UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Effective exercise modality for postmenopausal women with type 2 diabetes Cuff, Darcye J. 2004

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_2004-931129.pdf [ 8.48MB ]
JSON: 831-1.0077411.json
JSON-LD: 831-1.0077411-ld.json
RDF/XML (Pretty): 831-1.0077411-rdf.xml
RDF/JSON: 831-1.0077411-rdf.json
Turtle: 831-1.0077411-turtle.txt
N-Triples: 831-1.0077411-rdf-ntriples.txt
Original Record: 831-1.0077411-source.json
Full Text

Full Text

E F F E C T I V E E X E R C I S E M O D A L I T Y F O R P O S T M E N O P A U S A L W O M E N W I T H T Y P E 2 D I A B E T E S by Darcye J. Cuff B.Sc. (Hon. Kinesiology), Simon Fraser University, 1984 M.Sc. (Nutritional Sciences), University of Toronto, 1988 A THESIS S U B M I T T E D IN P A R T I A L F U L F I L L M E N T O F T H E R E Q U I R E M E N T S F O R T H E D E G R E E O F D O C T O R O F P H I L O S O P H Y in T H E F A C U L T Y O F G R A D U A T E STUDIES ( S C H O O L O F H U M A N KINETICS ) We accept this thesis as conforming to the required standard. T H E U N I V E R S I T Y O F BRITISH C O L U M B I A 2004 © Darcye Jean Cuff, 2004 Abstract The purpose of this study was to evaluate changes in insulin sensitivity assessed by hyperinsulinemic-euglycemic clamp after 16 weeks of aerobic plus resistance training (Ae + RT), aerobic only training (Ae only), or usual care in a group of 28 randomly assigned postmenopausal women with type 2 diabetes (T2D). A second purpose was to relate improvements in insulin sensitivity to changes in abdominal subcutaneous and visceral AT (AT) cross-sectional area and thigh muscle cross-sectional area and density as assessed by single-image computed tomography (CT) scans. Glucose disposal during the hyperinsulinemic-euglycemic clamp was significantly increased by 77% after Ae + RT training (pre 2.36 ± 0.33, post 4.17 + 0.65 mg/kg/min, p=0.007), with no significant change seen in either the usual care group (pre 2.29 ± 0.46, post 2.36 + 0.46 mg/kg/min) or the Ae only group (pre 2.78 + 0.49, post 3.33 ± 0.64 mg/kg/min). The Ae + RT group demonstrated significant losses of 10.5% from the visceral AT area (pre 251.1 ± 22.9 cm2, post 224.8 ± 20.0 cm2, p<0.006), significant increases of 5.9% in muscle cross-sectional area (pre 207.7± 10.0 cm2, post 213.6 ± 9.6 cm2, p=0.017) and significant increases in average muscle density (pre 43.9 ± 1.7, post 46.4 + 1.2 HU, p<0.015) and the proportion of normal density muscle (pre 156.6 ± 12.8, post 167.1 ± 11.4 cm2, p<0.001). The usual care group had a significant decrease in average muscle density (pre 47.1 + 1.1, post 46.0 ± 0.9 HU, p<0.004) and a significant increase in the proportion of low density muscle (pre 50.6 ± 4.2, post 53.8 ± 3.9 cm2, p<0.015). Across all subjects, improved glucose disposal during the euglycemic clamp was significantly negatively associated with change in total abdominal AT area (r=-0.62, p<0.001), abdominal subcutaneous ii AT area (r=-0.64, p<0.001), visceral AT area (r=-0.32, p<0.047), and positively associated with muscle cross-sectional area (r=0.43, p<0.011), average muscle density (r=0.50 p=0.003) and normal density muscle (r=0.52, p<0.002). Significant associations remained between change in glucose disposal and change in muscle cross-sectional area (r=0.43, p<0.01), and area of normal density muscle (r=0.42, p<0.02) after controlling total abdominal AT area. These findings suggest that aerobic plus resistance training is an effective exercise strategy to improve insulin sensitivity in postmenopausal women with type 2 diabetes. The improved insulin sensitivity is related to loss of abdominal subcutaneous and visceral AT and to increased muscle cross-sectional area and density. i i i Table of Contents Abstract iii Table of Contents iv List of Tables vii List of Figures viii Acknowledgments ix Chapter 1: Introduction 1 1.1 Health Implications of Obesity 1 1.2 Possible Mechanisms Linking Obesity with Health Outcomes 3 1.3 Physical Activity Prevents Diabetes and Improves Insulin Sensitivity 5 1.4 Possible Mechanisms Linking Physical Activity with Improved Insulin Sensitivity 6 1.5 Research Priorities 6 Chapter 2: Review of the Literature 8 2.1 Regional Adiposity 8 2.1.1 Regional AT Imaging 10 2.1.2 Regional Adipose Tissue Patterns Relative to Gender, Ethnicity and Age 15 2.1.2. Classification for Regional AT Compartments by Imaging Methods 17 2.2 Insulin Resistance 18 2.2.1 Measurement of Insulin Resistance 20 2.2.2 Prevalence of Insulin Resistance 24 2.2.3 Sequela of Insulin Resistance 25 2.2.4 Site of Insulin Resistance 26 2.3 Abdominal Obesity and the Pathogenesis of Insulin Resistance 27 2.3.1 Abdominal Obesity or Insulin Resistance as the Initiating Factor 31 2.4 Possible Mechanisms Linking Abdominal Obesity with Insulin Resistance 32 2.4.1 Dysregulation of Fatty Acid Metabolism - Elevated Free Fatty Acids 32 2.4.2 Ectopic Lipid Accumulation 34 2.4.3 Adipose Tissue as an Endocrine Organ 37 2.4.4 Neurohormonal or Hormonal Regulation of Lipid Metabolism 39 2.5 Physical Activity and Insulin Resistance 40 2.5.1 Physical Activity and the Prevention of Type 2 Diabetes 40 2.5.2 Exercise and Insulin Resistance 43 2.5.3 Mechanisms by Which Exercise Training Improves Insulin Sensitivity 47 2.6 Weight Loss and Exercise Interventions: Relationship of Insulin Sensitivity to Body Composition 49 iv 2.6.1 Change in Abdominal AT 54 2.6.2 Change in Muscle Characteristics 56 2.7 Summary 57 2.8 General Hypotheses 58 Chapter 3: Methods 60 3.1 Study Design 60 3.1.1 Subjects and Sample Size 60 3.1.2 Randomization and Intervention 62 3.2 Measurements 64 3.2.1 Dietary Assessment 64 3.2.2 Peak V0 2 Determination 64 3.2.3 Hyperinsulinemic-Euglycemic Clamp Studies 65 3.2.4 Computed Tomography Imaging 67 3.2.5 Blood Analyses 68 3.3 Statistical Analyses 68 3.4 Specific Hypotheses 70 Chapter 4: Results 72 4.1 Subjects 72 4.2 Body Weight 74 4.3 Dietary Assessment 75 4.4 Fitness and Resistance Training Measures 75 4.5 Fasting Insulin, Glucose and Glycosylated Hemoglobin 76 4.6 Hyperinsulinemic-Euglycemic Clamp Studies 77 4.7 Computed Tomography Scans - Abdominal AT Areas 79 4.8 Computed Tomography Scans - Muscle Cross-sectional Area and Density 82 4.9 Blood Lipids 84 4.10 Correlation of Glucose Disposal with Adipose Tissue and Muscle Characteristics 85 Chapter 5: Discussion 86 5.1 Exercise Training Modality and Insulin Sensitivity 88 5.1.1 Improvements in Insulin Sensitivity after Aerobic plus Resistance Training 88 5.1.2 Lack of Improvement in Insulin Sensitivity after Aerobic Only Training 90 5.2 Abdominal Adiposity Changes 93 5.2.1 Change in Total Abdominal Adipose Tissue 93 5.2.2 Change in Abdominal Subcutaneous and Visceral AT 95 5.3 Change in Muscle Cross-Sectional Area and Muscle Density 100 5.3.1 Change in Muscle Cross-sectional Area 100 v 5.3.2 Change in Muscle Density 101 5.4 Relationships of Change in Body Composition with Change in Insulin Sensitivity 106 5.5 Fasting Insulin, Glucose and Glycosylated Hemoglobin Levels 109 5.6 Lipoprotein Levels 110 5.7 Study Strengths and Limitations 112 Chapter 6: Summary, Conclusions and Recommendations 115 6.1 Summary 115 6.2 Conclusions 117 6.3 Recommendations for Future Research 118 References 120 Appendix A: Recruitment Letter for Mail-Out 136 Appendix B: Informed Consent 137 Appendix C: Initial Subject Characteristics 140 Appendix C: Individual Subject Attendance to Exercise Classes 141 Appendix C: Change in Resistance Training Loads 142 Appendix D: Carbohydrate Metabolism Measures 143 Appendix E: Sample of One Analyzed CT Scan 145 Appendix F: Abdominal AT Cross-sectional Areas 147 Appendix G: Thigh Muscle Characteristics 148 Appendix H: Blood Lipid Results 150 vi List of Tables Page Table 2.1 Imaging technologies in quantifying adipose tissue 12 Table 2.2. Sizes of regional adipose tissue depots 16 Table 2.3 NCEP and WHO definitions of the Metabolic Syndrome 19 Table 2.4 Methods of measuring insulin resistance 23 Table 2.5 Cardiovascular disease outcomes in insulin-resistant 26 conditions Table 2.6 Studies evaluating body composition changes related to 51 changes in Insulin sensitivity Table 3.1 Inclusion and exclusion criteria 62 Table 4.1: Initial Subject Characteristics 72 Table 4.2 Number of subjects using non-glycemic medical therapy 73 Table 4.3 Subject medical changes and potential effect on outcomes 74 Table 4.4 Initial and Change Data for Body Weight and Fitness 75 Table 4.5 Pre and Post glycosylated hemoglobin, fasting insulin and 76 fasting glucose Table 4.6 Change in Glucose Infusion Rates During Euglycemic Clamp 77 Table 4.7 Initial And Change Data of Abdominal AT Cross-sectional 80 Areas Table 4.8 Initial and Change Data of Mid-thigh Muscle Characteristics 83 Table 4.9 Initial and Change Data for Blood Lipids 84 Table 4.10 Pearson Correlation Coefficients between glucose infusion 85 rates, abdominal AT and muscle characteristics vii List of Figures Page Figure 3.1 Flowchart of patient recruitment 61 Figure 3.2 Exercise Group Class Structure 64 Figure 3.3 Sample Hyperinsulinemic-Euglycemic Study 66 Figure 4.1 Steady State Glucose Infusion Rates - Group Means 78 Figure 4.2 Steady State Glucose Infusion Rates - Individual Subject 78 Data Figure 4.3 Change in Adipose Tissue Cross-Sectional Areas 81 Figure 4.4 Change in Adipose Tissue Areas - Combining Exercise 81 Groups Figure 4.5 Change in Muscle Cross-sectional Area and Density 84 viii Acknowledgments I am indebted to the participants who donated blood, sweat and laughter; to Ellen Bjornson for her exceptional exercise leadership and to Chris Lockhart and Rosemary Torresani for their expertise and care during the euglycemic clamp studies. I was also fortunate to work with the radiology technologists at St. Paul's Hospital who were extremely professional and helpful in conducting the computed tomography scans. This research was supported by a grant from the Heart and Stroke Foundation of British Columbia and Yukon, under the investigative committee of Dr. Jiri Frohlich, Dr. Alan Martin and Dr. Andrew Ignaszewski, with the collaboration of Dr. Graydon Meneilly and Dr. Hugh Tildesley. The input of Dr. Diane Finegood during the grant application process was especially helpful. The preparation of this doctoral thesis was supervised and guided by Dr. Alan Martin, Dr. Ted Rhodes and Dr. Andrew Ignaszewski. ix Chapter 1: Introduction 1.1 Health Implications of Obesity Obesity has been recognized as the second leading cause of preventable death, after smoking1, and is regarded as today's principal neglected public health problem2. Twenty-nine percent of Canadian adults (age 20-64 years) are classified as overweight (body mass index (BMI) > 27 kg/m2), and 12% are obese (BMI > 30 kg/m2), based on the World Health Organization (WHO) criteria3. The prevalence of overweight and obesity increases with age, reaching 37% for overweight, and 16% for obesity in the older group (45-64 years). These numbers represent increases of approximately 10% over the time period of 1985-19964. This ever-increasing prevalence of overweight and obesity continues as a consequence of societal, economic and technological advances which have contributed to unhealthy diets and sedentary behaviour. Overweight and obesity occur when caloric intake exceeds expenditure, and it is believed that inherited tendencies to subtle disorders of weight-regulating mechanisms are magnified by poor diet and lack of physical activity5. Sixty to seventy percent of the variance in BMI can be attributed to environmental conditions such as high fat, energy dense diets and sedentary lifestyles6. Obesity increases the risk for many disorders that are associated with high mortality and morbidity, including diabetes, hypertension, coronary heart disease, dyslipidemia and some cancers7. Type 2 diabetes (T2D) is tightly associated with obesity, and prevalence rates of diabetes are high. In 2000, approximately 4.1 % of Canadians over the age of 20 years had diabetes (90% of these cases are T2D). Prevalence rates of diabetes are similar in men (4.4%) and women (3.9%), and increase with age, with up to 13% of adults over the age of 60 having the disease8. Prevalence rates may be underestimated by 30% as a result of subclincial undiagnosed diabetes9. The economic impact of physical inactivity, obesity and diabetes is large. Treatment of diabetes is estimated to cost as much as $9 billion dollars annually, or 11% of the total Canadian health care budget (1998 figures)10. Estimates from the United States suggest that the direct costs of inactivity and obesity account for 9.4% of American health care expenditures7. Prevalence rates and economic costs of physical inactivity, obesity and diabetes are likely to worsen as the population ages. The disturbing trend for physical inactivity and obesity is also apparent in younger age groups. Two-thirds of Canadian children are not sufficiently active for optimal growth and development, and 25% are considered overweight. The proportion of children who are obese has increased from 2% to 10% between 1981 and 1996 alone1 1. The risk of disease is also related to the pattern of body fat distribution. A pattern of upper body fat distribution as measured by waist-to-hip ratio (WHR) is independently associated with higher risk of developing diabetes and cardiovascular disease 1 2 , 1 3. Yet the WHR provides only a crude index of adipose tissue (AT) distribution14. The development of computed tomography (CT) 1 5 and magnetic resonance imaging (MRI)1 6 techniques has allowed the visualization and precise measurement of AT depots. Much work has utilized these techniques in investigating regional AT distribution, especially in the abdominal region. 2 Most investigators have concluded that visceral AT is more closely associated with metabolic abnormalities than other AT depots17, yet there is evidence that insulin resistance is also closely associated with subcutaneous abdominal or truncal A T 1 8 . More investigation of the contribution of different abdominal AT sites is needed. The metabolic abnormalities associated with excess abdominal fat distribution include an atherogenic lipoprotein profile, high fibrinogen levels, hypertension, insulin resistance, hyperinsulinemia, and glucose intolerance6. The term metabolic syndrome has recently been adopted and defined 1 9 2 0 to unify effort in identification, diagnosis and treatment of the health consequences of the syndrome. The metabolic syndrome includes: abdominal obesity, elevated fasting glucose levels, dyslipidemia (elevated triglyceride levels and/or reduced high density lipoprotein levels) and hypertension. The clustering of metabolic abnormalities is believed to exert an additive effect on the atherosclerotic process21. Individuals with the metabolic syndrome are at increased risk for coronary heart disease, even when they are not diabetic22 ,23. These findings have led to the belief that hyperinsulinemia and insulin resistance are early events which underlie the development of not only diabetes, but also hypertension and premature atherosclerosis. 1.2 Possible Mechanisms Linking Obesity with Health Outcomes Although abdominal adiposity is closely associated with insulin resistance and other metabolic disturbances, the mechanisms linking abdominal obesity and 3 insulin resistance remain a matter of debate24. It is not clear whether the relationship between abdominal obesity and insulin resistance is causal 2 5 , or whether these two attributes are markers of some other factor. The visceral-portal hypothesis has been used as a framework to explain the potential role of visceral obesity in insulin resistance. Elevated free fatty acid (FFA) levels have been shown to impair muscle glucose uptake and impair insulin's ability to suppress hepatic glucose production26. The drainage of visceral AT into the portal circulation and its high lipolytic capacity are thought to expose the liver to elevated levels of FFA 2 7 , therefore having a unique influence on hepatic function. With increasing evidence of the detrimental effects of regional obesity, new scientific paradigms have been described, not necessarily exclusive of the visceral portal hypothesis. The first, the endocrine hypothesis, reflects the newly appreciated role of AT as an endocrine organ system. AT releases peptides, hormones and cytokines, in a manner which varies according to the location and size of the AT depot17. These compounds, collectively referred to as adipokines (structurally similar to cytokines) are a subject of intense research interest and include proteins such as tumour necrosis factor alpha (TNF alpha), leptin, resistin, adipsin and adiponectin. These compounds may play important regulatory roles in processes including lipid metabolism, feeding behaviour, hemostasis, vascular tone, energy balance and insulin sensitivity28. 4 The second hypothesis is the ectopic fat storage syndrome, proposed to explain the excess lipid found in skeletal muscle, liver and other tissues29 and its association with insulin resistance. It proposes that in circumstances of excess energy storage, AT fails to sequester the excess lipid, resulting in ectopic lipid accumulations in other tissues 2 9 , 3 0. Impairment of the proliferation and differentiation capacities of adipocytes has been suggested as a precipitating factor in ectopic lipid storage30. An alternative explanation for the increased intracellular lipid is that insufficient mitochondrial oxidation of lipid, possibly due to dietary lipid load, may then lead to intracellular lipid accumulation31. Cellular lipid oxidation is regulated by exercise, neuro-endocrine mechanisms such as sympathetic nervous system activation and endocrine hormones including those from AT (e.g. adiponectin, leptin)32. 1.3 Physical Activity Prevents Diabetes and Improves Insulin Sensitivity There are clear links between physical activity and diabetes risk 3 3 3 4 . Higher levels of physical activity are associated with a 30-50% reduction in the risk of T2D 3 5 . Intervention studies have provided evidence of improved insulin sensitivity, glycemic control, and other metabolic risk factors with exercise programs. Large randomized trials3 6"3 8 have demonstrated that lifestyle interventions including dietary counseling, moderate weight loss and increased physical activity prevented or delayed the development of T2D. Yet, as with obesity and diabetes, levels of physical inactivity in Canadian society are at dangerous levels. The majority of Canadian adults (55%) are physically inactive, and thus face increased risk of 5 disease and premature mortality. Levels of physical inactivity are higher in women and older adults4. 1.4 Possible Mechanisms Linking Physical Activity with Improved Insulin Sensitivity Several possible biological mechanisms could explain the role of physical activity in the improvement of insulin resistance39. An indirect effect is the documented role of exercise, alone or in conjunction with caloric restriction, in lowering excess body adiposity40,41. More direct adaptations also may be at play in improving insulin resistance, the most important being an increased action of insulin at peripheral tissues due to increases in muscle blood flow, insulin-stimulated glucose uptake, oxidative capacity, and muscle mass39. 1.5 Research Priorities Although the value of physical activity in improving insulin resistance is clear, what remains are the questions of what type, how much, at what intensity, and how often should exercise be done to obtain optimal influences on health. Secondly, what is the interaction between changes in regional AT distribution and insulin sensitivity with exercise interventions? The purpose of this study was (i) to evaluate the role of resistance training combined with aerobic training as a potential exercise program to improve insulin sensitivity, (ii) to follow changes of the visceral and subcutaneous abdominal AT components after exercise training to better understand their relative contribution to changes in insulin sensitivity, (iii) to measure muscle cross-sectional area and muscle density and the response of these characteristics to exercise training and (iv) to evaluate the relationship of exercise-induced improvements in insulin sensitivity with corresponding abdominal AT and skeletal muscle changes. Chapter 2: Review of the Literature Obesity, and abdominal obesity in particular, are closely linked to risk of diabetes and cardiovascular d i s e a s e 1 3 ' 4 2 . The basis for this link is believed to be the ability of obesity to engender insulin resistance. Insulin resistance is a fundamental aspect of T2D and is also linked to an array of other metabolic characteristics which form the metabolic synd rome 3 2 . Studies utilizing imaging techniques of CT and MRI have demonstrated that the visceral A T component of obesity has a close association with several characteristics of the metabol ic syndrome, suggest ing that this A T depot may have a particular negative influence on health. Recent evidence suggests that other A T depots may also influence metabol ic health. This review will begin with an examinat ion of the measurement of regional adiposity using imaging techniques. The next section will deal with insulin resistance, its measurement and pathogenesis, and what is known about the relationship between insulin resistance and regional A T depots. The final section will review the literature on how regional A T depots, insulin resistance and their relationships are affected by exercise with and without weight loss. 2.1 Regional Adiposity Obesity increases the risk for many diseases associated with high mortality and morbidity, including diabetes, hypertension, coronary heart disease (CHD), dysl ipidemia, gallbladder disease and certain cancers 6 . BMI is the commonly used, readily available indicator for assessing obesity and correlates significantly with percentage of body fat, morbidity and mortal i ty 4 3 . The W H O BMI value of 25 k g / m 2 is the generally accepted threshold for identifying an increased risk of 8 obesity-related disease, notably T2D, hypertension and CHD. Health risk increases with the degree of obesity, up to levels of extreme obesity of BMI >40 kg/m2. More than 80% of obesity co-morbidities occur in those with a BMI of at least 30 kg/m2. In particular 6 1 % of the prevalence of T2D has been attributed to obesity43. Where excess body AT is located is important in evaluating health risks. A prospective study of Swedish men and women demonstrated that WHR, independent of total body fat, was predictive of an increased risk of CHD and T2D. The highest tertile of WHR combined with the highest fertile of BMI was associated with a 30 fold increase in the risk of developing T2D relative to those in the lowest tertile of both BMI and WHR 1 3. Yet the lowest tertile of BMI with the highest tertile of WHR also exhibited a six fold increased risk of developing T2D. Thus, assessment of WHR provides additional value in assessing health risks. However, waist circumference (WC) has been found to be a better indicator of abdominal AT distribution than WHR 4 0. WC is particularly useful in assessing risk for BMIs ranging from 25-34.9 kg/m2. Above this level, WC has little added predictive power of disease risk beyond that of BMI 4 0. Increased relative risk is associated with WC > 102 cm in men or > 88 cm in women4 3. The development of imaging techniques such as CT and MRI has improved our ability to localize AT to specific regions of the body 4 4 . The distribution of AT into the intra-abdominal, intra-thoracic and subcutaneous compartments can be visualized and quantified and has been the focus of many investigations. Imaging techniques have also been used to further subdivide these AT areas. For example, 9 subcutaneous AT of the lower trunk and gluteal-thigh region has a thin fascial plane dividing it into superficial and deep portions45, which appear to have different morphological and metabolic characteristics46,47. Furthermore, recent advances in imaging techniques have revealed that AT is deposited within tissues such as skeletal muscle, liver, heart and pancreas, a deposition pattern with negative associations with health. 2.1.1 Regional A T Imaging CT, MRI and dual-energy x-ray absorptiometry (DXA) technologies have been applied in the study of body AT/fat patterning and its relation to health. In this discussion of body adiposity and its measurement the distinction will be made between lipid (fat) and adipose tissue. Adipose tissue is connective tissue with adipocytes with many functions including: energy storage, thermal insulator, mechanical cushion, and more recently recognized as having an endocrine role. Adipose tissue contains approximately 80% of lipid (triglyceride), with protein, water, and minerals making up the remaining 20% 4 5 . Imaging methods of CT and MRI allow visualization of adipose tissue. In contrast, DXA quantifies lipid, usually in the form of triglyceride. This lipid is found primarily in adipose tissue, but also is within other tissues. The total mass of lipid and AT is not necessarily identical45. The validity and reproducibility, strengths and weaknesses of CT, MRI and DXA are compared in Table 2.1. A description of what each technology measures follows. Computed Tomography CT depends on differential absorption of x-rays by the body's tissues. The attenuation of the x-ray beam is determined by the physical properties of the tissue 10 including its density and the ratio of electrons to molecular mass. Increasing the density of a material increases its linear attenuation coefficient. The ratio of electrons to mass for carbon, nitrogen and oxygen is 0.5, whereas the ratio for hydrogen is 1.0, so the higher proportion of hydrogen in AT gives it a specific attenuation coefficient48. The generated two-dimensional gray scale image provides a visualization of internal tissues and organs. Reproducible within-subject tissue attenuation values, a stable attenuation measure, and the linearity of the attenuation coefficient scale make use of an external calibration phantom unnecessary. The attenuation is expressed as Hounsfield Units (HU), or CT units. The density of water is 0 HU, -1000 HU is air, +1000 HU is bone. CT discerns AT, muscle and bone because of their different attenuation characteristics. AT is less dense than water with attenuation values of -30 to -190 HU, and muscle is denser than water, in the 0 to 100 HU range. The density of a pixel (picture element) represents the average density through the scan thickness, usually set in the 1-10 mm range as part of the scanning parameters. Areas of AT, muscle or bone can be measured by either manually tracing regions of interest or allowing software to calculate areas of interest by defining specific attenuation ranges4 9 , 5 0. Due to ionizing radiation exposure, CT protocols are usually limited to single images for research purposes. Single scans at the intervertebral space between the third and fourth lumbar vertebra (L3-4) or the fourth and fifth lumbar vertebra (L4-5) to assess intra-abdominal AT areas demonstrate correlation coefficients with AT volumes obtained from multi-slice scans of >0.9551. 11 c/> sz c IS c o ro -*—* 'E < cm c -9 'o O Q. or C ( 0 zs O CO Q) 'S O c .c o 0 H cn c CD ( 0 T J ro > CN 0> ro cu 3 (/) (0 ~ o o 3 o o 2 3 o o CO c o CD N ro 3 CO > u 0) > 10 w CD ^ E "2 O O o 3 C O • i l l * •Jo a) co § l i * S go o co TJ . E CD O Q) CO CJ <z N x "° i_ . E o C X D> S "JS ~ .L= . sz £o co j= ~ E -° -i—• *— = TJ X l C CO CO •c — g co co c > Q) t CD § 2 co 3 co co c c CD CD ro o 9- c E co o c r " ° »~ s O r- ^ E U= CD T - O c < to CO E TJ e o. ro cu O co < 2 iS o ro 2. E o -§ io CD CD 3 > « - 3 £ T 3 O CO CM 3 C •s n i io a E < co C \ - X l | r o 2 £ o c j o o — C O oo CO CO (0 ro o o 8 .9° ! oi LU co LU <U CO I— 10 _ CO CO CO o CD CO CO 1° 11 2 § 1 S II O 77 m 3 CO > H O <ox to > CD > o (0 ~ CD ro w ro j» « O) c c N ' c o 75 5 ro E CO -*—' o CD _ CD ro TJ ° o co -~ CD C D - t i CO = E-° .E ro cB .9- o -O 3 £ -7 E E E CO CO CD CD COXt II o CO C L 9 " ° O n c u JS«ro <u ui i . y co ro o 3 o = ._ CD o t O o ro o — . ~ 3 i= Z . E . £ ro.E E 2 ' 3 CO c CO £ 3 8 §•= I o E ro 0 0 = Q - o ^ w o p or > E ^ > CD .55 co (0 j>. "Jo ro o c o co CD _ « E ro c c CD O -•-» — C . t i . _ ro ~ — E g-o _ ro CO CD — O) iS ro 5 .E ro T 0 CD sz c o c o c : F O T3 CO CD CO ro 2? « ° 8 o § •= 45 = CD ^ i5 E ro £ g s -' 5 § 2 LT E CD ro o — C 0 ± i E o x i ci: — co 3 o 2 S c ro >> o * II II i _ CJ) O CO * - X l CO o CD i i *— CD co r CD CO p ro £ — ro co tr ~ CD ^ <o o ro >> 0 E io c= 0 c — i i CD •Sro£ roS CO o co .E c OT c i | l CO ^ c co o o 1 > o CM J CO aj CD ro E CD CD CL £ CD > O CO 3 o CD c co ^—• 3 o XI 3 (0 ro 1 CD O CO •> LU LU — T O ) OO Q < C D CD CO _ ~ w- CO Q . O T3 cr> lit c5 2 CO 3 iSS • g 01 U . o O ID ( O o ^ ro g g°> ro c O v_ •J= o o CD (0 (0 •a CD §i ro "o ro s-o CD • 1 CO > or ro c O CO ro E co o 3 ^ 3 o -Q CD ro c co 3 cz o -XI T 3 3 C co ro LU w c n CD CO (0 ^ -ro °> c o ro 11 CD CD < ro o CO 3 O CD CD O CO 3£ g ro CM 2 .£ co ro TJ ro o 10 > o (0 3 E CM *— o ( - LU CO > S-2 XI ro CO<N < < ro c 'E o TJ X l ro x: ro .22 i= 3 o sz ro ro •5 c g) CO CO a) XI ro 1 CO 3 CO o CL x I— ® O •43 ro .55 £ H E § < 1 a) o CD a). r- CO o co ro — N ro ro ro ro MH- — 0 ro o XI ro "co • u > O TJ O XI ro —^< o co o ro 3 o O o , < ^ • CO a CD c O XI I ro CD c CO CD "o 2 3 O O ro CD ro CD O c CD g CD o o > o CD" ra E CO CD CD O \ o CD "E ro TJ tz ro ^—. CO LU LU CO o CO XI CO >s ro ro Q CO c CO ro E CD O c ro c o CO CD CD c CO ro E E >< sz CL ro co o E 2 TJ CD E o O o ro 51 12 CT imaging has also been applied to characterization of lipid within muscle. Ectopic lipid contained within normally lean, dense tissue such as muscle would decrease the attenuation coefficient. Lipid content of muscle biopsy samples56, phantom lipid concentrations and microscopic evaluation of histological lipid staining 5 7 have shown significant correlations with CT attenuation. Thus, decreased attenuation of skeletal muscle appears to reflect increased lipid content, but it does not allow differentiation of lipid from around muscle fibres (interstitial AT) from that lipid within the muscle cell (intramyocellular). Because muscle ectopic lipid is associated with obesity, insulin resistance and muscle weakness, CT scanning can provide information about muscle composition in health and disease states50. Magnetic Resonance Imaging MRI involves placing a subject within a magnetic field, which causes normally random magnetic moments of tissue protons to align in one direction. Their absorption and subsequent "relaxation" or release of energy from an applied radio frequency field, differ amongst tissues. The varying relaxation times are used to construct an image. Varying the time parameters of the applied radio-frequency pulse optimizes contrast between tissues52. Distinction between adipose and lean tissue can be made on the gray-scale image generated, similar to CT image analyses. MR whole body images can be acquired in approximately 30 minutes, although image analysis takes considerable time and sophisticated software is required. MR has a greater variability in assessing visceral AT, particularly if the 13 visceral AT area is small. In practice several images are taken to assess visceral AT 5 8 . Very recently MRI methods have been developed to quantify lipid content of muscle. Standard MRI allows visualization of interstitial AT, but does not allow quantification of the intramyocellular lipid (IMCL). Recent advances in MRI protocols have been used to measure IMCL5 9, although its use is limited. Recent advances in 1H-magnetic resonance spectroscopy (MRS) have provided the ability to image IMCL in vivo. This procedure detects proton signal differences between fatty acids which are within the muscle cell (IMCL) and those from surrounding AT 6 0 Dual-energy X-ray absorptiometry DXA measurements are made using an x-ray beam with two energy peaks, and the ratio of beam attenuation at the two energy levels is compared to a calibration value for water and lipid. DXA was originally developed for bone mineral assessment, and later refined for soft tissue. The principle underlying DXA is based on a three-compartment model in which fat, bone mineral and the remainder (which is fat-free, bone mineral-free tissue termed the DXA lean soft tissue) attenuate the x-ray beams in a tissue-specific manner. DXA measures of body fat include all fatty elements and are not exclusive to AT 5 0 . When DXA is used to assess abdominal adiposity it provides a measure of body lipid but does not differentiate the location of the lipid, or the pattern of AT depots. 