Exercise Performance and Effects of Exercise Training in Diabetes
There is a well established relationship between physical activity, metabolism, diabetes, and cardiovascular risk. In fact, numerous prospective epidemiological studies demonstrate an inverse correlation between physical activity and mortality, both cardiovascular and all cause mortality. This association is plausible when considered in the context of the impact of physical activity upon metabolic parameters that modulate cardiovascular risk such as blood pressure, dyslipidemia, inflammatory markers, and carbohydrate tolerance. Exercise is also pivotal for weight maintenance and prevention of obesity, a leading cause of new onset diabetes, which in turn contributes significantly to cardiovascular disease burden and mortality, as well as to noncardiac and all cause mortality. Prospective studies demonstrate the ability of diet and exercise to prevent progression from impaired glucose tolerance to diabetes. Despite the salutary effects of exercise on diabetes and cardiovascular risk, recent literature indicates that people with diabetes do not exercise as much as those without. This failure to exercise is likely behavioral and functional. Our recent work demonstrates that there are defects in both maximal and submaximal exercise function in persons with type 2 diabetes mellitus. In this chapter, we will review the cardiovascular and metabolic impacts of exercise, the relationship of exercise to diabetes prevention, and work from our lab examining the impact of diabetes on exercise capacity with some insights into the general mechanisms likely to be involved. The later chapters in this section will outline the impact of exercise on body composition and on cardiac, skeletal muscle, and endothelial function in additional detail .
Poor physical fitness is associated with increased morbidity and mortality. It has been observed consistently that low cardiorespiratory fitness and physical inactivity predict mortality in normal weight and obese men, in older men and women, and in men with Type 2 Diabetes Mellitus (T2DM) (1, 2, 3, 4, 5, 6, 7, 8, 9). Sedentary behavior has been clearly implicated as a factor leading to the development of diabetes as well as the worsening of cardiovascular (CV) outcomes of diabetes. Physical inactivity has become so common that one group has coined the term “sedentary death syndrome” (10). The sedentary death syndrome model proposes that evolution favored genes that support the physical activity required for long-term health in an agrarian society and that sedentary behavior is maladaptive.
Exercise has long been recognized as a cornerstone for the treatment of patients with T2DM. Over 80 years ago, Allen et al. reported that a single bout of exercise lowered the blood glucose concentration of persons with diabetes and improved glucose tolerance temporarily (11). Since that observation, numerous studies have confirmed the beneficial effects of exercise for persons with T2DM (12, 13, 14, 15, 16, 17). Paradoxically, despite extensive data indicating the importance of physical activity and exercise, 60–80% of adults with T2DM do not exercise sufficiently, and adherence to exercise programs is low in these patients (18, 19). One possible reason for this is that exercise performance is impaired in individuals with diabetes, even in early, uncomplicated T2DM (20, 21, 22, 23). This impairment will be discussed in detail in a later section.
4.2 BENEFITS OF ROUTINE PHYSICAL ACTIVITY
4.2.1 CV Disease and All-Cause Mortality
Results have been mixed in studies examining the effect of exercise interventions on lipid levels, as reviewed in (27). In general, studies with longer interventions (greater than 6 months) of higher intensity have been most likely to show increases in HDL levels and reductions in cholesterol and triglyceride (TG) levels. For instance, in 111 sedentary, overweight men and women with mild to moderate dyslipidemia, Kraus et al. found significant reduction in LDL and TG levels and improvement in HDL level with their highest intensity intervention and increases in LDL particle size in all exercise groups after 6 months (28). However, a recent study with a 9-month running intervention in young healthy adults showed only an insignificant trend toward LDL lowering and no effect on HDL or TG levels despite a 24% increase in peak VO2 (27). A significant reduction in apoB was reported, suggesting again that exercise may induce antiatherogenic changes in LDL particle size. A recent meta-analysis of studies of 2–12 months of exercise in subjects with T2DM found a significant decrease in TG levels, but no significant change in HDL or LDL (29). Another study with a 31-week exercise intervention in subjects with T2DM did demonstrate a significant increase in HDL in addition to decreased TG, but stable LDL (30). Unfortunately, the effects of diet have not been distinguished from those of exercise in most available studies.
Overall, HDL response to exercise training is variable and appears to depend upon multiple factors including dose, gender, and genetic background. As summarized by Ring-Dimitriou, studies support the need for longer duration, higher intensity exercise for significant HDL-raising effects (27). They also suggest that gender differences may exist with greater benefit occurring in men than in women. This may reflect higher baseline HDL in female subjects. Recently it was reported that there may be genetic determinants of whether people will respond to exercise by increasing HDL cholesterol. A polymorphism in the PPAR delta receptor (more common in Causcasians) was associated with a significantly greater improvement in HDL with exercise training (31). It is reasonable to conclude that exercise training may have a positive effect on lipids, but it should not be employed in lieu of lipid lowering phamacotherapy when indicated. At present it does represent one of the very few interventions, and arguably the safest intervention, with potential for raising HDL.
In addition to the well-established risk factors, an elevated level of plasma fibrinogen has also been reported to be a CV risk factor. Acute exhaustive exercise stimulates both thrombosis and fibrinolysis with a net neutral effect on hemostasis in most populations (32, 33). In the general population, the chronic and immediate postexercise responses in the thrombotic and fibrinolytic systems have been shown to be variable and reflect differing adaptations with ageing and responses to exercise protocols. In a recent study, investigators examined hemostatic variables including factor VII activity (FVIIa), tissue factor pathway inhibitor factor Xa complex (TFPI/Xa), and plasminogen activator inhibitor-1 (PAI-1) antigen activity after a high fat meal before and after exercise training (34). They observed reduction in the potential for coagulation and improved fibrinolytic potential in trained subjects with the meal stimulation suggesting that under certain conditions (e.g., postprandially) exercise may have cardiovascularly beneficial effects on hemostasis.
