Cardiometabolic Risk Factors in Children

Abstract

Considerable evidence has accrued that cardiovascular disease (CVD) has its beginnings during childhood, with some reports of endothelial damage occurring in early adolescence. Thus, early prevention and intervention on developing cardiometabolic risk factors is important. Presently, diet and exercise are strategic parts of any CVD prevention or treatment program. Although diet is important, the effects of exercise training or regular moderate-to-vigorous intensity physical activity go beyond simply modifying the traditional risk factors. For children, the majority of studies show that regular physical activity can lower elevated blood pressure, positively influence lipid profiles, reduce inflammation, and improve vascular functioning. These changes are most evident in children who exercise and lose weight. Despite these potential benefits of physical activity, not all studies on children have found positive effects of habitual physical activity on CVD risk factors. Reasons for these differences are discussed throughout this review. The literature also suggests that habitual physical activity, although meeting guidelines, may not be sufficient for all individuals to bring about changes; thus, to be an effective modality physical activity needs to increase beyond habitual levels or be of a moderate to vigorous intensity. This review provides recent evidence (since 2008) regarding the influence of physical activity on the cardiometabolic risk factors in children.

“Considerable evidence has accrued that cardiovascular disease (CVD) has its beginnings during childhood, with some reports of endothelial damage occurring in early adolescence.”

Considerable evidence has accrued that cardiovascular disease (CVD) has its beginnings during childhood, with some reports of endothelial damage occurring in early adolescence. ¹ Since the process begins early in life, childhood is the time of life during which prevention can be most effective. The challenge is to determine which health-related factors should be the focus of an intervention—environmental, genetic, lifestyle, or biologic. In adults, associations can be developed between death from CVD and health-related factors; thus, the decision is much easier as dyslipidemia, hypertension, and glucose intolerance have all been associated with CVD morbidity and/or mortality. In children, the disease has not yet progressed to the same endpoint as in adults. In addition, conditions such as glucose intolerance change with chronological age and/or pubertal stage and differ between the sexes in children. Consequently, the issues regarding prevention and selection of risk factors are complicated. ²

Many of the CVD risk factors are interrelated and can influence each other; thus, there are ongoing discussions of which risk factors are most important. For example, traditional risk factors such as blood pressure, insulin resistance, and obesity are interrelated ³ , but the more contemporary factors of oxidative stress (reactive oxygen species [ROS]), fibrinogen, plasminogen activator inhibitor-1 (PAI-1), and adipokines are also interrelated with obesity.^4-6 High levels of circulating lipids are associated with insulin resistance, ultimately resulting in the formation of atherosclerotic plaques. ¹ Finally, low-grade inflammation exacerbates all stages of the atherosclerotic process, including plaque development, disruption, and thrombosis. ¹ Thus, obesity could be considered an underlying factor; however, in some individuals these risk factors can exist in the absence of obesity.

Presently, interventions in children have focused their results on the traditional major risk factors for CVD: blood pressure, total cholesterol, low-density lipoprotein (LDL) cholesterol and triglycerides, high-density lipoprotein (HDL) cholesterol, glucose intolerance, and obesity. These factors have been recently termed cardiometabolic risk factors. ⁷ However, other more contemporary factors have become evident, including lipoprotein (a), markers of inflammation (C-reactive protein [CRP] and cytokines), homocysteine, ROS, elevated adipokines, elevated PAI-1, and possible Brain Derived Neurotropic Factor (BDNF). The majority of these nontraditional factors are related to obesity.^5,8

The goal of any intervention is to modify these risk factors most commonly through diet and increased physical activity (PA), and there have been numerous reports on the effects of these modalities on cardiometabolic risk factors in children. ⁹ Although diet has shown to be effective in improving these cardiometabolic risk factors, ¹⁰ PA also has a significant impact on these factors. In adults, the effects of exercise training, or regular moderate-intensity PA, on CVD go beyond modifying the traditional cardiometabolic risk factors by improving endothelial vascular function autonomic function and reducing oxidative stress. ¹¹ However, these relationships are more complex in children due to the influence of normal growth and maturation. Thus, the purpose of this review is to highlight recent evidence regarding the influence of PA on the cardiometabolic risk factors in children that has accrued since 2008, particularly focusing on the newer nontraditional or novel risk factors.