14 A combined approach is often used to assess total body adiposity: for example total body fat has been measured by DXA with single image CT scans to assess regional AT patterning. It has also been proposed that DXA assessments of central abdominal fat are of value in understanding health implications of regional adiposity61. 2.1.2 Regional Adipose Tissue Patterns Relative to Gender, Ethnicity and Age The imaging methods described above have been instrumental in providing information about regional AT patterning, and how it is related to genetics, gender and aging. As with generalized obesity, there seems to be a genetic predisposition to the development of visceral obesity. The Heritage Family study62 in 86 families found the maximal heritability for CT-assessed visceral AT was 48%. Heritability includes both genetic and familial environmental effects. The Quebec Family Study used CT measures of visceral AT in 382 individuals from 100 families, and attributed 60% of the variance in visceral AT size to heritable factors. Twin studies illustrate the role of genetic factors in abdominal AT patterning. Overfeeding of 6 pairs of monozygotic twins for a 22 day period was followed by CT measures of visceral AT. Consequently, 75% of the variance in site of AT deposition was found between pairs, with only 25% of variance within the pairs63. Gender differences are apparent in pattern of abdominal AT, as a central accumulation is more prevalent in men than premenopausal women. For any given level of body adiposity, men have on average twice the amount of visceral AT as premenopausal women64. Levels of visceral AT are also correlated with 15 increasing age in men and women. In women, the pattern of body AT deposition changes after menopause with the accumulation of visceral AT. This pattern seems to be avoided with the use of hormone replacement therapy65. Further work is necessary to establish to what extent the expanding visceral AT may be contributing to the deteriorating metabolic profile seen in postmenopausal women. Body AT distribution differs across ethnic groups. Mexican-Americans have a greater propensity to abdominal AT accumulation than Caucasians, who in turn have a greater abdominal adiposity than African-Americans65. These differences may modify the predictive validity of using WC as a surrogate for abdominal obesity. WC is a better predictor of disease risk than BMI in persons of Asian descent. WC also has greater value for estimating risk at older ages40. The availability of CT and MR scanning has allowed quantification of the AT depots in different regions. The relative sizes of the various depots are outlined in table 2.2. Table 2.2 Sizes of regional AT depots Depot kg Approximate size % of total AT Subcutaneous abdominal AT 2.4 -7.2 kg' 23-29% $ 3 20-23% S 1,2 Intra-abdominal AT Visceral 1.1 -3.5 kg1 2-4 .4% $ J 10- 13%<3 1,2 Retro-peritoneal 0.8-1.6 kg1 4.6 - 7.8%1 Non-visceral internal AT 3.2 - 3.7% $ J 1 from ref'O O D 39 men, BMI 23-37 kg/m' full body MR scanning 2 from ref 4 1, 52 healthy men, BMI 33 kg/m2, full body MR scanning 3 From ref66, 67 women ranging in BMI from 19-40 kg/nr, full body MR scanning 16 2.1.2. Classification for Regional A T Compartments by Imaging Methods Our insight into the role of excess AT and its impact on health has developed into the understanding that specific regional depots have varying associations with pathological processes. As research expands and imaging technologies becomes more widely available, variability in the terms and definitions of AT compartments hampers comparisons. Inconsistencies in the use of specific terms have been reviewed and a more accurate classification of AT compartments has been proposed 4 5 . The role of visceral AT has been of particular interest due to its associations with negative metabolic profiles. Yet definitions of visceral AT have varied, mostly referring to intra-abdominal or intra-abdominal plus intra-pelvic AT. Whole-body scans with MRI protocols can also measure intra-thoracic AT, which as a body cavity could also be considered "visceral'. The proposed classification system differentiates total body AT into subcutaneous and internal components. Internal AT is further separated into visceral and non-visceral components. Visceral AT should include AT in the chest, abdomen and pelvis (although in practice it is commonly used to refer to abdominal AT only), and non-visceral AT would include that which is within and between muscles. Visceral AT could be further divided into intraperitoneal and extraperitoneal regions. The intraperitoneal region includes the omental and mesenteric components which are of interest relative to their metabolic activity. Visceral AT is often measured from single slice images of the abdominal cavity, usually at the L4-L5 level. These images would include omental, mesenteric and retroperitoneal compartments. For consistency, anatomical boundaries used for measuring visceral AT should use the inner boundary of abdominal muscle wall as the limit, which would exclude the intra-muscular AT. 2.2 Insulin Resistance Insulin resistance refers to a metabolic state characterized by decreased tissue responsiveness to physiologic concentrations of insulin, i.e. a reduced insulin sensitivity67. In practice insulin sensitivity, a more specific term than insulin resistance, is defined as the insulin concentration needed to produce half the maximal rate of glucose uptake6 8 , 6 9. Less than normal insulin sensitivity presents as insulin resistance, but the limits of what may be considered normal insulin sensitivity are not defined70. Furthermore, it is not yet clear whether the resistance to insulin's action may also be apparent in it's other metabolic roles (e.g. fat and protein metabolism, ionic transport, cell growth and differentiation). In this discussion insulin resistance is used in its general application as reduced insulin sensitivity. In 1936, Himsworth71 recognized that some cases of diabetes were a result of a lack of "insulin-sensitizing factor", rather than a lack of insulin. This is now referred to as insulin resistance and the diabetes to which it is related is T2D. More recently, the concept was reintroduced and expanded, and now insulin resistance is seen as an underlying cause of not only T2D but also a constellation of risk factors related to C V D 2 2 . The clustering of features was initially termed Syndrome X, and included impaired glucose tolerance, hyperinsulinemia, dyslipidemia and hypertension 2 2 . 18 Since then the syndrome has continued to evolve and be renamed and redefined. Currently, two descriptions of the syndrome associated with insulin resistance are in use (Table 2.3), both of which refer to it as the metabolic syndrome. Table 2.3 NCEP and WHO definitions of the Metabolic Syndrome NCEP Definition Modified WHO Definition At least 3 of the following: Hyperinsulinemia or fasting plasma glucose > -Fasting plasma glucose > 110 mg/dl (6.1 110 mg/dl (6.1 mmol/l) mmol/l) AND - Waist circumference >102 cm men, >88 cm At least 2 of the following: women - Abdominal obesity - Serum triglycerides > 150 mg/dl (1.7 mmol/l) definition 1 - waisthip ratio >0.9 or BMI >30 - Serum HDL-cholesterol < 40 mg/dl (1.04 definition 2 - waist circumference > 94 cm mmol/l) - Dyslipidemia - serum triglycerides > 150 mg/dl - Blood pressure > 130/85 mm Hg or medication (1.7 mmol/l), or HDL cholesterol < 35 mg/dl (0.91 mmol/l) - Hypertension - blood pressure > 140/90 mm Hg or medication Adapted from refu The WHO definition focuses on insulin resistance as an underlying cause of the metabolic syndrome, with a greater emphasis on the genetic basis of the syndrome, and pharmacological approaches to treatment. By definition, insulin resistance (determined during a hyperinsulinemic-euglycemic clamp) is part of the WHO definition. The National Cholesterol Education Program - Adult Treatment Panel III (NCEP) definition approaches the syndrome as having a behavioral or 19 lifestyle basis, where the focus is on prevention and treatment of obesity'". According to the NCEP definition, insulin resistance is usually present, but not necessary to qualify as the metabolic syndrome. The portion of obese individuals with multiple risk factors without insulin resistance is not known. The criteria provided by either the WHO definition or the NCEP definition can be applied in most clinical settings, and allow insulin resistance to be recognized by some of its key features. The concept of an insulin-resistant/metabolic syndrome integrates the many divergent aspects of insulin resistance, and is useful in understanding the link between disorders such as T2D, atherosclerosis, hypertension, and hyperlipidemia74. Thus, insulin resistance and the compensatory hyperinsulinemia that are part of this syndrome may underlie the development and progression of several of these cardiovascular disease risk factors. 2.2.1 Measurement of Insulin Resistance The gold standard measurement of IR is the hyperinsulinemic euglycemic clamp. During this procedure, insulin is infused at physiological (10 or 40 mU/m2/min) or supra-physiological levels (>100 mU/m2/min) to achieve steady-state insulin concentrations. Glucose is infused at a variable rate to maintain euglycemia until a steady state level is reached between 2- 3 hours. The glucose infusion rate represents glucose uptake rate, as hepatic glucose production is completely suppressed. Higher glucose infusion rates indicate higher whole-body insulin sensitivity. 20 In the hyperglycemic clamp technique, plasma glucose is acutely raised and maintained, and the plasma insulin response is measured. The relationship of the amount of glucose infused to the mean insulin concentration is used to estimate insulin sensitivity. Both the hyperinsulinemic-euglycemic and hyperglycemic clamps can also use tracer techniques to differentiate hepatic and peripheral insulin sensitivity. Both clamp procedures are labour intensive, but are the most sophisticated, accurate and widely used methods to measure insulin sensitivity75. Other techniques are available, providing different interpretations (Table 2.4). Two intravenous glucose tolerance tests, the frequently sampled intravenous glucose tolerance test (FSIGTT) and the intravenous glucose tolerance test (IVGTT), were developed to quantify whole-body glucose metabolism and can be used in large population studies. A bolus injection of glucose is followed by blood sampling over the next 3 hours. A measurement of insulin sensitivity is derived from computer-modeling of glucose and insulin kinetics75. These models represent a composite of hepatic and whole-body insulin sensitivity. The oral glucose-tolerance test (OGTT) follows plasma glucose and an insulin response to an oral glucose challenge (usually 7 5 g glucose) and is a reflection of the composite of hepatic and whole-body insulin sensitivity. The homeostasis-model assessment (HOMA) incorporates fasting plasma glucose and insulin levels in derived equations to arrive at an insulin sensitivity index, and is more reflective of hepatic rather than whole body insulin sensitivity. OGTT and HOMA 21 assessment models are relatively crude, and do not directly assess tissue insulin responsiveness, but are simple to administer with a large body of established research utilizing these methods70 ,76. 22 CO CD m B c ro 1 CO co CD > CO cz CD a. X a> Q) CO > CZ co C C CD CD CO 4—1 i ro cu to ro CD x ° g T J Q - CD 11 2 CD J Z o_ cu CL CO > CO I & ro c ro l § ro o c ro CO 3 CA C c 3 CO CO Cl) CN CC 1-ro CL CD »n CO CO O  CO CO 73 . cz co o co l CD O CO i= o + .2 ro CD T J J Z cz Q . ro ai lo CL CO CD .« 8 135 CO CD CO CO > CD ™ ••-» J Z T J aj *i ™ X CO 0 cz o Q) I C O O '••= CD 7S CO c], 3 r o uj ^ CD "O W T J CO " X 8 l 8 1 2 + 2 o O ) 1 C D C T J c ro >, I a? i t * ro <2 >. CZ CD = CO J3 s o " ry- zi r- T J T - O I fc; in CL CN 2? CO CD 4= CD CD > CO c CD CL X CD CD" CO > £. CO ti C CD CD CO H 3 ro o CD o co ro fD x S CL CD 11 3 CO 2 ° I! CO Q -0) . CO 3 > O CD co B c= Q. e g o 3 — fc; J= >-Q - ro o «> E CD CD UJ CD CL CO CO CD CD > • -i t: o 8 * CD - K , co 9^ V 8 ro 3 CD CO -Q T J CO "J  " — ' T J ZI N 8 N CD CO C • = — z. = > co ro w CD j ^ T J CO ^ CD CO CD | CZ CO co cz O CD 1 CO CD T J !E — T J c co .2 E O CD — 4^  5= CO 3 >* CO CO >,<=> :> >> 55 a cz ro CD cz CO .E cz X I g 3 Q_ CO CL .E ro -°§§ CD .2 = o t 0 ) i 8-CL CL CO fZ >• 3 JS X CD O O 'E CD O >» 75ie CD £ CL E >. ro X o o 4— 0 o CD CL CO *—< o cz CD \ CD 00 co™ >. cz ro = CO 3 CO CO co E o co -2 § CD CO CO i l l co I 'gj? CD > CO c CD CL X CD c CD-CO CD CO CO U J O 5 CD CD '5 o 0 C CL E T J CD CD CO CO CD 32 CO C „ CO ro co 0 CO o o CD O o f— c o ro^ CO CD O CL E ro o CO CO ro co o CD E o X co "c "5 co c X I ~ ro o o — ro > CO 3 > T J CZ 2 c c ro o CD o c CD 3 CT CD CO c o o CD J Z CO £ 3 CO CO CD J Z o CD CO = CO ro 3 > ^  > '*= TJ o cz CD — o- o eo u co g> P S CD cz >, 5 CD CO cz O -CL CD 9? co c rn m c °i CO ™ CD O o T J JO c co .2 cz o ™ ro i= CO J Z 4—1 c o • 0 CD CO co J3 JZ 2 ro CD (A CD >% T J O J3 CD E> ro CD_ ™ if 4- O O T J CD O ro CD CP J Z c CD E CO CO CD 11 CO c ro « CL 3 X o CD CD ~ CL I I — o h o O o • 10 CO fl> 8 CD ^ c z P O ) CO K ro CD CD 0 2° CD 4—• m 3 O 3 CO tz cz CD CD CD CD J3 3 CO _ CO co cy E o | T 5 c ro o o cz cz ro - ' § . 2 co .•ti co g=5 S o S = S CO o 4— CD T J CZ ro tz o CO CD co t ; — cu 2- CD •fc:  3 CD CO O CZ T J CD CO o o 3 Z co CD J Z CO ro CL CD c o o CD w 61 cz cz 1 - o CO CD « ro E "co CD a) E5 . cz 1 = S o S E ^ ro CD CO > CD CO 3 S 5 -»-» •— CD CO CO .E o c g 'co CO CD ( CL CL 3 CO C -•U .2 o "S tj c CL 3 CO CZ T J CD = 0 0 O fc: T J O CL T 3 O o ro CD T J •> o CD C "D CD CZ J Z •— co — CZ T J CD CD w ro cz 3 ro t CD O CO > JD ' o O C O O 3 CO "43 T J CZ £ S CO O Q . a 8 s Q. .E .= CD 3 CO fD CZ X J CD O CZ CO CD I— cz CD y CD CO > > r> to CD ro o i= 3 T J i z CO o c: L O .55 • i _ o ro ro > CD CZ T J r~ CD CD CD CD E CO ,•- ro LL. LL. CO t Z -2.2.2 Prevalence of Insulin Resistance Prevalence estimates of insulin resistance (IR) vary depending on its definition and the population being tested. IR is almost universal in those with established T2D (80-88%), and highly prevalent in those with impaired glucose tolerance or impaired fasting glucose (50-59%)81. But IR is also present in populations who have normal glucose tolerance. For example, hyperinsulinemic-euglycemic clamp data demonstrated that IR existed in 25% of normal glucose-tolerant Swedish and Finnish adults82. Prevalence estimates of the metabolic syndrome, the syndrome associated with insulin resistance, are 24% in U.S. adults, and 45% in older adults (50-70 years of age) using the NCEP criteria83. Canadian estimates suggest that the metabolic syndrome exists in 17% of Canadians84. IR is associated with advancing age. This is clear from the results of the European Group for the Study of Insulin Resistance (EGIR) who measured insulin action by hyperinsulinemic-euglycemic clamp in 1146 healthy Caucasians85. A significant decline in insulin action was seen starting with subjects aged 50 years up to age 70 years, with no sex difference seen. Increased IR with aging is related to certain anthropometric and lifestyle factors86. Body compositional changes with aging include a decrease in lean body mass (especially muscle tissue) and an increase in adipose mass, with progressive redistribution of AT to intra-abdominal AT, away from limb subcutaneous A T 8 7 . Intra-abdominal AT accounted for over 50% of the variance in insulin sensitivity, whereas age per se was not associated with insulin sensitivity87, suggesting the importance of anthropometric changes vs. 24 obligatory aging changes. Age-related variations in physical activity play a key role, as aged subjects undergoing exercise training increase whole-body insulin action86 2.2.3 Sequela of Insulin Resistance A high and increasing prevalence rate of the metabolic syndrome and IR has serious health and economic ramifications. It is felt that IR, and compensatory hyperinsulinemia, are acquired early, and begin a cascade of metabolic changes that lead to diabetes, dyslipidemia, hypertension, hypercoagulability, and eventually cardiovascular disease21. But it is the clustering of metabolic factors that exerts an additive effect on atherosclerosis 2 0 . The impact of the metabolic syndrome can be illustrated by the relationship of diabetes to atherosclerotic disease. The presence of diabetes markedly increases the risk for coronary artery disease and cardiovascular disease mortality88. Risk for a CVD event in individuals withT2D without established CVD is equivalent to the risk in individuals with established coronary disease. Similarly increased CVD risk is seen in other insulin-resistant conditions such as impaired glucose tolerance, the metabolic syndrome and obesity (Table 2.5). These findings suggest that the risk of complications commences many years before the onset of clinical diabetes. This radical change in the understanding of insulin-resistant conditions influences contemporary therapy, suggesting an earlier and far more aggressive approach to the treatment of the hyperglycemia as well as CVD risk factors such as hypertension, dyslipidemia and central obesity, in the hope of significantly reducing cardiovascular morbidity and mortality89. 25 obligatory aging changes. Age-related variations in physical activity play a key role, as aged subjects undergoing exercise training increase whole-body insulin action86. 2.2.3 Sequela of Insulin Resistance A high and increasing prevalence rate of the metabolic syndrome and IR has serious health and economic ramifications. It is felt that IR, and compensatory hyperinsulinemia, are acquired early, and begin a cascade of metabolic changes that lead to diabetes, dyslipidemia, hypertension, hypercoagulability, and eventually cardiovascular disease21. But it is the clustering of metabolic factors that exerts an additive effect on atherosclerosis 2 0 . The impact of the metabolic syndrome can be illustrated by the relationship of diabetes to atherosclerotic disease. The presence of diabetes markedly increases the risk for coronary artery disease and cardiovascular disease mortality 8 8 . Risk for a CVD event in individuals withT2D without established CVD is equivalent to the risk in individuals with established coronary disease. Similarly increased CVD risk is seen in other insulin-resistant conditions such as impaired glucose tolerance, the metabolic syndrome and obesity (Table 2.5). These findings suggest that the risk of complications commences many years before the onset of clinical diabetes. This radical change in the understanding of insulin-resistant conditions influences contemporary therapy, suggesting an earlier and far more aggressive approach to the treatment of the hyperglycemia as well as CVD risk factors such as hypertension, dyslipidemia and central obesity, in the hope of significantly reducing cardiovascular morbidity and mortality89. 25 Table 2.5 Cardiovascular disease outcomes in insulin-resistant conditions CVD outcome Insulin-resistant Condition T2D -t2x risk of CHD & stroke in men t3-4x risk of CHD & stroke in women - post Ml - 4x risk of re-infarction, greater risk for heart failure, higher mortality rates - 80% of diabetic mortality due to atherosclerosis - 75% of diabetic complication hospitalizations due to atherosclerosis - 50% of newly diagnosed diabetics have CAD Metabolic Syndrome 3 -4.3 fold increased risk for CVD mortality ,72 Impaired glucose tolerance -1 .3 -2 fold increase relative risk in CAD mortality Obesity 2-3 x relative risk CVD cardiovascular disease, CAD coronary artery disease, Ml myocardial infarction From ref38 and 9 0 2.2.4 Site of Insulin Resistance Resistance to the action of insulin has been clearly demonstrated in both peripheral tissue and the liver. Hepatic IR is manifested by overproduction of glucose, despite fasting hyperinsulinemia. The excessive hepatic glucose production results from augmented gluconeogenesis and glycogenolysis. In the insulin-stimulated state, peripheral tissues, primarily muscle, are the major site of IR91 and are responsible for postprandial hyperglycemia with compensatory hyperinsulinemia. The major defect in insulin action at the cellular level is post-receptor; with alterations in insulin signal transduction which diminish glucose transport and phosphorylation92. IR is not limited to skeletal muscle and liver, but is also evident in AT. IR in adipocytes lead to chronically elevated levels of free fatty acids (FFA) in the blood, which induce IR in muscle and liver and impair insulin 26 secretion60. Other tissues, such as the gastrointestinal tract and the glucagon secreting (alpha) cells of the pancreas have demonstrated alterations in function in diabetes, which may be involved in the pathogenesis of the disease91. The familial nature of IR suggests genetic links, but to date there do not appear to be any single or multiple common genetic mutations which underscore IR74. Family resemblance studies comparing the pattern of IR between monozygotic twins and studying nuclear families or extended pedigrees have suggested that genetic factors influence IR. However, this approach does not distinguish shared environmental influences from the influence of genes per se9 3. Foremost among the acquired factors that cause or aggravate IR are obesity, especially central or abdominal accumulation of AT 9 4 " 9 6 , and physical inactivity. These two factors will be discussed in the following sections. 2.3 Abdominal Obesity and the Pathogenesis of Insulin Resistance Obesity is related to IR and the relationship is even stronger if a high proportion of AT is centrally located (as assessed by waist circumference or W H R ) 1 3 ' 1 7 2 5 9 7 . One approach to investigating respective contributions of abdominal adiposity to metabolic risk involved matching for total body fat in obese subjects, who had low or high visceral AT areas on CT images98. Obesity was associated with moderate increases in fasting plasma insulin and triglyceride concentrations. However, obesity with a high proportion of visceral AT was accompanied by a cluster of metabolic abnormalities including hypertriglyceridemia, low HDL concentrations, 27 hyperinsulinemia, and a deterioration of glucose tolerance. It appears that obese subjects with the highest accumulation of visceral AT have the most deleterious metabolic risk profile. Most investigators have concluded that visceral AT is more closely associated with metabolic abnormalities than subcutaneous AT 1 7 . Visceral AT was associated with glucose and insulin responses to an oral glucose challenge in obese men" and premenopausal, obese women100, and was found to be a better correlate of metabolic complications than body fatness per se 9 9 . Much of this evidence relies on post-challenge glucose and insulin measures, which are an indirect assessment of insulin sensitivity. The associations between visceral AT and the glycemic response to an oral challenge are not large (accounting for less than ~30% of variance in glucose tolerance)99"101, and relationships with other abdominal or central AT sites were also evident (abdominal subcutaneous AT, anthropometric measures of central adiposity (sum of trunk skinfolds), and CT scans of the chest to thigh region). Further studies employing more direct measures of insulin sensitivity, i.e. a hyperinsulinemic euglycemic clamp, strengthened earlier relationships with central body adiposity 6 6 9 4 ' 1 0 2 . insulin action had closer relationships with central body fat by DXA measurement (r= -0.64 to -0.89), by trunk skinfolds (r=-0.72), and by abdominal subcutaneous AT (CT imaging, r=-0.61, MR imaging r= -0.62), whereas visceral AT demonstrated a significant but lesser correlation coefficient (r=-0.52). 28 Trying to identify which regional AT depot is more closely associated with insulin resistance is problematic due to the relationship of subcutaneous and visceral AT to each other (r ~ 0.72-0.92)24, and many correlation studies have been confounded by this relationship. It is clear that an association exists between abdominal obesity and insulin resistance; some results show visceral AT as a closer correlate, while others support subcutaneous truncal or abdominal AT as having a closer link. More evidence on this issue is needed. As the hyperinsulinemic euglycemic clamp was adopted for use, more accurate measurements of IR were made, allowing deeper investigation into the relationship between IR and various regional AT depots. Visceral and other AT depots were found to be associated with IR. Euglycemic clamp studies103 and IVGTT 1 0 4 in both lean and obese women found that IR was related to visceral AT. In contrast, subcutaneous abdominal AT on MR imaging provided incremental prediction of IR (assessed by euglycemic clamp), whereas intraperitoneal AT (a component of visceral AT) did not increase the predictive power in healthy men 6 6 and men with type 2 diabetes18. Another study of men and women with varying levels of adiposity found subcutaneous AT on CT scans to have the strongest association with IR94. This relationship retained significance after controlling for visceral AT, yet visceral AT did not retain significance after controlling for subcutaneous abdominal AT. Other investigations have utilized DXA measurements to investigate regional fat distribution and IR. DXA imaging assesses total fat in a region of interest, but cannot distinguish visceral from subcutaneous AT. Strong 29 relationships have been reported between DXA measurements of central (approximately 1s t to the 5 th lumbar vertebrae) abdominal AT mass and euglycemic clamp-assessed IR 1 0 2. Truncal skinfolds have also been reported as stronger correlates of clamp- assessed IR than MR measures of visceral A T 6 6 ' 1 0 4 . The above data questions the singular role of visceral AT in the development of metabolic disease, and is a topic of debate 2 4 , 1 0 5. Insulin sensitivity may be a function of visceral and subcutaneous AT on the trunk, and potentially of other yet unrecognized adipose deposition sites. Also, relative sizes of AT depots may play a role in their potential metabolic impact. By virtue of size alone, subcutaneous abdominal or truncal AT might impart an influence on insulin sensitivity. The finding that non-visceral internal AT volumes can be greater than visceral AT volume is intriguing106; representing an AT depot whose physiological function is to be determined. Other investigators have suggested that further depots of AT are also related to IR. For example, abdominal subcutaneous AT can be subdivided into a more superficial and a deeper layer. The deeper area of subcutaneous abdominal AT, which is anatomically separated from the more superficial layer by a band of fascia and visually distinct on CT images, was closely related to glucose disposal during a euglycemic clamp in lean and obese men and women, whereas the area of the more superficial layer was not related46. In a similar approach, subcutaneous thigh AT can be subdivided into deeper and more superficial layers according to the 30 band of fascia. As with subcutaneous abdominal AT, the deeper area of AT was a significant correlate with glucose disposal during a euglycemic clamp, while the more superficial layer was not107. The use of CT and MR imaging has expanded our understanding of body AT patterning relative to metabolic variables. AT depots of interest are no longer limited to the visceral depot but should include other components of upper body obesity such as subcutaneous abdominal and subcutaneous truncal regions. Much of the above evidence is cross-sectional so a causal relationship between IR and AT regions can only be suggested. More evidence of a prospective nature would be helpful in following changes in regional AT depots and changes in IR. 2.3.1 Abdominal Obesity or Insulin Resistance as the Initiating Factor As research began to reveal the pattern of metabolic abnormalities associated with upper body adiposity, the widely-held belief was that obesity was the underlying factor108. Others held that insulin resistance was the underlying factor, and that once insulin resistance had developed, the other characteristics of the syndrome would follow, if the individual was genetically susceptible25. In support of this belief, it was shown that baseline insulin levels, independent of body weight or AT distribution, were predictors of the development of metabolic disorders109. The exact nature of the relationship between insulin resistance and abdominal obesity, whether causal or not, is yet to be determined. It is possible that an unknown common genetic or environmental factor produces both insulin resistance and 31 central obesity Some of the possible mechanisms linking these two factors will be reviewed in subsequent sections. 2.4 Possible Mechanisms Linking Abdominal Obesity with Insulin Resistance What are the possible mechanisms linking excess abdominal AT with the degenerative metabolic aspects of these diseases? There are several lines of research implicating these possible links, with some findings indicating regional AT differences. The approach to IR and associated conditions has shifted in recent years from a glucocentric to a lipocentric view60 to reflect the growing understanding of the role of lipid metabolism in the pathogenesis of IR. Some of the links between dysregulated fatty acid metabolism and IR are discussed below. 2.4.1 Dysregulation of Fatty Acid Metabolism - Elevated Free Fatty Acids In insulin-resistant states such as T2D, obesity, and the metabolic syndrome, elevated FFA levels from excess AT lipolysis are strongly implicated in the pathogenesis17 of IR. For several decades, elevated FFA levels and FFA oxidation have been thought to compete with glucose for substrate oxidation (Randle cycle, or glucose-fatty acid cycle) and it was speculated that increased lipid oxidation might cause the IR associated with diabetes and obesity110. FFA may contribute to IR by inhibiting glucose uptake, glycogen synthesis and glucose oxidation111. Elevated FFAs are also associated with a blunting of insulin-stimulated recruitment of GLUT 4 glucose transporters to the plasma membrane17. Thus, FFA-induced IR may be the result of an alteration in insulin signaling. 32 The predominant paradigm to explain the link of abdominal obesity with IR has been the visceral-portal theory which suggests that visceral AT has deleterious effects due to its anatomical site and pattern of venous drainage. The portal drainage exposes the liver to excess FFA, impacting its regulation of glucose and lipoproteins. Although in vitro evidence had suggested higher lipolytic activity of visceral adipocytes 1 1 2 , isotopic dilution techniques in vivo have demonstrated that increased FFA release in upper body obesity is primarily (75%) from upper body subcutaneous A T 1 7 . Increased lipolytic rates from abdominal subcutaneous adipocytes are thought to be related to greater beta-adrenoreceptor density or sensitivity113 and/or a larger adipocyte size17. Larger adipocyte size, as found in abdominal subcutaneous adipocytes in individuals with upper body obesity, is correlated with both basal and stimulated lipolysis rates17, and has been found to be associated with insulin resistance114. Elevated FFA levels have documented effects on many tissues. At the liver they increase hepatic glucose production, contributing to hyperglycemia, and stimulate VLDL production with the resultant dyslipidemia patterns (elevated triglycerides, low HDL-cholesterol and the increase in small, dense LDL particles) characteristic of insulin-resistant individuals17. Vasoconstriction is affected by FFA levels and may be the link between the elevated FFA levels and the hypertension seen in the metabolic system. FFA levels may also modulate the insulin regulation of skeletal muscle blood flow and vasorelaxation, which may contribute to hypertension of the 33 metabolic syndrome. The pancreatic islets may also be susceptible to elevated FFA levels, with consequent impaired insulin secretion. Thus many of the metabolic abnormalities of IR are suspected to be related to an elevated FFA release that occurs in upper body AT distribution17. 