs-CRP (mg/dL)	T'ai chi	36	0.140±0.141	0.140±0.246	0.130±0.153	0.833	0.531	0.225
	Control	35	0.111±0.063	0.114±0.123	0.214±0.459
CIMT (mm)	T'ai chi	37	0.64±0.13	0.63±0.11	0.62±0.10	0.027	0.936	0.299
	Control	35	0.70±0.15	0.71±0.19	0.71±0.18

HOMA-IR, homeostasis model assessment-estimated insulin resistance; HDL-C, high density lipoprotein-cholesterol; LDL-C, low density lipoprotein-cholesterol; HbA_1C, hemoglobin A_1c; hs-CRP, high-sensitivity C-reactive protein; CIMT, carotid intima-media thickness.

p-values were calculated using two-way repeated-measures general linear model adjusted for baseline measurements of body mass index and waist circumference.

Discussion

To our knowledge, this is the first study to explore the long-term effects of habitual TC exercise on adiponectin responses in individuals with cardiovascular risk factors. Our results showed that 24 months of habitual TC exercise led to a significantly greater increase of Ln adiponectin compared to the control group. In a recent meta-analysis of 17 case–control studies in Chinese populations, low serum total adiponectin levels were associated with increased risk of a first cardiovascular event.²² In another meta-analysis of 13 prospective studies on diverse populations, higher adiponectin levels were found to be monotonically associated with a lower risk of type 2 diabetes.²³ Therefore, habitual TC exercise may confer beneficial effects on reducing the risk of both cardiovascular events and type 2 diabetes in individuals with cardiovascular risk factors.

Since this is the first study exploring the effect of habitual TC on adiponectin response, no previous data are available for comparison. Studies on the long-term effects of exercise on adiponectin levels have shown discrepant results. A 9-month randomized controlled trial of a home-based exercise intervention in middle-aged Taiwanese adults with at least one diabetic risk factor found no significant improvement in adiponectin levels.²⁴ However, a 6-month study on untrained middle-aged individuals reported that progressive aerobic training was associated with significant increase in adiponectin.²⁵

Our study found that long-term habitual TC exercise was not associated with changes in levels of HOMA-IR, insulin, glucose, and HbA_1C. This is in contrast to the findings from the acute effects of TC in which a single bout of TC was significantly associated with a decrease in HOMA-IR and glucose.⁶ Our findings are in agreement with two randomized controlled trials showing TC did not improve glucose homeostasis or insulin sensitivity in patients with type 2 diabetes mellitus.^26,27 Regarding HbA_1C, a single group 12-week pre–post study, HbA_1C levels were reported to significantly decrease with TC exercise in type 2 diabetic patients.²⁸ Another 6-month TC exercise program compared individuals with type 2 diabetes who had completed 80% of the TC sessions with those who had not. The TC adherent group showed a significant greater decline in HbA_1C than the nonadherent group.²⁹ However, a community-based randomized controlled trial of 53 adults with type 2 diabetes and a baseline HbA_1C of 7% or more reported that a 6-month TC program was unable to improve HbA_1C. The authors of the study concluded that longer duration or increased number of TC sessions per week might be required to demonstrate changes in metabolic or cardiovascular parameters.²⁷

In addition, findings from our study indicated that long-term habitual TC exercise was not associated with changes in lipid profiles. Although blood lipid profiles can be altered at low exercise training volumes, the threshold of between 1200 to 2200 kcal of energy expenditure per week has to be met before favorable effects on HDL-C and triglyceride levels can be observed. Levels of total cholesterol and LDL-C cannot be altered by exercise training unless dietary fat intake is reduced along with body weight loss.³⁰ In a study of 70 patients with dyslipidemia in which participants were self-selected into either a 12-month TC program group or a control group, significant improvements in total cholesterol, triglyceride, and LDL-C were observed in the TC group compared to those in control group. However, the mean body weight was also significantly lower in the TC group compared to that of the control group. The reduction of body weight might partially explain the beneficial effect of TC exercise on blood lipid profiles.³¹ In our study, the lack of significant changes of HDL-C and triglyceride levels between TC and control group over time might be due to the similar total physical activity levels between the two groups. Although participants in the control were not engaged in the TC program, they reported a significantly higher levels of walking compared with the TC group.

Findings from our study indicated that long-term habitual TC exercise was not associated with indicators of atherosclerotic burden. Studies examining the effect of exercise training on hs-CRP have produced conflicting results. In a study comparing different exercise modalities in 82 patients with type 2 diabetes and metabolic syndrome, a significant reduction of hs-CRP was associated with long-term aerobic and resistance physical activity.³² However, in another study of 421 sedentary, overweight, and obese postmenopausal women with elevated blood pressure, 6 months of aerobic exercise training did not improve hs-CRP.³³ A meta-analysis of five randomized controlled trials also reported that aerobic interventions failed to reduce hs-CRP levels in adults.³⁴ The differences in findings may partly be related to the type of exercise training employed in the studies. Resistance exercise training appeared to reduce hs-CRP more effectively than aerobic exercise training.³⁵ Regarding the findings from the noninvasive CIMT measurement for evaluation of subclinical atherosclerosis, we also observed no significant differences between TC and control group over the study period. Therefore, findings in hs-CRP and CIMT from the current study add to the literature that long-term moderate-intensity TC exercise does not appear to change indicators of atherosclerotic burden. One explanation for the lack of significant results is that the CIMT of the participants were within normal ranges at baseline. Additional studies will be required to investigate whether TC exercise can reduce progression of CIMT in individuals with higher initial CIMT levels.