Defining Physical Activity and Differentiating It From Aerobic Fitness

Studies examining the impact of exercise on cardiometabolic risk factors have taken differing approaches. One approach has been to focus on cardiorespiratory fitness or aerobic power, of which maximal oxygen uptake (Vo_2max) is a key indicator. These studies have shown that higher levels of Vo_2max are related to improved cardiometabolic risk profiles.^9,12-15 The problem with this approach is that aerobic power is a physiologic trait, not a behavior. Although exercise training can improve this trait, the phenotype and its trainability are influenced by genetics. ¹⁶ An underlying assumption is that high aerobic fitness occurs in the most physically active children. Studies investigated this relationship in a cohort of young children and adolescents and found that PA levels accounted for only 2% to 9% of the variance in VO_2max for young or adolescent boys.^13,17 There was no clear relationship for young girls, ¹⁷ but a significant relationship existed for older girls. ¹³ Kristensen et al have shown that in the least active children, increased levels of moderate to vigorous PA were related to increased aerobic fitness. ¹³ Another issue regarding the relationship between CVD risk factors and Vo_2max is that the most common unit of measurement is mL O₂/kg body mass/minute. Body mass is made up in part by body fat—a factor directly associated with the cardiometabolic risk factors.^15,18,19 Since the relationship between PA and aerobic fitness is somewhat tenuous and in need of further clarification, we are focusing on PA as a behavior and not a trait.

Another approach for exercise studies has been to evaluate participation in PA (either habitual levels or in an exercise training program). Participation in PA is a behavior—a choice made by the individual. PA includes all movements that increase energy expenditure above resting level, not only those intended to increase fitness (eg, moderate-to-vigorous intensity PA), as is the case for exercise. As a result, participation in PA may not necessarily lead to improvement in aerobic power, although it has been related to improvements in cardiometabolic risk factors in some studies of children, but not all.^9,14,20 Thus, the modifiable behavior of PA in children is the focus of this review.

For a long time PA has been measured using questionnaires. These surveys require the children to remember what they did the previous day, previous 3 days, or even the entire week. The problem with this approach has been children’s ability to accurately recall their PA, ²¹ and in addition, overweight children tend to overestimate their activity levels. ²¹ Although other methods to measure PA have been tried (eg, heart rate, thermoband), accelerometry has proven to be a more precise, objective means of estimating PA. A good resource on the topic is the November supplement to Medicine & Science in Sports & Exercise, Volume 37, 2005. Although accelerometry is a better method for estimating PA, it has its problems especially in activities that are nonambulatory, such as swimming, cycling, rowing, or weight lifting. The differing approach to measuring PA could lead to differing results, as will be noted in the later discussions.

Potential Benefits of Physical Activity

Physical activity has the potential to reduce cardiometabolic risk factors via several mechanisms.^11,22 Exercise increases metabolic rate and fat catabolism. This results in a reduction of basal triglyceride levels. Regular exercise improves the lipid enzyme profile (cholesterol ester transfer protein, lecithin, cholesterol acyl transferase, and lipoprotein lipase), which, in turn, can result in modest reduction in LDL cholesterol and elevations of HDL cholesterol. Exercise can also improve insulin sensitivity and blood glucose control—improving glucose tolerance. The elevated metabolic rate can also result in loss of fat mass. The loss of fat mass in an overweight child can decrease the vasculature needed to support the adipose tissue and, in accordance to LaPlace’s Law, should result in lower blood pressure. Reduced fat mass can also reduce the amount of adipose-derived inflammation (CRP, interleukins, PAI-1). In addition, regular participation in exercise increases the enzymes that counter free radicals (SOD, catalase, glutathione reductase), thereby reducing potential free radical damage. Regular participation in exercise (and weight loss) causes a reduction in resting sympathetic and adrenal outflow. This has the potential to reduce elevated blood pressure by reduced α-adrenergic vasoconstrictive effects, reduce resting heart rate, and, in some cases, improve endothelial functioning. Regular exercise can also improve the adipokine profile, which minimizes endothelial damage, ²² reduces the risk of insulin-producing β-cell apoptosis, and improves muscle cell insulin signaling. Finally, regular exercise has the potential to alter vascular remodeling, which increases angiogenesis and arteriogenesis. ²² Thus, the potential effects of exercise can be far reaching for general health, as well as slowing the development of metabolic and CVDs. One caveat is that the effects of PA and exercise on the cardiometabolic risk factors are more noticeable in children who exhibit elevated risk factors and particularly in obese children. In children with normal risk factor values (or normal weight), there may be little or no effects. This may have caused some of the ambiguity in the literature.