2.4.2 Ectopic Lipid Accumulation Accumulation of lipid and products of fatty acid metabolism inside muscle cells has been implicated in the cause of IR. In genetically obese diabetic animal models the syndrome of IR, dyslipidemia, T2D and CVD is referred to as lipotoxicity and is attributed to products of excess fatty acid oxidation that build up in skeletal muscle, pancreatic islets and myocardium30. Accumulation of high levels of these metabolic products leads to complications by disrupting cell function and ultimately by promoting programmed cell death (apoptosis). The possibility that the complications of human obesity are similar to those of the obese rat model has drawn research interest to the area of ectopic lipid deposition and its relevance to human disease. The "ectopic lipid storage" hypothesis proposes115 that excess dietary lipid causes adipocyte hypertrophy and overload with resultant lipid storage in ectopic sites, such as liver, skeletal muscle, and pancreatic beta cells. AT accommodates increased lipid storage by hypertrophy and by increasing differentiation of pre-adipocytes into lipid storing cells. A nuclear hormone receptor, peroxisome proliferator-activated receptor-y (PPAR-y), regulates adipocyte differentiation, and PPAR-y in turn is regulated by prostaglandins, cytokines, and hormones (e.g. 34 corticosteroids, insulin). Impairment of the proliferation and differentiation system of adipocytes may be a precipitating factor in the shunting of lipid to deposition in tissues other than AT. Of interest also, is the observation that subcutaneous pre-adipocytes may have a greater differentiation capacity than those from the visceral Ectopic or intracellular lipid is thought to decrease insulin action in muscle and liver, and impair insulin secretion from the pancreatic beta cell 1 1 6. Increased lipid content in skeletal muscle appears as reduced attenuation on CT imaging. A lower muscle attenuation in obesity and T2D has been reported 9 4' 1 0 3 , 1 1 7' 1 1 8, and was a strong predictor of insulin sensitivity, adding independent predictive value to that of visceral A T 9 4 . IMCL, as assessed by MRS, also demonstrated a close relationship with IR in glucose tolerant subjects1 1 9 , 1 2 0, impaired glucose tolerant subjects, and insulin-resistant offspring of T2D parents121 or those with T2D 1 2 2 . The appearance of higher levels of IMCL early in the development of IR (i.e. in non-obese, insulin-resistant offspring) suggests that deposition of lipid within skeletal muscle may be an early body composition abnormality, rather than a late complication of excess adiposity123. Studies have examined whether manipulating IMCL would cause reciprocal changes in insulin sensitivity. Artificial elevation of circulating FFA can result in a suppression of insulin-mediated glucose uptake in healthy people, 35 which corresponds with the accumulation of triglyceride in skeletal muscle cells (measured by MRS). The thiazolidinediones (TZD's) are a class of anti-diabetic agents which improve hyperglycemia, at least in part by improving insulin sensitivity124. Their mechanism of action is thought to be by activating molecular nuclear receptors called peroxisome proliferator-activated receptors gamma (PPARy). Once activated, the PPARs regulate genetic transcription and translation of proteins involved in glucose and lipid metabolism in AT, skeletal muscle, and liver, although the greatest expression of PPARy is in adipocytes32. One of the actions of PPARy is to stimulate preadipocyte differentiation and redistribution. PPARy activation may increase the number of small adipocytes, which are thought to take up more glucose than large adipocytes125. Ex-vivo data suggests that TZD-stimulated differentiation is greater in subcutaneous than visceral depots. This may be the mechanism behind the observation that TZD therapy results in redistribution of visceral to subcutaneous A T 1 2 6 . PPARy activation by the TZD agents as anti-diabetic therapy presents a paradox as the agents enhance insulin sensitivity yet increase lipid mass (through promotion of adipocyte differentiation). The adipogenesis may be selective in recruiting subcutaneous depots rather than those in the visceral region. The therapeutic effects of TZDs suggest that PPARy which is predominantly expressed in AT, plays a role in insulin sensitivity, although much remains unknown about the mechanisms of action. 36 An alternate approach suggests that ectopic lipid deposition arises from impaired energy production within the cell, rather than from increased substrate delivery to the cell as described above. Mitochondrial bioenergetic function, size31, and oxidative enzyme levels103 were seen to be diminished in obesity and T2D, findings which correlated with IR. Disturbances in mitochondrial function in muscle and other tissue may lead to lipid accumulation, which in turn may cause or aggravate IR 1 2 3. 2.4.3 Adipose Tissue as an Endocrine Organ The adipocyte, traditionally understood as a passive fuel storage depot, is now thought to also play an active role in body weight and carbohydrate regulation systems through its secretion of a number of products such as leptin, interleukin-6, angiotensin II, adiponectin, resistin and TNF- alpha1 2 7, which are referred to as adipokines. These factors influence distant tissues, and thus AT operates as an endocrine organ. As adipocytes respond to metabolic, hormonal and neural stimuli by releasing various factors an alternate hypothesis has been suggested to account for this role which has been termed the endocrine hypothesis115. Adipocytes are likely to exhibit differences in their endocrine roles according to their size 1 1 5 , or to their location112. Some of the factors secreted by adipocytes are discussed below. Leptin is an adipocyte-derived peptide hormone having effects on satiety, energy expenditure and neuroendocrine function. Its major site of action is the hypothalamus, although it may also have important effects on peripheral target 37 cells of the pancreas, liver, muscle and AT". Increased AT levels are associated with increased leptin levels which act to reduce food intake. In general, obese individuals are leptin-resistant, and it is possible that leptin exerts its modulating effect on metabolism and satiety through actions at the hypothalamus32 1 2 8 . Circulating levels of leptin show a higher correlation with subcutaneous adiposity than with visceral adiposity, and leptin secretion is greater from subcutaneous than from visceral AT. This suggests that subcutaneous AT may have a greater effect on the regulation of appetite via its secretion of leptin. Leptin may also have insulin-sensitizing effects on skeletal muscle and other tissues, although most human studies have not shown an association between leptin and insulin sensitivity17. Adiponectin is an adipocyte-derived peptide, whose levels correlate with insulin sensitivity. Animal studies suggest that treatment with adiponectin may reduce IR28. The role of adiponectin in insulin sensitivity and its site of action remain controversial129. Resistin is the most recently discovered adipocyte peptide hormone, with an association to IR in mice, although it is uncertain whether a human form of the hormone exists129. Abdominal obesity is associated with atherothrombotic events. Plasminogen-activator inhibitor-1 (PAI-1) is a major circulating inhibitor of thrombolysis, and is produced by visceral and subcutaneous adipocytes. Levels of secretion of PAI-1 are increased in obese individuals, although the few studies to date have been 38 conflicting as to the relative contribution of the subcutaneous and visceral AT masses, and further examination is required112. This peptide provides a possible link between central obesity and the thrombotic component of cardiovascular risk. TNF-alpha is a cytokine produced by adipocytes, and suggested to be a factor in the IR of obesity although its role remains controversial17. Its actions on adipocyte function include an inhibition of lipogenesis and an increase of lipolysis32. AT TNF-alpha messenger RNA is positively correlated with obesity, central obesity and fasting insulin, and negatively associated with glucose disposal17. Differences exist in TNF-alpha levels between visceral and subcutaneous adipocytes112. 2.4.4 Neurohormonal or Hormonal Regulation of Lipid Metabolism Abdominal adiposity and metabolic disorders such as IR may not be causally linked but may arise from common antecedents. The initiating problem might be IR, or reduced insulin secretory response that leads to central or visceral adiposity112. Other factors which have been related to a central AT distribution include hormonal levels such as the sex steroids and glucocorticoids130. Bjorntorp130 has suggested that visceral obesity is a maladaptive response to stress, involving activation of the hypothalamic-pituitary-adrenal axis and dysregulation of glucocorticoids and gonadal steroid levels131. More recently others have expanded on the role of the central nervous system, hypothesizing that obesity and T2D occur together because carbohydrate metabolism and body weight regulation rely on common hormonal (insulin, leptin, adrenal steroids) and CNS (hypothalamic amine and neuropeptide) signaling systems132. In this model 39 the hypothalamic centers of the brain coordinate energy homeostasis through regulation of hunger, satiety, energy expenditure, and hormone regulators of substrate interconversion, storage and mobilization (e.g. insulin). 2.5 Physical Activity and Insulin Resistance Physical activity has long been considered therapeutic in those with diabetes but the accumulation of substantial scientific evidence about its role in the prevention of IR and consequent diabetes has been more recent. The following two sections will discuss the evidence of the role of physical activity in improving insulin resistance. Physical activity is defined as physical movement that is produced by contraction of skeletal muscle which increases energy expenditure133, and is often the measure in epidemiological studies. Exercise, a type of physical activity, is defined as planned, structured, and repetitive movement done to improve or ..maintain physical fitness133, and is used in the description of intervention studies. 2.5.1 Physical Activity and the Prevention of Type 2 Diabetes A recent review of prospective observation studies found a marked risk reduction for T2D in physically active versus sedentary individuals, with most of the studies demonstrating risk reduction in the order of 30-50%. Similar benefits of physical activity were seen in women and men, and in older and younger groups. The benefit of physical activity was evident after adjusting for age, BMI and other potentially confounding factors35. One of the more recent prospective studies specifically examined the dose-response relationship of physical activity and the prevalence of T2D in a predominantly Caucasian cohort134. A lower incidence of 40 T2D was found with higher levels of physical activity ranging from regular walking or cycling to recreational to sporting or vigorous activities. A clear dose-response effect persisted after adjustment for age and BMI. Recent results from another prospective trial of physical activity and mortality in men with T2D demonstrated that low cardiovascular fitness and physical inactivity independently predicted all-cause mortality135. These results suggest that the effect of physical activity in diabetes extends beyond the limited assessments of insulin action and glycemic control. Large scale intervention trials have been consistent in demonstrating the benefits of increased levels of physical activity on the prevention of T2D. In a non-randomized 5 year study of Swedish men (181 with impaired glucose tolerance, 41 with newly diagnosed diabetes, age 40-49 years), a structured diet and exercise program was compared to a non-intervention group136. In the intervention group, the relative risk of T2D was significantly reduced to 0.37. In subjects with diabetes at baseline, approximately 54% no longer had glucose levels diagnostic for diabetes. Three randomized, controlled studies in individuals with impaired glucose tolerance (IGT) have confirmed these findings. The Da Qing IGT and Diabetes study in China followed 577 men and women with IGT (mean age 45 years, mean BMI 25.8 kg/m2). Subjects were randomized to a control group or one of three intervention groups: diet only, exercise only, or diet plus exercise. After 6 years of 41 follow up and adjustment for baseline BMI and fasting glucose levels, the reduction in risk of developing diabetes was 31% in the diet only group, 46% in the exercise only group, and 42% in the diet plus exercise group37. The Finnish Diabetes Prevention Study randomized 522 overweight (mean BMI 31 kg/m2) men and women (mean age 55 years) to a control group or an intervention group, with a mean duration of follow-up of 3.2 years. The intervention group received individualized counseling aimed at decreased dietary saturated lipid and body weight and increased dietary fibre intake and physical activity. During the trial the risk of diabetes in the intervention group was reduced by 58%38 as compared to the control group. The Diabetes Prevention Program randomly assigned 3234 multiethnic men and women with impaired fasting and post-load glucose levels to placebo, metformin, or a lifestyle modification program. The lifestyle modification program had the goal of a seven percent weight loss and at least 150 minutes of physical activity per week. The mean age of the participants was 51 years, and the mean BMI was 34 kg/m2. After an average follow-up of 2.8 years, the lifestyle intervention reduced the incidence of diabetes by 58%, and metformin reduced the risk by 31%, compared to the placebo group36. The data from these studies suggest that lifestyle changes including increased physical activity, dietary changes and weight loss prevent or postpone the onset of 42 T2D. Furthermore, the reduction in risk of developing T2D was seen with 150-210 minutes/week of moderate intensity activity and modest weight reductions (4.2 kg in the Finnish study, 7% loss of initial weight in the Diabetes Prevention Program). At present, data are limited as to the minimal or optimal intensity, frequency and duration of exercise for reducing the risk of T2D 3 5 . 2.5.2 Exercise Training and Insulin Resistance A single bout of exercise enhances insulin-mediated glucose disposal in normal subjects137, in insulin-resistant first-degree relatives of T2D 1 3 8 , in obese subjects with IR, and in subjects with T2D 1 3 9 . This improvement is of short duration, lasting for approximately three days 1 3 9 1 4 0 . Whether the magnitude of this acute effect is affected by exercise mode, duration or intensity is not clear1 4 1. Repeated exercise sessions, i.e. exercise training, also produce beneficial effects pn insulin-stimulated glucose disposal138. This has been seen in normal glucose-tolerant humans 1 4 2 - 1 4 4, in obese non-diabetic subjects145 and in patients with J2Q146,147 T n e m a g n j t U C | e 0 f improvement in insulin-stimulated glucose disposal as a result of endurance training is 11-36% in both healthy and insulin-resistant populations148"150. Endurance training programs which have induced such changes include the following characteristics: exercise intensity of 50-80% of maximal capacity, duration of 45-70 minutes per session, a frequency of 3-6 times per week, for a 6 week to 6 month time period 1 2' 4 1' 1 4 0' 1 4 6' 1 4 9' 1 5 1 . 43 In the trained state the effect of an acute bout of exercise may be more pronounced. Insulin-stimulated glucose disposal was shown to increase 22% after the first exercise session, with a 42% increase after 6 weeks of exercise training, suggesting an effect of training in addition to the effect of an acute exercise bout138. This is supported by a study in which subjects performed one-legged cycle training for ten weeks. Insulin-stimulated glucose uptake across the trained leg muscle bed 16 hours after the last exercise bout was increased. Subsequently glucose clearance 16 hours after a single exercise session was measured in the untrained leg, with no increase in glucose clearance rates observed140. From these data the authors concluded that increased insulin action in skeletal muscle was due to adaptive mechanisms elicited by repeated, regular physical exercise, and it could not be invoked by a single exercise bout. Although it is generally accepted that the effect of training on insulin sensitivity is rapidly lost, it may be that training potentiates the acute effect of exercise on insulin sensitivity141, i.e. training would provide an additional effect beyond that of a single exercise bout on increased insulin sensitivity. In order to optimize insulin sensitivity, it is advisable to perform exercise on a regular basis to take advantage of both training and acute bouts of exercise. Characteristics that would constitute a minimal or optimal exercise strategy have not been investigated, and exercise program design including type, amount and duration of training, as well as modulatory factors such as degree of prior IR and insulin deficiency, and concurrent changes in body composition have not been well studied. The limited data available are discussed below. 44 Several studies have compared training programs of differing aerobic intensities, with variable results. In healthy young men and women, a high intensity training program, but not a low intensity program, resulted in an improved response to an insulin tolerance test152. Seven consecutive days of higher intensity exercise, but not lower intensity exercise, improved insulin response to an oral glucose challenge in obese men, although a group of obese men with type 2 diabetes did not respond to either program153. Another investigation found improved glucose disposal during a euglycemic clamp after both a moderate and a more intense training program in subjects with impaired glucose tolerance. No difference was evident between the training regimes149. In general, studies demonstrating improved insulin sensitivity utilized training programs with intensities in the range of 50-80% of maximal capacity. These few studies with negative results used minimal exercise interventions, for example: very low intensity exercise (40%:20% exercise:rest intervals)152, or limited duration of training (7 consecutive days)1 5 4. It may be that improved insulin sensitivity can be achieved with a range of exercise intensities and durations155. Although the majority of training programs have also elicited improvements in aerobic capacity 1 2 , 1 4 6' 1 4 9 , 1 5 6" 1 5 9, changes in insulin responsiveness are evident either without concomitant changes in cardiorespiratory fitness 1 4 0 , 1 6 0 or are statistically independent of fitness changes161 Thus, improvements may be elicited without necessarily increasing cardiorespiratory capacity, and other factors such as total amount of work performed may be a primary determinant of improvement in insulin sensitivity141. 45 Skeletal muscle is the largest insulin-sensitive tissue and dominates clearance of oral and intravenous glucose load 1 6 2. Maintenance or increase of muscle mass could increase glucose storage area and facilitate glucose clearance from the circulation. Resistance training, via its hypertrophic effect on skeletal muscle mass, may favourably impact insulin sensitivity. This may be particularly relevant with inactive older groups as there is a steady decline of muscle mass past the age of 50 years1 6 2. A 16 week resistive training program increased glucose disposal during a hyperinsulinemic-euglycemic clamp by 24% and 16% in healthy men 1 6 3 and postmenopausal women1 6 4 respectively. In subjects with T2D, response to an oral glucose challenge was improved after a single resistance training session 1 6 5 and after an 8 week training program166. A 23% improvement in glucose disposal during a euglycemic clamp was reported after resistance training in men with impaired glucose tolerance159. These data suggest that improvements in insulin sensitivity after resistance training are similar to those achieved with aerobic exercise training. Further investigation is warranted, utilizing direct measures of insulin sensitivity (i.e. glucose clamp studies), and in broader study populations. Training programs that combine the two modalities, aerobic and resistance training may be most advantageous because they combine different mechanisms of action. Little evidence is available in this regard. It is notable that resistance training was included in the lifestyle changes of the Finnish Diabetes Prevention Study; the lifestyle intervention group demonstrating a 58% reduction in risk of developing diabetes38. One randomized controlled trial compared an endurance training program with a combined endurance plus resistance training 46 program in sedentary men with hyperinsulinemia. The combined program resulted in greater reductions in glucose and insulin levels than the endurance program167. Although the data did not assess insulin sensitivity directly, they suggest that a combined program may have greater potential to reduce IR than either exercise regime alone. 2.5.3 Mechanisms by Which Exercise Training Improves Insulin Sensitivity Skeletal muscle is the primary tissue responsible for glucose disposal after a glucose or insulin challenge, or during an exercise bout. Increased uptake and utilization of glucose during acute exercise are regulated by increased blood flow to the working muscles, and increased glucose transport. The latter is considered to be the rate-limiting step in glucose utilization during exercise168. Glucose transport into the muscle cell is regulated by insulin and insulin-like factors through the activation of a series of intracellular proteins. These steps result in the translocation of the glucose transporter protein GLUT-4, and it is believed that it is the magnitude of the GLUT-4 translocation that determines skeletal muscle's capacity to transport glucose. Glucose transport is also stimulated by an insulin-independent mechanism which is activated by contractions and other physiologic stimuli169. The enhancement of glucose transport after acute exercise can last for several hours, and seems to be related to the degree of glycogen depletion during the preceding exercise. It is believed that GLUT-4 translocation from exercise is drawn from a separate intracellular pool than GLUT-4 translocation from insulin stimulation168. 47 In T2D, insulin stimulation fails to induce normal GLUT-4 translocation to the muscle cell membrane, so that an impairment of insulin-stimulated glucose transport is felt to be responsible for the resistance to insulin's a c t i o n 1 7 0 . Yet the contraction-stimulated pathway of glucose transport appears to be normal. In humans with obesity or T2D, defects in insulin signaling are thought to be involved in IR. Exercise training is associated with increased GLUT-4 protein expression across non-diabetic, IR and diabetic groups. The enhanced GLUT-4 expression is accompanied by modest increases in insulin-stimulated glucose disposal during a euglycemic c lamp and in reduced glucose and insulin responses during a glucose tolerance test. These changes do not fully explain the increased insulin sensitivity, suggesting that other factors may be invo lved 1 6 8 . Other work in insulin-resistant subjects has demonstrated that there is an increased response of enzymes involved in glucose phosphorylat ion and o x i d a t i o n 1 3 8 , 1 6 9 with exercise training. Exercise training has a variety of other morphological effects which may play a role in improving insulin sensitivity. Endurance exercise training increases muscle capillary density and increases the conversion of type lib to type Ma fibers. Type Ha fibers have a greater capillary density and a higher concentration of glucose transporters and are more insulin responsive. These morphological changes in muscle have been associated with changes in fasting insulin and glucose tolerance and insulin-stimulated glucose d isposa l 3 9 . A n increase in muscle mass 48 may also improve IR by increasing glucose storage as discussed earlier. Exercise training also has indirect effects on IR via changes in body composition, including decreasing body adiposity and maintaining skeletal muscle mass. The effects of exercise on body composition and subsequent effects on IR are addressed in the following sections. 2.6 Weight Loss and Exercise Interventions: Relationship of Insulin Sensitivity to Body Composition The role of the relationships of different sites of AT deposition (visceral, subcutaneous abdominal, and intermuscular) with metabolic variables has been evaluated in cross-sectional work; abdominal obesity is a clear correlate, although debate exists as to the relative roles of visceral vs subcutaneous abdominal AT. Newer evidence suggests that ectopic lipid may be an early defect in the development of IR. Intervention studies, by evaluating change in IR related to the change in AT patterning, can help elucidate the role of body composition characteristics in IR. A variety of interventions have been used including: diet-induced weight loss 4 1- 1 7 1 - 1 7 6, exercise-induced weight loss 4 1 , 1 6 1* 1 7 1 , 1 7 2 , 1 7 4" 1 7 7 , and exercise without weight | O S S 4 1 . 1 4 8 1 5 1 1 7 8 . 1 7 9 Most studies used obese, normoglycemic groups and involved substantial weight losses (approximately 7-16 kg). Regional AT and muscle characteristics were assessed using either MRI or CT measures. Some evidence is available linking glycemic responses to body composition changes, and limited data is available linking euglycemic clamp 49 assessments to body composition changes. The studies are outlined in Table 2.6, and will be discussed below. 50 CO cz o -4—' c CO CD CO CO o D) CD l-o 0) CO CO l -Q •4—• CO l b > "to c CU co CO cz 0 E J Z l ie CD N E o CN CN o 0 cz ro Z3 T J JD ro < o CO > < c cu E CO CO cu CO < CO -*-« o 1.2. CO ro CO o CL CO CU CO o o Z J O) in CM co o Ji I-< CO ! JD o Z J ^ E o CN E ~ r^ - o CO CO o E o CD CO o. E ro o o 'E cu o >% h - CO t— co Z J _ CD lO ' 0 "co ro CL I-9 ca cn ,9, CD CD CD -r-CN T- CO CD CO E o CD -cr o cu O) ro m Of o CO co 1- ^ CO - » 2 cu T J > .CO ro o co > * ^ CN E o CO co" o -> r--00 CO W C D , o CD •9 CD E ro _: N ro cu oo CO CO 8 CD 2 LL co~ cu co cz o CL CO 0 CD m CN ^ C N •<- CO co c "•4-» .—-52 CD J5 CN t != 5 I E c ro 0 cn ro CN ^ , co co H _ CL h-oo 2 o cn t- _ r 0 co C N F2 ro 0 t 0 I 5 CM CO ES? O co CO CD 1-5 G-O -3" 0 cn co CO I m o+ CO o CM PP CO •<- ^ CO -S» 5 0£ CZ 0 0 0 CD £ CD 0 ^ 0 _: _ J CO cz ' £ CD •2 ii' o IL co *—-3 I—*>-55 ro n *- cz • J cu fc co CD > J2 *§ cz 0 0 i i -e o CL o J Z "> co cu § ? Si? 0 o in i 0s T _ J - o CO -—-CN_ CT) cn E o LO CN CD CD —» 00 >. CD CN CL| E • J5 O) o CQ t--o" co -> or co ^5 co o co o ° o Ql CN TJ) CM CM CD - » E CD CN 00 -5» co E o T— lO CO CD m o" ^ 5. 0 3 co o = , ro cb 0)0 R" 0 0 n F CO 4— CD LC - ° ^ 0 CN O CO — O —i CO CM -I—' '-*—' 'co c CD to to cz 0 ) o to 3 E JZ CO lie to o CD tz is Z3 |xi CD CO L_ CD O (O > TJ CD ZI c I CD i CM _CU XI CO c 'c CD c: CD E CO to CD CO (O < w o CD x? 3 CO CD CO CD X LU CD CO CD 2 CD 2 < -3£ CO R 5 ^ E CN O IO £ ' i o T » 2 CO 0 £ « co cn Q CD i -CD oo co co o o 3 IL c o ^ 8 to > CO < CD I D -<fr CN rvi CO LO 5-CD O -ST CO CO <D t CO ° CO o CO c o ^ ^ 8 CN < s W c S 3 CO <2 <3 .£ ^ E ~ CN <P X ) 2 0 CD co o CD » co CD -* o CD * . ! ? C N E * ° to 8 £ <i> 0 t CD 2 Or CO T— c CD CD i ig 9- <-> sz co to 0 9 c? Isz CD O zl £ E JzT<N I ii •g ^ *3 Ec? o S-LO CO i_ >> 00 LO CD CO CO CN CO CO CO Z2 a. i= g> C 0 « ~ o 5 CD H CD & E, CD co o >>_ o or co C N § CO 9J <o £ JL CD C X I _ to C L CD o o CO ~ CD co co 0  co o ro CO 3 o 00 <" ^ i o E -» 0 Ec? CN CD CN E Co cn csi CO i i -—> ^ ° CD co CO y oo £ CO o 0 0 co ro ro + =5-> S 0 P! ro o —> 0 CN a> So % CO 0 s. .9-0 o o to J = 0 co co 0 o z CO ro CO 3 o CO 3 -> 0 CN E C ° ° tr* CD (SJ CM CO 0 — CJ co E. CO 1 X 5 "to co 2" 0 $ CO 1= , r o + I — "O O j"?0 °E~ C CM 0 $ _ C N c n o ro o -3 0 CN cn •«* r IL c 'E E ^ co — ^ ro co 0 c to CM > r^ . < 8 c CM oci CM CO LO 1^  oo CM eg E o CM LO % E Q. o o <N CO T-o 3 h = 10 12 CO ro or 0 to o CO S B —• ° DT ro CM E ^Co ^ LO CN E _ O vP co oo LO oo ' ' CO CD 0 c ro co 1 - co co o i_ ^ ° L co cb E 0 Is-1 s > _ o — CO C N CO o c CO .S "CO j^. 0 CO 0 •c e s ^ CD 5 1 o | -2 o c5 co a 8 0 C g t ro o CD -c S O o o S 0 ~-0 c 5 2 g < to o C L co o to *~ c •o y -5 io S o 8 I s 2- ro !s | c^  e g v. O =3 .3 O S c c 0 la if-52 CD O C L c O E ^ K TO S O O Q.-Q « o S o c - CO ^ CO c O P ^ 0 c o 0 CO o . £ § 3 52 co cz 0 CO cz CO cz o CO 3 E JZ cn l ie CO o 0 c ro 1 CO I JD ro CD O c CO c c CO E O) CD 2 CD O CO > TJ CD r» c *•*-» c o C D i CM 0) JD CTJI CO c o "c CD c! 0) 05 1— •4—' (D CO o CD X 1X1 cz 0 E CO CO 0 CO CO < £ o 0 J D 1 3 co I f 0 LL S H | < C N V ro L O 0 cz CO CO E > CD < + o 4 oo ^ T - CZ CD CM CO OQ ( j . JD CO o 0 L O CD S "^ « o •^f CM Si ro 0 -CO o CO o Q i CM CD •3-oo CD II TD CM ^ ^ Si CD cB E-5 o ° * CM CD CO oo CO S CO E o CD O 0 CO ro OQ o+Q CD CD CM O JD O \— CU ro co — I l 0 ? i - r--0 cn cz cn 3 T— o ^ ro E Co T- co L O s i •c— oo L O E o CD oS N E o CO CO 4 0 0 0 O CO CZ co ro CO CO CD m J2 2 £ o 00 ^—' I— _ L O £ CT) 3 oo £ L I— ro CD — a ro T I 0 IS 0 ro CO o CD-CO T J 0 CO o o i t •c— Z) X CM cz g ro 3 CZ 0 < G ) CM 0 TO CO - O CN JD CM CL E ro o h-O CM ai 0 o co CO F CO ^ CZ S CD o s-° ~ 0 ro o 2 = o ro cz X cz .2 C° •s °^ co J^-c 0 , 0 0 | CM 0 TO 0 CO " CN CO ™ .<2 Q . = CO =f 5 0 E OQ 12 Of o o V- E i -CD O o o o CM ro 0 EC? r-. co E Co CO S. CO CO 0 TO CO CN JD - 2 0 03 CO - co Of-g a> > CD o CN CM O 0 0 CD 2.8 CO o c s co-S Q . L -8 CD coi» CO 0 o co | CD s O "O "7 -2 0 CO § $ § p f s O CD h- O) fei O " 3) Z-co . P 0 •S D) 0 C o ro c CO o CO •Si 0 O c .O D) S co t s I? 0 .o 5 s H i o -EE p •p 0 3 CO c c 0 O Q. h -2 O cj s" ? •a .£ •S » CO JO E; 0 T?S O 0 J3 CO O l | OQ pj 53 2.6.1 Change in Abdominal AT In weight loss interventions, induced by diet and/or exercise, changes in the visceral AT have ranged from 17-44%, and changes in subcutaneous abdominal AT have ranged from 15-33%. Within each study and expressed relative to its original size (i.e. as a percentage), the loss from the visceral AT was greater than from abdominal subcutaneous AT. Expressed on an absolute basis, losses from the visceral AT range from 19-93 cm2, and losses from subcutaneous abdominal AT range from 26-145 cm2. Within each study, and compared on an absolute basis, losses from subcutaneous abdominal AT are consistently greater than losses from visceral AT. Few studies have used exercise interventions without an accompanying body weight loss. In a group of predominantly middle-aged men with T2D, response to an aerobic training program resulted in a relative and absolute greater loss from the visceral AT (76 cm2, or 48%) than from subcutaneous abdominal AT (41 cm2, or 18%) on MR scanning148. In response to a resistance training program, postmenopausal women of normal weight demonstrated small changes in both visceral AT (14 cm 2 or 10%) and subcutaneous abdominal AT (18 cm 2 or 6%) on CT scans 1 7 8. Therefore, when expressed relative to its original size, visceral AT losses appear greater, but when expressed as absolute changes, similar or larger losses occur from subcutaneous abdominal AT which is typically a larger depot than visceral AT. This pattern is consistent when larger weight losses are induced in response to either diet or exercise programs. When the intervention consists of exercise training without weight loss, small changes are evident in AT proportions, 54 although the data are limited. These studies were not designed to determine whether there was a preferential reduction in visceral or abdominal AT, so no data are available on that comparison within individual studies. Published guidelines on the prevention and treatment of obesity40 have concluded that physical activity alone reduces total and abdominal AT modestly, or not at all. Yet a recent investigation41 found that abdominal and visceral AT were decreased in obese men after exercise training without concurrent weight loss. Improved insulin sensitivity accompanied the abdominal AT loss. These results support the recommendation that exercise alone without caloric restriction is an effective strategy for obesity reduction181, and that reductions are seen in the most deleterious abdominal depots. Furthermore, exercise-induced reductions in visceral and subcutaneous abdominal AT occur without a corresponding change in waist circumference182, suggesting that external anthropometric measurements may be misleading. Evaluating changes in regional adiposity relative to changes in glycemic response or IR would provide insight into their relative contribution to the pathogenesis of IR. Several investigations (Table 2.6) related the regional adiposity changes to changes in glycemic responses to an oral challenge. Some reports found that change in glucose response was related to the change in visceral A T 4 1 , 1 7 4 , and the change in fasting insulin was related to the change in subcutaneous abdominal A T 1 7 4 . Another study reported no relationship between change in glycemic responses and change in body AT depots in obese women1 7 2. Two studies 55 assessed IR directly by euglycemic clamp, and found changes in IR were correlated to changes in visceral AT 4 1 , 1 7 1 , 1 7 2 . Although exercise training as a single intervention increases glucose disposal, little work has evaluated the associations with body composition changes under these circumstances. 