In people with diabetes the results are similarly variable. Fibrinogen level is elevated in men and women with T2DM (35). Results to date are not clear as to whether exercise training decreases fibrinogen level in the person with T2DM. Schneider et al. found that although VO2max increased by 8% with 6 weeks of training, fibrinogen level did not change significantly in a group of sedentary persons with T2DM (36). Conversely, Hornsby et al., found that a 12.5% increase in VO2max after 12–14 weeks of training was associated with a significant decrease in fibrinogen level in sedentary persons with T2DM (37). In the large Finnish Diabetes Prevention Study a combined diet and exercise intervention decreased PAI-1 level consistent with improved fibrinolysis (38). Another recent study found improved fibrinolysis after 6 months of aerobic training in overweight to obese men and women (39). Interestingly, improvements were significantly greater in men than in women and correlated closely with abdominal fat. Thus, the effect of exercise training on fibrinolysis appears to be generally salutary and may be mediated by changes in body composition, but this relationship requires further investigation.
4.2.4 Blood Pressure
High blood pressure is a leading contributor to CV mortality, and there is a consistent inverse relationship between physical activity and blood pressure in cross-sectional studies. The first study to examine the impact of training upon blood pressure was conducted by Jennings with a very rigorous exercise program in sedentary men (40). Over the last few decades a dose–response effect of exercise on blood pressure has been observed in both men and women, including those with CV and metabolic comorbidities. A recent meta-analysis assessed 72 longitudinal intervention studies to determine the impact of exercise training on blood pressure (41). Studies included both hypertensive and normotensive subjects. Overall the analysis demonstrated a small (3 mmHg), but clinically and statistically significant, decline in both systolic and diastolic average blood pressure, with a greater reduction in hypertensive subjects. They concluded that endurance training decreases blood pressure through a reduction in systemic vascular resistance secondary to decreased sympathetic nervous system and renin–angiotensin system activity. In a recent study of 30 obese T2DM subjects, a 3-month exercise intervention improved both systolic and diastolic blood pressure (42). Overall, improvement of blood pressure with exercise training is the most consistently demonstrated benefit of physical activity on CV health.
4.2.5 Endothelial Function
Coronary and peripheral artery endothelial dysfunction (ED), most often measured as impaired vasodilator response to mechanical or pharmacologic stimuli, have been shown to correlate with CVD risk, cardiac events in known CVD, and poor prognosis in CVD (43, 44, 45, 46, 47, 48). Exercise training improves endothelial function in the context of metabolic disease and CVD. For instance, in a study of patients with congestive heart failure (CHF), 4 weeks of lower leg exercise training significantly improved upper extremity endothelium-dependent vasodilation, but not endothelium-independent responses (49). Another study demonstrated improved flow-mediated dilatation (FMD) with a 12-week treadmill training program in hypertensive men (50). Such responses to exercise training have also been demonstrated in insulin resistant and diabetic subjects. Kelly et al. showed that 8 weeks of stationary bike training significantly improved brachial artery FMD in a group of overweight children relative to a sedentary control group (51). Similarly Meyer et al. found improved ED after 6 months of endurance training in obese sedentary children (52). Two recent studies demonstrated improvement in endothelium dependent vascular reactivity after 8 weeks of exercise training in overweight and T2DM adult subjects (53, 54). A third showed improvement in biomarkers of ED after 6 months of training in older patients with T2DM (55). In contrast, in a study of T2DM subjects no improvement was seen in microvascular function as measured by maximum skin hyperemia after 6 months of aerobic exercise training (56). Other studies have failed to find benefits of exercise on endothelial function in healthy individuals without baseline ED, for instance in healthy relatives of T2DM individuals (57) and in healthy middle aged men (58). The weight of evidence suggests that exercise training does significantly improve impaired endothelial function, but has no significant impact in normal vessels.
4.2.6 Inflammation and Immunity
The relationship between exercise and immune function is reported to be a “J” shaped curve wherein increasing from sedentary to moderate activity improves immune function but exercise training in elite athletes may diminish immune function (especially in the first 24 h after a bout of exhaustive exercise) (59). Regular performance of about 2 h of moderate exercise per day is associated with a reduction of risk for common viral infections of 29% compared to sedentary subjects (60). In contrast, exhaustive exercise such as a marathon is associated with a 100–500% increase in the risk of viral infection (61). It is worth noting that it is the rare individual who will exercise rigorously greater than 2 h per day, so the potential deleterious effects of exhaustive exercise are not likely to be observed in the general population.
Inflammation, the other face of the immune spectrum, is one of the universal mechanisms contributing to the initiation and progression of atherosclerosis (62) and the development of T2DM (63, 64). In general short term moderate intensity exercise interventions have a modest positive impact on some subset of circulating cytokines such as IL-1, -6, and -18, CRP, and TNF-; presumed anti-inflammatory markers such as adiponectin (and IL6?, see later); and inflammation-related cell adhesion molecules such as VCAM, ICAM, and the selectins (55, 57, 65, 66, 67, 68), but exact methods and results have been mixed. For instance, Zoppini et al. found stable CRP and decreased ICAM and P-selectin following 6 months of aerobic exercise in older, sedentary, overweight diabetics (55). In contrast, Olson et al. found reduced CRP and increased adiponectin, but stable cell adhesion markers after 1 year of resistance training in overweight women (65). In addition, there are studies that do not demonstrate any exercise-induced change in circulating inflammatory markers (69). Thus, evidence regarding the effect of exercise on inflammation is mixed and apparently heavily dependent upon the baseline status of the population and on the nature, intensity, and regularity of the exercise intervention. It is likely that a muscle damaging level of exercise can cause inflammation while more modest or habitual exercise reduces systemic inflammation to some degree. The net result is therefore a balance of these two opposing forces. Further studies are clearly necessary for a better understanding of the effects of exercise on systemic inflammation.