Despite these potential benefits of exercise, not all studies of youth have found positive effects of exercise on cardiometabolic risk factors.^2,23,24 Thus, the purpose of this review is to examine the recent findings on the effects of regular participation in PA on the cardiometabolic risk factors in children using cross-sectional and longitudinal studies. Although the traditional risk factors are discussed, the majority of the review focuses more on the nontraditional or novel risk factors. An attempt will be made to translate the “theory” (as described in the Introduction) into the “reality.” The review is arranged to examine what is known regarding each of the cardiometabolic risk factors separately. Within each of the factors, the cross-sectional study results will be presented first followed by the longitudinal studies. The longitudinal studies include both exercise training studies and studies of children whose habitual PA levels have been followed over time. Finally, the review will culminate with a short summary and recommendations.

Obesity

The significance of obesity on the development of CVD in youth cannot be over stated. ²⁷ Childhood obesity has been linked to elevated blood pressure, hypertriglyceridemia, insulin resistance, and type 2 diabetes mellitus, hypercoagulability, mild inflammation, and increased levels of soluble adhesion molecules, which initiate the atherosclerotic process.^1,25-28 Thus, for children, the prevention or treatment of obesity is a prime consideration. Although the literature on this topic is extensive, far too much to be included in this review, the majority of studies show that participation in an exercise programs can potentially induce weight loss and that habitually active individuals usually have a lower risk of obesity.

Blood Pressure

Elevated resting blood pressure, or hypertension, has long been recognized as a key risk factor for CVD. Measures of systolic and diastolic pressures indicate the degree of strain placed on the blood vessels during the contraction and relaxation cycle, respectively. Several recent studies have demonstrated that blood pressure is related to body composition, physical fitness, and PA levels in youth. For example, in a large study of 11- to 12-year-old youth, systolic blood pressure (SBP) was inversely related to total and moderate-to-vigorous intensity PA as measured by accelerometry. ²⁹ However, when the amount and intensity of PA were analyzed in the same model, it appears that amount of PA was more influential than the intensity at which it was performed. ²⁹ This was one of the first studies to show that the volume may be more important than intensity of PA. Another large study of 11- to 18-year-old Chinese youth provided support for the relationship between blood pressure and frequency of structured exercise (excluding physical education classes) in youth. ³⁰ These authors reported an odds ratio of 0.63 (P = .002) for hypertension between youth who took part in structured PA at least 2 times per week and those who did not participate, even after adjusting for age, sex, family history of hypertension, sleep duration, and body mass index (BMI). These relationships were nearly identical for youth who exercised at least 3 times per week (odds ratio = 0.62), but were noticeably lacking when youth only exercised once per week. This suggests that for high blood pressure to be affected, youth must engage in structured PA at least twice per week, in addition to their schools physical education curriculum.

Intriguingly, a cross-sectional investigation in youth (~11.3 years) revealed that blood pressure was more strongly related to fatness than it was to physical fitness. ³¹ In this study, SBP was significantly higher in obese youth compared with normal weight youth, whereas diastolic blood pressure (DBP) differed significantly among obese, overweight, and normal weight youth. ³¹ SBP did not differ according to physical fitness quintiles in youth, but there were trends for lower DBP for the most fit boys and girls. These results suggest that fatness may be more important in controlling blood pressure in youth than having a high level of physical fitness. However, this should not be misinterpreted to imply that that being physically active is not important.

There have also been several recent longitudinal studies investigating the relationship between blood pressure and exercise in youth. For example, at the conclusion of a 3-year intervention aimed at increasing physical education from 90 to 180 minutes per week, researchers found smaller increases in SBP for boys in the intervention group compared with boys in the control group. ³² This occurred despite similar mean PA levels and body fat levels between the groups. ³² These BP trends were not observed in girls. Another exercise intervention conducted in the school setting for overweight youth resulted in significant reductions in numerous cardiometabolic risk factors including SBP. ³³ The 15-week intervention involved 60-minute sessions of moderate-to-vigorous intensity PA 3 times each week. An encouraging finding was the 50% reduction in the prevalence of hypertension among the sample at the end of the intervention—10 children (21% of the sample) at baseline and 5 at the conclusion. ³³ Likewise, promising results were found in a study examining whether the benefits of a PA intervention for obese youth were maintained 2 years after its conclusion. ³⁴ While BMI z-score and PA levels remained stable after the conclusion of the intervention, the rates of hypertension declined (50% to 28% for SBP and 42% to 6% for DBP) in the 2 years following the intervention. These results suggest that favorable changes in cardiometabolic risk factors associated lasted well beyond the conclusion of the intervention. Taken together, these longitudinal studies provide evidence that well-planned PA interventions in youth can have significant reductions in SBP and DBP, especially in those children with elevated pressures.