2.6.2 Change in Muscle Characteristics Changes in muscle characteristics occur in response to exercise and/or diet-induced weight loss. When weight loss is induced by diet alone, overall muscle mass is decreased 4 1 ' 1 7 1 1 7 2 1 7 4 along with a decrease in low attenuation muscle 1 7 1 , 1 7 2. If weight loss is accomplished through diet and exercise programs, then skeletal muscle has been reported to exhibit no change 4 1 , 1 7 4; to increase cross-sectional area 1 7 8 , 1 8 3 ; to increase normal density component and decrease in low density component on CT scans 1 8 0 , 1 8 3; or to decrease intermuscular AT as assessed by MRI 1 7 2 . Few studies have evaluated changes in muscle characteristics (density, or attenuation on CT or IMCL by MRS) in conjunction with directly assessed insulin sensitivity. The only report to date found that a substantial weight loss (15 kg) resulted in a decrease in muscle cross-sectional area and low density muscle, but these changes did not correlate with changes in IR 1 7 1. The role of exercise as an intervention in attenuating muscle lipid accretion, and potential associations with IR have not been evaluated. 56 2.7 Summary IR is associated with risk for metabolic syndrome, T2D and atherosclerosis, although initiating factors in its development are not clear. Abdominal obesity, and in particular the visceral component of abdominal obesity, has been closely associated with the negative metabolic characteristics of IR. Several possible mechanisms linking IR and regional obesity may be at play, including abnormal lipid metabolism leading to elevated FFA levels, ectopic lipid accumulation, and adipose-tissue derived hormones. Alternatively, abdominal AT accumulation and/or IR may both be caused by another system disturbance, such as disrupted neuroendocrine function. Although abdominal obesity is strongly implicated in the negative health consequences associated with IR, there are gaps in our knowledge in the area of regional AT, IR and exercise-induced improvements. Available evidence suggests that exercise with and without weight loss is associated with reductions in both visceral and abdominal subcutaneous AT, although the question of whether visceral AT is preferentially reduced has not been the objective of published work. As exercise is a common feature of prevention and treatment for obesity and IR, it is important to clarify whether exercise interventions are associated with reductions in abdominal subcutaneous and/or visceral obesity. Recent work suggests that not only visceral AT but also the subcutaneous component of abdominal obesity may have a deleterious impact on IR. There is also emerging evidence that ectopic lipid in other tissues, such as 57 skeletal muscle, is also a contributor to IR. Few data are available with respect to how intermuscular lipid changes after exercise training, and how these changes may be related to improved insulin sensitivity. Physical activity has a beneficial effect on insulin sensitivity in normal as well as insulin-resistant populations95,184. Exercise training may result in moderate decreases in AT, including abdominal adipose tissue, and increases in muscle mass 1 5 5 , 1 8 5 . These body composition changes may be contributing to the increased insulin sensitivity, but part of the training effect appears to be independent of changes in body composition1 5 5 , 1 8 6. 2.8 General Hypotheses This comprehensive review of literature reveals important questions that remain unanswered. Some of these form the basis of this study and general hypotheses are outlined below. The specific and null hypotheses are detailed after the chapter on study methodology. In a group of postmenopausal women with T2D: 1. a program of aerobic plus resistance training will be more effective than aerobic training alone in improving insulin sensitivity. 2. both aerobic plus resistance training and aerobic only training will induce similar losses in abdominal AT. Losses from the abdominal subcutaneous AT will be similar to or greater than those from the abdominal visceral depot. 58 3. a combined aerobic and resistance training program will induce changes in skeletal muscle size and density to a greater extent than an aerobic only program. 4. improvements in insulin-stimulated glucose disposal after aerobic plus resistance training and after aerobic only training will be related to changes in abdominal AT, as well as to changes in skeletal muscle characteristics. 59 Chapter 3: Methods 3.1 Study Design 3.1.1 Subjects and Sample Size Subject recruitment was initially through the Diabetes Teaching and Training Centre of St. Paul's Hospital. Basic demographic and medical criteria were used to generate a list of potential subjects from the database of all patients of the Diabetes Centre (approximately 10,000)(Figure 3.1). This search generated 278 cases whose charts were reviewed for study eligibility. Patients eligible as per inclusion and exclusion criteria (Table 3.1) were contacted by mail (146 cases). Thirty-three respondents underwent telephone screening for compliance with inclusion and exclusion criteria, which resulted in 21 eligible individuals who completed and signed informed consent (Appendix A), medical exam and exercise stress testing. A secondary recruitment process was through advertisement in local newspapers and a hospital and university newsletter, which generated 19 responders, 11 of whom were eligible for the study. As a result of both the mail-out and the advertising, 32 individuals were eligible for the study, 2 were excluded after medical exam, and another 2 dropped out of the study due to recurrent infection illness. Twenty-eight individuals completed the study. Sample size calculations were based on published effect sizes for change in muscle cross-sectional area (0.49-0.55 1 7 8 1 8 3 ) and changes in insulin sensitivity (0.47-0.77140'149'150'159) after exercise training. Using the variables of effect size of 60 Figure 3.1 Flowchart of Patient Recruitment Diabetes Centre Database search M 0.000 patients) Search criteria: 1 s t visit> 1995 Age: >50yrs. <80 yrs. BMI >27 Onset of diabetes >1995 Activity = sedentary Angina *yes Therapy * insulin Ml * yes Gender = female Residence = Greater Vancouver Diabetes = type 2 area 278 cases Chart Review excluded: Heart disease Smoking Insulin therapy Diabetes > 10 yrs Change address Metabolic disease Severe musculoskeletal limitations Language, cognitive barriers Highly active 146 cases Mail out recruitment letters -33 responders 21 eligible after phone screening Ads placed in university, hospital and Canadian Diabetes Association newsletters and in Vancouver community newspapers ~ 19 responders 11 eligible after phone screening 32 cases 2 ineligible after medical exam Usual care group N=8 Ae + RT group N=11 1 transfer in from Ae + RT group 1 transfer - relocate out of area - complete study as control 9 complete study 10 complete study Ae only group N=11 2 withdraw -infectious illness 9 complete study 61 0.50, desired statistical power of 0.85-0.91, and significance level of p=0.05, a sample size of 12 subjects for each of three groups was calculated using Datasim software (Lewiston, ME) 1 8 7 . This statistical power is the probability of correctly rejecting a false null hypothesis, i.e. to be able to detect if a difference exists between the groups. Table 3.1 Inclusion and Exclusion Criteria . Inactive lifestyle (less than 20 min . Inability to participate in a structured of moderately vigorous activity, exercise program due to orthopedic twice per week, over the past year) limitations or scheduling problems . Ability to give informed consent 3.1.2 Randomization and Intervention Eligible subjects were randomized by use of a table of random numbers188 to one of three groups. 1. Usual Care - Subjects were followed on the usual basis by their caregivers. This included any regularly scheduled visits to their physician. No changes in therapy occurred in this group during the course of the study. 2. Aerobic plus resistance training (Ae + RT) group - Subjects attended a 75 minute exercise class, three days per week, for 16 weeks. Each class consisted of a warm-up, aerobic exercise, resistance exercise, and a cool-down phase. Aerobic training consisted of individually prescribed exercise intensity at 60-75% of heart rate reserve (difference between maximal and Inclusion Criteria Exclusion Criteria Female Postmenopausal Age < 75 yrs. Diagnosis of type 2 diabetes mellitus within the past 10 years Diet control or oral agents Waist circumference > 90 cm Current smoking habit Current insulin therapy Objective or subjective evidence of coronary artery disease Metabolic disease (thyroid, renal disorders) Unstable body weight 62 resting heart rate during exercise tolerance test)133, for 30 minutes on a variety of equipment including: treadmills, stationary bicycles, recumbent steppers, elliptical trainers and rowers. The RT program consisted of exercises to train most large muscle groups (leg press, leg curl, hip extension, chest press, latissimus pull-down) for 2 sets of 12 repetitions on stack weight equipment (Apex, Saanichton, B.C.). The acclimation technique133 was used for determining resistive training loads, as follows. The subject started with a light load and monitored for 10-12 repetitions with the aim of working at a rating of perceived exertion of no greater than 3-4 on the Borg scale ("moderate" to "somewhat hard"). The resistance was then progressed each 1-2 weeks while maintaining appropriate technique and a Borg rating of 3-4. After a two week learning phase, training loads were recorded until the last week. Increase in training load was taken as the difference between final load and the initial load. 3. Aerobic training (Ae only) group - Subjects attended a 75 minute exercise class, three days per week, for 16 weeks. Each class consisted of a warm-up phase, aerobic phase, and a cool-down phase. Aerobic exercise intensity was prescribed as for the Ae + RT group, and the equipment utilized was the same. Low impact, low intensity, dynamic movement replaced the resistance training component in the combined training group (Figure 3.2). Published values for the caloric cost of resistance training have been estimated to be 0.1 kcal/min/kg 1 8 9 1 9 0 . Low impact aerobics is an activity which incurs a similar energy cost 1 7 4 ' 1 9 \ thus low-impact dynamic 63 aerobic movements replaced resistance training in the Ae only group, resulting in a somewhat longer warm-up and cool-down phase. The replacement of resistance training with low impact dynamic movement was done in an effort to minimize differences in overall caloric expenditure. Figure 3.2 Exercise Group Class Structure Aerobic traininc class 20' dynamic warm-up 30' aerobic training 25' dynamic cool-down = 75' total Aerobic + Resistance Class 12' dynamic 30' aerobic training 20' resistance 13' cool- = 75' total warm-up training down 3.2 Measurements 3.2.7 Dietary Assessment Subjects were asked to continue with their current diet and physical activity pattern outside the study protocol. Dietary records were assessed by three day food records for caloric, macronutrient and micronutrient intake, before and after the study and analyzed with the Nutritionist IV software (First Data Bank, San Bruno, California) by a dietitian at St. Paul's Hospital. 3.2.2 Peak V02 Determination Peak oxygen uptake (V02) was determined before entering the study, and at the end of the 16 weeks. Respiratory gas analyses were carried out during a progressive Naughton protocol treadmill test133 to voluntary exhaustion. Peak V 0 2 64 was determined using a Beckman metabolic cart (Sensormedics, Yorba Linda, California). Twelve-lead ECG analysis continued throughout the test. 3.2.3 Hyperinsulinemic-Euglycemic Clamp Studies Hyperinsulinemic-euglycemic clamp studies were performed according to established methodology192. Subjects were asked to report to Vancouver Hospital and Health Sciences Centre after a 12 hour overnight fast, and to refrain from strenuous physical activity for the preceding day. In the 30 minutes prior to the start of the euglycemic clamp, four blood samples were taken to measure basal insulin and glucose. At time 0 the euglycemic clamp study was started and continued to 180 minutes. Regular human insulin (Humulin R, Eli Lilly, Indianapolis, In.) was infused at a constant rate of 40 mU/m2/min via 18 gauge catheter inserted into an antecubital vein. Glucose levels were allowed to fall from fasting level to a euglycemic level (aprox. 5.5 mmol/L). Blood levels were maintained, or clamped, at this level by a variable glucose infusion rate for the remainder of the study. Glucose infusion rates were adjusted according to 5 minute interval sampling of arterialized venous blood from a contralateral hand vein. Blood samples were collected in heparinized syringes. Plasma glucose was analyzed immediately using a YSI Glucose Analyzer (Yellowsprings Instruments, Yellowsprings, OH). The remaining blood was placed in pre-chilled test tubes containing aprotonin (400 KlU/ml) and EDTA (1.5 mg/ml) and centrifuged at 4° C. Samples were stored in a -70 °C freezer. Subjects in the exercise groups underwent their repeat glucose clamp at 60 hours after the last exercise class. Results of the euglycemic clamps are expressed as the glucose infusion rate 65 averaged over the last 60 minutes of the clamp. A graphical representation of a euglycemic clamp study is presented in Figure 3.3. Within lab values for intra-subject variation in glucose infusion rates are not available, published values of the intra-individual coefficient of variation range from 2.4-17%69. 6 I IV ro or 5 II c ^ 4 Q, § » to £= O o 3 CD O k Euglycemic Clamp Study insulin infusion (40 mU/m /min) Glucose disposal - last hravg = 2.78 mg/kg/min 180 160 140 120 ^ cu 100 0) tn — o 5 « o> 80 CD E •a 60 M 40 20 \ * <£> «$> <$> # # ^ # ^ ^ ^ ^ ^ ^ ^ ^ ^ Time (minutes) S3 glucose infusion rate — • — blood glucose level (mg/dl) Figure 3.3 Sample euglycemic clamp study. >4f time 0 insulin infusion begins, which brings glucose levels down. When glucose levels reach approximately 100 mg/dl (5.5 mmol/L), glucose infusion begins and is adjusted at 5 minute intervals to "clamp" glucose levels at that level. Glucose infusion rates averaged over the last hour are taken to represent insulin sensitivity. 66 3.2.4 Computed Tomography Imaging CT scans were performed on a General Electric CT/i scanner (General Electric Medical Systems, Milwaukee, Wisconsin), before and after the intervention. An initial lateral scout, centered at the level of the iliac crest allowed accurate positioning of a single, un-angled axial image at the L4/L5 vertebral disc. A 10 mm single image scan was taken using the parameters of 120 kVp, 300 mA, for 1 second duration of a 512 by 512 matrix. Cross-sectional areas of abdominal subcutaneous and visceral AT were measured from this scan. A second, anterior-posterior scout covered the iliac crest to just below the knee to allow accurate positioning of a second image at the mid-point of the femur, from the superior rim of the femoral head to the inferior surface of the femoral condyles. A single un-angled axial image was taken at this position for skeletal muscle analyses. Within lab intra-individual coefficient of variations are not available, although published values utilizing similar protocols are approximately 2% for total abdominal, subcutaneous abdominal, and visceral A T 4 5 , 4 9 CT images were analyzed using Slice-O-Matic software (v. 4, Tomovision, Montreal, Qo) Standard attenuation ranges were used to visualize and quantify AT (-190 to-30 HU) and muscle tissue (0 to 100 HU). Muscle parameters measured were total cross-sectional area of muscle, the average density of the muscle tissue (HU) and the cross-sectional areas of low density muscle (0 to 34 HU) and high density muscle (35 to 100 HU) 1 0 3 , 1 1 7 . Correlation coefficients of intra-observer repeat analyses ranged from r=0.86 to r=0.99. 67 3.2.5 Blood Analyses Venous blood sampling was done after a 14 hour fast, with no alcohol for the preceding 3 days. Samples were analyzed for fasting insulin, glucose, glycosylated hemoglobin (HbA1c), total cholesterol (TC), low density lipoprotein (LDL-C), high density lipoprotein (HDL-C), triglyceride (TG) and apolipoprotein B (apo B). All laboratory measures were conducted in St. Paul's Hospital Laboratory, accredited by the Diagnostic Accreditation Program of British Columbia. Serum TC and HDL-C were assessed by the enzymatic colorimetric test using Boehringer Mannheim Systems reagents in a Hitachi 911 system. Serum TG was measured by the enzymatic colorimetric test using Bayer reagents in a Hitachi 911 system. LDL-C (mmol/L) was calculated using the equation (LDL-C = TC - HDL-C -TG/2.22) 1 9 3 3.3 Statistical Analyses Results are reported as group means, plus and minus the standard error of the mean, unless otherwise indicated. Pre-intervention differences in metabolic and CT imaging data were investigated by analysis of variance (ANOVA). Changes in metabolic and imaging variables were compared between groups by repeated measures ANOVA, with Tukey's post hoc tests applied if a significant overall group comparison was found. Pearson product-moment correlations were used to determine the simple relationship between imaging results and euglycemic clamp results. Statistical tests were performed using SPSS software (v. 10). Statistical significance was set at p<0.05 level. 68 The study proposal was approved by the Medical Ethics Committee of St. Paul's Hospital, and funded by the Grant-ln-Aid program of the Heart & Stroke Foundation of B.C. and Yukon. 69 3.4 Specific Hypotheses 1. Research Hypothesis: There will be a significant increase in insulin-stimulated glucose disposal over the intervention period for the Ae + RT group and for the Ae only group, but not for the usual care group. The change in the Ae + RT group will be significantly greater than the change in the Ae only or usual care group. Null Hypothesis: There will be no significant difference in insulin-stimulated glucose disposal over the intervention period for the Ae + RT, Ae only or usual care groups. 2. Research Hypothesis: There will be a significant decrease in cross-sectional area of total abdominal AT after the intervention on single-image CT scans (L4-L5 vertebral level) in the Ae + RT group and the Ae only group but not in the usual care group. There will be significant decreases in the cross-sectional area of abdominal subcutaneous AT and visceral AT in the Ae + RT and the Ae only groups but not in the usual care group. Null Hypothesis: There will be no significant difference in cross-sectional area of total abdominal AT after the intervention on single-image CT scans (L4-L5 vertebral level) in the Ae + RT group, Ae only group, or usual care group. There will be no significant differences in cross-sectional area of abdominal subcutaneous AT or visceral AT in the Ae + RT group, Ae only group, or usual care group. 3. Research Hypothesis: There will be a significant increase in mid-thigh muscle cross-sectional area in the Ae + RT group but not in the Ae only or usual care 70 group. There will be a significant decrease in low density muscle cross-sectional area in the Ae + RT group, and in the Ae only group but not in the usual care group. Null Hypothesis: There will be no significant differences in mid-thigh muscle cross-sectional area after the intervention in Ae + RT group, Ae only group, or usual care groups. There will be no significant differences in low density muscle cross-sectional area after the intervention in the Ae + RT group, Ae only group, or usual care groups. 4. Research Hypothesis: The improvement in insulin-stimulated glucose disposal across all subjects will be significantly inversely related to decreases in abdominal subcutaneous AT and visceral AT, and significantly positively related to increases in muscle cross-sectional area and muscle low density cross-sectional area. Null Hypothesis: There will not be a significant relationship between insulin-stimulated glucose disposal and abdominal subcutaneous AT, visceral AT, muscle cross-sectional area or muscle low density cross-sectional area. 71 Chapter 4: Results 4.1 Subjects Fifty-two women responded to the recruitment letters and advertisements, of whom thirty-two postmenopausal women were eligible according to the inclusion criteria and completed the initial medical exam and exercise stress testing. Two women failed the screening due to ECG evidence of CHD during the exercise stress test and two women in the exercise groups dropped out due to infectious illnesses unrelated to the study. Twenty-eight women started and completed the study. Group subject characteristics and therapeutic regimes are outlined in Table 4.1, individual subject data are listed in Appendix B. Initial values of age, duration of diabetes, weight, BMI, and levels of cardiorespiratory fitness (peak V0 2 ) did not differ between the groups. Table 4.1 Initial Subject Characteristics Control group Ae + RT Ae only n 9 10 9 Age (years) 58.9+ 2.8 63.4 ± 2.2 59.4 ± 1.9 Duration of diabetes (years) 4.7 ± 1.2 3.7 ±0.9 3.2 ± 0.6 Number of subjects treated with: Diet only Single oral agent >2 oral agents 2 3 4 6 1 3 6 2 1 BMI (kg/m') 36.7+ 2.0 33.3 ±1.5 32.5 ± 1.4 Waist circumference (cm) 119.0 ± 2.1 111.1 ±2.7 106.1 ± 2.3 Data are means ± SEM 72 Therapeutic management of blood glucose in approximately half of the subjects was by diet alone (n=14), six (21.4%) had single agent pharmacotherapy (sulfonylurea or biguanide), and eight (28.6%) were on combination therapy (sulfonylurea plus biguanide, one subject was on a biguanide plus a thiazolidenidione). The use of non-glycemic medications is documented in Table 4.2. Some of these classes of medications do not alter glycemic control, such as the statins, the angiotensinogen converting enzyme inhibitors (ACE inhibitors) or the calcium channel blockers. The other groups, such as the diuretics, hormone replacement therapy, and thyroid replacement therapy, could affect glucose levels Table 4.2 Number of Subjects Using Non- Glycemic Medical Therapy Medication Usual care group Ae + RT group Ae only group Lipid lowering (statins) 2 1 4 ACE inhibitors 1 2 2 Calcium channel blockade 0 0 2 Diuretic (HCTZ) 1 3 2 HRT 2 1 3 Thyroid replacement 0 1 3 ACE angiotensinogen converting enzyme, HCTZ hydrachlorothiazide, HRT hormone replacement therapy Several changes occurred in medication regimes during the course of the study. These changes and their known influences on glycemic control or insulin sensitivity are outlined in Table 4.3. No changes in medication regimes occurred in the usual care group. 73 Because one or two subjects in each group are involved, and the medication changes within a group are in opposition, they are unlikely to have had a significant influence on glucose or insulin levels. These medications are nOt known to influence whole body insulin sensitivity or regional AT changes, the major outcomes of this investigation. Table 4.3 Subject Medication Changes and Potential Effect on Outcomes Group Individual Possible effect of this change Strengthen or Weaken Medication Change Anticipated Effect of Exercise Ae + d/c HCTZ I glucose levels short term Strengthen glycemic effect RT group I amylodipine dose No known effect (5 to 2.5 mg OD) I metformin dose t glucose levels via hepatic Weaken glycemic effect glucose output d/c atorvastatin No known effect Ae d/c glyburide I insulin levels Strengthen hypoinsulinemic only (from 2.5 mg OD) effect group tglucose levels Weaken glycemic effect THCTZ I glucose levels short term Strengthen glycemic effect (2.5 - 5 mg OD) d/c discontinue, HCTZ hydrachlorothiazide, OD once daily, BID twice daily, TID three times daily Known effects of medications are: HCTZ increases shod term glucose levels, metformin acts to decrease glucose levels via decreased hepatic output, glyburide acts to increase insulin secretion & thereby decrease glucose levels194 4.2 Body Weight After the intervention, within group body weights did not exhibit significant differences. There were significant differences between groups in that the loss in weight exhibited by the Ae + RT group (-2.9 ± 1.3 kg) was significantly different from a weight gain in the usual care group (+ 2.0 ± 1.2 kg, p<0.05). Individual subject data are detailed in Appendix B. 74 Table 4.4 Initial Values and Change Data for Body Weight and Fitness Usual Care group Ae + RT Ae only pre post pre post pre post Weight (kg) 95.6 ± 97.6 ± 89.5 ± 86.7 ± 81.2 ± 79.9 ± 6.54 7.44 1.50 4.33f 3.8 4.22 Peak V 0 2 1.57 ± 1.53 ± 1.64 + 1.81 ± 1.67 + 1.90 + (L/min) 0.11 0.11 0.11 0.11 0.11 0.13 Peak V 0 2 17.90 ± 17.01 ± 18.53 + 21.32 20.85 ± 23.76± (ml/kg/min) 1.87 1.52 1.50 ±2.00f 1.45 1.23f Data are means ± SEM, group means represent means of subjects who completed both a pre and post intervention V 0 2 test. *Significant within-group change (p<0.017, repeated measures ANOVA with Bonferonni adjustments). fSignificant difference in change as compared to usual care group. 4.3 Dietary Assessment Compliance with dietary record return was poor, with 14 of the 28 subjects (50%) returning both a pre- and post-intake three-day food record. Statistical analyses on the limited records indicated no significant differences in the intake of calories or macronutrients (protein, fat, carbohydrate) across the groups. 4.4 Fitness and Resistance Training Measures Attendance for the exercise sessions averaged 92.7% (44.5 out of 48 classes) for both training groups, with absences distributed consistently throughout the course of the study (Apppendix C). After the 16 weeks of training the within-group changes in V 0 2 were not significantly different (Table 4.4), although between group analyses of relative oxygen consumption (i.e. expressed per kg of body weight) resulted in a significant difference between the Ae +RT group (+2.79 ± 0.79 ml/kg/min) and the usual care group, and between the Ae only group (+2.92 ± 0.95 ml/kg/min) and the usual care group (+0.9 ml/kg/min, p<0.05 for both comparisons)(Table 4.4). If the exercise groups were combined for analyses of cardiorespiratory fitness changes, then the combined exercise group had a 75 significant pre- to post-study improvement in V 0 2 expressed on an absolute basis (+0.19 l/min, p<0.17) or relative to weight basis (+2.86 ml/kg/min, p<0.17). In the Ae + RT group upper body resistive training loads increased by 42%, and lower body training loads increased by 63% from baseline (Appendix C). Training loads were not assessed in the Ae only and usual care groups. 4.5 Fasting Insulin, Glucose and Glycosylated Hemoglobin Initial values of fasting glucose, insulin and HbA1c did not differ across the three groups (p>0.05). The post-study fasting glucose, insulin and HbA1c did not differ from pre-study values in any of the groups (Table 4.5, Appendix D). One subject in the usual care group had much higher pre and post insulin levels as compared to the other subjects in the control group (pre value of 659.9 and post value of 503.7 pmol/L), with a comparatively large change value as well. When that subject was excluded from the statistical analyses, the results remained non-significant. Analysis of the pattern of change between groups did not reveal any significant differences. Table 4.5 Pre and Post HbA1c, Fasting Insulin and Fasting Glucose group Usual care Ae + RT Ae only pre post pre post pre post HgA1c (fraction of total) 0.069± 0.004 0.071±0.006 0.069+0.004 0.069+0.004 0.063±0.002 0.064±0.004 Fasting insulin (pmol/L) 248 ± 53 228 ±3 6 177 ±9.7 160 + 10.4 217 ± 14.4 201 ± 13.1 Fasting glucose (mg/dl) 8.5 ± 1.2 8.1 ± 1.1 7.9 ± 1.0 7.4 ±0.8 7.2 ±0.4 7.3 ± 0.6 Means ± sem 76 4.6 Hyperinsulinemic-Euglycemic Clamp Studies Initial values of steady state glucose infusion rates were similar across groups (p>0.05). The only group to demonstrate a significant pre- to post-study difference in glucose infusion was the Ae + RT group (+1.82 ± 0.52 mg/kg/min, p<0.17). The changes in steady state glucose infusion rates were significantly different across the groups with the increase in glucose infusion rates in the Ae + RT group (+1.82 mg/kg/min) significantly greater than the usual care group (0.07 ± 0.83 mg/kg/min, p<0.05). The glucose infusion rate in the Ae + RT group approached, but did not attain, statistical significance as compared to the Ae only group (0.55 ± 0.36 mg/kg/min)(Table 4.6, Figure 4.1). Individual data are detailed in Appendix D. Individual pre and post intervention steady state glucose infusion rates are displayed in Figure 4.2. Table 4.6 Changes in glucose infusion rates during euglycemic clamp. Control group Ae + RT Ae only Absolute % Absolute % Absolute % G i 1 0.07 3.1 1.82 77.1 0.55 19.8 ± (mg/kg/min) + 0.28 ± 12.2 ± 0.52* t ±22.0 ±0.36 12.9 Data are means ± SEM. 1 glucose infusion rate. *Significant within-group change (p<0.017, repeated measures ANOVA with Bonferonni adjustments), f Significant difference in change as compared to usual care group (p<0.05 repeated measures ANOVA). 77 Figure 4.1 Steady State Glucose Infusion Rates - Group Means CD -I—» CO - I , S E 2 O o O) 0 H usual care SlAe + RT DDAeonly *t 4.17 2.78 2.17 2.36 ^ 2.36 3.33 pre post Data are means ± SEM. 1 glucose infusion rate. *Significant within-group change (p<0.017, RM ANOVA with Bonferonni adjustments). tSignificant difference in change as compared to usual care group (p<0.05 RM ANOVA). ^Significant difference in change as compared to Ae only group (p<0.05 RM ANOVA). Figure 4.2 Steady State Glucose Infusion Rates - Individual Subjects Ae + RT Usual Care Ae only 8 ^ 6 •si - 4 CD p <f) fc o o Z3 CO 9 Pre Post 8 'ZX Pre Post 8 I • -Pre Post 78 4.7 Computed Tomography Scans - Abdominal AT Areas Group means of cross-sectional areas of abdominal AT from the CT scans are outlined in Table 4.7 and Figure 4.3, with individual subject values in Appendix F. Sample abdominal scans are included in Appendix E. Initial levels of total abdominal AT (p=0.053) and subcutaneous abdominal AT (p=0.057) displayed a trend to differ across groups. Initial areas of visceral AT did not differ across groups (p>0.05). None of the groups demonstrated a pre-study to post-study change in total abdominal AT. Between group analysis demonstrated that the pattern of change in the Ae + RT group (-48.3 cm2, -6.7%) differed from the usual care group (+17.1 cm2, +2.1%, p<0.05). No significant difference was apparent between the usual care group and Ae only group (-17.0 cm2, -2.8%) or between Ae only and Ae + RT groups. Visceral AT was significantly reduced only in the Ae + RT group after the intervention (-26.3 cm2, -10.5%). No significant difference was apparent in subcutaneous abdominal AT either within or between groups (Figure 4.3). When the two exercise groups were combined for analyses, the significant within-group changes included a reduction in total abdominal AT and visceral AT. Between-group analyses demonstrated a significant difference in pattern of change in subcutaneous AT (-15.5 cm2, -3.5%) in the combined exercise group as compared to the usual care group (+17.4 cm2, +3.2%)(Figure 4.4). 