A further complication to this question arises from the fact that one of the cytokines that is frequently measured in studies of inflammation in metabolic syndrome, obesity, and diabetes is IL-6. A large body of recent data suggests that IL-6 has pleiotropic effects that include significant metabolic and insulin sensitizing effects, as well as possible anti-inflammatory effects [reviewed in (70, 71)]. IL-6 is produced by skeletal muscle during sustained exercise and plasma levels of IL-6 are transiently dramatically elevated in response to exercise. Studies of IL-6-deficient mice have demonstrated that these animals have decreased exercise endurance, decreased O2 consumption during exercise, and impaired fatty acid oxidation in response to exercise (72). These effects appear to be mediated by a decrease in induction of AMP kinase activity and of fatty acid oxidation pathways in exercising muscle, by decreased lipolysis in adipocytes and glucose release from the liver, and by a decrease in sympathetic outflow during exercise (70). Overall the literature is consistent with a crucial role for IL-6 in exercise performance and in the generation of a high turnover metabolic state during exercise and other forms of physical stress. Interestingly, by 9 months of age the IL-6-deficient mice are obese and have several features of metabolic syndrome including impaired glucose tolerance. Clearly, a full understanding of IL-6’s complex role in exercise, metabolism, diabetes, and inflammation awaits further studies.
Obesity is a common problem for persons with T2DM. Exercise conditioning may serve as an adjunct therapy especially when linked to diet. However, in the absence of diet exercise does not consistently lead to weight loss although body composition may be improved (73). These changes in body composition include decreased visceral adiposity and thus may have significant beneficial effects on CV risk factors. However, exercise training without dietary change results in minimal absolute weight loss despite greater than 60–90-min a day of moderate activity (74). In contrast to the limited impact of isolated exercise for weight loss, exercise is very effective for prevention of weight gain, acceleration of weight loss in combination with diet, and most importantly maintenance of weight loss. In a community-based study, introduction of walking and healthy snacks prevented weight gain (75). Similarly, in the National Weight Control Registry, comparison of a group of subjects who have maintained a substantial weight loss for greater than 12 months with those who regained weight suggests that physical activity of greater than 2,000 calorie per week is a crucial element of long-term success (76). When exercise mediation of weight loss has been examined prospectively, similar results are reported. For example, an intervention with diet with or without exercise for 12 weeks resulted in a weight loss of 10 kg with diet alone and 14 kg with diet plus exercise. After 12 weeks the dietary intervention was discontinued but the exercise intervention continued. At 36 weeks the diet group had regained all but 4 kg whereas the exercise group maintained 12 kg of weight loss (77). It is critical to understand that exercise alone does not lead to weight loss and to convey this to patients so that they will have an appreciation of the role of exercise and not be discouraged by an apparent lack of weight loss results from their exercise regimen.
4.2.8 Glucose Regulation and Insulin Sensitivity
Glucose metabolism in response to exercise has been extensively studied as it poses an important clinical challenge. Exercise has two different impacts on carbohydrate metabolism, the bout effect and the training effect. The bout effect refers to the direct impact of an episode of exercise on glucose during the exercise and for an interval of 1–72 h after the exercise is complete. Exercise training is typically considered routine physical activity that increases functional exercise capacity for which the gold standard is maximal exercise capacity (VO2max). Exercise training usually also effects body composition, especially lean body mass. The benefits of exercise for glycemia likely result from a combination of the bout and training effects.
It is well established that even a single bout of exercise has a pronounced effect on the metabolism of the person with T2DM. In fact, much of the benefit of training may be due to the most recent bout of exercise (78, 79). In support of the concept that single bouts of exercise affect metabolic parameters, Devlin and others reported that a single bout of glycogen-depleting exercise in patients with T2DM significantly increased glucose disposal for up to 12–16 h postexercise due to an enhanced rate of nonoxidative glucose disposal (78). This increase occurs at the level of both liver and muscle tissue (80). Others have found that exercise conditioning for 1 week increases whole body insulin-mediated glucose disposal (81) and glucose tolerance (12) in patients with T2DM. It is not completely clear how much metabolic benefit is derived from a single bout of exercise versus the effect of cumulative bouts, but it is clear that the benefit of a bout of exercise is lost rather quickly so that repeated exercise, probably daily, is needed for long-term, bout effect benefits on glucose metabolism. In addition, a very brief period of exercise such as a single bout or even a week of exercise is clearly insufficient to cause increases in maximal oxygen consumption, changes in body composition, or improvement in other CV parameters, which are affected by longer periods of training and have clear independent mortality benefits, as well as potential independent effects on glucose metabolism.
The effects of exercise training or routine physical activity on insulin sensitivity are likely to be complex and multifactorial, and the relative roles of decreased visceral fat, CV fitness, and cumulative bout effects of exercise have yet to be defined [reviewed in (82)]. Recent studies clearly demonstrate that exercise training leading to increased fitness (generally defined as an increase in VO2max) also results in improved insulin sensitivity as measured by the gold standard hyperinsulinemic euglycemic clamp (83, 84). These studies also compared exercise regimens consisting of moderate versus high intensity activity but with equal exercise energy expenditure and found greater effects on insulin sensitivity with higher intensity physical activity despite similar effects on VO2max. These results suggest that fitness per se may not correlate directly with insulin sensitivity. Others have asked whether the benefits of long-term exercise training (as opposed to the bout effect) on insulin sensitivity can be completely accounted for by changes in visceral adiposity and have had mixed results [reviewed in (82)]
The clinical implications were recently assessed in a meta-analysis that concluded that at least 12 weeks of exercise training, either aerobic, resistance, or combination training, results in a reduction in hemoglobin A1c of 0.8%, an effect that is comparable to the improvement typically achieved by dietary or single agent drug therapies (85).