Lipids

Alterations in blood lipids indicating increased risk for CVD include elevated total and LDL cholesterol and triglycerides and decreased HDL cholesterol levels. Data from longitudinal research has shown that altered blood lipids are associated with changes in PA and body weight in youth. For example, in a recent study of overweight youth involved in a weight management program that included a significant exercise component, 65% had dyslipidemia at baseline, and this was the most common comorbidity. ³⁵ A closer examination of their pre-intervention data showed that 52% of participants had elevated triglycerides, 47% had elevated LDL cholesterol, 41% had elevated total cholesterol, and 29% had low HDL cholesterol levels. ³⁵ After the intervention, significant improvements were found in each of these measures: median values of −12.5 mg/dL for cholesterol, −15 mg/dL for triglycerides, −6 mg/dL for LDL cholesterol, and 3 mg/dL for HDL cholesterol. ³⁵

Several exercise interventions also reported significant changes in blood lipids for obese youth. Kovács et al found a small but significant reductions in LDL cholesterol in obese children (ages 6.5-12.5 years) following a 15-week school-based exercise intervention. ³³ Likewise, the number of participants with elevated triglycerides declined by 22% following the training program. Similarly, another study reported significant, favorable changes in total cholesterol, triglycerides, HDL cholesterol, and LDL cholesterol in youth (ages 10-13 years) who participated in an intervention aimed at increasing PA levels and increasing their knowledge about cardiovascular health. ³⁶ Another intervention aimed at increasing PA in obese youth also resulted in significant reductions in total cholesterol, LDL and VLDL cholesterol, and triglyceride levels as well as increases in HDL cholesterol, compared with the control group, following 12-weeks of activity. ³⁷ Finally, in a study comparing adolescents who traveled to and from school via bicycle versus other modes of transportation, those who cycled had lower triglyceride levels compared with those who did not. ³⁸ Interestingly, when these participants were first studied 6 years prior (at age 9 years), there were no significant differences in triglyceride levels among cyclists and noncyclists. These results suggest that something as simple as cycling to and from school can reduce an adolescent’s risk for CVD. Taken together, these studies highlight the importance of achieving healthy body weight and increasing PA levels in an effort to achieve healthy blood lipid values during childhood and adolescence.

Glucose Regulation

A remarkable number of youth are developing prediabetes—diagnosed by elevated fasting glucose levels and varying degrees of resistance to insulin. ³⁹ Unfortunately, most of these youth will go on to develop type 2 diabetes unless intervention occurs. ⁴⁰ The most common method of regaining control of glucose and insulin levels is through attainment of a normal body weight by increased PA and healthy nutrition choices. Numerous studies in youth have demonstrated the relationship between glucose and insulin with body weight and PA, although the strength of the relationships varies depending on the population studied and methods used. ⁴¹ Longitudinal studies investigating the relationship between weight change, PA, and glucose or insulin have shown favorable results in youth. The majority of these studies involved interventions lasting at least 12 weeks and targeted at obese youth. For example, fasting insulin and HOMA-IR decreased in youth who took part in a 12-week PA intervention. ³⁶ Likewise, HOMA-IR decreased in obese Korean youth following a 12-week intervention, while these changes were not seen in the control group. ⁴² The intervention group took part in 90 minutes of PA 2 times per week and this stimulus elicited significant increases in aerobic fitness compared with the control group. In a similar 12-week exercise intervention, insulin levels were significantly lower in the exercise group compared with the control group. ³⁷ Finally, a study by Jekal and colleagues ⁴³ involved a 12-week exercise intervention of obese, unfit youth in Korea, and the results were promising: insulin resistance improved, as well as the composite CVD risk score, body mass, and Vo_2max.

Additional support for the importance of PA was shown in a study where boys who were a part of a 3-year physical education intervention had a smaller increase in their HOMA-IR score compared with boys who were in a control group. ³² These trends were not observed for girls. The aforementioned study investigating the role of cycling to and from school reported that youth who cycled had significantly lower glucose, insulin, and HOMA compared with youth who traveled by other means. ³⁸ Similar to the trends found for triglycerides, these differences were only seen in the participants when they were 15 years old and not when they were 9 years old. Perhaps the most encouraging results from this investigation were the significantly lower glucose, insulin, and HOMA values in youth who were noncyclists at baseline and became cyclists 6 years later, compared with those who remained noncyclists. These results demonstrate that important metabolic changes can be achieved by choosing a more active mode of transportation to and from school.