Several limitations of this study should be noted. First, the participants were not assigned to either TC or control group by randomization and therefore, differences in characteristics of participants such as dietary habits may account for the observed results, including the lack of differences between the two groups. Findings from our observational studies should be interpreted with caution and need to be confirmed in randomized controlled trials. Second, it can be observed that the mean total energy expenditure of the control group was similar to that of the TC group. Participants in the control group were also engaged in exercise other than TC rather than living a sedentary lifestyle. This potentially can dilute any effect of TC exercise on the observed parameters. Therefore, the observed effects of TC exercise should be interpreted as its effect in addition to other normal daily physical activity rather than comparing to no exercise. Finally, different oligomers of adiponectin possess distinct biological activities, but they were not separately measured in our study because commercially available radioimmunoassay, at present, is unable to distinguish between them. Therefore, we cannot conclude which oligomers of adiponectin have led to the observed changes in total adiponectin levels.

Conclusions

The findings of this study show increased levels of adiponectin in individuals with at least one cardiovascular risk factor who practiced TC over a 24-month period, compared to age-matched controls with similar risk factors who did not practice TC. The lack of associations between other measured parameters suggested that habitual TC may have a lesser role to play in affecting glucose homeostasis, blood lipid profiles, and atherosclerotic burden.

Footnotes

Acknowledgments

This study was partially supported by a grant from the Department of Medical Research of Chia-Yi Christian hospital, Taiwan. We also acknowledge Mr. Fu-Chu Lin for his voluntary contribution as the t'ai chi instructor during the study.

Disclosure Statement

No competing financial interests exist.

References

Jahnke

, Larkey

, Rogers

et al. A comprehensive review of health benefits of qigong and tai chi. Am J Health Promot, 2010; 24:e1–e25.

Tsai

, Wang

, Chan

et al. The beneficial effects of tai chi chuan on blood pressure and lipid profile and anxiety status in a randomized controlled trial. J Altern Complement Med, 2003; 9:747–754.

Yeh

, Wang

, Wayne

, Phillips

. The effect of tai chi exercise on blood pressure: A systematic review. Prev Cardiol, 2008; 11:82–89.

Taylor-Piliae

, Haskell

, Froelicher

. Hemodynamic responses to a community-based Tai Chi exercise intervention in ethnic Chinese adults with cardiovascular disease risk factors. Eur J Cardiovasc Nurs, 2006; 5:165–174.

Wang

, Bannuru

, Ramel

et al. Tai Chi on psychological well-being: Systematic review and meta-analysis. BMC Complement Altern Med, 2010; 10:23.

Chang

, Koo

, Ho

et al. Effects of Tai Chi on adiponectin and glucose homeostasis in individuals with cardiovascular risk factors. Eur J Appl Physiol, 2011; 111:57–66.

Kato

, Kashiwagi

, Shiraga

et al. Adiponectin acts as an endogenous antithrombotic factor. Arterioscler Thromb Vasc Biol, 2006; 26:224–230.

Choi

, Kim

, Yang

et al. Association of adiponectin, resistin, and vascular inflammation: Analysis with 18F-fluorodeoxyglucose positron emission tomography. Arterioscler Thromb Vasc Biol, 2011; 31:944–949.

Nilsson

, Engstrom

, Hedblad

et al. Plasma adiponectin levels in relation to carotid intima media thickness and markers of insulin resistance. Arterioscler Thromb Vasc Biol, 2006; 26:2758–2762.

10.

Matsubara

, Maruoko

, Katayose

. Decreased plasma adiponectin concentrations in women with dyslipidemia. J Clin Endocrinol Metab, 2002; 87:2764–2769.

11.

Hotta

, Funahashi

, Arita

et al. Plasma concentrations of a novel adipose-specific protein, adiponectin, in type 2 diabetic patients. Arterioscler Thromb Vasc Biol, 2000; 20:1595–1599.

12.

Rothenbacher

, Brenner

, Marz

, Koenig

. Adiponectin, risk of coronary heart disease and correlations with cardiovascular risk markers. Eur Heart J, 2005; 26:1640–1646.

13.

Kazumi

, Kawaguchi

, Hirano

, Yoshino

. Serum adiponectin is associated with high-density lipoprotein cholesterol, triglycerides, and low-density lipoprotein particle size in young healthy men. Metabolism, 2004; 53:589–593.

14.