79 CO CL ZJ o co CD (O o CD X CD T J CD c !a E o O _>,| c o CD < c co 1 Z O E CD j COS CZ co tr + CD < CL =3 2 CO CO O ro ZJ co ZD E CM s CJ , _ ^ CD - ° T ~ +1 o LO CO CO CO CO + 1 CD CO . , O - o 0 0 r- c\i i + i 00 cb CD c d CD CN CD v 9 + 1 + | J " ? ^ CO 1^ GO CD < ro c E ro o •ti ^ O X I I- CO CN O) + 1 o iri LO CO LO CN + 1 00 CO CO CN + 1 LO CO co csi + 1 + 1 CN P 00 CN 00 oci CN + 1 CN O LO r- CO + 1 + 1 o CSI CN CO o CD c co ZJ CJ CO < O r-LO csi + 1 + 1 O r -od i^ ; oo LO + 1 co co CN LO csi + i + i oo T oo" t CSI + 1 h-LO T— CN + 1 co cb CN CT) Csi + 1 LO o CT> CT) csi cb csi CN CSI + i + 1 + 1 o CT) CD CO CN CD LO N- CN O o o 00 CN CO CD CN , ^ T-CN T— ^ + 1 + 1 + 1 + 1 + 1 CN CN CO o • o oo 00 o -<r CD LO co + 1 + 1 + 1 co oci o i CT> o LO CO LO CN C D -H < CD O CO Figure 4.3 Change in abdominal adipose tissue cross-sectional areas. 17.1 a usual care Si Ae+RT OJ Ae only 17.4 total abdominal AT visceral AT subcutaneous abdominal AT Data are means + SEM. *Significant within-group change (p<0.017, RM ANOVA with Bonferonni adjustments). tSignificant difference in change as compared to usual care group (p<0.05 RM ANOVA). Figure 4.4 Change in abdominal adipose tissue areas - combining exercise groups 40 * 20 _c> co co 0 o cu 2 o c cu o -20 ^0 -60 -80 17.0 -33.5 total abdominal AT • usual care a combined exercise groups 17.4 -0.4 -18.0 visceral AT -15.5 subcutaneous abdominal AT Data are means ± SEM. 'Significant within-group change (p<0.017, RM ANOVA with Bonferonni adjustments). tSignificant difference in change as compared to usual care group (p<0.05 RM ANOVA). 81 4.8 Computed Tomography Scans - Muscle Cross-sectional Area and Density Group means for muscle characteristics assessed from the CT scans are displayed in Table 4.8 and changes in these characteristics represented in Figure 4.5. Individual subject data are in Appendix G. Initial muscle cross-sectional areas did not differ across the groups (p>0.05). The cross-sectional area of muscle significantly increased in the Ae + RT group (5.9 cm2, 2.8%, p=0.017). Average muscle density significantly increased in the Ae + RT group (+2.4 HU, 5.5%, p=0.015), but significantly decreased in the usual care group (-1.2 HU, 2.5%, p=0.004). Low density muscle cross-sectional area significantly increased in the usual care group (+3.2 cm2, 6.3%) after the study. The cross-sectional area of normal density muscle increased in the Ae + RT group (+10.5 cm2, 6.7%, p<0.017). Between group analyses demonstrated significantly different patterns of change in low density muscle area between the Ae + RT group (-4.1 cm2, -8.0%, p<0.05) and the usual care group (+3.2 cm2, 6.3%). Between-group analyses also demonstrated that the pattern of change of normal density muscle cross-sectional area was significantly different in the Ae + RT group (+10.5 cm2, 6.7%, p<0.05) as compared to both the usual care group (-3.5 cm2, 2.0%, p<0.05) and the Ae only group (2.3 cm 2, 1.3%, p<0.05). 82 Table 4.8 Initial and Change Data of Mid-thigh Muscle Characteristics. Usual care group Ae + RT Ae only Initial Change (%) Initial Change (%) Initial Change (%) Cross-sectional area - cm 2 225.8 ± 8.9 0.7 ± 1.6 (0.3 ± 0.7) 207.7 ± 10.0 5.9 ±2.0* (2.8 ± 1.0) 224.1 ± 15.8 0.9 ±2.1 (0.4 ±0.9) Average muscle density- HU 47.1 ± 1.1 -1.2 ±0.3* (-2.5 ± 0.6) 43.9 ±1.7 2.4 + 0.8* t (5.5 ± 1.8) 46.1 ± 1.4 0.4 ±0.8 (0.9 ± 1.7) Low density muscle area-cm 2 50.6 ±4.2 3.2 ± 1.0* (6.3 ± 2.0) 51.3 ±6.7 -4.1 ±2.5* t (-8.0 ±4.9) 49.3 ± 7.3 -1.4 ±0.8* (2.8 + 2.6) Normal density area-cm 2 175.3 ±7.5 -3.5 ± 1.7 (-2.0 ± 1.0) 156.6 ± 12.8 10.5±2.0*f % (6.7 ± 1.3) 174.8 ± 11.1 2.3 ± 1.8 (1.3 ± 1.0) Data are means ± SEM. ( ) Percentage change in parentheses. 'Significant within-group change (p<0.017, RM ANOVA with Bonferonni adjustments). tSignificant difference compared to usual care group (p<0.05 RM ANOVA). tSignificant difference compared to Ae only group (p<0.05 RM ANOVA). 83 Figure 4.5 Changes in cross-sectional area and density of muscle 15 10 5 to 8 o o c -10 * 5.87 0.7C 0.89 LU HJ * 3.21 m. •1.40 cross-sectional area t r 10.48 • usual care Si Ae+RT trj Ae only 2.31 -4.10 t low density muscle 332 normal density muscle 4.9 Blood Lipids Group means for blood lipid results are outlined in Table 4.9, with individual subject data in appendix H. Initial values of TC, LDL, HDL, triglycerides, and apo B did not differ between the groups (p>0.05). None of the lipid sub-fractions demonstrated a significant within- or between-group change. Table 4.9 Initial and Change Data for Blood Lipids Contra group Ae + RT Ae only pre change pre change pre change TC (mmol/L) 5.18 + 0.30 -0.19 ±0.22 5.07 ± 0.22 0.21 ± 0.23 4.55 ±0.38 0.18 ±0.15 LDL (mmol/L) 3.05 ± 0.34 -0.14 ±0.17 3.0 ±0.18 0.24 ± 0.20 2.53 ± 0.34 0.01 ±0.15 HDL (mmol/L) 1.05 ±0.06 -0.06 ± 0.03 1.37 ±0.10 -0.01 ± 0.03 1.27 ±0.11 -0.03 ± 0.06 TG (mmol/L) 2.36 ±0.31 0.05 ±0.15 1.60 ±0.18 -0.05 ±0.14 1.64 ±0.23 0.16 ±0.14 Apo B (g/L) 1.10 ±0.08 -0.01 ± 0.05 0.97 ±0.04 0.05 ±0.03 0.86 ±0.11 0.06 ± 0.05 Data are means ± SEM. 84 4.10 Correlation of Glucose Disposal with Adipose Tissue and Muscle Characteristics Pearson correlation coefficients calculated for all subjects for the change in glucose disposal rates and the change in abdominal AT areas and muscle characteristic variables are outlined in Table 4.10. The changes in total abdominal AT, subcutaneous abdominal AT, visceral AT, muscle cross-sectional area, and muscle density were all significantly correlated with the change in glucose infusion rates. After controlling for the change in total abdominal AT area, the changes in muscle cross-sectional area (r=0.43, p<0.013), average muscle density (r=0.36, p<0.035), and normal density muscle (r=0.42, p<0.015) retained significant correlations with glucose infusion rates. When the change in visceral AT was controlled glucose disposal remained correlated with the changes in muscle CSA (r=0.41, p<0.05), average muscle density (r=0.50, p<0.01), and normal density muscle area (r=0.47, p<0.01). When the change in subcutaneous abdominal AT was controlled, then glucose disposal remained correlated with the change in average muscle density (r=0.47, p<0.01) and the change in normal density muscle area (r=0.46, p<0.01). Table 4.10 Pearson Correlation Coefficients between glucose infusion rates and abdominal AT and muscle characteristics A G , A total abdominal AT (cm2 ) A visceral AT (cm2 ) subcutaneous abdominal AT (cm2 ) A muscle CSA (cm2 ) A average muscle density (HU) A normal density muscle CSA (cm2 ) Pearson r R2 -0.620 -0.32 -0.64 0.43 0.50 0.52 0.38 0.10 0.41 0.18 0.25 0.27 P values <0.001 0.047 <0.001 0.011 0.003 0.002 G | glucose infusion rate, A change, CSA cross-sectional area 85 Chapter 5: Discussion IR is associated with risk for metabolic syndrome, T2D and atherosclerosis. Aerobic exercise training has proven benefit in improving insulin sensitivity in glucose-tolerant humans 3 2 ' 1 4 4 1 4 5 i and in patients with T2D 1 4 6 , 1 4 7. Limited data suggest that resistance training programs may also increase insulin sensitivity 1 5 9 , 1 6 3 , 1 6 4 , 1 6 6 . The role of resistance training warrants further investigation utilizing direct measures of insulin sensitivity (i.e. glucose clamp studies), and in broader study populations. One study is available on training programs that combine the two modalities of aerobic and resistance training. A combined aerobic plus resistance training program may be advantageous as they combine different mechanisms of action. In practice, a program which includes both aerobic plus resistance training is likely to be prescribed for individuals given the well-documented array of health benefits of aerobic training and the increasingly recognized health and functional benefits of resistance training. Thus, one of the aims of this study was to compare the changes in insulin sensitivity and body composition in response to an aerobic only training program as compared to an aerobic plus resistance training program. Abdominal obesity, and in particular the visceral component of abdominal obesity, has been closely associated with the negative metabolic characteristics of | R 9 9 , 1 0 1 , However, visceral AT may not have a unique effect as abdominal subcutaneous A T 1 8 , 6 6 ' 9 4 and IMCL in skeletal muscle94 have also been shown to be positively associated with metabolic risk. 86 Some of the benefit from aerobic and/or resistance training may be achieved by concomitant changes in body composition such as reduced abdominal AT or intermuscular AT and increased skeletal muscle mass. Only two studies have evaluated changes in abdominal visceral, subcutaneous and IMCL regions relative to insulin sensitivity after aerobic training 1 5 1 , 1 8 1 or resistance training151. Thus the second aim of this study was to evaluate changes in abdominal visceral AT, abdominal subcutaneous AT, IMCL and skeletal muscle size relative to changes in insulin resistance. Postmenopausal diabetic women have been underrepresented in study populations evaluating exercise and insulin sensitivity. Research studies investigating groups with diabetes or glucose intolerance have used predominantly middle-aged men 1 2 , 1 4 0 ' 1 4 6 , 1 4 8 , 1 9 5 ; postmenopausal women have been underrepresented. Only four studies have included middle-aged diabetic women and their numbers have been small (3,4,4 and 11 subjects) 1 6 6 , 1 4 8 , 1 5 0 , 1 4 9 , while more recent studies have examined young (average age 28 years) normoglycemic women151, or middle-aged normoglycemic men 1 8 1. Individuals with T2D are characterized by IR and abdominal obesity, and are an important group to test the effect of exercise training on insulin responsiveness and regional AT variables. Despite the well known general benefits of exercise 67% of older Canadian women are currently physically inactive196, and 13% of women over the age of 60 have T2D. Physical activity reduces the incidence of T2D 3 6" 3 8, improves metabolic control in those with established diabetes197, and reduces the risk of all-cause and cardiovascular disease in individuals with diabetes198. These high prevalence rates of 87 physical inactivity and T2D in Canadian older women and the serious morbidity outcomes of these lifestyle practices underscore the value of investigating effective interventions for this group. Thus we chose to evaluate postmenopausal women with T2D and their metabolic and body composition responses to aerobic only training and aerobic plus resistance training programs. 5.1 Exercise Training Modality and Insulin Sensitivity 5.1.1 Improvements in Insulin Sensitivity after Aerobic plus Resistance Training The first hypothesis of this thesis was that glucose disposal during a euglycemic clamp would be significantly improved in the Ae+RT group and the Ae only group, but not in the usual care group. The Ae + RT training program significantly improved glucose disposal by 77% (1.82 mg/kg/min), whereas the increase in glucose disposal in the Ae only group was not significant (0.55 mg/kg/min, 19.8%). No studies have evaluated the effect of aerobic plus resistance training on insulin sensitivity and regional AT, so these results will be compared to data from resistance training only programs. Three reports are available on the effect of resistance training on insulin sensitivity as measured by hyperinsulinemic-euglycemic clamp. In overweight (BMI > 27 kg/m2), postmenopausal women a 16 week resistance training plus weight loss program resulted in a 52% increase in insulin sensitivity164. In the same study, a group of normal weight (BMI < 27 kg/m2) women completing the same resistance training program without weight loss, did not demonstrate a significant improvement in insulin sensitivity164. In men with impaired glucose tolerance, a 10 week resistance training program resulted in a 23% improvement in insulin sensitivity159, and in a group of older men and postmenopausal women a 6 month resistance training 88 program tended to improve insulin sensitivity during a euglycemic clamp1 9 9. In young, normal weight (BMI of 22 kg/m2) women a 6 month resistance training program brought about an increase in glucose disposal during a clamp of 9%151. Thus, our results of improved insulin sensitivity in response to an Ae + RT training program support the value of this exercise modality in improving insulin sensitivity. Evidence from studies evaluating glucose and insulin responses to resistance training also suggest that resistance training has beneficial effects on insulin action 1 6 6 , 1 7 2 , 1 7 4 . In men and women with T2D, circuit weight training decreased the insulin response to an oral glucose challenge166. In overweight men 1 7 4 and women1 7 2 a combined weight loss and resistance training program decreased fasting insulin levels and insulin responses to an oral challenge, a reflection of improved insulin action. Our results are the first to demonstrate improved insulin sensitivity in postmenopausal women with T2D in response to a program of aerobic and resistance training. Improvement in insulin action following a program which includes resistance training may result from an increase in skeletal muscle tissue. Individual muscle fiber hypertrophy would allow an increase in the amount of glucose that is stored per fiber, facilitating insulin binding and clearance200. Other mechanisms which may play a role include an increased GLUT-4 expression, as well as adaptive responses of enzymes involved in glucose phosphorylation and oxidation39 ,169. It is not known how these adaptations may differ in response to aerobic as compared to aerobic plus resistance training programs. 89 5.1.2 Lack of Improvement in Insulin Sensitivity after Aerobic Only Training The aerobic group of this study did not achieve a significant increase in glucose disposal, in contrast to expectations. Improvement in insulin sensitivity has been reported after aerobic exercise training in men with glucose intolerance or T2D without weight loss 1 2 , 1 4 0 ' 1 4 6 , 1 5 1 , with increased glucose disposal rates in the range of 16-44%. When exercise training is accompanied by weight loss (11 kg, or 11% of body weight), insulin sensitivity in men and women is also improved 1 4 9 ' 1 5 0 in the range of 11 -27%. Diet-based weight loss interventions in overweight men and women also result in increased insulin sensitivity of 25%171 (15 kg loss) to 43% (7.4 kg loss)181. From these reports it appears that exercise training without weight loss, weight loss alone, or exercise training in combination with weight loss induces improvements in insulin sensitivity. Further comparison of the magnitude of improvement of the different interventions is difficult as each study was different with respect to the amount of weight lost, the subjects studied (i.e. male/female, overweight or normal weight, glucose tolerant or diabetic), characteristics of the hyperinsulinemic-euglycemic clamp (i.e. level of hyperinsulinemia), and the time elapsed between the clamp study and the last exercise session. The results of a study of obese men comparing the independent effects of a 3 month diet-induced weight loss, exercise-induced weight loss, or exercise without weight loss1 8 1 are helpful in comparing the effectiveness of weight loss compared to exercise. Significant increases in glucose disposal during a euglycemic clamp were similar in the diet-induced (43%) and exercise-induced (64%) weight loss groups. Exercise without weight loss was associated with a 30% increase in glucose disposal which did not reach statistical significance. Because the positive effect of exercise on insulin 90 sensitivity attenuates quickly and the euglycemic clamp studies were performed four days after the last exercise session, the results reinforce the value of regular adherence to exercise41. The fact that the aerobic plus resistance training group in our study had significant improvement in insulin sensitivity without significant weight loss suggests that lack of substantial weight loss is not the reason for a lack of improvement in glucose disposal in the Ae only group of our study. Lack of improvement might be related to the timing of insulin sensitivity assessment. Insulin sensitivity declines with training cessation and measures of insulin sensitivity have been carried out at 48 hours post exercise and up to 3-6 days post exercise. In men with T2D a 10 week aerobic training program increased glucose disposal by 39% during a hyperinsulinemic-euglycemic clamp measured within 24 hours post exercise; this effect was lost when re-measured 6 days post training140, while a group of normoglycemic men in the same study retained improved insulin sensitivity when measured 6 days after training (values not reported). In normal weight women, 6 months of aerobic or resistance training increased glucose disposal significantly (16 and 9%, respectively) when measured 4 days after the last exercise session151, and a group of men and women with diabetes demonstrated improved glucose disposal (44%) when measured 4 days post exercise12. Our glucose disposal measures were obtained approximately 60 hrs after the last exercise session, which should have captured the effects of exercise before its diminution. Our results may reflect the impact of the most recent exercise session, or a training adaptation or an interaction of both. In practical application, this suggests the value of consistent exercise frequency to maintain the beneficial effects. The documented improvement in our 91 aerobic plus resistance training group suggests that elapsed time is not the main factor in explaining the lack of results. Aerobic training program characteristics might be considered in evaluating differences across studies. This program utilized exercise program features comparable in intensity, frequency, duration as other reports 1 2 ' ' 4 0 ' 1 4 9 , 1 5 0 , 1 5 1 , 1 5 9 . The study sample size may have been inadequate to demonstrate differences of glucose disposal in the Ae only training group as compared to the other groups. Effect sizes (change after intervention divided by the standard deviation) in the literature for insulin sensitivity increases after aerobic or resistance exercise training in men and women with glucose intolerance or T2D ranged from 0.47 to 0.77 1 4 0' 1 5 0' 1 5 9. The effect size for changes in CT-assessed thigh muscle cross-sectional area after resistance training in postmenopausal women was 0.49-0.55 1 7 8 , 1 8 3' 2 0 1. In planning the study, the smaller effect size of 0.50 (similar for change in insulin sensitivity and for change in muscle cross-sectional area) was used in power calculations to determine the necessary sample size, with the result of 12 subjects per group to provide statistical power of 0.85-0.91. A sample size of 10 was calculated to provide power of 0.76-0.84. Recruitment for the study was closed at a total of 28 subjects due to difficulty in maintaining a sufficient recruitment rate to justify continuation of the costs of exercise training and testing for the study. The effect size from this study's data of insulin-stimulated glucose disposal for the Ae only group was 0.51. The effect size for the Ae + RT group was 1.1. Thus it is possible that the sample size achieved in this study was too small to see the magnitude of effect on insulin sensitivity in the Ae only group. 9 2 Subject characteristics might also play a role. Nonobese, obese, or impaired glucose tolerance subjects have been used as models of IR in which training improvements can be demonstrated, but established diabetes may represent a model of IR further advanced in the natural history of the disease and less amenable to improvement. Our subjects demonstrated a greater IR (2.5 mg/kg/min insulin-stimulated glucose disposal) than other studies in type 2 diabetes (4.5 mg/kg/min12 to 5.47 mg/kg/min146). A population with more severe IR may not respond in a similar time-frame to the aerobic training stimulus, although it appears that using a more effective exercise stimulus, i.e. one which includes resistance training, may bring about changes within the time-frame investigated. Our results of improved insulin sensitivity in response to a combined aerobic plus resistance training program suggest the effectiveness of such a dual-modality approach. This may be especially true in light of the lack of significant response to an aerobic training only program in this study population. The beneficial results seen in the program which included resistance training may be due to increased muscle mass, as skeletal muscle is the primary site of glucose disposal. This is supported by our finding of an increased muscle cross-sectional area in the Ae + RT group (discussed in the section on Change in Muscle Characteristics). 5.2 Abdominal Adiposity Changes 5.2.7 Change in Total Abdominal Adipose Tissue None of the study groups demonstrated a significant weight change from baseline values. The modest weight loss in the Ae + RT group (2.8 kg, 3.1%) was significantly 93 different from a modest gain in the usual care group (+2.0 kg, 2.1%). A meta-analysis on exercise and weight loss concluded that exercise without caloric restriction results in modest losses of body weight (2.9 kg, or 0.2 to 0.26 kg/wk) 1 8 1 , 2 0 2 in overweight adults over similar time periods. The second hypothesis of this thesis was that total abdominal AT would be significantly decreased in each of the exercise training groups but not in the usual care group and that changes would not differ between the exercise groups. In pre to post study comparisons, none of the groups showed significant change in total abdominal AT area. Between group comparisons demonstrated a significant difference in the pattern of total abdominal AT loss in the Ae + RT group, as compared to a pattern of gain in the usual care group. The exercise groups were combined, and with this analysis, total abdominal AT area was significantly reduced by 33.5 cm 2 (5%) after the study. These findings of a small loss of abdominal adiposity were similar to changes reported in overweight postmenopausal women1 7 7 after a twelve month walking program. These authors reported decreases (of 29.7 cm2, or 5.5% of initial levels) in cross-sectional area from a single-image CT scan at the L4-L5 vertebral disc space. Weight loss in this study was 1.3 kg, similar to our results. Two exercise studies which measured regional AT with MRI reported losses in abdominal obesity with weight maintenance. In overweight men4 1 after a three-month aerobic exercise program, losses were seen in abdominal AT as assessed by multiple MRI of the abdominal region, although specific values were not reported. In overweight men with T2D, after an 8 week aerobic exercise program with no weight 94 loss1 4 8, a loss in abdominal AT by single image MR at the level of the umbilicus was seen. The decrease in abdominal area was 116.3 cm2, or 30.3% of initial area. Thus, the finding in the present study confirm other reports which suggest that exercise without caloric restriction is an effective strategy for reducing abdominal obesity. 5.2.2 Change in Abdominal Subcutaneous and Visceral AT The second component of the hypothesis relating to abdominal AT changes was that losses from abdominal subcutaneous AT and visceral AT losses would be similar in magnitude. The objective was to follow adipose patterning changes after an exercise intervention to assess whether there would be preferential mobilization of AT from the visceral AT depot. After the study a significant loss was seen in the visceral AT area of the Ae + RT group, while the subcutaneous depot did not change significantly in any individual group. If the two exercise groups were combined, the loss in visceral AT in the exercise group remained significant. There was a significant difference in subcutaneous abdominal AT change between the combined exercise group, which demonstrated a pattern of loss, as compared to a pattern of gain in the usual care group. These results do not support the hypothesis, as the visceral AT depot significantly decreased and the subcutaneous abdominal AT was not significantly reduced. Of interest is the finding that the usual care group displayed a tendency to gain AT in the abdominal subcutaneous depot. This raises a future research question of whether this represents a pattern of preferential weight gain in the subcutaneous abdominal depot. Review of published reports of exercise training without weight loss has revealed variable results. In a study of single-image abdominal CT scans in postmenopausal 95 women after 16 weeks of resistance training, significant losses were reported from the visceral AT (13.9 cm2, or 9.6% of initial levels) but not from the subcutaneous abdominal AT region178. Losses from both regional depots were described after an 8 week aerobic training program in men with T2D 1 4 8 as assessed by single image MRI studies. If exercise training is accompanied by substantial weight loss, decreases in both abdominal regions have been documented in obese women by single image CT analyses180 and by multiple image MRI analyses of the abdominal region 1 7 2, and in obese men 4 1 , 1 7 4 . The variable differences across studies may be attributable to small adiposity changes (in studies with no weight change), small sample sizes, and the large variability in these depots seen in most studies. Our study group was characterized by abdominal obesity (BMI of ~34 kg/m2 and waist circumference of ~112 cm) by selection criteria, but considerable heterogeneity in abdominal AT patterning was evident based on inter-individual coefficients of variation of 33.7% for visceral AT and 27.0% for subcutaneous abdominal AT. Two studies have utilized large sample sizes to better assess body AT pattern changes with exercise. Wilmore et al. (1999) studied a very large sample size (n=299) of young (age 33 years), lean (BMI 24.9 kg/m2) women using single-image CT at the 4 th-5 th lumbar vertebral space. Small losses in both visceral (3.1 cm2, 4.6%) and subcutaneous (8.9 cm2, 3.2%) abdominal AT depots were seen after 5 months of aerobic exercise without weight loss1 7 9. A second study in overweight, postmenopausal women (n=173, age 61 years, BMI 30.5 kg/m2) found that a walking-based program with a small weight loss (1.4 kg) brought about significant losses from both visceral (8.6 cm2, 6.9%) and subcutaneous 96 abdominal (28.8 cm2, 7.4%) AT regions177 on single-image CT scans at L4-L5. The values from our results are similar with visceral AT losses of 18 cm 2 or 7.7% of initial levels and losses of subcutaneous abdominal AT losses of 15.5 cm 2 or 3.5% of initial levels. It appears that weight loss and/or exercise interventions induce losses of both visceral and subcutaneous abdominal AT. Reviewing the data summarized in Table 2.6 indicates that in the majority of studies absolute losses from the subcutaneous abdominal AT region are equal to or greater than losses from the visceral Aji72,174,179,180 t n 0 U gh the statistical comparison of the magnitude of loss between these two depots is rarely made within studies. This pattern seems to apply whether the AT losses are from diet-based weight loss interventions, or from exercise training with and without weight loss interventions. Frequently, adiposity changes are expressed as a percentage of the initial level, often leading to the limited interpretation that visceral AT losses are greater than abdominal subcutaneous AT losses. Losses from AT may be better understood if the magnitude of the loss is considered both as absolute and relative change. Several factors should be considered in the question of preferential regional mobilization of AT. Adaptation of the imaging technologies of CT and MRI has permitted research of this issue in the past 5-10 years. Significant evidence from cross-sectional studies supports an association of visceral obesity with insulin sensitivity, but less prospective data is available following change in insulin sensitivity with change in CT or MRI imaging of regional body composition 4 1 , 1 5 1 , 1 7 1. The populations studied have for the most part been normoglycemic, middle aged (33-45 9 7 years) obese (BMI ranging from 30-40 kg/m2) men and women. Insulin sensitivity demonstrates an inverse relationship with increasing adiposity, as demonstrated by a group of middle-aged men ranging in adiposity (BMI 23-37 kg/m2)66. Total DXA-measured body fat was significantly inversely correlated with glucose disposal during a euglycemic clamp (r=-0.61), suggesting that total body fat explains approximately 37% of the variation in insulin sensitivity. The effect of increasing adiposity on insulin sensitivity appears to reach a plateau at a BMI of 30 kg/m2 6 6 . This suggests that change in insulin sensitivity may be blunted at higher body fat values, also blunting associations with regional body composition variables. Similarly, the relationship between insulin sensitivity and adiposity may differ in lean subjects. Cross-sectional areas above 135 cm 2 of visceral AT have been associated with adverse metabolic profiles203. These findings raise the question of whether there are thresholds and ceilings for the putative negative effects of individual regional adipose depots on insulin sensitivity. Gender differences may exist in regional adiposity changes with exercise interventions. A large study179 found that women (mean age 33 years) had lower initial levels of visceral AT on single-image CT scans than men. After 5 months of aerobic exercise training, women lost less subcutaneous AT (measured by skinfolds), less abdominal visceral AT, and less total fat than men. These results demonstrate a gender difference in weight and body adiposity loss in response to exercise. Other investigators have been able to induce similar adiposity losses by diet and exercise in separate groups of men 1 7 4, premenopausal women172, and postmenopausal women1 8 0 with improvements in insulin action during an oral glucose tolerance test reported in men and postmenopausal women, but not in premenopausal women. 98 Some of the evidence on the association between regional AT depots and insulin sensitivity reviewed is from single slice CT protocols. High correlation values between single image cross-sectional areas of AT with cadaver planimetry and multiple slice protocols support the use of single slice methodology52, especially in the context of radiation exposure of the scanning. However, the limitation of single slice imaging has recently been raised. Comparisons of 4 contiguous cross-sectional abdominal CT images (L2-L4) indicated considerable intra-individual variability in AT deposition patterning in normal weight, premenopausal women61. These findings suggest that single slice scanning may be of limited value in comparative studies61, which is a large part of the evidence linking visceral AT with abnormal metabolic variables. A thorough evaluation of intra-individual variability in body AT deposition was done in 54 healthy women who ranged widely in body weight and ponderosity (BMI 19-40 kg/m2), utilizing full body MR imaging106. Whole-body results indicated significant variation in visceral and non-visceral internal AT volumes which could not be predicted by either indirect methods (standard anthropometic measurements) or direct measurements (MRI) of total body AT or subcutaneous AT. Although a single-image CT analysis is highly correlated (r=0.83)52 with cadaver-assessed total intra-abdominal AT, it explains a large portion (R2 =69%), but not the whole variance. Single slice imaging data can provide a rough estimate of internal (including visceral) AT, but potentially at the loss of more subtle differences. These compromises may be important in studies evaluating the relationship between AT and metabolic markers, but may have less impact when serial measurements are taken in intervention studies. The variability in AT patterning indicated by these studies may be a factor in 99 limiting information based on single slice methodologies, most commonly used in CT imaging. In summary, our results indicate a statistically significant reduction in total and visceral AT in the combined exercise groups after the intervention. In addition, the usual care group demonstrated a significant gain in subcutaneous AT compared to the combined exercise group. A review of published prospective studies (Table 2.6) suggests that losses occur from both subcutaneous and visceral components of abdominal AT. Losses from the subcutaneous depots are numerically similar or larger, although a statistical comparison within groups has rarely been done. 5.3 Change in Muscle Cross-Sectional Area and Muscle Density 5.3.1 Change in Muscle Cross-sectional Area It was hypothesized that the Ae+ RT intervention would bring about a greater increase in mid-thigh muscle cross-sectional area than the Ae only group or the usual care group. Muscle cross-sectional area did not change after aerobic training, but increased after aerobic plus resistance training. An increase in muscle cross-sectional area after the Ae + RT training would be expected based on other findings of muscle hypertrophy after resistance training, but no change with aerobic training only 411si,174,178,183 A n j n c r e a s e j n m u s c | e cross-sectional area may have a beneficial impact on glucose disposal, as muscle mass is the predominant site of oral glucose load clearance39. This is supported by the finding that in non-obese younger women, improved insulin sensitivity after resistance training did not persist if the glucose disposal rate was expressed on a fat-free mass basis1 5 1. 