4.2.9 Prevention of Diabetes
The role of exercise in the prevention of diabetes is unequivocal but has been most often and best studied in the context of a combined diet and exercise intervention. Early epidemiological and sociological evidence demonstrated a strong inverse correlation between habitual physical activity and incidence of diabetes. This evidence included the change in incidence of diabetes with a move from a rural lifestyle, observed in American versus Mexican Pima Indians. This relationship has been observed across diverse populations including male college alumni, female college alumni, registered nurses, and British men [reviewed in (86)]. These observations were followed by a set of prospective studies, the Finnish Diabetes Prevention Study (87), Da Qing Study (88), and the Diabetes Prevention Program (89). In all of these studies a diet and exercise intervention prevented transition from impaired glucose tolerance to diabetes in 50–60% of individuals. Only the Da Qing Study included an exercise alone arm. The preventative effect of exercise in this arm was similar to that observed with diet alone and was independent of weight loss, though body composition was not addressed. The success of exercise in diabetes prevention is likely to result from one or more of the effects described earlier, specifically improved insulin sensitivity, decreased visceral adiposity, and/or modulation of inflammation and oxidative stress.
4.3 EFFECTS OF T2DM ON EXERCISE PERFORMANCE
Persons with T2DM are at higher risk than nondiabetics for coronary artery disease, stroke, and peripheral arterial disease due to accelerated atherosclerosis (90). Exercise conditioning is thus likely to be especially beneficial in these individuals through the modification of CV risk factors discussed earlier (79, 91). Furthermore, the observed beneficial effects of physical activity on insulin sensitivity and glucose metabolism make it clear that, in addition to reducing CV morbidity and mortality, exercise training, or even an increase in the level of habitual physical activity, has a key role in the management of diabetes. Yet the population studies described earlier indicate that people with T2DM are generally less active than nondiabetic people. While some aspects of this behavior may be accounted for by lifestyle choices that contribute to the initial development of diabetes, recent evidence suggests that pathophysiological factors may also contribute to this decrease in activity. This section will primarily address changes observed in subjects with T2DM in CV or cardiopulmonary exercise performance, defined by maximal oxygen consumption (VO2max) and by kinetics of oxygen consumption during submaximal exercise. These data suggest that the cause and effect relationship of the correlation between low physical activity and diabetes may be bidirectional.
4.3.2 Maximal Exercise Capacity
Maximal Exercise Capacity
36 ± 6
37 ± 6
42 ± 7
Fat free mass (kg)
42 ± 7
48 ± 5
47 ± 5
6.0 ± 0.6
5.3 ± 0.5
9.0 ± 0.4*
Maximal exercise response
25.1 ± 4.7
21.8 ± 2.9
17.7 ± 4.0*
26.0 ± 6.0
23.0 ± 1.8**
22.4 ± 5.5**
1.13 ± 0.08
1.12 ± 0.06
1.16 ± 0.13
4.3.3 Submaximal Exercise Tolerance and Oxygen Uptake Kinetics (VO2 Kinetics)
The exercise abnormality observed at maximal exercise in T2DM is also observed during less vigorous physical activity (i.e., submaximal exercise). During the early stages of an incremental exercise test, oxygen uptake (VO2) increases with each increase in work rate. In nondiabetic individuals, there is a predictable increase in VO2 to meet the metabolic demand for a given increase in workload (e.g., ~10.1 ml/min/W) (97). The VO2 to work load relationship thus describes an individual’s overall ability to adjust to the exercise stress, and reductions in the slope of this relationship have been shown to effectively indicate abnormalities of cardiac output and gas exchange in cardiopulmonary and vascular diseases (98).
Similar to persons who have overt CVD, the increase in VO2 per unit of increase in workload is reduced in people with T2DM compared with healthy controls (22). Potential mechanisms for this abnormal response include a decrease in oxygen delivery and decreased cardiac function, and/or an abnormality of muscle oxidative metabolism. To further evaluate this possibility, submaximal constant-load exercise has been employed. Unlike graded or incremental exercise, constant-load exercise is performed at a moderate workload below the individual’s lactate threshold, where a steady-state VO2 for a given work rate can be obtained.
Following the onset of exercise, VO2 rises exponentially to steady state, the time course of which represents the VO2 kinetic response. The VO2 kinetics are determined by the systemic integration of muscle VO2, CV adaptations of oxygen delivery, and pulmonary gas exchange. Three phases of the pulmonary VO2 response to the change from rest to moderate constant-load exercise have been proposed (99, 100). At the onset of exercise, pulmonary VO2 in the lungs increases abruptly for the first 15–20 s as cardiac output and pulmonary blood flow initially increase (cardiodynamic phase or phase 1). Following a circulatory transit delay (usually about 20–40 s), VO2 then increases exponentially (phase 2), reflecting the increase of muscle VO2 as tissue oxygen extraction and blood flow increases to meet the exercise demand (101, 102). This is the primary component of VO2 kinetics and is described by a time constant (tau) reflecting the time to reach ~63% of the increase in VO2. Phase 2 ends as muscle VO2 and pulmonary gas exchange reach a steady state. Phase 3 is the steady-state VO2 during moderate exercise.