Finally, the effects of a lifestyle intervention alone and in combination with metformin were assessed in obese, insulin-resistant youth (age ~ 13 years). In this investigation, researchers reported that after 6 months of lifestyle intervention, HOMA was significant reduced, due to a lowering of insulin, but fasting glucose did not significantly decline. ⁴⁴ The reductions in HOMA in the group that received a lifestyle intervention and metformin, however, were not significant. The lifestyle intervention included nutritional guidance, assessment of physical fitness and assistance developing goals and a PA plan, analysis of the youth’s support system and suggestions for overcoming barriers, as well as strategies to assist their parent or guardian in helping the children achieve their goals. These data are encouraging as they demonstrate the benefits of lifestyle changes for improving insulin resistance, independent of metformin. ⁴⁴

Plasminogen Activator Inhibitor-1

Plasminogen activator inhibitor-1 is involved in the atherogenic process and is a significant marker for recurrent myocardial infarction. ⁴⁵ PAI-1 inhibits fibrinolysis and increased levels of PAI-1 impair fibrinolytic function and enhance thrombogenesis. PAI-1 is normally secreted by endothelial cells, vascular smooth muscle cells, hepatocytes platelets, and adipocytes. In addition, TNF-α, which can also be produced by the adipocyte, stimulates PAI-1 production. ⁴⁶ Plasma leptin levels are correlated with PAI-1 levels, independent of insulin and fat mass, although a regulatory effect remains to be clarified. ⁴⁷

Studies on children agree that overweight and obese children have higher levels of PAI-1 than normal weight children.^48-51 Akanji et al found that levels of PAI-1 were 4 to 8 times higher in overweight adolescents compared with normal weight adolescents. ⁴⁸ Balagopal et al noted that PAI-1 levels were 5 times higher in the obese adolescents compared with normal weight adolescents, ⁴⁹ whereas Nienaber et al found PAI-1 levels to be approximately 50% higher in obese versus nonobese adolescents. ⁵⁰ Finally, researchers found a relationship between waist circumference and PAI-1 in children as young as 9 years of age, but there appears to be no information on PAI-1 in young children. ⁶

With respect to exercise, Nienaber et al reported that PAI-1 levels were 46% higher in active adolescents—the opposite of what one would expect. ⁵⁰ PA levels were estimated from a questionnaire. In contrast, Akanji et al, using a rudimentary questionnaire, found that PAI-1 did not differ between those that said they exercise and those who did not exercise. ⁴⁸ Few recent exercise intervention studies have examined PAI-1. Balagopal et al investigated the effects of a 3-month PA-based intervention in obese adolescents. ⁴⁹ The adolescents participated in a supervised PA program of 45 to 60 minutes, 3 times per week. Although PAI-1 levels were 5 times higher in the obese adolescents compared with normal weight adolescents, participation in the 3-month program did not have any effect on PAI-1. The lack of an effect of the program could be related to insufficient exercise or failure of the exercise to normalize glycemic control.

Hypercoagulability

Fibrinogen is a glycoprotein required for blood clotting. Higher levels (>3.4 g/L) are associated with CVD because of the increased clotting capacity of blood. Fibrinogen is typically elevated in obese children,^25,48 suggestive of a state of hypercoagulability. There are few studies that have examined fibrinogen and PA in children. An early study by Zahavi et al ⁵² and a more recent study by Nienaber et al, ⁵⁰ using a PA questionnaire, suggested that there was a negative (unspecified) relationship between PA levels and fibrinogen. In contrast, several researchers found no significant relationship between PA levels and fibrinogen.^53-55

The influence of exercise training on fibrinogen concentration in children is equivocal. Balagopal et al showed that a 3-month lifestyle intervention (diet and exercise) reduced fibrinogen and also adiposity. ⁵¹ Meyer et al reported that a 6-month exercise program (1 hour, 3 times per week) decreased the fibrinogen levels of obese adolescents. ⁵⁶ However, the participants lost body fat and the investigators did not determine whether the change in fibrinogen was independent of the weight loss. Barbeau et al examined the effects of an exercise program on fibrinogen levels of a small cohort of obese adolescents. ⁵⁷ They found that 12 weeks of high-intensity training (1047 kJ or 250 kcal/session at >70% maximal capacity) did not significantly reduce fibrinogen. Similarly, Ferguson et al found no overall effect of a 4-month exercise program on the fibrinogen levels of obese children. ⁵⁸ The authors noted that the children with greatest body fat and fibrinogen levels at baseline had the greatest reductions in fibrinogen at the end of the 4 months. The combined effects of diet and exercise on hemostatic function were examined by Fritsch et al. ⁵⁹ They found that a 1-year program resulting in weight loss completely normalized fibrinogen levels. Thus, a longer intervention that led to a greater decrease in adiposity may be required to cause significant changed in fibrinogen.