Shetty

, Economides

, Horton

et al. Circulating adiponectin and resistin levels in relation to metabolic factors, inflammatory markers, and vascular reactivity in diabetic patients and subjects at risk for diabetes. Diabetes Care, 2004; 27:2450–2457.

15.

De Backer

, Ambrosioni

, Borch-Johnsen

et al.

European guidelines on cardiovascular disease prevention in clinical practice: Third joint task force of European and other societies on cardiovascular disease prevention in clinical practice (constituted by representatives of eight societies and by invited experts)

Eur J Cardiovasc Prev Rehabil, 2003; 10:S1–S10.

16.

Grundy

, Pasternak

, Greenland

et al. Assessment of cardiovascular risk by use of multiple-risk-factor assessment equations: A statement for healthcare professionals from the American Heart Association and the American College of Cardiology. Circulation, 1999; 100:1481–1492.

17.

Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults. Executive Summary of The Third Report of The National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, And Treatment of High Blood Cholesterol In Adults (Adult Treatment Panel III)

JAMA, 2001; 285:2486–2497.

18.

Buckley

, Fu

, Freeman

et al. C-reactive protein as a risk factor for coronary heart disease: A systematic review and meta-analyses for the U.S. Preventive Services Task Force. Ann Intern Med, 2009; 151:483–495.

19.

Lorenz

, von Kegler

, Steinmetz

et al.

Carotid intima-media thickening indicates a higher vascular risk across a wide age range: Prospective data from the Carotid Atherosclerosis Progression Study (CAPS)

Stroke, 2006; 37:87–92.

20.

Liou

, Jwo

, Yao

et al. Selection of appropriate Chinese terms to represent intensity and types of physical activity terms for use in the Taiwan version of IPAQ. J Nurs Res, 2008; 16:252–263.

21.

Matthews

, Hosker

, Rudenski

et al. Homeostasis model assessment: Insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia, 1985; 28:412–419.

22.

Zhang

, Liu

, Che

, Xu

. Serum total adiponectin level and risk of cardiovascular disease in Han Chinese populations: A meta-analysis of 17 case-control studies. Clin Endocrinol, 2011, 10.1111/j.1365-2265.2011.04260.x[Epub ahead of print]

23.

, Shin

, Ding

, van Dam

. Adiponectin levels and risk of type 2 diabetes: A systematic review and meta-analysis. JAMA, 2009; 302:179–188.

24.

, Hwang

, Chen

, Chuang

. Home-based exercise for middle-aged Chinese at diabetic risk: A randomized controlled trial. Prev Med, 2011; 52:337–343.

25.

Mujumdar

, Duerksen-Hughes

, Firek

, Hessinger

. Long-term, progressive, aerobic training increases adiponectin in middle-aged, overweight, untrained males and females. Scand J Clin Lab Invest, 2011; 71:101–107.

26.

Tsang

, Orr

, Lam

et al. Effects of tai chi on glucose homeostasis and insulin sensitivity in older adults with type 2 diabetes: A randomised double-blind sham-exercise-controlled trial. Age Ageing, 2008; 37:64–71.

27.

Lam

, Dennis

, Diamond

, Zwar

. Improving glycaemic and BP control in type 2 diabetes. The effectiveness of tai chi. Aust Fam Physician, 2008; 37:884–887.

28.

Yeh

, Chuang

, Lin

et al. Tai chi chuan exercise decreases A1C levels along with increase of regulatory T-cells and decrease of cytotoxic T-cell population in type 2 diabetic patients. Diabetes Care, 2007; 30:716–718.

29.

Song

, Ahn

, Roberts

et al. Adhering to a t'ai chi program to improve glucose control and quality of life for individuals with type 2 diabetes. J Altern Complement Med, 2009; 15:627–632.

30.

Durstine

, Grandjean

, Cox

, Thompson

. Lipids, lipoproteins, and exercise. J Cardiopulm Rehabil, 2002; 22:385–398.

31.

Lan

, Su

, Chen

, Lai

. Effect of T'ai chi chuan training on cardiovascular risk factors in dyslipidemic patients. J Altern Complement Med, 2008; 14:813–819.

32.

Balducci

, Zanuso

, Nicolucci

et al. Anti-inflammatory effect of exercise training in subjects with type 2 diabetes and the metabolic syndrome is dependent on exercise modalities and independent of weight loss. Nutr Metab Cardiovasc Dis, 2010; 20:608–617.

33.

Stewart

, Earnest

, Blair

, Church

. Effects of different doses of physical activity on C-reactive protein among women. Med Sci Sports Exerc, 2010; 42:701–707.

34.

Kelley

, Kelley

. Effects of aerobic exercise on C-reactive protein, body composition, and maximum oxygen consumption in adults: A meta-analysis of randomized controlled trials. Metabolism, 2006; 55:1500–1507.

35.

Donges

, Duffield

, Drinkwater

. Effects of resistance or aerobic exercise training on interleukin-6, c-reactive protein, and body composition. Med Sci Sports Exerc, 2010; 42:304–313.