In the healthy individual, VO2 kinetics may be limited by either a maldistribution of blood flow to the working tissues limiting O2 transfer or by the inertia of oxidative metabolism (101, 103). In disease states where oxygen delivery is compromised, as with CVDs, VO2 kinetics are limited by the body’s ability to deliver oxygen to working muscle, and therefore may directly reflect impaired oxygen delivery (104, 105). Since impaired cardiac output and/or local distribution of blood flow to exercising muscles are components of the O2 delivery process, VO2 kinetics may thus provide a measure the effectiveness of the CV system in delivering sufficient oxygen to satisfy the requirements of muscle during exercise (106). In this regard, the time constant of phase 2 VO2 kinetics is prolonged in patient groups with abnormal CV responses to exercise, and in general is sensitive to alterations in oxygen exchange at the lungs, cardiac output, oxygen diffusion, and rates of tissue oxygen consumption.
Submaximal Exercise Kinetics
°20 W Tau (s)
21.4 ± 8.9
18.4 ± 9.9
42.6 ± 23.8*
°30 W Tau (s)
28.8 ± 5.3
27.8 ± 8.9
36.8 ± 6.2*
°80 W Tau (s)
42.8 ± 7.5
41.2 ± 8.2
55.7 ± 20.6
Heart rate kinetics
°20 W Tau (s)
8.5 ± 4.6
10.6 ± 8.2
23.8 ± 16.2*
°30 W Tau (s)
23.9 ± 13.8
14.2 ± 8.0
40.7 ± 11.9*
°80 W Tau (s)
41.2 ± 14.8
43.3 ± 11.3
72.3 ± 21.5*
4.4 POTENTIAL MECHANISMS LEADING TO EXERCISE IMPAIRMENT
There are several potential pathogenic mechanisms that may contribute to the decreased capacity for exercise in T2DM. These include metabolic and nonmetabolic sequelae of diabetes in the vasculature and in cardiac and skeletal muscle. These are discussed in the following sections.
The relationship between markers of glucoregulation and exercise has been investigated to determine whether these factors are likely determinants of exercise performance. To date, associations have not been found between hemoglobin A1C or fasting serum glucose concentration and exercise performance (23, 94, 95, 108, 109). In other words, although a single bout of exercise improves glycemic control (albeit temporarily), changes in glycemic control, per se, do not appear to affect exercise performance.
4.4.2 Insulin Resistance
In contrast to hyperglycemia, various reports have suggested that insulin resistance (IR) is associated with reduced VO2max in T2DM (110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121). IR has also been reported to be inversely correlated with VO2max in several disease states in addition to diabetes, including heart failure and chronic renal failure (113, 114). That this decrease in exercise capacity is independent of other complications of diabetes or of the systemic illness associated with heart and renal failure is further supported by the recent finding of exercise defects in nondiabetic women with polycystic ovarian syndrome (PCOS) (115) and of exercise defects in the metabolic syndrome (116). The significant decline in VO2max in subjects with PCOS compared to age- and weight-matched controls correlated with all measures of IR, but not with other reported measures including blood pressure, cholesterol, and androgen levels. In addition, there is an association between IR, and low physical fitness level in normotensive men with a family history of hypertension (117).
The cause and effect relationship between IR and impaired exercise performance is not well understood and has been further addressed through the use of a pharmacological intervention to improve insulin sensitivity. In a study of 20 women with early, uncomplicated T2DM randomized to rosiglitazone or placebo, rosiglitazone treatment resulted in a significant improvement in VO2max of 7%. This improvement correlated with both increased insulin sensitivity and improved endothelial function (110).
While the positive effects of exercise on insulin sensitivity are clear, the earlier results support the hypothesis that IR in turn negatively effects exercise capacity. Other literature lends support to multiple possible mechanisms for such a relationship including IR at the level of the vasculature leading to ED (in both peripheral and cardiac circulation), IR at the level of the muscle (cardiac and skeletal) leading to a decline in mitochondrial content and/or function, and IR at the level of the heart and/or skeletal muscle leading to inefficient substrate utilization. Recent attention has been focused on changes in substrate utilization and metabolic inflexibility in IR. Simply stated, insulin promotes carbohydrate utilization. In the absence of sufficient insulin signaling in IR, metabolism relies more heavily on fatty acids, a less oxygen efficient fuel source. These mechanisms and their potential relationship to exercise capacity are discussed further in the following sections.
220.127.116.11 Endothelial Dysfunction
One possible mechanism for the exercise abnormalities observed in persons with T2DM invokes ED as a contributing factor. The exercise abnormalities observed could reflect a deficient endothelial dilator response to metabolic demand in heart as well as peripheral skeletal muscle. In this scenario, exercise capacity would be limited by peripheral and/or coronary blood flow. It is well established that peripheral endothelial function and vascular reactivity in response to pharmacological vasodilators and to cuff ischemia at rest (118, 119) as well as in response to exercise are abnormal in adults with T2DM compared to nondiabetic controls (120, 121). Furthermore, insulin’s physiologic ability to enhance endothelium-dependent vasodilation is markedly impaired in diabetic individuals compared with that of lean control subjects, and it has been proposed that IR at the level of the endothelial cell is invariably associated with ED (122). This is supported by the observation that obese subjects with and without T2DM have endothelium-dependent vasodilation that is reduced by 40–50% compared with lean control subjects (123). In addition, every insulin resistant state studied to date has been found to have associated ED (122). Thus, IR in T2DM results in ED and to impaired demand-mediated increases in muscle (and probably cardiac) blood flow, in addition to decreased glucose transport into muscle. Alternatively, the vascular dysfunction of T2DM and insulin resistance may be a direct result of insulin resistance at the level of the vascular smooth muscle cell causing altered substrate utilization with a greater reliance on less efficient fuels (fatty acids). Finally, ED may result from the systemic inflammation and oxidative stress associated with IR and obesity. Prompted in part by findings in other disease states, such as heart failure, where an association between exercise performance and endothelial function has been reported (124), the relationship between endothelial function and the exercise abnormalities of T2DM is being investigated further.