Another study examined the influence of weight loss, induced by a diet and exercise program, on fibrinogen. Over the 6 months the participants lost approximately 5 kg, and circulating fibrinogen levels significantly declined. ³ Unfortunately, the specific effect of exercise cannot be determined, but the results suggest that exercise can be a vital part of a weight loss program that improves coaguability of blood.

Inflammatory Markers

C-reactive protein is emerging as one of the most powerful predictors of CVD since it is correlated with coronary heart disease in several large cross-sectional and prospective population studies.^60-63 Its physiological role is to bind on the surface of dead or dying cells in order to activate the complementary part of the immune response. CRP can be produced in the adipocyte, ⁶² but is synthesized mainly by the liver in response to inflammation and also to factors released by adipocytes (tumor necrosis factor-α [TNF-α], interleukin-6 [IL-6]). CRP is significantly associated with established CVD risk factors including obesity, hypertension, and dyslipidemia.^26,64 In addition, Lau and associates note that CRP directly participates in the process of atherogenesis by inducing the expression of VCAM-1 and ICAM-1. ⁶² CRP attenuates basal and stimulated endothelial nitric oxide production, which, in turn, may inhibit angiogenesis, an important compensatory response in chronic ischemia. Furthermore, CRP upregulates angiotensin type 1 receptor that facilitates angiotensin II–induced vascular smooth muscle cell migration and proliferation and vascular remodeling. Interestingly, the of CRP on endothelial dysfunction is potentiated by hyperglycemia. ⁶⁵ In children, CRP levels are clinically higher in overweight and obese youth as well as those with insulin resistance and elevated triglycerides.^6,50,65,66 In addition, youth with increased intima thickness had higher CRP levels. ²⁶

Studies on CRP and PA in children are sparse but have been occurring for over a decade and have produced mixed results. Early cross-sectional studies,^50,64,67,68 using PA questionnaires, have shown that higher levels of exercise are associated with lower CRP. In contrast, many other researchers noted no relationship between CRP and PA levels (overall, moderate intensity, or vigorous intensity) when measured by questionnaire or accelerometry. For example, Nienaber et al found CRP levels to be similar for active and inactive adolescents when they evaluated activity levels using the PDPAR questionnaire. ⁵⁰ Ruiz and colleagues, using accelerometry to estimate PA levels, also reported no relationship between PA and CRP levels in 9- to 10-year-old youth participating in the European Youth Heart Study. ⁵⁵

With respect to longitudinal exercise training studies, Carrel et al introduced a fitness orientated physical education class for an entire school year in adolescents and found no change in CRP by the end of the school year. ⁶⁹ Similarly, Kelly had obese adolescent participate in an 8-week stationary cycling program (50% to 70% of maximal capacity) and found no change in CRP. ²⁴ In contrast, other studies have noted a decline in CRP with a 12-week, 90 min/session, 2 session/week intervention ⁴² and as little as a 2-week program of 3.5 hours of exercise per day. ⁶³ Balagopal et al had obese youth walk briskly for 45 minutes, 3 times a week for 3 months, with some dietary restriction. ⁴⁹ They noted that CRP declined about 33%; concomitantly the participants lost approximately 1.2 kg in body mass. However, it was not clear whether it was the exercise or the weight loss that led to the decline in CRP.

It is possible that for some children habitual levels of PA (even though high) are not sufficient to modify CRP and an increase in PA over habitual levels or a reduction in weight, or increase in cardiometabolic fitness may be necessary to bring about a change in CRP. ⁵⁵ In support, Rosenbaum et al ⁷¹ had 20 adolescents complete a 4-month program, whereas Meyer et al ⁵⁶ had 33 obese adolescents complete a 6-month exercise program, and in both studies of sedentary adolescents CRP levels significantly declined. Neither program reported the intensities of their exercise. Rosenbaum did not provide information on their program other than to say it was 3 times per week, whereas Meyer indicated that the program was 3 times a week for 60 minutes. Garanty-Bogacka and associates exercised 50 obese children for 6 months and simultaneously restricted caloric intake. ³ They noted that CRP levels declined but did not determine if the exercise, caloric restriction, or the combined effect on weight loss (~5 kg) was the most significant component. Balagopal et al provided a 45-minute, 3 times per week exercise program for 3 months (with a diet and reduced TV viewing) and found that the majority youth (7 out of 8) reduced CRP. ⁵¹ In contrast, no significant effect of a 12-week aerobic training program on CRP was found in a study of overweight girls. ⁷² Similarly, Barbeau et al reported no change in CRP concentration following an 8-month exercise program in obese youth. ⁵⁷