100 5.3.2 Change in Muscle Density A second component of the hypothesis being tested stated that Ae + RT training and Ae only training would induce a greater decrease in mid-thigh muscle density than the usual care group. Muscle density was assessed by assessing the overall average muscle density on the CT image, as well as by the proportion of muscle cross-sectional area which was of normal density (35-100 HU) or low density (0-34 HU). This hypothesis was supported by the findings that average muscle density was increased in the Ae + RT group, normal density muscle area was significantly increased by 10.5 cm 2 (6.7%) in the Ae + RT group, an increase which was significantly greater than both usual care group and the Ae only group. The usual care group displayed an increase in low density muscle, which was significantly different than the Ae + RT group. These results partially support the hypothesis, as the Ae + RT training program increased muscle density, but Ae only training did not. As in abdominal adiposity, the usual care group appeared to worsen as muscle density decreased over the study period. Low density muscle tissue on CT images is thought to contain greater amounts of intermuscular lipid and had been linked to IR in obesity and T 2 D 9 4 , 1 1 7 , 1 1 8 , 2 0 4 It has been unclear how exercise training might affect this aspect of muscle composition. Decreases in muscle density on CT imaging have been reported after aerobic training with1 8 0 or without weight loss1 5 1, and after resistance training 1 5 1 , 1 8 3. These findings are thought to reflect changes in intermuscular lipid content. The findings of reduced muscle attenuation are consistent with increased muscle lipid content determined histochemically and from phantom models (solutions of varying lipid 101 content) in which an increase of 1% in lipid content resulted in a predictable 1 HU decrease in attenuation48. Other changes in muscle tissue composition, such as water content or an increase in contractile proteins, may be a factor in CT assessed muscle density. The latter component of muscle is likely to increase after resistance training although little data is available on how this change might impact CT measurements. Thus, CT assessment of muscle density is interpreted to represent an increased lipid content of muscle, although potentially changes in lean component of muscle may also occur. CT attenuation changes cannot distinguish lipid within interstitial AT from lipid stored within muscle cells i.e. IMCL 4 8. Standard MRI allows visualization of the interstitial AT in muscle59, i.e. the AT which is intertwined between bundles of muscle fibres172, thus it can be used to characterize and distinguish the interstitial AT content of lean and obese muscle44. Strong correlations between MRI (r=0.92) measurements of interstitial AT and cadaver section analysis53 support the use of MRI to assess interstitial AT in muscle and estimate that interstitial AT represents 24% of skeletal muscle area5 3. In obese individuals an increase in AT below the muscle fascia and interspersed between muscle groups was associated with IR 3 7 , 1 0 7 . Studies using MRI have also reported that interstitial AT is reduced after aerobic or resistance training (with weight loss)172. Recent advances in MRS appear to be able to distinguish IMCL from interstitial AT, and have documented that IMCL is related to the insulin resistance in obesity122 and in first degree relatives of those with T2D 1 2 1 . IMCL has also been examined by directly visualization using microscopy. In situ staining of lipid and examination with microscopy localizes the lipid to within the myocyte, and allows quantification of the 102 volume of IMCL. The volume of the myocyte occupied by lipid has been reported to be 1.5% in lean subjects, 3-4% in obesity and slightly greater again in T2D 2 0 5 . A seemingly paradoxical relationship exists between IMCL and IR as demonstrated by the finding that endurance-trained athletes have an enhanced storage of IMCL along with high insulin sensitivity, relative to sedentary individuals. A comparison206 of i) sedentary lean, ii) sedentary obese, iii) obese subjects with T2D and iv) lean exercise-trained men found that insulin sensitivity assessed with a euglycemic clamp was greater in the lean and trained individuals than the obese and diabetic individuals. Quantitative histochemical analyses of muscle biopsies showed that IMCL was higher in those with diabetes compared to lean subjects, but was also higher in the trained subjects when compared to the lean 2 0 6. This demonstrated that endurance trained individuals were markedly insulin sensitive, despite having elevated IMCL content. The endurance trained subjects also had high cellular oxidative capacity as indicated by succinate dehydrogenase activity. The authors suggest that in endurance trained subjects the periodic utilization (oxidation) of the IMCL as an energy substrate during regular exercise may overcome detrimental effects of IMCL accrual. IMCL accumulation in sedentary, obese and diabetic subjects was associated with lower oxidative capacity. Under these circumstances IMCL accumulation may be resulting from deficient utilization, and confer insulin resistance. Elevated IMCL content in sedentary, obese and diabetic subjects may have detrimental effects via intermediaries of non-oxidative IMCL metabolism, such as long chain acyl CoA, diacylglycerol and ceramide 3 0 , 1 2 3. IMCL accumulation in obesity and diabetes may be a surrogate for the underlying problem of impaired lipid 103 oxidation, whereas IMCL in the trained state may not have detrimental effects due to its regular utilization and turnover as energy substrate. The accumulation of IMCL is an important aspect of IR in muscle in obesity and T2D, and is related to changes in the use of lipid and carbohydrate fuels. The activities of several marker enzymes of oxidative and glycolytic capacity are altered in obesity, such that the capacity for lipid oxidation is reduced while glycolytic potential is increased 2 0 7 , 2 0 8. This is supported by the findings from leg balance studies which assessed fuel utilization across the leg muscle bed. Under rest conditions, leg muscle in obese subjects demonstrated a^decreased reliance on lipid oxidation, a finding which correlated with insulin resistance during a euglycemic clamp2 0 9. Under insulin-stimulated conditions, obese subjects did not suppress lipid oxidation to the same degree as lean subjects, and this lack of suppression of lipid oxidation was related to insulin resistance. This indicates that the responsiveness to insulin modulation of muscle fuel (lipid) use is related to its capacity to respond to insulin stimulation of glucose uptake. This biochemical characteristic of insulin-resistant muscle to deal with lipid oxidation has been referred to as metabolic inflexibility123. If defects in the capacity for lipid oxidation are factors in the pathogenesis of IR, any effort to modify and increase lipid oxidation, such as exercise training, should be considered a potential form of treatment123. Endurance training effects could include depletion of intermuscular AT through an enhanced capacity for lipid oxidation210. These biochemical changes may underlie the observation of a 15% lower maximal oxygen uptake in relatives of individuals with T2D vs. controls 2 0 7 , 2 0 8. Women in the current study demonstrated an increase in peak V 0 2 of 11.4% after exercise training, similar to findings of improved maximal V0 2 , and improved oxygen kinetics in 104 premenopausal women with T2D after a three month exercise training program2 1 1. These results suggest benefits in oxygen delivery and utilization, although it is not known whether the mechanisms improved in these studies are the same oxidative and glycolytic impairments described above. Thus, increased lipid content in skeletal muscle is associated with increased resistance to the action of insulin. This relationship is evident when muscle lipid content is assessed by reduced attenuation on CT images, a technique which does not distinguish IMCL from lipid of interstitial AT. MRI studies support the association of interstitial AT with IR, and MRS studies1 2 1 and histological examination of biopsy samples1 2 3 of IMCL support the association of intramyocyte lipid accumulation with IR. It is not clear if these different sites of lipid deposition have similar mechanisms of action of effects on tissue insulin sensitivity. Insulin-resistant muscle is characterized by lower oxidative capacity, and lower rates of oxidation of FFA 2 0 9 , 2 1 2 , findings which led to the proposal that an impaired lipid oxidative capacity of muscle contributed to lipid accumulation within muscle31. The deficient utilization and subsequent accumulation of lipid within the muscle cell may contribute to IR. A normal response to exercise training is an enhancement of insulin sensitivity and an increased capacity for lipid oxidation213. Our results of reduced attenuation on CT images of muscle suggest the possibility that IMCL has been reduced after the Ae + RT program. 105 5.4 Relationships of Change in Body Composition with Change in Insulin Sensitivity Across all subjects (n=28), change in insulin sensitivity was inversely correlated with change in subcutaneous abdominal AT (r=-0.64, p<0.001) and with change in visceral AT (r=-0.32, p<0.05). Similarly, change in insulin sensitivity was positively correlated with change in muscle cross-sectional area (r=0.43, p<0.01), average muscle density (r=0.50 p<0.01) and the proportion of normal density muscle (r=0.52, p<0.01). When total abdominal AT was controlled, the increased muscle area, average muscle density, and normal density muscle retained independent, significant correlations with glucose disposal. Thus, abdominal AT and muscle area and density are independently related to improved insulin sensitivity after exercise training. Substantial evidence on the unique role of visceral AT was based on responses to an oral challenge, where visceral AT explained 19% of the variance in glucose area and 32% of the variance in insulin response9 9 , 1 0 0. Change in visceral AT also emerged as a significant correlate with glucose disposal after weight loss with or without exercise in obese men4 1. Limited reports have described the relationship between directly assessed insulin sensitivity, i.e. glucose disposal during a euglycemic clamp, relative to CT or MRI body composition changes after an exercise training program 1 4 8 , 1 5 1 , 1 8 1 . These reports supported previous observations of visceral AT as a significant correlate of glucose disposal during a euglycemic clamp in obese men after exercise and/or weight loss, and in men and women with T2D after exercise training 1 4 8 Thus previous reports have documented relationships with visceral AT after an intervention of exercise training and/or weight loss, although possible relationships with 106 subcutaneous AT or lipid within muscle remain unclear. Muscle mass assessed by multiple-slice MRI did not change after aerobic exercise training in one study of overweight men 1 8 1, although a second study reported a 23% increase in muscle size along with a 46% increase in insulin tolerance (intra-venous insulin tolerance test)148. Increased muscle attenuation, and no change in subcutaneous or visceral abdominal AT areas on CT imaging were accompanied by increased insulin sensitivity in young, non-obese women after both endurance and resistance training151. In our results an increased insulin sensitivity was correlated with increased muscle cross-sectional area (r=0.43) and increased normal density muscle (r=0.52). The significant correlations remained after controlling for the effect of total abdominal AT (r=0.43 and r=0.36 for muscle cross-sectional area and average muscle density respectively). The difference between our results and other findings may be due to the added effectiveness of muscular strength training in stimulating changes in muscle area or function. An improvement in glucose disposal in young, nonobese women seen after strength training disappeared when indexed for DXA-assessed whole body fat free mass 1 5 1, suggesting that the increase in lean body tissue (including skeletal muscle) contributed to the improvement. An endurance trained group in the same study retained insulin sensitivity improvements after the indexation for fat free mass. These results suggest that aerobic training and resistance training improve insulin sensitivity but by different mechanisms. The comparison of results from studies utilizing resistance training highlights two issues. The first is the variable expression of glucose disposal hampers comparisons across studies, as it has been expressed as an absolute rate (mmol/min), indexed for fat free mass assessed by DXA or hydrodensitometry (mg/kg lean body mass/min), 107 or indexed for MRI assessed skeletal muscle (mg/kg muscle/min). Expressing glucose disposal during a euglycemic clamp as an absolute rate will allow changes due to increased skeletal muscle mass to be more evident. Secondly, the baseline characteristics of the study group may affect outcomes. For example, our study cohort had lower initial muscle density (45-46 HU vs. 49 HU) than that reported in the study of young, non-obese women; a substantial difference when the magnitude of change in muscle density with exercise training is in the order of 2 HU. A lower initial level may allow improvements to be more easily attained. These results provide a broader evaluation of several body composition characteristics and their relative contribution to improved insulin sensitivity with exercise training. They add support to cross-sectional evidence demonstrating that subcutaneous abdominal AT, as well as visceral AT, demonstrates a strong relationship with IR 9 4 , 2 1 4 Furthermore, these results also provide evidence that increased skeletal muscle cross-sectional area and increased muscle density (assumed to reflect decreased intermuscular lipid) are independent contributors to improved insulin sensitivity. The associations between the regional AT depots and muscle area explain a portion of the improvement in insulin sensitivity, with R 2 values ranging from 10% (visceral AT) to 41% (subcutaneous abdominal AT). The unexplained variance in the improvement in insulin sensitivity could be due to metabolic improvements which occur with exercise training39, but which were not assessed in this study. As IR is associated with reduced FFA utilization by muscle and diminished activity of key enzymes of FFA metabolism, exercise training adaptations, including improved capillary density, activity levels of regulatory 108 enzymes of lipid metabolism and mitochondrial volume , could be contributing to improvements seen in this study, in addition to body composition changes. 5.5 Fasting Insulin, Glucose and Glycosylated Hemoglobin Levels HbA1c is a frequently used tool in the management of T2D to evaluate glycemic control. Glycosylated hemoglobin reflects glucose control over the previous two to three months, and the goal for treatment is to achieve levels of less than 0.07 (or 7% of total hemoglobin) according to Canadian medical guidelines90. A recent review and meta-analysis of glycosylated hemoglobin and exercise training has found that a small to modest decrease of 0.05-1.0% in glycosylated hemoglobin can be expected in response to exercise training 1 9 7 , 2 1 5. Variable findings have been attributed to small sample sizes and the complex pathophysiology of type 2 diabetes197. A reduction of this size is similar to the reduction achieved in the intensive glucose-lowering treatment group of the United Kingdom Prospective Diabetes Study (UKPDS), which was related to a significant reduction of diabetes-related clinical end points215, suggesting therapeutic benefit. Resistance training interventions appeared to have a similar effect to aerobic training on glycemic control although only two randomized, controlled studies were available using this exercise modality215. In the above meta-analyses the effect of exercise training (aerobic or resistance) was estimated from data obtained from middle-aged participants with T2D, with only one study in the meta-analysis including participants who were older than 65 years. This limits the generalizability of these findings to older adults. Fasting plasma glucose levels are also used in assessing glycemic control. A recent review of published literature found that improvements in fasting glucose followed a 109 pattern similar to improvements in HbA1c i.e. that modest improvements were noted in many, but not all, studies197. A larger sample size than used in our study may be necessary to see modest changes. Although the effect of exercise on glycemic control as indicated by HbA1c and fasting glucose levels is modest, there are other positive effects. Foremost is the well-documented improved insulin sensitivity which results from exercise training, which is a topic of this investigation. Exercise has also been found to improve postprandial hyperglycemia, and may acutely lower hepatic glucose production197. Although the reported effects of exercise interventions on glycemic control are variable, there appears to be consensus that exercise training brings about a reduction of fasting insulin levels, reflecting an improvement in insulin sensitivity174,197. In obese individuals, small changes in insulin sensitivity have been found to be associated with relatively large changes in fasting insulin levels, suggesting the sensitivity of this measure in obese individuals216. 5.6 Lipoprotein Levels The present study was unable to show a significant effect of either exercise training program on lipoprotein patterns. Several reviews are available on the effect of exercise on lipoproteins 1 9 7 , 2 1 7 , 2 1 8. In healthy populations, endurance exercise training studies indicate that plasma TG concentrations are usually decreased, with an approximate mean change of 20%217. This decrease is related to baseline concentrations, with greater reductions reported in persons previously inactive with high baseline concentrations. Changes in plasma TC levels are not seen after endurance training, unless the training program is accompanied by a reduction in body weight, body AT, or dietary lipid. Similarly, plasma LDL is not found to be 110 lowered after exercise training, and insufficient evidence exists to suggest that the small, dense LDL particles are impacted by endurance exercise2 1 7. HDL levels are generally responsive to endurance training. Exercise-induced changes in HDL are approximately 10%. Meta-analyses of exercise induced change suggest an inverse correlation between initial HDL levels and the change in HDL after exercise training. Twelve weeks of exercise training are usually needed to induce an increased HDL concentration, and a strong relationship exists between exercise volume and HDL. The favourable changes in HDL levels are reported without altered body weight or composition, but this increase is augmented by body fat loss. Our lack of change in HDL may be due to a relatively short intervention period of 16 weeks. Lipoprotein changes are not usually found following resistance training. The reason for the lack of effect is likely related to the overall caloric expenditure and the exercise volume completed2 1 7. A review of the effect of exercise on lipoproteins in T2D has found that aerobic exercise increases HDL, with some studies also reporting a lowering of LDL, with changes in these lipoprotein fractions in the order of 10% from baseline 1 9 7 ' 2 1 5. The majority of clinical trials which have examined metabolic effects in T2D have examined the effect of aerobic exercise. Only three studies have examined the effect of resistance training in subjects with T2D, and have reported little to modest effect on blood lipids. In non-diabetic subjects a significant increase in HDL is a common finding, with a mean increase of 4.6% reported across studies2 1 8. Significant reductions in LDL, TG and TC with exercise were observed less frequently Our subjects had lipid values that were for the most part at target levels90, and at these levels smaller changes might be expected. Alternatively, modest changes could require a larger sample size to demonstrate an effect. i l l 5.7 Study Strengths and Limitations The main objective of this study was to compare the effect of two different exercise regimes on insulin sensitivity, abdominal AT changes and skeletal muscle characteristics. One of the strengths of this study was the randomized, controlled design. Initial subject characteristics across the three groups did not differ statistically, and initial values for glucose disposal during the euglycemic clamp studies, and glycemic control were very similar, suggesting a successful randomization. Compliance with the exercise programs was very good (93%) and post-training improvements of aerobic fitness (mean increase in peak V02 of 11.4% across both exercise groups) and resistance training loads (in the Ae + RT group) suggest the exercise programs were effective in increasing cardiovascular fitness and muscular strength. The other strengths of this study included the use of hyperinsulinemic-euglycemic clamps, the "gold standard" of assessment of insulin action and CT imaging was used to assess abdominal AT compartments and muscle cross-sectional area, both accurate measures. Exercise programs were designed to minimize differences in energy expenditure between the training groups although energy expenditure was not directly assessed. Participants were instructed to maintain current dietary habits, but dietary intake was not rigorously controlled, and compliance with dietary record taking was poor. Thus it is not known if differences in dietary intake or energy expenditure may have occurred between the groups. Without these controls on dietary intake and energy expenditure we cannot ascertain which was responsible for the modest AT losses. 112 Evaluating regional body composition changes and their relationship to metabolic features such as IR are best understood in the context of the whole body. In this study our only assessment of total body adiposity was the relatively crude measure of BMI. Having a total body assessment, for example DXA, would have allowed us to evaluate abdominal adiposity and muscle changes relative to whole body changes. In addition, using a single slice CT protocol has limitations. Single slice abdominal imaging may be missing subtle differences in abdominal adipose patterning which could be important relative to biological markers such as insulin sensitivity. To avoid this problem, some research groups have used a combination of abdominal DXA and single image CT 6 1 . CT imaging of changes in skeletal muscle density, although assumed to reflect change in muscle lipid content, may reflect changes in other tissue constituents such as protein or water. MRI can be used to quantify interstitial AT in muscle, although the standard error of the estimate is quite high (30%)53. MRS techniques have corroborated the association of IMCL with IR59, although little data is available with respect to changes with exercise or specifically with resistance training using this technique. Furthermore, CT images of muscle cross-sectional area at the thigh may not represent a comprehensive picture of muscle changes as exercise in both the Ae only and the Ae + RT group included upper extremity musculature involvement. Initial levels of total and subcutaneous AT area displayed a trend to differ across groups, the usual care group displaying a trend towards a greater area of total and subcutaneous abdominal AT with change scores for AT areas and muscle density 113 appearing to worsen. This trend may reflect a poorer initial metabolic status, although insulin sensitivity values from glucose clamp studies were similar. Our results are limited to pre and post intervention measures only. Serial measurements during the course of the study would have provided more information on the time course of changes in insulin sensitivity and regional adiposity and if these changes followed a similar pattern. However, the prohibitive cost of the measurements limited the number of times each subject was assessed. 114 Chapter 6: Summary, Conclusions and Recommendations 6.1 Summary This study tested the effect of adding resistance training to an aerobic training program on insulin sensitivity, abdominal obesity and skeletal muscle characteristics in postmenopausal women with T2D. We hypothesized that the combination of aerobic and resistance training would prove an effective exercise modality to improve insulin sensitivity. We also evaluated accompanying changes in abdominal AT and muscle in order to ascertain their relative influences on improved insulin sensitivity. The principal findings are that a combined aerobic plus resistance training program elicited significant improvements in insulin sensitivity although the aerobic only training did not. Significant reductions in visceral AT were seen in the Ae + RT group after the study. When both exercise groups were combined for abdominal AT analyses, significant reductions after training were seen in total abdominal AT and visceral AT. Furthermore, Ae + RT training reduced average muscle density, and increased the area of normal density muscle. Increased glucose disposal during euglycemic clamp studies was inversely related to decreased abdominal subcutaneous and visceral adipose and directly related to increased muscle cross-sectional area and muscle density. These results have potential implications for the prevention and treatment of T2D in postmenopausal women and other insulin-resistant populations. Diabetes is a highly prevalent condition ( 1 1 % of North Americans over the age of 652 1 9) and results in substantial health care and lost productivity costs (over 1 billion dollars annually in Canada1 9 6). The cost of physical inactivity and obesity, two major contributing factors 115 to the development of diabetes, has been estimated to be a min imum of 4.2% of direct health care costs for T2D (U.S.) 7 . Those with diabetes incur greater medical costs than those without diabetes, with cost ratios ranging f rom 2.3-7.8 that of an individual without diabetes. The economic estimates represent only part of the burden of diabetes. Further substantial costs are incurred in indirect costs which include lost productivity f rom diabetes related morbidity and premature mortality. Significant costs associated with the co-morbid condit ions of diabetes (e.g. retinopathy, amputat ion, hypertension, heart disease, peripheral vascular disease and infections) are not always attributed to the underlying d iabe tes 2 1 9 . Economic analyses also do not necessarily include out-of-pocket expenses borne by the individual. Implementing physical activity and modest weight loss programs have been shown to reduce the risk of developing T2D by over 5 0 % 3 6 , and physically active individuals have lower annual direct medical costs than inactive people, a saving estimated to be 76.6 billion do l l a r s 2 2 0 (in the U.S.). The biggest potential saving in direct medical costs would be among women 55 and o lde r 2 2 0 ; indicating that the importance of physical activity in this group. It is clear that the economic costs of diabetes are large, and significant savings can be made if the most inactive portion of the population achieves modest levels of physical activity. Furthermore, prospective data on the prevention of diabetes indicate that a dose-response relationship exists between exercise and development of diabetes, i.e. more vigorous exercise, or higher levels of physical activity in the form of walking, further reduces the relative risk of developing d i a b e t e s 1 3 4 , 2 2 1 . The results of the current study suggest that further gains might be 116 made as we learn more about the optimal type, amount, duration and frequency of exercise to offset deleterious health effects. 6.2 Conclusions 1. A 16 week exercise program which includes aerobic and resistance training components was effective in improving insulin sensitivity in postmenopausal women with T2D, whereas a program of aerobic only training was not. 2. Ae + RT training resulted in a significant reduction of visceral AT after the training program. The combined exercise groups demonstrated a significant reduction in total abdominal AT and visceral AT. The usual care group showed a significant increase in subcutaneous abdominal AT as compared to the combined exercise groups. 3. After Ae + RT training, average muscle density was significantly increased, as was the proportion of normal density muscle. The usual care group displayed a significant decrease in average muscle density, and a significant increase in the proportion of low density muscle. Low density muscle is associated with IR and is thought to reflect ectopic lipid stored within muscle tissue. 4. Improvements in insulin sensitivity are inversely related to changes in total, subcutaneous and visceral abdominal AT, and positively correlated to changes in muscle cross-sectional area and density. The relationship of insulin sensitivity with muscle cross-sectional area and density remained after controlling for the change in 117 total abdominal AT. These results suggest that both abdominal AT area and muscle area and density are factors in IR. 6.3 Recommendations for Future Research The results of the present study suggest that including resistance training with aerobic training can be an effective exercise program to improve IR in postmenopausal women with T2D. As improved insulin sensitivity was related to both a reduction in abdominal AT and increased muscle area and density, it appears that metabolic benefits can be attained not only through a reduction in AT, but also through an increase in muscle tissue and muscle density. Further understanding of this area could be gained if future research investigated the following areas: 1. Total AT is not a homogeneous compartment, but rather a tissue with specific regional depots with varying metabolic functions. These specific depots extend beyond simple subcutaneous and visceral AT, and may include deep and superficial portions of subcutaneous A T 4 7 , 1 0 7 . Ectopic lipid, such as lipid content within muscle, either as interstitial AT or lipid within the myocyte, has been linked to several disease processes, and may also include lipid within other tissues106 such as liver 2 2 2 , 2 2 3, heart and pancreas30. Following the changes in these specific depots after an intervention would help in our understanding of the role of AT in metabolic outcomes such as insulin sensitivity. 