Support for the ability of ED alone to cause exercise defects comes from the studies of Jones et al. using N-nitro-l-arginine methyl ester (l-NAME) to reduce nitric oxide (NO) levels prior to performing exercise. They found a decrease in maximal oxygen uptake (VO2max), which correlated with the expected reduction in vasodilation and decreased perfusion of large muscle groups (125). However, in contrast to our studies with T2DM subjects, l-NAME induced an acceleration of the rate at which oxygen consumption increased with exercise (VO2 kinetics) (125, 126). This could be explained by recent studies in animals and man demonstrating a role for NO in regulation of myocardial substrate utilization. Inhibition of NO synthase in dogs with l-NAME results in a marked increase in glucose oxidation and a decrease in fatty acid metabolism (127). It has been proposed that NO interferes with oxidative metabolism by competing for O2 binding at cytochrome c oxidase in the mitochondrial electron transport chain (128, 129, 130). The result of such NO-mediated mitochondrial inhibition is to modulate muscle oxidative phosphorylation and muscle VO2 kinetics. Thus, inhibition of NO synthesis alone appears to decrease VO2max, but may speed VO2 kinetics via the removal of NO-mediated effects on mitochondrial oxidative metabolism. The fact that both parameters are affected negatively in diabetes implies that changes in exercise parameters in diabetes cannot be fully explained by changes in NO synthesis or, presumably, ED alone.
18.104.22.168 Myocardial Dysfunction
There are also likely to be cardiac factors contributing to the exercise abnormalities of T2DM. Evidence has accumulated for the existence of myocardial dysfunction that is unrelated to coronary artery disease in many individuals with diabetes, even early uncomplicated diabetes (e.g., 131, 132, 133, 134, 135, 136, 137, 138). This condition has been termed “diabetic cardiomyopathy” and generally refers to a finding of subclinically impaired left ventricular (LV) function at rest (131, 134, 136, 137, 139) and/or during exercise (133, 135) in the absence of major coronary disease or hypertension. The earlier studies have demonstrated a predominant component of diastolic dysfunction in diabetic cardiomyopathy. Clinically it has been shown that cardiac diastolic dysfunction correlates closely with impairments in CV exercise capacity in heart failure (124), in diabetes (135, 139), and in normal subjects (177). In our studies of exercise dysfunction in T2DM we have found a reduced cardiac output by right heart catheterization during exercise in persons with diabetes compared to nondiabetic, healthy, age- and weight-matched controls (140). In addition, we have observed that pulmonary capillary wedge pressure rises more steeply and to a greater level with exercise in T2DM than in controls consistent with significant diastolic dysfunction during exercise (140) and that this presumed diastolic dysfunction correlates with the observed decrease in exercise capacity. Thus while the prevalence, etiology, and clinical significance remain unclear, it is possible that diabetic cardiomyopathy plays a significant role in the exercise defects seen in T2DM.
Finally, we have also observed that the cardiac diastolic dysfunction, which correlates with the decrease in exercise capacity in uncomplicated T2DM, also correlates with reduced myocardial perfusion (Regensteiner and Reusch, unpublished results). Based on these studies, impaired coronary artery endothelial function may be the mechanism for exercise impairment in T2DM via adverse effects on cardiac function. However, other data in the literature suggest the alternative or additional mechanisms discussed later.
4.4.3 Cardiac Substrate Utilization in Insulin Resistance
Cardiac energy production via preferential use of fat over glucose could contribute to exercise defects in diabetes. This model is supported by recent studies of cardiac substrate utilization in diabetes. Studies examining cardiac fuel utilization in IR DM rodents demonstrated a fixed, excess reliance on inefficient fat oxidation in the diabetic myocardium indicating metabolic inflexibility relative to nondiabetic controls (141). Mazumder et al. characterized cardiac substrate utilization in mice and found that basal and palmitate-stimulated FFA utilization were 1.5–2-fold higher in IR ob/ob mice than in wild-type mice (142). This fuel preference occurred at the expense of cardiac glucose oxidation and was accompanied by increased myocardial oxygen consumption with less ATP produced per unit of O2 consumed, and impaired cardiac efficiency (142). Similar results have been obtained in other IR animal models (db/db and ZDF) and in human subjects [reviewed in (143)]. For example, Peterson et al. demonstrated increased myocardial oxygen consumption, decreased cardiac efficiency, and increased cardiac fatty acid utilization in obese women compared to controls (144). However, Knuuti et al. did not find changes in cardiac fatty acid utilization in a small study of men with impaired glucose tolerance (145). Human studies demonstrate that a few days of high fat diet enhance fat oxidation and decrease mitochondrial efficiency (Ravussin E, Baton Rouge, LA, Personal communication). This is similar to the skeletal muscle mitochondrial dysfunction and inefficient glucose oxidation observed in subjects with T2DM and their relatives (146, 147). Inefficient myocardial function usually leads to diastolic dysfunction, which is the defect our group has implicated in T2DM subjects with exercise intolerance.