Cytokines

Cytokines are proteins, peptides, or glycoproteins that are secreted by numerous cells and are a category of signaling molecules used extensively in intercellular communication. Cytokines typically are referred to as interleukins and interferons. The majority of cytokines are secreted by T cells and macrophages to stimulate an immune response, but they are also produced in adipose tissue. The most studied cytokines in relation to CVD are IL-6 and TNF-α. One reason for this focus is that both cytokines are related to adipose tissue and there is some degree of autocrine communication between the 2 cytokines. ⁷³

The adipocyte produces IL-6 relative to the level of obesity. However, IL-6 is also secreted by myocytes in response to exercise. ⁷⁴ The role of IL-6 in insulin resistance and development of CVD is equivocal, since studies have shown both pro- and anti-inflammatory effects of IL-6.^74,75 Due to its direct stimulatory effect on the production of acute-phase reactants, including CRP, and its correlation with inflammation, it is generally considered pro-inflammatory. Also, IL-6 exacerbates ROS through its effects on CRP. The anti-inflammatory effects of IL-6 include inhibition of TNF-α and stimulation of the production of the anti-inflammatory IL-1 receptor antagonist and IL-10. ⁷⁴ Cross-sectional studies have related IL-6 to all stages of the pathogenesis of CVD,^76-78 but causality in the development of CVD has not been established. IL-6 has also been consistently related to insulin resistance or impaired glycemic control^79,80 and obesity.^25,81

TNF-α is a cytokine produced primarily by macrophages and also by a variety of other cells, including the adipocyte. TNF-α has a role in cellular uptake of glucose by impairing the insulin signaling cascade (IRS-1, PI3K), decreasing GLUT-4 content. It also decreases PPAR-γ production by adipose tissue cells, which is an insulin sensitizing protein. TNF-α is involved in systemic inflammation and stimulates the acute-phase reaction in inflammation. Chronic elevation of TNF-α has been found in obesity, including childhood obesity.^6,25,82 High circulating levels are related to CVD risk. ⁸ Although the specific role of TNF-α in the development of CVD is still not well understood, TNF-α was found to induce the expression of intercellular adhesion molecule (ICAM)-1, an inflammatory molecule involved in the early process of atherosclerosis. ⁸ Elevated TNF-α has also been associated with insulin glycemic control^82,83 and lipid metabolism. ⁸⁴

Higher levels of habitual PA are associated with lower circulating IL-6 and TNF-α. ⁸⁵ Conversely, IL-6 concentrations were significantly higher in sedentary adolescent girls. ⁸⁶ Furthermore, exercise training has been shown to reduce circulating levels of both cytokines. For example, Carrel et al introduced a school-based gym class aimed at improving fitness for obese children and the 9-month program significantly decreased TNF-α, without a reduction in weight. ⁶⁹ Balagopal et al had obese youth complete a brisk walking program (45 minutes, 3 times a week) and diet for 3 months. ⁸⁷ They found that IL-6 was reduced by approximately 25%; however, the participants also lost ~1.2 kg body weight. Similarly, a 6-month diet and moderate exercise program reduced circulating IL-6 by 55%. ³ Once again there was considerable weight loss (~5.3 kg) complicating the specific cause of the reduction. In contrast to the aforementioned studies, Kelly had a small number (n = 9) of obese youth complete 8 weeks of supervised stationary cycling, 4 times per week for 30 to 50 minutes at 50% to 70% of maximal capacity. ⁸⁸ At the end of the 8 weeks there were no significant changes in IL-6 or TNF-α, despite improvement in exercise capacity. However, there was no weight loss, which suggests the importance of weight loss over exercise for inducing changes in these cytokines.

Adipokines

Adipokines are a general class of proteins secreted by adipose tissue. Originally adipokines were thought of as immunomodulating agents, but now the term includes several hormones released from adipose tissue. These include IL-6, TNF-α, and PAI-1, which were discussed in earlier sections of this review. Besides these aforementioned inflammatory factors, the most heavily studied adipokines are adiponectin, resistin, leptin, and visfatin; the first 3 of which are actually adipose-derived hormones. Also included in this category are apelin, chemerin, and retinol binding protein-4. Although their specific physiology is still under intense investigation, they may be involved with immune functioning, atherogenesis, and endothelial function.