118 2. Due to the large intra-individual variability in AT patterning61, multiple-image CT or MRI protocols would increase accuracy106 of change in regional AT patterns relative to change in metabolic measures. 3. Larger sample sizes would be necessary to discriminate the magnitude of improvements in insulin sensitivity induced by different exercise training protocols. 4. A more direct measure of change in intermuscular lipid would help increase our understanding of the role of change in muscle in improved insulin sensitivity with exercise training. Low muscle density in CT image analyses has been interpreted in the literature to reflect excess intermuscular lipid and has been associated with IR, but it cannot quantify the content of lipid, or pinpoint its location to intra- or extra-cellular sites. Recent developments in MRI may prove useful in quantifying interstitial AT within muscle50. 5. The results of this study demonstrate the value of the addition of resistance training to an aerobic training program. Further study is needed to evaluate the independent and combined contribution of exercise modality i.e. resistance training only, aerobic training alone and a combined aerobic plus resistance training program. 119 Hill, J., Wyatt, H, Outpatient managment of obesity: A primary care perspective. Obesity Research, 2002.10: p. 124S-130S. Obesity: Preventing and Managing the Global Epidemic. Report of a WHO Consultation, Geneva, Switzedand. 2000, World Health Organization: Geneva, p. WHO Technical Report Series. World Health Organization, Obesity: preventing and managing the global epidemic of obesity. Report of the WHO consultation of Obesity. Geneva, Switzedand, June 3-5, 1997. see http://www.cflri.ca/cflri/pa/survevs/2001survev/2001survey.html (accessed Oct 23, 2003) Aronne, L, Obesity as a disease: Etiology, treatment, and managment considerations for the obese patient. Obesity Research, 2002.10: p. 95S-104S. Pi-Sunyer, F., The obesityepidmeic: pathophysiology and consequences of obesity. Obesity Research, 2002. 10: p. 97S-104S. Colditz, G., Economic costs of obesity and inactivity. Medicine and Science in Sports and Exercise, 1999. 31: p. S663-S667. Canadian Diabetes Strategy, Draft Blueprint for Action for the National Diabetes Strategy. Diabetes Datahttp://www.hc-sc.qc.ca/pphb-dqspsp/ccdpc-cpcmc/ndss-snsd/enqlich/diabetes data/index Health Canada, Diabetes in Canada (2nd ed.). 2002. Tremblay, M., Katzmarzyk, PT, Willms, JD, Temporal trands in obesity and overweight in Canada 1981-1996. International Journal of Obesity, 2002. 26: p. 538-543. Krotkiewski, M., Lonnroth P., Mandroikas K., Wroblewski Z, Rebuffe-Scrive M., The effects of physical training on insulin secretion and effectiveness and on glucose metabolism in obesity and Type 2 (non-insulin-dependent) diabetes mellitus. Diabetologia, 1985. 28: p. 881-890. Ohlson, L, Larsson, B, Svardsudd, K, Wilin, L, Eriksson, H, Wilhelmsen, L, Bjorntorp, P, Tibblin, G., The influence of body fat distribution on the incidence of diabetes mellituse: 13.5 years of follow-up of the participants in the study of men born in 1913. Diabetes, 1985. 34: p. 1055-1058. Pouliot, M.-C, Despres, J-P, Lemieux, S, Moorjani, S, Bouchard, C, Tremblay, A, Nadeau, A, Lupien, P.J, Waist circumference and abdominal sagittal diameter: best simple anthropometric indixes of abdominal visceral adipose Tissue Accumulation and Related Cardiovascular Risk in Men and Women. The American Journal of Cardiology, 1994. 73: p. 460-468. Ashwell, M., Cole, TJ, Dixon, AK, Obesity: New insight into the anthropometric classification of fat distribution shown by computed tomography. British Medical Journal, 1985. 290: p. 1692-1694. 120 16. Ross, R., Leger, L, Morris, D, de Guise, J, Guardo, R., Quantification of adipose tissue by MRI: relationship with anthropometric variables. Journal of Applied Physiology, 1992. 72: p. 787-795. 17. Sheehan, M., Jensen, MD, Metabolic Complications of Obesity. Medical Clinics of North America, 2000. 84. 18. Abate, N., Garg, A, Peshock, RM, Stray-Gundersen, J, Adams-Huet, B, Grundy, SM, Relationship of generalized and regional adiposity to insulin sensitivity in men with NIDDM. Diabetes, 1996. 45: p. 1684-1693. 19. Alberti, K., Zimmet PZ, Definition, diagnosis and classification of diabetes mellitus and its complications. Pad 1: Diagnosis and classification of diabetes mellitus provisional report of a WHO consultation. Diabetic Medicine, 1998.15: p. 539-553. 20. Expert Panel on Detection, E., and Treatment of High Blood Cholesterol in Adults. Executive summary of the thrid report of the National Cholesterol Education Program Expert Pant on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). Journal of the American Medical Associaltion, 2001. 285: p. 2486-2497. 21. Haffner, S., M. Stern, and H. Hazuda, Cardiovascular risk factors in confirmed prediabetic individuals. Does the clock for coronary head disease start ticking before the onset of clinical diabetes? Journal of the American Medical Association, 1990. 282: p. 1519-22. 22. Reaven, G., Role of Insulin Resistance in Human Disease. Diabetes, 1988. 37: p. 1595-1607. 23. Ruderman, N., Chisholm, D, Pi-Sunyer, X, Schneider, S., The Metabolically Obese, Normal-Weight Individual Revisited. Diabetes, 1998. 47: p. 699-713. 24. Frayn, K., Visceral fat and insulin resistance - causative or correlative? British Journal of Nutrition, 2000. 83: p. S71-S77. 25. DeFronzo, R., Ferraninni, E., Insulin Resistance: A multifaceted syndrome responsible for NIDDM, obesity, hypertension, dyslipidemia, and atherosclerotic cardiovscular disease. Diabetes Care, 1991.14: p. 173-194. 26. Basu, A., et al., Systemic and regional free fatty acid metabolism in type 2 diabetes. American Journal of Physiology - Endocrinology and Metabolsim, 2001. 280: p. E1000-1006. 27. Bjorntorp, P., Portal adipose tissue as a generator of risk factors for cardiovascular disease and diabetes. Arteriosclerosis, 1990.10: p. 495-496. 28. Saltiel, A., You are what you secrete. Nature Medicine, 2001. 7: p. 887-888. 29. Ravussin, E. The role of ectopic fat deposition and the development of insulin resistance, in American Diabetes Association. 2002. San Francisco. 30. Unger, R., Orci, L, Diseases of liporegulation: new perspective on obesity and related disorders. FASEB Journal, 2001.15: p. 312-321. 1 2 1 31. Kelley, D., He, J, Menshikova, EV, Ritov, VB, Dysfunction of mitochondria in human skeletal muscle in type 2 diabetes. Diabetes, 2002. 51: p. 2944-2950. 32. Kahn, B., Flier, JS, Obesity and insulin resistance. Journal of Clinical Investigation, 2000.106: p. 473-481. 33. Riddle, M., Overview of Current Therapeutic Options: Session summary. Diabetes Care, 1999. 22: p. C76-C78. 34. Physical Activity and Health: A report of the surgeon general. Executive Summary, Centers for Disease Control and Prevention. U.S. Deptartment of Health and Human Services. 35. Skerrett, P., Manson, JE, Reduction in Risk of Coronary Heart Disease and Diabetes, in Handbook of Exercise in Diabetes, J.T.D. N. Ruderman, S. H. Schneider, A. Kriska, Editor. 2002, American Diabetes Association: Alexandria, Virginia, p. 155-181. 36. Diabetes Prevention Program Research Group, Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin. New England Journal of Medicine, 2002. 346: p. 393-403. 37. Pan, X.-R., Li, G-W, Hu, Y-H, Wang, J-X, Yang, W-Y, An, Z-X, Hu L-X, Lin, J, Xiao, J-Z, Cao, H-B, Liu, P-A, Jiang, X-G., Jiang, Y-Y, Wang, J-PI, Zheng, H., Zhang, H., Bennet, P. H., Howard, B.V., Effects of diet and exercise in preventing NIDDM in people with impaired glucose tolerance. Diabetes Care, 1997. 20: p. 537544. 38. Toumilehto J., L., J., Eriksson, JG, Vallet, T.T, Hamalainen H., Ilanne-Parikka, P., Keinanen-kiukaaniemi, S., Laakso, M., Louheranta, A., Rastas, M, Salminen, V., Uusitupa, M.,, Prevention of type 2 diabetes mellitus by changes in lifestyle among subjects with impaired glucose tolerance. New England Jouranl of Medicine, 2001. 344: p. 1343-1350. 39. Ivy, J.L., Exercise Physiology and Adaptations to Training, in Handbook of Exercise in Diabetes, N. Ruderman, Devlin, J.T., Schneier, S.H., Kriska, A., Editor. 2002, American Diabetes Association: Alexandria, p. 23-62. 40. Clinical Guidelines on the identification, evaluation, and treatment of overweight and obesity in adults: the evidence report, in Obesity Research. 1998, National Institutes of Health, p. 51S-209S. 41. Ross, R., Dagnone, D, Jones, PJH, Smith, H, Paddags, A., Hudson, R., Janssen, I., Reduction in obesity and related comorbid conditions after diet-induced weight loss or exercise-induced weight loss in men. Annals of Internal Medicine, 2000.133: p. 92-103. 42. Krotkiewski, M., Bjorntorp, P., Muscle tissue in obesity with different distribution of adipose tissue. Effects of physical training. International Journal of Obesity, 1986.10: p. 331-341. 43. Aronne, L., Classification of obesity and assessmnet of obesity-related health nsks. Obesity Research, 2002.10: p. 105S-115S. 44. Ross, R., Goodpaster, B, Kelley, D, Boada, F, Magnetic resonance imaging in human body composition research: from quantitative to qualitative tissue measurement. Annals of New York Academy of Science, 2000. 904: p. 12-17. 122 45. Shen, W., Wang, Z, Punyanita, M, Lei, J, Sinav, H, Krai, JG, Imielinska, C, Ross, R, Heymsfield, SB, Adipose tissue quantification by imaging methods: a proposed classification. Obesity Research, 2003.11: p. 5-16. 46. Kelley, D., Thaete, FL, Troost, F, Huwe, T, Goodpaster, BH, Subdivisions of subcutaneous abdominal adipose tissue and insulin resistance. American Journal of Physiology, Endocrinology and Metabolism, 2000. 278: p. E941-E948. 47. Smith, S., Lovejoy, JC, Greenway, F, Ryan, D, deJonge, L, de la Bretonne, J, Volafova, J, Bray, GA, Contributions of total body fat, abdominal subcutaneous adipose tissue compadments, and visceral adipose tissue to the metabolic complications ofobesity*1. Metabolism, 2001. 50: p. 425-435. 48. Goodpaster, B.H., Thaete, FL, Kelley, DE, Composition of skeletal muscle evaluated with computed tomography. Annals New York Academy of Science, 2000. 904: p. 18-24. 49. Plourde, G., The role of radiologic methods in assessing body composition and related metabolic parameters. Nutrition Reviews, 1997. 55: p. 289-296. 50. Goodpaster, B., Measuring body fat distribution and content in humans. Current Opionions in Clinical Nutrition and Metabolism Care, 2002. 5: p. 481-487. 51. van der Kooy, S., J.C, Techniques for the measurement of visceral fat: a practical guide. International Journal of Obesity, 1993.17: p. 187-196. 52. Despres, J.-P., Imaging techniques applied to the measurement of body composition., in Human Body Composition, A. Roche, Heymsfield, S, Lohman, T, Editor. 1996, Human Kinetics: Windsor, Ont. p. 149-166. 53. Mitsiopoulos, N., et al., Cadaver validation of skeletal muscle measurement by magnetic resonance imaging and computerized tomography. J Appl Physiol, 1998. 85: p. 115-122. 54. Abate, N., Burns D, Peshock, RM, Garg, A, Grundy SM, Estimation of adipose tissue mass by magnetic resonance imaging: validation against dissection in human cadavers. Journal of Lipid Research, 1994. 35: p. 1490-1496. 55. Kohrt, W., Kirwin, JP, Staten MA, Bourcy RE, King DS, Hollowszy, JO, Body composition by DXA: tried and true? Medicine Science in Sports and Exercise, 1995. 27: p. 1349-1353. 56. Phillips, D., Caddy, S, llic, V, Fielding, BA, Frayn, KN, Borthwick, AC, Taylor, R., Intramuscular triglyceride and muscle insulin sensitivity: evidence for a relationship in nondiabetic subjects. Metabolism, 1996. 45: p. 947-950. 57. Goodpaster, B., Kelley, DE, Thaete, FL, He, J, Ross, R, Skeletal muscle attentuation determined by computed tomography is associated with skeletal muscle lipid content. Journal of Applied Physiology, 2000. 89: p. 104-110. 58. Elbers, J., Haumann, G, Asscheman, H, Seidell, JC, Gooren, LJG, Reproducibility of fat area measurements in young, non-obese subjects by computerized analysis of magnetic resonance images. International Journal of Obesity, 1997. 21: p. 1121-1129. 123 59. Schick, F., Machman, J, Bretchtel, K, Strempfer, A, Klumpp, B., Stein, DT, Jacob, S, MRI of Muscular Fat. Magnetic Resonance in Medicine, 2002. 47: p. 720-727. 60. McGarry, J., Banting Lecure 2001. Dysregulation of fatty acid metabolism in the etiology of type 2 diabetes. Diabetes, 2002. 51: p. 7-18. 61. Greenfield, J., Samaras, K, Chishom, DJ, Campbell, LV, Regional Intra-subject variability in abdominal adiposity limits usefulness of computed tomography. Obesity Research, 2002.10: p. 260-265. 62. Rice, T., Despres, JP, Daw, EW, Gagnon, J, Borecki, IB, Perusse, L, Leon, AS, Skinner, JS, Wilmore JH, Rao, DC, Bouchard, C, Familial resemblance for abdominal visceral fat: the HERITAGE family study. International Journal of Obesity, 1997. 21: p. 1024-1031. 63. Bouchard, C , J. Despres, and P. Mauriege, Genetic and nongenetic determinants of regional fat distribution. Endocrinology Reviews, 1993.14: p. 72-93. 64. Lemieux, S., Prud'homme, D, Bouchard C, Tremblay, A, Despres, JP, Sex differences in the relation of visceral adipose tissue accumulation to total body fatness. American Journal of Clinical Nutrition, 1993. 58: p. 463-467. 65. Despres, J., Couillard, C, Bergeron, J, Lamarche, B, Regional Body fat distribtuion, the insulin resistance-dyslipidemic syndrome, and the risk of type 2 diabetes and coronary head disease., in Handbook of Exercise in Diabetes, N. Ruderman, Devlin, JT, Schneier, SH, Kriska, A, Editor. 2000, American Diabetes Association: Alexandria. 66. Abate, N., Garg, A, Peshock, RM, Stray-Gundersen, J, Grundy, SM, Relationships of generalized and regional adiposity to nsulin sensitivity in men. Journal of Clinical Investigation, 1995. 96: p. 88-98. 67. Ostenson, C.-G., The Pathophysiology of type 2 diabetes. Acta Physiologica Scandinavica, 2001. 171: p. 241-247. 68. Kahn, B., Insulin resistance, insulin insensitivity, and insulin unresponsiveness; a necessary distinction. Metabolism, 1978. 27: p. 1893-1902. 69. Scheen, A., Paquot, N, Castillo MG, How to measure insulin action in vivo. Diabetes /Metabolism Reviews, 1994.10: p. 151-188. 70. Starke, A.A.R., Determination of insulin sensitivity: methodological considerations. Journal of Cardiovascular Pharmacology, 1992. 20: p. S17-S21. 71. Himsworth, H., Diabetes mellitus - its differentiation into insulin sensitive and insulin insensitive types. Lancet, 1936.1: p. 127-130. 72. Lakka, H., Lakka TA, Tuomilehto J, Salonen JT, Abdominal obesity is associated with increased risk of acute coronary events in men. European Heart Journal, 2002. 23: p. 706-713. 124 73. Haffner, S., Diabetes, metabolic syndrome, and coronary head disease:. 2003, American Diabetes Meeting: New Orleans. 74. Kelley, D., Goodpaster, B, Wing RR, Simoneau, JA, Overview: what is insulin resistance? Nutrition Reviews, 2000. 58: p. S2-S3. 75. Ryan, A., Insulin resistance with aging: effects of diet and exercise. Sports Medicine, 2000. 30: p. 327-346. 76. Elahi, D., In praise of the hyperglycemic clamp. Diabetes Care, 1996.19: p. 278-286. 77. DeFronzo, R., Tobin, JD, Andres, R, Glucose clamp technique: a method for quantifying insulin secretion and resistance. American Journal of Physiology, 1979. 237: p. E214-E223. 78. Haffner, S., Mykkanen, L, Valdez, RA, Stern, MP, Evaluation of two insulin assays in insulin resistance syndrome (syndrome X). Arteriosclerosis and Thrombosis, 1994.14: p. 1430-1437. 79. Matthews, D., Hosker, JP, Rudenski. AS, Naylor, BA, Treacher, D., Turner, RC, Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia, 1985. 28: p. 412-419. 80. Bonora, E., Kiechl, S, Willeit, J., Oberhollenzer, F, Egger, G, Targher, G, Alberiche, M, Bonadonna, RC, Muggeo, M., Prevalence of insulin resistance in metabolicdisorders: The Bruneck Study. Diabetes, 1998. 47: p. 1643-1649. 81. Sievenpiper, J., Jenkins AL, Whitham, DL, Vuksan, V., Insulin Resistance: concepts, controversies, and the role of nutrition. Canadian Journal of Dietetic Practice and Research, 2002. 63: p. 20-32. 82. Isomaa, B., Almegren P, Tuomi T, et al., Cardiovascular moridity and modality associated with the metabolic syndrome. Diabetes Care, 2001. 2001: p. 683-689. 83. Ford, E., Giles, WH, Dietz, WH., Prevalence of the metabolic syndrome among US adults. Journal of the American Medical Assocication, 2002. 287: p. 356-359. 84. Ekoe, J., Epidemiology of type 2 diabetes mellitus in North America. CDJC, 1999. 23: p. 43-46. 85. Ferrannini, E., Vichi, S, Beck-Nielson, H, Laakso, M, Paoliso, G., Smith, J, Insulin action and age. Diabetes, 1996. 45: p. 947-953. 86. Barbieri, M., Rizzo, MR, Manzella, D, Paolisso, G, Age-related insulin resistance: is it an obligatory finding? The lesson from health centenarians. Diabetes/Metabolism Research and Reviews, 2001.17: p. 19-26. 87. Celafu, W., Wang ZQ, Werbel.S., Contribution of visceral fat mass to the insulin resistance of aging. Metabolism, 1995. 44: p. 954-959. 88. Ginsberg, H., Treatment for patients with the metabolic syndrome. American Journal of Cardiology, 2003. 91 (suppl): p. 29E-39E. 125 89. Zimmet, P., Albert', KGMM, Shaw, J, Global and societal implications of the diabetes epidemic. Nature, 2001. 414: p. 782-787. 90. Meltzer, S., Leiter, L, Daneman, D, Gerstein, HC, Lau, D, Ludwig, S, Yale, J-F, Zinman, B, Lillie, D, 1998 Clinical Practice Guidelines for the Management of diabetes in Canada. Canadian Medical Association, 1998.159: p. S1-S29. 91. Defronzo, R., Pathogenesis of type 2 diabetes: metabolic and molecular implications for identifying diabetes genes. Diabetes Reviews, 1997. 5: p. 177-269. 92. Shulman, G., Cellular mechanisms of insulin resistance in humans. American Journal of Cardiology, 1999. 84(1A): p. 3J-10J. 93. Stern, M., P, Strategies and prospects for finding insulin resistance genes. Journal of Clinical Investigation, 2000.106: p. 323-327. 94. Goodpaster, B., Thaete, FL, Simoneau, J-A, Kelley, DE, Subcutaneous abdominal fat and thigh muscle composition predict insulin sensitivity independently of visceral fat. Diabetes, 1997. 46: p. 1579-1585. 95. Eriksson, J., Taimela S, Koivisto, VA, Exercise and the Metabolic Syndrome. Diabetologia, 1997.40: p. 125-135. 96. Kissebah, A., Drakower, GR, Regional adiposity and morbidity. Physiological Reviews, 1994. 74: p. 761-805. 97. Bjorntorp, P., Metabolicimplications of body fat distribution. Diabetes Care, 1994.14: p. 1132-1143. 98. Despres, J., S. Moorjani, and T. Lupien PJ, A, Nadeau, A, Bouchard, C, Regional distribution of body fat, plasma lipoproteins, and cardiovascular disease. Arteriosclerosis, 1990.10: p. 497-511. 99. Pouliot, J.-C, Despres, J-P, Nadeau, A, Moorjani, S, Prud'homme, D, Lupien, J J, Tremblay, A, Bouchard, C , Visceral obesity in men: associations with glucose tolerance, plasma insulin, and lipoprotein levels. Diabetes, 1992. 41: p. 826-834. 100. Despres, J.-P., Nadeau, A, Tremblay, A, Ferland, M, Moorjani, S, Lupien, JJ, Theriault, G, Pinault, S, Bouchard, C , Role of deep abdominal fat in the association between regional adipose tissue distribution and glucose tolerance in obese women. Diabetes, 1989. 38: p. 304-309. 101. Despres, J.-P., Pouliot, M-C, Moorjani, S, Nadeau, A, Tramblay, A, Lupien, PJ, Theriault, Bouchard, C, Loss of abdominal fat and metabolic response to exercise training in obese women. American Journal of Physiology, 1991. 261: p. E159-E167. 102. Carey, D., Jenkins, AB, Campbell, LV, Freund, J, Chisholm, DJ, Abdominal fat and insulin resistance in normal and overweight women:direct measurements reveal a strong relationship in subjects at both low and high risk of NIDDM. Diabetes, 1996. 45: p. 633-638. 1 2 6 103. Simoneau, J.-A., Colberg, S R, Thaete, L, Kelley, D E, Skeletal muscle glycolytic and oxidative enzyme capacities are determinants of insulin sensitivity and muscle composition in obese women. FASEB Journal, 1995. 9: p. 273-278. 104. Albu, J., Kovera AJ, Johnson, J, Fat distribution and health in obesity.: p. 491-501. 105. Seidell, J., Bouchard, C, Visceral Fat in relation to health: is it a major culprit or simply an innocent bystander? International Journal of Obesity, 1997. 21: p. 626-631. 106. Thomas, E., Saeed, N, Hajnal, JV, Brynes, A, Godstone, AP, Frost, G, Bell JD, Magnetic resonance imaging of total body fat. Journal of Applied Physiology, 1998. 85: p. 1778-1785. 107. Goodpaster, B., Thaete, FL, Kelley, DE, Thigh adipose tissue distribution is associated with insulin resistance in obesity and in type 2 diabetes mellitus. American Journal of Clinical Nutrition, 2000. 71: p. 885-892. 108. Kaplan, N., The deadly quartet: upper-body obesity, glucose intolerance, hypertriglyceridemia, and hypertension. Archives of Internal Medicine, 1989.149: p. 1514-1520. 109. Haffner, S.M., Valdez, RA, Hazuda, HP, Matchell, BD, Morales, PA, Stern, MP, Prospective analysis of the insulin-resistance syndrome (syndrome X). Diabetes, 1992. 41: p. 715-722. 110. Randle, P., Regulatory interactions between lipids and carbohydrates: the glucose fatty acid cycle after 35 years. Diabetes/Metabolism Reviews, 1998. 14: p. 263-283. 111. Bergman, R., Ader, M, Free fatty acids and pathogenesis of type 2 diabetes mellitus. Trends in Endocrinolology and Metabolsim, 2000.11: p. 351-356. 112. Montaque, O, O'Rahilly, S., The Perils of Portliness. Diabetes, 2000. 49: p. 883-888. 113. Mauriege, P., Brochu, M, Prud'homme, Tramblay, A, Nadeau, A, Lemieux, S, Despres, J-P, Is Visceral adiposity a significant correlate of subcutaneous adipose cell lipolysis in men? Journal of Clinical Endocrinology and Metabolism, 1999. 84: p. 736-742. 114. Weyer, C , Foley, JE, Bogardus, C, Tataranni, PA, Pratley, RE, Enlarged subcutaneous abdominal adipocyte size, but not obesity itself, predicts Type II diabetes independent of insulin resistance. Diabetologia, 2000. 43: p. 1498-1506. 115. Ravussin, E., Smith, SR, Increased fat intake, impaired fat oxidation, and failure offatcell proliferation result in ectopic fat storage,insulin resistance, and type 2 diabetes mellitus. Annals of the New York Academy of Science, 2002. 967: p. 363-378. 116. Moller, D., New drug targets for type 2 diabetes and the metabolic syndrome. Nature, 2001. 414: p. 821-827. 117. Kelley, D., Slasky, BS, Janosky, J, Skeletal muscle density: effects of obesity and non-insulin-dependent diabetes mellitus. American Journal of Clinical Nutrition, 1991. 54: p. 509-515. 127 118. Pan, D., Lillioja, S, Milner, MR, Kriketos, AD, Baur, LA, Bogardus, C, Storlien, L H, Skeletal muscle membrane lipid composition is related to adiposity and insulin action. Journal of Clinical Investigation, 1995. 96. p. 2802-2808. 119. Stein, D., Dobbins, R., Szczepaniak L, Malloy, C, McGarry, JD, Skeletal muscle triglyceride stores are increased in insulin resistance (abstract). Diabetes, 1997. 46 (suppl 1): p. 23A. 120. Levin, K., Schroeder, HD, Alford FP, Beck-Nielsen, H, Morphometric documentation of abnormal intramyhocellular fat storage and reduced Igycogen in obese patients with type 2 diabetes. Diabetologia, 2001. 44: p. 824-833. 121. Perseghin, G., Scifo, P, De Cobellik, F, Pagliato, E, Battezzati, A, Arcelloni, C, Vanzullli, A, Testolin, G, Pozza, G, Del Machio, A, Luzi L, Intramhyocellular triglyceride content is a determinant of in vivo insulin resistance in humans. Diabetes, 1999. 48: p. 1600-1607. 122. Szsczepaniak LS, B., EE, Schick, F, Dobbins, RL, Garg, A, Burns DB, McGarry JD, Measurement of intracellular triglyceride stores by 1H spectorscopy: validation in vivo. American Journal of Physiology, 1999. 276: p. E977-E989. 123. Kelley, D., Goodpaster, BH, Storlien, L, Muscle triglyceride and insulin resistance. Annual Reviews Nutrition, 2002. 22: p. 325-346. 124. Olefsky, J., Treatment of insulin resistance with peroxisome proliferator-activated receptor gamma agonists. Journal of Clinical Investigation, 2000.106: p. 467-472. 125. Aroda, V., Henry, RR, Thiazolidinediones: Potential link between insulin resistance and cardiovascular disease. Diabetes Spectrum, 2003.16: p. 120-125. 126. Miyazaki, Y., Mahankali, A, Matsuda, M, Mahankali, S, Cusi, K, Mandarino LJ, DeFronzo, RA, Effect of pioglitazone on abdominal fat distribution and insulin sensitvity in type 2 daibetic patients. Journal of Clinical Endocrinology and Metabolism, 2002. 87: p. 2784-2791. 127. Flier, J., The adipocyte: storage depot or node on the energy information superhighway? Cell, 1995. 80: p. 15-18. 128. Friedman, J., Obesity in the new millennium. Nature, 2000. 404: p. 632-634. 129. Saltiel, A., Kahn, R, Insulin signalling and the regulation of glucose and lipid metabosm. Nature, 2001. 414: p. 799-806. 130. Bjorntorp, P., Neuroendocrine factors in obesity. Journal of Endocrinology, 1997.155: p. 193-195. 131. Bjorntorp, P., Metabolic Implications of body fat distribution. Diabetes Care, 1991.14: p. 1132-1143. 132. Porte, D., Seeley, RJ, Woods, SC, Baskin, DG, Figlewicz, DP, Schwartz, MW, Obesity, diabetes and the central nervous system. Diabetologia, 1998. 41: p. 863-881. 128 133. American College of Sports Medicine, ACSM's Guidelines for Exercise Testing and Prescription. 6th ed. ed, ed. B.A. Franklin. 2000, Baltimore: Lippincott, Williams & Wilkins. 368. 134. Wannamethee, S., Shaper, AG, Alberti, GMM, Physical activity, metabolic risk factors, and the incidence of coronary heart disease (CHD) and type 2 diabetes. Archives of Internal Medicine, 2000.160: p. 2108-2116. 135. Wei, M., Gibbons, LW, Kampert, JB, Nichamean, MZ, Blair, SN, Low Cardiorespiratory Fitness and Physical Inactivity as Predictors of Mortality in Men with Type 2 Diabetes. Annals of Internal Medicine, 2000.132: p. 605-611. 136. Eriksson, K.-F., Lindgarde, F, Prevention of Type 2 (non-insulin-dependent) diabetes mellitus by diet and physical exercise. The 6-year Malmo feasibility study. Diabetologia, 1991. 34: p. 891-898. 137. Mikines, K.J., Sonne, B, Farrell, PA, Tronier, B, Galbo, H, Effect of physical exercise on sensitivity and responsiveness to insulin in humans. American Journal of Physiology, 1988. 254: p. E248-E259. 138. Perseghin, G., Price, TB, Petersen, KF, Roden, M, Cline, GW, Gerow, K, Rotheman, DL, Shulman, Gl, Increased glucose transport-phosphorylation and muscle glycogen sythesis after exercise training in insulin-resistant subjects. New England Journal of Medicine, 1996. 335: p. 1357-1362. 139. Devlin, J., Hirshman, M , Horton, ED, Horton, ES, Enhance peripheral and splanchnic insulin sensitivity in NIDDM men after single bout of exercise. Diabetes, 1987: p. 434-439. 140. Dela, F., et al., Insulin-Stimulated Muscle Glucose Clearance in Patients with NIDDM. Diabetes, 1995. 44: p. 1010-1020. 141. Borghouts, L. and H. Keizer, Exercise and Insulin Sensitivity: A Review. Int J Sports Med, 2000. 20: p. 1-12. 142. Kahn, S., Larson, VG, Beard, JC, Cain, KC, Fellingham, GW, Schwartz, RS, Veith, RC, Stratton, JR, Cerqueira, MD, Abrass, IB, Effect of exercise on insulin action, glucose tolerance, and insulin secretion in aging. American Journal of Physiology, 1990. 258: p. E937-E934. 143. Kirwan, J., Kohrt, WM, Wojta DM, Bourey RE, Holloszy, JO, Endurance exercise training reduces glucose-stimulated insulin levels in 60-70 year old men and women. Journal of Gerontology, 1993. 48: p. M84-M90. 144. Tonino, R., Effect of physical training on the insulin resistance of aging. American Journal of Physiology, 1989. 256. p. E352-E356. 145. DeFronzo, R., Sherwin, RS, Kraemer, N, Effect of physical training on ilnsulin action in obesity. Diabetes, 1987. 36: p. 1379-1385. 146. Trovati, M., Carta, Qu, Cavalot, F, Vitali, S, Banaudi, C, Gucchina, P G., Fiocchi, F, Emanuelli, G, Lenti, G., Influence of physical training on blood glucose control, glucose tolerance, insulin 129 secretion, and insulin action in non-insulin-dependent diabetic patients. Diabetes Care, 1984. 7: p. 416-420. 147. Reitman, J., Vasquez B, Klimes I, Nagulesparan M., Improvement of glucose homeostasis after exercise training in non-insulin-depehdent diabetes. Diabetes Care, 1984. 7: p. 434-441. 148. Mourier, A., Gautier, J-F, De Kerviler, E, Bigar, AX, Villette, J-M, Gamier, J-P, Duvallet, A, Guezennec, CY, Cathelineau, G., Mobilization of visceral adipose tissue related to the improvement in insulin sensitivity in response tophysical training in NIDDM. Diabetes Care, 1997. 20: p. 385-391. 149. Hughes, V., Maria A. Fiatarone, et al., Exercise increases muscle GLUT-4 levels and insulin action in subjects with impaired glucose tolerance. American Journal of Physiology, 1993. 264: p. E855-862. 150. Bogardus, C , Ravussin E, Robbins, DC, Wolfe RR, Horton ES, Sims EAH, Effects of physicaltraining and diet therapy on carbohydrate metabolism in patients with glucoseintolerance and non-insulin-dependent diabetes mellitus. Diabetes, 1984. 33: p. 311-318. 151. Poehlman, D., RV, DeNino, WF, Brochu, M, Ades, PA, Effects of resistance training and endurance training on insulin sensitivity in nonobese, young women: a controlled randomized trial. Clinical Endocrinology and Metabolism, 2000. 85: p. 2463-2468. 152. Borghouts, L, Backx, K, Mensink, MF, Keizer, HA, Effect of training intensity on insulin sensitivity as evaluted by insulin tolerance test. European Journal of Applied Physiology, 1999. 80: p. 461-466. 153. Kang, J., Robertson, RJ, Hagberg, JM., Delley, DE, Goss, FL, DaSilva, SG, Suminski, RR, Utter, AC, Effect of exercise intensity on glucose and insulin metabolism in obese individuals and obese NIDDM patients. Diabetes Care, 1996.19: p. 341-349. 154. Kang, J., Kelley, DE, Robertson, RJ, Goss, FL, Suminski, RR, Utter, AC, DaSilva, SG, Substrate utilization and glucose turnover during exercise of varying intensities in individuals with NIDDM. Medicine Science in Sports & Exercise, 1998. 31: p. 82-89. 155. van Baak, M., Borghouts, LB, Relationships with physical activity. Nutrition Reviews, 2000. 58: p. S16-18. 156. Saltin, B., Lindgarde, F, Houston, M, Horlin, R, Nygaard, E, Gad, P, Physical training and glucose tolerance in middle-aged men with chemical diabetes. Diabetes, 1979. 28: p. 30-32. 157. Ruderman, N., Ganda OP, Johansen K, The effect of physical training on glucose tolerance and plasma lipids in maturity-onset diabetes. Diabetes, 1979. 28: p. 89-92. 158. Holloszy, J., Schultz, J, Kusnierkiewicz, J, Hagberg, JM, Ehsani, AA, Effects of exercise on glucose tolerance and insulin resistance. Acta Medica Scandinavia, 1986. [suppl]711: p. 55-65. n o 159. Eriksson, J., Tuominen, J, Valle, T, Sundberg, S., Sovijarvi, A, Lindholem, H, Tuomilehto, J, Koivisto, V, Aerobic Endurance Exercise or Circuit-Type Resistance Training for Individuals with Impaired Glucose Tolerance? Hormones and Metabolism Research, 1998. 30: p. 37-41. 160. Rogers, M., Yamamoto C, King, DS,, Improvement in glucose tolerance after 1 week of exercise in patients with NIDDM. Diabetes Care, 1988.11: p. 613-618. 161. Despres, J., Pouliot, MC, Moorjani, S, Loss of abdominal fat and metabolic response to exercise training in obese women. American Journal of Physiology, 1991. 261: p. E159-E167. 162. Ivy, J.L., Zderic, T.W., Fogt, D.L., Prevention and Treatment of Non-lnsulin-Dependent Diabetes Mellitus, in Exercise and Spod Sciences Reviews, J.O. Hollowsay, Editor. 1999, Williams & Wilkins: Baltimore. 163. Miller, J., Pratley, RE, Goldberg, AP, Gordon, P, Rubin, M, Treuth, MS , Ryan, A S, Hurley, B., Strength training increases insulin action in healthy 50-65 year old men. Journal of Applied Physiiology, 1994. 77: p. 1122-1127. 164. Ryan, S., Pratley, RE, Goldberg, AP, Elahi, D., Resistive training increases insulin action in postmenopausal women. Journal of Gerontology, 1996. 51A: p. M199-M205. 165. Fluckey, J., Hickey, MS, Brambrink, JK, Hart, KK, Alexander, K, Craig, BW, Effects of resistance exercise on glucose tolerance in normal and glucose-intolerant subjects. Journal of Applied Physiology, 1994. 77: p. 1087-1092. 166. Dunstan, D., Puddey, IB, Beilin, LJ, Burke, V, Morton, AR, Stanton, KG, Effects of short-term circuit weight training program on glycaemic control in NIDDM. Diabetes Research and Clinical Practice, 1998. 40: p. 53-61. 167. Wallace, M., Mills, BD, Browning, CL, Effects of cross-training on markers of insulin resistance/hyperinsulinemia. Medicine and Science in Sports and Exercise, 1997. 