Interestingly, increased fatty acid levels and utilization at the expense of glucose oxidation have also been demonstrated in ischemic myocardium in both animal models and humans [reviewed in (148)]. This fuel utilization preference has been shown to contribute to cellular acidosis and decreased cardiac efficiency in the ischemic heart and it is thought to play a role in ischemic and reperfusion injury. Pharmacological stimulation of glucose oxidation with dichloroacetate, an activator of the pyruvate dehydrogenase complex, rescues these defects in rat ischemic myocardium (149). Similarly, agents that inhibit fatty acid oxidation (obliging reliance on glucose) decrease infarct size and troponin release in a rat ischemia/reperfusion model (150), improve cardiac efficiency (148), and are currently under investigation as antianginal agents (151). Thus, the model of insulin resistance-induced myocardial substrate shifts may provide a mechanism not only for impaired exercise capacity but also for the worsened outcomes of acute coronary events in diabetes.
22.214.171.124 Skeletal Muscle Changesin Diabetes
The role of skeletal muscle in the impaired exercise responses of persons with T2DM has not been specifically elucidated. However, as skeletal muscle plays an integral role in IR, it is likely that changes in skeletal muscle structure and function may be associated with diminished exercise function. In our study of persons with T2DM in which VO2max was lower, cardiac index was reduced by about 15% and yet arteriovenous oxygen extraction was the same in the T2DM subjects compared to obese controls (140). Baldi et al. (20) also reported a reduced VO2max, a trend toward lower cardiac output as measured by rebreathing techniques, and lower arteriovenous oxygen extraction in T2DM patients compared to controls. In their study VO2max correlated with the arteriovenous oxygen difference, but not with cardiac output. The findings from both groups are interesting since even a modest reduction in cardiac output should increase reliance on arteriovenous oxygen extraction. The absence of an increase in this measure suggests that defects in oxygen transport to and/or oxidative capacity of the exercising skeletal muscle exist in T2DM and may contribute to the exercise defects seen in this population.
Related to these mechanisms, capillary density is reduced in T2DM skeletal muscle (152), and basement membrane structures are altered (153). These structural changes could directly contribute to alterations in microvascular hemodynamics that impair O2 exchange from capillary to myocyte as suggested by the work in diabetic rodent models (154, 155, 156). Indeed, the relationship between oxygen diffusion (potentially decreased in T2DM) and exercise performance in T2DM has not been extensively explored (157, 158). However, microvascular complications of T2DM have been associated with abnormal vascular function and lowered exercise capacity (159) further suggesting this mechanism as a component to the exercise dysfunction in T2DM. There is currently debate regarding the potential for abnormalities of mitochondrial function (21, 160, 161, 162) and whether they relate to functional defects in exercise performance or simply reflect reduced content secondary to detraining (163). To date, the available data are inconclusive, but rigorous studies are lacking. Nevertheless, adults with T2DM have been shown to demonstrate reduced skeletal muscle oxidative enzyme activity (162), lower mitochondrial content (161, 164), and an increased ratio of type IIb-to-type 1 muscle fiber ratio (165) compared with healthy subjects. Any of the factors may lead to reduced fractional oxygen extraction. Other, noncardiac components of oxygen delivery could also cause impairment in exercise performance in T2DM. Increased blood viscosity has been reported in persons with T2DM compared to nondiabetic individuals (157, 158). However, we found that while average whole blood viscosity was higher in persons with T2DM than nondiabetic controls, there was not a statistical relationship between viscosity and exercise performance (23). Overall it appears that the ability to deliver oxygen to the skeletal muscle as well as the ability of the muscle to utilize oxygen during exercise may be compromised in T2DM, and that this is another potential mechanism underlying the exercise defects seen in T2DM.
4.5 GENDER SPECIFICITY OF EFFECTS OF T2DM ON EXERCISE CAPACITY
Few previous studies have separately examined exercise performance in women with T2DM although it has been noted that they appear to have a reduced exercise performance compared to nondiabetic women (121). In prior studies, we have observed that women with T2DM had a more impaired exercise performance relative to their nondiabetic female counterparts, than the men with T2DM compared to their nondiabetic male counterparts (96). Maximal oxygen consumption was 26% lower in women with T2DM than nondiabetic women compared to 18% lower in men with T2DM compared to the nondiabetic men (P < 0.05). Although the small sample size makes these findings preliminary, the results are suggestive of possible sex-based differences in exercise performance between men and women with T2DM.
4.6 EFFECTS OF EXERCISE TRAINING ON EXERCISE PERFORMANCE IN T2DM
Exercise training can substantially improve exercise performance in individuals with T2DM (15, 109, 166). Improvements in VO2max in men and women with diabetes ranging from 8 to 30% have been documented (92, 166, 167). In addition, a decreased heart rate per submaximal workload has been reported (109) suggesting an improved exercise efficiency, again similar to results in nondiabetic persons. Oxygen uptake kinetics and heart rate kinetics became faster after 4 months of exercise training in persons with T2DM although not in nondiabetic controls suggesting improvement in the rate of circulatory adjustment to the beginning of exercise (92).
4.6.1 Exercise Training: Mechanisms of Improvement
Metabolic benefits in terms of how exercise improves insulin sensitivity are likely related to increased tissue sensitivity to insulin due to regular exercise conditioning (168, 169). Studies have shown that insulin binding to monocytes (170, 171) and erythrocytes (172) is increased by exercise conditioning and decreased with inactivity. It is possible that exercise conditioning causes a diminished secretion of insulin in response to a particular glucose concentration (173). Studies have suggested that exercise conditioning magnifies insulin-induced increases in the intrinsic activity of plasma membrane glucose transporters (174).