29: p. 1170-168. Goodyear, L. and B. Kahn, Exercise, gluccose transport, and insulin sensitivity. Annual Reviews of Medicine, 1998. 49: p. 235-261. 169. Henriksen, E., Effects of acute exercise and exercise training on insulin resistance. Journal of Applied Physiology, 2002. 93: p. 788-796. 170. Shepherd, P., Kahn, BB, Glucose transporters and insulin action - implications for insulin resistance and diabetes mellitus. New England Journal of Medicine, 1999. 341: p. 248-257. 171. Goodpaster, B., Kelley, DE, Wing, RR, Meiet, A, Thaete, FL, Effects of weight loss on regional fat distribution and insulin sensitivity in obesity. Diabetes, 1999. 48: p. 839-847. 172. Janssen, I., Fortier, A., Hudson, R., Ross, R., Effects of an energy-restnctive diet with or without exercise on abdominal fat, intermuscular fat, and metabolic nsk factors in obese women. Diabetes Care, 2002. 25: p. 431-438. 131 173. Leenen, R., van der Kooy, K, Droop, A, Seidell, JC, Deurenberg, P, Weststrate, JA, Hautvast, JGAJ, Visceral fat loss measured by magnetic resonance imaging in relation to changes in serum lipid levels of obese men and women. Arteriosclerosis and Thrombosis, 1993.13: p. 487^94. 174. Rice, B., Janssen, I, Hudson, R, Ross, R, Effects of aerobic or resistance exercise and/or diet on glucose tolerance and plasma insulin levels in obese men. Diabetes Care, 1999. 22: p. 684-691. 175. Zamboni, M., Armellini, F, Turcato, E, Todesco, T, Bissoli, L, Bergamp-Andreis, I.A., Bosello, O., Effect of weight loss on regional body fat distribution in premenopausal women. American Journal of Clinical Nutrition, 1993. 58: p. 29-34. 176. Van der Kooy, K., Leenen. R, Seidell, JC, Deurenberg, P, Droop, A, Bakker, CJG, Waist-Hip ratio is a Poor Predictor of Changes in Visceral Fat. American Journal of Clinical Nutrition, 1993. 57: p. 327-333. 177. Irwin, M., Yasui, Y., Ulrich, CM, Bown, D., Rudolph, R.E., Schwartz, R.S., Yukawa, M., Aiello, E., Potter, J.D., McTiernan, A., Effect of Exercises on Total and Intra-abdominal Body Fat in Postmenopausal Women. JAMA, 2003. 289: p. 323-330. 178. Treuth, M., Hunter GR, Kekes-Szabo, Weinsier, RL, Goran, Ml, Berland.L, Reduction in intra-abdominal adipose tissue after strength training in older women. Journal of Applied Physiology, 1995. 78: p. 1425-1431. 179. Wilmore, J., Despres, J-P, Stanforth, PR., Mandel, S, Rice, T, Gagnon, J, Leon, A.S., Rao, DC, Skinner, JS, Bouchard, C , Alterations in body weight and composition consequent to 20 wk of endurance training: the HERITAGE Family Study. American Journal of Clinical Nutrition, 1999. 70: p. 346-352. 180. Ryan, A., Nicklas, BJ, Berman, DM, Dennis, KE, Dietary restriction and walking reduce fat-deposition in the midthigh in obese older women. American Journal of Clinical Nutrition, 2000. 72: p. 708-713. 181. Ross, R., Freeman, JA, Janssen, Ian, Exercise alone is an effective strategy for reducing obesity and related comorbidities. Exercise and Sport Science Reviews, 2000. 28: p. 165-170. 182. Ross, R., Janssen, I, Is abdominal fat preferentially reduced in response to exercise-induced weight loss? Medicine & Science in Sports & Exercise, 1999. 31: p. S568-S572. 183. Sipila, S., Suominen, H., Effects of strength and endurance training on thigh and leg muscle mass and composition in eldedy women. Journal of Applied Physiology, 1995. 78: p. 334-340. 184. Borghouts, L., Keizer, HA, Exercise and insulin sensitivity: a review. Internal Journal of Sports Medicine, 2000. 21: p. 1-12. 185. Ross, R., Jannssen, I, Physical activity, total and regional obesity: Dose-response considerations. Medicine and Science in Sports and Exercise, 2001. 33: p. S521-S527. 186. Kelley, D., Goodpaster, BH, Effects of physical activity on insulin action and glucose tolerance in obesity. Medicine & Science in Sports & Exercise, 1999. 31: p. S619-S623. 132 187. Bradley, D., DATASIM. 1988, Desktop Press: Lewiston, ME. 188. Thomas, J., Nelson, JK, Research Methods in Physical Activity. 1990, Human Kinetics Books: Champaign, Illinois. 189. Ainsworth, B., Haskell, WL, Leon, AS, Jacobs, DR, Montoye HJ, Sallis, JF, Paffenbarger, RS, Compendium of Physical Activities: classification of energy costs of human physical activities. Medicine and Science in Sports and Exericse, 1993. 25: p. 71-80. 190. McArdle, W., Katch, Fl, Katch, V., Exercise Physiology: Energy, Nutrition, and Human Pedormance. 3rd ed. 1991, Philadelphia: Lea & Febiger. 853. 191. Ross, R., Rissanen, J, Mobilization of visceral and subcutaneous adipose tissue in response to energy restriction and exercise. American Journal of Clinical Nutrition, 1994. 60: p. 695-703. 192. Meneilly, G., Elliiott, T., Metabolic alterations in middle-aged and eldedy obese patients with type 2 diabetes. Diabetes Care, 1999. 22: p. 112-118. 193. Friedewald, W., Levy, Rl, Fredrickson, DS, Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clinical Chemistry, 1972.18: p. 499-502. 194. American Diabetes Association, Medications for the Treatment of Diabetes, ed. J.W. RK Campbell. 2000, Alexandria: American Diabetes Association. 195. Segal, K., Edana, A, Abalos, A, Effect of exercise training on insulin sensitivity and glucose metabolism in lean, obese, and diabetic men. Journal of Applied Physiology, 1991. 71: p. 2402-2411. 196. 2001 Physical Activity Monitor.Canadian Fitness and Lifestyle Research Insititute.available at http://www.cflri.ca/cflri/news/99/9903 1 .htm , 197. Kelley, D., Boodpaster, BH,, Effects of exercise on glucose homeostasis in Type 2 diabetes mellitus. Medicine and Science in Sports and Exercise, 2001. 33: p. S495-S501. 198. Gregg, Arch Intern Med, 2003.163: p. 1397-. 199. Ryan, A.S., Hurlbut, DE, Lott, ME, Ivey, FM, Fleg, J, Hurley, BF, Goldberg, AP, Insulin action after resistive training in insulin resistant older men and women. Journal of the American Geriatric Society, 2001. 49: p. 247-253. 200. Soukup, J.T., Kovaleski J E, A review of the effects of resistance training for individuals with diabetes mellitus. The Diabetes Educator, 1993.19: p. 307-312. 201. Nelson, M.E., et al., Analysis of body composition techniques and models for detecting change in soft tissue with strength training. American Journal of Clinical Nutrition, 1996. 63: p. 678-686. 133 202. Miller, W., Koceja, DM, Hamilton, EJ, A meta-analysis of the past 25 years of weight loss research using diet, exercise or diet plus exercise intervention. International Journal of Obesity Research, 1997. 21: p. 941-947. 203. Despres, J.-P., Visceral obesity,insulin resistance, and dyslipidemia: contribution of endurance exercise training to the treatment of the plurimetabolic syndrome, in Exercise and Spod Sciences Reviews, J.O. Hollowszy, Editor. 1997, Williams and Wilkins: Baltimore. 204. Simoneau, J.-A., Adaptation of human skeletal muscle to exercise-training. International Journal of Obesity, 1995.19: p. S9-S11. 205. Goodpaster, B., Theriault, R, Watkins, SC, Kelley, DE, Intramuscular lipid content is increased in obesity and decreased by weight loss. Metabolism, 1999. 49: p. 467-472. 206. Goodpaster, B., He, J, Watkins, S, Kelley, DE, Skeletal muscle lipid content and insulin resistance: evidence for a paradox in endurance-trained athletes. The Journal of Clinical Endocrinology and Metabolism., 2001. 86: p. 5755-5761. 207. Nyholm, B., Mengel, A, Nielsen, S, Skjaerbaek, C, Moller, N, Alberti, KGMM, Schmitz, O, Insulin Resistance in relatives of NIDDM patient: the role of physical fitness and muscle metabolism. Diabetologia, 1996. 39: p. 813-822. 208. Regensteiner, J., Sippel, J, McFarling, ET, Wolfel, EE, Hiatt, WR, Effects of non-insulin-dependent diabetes on oxygen consumption during treadmill exercise. Medicine and Science in Sports and Mexercise, 1995. 209. Kelley, D., Goodpaster, B, Wing RR, Simoneau, JA, Skeletal muscle fatty acid metabolism in association with insulin resistance, obesity and weight loss. American Journal of Physiology, Endocrinology and Metabolism, 1999. 277: p. E1130-E1141. 210. Turcotte, L, Richter, PE, Kiens, B., Increased plasma FFA uptake and oxidation during prolonged exercise in trained vs. untrained humans. American Journal of Physiology, 1992. 262: p. E159-167. 211. Brandenburg, S., Reusch, JE, Bauer, TA, Jeffers, BW, Hiatt, WR, Regensteiner, JG, Effects of exercise training on oxygen uptake kinetic responses in women with Type 2 diabetes. Diabetes Care, 1999. 22: p. 1640-1646. 212. Simoneau, J.-A., Kelley DE, Altered glycolytic and oxidative capacitites of skeletal muscle contribute to insulin resistance in NIDDM. Journal of Applied Physiology, 1997. 83: p. 166-171. 213. Phillips, S., Green, HJ, Tarnopolsky, MA, Heigenhauser, GF, Hill RE, Grant, SM, Effects of training duration on substrate turnover and oxidation during exercise. Journal of Applied Physiology, 1996. 81: p. 2182-2191. 214. Abate, N., Insulin resistance and obesity; the role of fat distribution pattern. Diabetes Care, 1996.19: p. 292-294. 134 215. Boule, N., Haddad, El, Kenny GP, Wells, GA, Sigal RJ, Effects of exercise on glycemic control and body mass in type 2 diabetes mellitus: a meta-analysis of controlled clinical trials. Journal of the American Medical Association, 2001. 286: p. 1218-1227. 216. Kahn, S., Prigeion, RL, Mcculloch, DK, Boyko, J, Bergman, RN, Schwartz, MW, Neifing, JL, Ward, WK, Beard, JC, Palmer, JP, Porte, D, Quantification of the relationship between insulin sensitivity and beta-cell function in human subjects. Diabetes, 1993. 42: p. 1663-1672. 217. Durstine, J., Grandjean, PW, Cox, CA, Thompson, PD, Lipids, Lipoproteins, and Exercise. Journal of Cardiopulmonary Rehabilitation, 2002. 22: p. 385-398. 218. Leon, A., Sanchez, OA, Response of blood lipids to exercise training alone or combined with dietary intervention. Medcine and Science in Sports and Exercise, 2001. 33: p. S502-S515. 219. Javitt, J., Chiang, Y-P, Economic Impact of Diabetes, N.D.D. Group, Editor. 1995. p. chap 30. 220. Lower Direct Medical Costs Associated with Physical Activity.National Center for Chronic Diasese Prevention and Health Promotion.available at: http:/www/cdc.gov/nccdphp/dnpa/press/archive/lower_cost.htm 221. Hu, F., Sigal, RJ, Rich-Edwards, JW, Colditz, GA, Soomon, CG, Willett, WC, Speizer, FE, Manson, JE, Walkiing compared with vigorous physical activity and risk of type 2 diabetes in women: a prospective study. JAMA, 1999. 282: p. 1433-1439. 222. Kelley, D., McKolanis, TM, Hegazi, RAF, Kuller, LH, Kalhan, SC, Fatty liver in type 2 diabetes mellitus: relation to regional adiposity, fatty acids, and insulin resistance. American Journal of Physiology, Endocrinology and Metabolism, 2003. 285: p. E906-916. 223. Knobler, H., Schattner, A, Zhornicki, T, Malnick, SDH, Keter, D, Sokolovskaya, N., Lurie, Y, Bass, DD, Fatty liver- an additional and treatable feature of the insulin resistance syndrome. Q J Med, 1999. 92: p. 73-79. 135 blood taken is about 4 teaspoons (20 ml.). This sample will be analyzed for different fractions of blood cholesterol, and for insulin and glucose (sugar) levels. Exercise testing: You will be asked to walk on a treadmill, at gradually increasing speed and grade, to the maximal effort that you can give. This test is designed to be short, total time of approximately 12 minutes, with the last 3 or 4 minutes requiring vigorous effort. During the test you will be breathing through a mouthpiece that analyzes your breath. The result of the test, your "maximal oxygen uptake" provides information of your level of fitness or ability to do physical exercise - this is like measuring the horsepower of an engine. Computed tomography (CT) scans: Computed tomography involves a thin beam of x-rays focused on a specific part of the body. For this study, the C T scan will be done on one thin slice of the abdomen, and on one thin slice of the thigh muscle. The scans will be done at the St. Paul's Hospital Radiology department. The procedure is painless, as you do not feel the x-rays. After these measurements, participants will be assigned on a random basis (like pulling a number from a hat) to one of the three groups: 1. aerobic exercise classes, three times per week, 75 minutes per class, for four months 2. aerobic plus strength training classes, three times per week, 75 minutes per class, for four months 3. O R to be followed with the usual care of their physician. Y o u have an equal chance of being assigned to any of these groups. After four months of participation in the exercise classes O R the usual care of your physician, the blood sample, exercise test and computed tomography scans will be repeated. During your participation in the study you will be asked not to become pregnant, and to keep the study investigators informed of any change in medication that your doctor prescribes for you. Y o u will also be asked, once at the beginning and again at the end of the study, to keep a three day diary of your food intake. Risks Involved with the Study: As with any physical activity, there exists the possibility during exercise of adverse changes including abnormal blood pressure, fainting, disorders of heart rhythm and in very rare instances, heart attack, stroke, or death. There is also the chance of muscle or joint soreness, or strain, resulting from exercise, or times of low blood sugar. A l l efforts will be made to avoid these risks, Computed tomography involves exposure to radiation. Extended exposure to radiation carries risks, although the exposure involved in this study protocol is less than a dental or chest x-ray, and less than the level of radiation we are exposed to in 1 week of urban living, or that of a four hour plane flight. Potential Benefits Exercise is considered important in the management of diabetes, and its associated impact on health. Participating in this study will help you learn and establish an appropriate exercise program that can aid in better control of blood sugar, blood pressure, body fat and blood 138 cholesterol. It can also help improve overall mood and sense of well-being. The support of professional advice and the support of a group of people in similar conditions is also a positive benefit from participation in this study. There is no monetary payment for participating in this study, although parking costs may be re-imbursed. Confidentiality Your medical record will be kept secure, and access to it limited to the investigators, the study coordinator, and your physician(s). Data collected from the study will not be linked to any individual by name, but by code. Only the investigators and coordinator will have access to the document that links names with code numbers. If you have questions, or are unclear about any of the study procedures, risks, or benefits, you are free at any time to contact the study coordinator or investigators at the numbers listed at the top of this consent form. Participant Consent I understand that participation in this study is entirely voluntary and that I may refuse to participate or I may withdraw from the study at any time without any consequences to my continuing medical care. I have received a copy of this consent form for my own records. I consent to participate in this study. Participant's Signature Date Witness's Signature Date 139 Appendix C: Initial Subject Characteristics Subject age (yrs) yrs x DM1 (cm) BMI (kg/m2) wt(kg) V0 2 (L/min) pre change pre change control group 1 52 3 - 48.3 133 6.6 2.20 -2 44 2 - 37.2 105 6.5 2.00 0.00 3 58 10 - 29.6 71 -3.1 1.62 -0.22 4 64 4 - 34.6 81 -0.5 1.90 -0.15 5 65 7 - 39.3 110.9 3.1 1.40 0.20 6 74 10 112.0 30.2 77.3 3.5 1.20 -0.20 7 61 1.5 121.0 35.0 93 -1.9 1.30 0.10 8 56 1 124.0 42.1 104.5 3.2 2.10 -9 56 3.5 34.3 84.6 0.3 2.00 -avg: 58.9 4.7 119.0 36.7 95.6 2.0 1.75 -0.05 std.dev. 8.5 3.5 6.2 5.9 19.6 3.5 0.37 0.17 sem 2.8 1.2 2.1 2.0 6.5 1.2 0.12 0.07 Ae + RT group 1 56 1.5 115.0 34.0 97 -9.2 1.75 -0.08 2 71 3 105.0 35.4 97.5 1.7 1.30 0.00 3 73 4 96.0 27.3 66.4 -2 1.34 0.37 4 60 3 112.0 29.0 80.5 -0.5 2.05 -0.09 5 65 4.5 127.0 41.0 109.7 2.4 1.46 0.21 6 56 1 - 28.2 81.5 -9.1 1.90 0.40 7 71 1 113.0 40.1 92 -1 1.40 0.20 8 65 2 113.0 30.5 89.8 -1.5 1.40 0.36 9 63 7 104.5 32.0 82.5 -6.5 1.60 -10 54 10 114.5 35.7 98.5 -2.9 2.20 0.15 avg: 63.4 3.7 111.1 33.3 89.5 -2.9 1.64 0.17 std.dev. 6.9 2.9 8.6 4.8 12.2 4.1 0.32 0.19 sem 2.2 0.9 2.7 1.5 3.9 1.3 0.10 0.06 Ae only group 1 61 1 102.0 29.3 75 -0.5 1.72 0.11 2 55 5 118.0 36.6 103.3 2.2 1.78 0.90 3 55 3 118.0 36.8 94.2 -0.2 1.84 0.06 4 68 4 103.0 28.2 69.6 0 1.20 0.22 5 59 4 100.0 29.1 72.7 -1.2 1.27 0.01 6 58 6.5 102.0 39.6 88 -1.5 1.50 0.50 7 67 2.5 106.0 30.1 78 -2.3 1.85 0.09 8 50 1 102.0 31.9 75.6 -5.1 2.26 -0.18 9 62 1.5 103.5 30.4 74 -2.5 1.67 0.27 avg: 59.4 3.2 106.1 32.5 81.2 -1.2 1.68 0.22 std.dev. 5.8 1.9 7.0 4.1 11.4 2.0 0.32 0.32 sem 1.9 0.6 2.3 1.4 3.8 0.7 0.11 0.11 combined exercise groups avg. std.dev. sem ' years since diagnosis of diabetes 2 waist circumference dash indicates data not availalble 140 Appendix C: individual Subject Attendance to Exercise Classes Exercise Class Attendance 100 n 90 -Q . 3 O 80-o> c 70 -15 •*-* 60 -o **— o 50 . cp 40 -0) O C ra 30 -x» e 0) 20 -*s ra 10 -0 -Q O ooo •oo*o**# O Q O * o»»»oo •o*o • ooo» • **dd o ooo•doodd* od oo 6 o o •ooodo* • • o • • • • • Ae + RT group o Ae only group 5 10 15 20 25 30 35 40 45 50 class #1-48 141 Ui T3 (0 O _1 Ui c "E "<5 H 0) o c (0 (0 (7) d) 01 D) c re O 6 c 0> a a < &a> i^cu ™ co a. fc 75 aj a> o. a) t£ .E o cu cu m ,- |v- CO CO co co CN CN 5) •sT CN CO 00 cu co =; tr -== a> > §> q C vO Q CO o C ^ — o O 7 3 T 3 C II ro <"> ,® in ~ co c 52 S E co i H P ro .E o *. * — CO O C cn 3 a> g. g ^ to to cu c Q . — CO ^ CO CO cu -a O ) o cu C L o cu !£? CO o C L o 00 cu C v P ro ^ sz o CO Q cu ro °^  o co •sir o o •sr CD O cu I— CL CO CO w so co to C ) W CO o o o o o O o CO GO 00 o in o in d CO CO •sr CO o o o CO CO T— o no o OO a to CO to cu o d •sT o in" •sr o o o in o in CO CO CO cn o o o CO o o m T— d CN CO CN CO o CL C L S> q c v ? o ro ^ m O CO cu CL CU ro °^ iv. ~ o o § CD O od co CO CO IO io o o o r-" (v." m" o" d co co •sr -sr •<* m o o o o CO O d in CN o o o CO co o d co co co in o o O o o o o o i in th d in i n 1 1 d d CD CO CN co CO CO SO CM O o o o o in" d d d in" CO in m m co o d in o d co o o d m" m T -tv. co CN co 00 o co CO CO oo ro N. 00 CO ob •sr d r--oci 00 00 m ih •sr CO 00 in CO in d in CN co —^ "ST CO o •sr o in o o rv. o o CO CO CO co ih CN d CN co d d in ih co d CO d co CO 00 io o o o o o o m in o o d oo d in d CO d CD ih in ih in CO •sr d •sr ih CO o o o o o o o o o o d CO d •sr ih CO d in d •sr d in ih CN ih CO ih CN d in o o CO CO CO CN in in o o CN CN ih d T— T— o o T— T— T— o 00 o ih •sr o o in o o ih d CN d ch •sr CO m co CN CO co CO o ih o in •sr o d o o o o o d ih ih CN o co o o o o co in o ro o o CN CN CO a i od •sr m ro •sr io m CO in o r-- CO o T -cb co in CN • d oo co co CNI d CO in o d CN o o ih co in o in in o o m •sr o m d r-' •sr •«-CN CN o -r-CN T -m o m CN t^ . CN CO 00 O CN in m o c N c o - s r i n c o r v - o o r o co co •sr •sr CN C O > T 3 < CO I O o >^ 3 1 T3 CO "P i t a> o SB C a-B T3 co CU CD n co CO T3 CC 'r-CO CO C CO 5 -° * ± CO CO CO a co -2 CO aj S 32 I 8 CO co a> IU cu CO CO ^ O co o § 0 5 co cz 8 E o o od E cu CO CO 1 a> T3 T3 _ CU iS to cu cu co £ ^ 5 u cu if E to o s= cu 3 CO •o E •a cu "55 "> 3 CO S" <" 2> co "a o cu CU co S" = 3 O CO CU !i= -a o cu a. i— o cu 4—1 CO c _ cu -^ «» CO — c CO cu -o S CD tO 3 § O CO CO © •R io CO o 3 o _ CO CO t 1 4 2 Appendix D: Carbohydrate Metabolism Measures HgA1C G,1 (mg/kg/min) Subject pre post change pre post change control 1 - 0.087 - 0.53 1.1 0.57 2 0.068 0.062 -0.006 3.78 2.78 -1.00 3 0.067 - - 2.48 3.27 0.79 4 0.064 0.065 0.001 1.68 1.08 -0.60 5 0.089 0.095 0.006 0.02 0.02 0.00 6 0.066 0.062 -0.004 4.13 3.21 -0.92 7 0.069 0.067 -0.002 2.23 2.07 -0.16 8 0.058 0.061 0.003 2.51 3.99 1.48 9 - - - 3.21 3.71 0.50 avg. 0.069 0.071 -0.0003 2.29 2.36 0.07 std.dev. 0.010 0.016 0.005 1.38 1.37 0.83 sem 0.004 0.006 0.002 0.459 0.456 0.28 Ae + RT 1 0.062 0.059 -0.003 3.46 6.67 3.21 2 0.057 0.057 0 0.69 3.65 2.96 3 0.088 0.075 -0.013 3.17 8.32 5.15 4 0.071 0.072 0.001 2.44 2.08 -0.36 5 0.06 0.061 0.001 1.98 3.80 1.82 6 0.06 0.058 -0.002 3.38 5.30 1.92 7 0.057 - - 3.37 3.18 -0.19 8 0.062 0.06 -0.002 2.69 4.21 1.52 9 0.073 0.077 0.004 1.11 2.07 0.96 10 0.095 0.098 0.003 1.26 2.43 1.17 avg. 0.069 0.069 -0.001 2.36 4.17 182 std.dev. 0.013 0.014 0.006 1.04 2.05 1.65 sem 0.004 0.004 0.002 0.329 0.649 0.52 Ae only 1 0.059 0.057 -0.002 2.13 4.45 2.32 2 0.07 0.074 0.004 3.89 3.49 -0.40 3 0.076 0.079 0.003 1.12 1.94 0.82 4 0.06 0.059 -0.001 4.38 5.99 1.61 5 - 0.081 - 1.59 0.19 -1.40 6 0.059 0.057 -0.002 2.12 2.65 0.53 7 0.066 0.063 -0.003 1.83 2.10 0.27 8 0.055 0.052 -0.003 2.47 3.19 0.72 9 0.059 0.057 -0.002 5.45 5.96 0.51 avg. 0.063 0.064 -0.001 2.78 3.33 0.55 std.dev. 0.01 0.01 0.003 1.46 1.91 1.07 sem 0.002 0.004 0.001 0.485 0.635 0.36 dash indicates data not available 1 glucose infusion rate 143 Appendix D: Carbohydrate Metabolism Measures fasting insulin (pmol/L) fasting glucose (mg/dl) Subject pre post change pre post change control 1 210.0 199.6 -10.4 11.6 10.3 -1.3 2 192.3 198.5 6.2 7.7 7.9 0.2 3 239.5 203.1 -36.4 7.3 6.6 -0.7 4 659.9 503.7 -156.2 8.1 8.5 0.4 5 265.1 249.2 -15.9 16.5 15.4 -1.1 6 137.2 150.9 13.7 6 5 -1 7 195.9 191.6 -4.3 6.9 7.1 0.2 8 196.5 218.1 21.6 5.8 6.3 0.5 9 139.5 138.7 -0.8 6.3 5.7 -0.6 avg. 248.4 228.2 -20.3 8.5 8.1 -0.4 std.dev. 159.7 108.5 53.7 3.5 3.2 0.7 sem 53.2 36.2 17.9 1.2 1.1 0.2 Ae + RT 1 150.3 122.9 -27 A 6.4 6.5 0.1 2 235.6 173.1 -62.5 7.9 6.5 -1.4 3 170.2 113.6 -56.6 11.3 8.1 -3.2 4 169.1 162.7 -6.4 9.1 8.8 -0.3 5 152.8 188.9 36.1 4.9 5.4 0.5 6 157.6 147.9 -9.7 5 5.1 0.1 7 184.8 169.9 -14.9 4.8 5.2 0.4 8 189.6 171.3 -18.3 6.3 6.1 -0.2 9 217 221.7 4.7 8.7 9.9 1.2 10 139.4 128.8 -10.6 14.2 12.4 -1.8 avg. 176.6 160.1 -16.6 7.9 7.4 -0.5 std.dev. 30.6 32.8 28.4 3.1 2.4 1.3 sem 9.7 10.4 9.0 1.0 0.8 0.4 Ae only 1 159.5 169.2 9.7 6.6 7 0.4 2 195.6 205.1 9.5 8.7 9.3 0.6 3 208.8 171 -37.8 8.2 7.2 -1 4 181.8 153.3 -28.5 6 5.4 -0.6 5 216.2 242.4 26.2 9.1 10.9 1.8 6 227.7 212.3 -15.4 5.7 5.7 0 7 299 276.9 -22.1 8 7.9 -0.1 8 268.1 201.2 -66.9 6.4 5.8 -0.6 9 197.2 177.9 -19.3 6.3 6.7 0.4 avg. 217.1 2010 -16.1 7.2 7.3 0.1 std.dev. 43.2 39.3 28.2 1.3 1.8 0.8 sem 14.4 13.1 9.4 0.4 0.6 0.3 1' glucose infusion rate, dash indicates data not available 144 Appendix E: Sample of One Analyzed CT Scan Analyzed Abdominal Scan Abdominal image at L4-5 level. Darker gray areas are AT, intermediate lighter areas are muscle and organ tissue, and bright white areas represent bone i.e. vertebra and top of iliac (pelvic) bone. Abdominal image from above subject, pixels are coloured (or "tagged") according to their density. Red represents pixels within density range of -190 to -30 HU, which corresponds to AT. The red pixels in this image represent 573 cm2. The same image after removing all coloured pixels which do not fall within the inner boundary of the abdominal wall. In this image the intra-abdominal AT area is 173 cm2. 145 -Thiqh Muscle Mid-thigh image displaying darker gray areas as AT, intermediate lighter area is skeletal muscle, and the bright white area is bone. Image analyses where pixels of the density range corresponding to AT (-190 to -30 HU) are coloured red. Pixels within the 0 to 100 HU range correspond to muscle and are coloured blue. Tissue of greater density, e.g. bone, is coloured a purple. In this image AT area is 204.8 cm2, and skeletal muscle is 247.9 cm2. Image analyses subdivides the muscle into a low density range (0 to 33 HU) coloured green, and a normal density range (34 to 100 HU) which is coloured blue. Low density muscle comprises 71.1 cm2, and normal density muscle makes up176.9 cm' muscle area. of the total 146 Appendix F: Abdominal AT Cross-sectional Areas total abdominal AT (cm2) visceral AT (cm2) subcutaneous AT (cm2) subject pre post change pre post change pre post change control 1 1063.1 1123.9 60.8 248.0 278.0 30.0 815.1 845.9 30.7 2 920.3 1011.9 91.6 260.7 311.2 50.5 659.6 700.8 41.1 3 544.8 481.4 -63.4 189.6 148.7 -40.9 355.2 332.8 -22.5 4 663.1 679.4 16.2 232.2 252.1 19.9 430.9 427.3 -3.6 5 1073.8 1063.9 -9.9 484.0 461.2 -22.8 589.8 602.7 12.9 6 525.5 588.7 63.2 111.2 126.0 14.8 414.3 462.7 48.4 7 748.2 759.9 11.7 276.5 293.8 17.3 471.7 466.0 -5.6 8 909.3 951.5 42.2 206.4 194.8 -11.5 702.9 756.7 53.8 9 827.2 768.1 -59.0 323.3 262.8 -60.5 503.9 505.3 1.4 avg 808.4 825.4 17.0 259.1 258.7 -0.4 549.3 566.7 17.4 std.dev. 203.9 223.6 53.9 103.3 99.5 36.0 152.5 170.4 27.1 sem 68.0 74.5 18.0 34.4 33.2 12.0 50.8 56.8 9.0 Ae+RT group 1 839.6 670.8 -168.9 281.1 220.1 -61.0 558.6 450.7 -107.9 2 886.0 816.0 -70.0 294.1 285.3 -8.8 591.9 530.8 -61.2 3 542.1 488.7 -53.4 135.1 138.7 3.6 407.0 350.0 -57.0 4 628.1 650.8 22.7 164.6 164.7 0.1 463.5 486.2 22.7 5 958.9 952.6 -6.2 375.3 337.1 -38.2 583.5 615.5 32.0 6 622.7 509.9 -112.8 231.5 173.9 -57.6 391.3 336.1 -55.2 7 772.5 746.8 -25.7 285.1 248.4 -36.7 487.4 498.4 10.9 8 655.3 635.0 -20.3 287.5 249.2 -38.4 367.8 385.8 18.0 9 652.6 598.0 -54.6 274.9 265.3 -9.6 377.7 332.7 -45.0 10 642.3 648.8 6.5 181.9 166.0 -15.9 460.4 482.8 22.4 avg 720.0 671.7 -48.3 251.1 224.8 -26.3 468.9 446.9 -22.0 std dev 135.8 138.6 58.1 72.5 63.4 23.2 85.1 93.9 48.7 sem 43.0 43.8 18.4 22.9 20.0 7.4 26.9 29.7 15.4 Ae only group 1 573.1 527.5 -45.6 172.6 169.0 -3.6 400.4 358.5 -42.0 2 705.0 740.7 35.7 141.6 160.0 18.4 563.4 580.7 17.2 3 822.4 768.5 -53.9 318.6 314.0 -4.5 503.9 454.5 -49.4 4 626.0 613.5 -12.5 297.0 300.6 3.6 329.0 313.0 -16.1 5 431.3 446.8 15.5 207.0 194.3 -12.8 224.3 252.6 28.3 6 760.1 747.5 -12.6 168.8 133.1 -35.7 591.3 614.4 23.1 7 532.8 490.0 -42.7 268.1 239.0 -29.2 264.6 251.1 -13.6 8 507.7 510.3 2.6 174.3 167.2 -7.1 333.4 343.1 9.7 9 593.2 553.5 -39.6 192.9 184.8 -8.2 400.2 368.8 -31.5 avg 616.8 599.8 -17.0 215.7 206.9 -8.8 401.2 392.9 -8.2 std dev 125.8 123.1 30.9 63.0 63.8 16.2 128.8 131.6 29.1 sem 41.9 41.0 10.3 21.0 21.3 5.4 42.9 43.9 9.7 147 Appendix G: Thigh Muscle Characteristics Ms CSA1 low density ms normal density ms (cm2) (cm2) (cm2) Subject pre post change pre post change pre post change control 1 258.6 259.4 0.8 63.1 65.7 2.6 195.0 187.1 -7.9 2 237.9 233.6 -4.3 48.1 50.3 2.2 190.3 182.7 -7.6 3 199.3 203.6 4.2 24.8 31.6 6.8 174.5 172.0 -2.6 4 239.2 238.2 -1.0 49.7 48.2 -1.5 188.6 189.1 0.5 5 254.8 250.5 -4.3 60.1 66.5 6.4 194.7 184.0 -10.6 6 193.0 190.9 -2.1 62.9 67.7 4.8 130.0 123.2 -6.9 7 232.9 230.9 -2.0 57.8 56.2 -1.6 175.9 174.6 -1.3 8 230.4 239.8 9.4 50.1 53.8 3.7 181.1 186.0 4.9 9 186.2 192.0 5.8 39.1 44.6 5.5 147.6 147.4 -0.2 avg. 225.8 226.5 0.7 50.6 53.8 3.2 175.3 171.8 -3.5 std.dev. 26.6 25.1 4.8 12.5 11.8 3.1 22.4 22.3 5.0 sem 8.9 8.4 1.6 4.2 3.9 1.0 7.5 7.4 1.7 Ae + RT group 1 233.7 232.0 -1.7 45.6 40.7 -4.9 188.1 191.3 3.2 2 171.7 187.7 16.0 55.6 59.2 3.6 116.2 128.5 12.4 3 202.5 215.1 12.6 36.5 37.9 1.4 166.2 177.2 11.1 4 232.9 243.3 10.4 26.4 28.7 2.2 207.0 214.2 7.2 5 221.0 227.6 6.6 53.6 57.8 4.2 167.4 175.9 8.5 6 252.9 251.8 -1.1 32.3 27.8 -4.5 220.6 224.0 3.4 7 205.6 210.6 5.0 53.2 49.0 -4.2 153.3 161.6 8.3 8 171.3 179.9 8.6 53.4 48.4 -5.0 117.9 131.5 13.6 9 158.9 156.0 -2.9 52.7 38.1 -14.6 106.1 118.0 11.9 10 226.9 232.0 5.1 103.4 84.2 -19.2 123.5 148.6 25.1 avg 207.7 213.6 5.9 51.3 47.2 -4.1 156.6 167.1 10.5 sd 31.5 30.5 6.3 21.0 16.9 7.7 40.3 36.1 6.3 sem 10.0 9.6 2.0 6.7 5.3 2.4 12.8 11.4 2.0 Ae only group 1 201.8 211.0 9.2 .32.8 33.4 0.6 169.0 177.6 8.6 2 297.0 295.2 -1.8 79.8 83.7 3.9 217.2 212.5 -4.7 3 247.9 245.2 -2.7 71.1 68.9 -2.2 176.9 177.3 0.5 4 181.9 178.1 -3.7 58.1 49.3 -8.8 123.8 128.0 4.2 5 195.6 199.6 4.1 31.5 34.2 2.7 168.8 165.5 -3.4 6 233.7 224.1 -9.5 34.0 28.5 -5.5 199.3 195.7 -3.6 7 213.0 221.2 8.2 38.2 38.9 0.7 175.0 181.7 6.6 8 243.3 242.0 -1.3 43.7 42.2 -1.5 195.5 199.8 4.4 9 202.3 208.0 5.7 55.0 52.5 -2.5 147.3 155.5 8.2 avg 224.1 224.9 0.9 49.3 47.9 -1.4 174.8 177.1 2.3 sd 35.4 33.5 6.2 17.6 18.2 4.0 28.0 25.4 5.3 sem 11.8 11.2 2.1 5.88 6.05 1.3 11.1 10.1 1.8 148 Appendix G: Thigh Muscle Characteristics Thigh Subcutaneous AT Average density (cm2) (HU) Subject pre post change pre post change control 1 471.9 463.7 -8.3 49.3 47.5 -1.8 2 436.3 442.2 5.9 49.0 48.1 -0.9 3 136.7 131.1 -5.6 52.0 49.6 -2.5 4 233.0 224.6 -8.4 46.6 47.0 0.5 5 310.1 317.1 7.1 47.1 45.4 -1.6 6 270.4 287.6 17.2 40.2 39.5 -0.6 7 237.0 213.9 -23.2 45.3 44.7 -0.6 8 499.0 519.8 20.8 47.6 46.4 -1.2 9 273.3 266.5 -6.8 47.2 45.4 -1.8 avg. 318.6 318.5 -0.1 47.1 46.0 -1.2 std.dev. 123.2 130.1 14.0 3.3 2.8 0.9 sem 41.1 43.4 4.7 1.1 0.9 0.3 Ae + RT 1 288.5 249.8 -38.8 46.9 49.0 2.1 2 404.6 434.8 30.2 40.9 41.8 0.9 3 203.0 174.9 -28.1 43.0 47.3 4.3 4 188.0 190.1 2.1 52.7 51.3 -1.4 5 422.8 424.9 2.1 45.8 45.4 -0.4 6 149.7 122.1 -27.5 50.4 52.7 2.3 7 320.5 317.1 -3.4 45.4 46.4 1.0 8 240.6 228.0 -12.6 40.2 43.3 3.0 9 304.9 224.7 -80.2 39.2 45.9 6.7 10 289.0 264.6 -24.5 34.8 40.4 5.6 avg 2812 263.1 -18.1 43.9 46.4 2.4 sd 89.0 102.4 29.6 5.4 3.9 2.6 sem 28.1 32.4 9.4 1.7 1.2 0.8 Ae only 1 206.7 188.2 -18.5 46.7 47.4 0.7 2 322.4 343.2 20.8 44.6 44.3 -0.3 3 204.8 195.2 -9.5 41.5 43.9 2.4 4 211.8 195.7 -16.1 40.9 43.3 2.5 5 124.7 121.3 -3.5 49.1 45.3 -3.8 6 401.5 378.1 -23.4 51.4 52.5 1.1 7 139.9 120.2 -19.7 48.1 47.5 -0.6 8 190.1 190.6 0.5 51.3 49.5 -1.8 9 190.1 178.5 -11.6 41.7 45.1 3.5 avg 221.3 212.3 -9.0 46.1 46.5 0.4 sd 70.5 81.4 13.6 4.2 3.0 2.3 sem 23.5 27.1 4.5 1.4 1.0 0.8 149 Appendix H: Blood Lipid Results TC (mmol/L) LDL (mmol/L) HDL (mmol/L) LD pre post change pre post change pre post change control na 6.32 na na 3.36 na na 1 na 1 4.11 3.64 -0.47 2.27 2.15 -0.12 1.21 1 -0.17 2 4.48 na na 1.89 na na 1.16 na na 3 4.77 4.76 -0.01 2.45 2.27 -0.18 0.75 0.8 0.03 4 5.23 5.39 0.16 3.10 3.26 0.16 0.97 0.9 -0.06 5 5.31 4.44 -0.87 3.31 2.51 -0.8 1.1 1 -0.07 6 6.29 5.76 -0.53 3.98 3.66 -0.32 1.05 1 -0.01 7 6.09 6.69 0.60 4.32 4.73 0.41 1.14 1.1 -0.08 8 - - - - - - - - -avg. 5.18 5.29 -0.19 3.05 3.13 -0.14 1.05 0.98 -0.06 sd 0.80 1.08 0.54 0.90 0.91 0.42 0.16 0.10 0.07 sem 0.30 0.41 0.22 0.34 0.34 0.17 0.06 0.04 0.03 Ae + RT group 1 5.06 4.71 -0.35 3.09 3.03 -0.06 1.09 1.2 0.07 2 5.23 5.40 0.17 2.74 3.04 0.3 1.72 1.7 -0.05 3 5.28 5.21 -0.07 3.39 3.21 -0.18 1.58 1.6 0.04 4 4.49 4.10 -0.39 2.14 1.91 -0.23 1.57 1.6 -0.01 5 5.56 6.12 0.56 2.92 3.45 0.53 1.43 1.4 -0.05 6 6.29 6.33 0.04 3.84 4.09 0.25 1.63 1.6 0.01 7 4.58 6.67 2.09 2.81 4.52 1.71 1.08 1.2 0.08 8 5.82 5.52 -0.30 3.66 3.51 -0.15 1.6 1.5 -0.15 9 4.20 4.32 0.12 - 2.89 - 0.81 0.7 -0.1 10 4.21 4.44 0.23 2.45 2.46 0.01 1.15 1.2 0.09 avg. 5.07 5.28 0.21 3.00 3.21 0.24 1.37 136 -0.01 sd 0.70 0.89 0.72 0.55 0.75 0.61 0.31 0.30 0.08 sem 0.22 0.28 0.23 0.18 0.24 0.20 0.10 0.09 0.03 AE group 1 4.46 5.09 0.63 3.08 3.59 0.51 0.91 1.1 0.15 2 3.83 4.32 0.49 2.16 2.33 0.17 1.16 1.1 -0.1 3 6.22 6.11 -0.11 3.57 3.37 -0.2 1.51 1.6 0.04 4 5.33 5.28 -0.05 2.77 2.72 -0.05 1.56 1.8 0.19 5 - 5.27 - - 3.66 - - 1 -6 4.29 4.72 0.43 1.88 1.43 -0.45 1.43 1.2 -0.26 7 4.25 3.71 -0.54 2.61 2.03 -0.58 0.93 0.9 0 8 5.36 6.00 0.64 3.48 4.07 0.59 1 0.9 -0.1 9 2.69 2.60 -0.09 0.67 0.78 0.11 1.69 1.5 -0.18 avg. 4.55 4.79 0.18 2.53 2.66 0.01 1.27 1.21 -0.03 sd 1.08 1.12 0.43 0.95 1.11 0.42 0.31 0.31 0.16 sem 0.38 0.37 0.15 0.34 0.37 0.15 0.11 0.10 0.06 dash indicates data not available 150 Appendix H: Blood Lipid Results TG (mmol/L) apo B (g/L) pre post change pre post change control 1 na 4.2 na na 1.5 na 2 1.37 1.1 -0.32 0.75 0.68 -0.07 3 3.14 na na 1.06 na na 4 3.41 3.7 0.31 1.09 1.16 0.07 5 2.52 2.7 0.14 1.31 1.27 -0.04 6 1.96 2 0 0.98 0.85 -0.13 7 2.73 2.3 -0.43 1.3 1.21 -0.09 8 9 1.38 2 0.57 1.24 1.45 0.21 avg. 2.36 2.55 0.05 1.10 1.16 -0.01 sd 0.81 1.09 0.38 0.20 0.30 0.13 sem 0.31 0.41 0.15 0.08 0.11 0.05 Ae + RT 1 1.91 1.1 -0.77 1.05 1.00 -0.05 2 1.68 1.5 -0.17 0.94 0.98 0.04 3 0.67 0.8 0.16 0.94 0.95 0.01 4 1.67 1.4 -0.29 0.69 0.65 -0.04 5 2.64 2.8 0.16 1.05 1.31 0.26 6 1.78 1.3 -0.47 1.2 1.21 0.01 7 1.51 2.2 0.64 0.91 - -8 1.22 1.2 -0.01 1.09 1.15 0.06 9 _ 1.6 - 0.98 1.08 0.1 10 1.33 1.6 0.28 0.84 0.87 0.03 avg. 1.60 1.55 -0.05 0.97 1.02 0.05 sd 0.54 0.56 0.42 0.14 0.20 0.09 sem 0.18 0.18 0.14 0.04 0.07 0.03 Ae only 1 1.03 1 -0.08 0.92 1.09 0.17 2 1.11 2 0.91 0.65 0.89 0.24 3 2.48 2.6 0.11 1.12 1.12 0 4 2.18 1.8 -0.41 1.06 1.09 0.03 5 - 1.4 - - - -6 2.13 2.5 0.37 0.82 0.95 0.13 7 1.54 1.6 0.08 0.97 0.84 -0.13 8 1.92 2.3 0.33 1.09 - -9 0.71 0.7 -0.03 0.23 0.23 0 avg. 1.64 1.75 0.16 0.86 0.89 0.06 sd 0.64 0.66 0.39 0.30 0.31 0.12 sem 0.23 0.22 0.14 0.11 0.11 0.05 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items