As discussed earlier, T2DM may adversely affect exercise performance in part because of detrimental effects on diastolic function. Exercise training might be expected to improve diastolic dysfunction based on animal studies (175), but randomized studies of exercise training are needed to confirm this benefit and provide the responsible mechanism.
4.6.2 Exercise and Endothelial Vasodilator Function
The beneficial effects of exercise on endothelial function have been suggested by both animal and human studies where exercise was associated with improved endothelial vasodilator function (54, 176). In humans with T2DM, reactive hyperemic brachial artery vasodilation and forearm blood flow have been improved by exercise in contrast to the response in nondiabetic controls (176). It is thought that the improvements represent a systemic rather than a local benefit of exercise since while the exercise was done using the lower body muscles, improvement in brachial artery reactivity in the arm was a primary outcome. Further research in this exciting area is underway.
The relationship between CV exercise capacity and diabetes is complex and involves multiple physiological systems. Furthermore, the relationship is likely to represent bidirectional causality. The benefits of exercise (and, conversely, the ill effects of sedentary behavior) on CV risk factors, endothelial function, insulin sensitivity, diabetes prevention, and CV and all-cause mortality are clear. Other benefits including maintenance of mitochondrial health and number, and effects on hemostasis and systemic inflammation are likely, but less well defined. It also seems likely that further research will reveal other areas of benefit derived from regular exercise. On the other hand, individuals with T2DM who would be expected to benefit disproportionately from exercise have been shown to be relatively inactive and cardiovascularly unfit. While the increased risk of diabetes in sedentary individuals is undoubtedly one contributor to this relationship, recent evidence suggests that diabetes may itself cause defects in CV exercise capacity. These defects in turn may make exercise more difficult and uncomfortable and thus encourage sedentary behavior in the very population that would most benefit from exercise. The mechanism of decreased exercise capacity in T2DM is poorly understood, but appears to involve impaired oxygen delivery through cardiac and vascular mechanisms, as well as impaired oxygen utilization at the tissue level. A better understanding of these mechanisms and of the benefits of exercise in this population is essential and awaits further research.
- 11.11. Allen F, Stillman E, Fitz, R. Total dietary regulation in the treatment of diabetes. Exercise, Vol 11. NY: Rockefeller Institute of Medical Research; 1919:486–499.Google Scholar
- 31.31. Hautala AJ, Leon A, Skinner JS, Rao DC, Bouchard C, Rankinen T. Peroxisome proliferator-activated receptor delta polymorphisms are associated with physical performance and plasma lipids: the HERITAGE family study. Am J Physiol Heart Circ Physiol. Jan 26 2007;292: H2498–H2505.PubMedGoogle Scholar
- 63.63. Festa A, D'Agostino R, Jr, Rich SS, Jenny NS, Tracy RP, Haffner SM. Promoter (4G/5G) plasminogen activator inhibitor-1 genotype and plasminogen activator inhibitor-1 levels in blacks, Hispanics, and non-Hispanic whites: the insulin resistance atherosclerosis study. Circulation. May 20 2003;107(19):2422–2427.PubMedGoogle Scholar
- 65.65. Olson TP, Dengel DR, Leon AS, Schmitz KH. Changes in inflammatory biomarkers following one-year of moderate resistance training in overweight women. Int J Obes. Feb 13 2007;31(6):996–1003.Google Scholar
- 91.91. Schneider SH, Vitug A, Ruderman N. Atherosclerosis and physical activity. Diabetes Metab Rev. 1986;1(4):513–553.Google Scholar
- 97.97. Wasserman K, Hansen JE, Sue DY, Whipp BJ, Casaburi R. Principles of Exercise Testing and Interpretation, Second ed. London: Lea & Febiger, 1994.Google Scholar
- 100.100. Whipp B, Mahler M. Dynamics of pulmonary gas exchange during exercise In: West J, ed. Pulmonary Gas Exchange, Vol 2. New York: Academic; 1980:33–96.Google Scholar
- 121.121. Regensteiner JG, Popylisen S, Bauer TA, et al. Oral l-arginine and vitamins E and C improve endothelial function in women with T2DM. Vasc Med. 2003;8(3):169–175.Google Scholar
- 133.133. Mustonen JN, Uusitupa MI, Tahvanainen K, et al. Impaired left ventricular systolic function during exercise in middle-aged insulin-dependent and noninsulin-dependent diabetic subjects without clinically evident cardiovascular disease. Am J Cardiol. Dec 1 1988;62(17):1273–1279.PubMedGoogle Scholar
- 138.138. Robillon JF, Sadoul JL, Jullien D, Morand P, Freychet P. Abnormalities suggestive of cardiomyopathy in patients with T2DM of relatively short duration. Diabetes Metab.Sep–Oct 1994;20(5):473–480.Google Scholar
- 140.140. Regensteiner J, Groves, BM, Bauer, TA, Reusch JEB, Smith, SC, Wolfel, EE. Recently diagnosed T2DM mellitus adversely affects cardiac function during exercise. Diabetes. 2002;51 (Suppl):A59.Google Scholar
- 157.157. MacRury SM, Small M, MacCuish AC, Lowe GD. Association of hypertension with blood viscosity in diabetes. Diabetes Med. Dec 1988;5(9):830–834.Google Scholar
- 166.166. Holloszy JO, Schultz J, Kusnierkiewicz J, Hagberg JM, Ehsani AA. Effects of exercise on glucose tolerance and insulin resistance. Brief review and some preliminary results. Acta Med Scand Suppl. 1986;711:55–65.Google Scholar
- 168.168. Berntorp K, Lindgarde F, Malmquist J. High and low insulin responders: relations to oral glucose tolerance, insulin secretion and physical fitness. Acta Med Scand. 1984;216(1):111–117.Google Scholar