Adiponectin increases fatty acid oxidation and improves glucose uptake in skeletal muscle. It also blocks some of the metabolic pathways (NF-κB) directly related to atherosclerosis. Thus, adiponectin is “protective” against CVD and diabetes. ⁸⁹ Most studies have found that children with large amounts of fat mass have low circulating adiponectin levels,^90-93 although there is no total agreement among researchers. ⁶ Resistin is another adipokine, and although its functions are not well understood, it is known to be involved with glucose transport into the cell (by antagonizing insulin) and gluconeogenesis. It has also been shown to be associated with heart failure, later in life. ⁹⁴ Its presence in the blood is elevated by obesity. ⁸⁶ Adiponectin and resistin are also associated with insulin resistance in children. ⁸³ Leptin, classified as a hormone, has a role in modulating the immune response to atherosclerosis, although its primary purpose appears to be suppression of appetite. Visfatin is related to insulin resistance and oxidative stress and is higher in obese children. ⁹⁵

In general, habitual levels of exercise may not moderate the association between these cytokines and insulin resistance, except for children in the most extreme of activity levels—most sedentary versus most active.^96,97 For example, Rubin et al have shown that higher levels of VPA were associated with higher circulating levels of adiponectin and lower resistin levels. ⁸³

The effects of exercise interventions on adipokines are equivocal. On the favorable side, Carrel et al introduced a fitness-orientated exercise program to school physical education classes and at the end of the school year found that adiponectin levels increased. ⁶⁹ Reinehr et al had obese youth complete a 1-year exercise, diet and behavior modification program and found an increase in adiponectin. ⁹⁸ Koncsos et al had obese children complete a 2-week diet and exercise program and found increases in adiponectin immediately following the intervention. ⁷⁰ Conversely, several studies have found no effects of exercise programs last 8 weeks to several months on adipokines.^42,88,99 The problem with these studies is that the results are confounded by weight loss. To clarify the effects of exercise versus diet and weight loss, Shalitin and associates had 162 obese children complete one of three, 12-week interventions; exercise only (90 minutes moderate exercise 3 days/week), diet only (balanced hypocaloric), or combined diet + exercise. ¹⁰⁰ Weight loss was greater in the diet and diet plus exercise groups compared with the exercise-only group. Concomitantly, the adiponectin levels increased more in these 2 groups compared with the exercise-only group. Their results suggest that exercise alone may not be sufficient to modify cytokines, unless there an accompanying loss of weight. Clearly more studies are needed to unravel this conundrum.

The relationship between exercise and resistin has been less examined in youth. Rubin et al, using a cross-sectional design, found that high levels of vigorous PA were associated with lower circulating resistin. ⁸³ However Kelly et al found that 8 weeks of stationary cycling for 30 to 50 minutes at a moderate intensity did not significantly lower resistin levels of obese youth. ⁸⁸ Once again, more studies on resistin are needed.

The results of exercise studies on leptin are similar in that most interventions that result in weight loss also reduce leptin.^{70,87,88,98,99,101,102} Intriguingly, one study has shown that leptin responds to the intensity of exercise training, such that higher levels of leptin occur during harder levels of training. ¹⁰³ Therefore, in general, exercise that is greater than habitual levels and of sufficient intensity and duration to induce weight loss may be necessary to improve adipokine profiles in youth.

Oxidative Stress

The production of ROS, sometimes referred to as free radicals, naturally occurs with metabolism (mitochondrial in origin) as well as exposure to sunlight, pollutants, and as a by-product of certain disease states. These ROS are normally countered by enzyme systems within the body (superoxide dismutase, catalase, glutathione peroxidase). However, in some cases the production of the ROS is greater than the capacity to neutralize them— hence oxidative stress. This can result in damage to the molecular structure of cells, including the DNA. Not surprisingly, obesity increases oxidative stress.^4,104 ROS have been linked to CVD, since oxidation of LDL in the vascular endothelium is a precursor to plaque formation. ROS also plays a role in both strokes and heart attacks. Exercise training increases the counterregulatory enzymes, but how this affects ROS in children has not been studied. Such studies are difficult and the biochemistry involved is quite intricate. A need for further study on ROS exists, but at the present time studies on ROS may be more beneficial in exploring mechanisms of the development of CVD rather than as a marker for CVD. ⁹⁵

Homocysteine

Homocysteine is an amino acid usually synthesized from methionine. Although it has been related to CVD, the relationship is controversial, because deficiencies of folic acid (B₆), pyridoxine (B₆), or B₁₂ (cobalamin) can lead to high homocysteine levels. In addition, intense, long-duration exercise may increases plasma homocysteine levels. ¹⁰⁵ Also, chronic alcohol consumption results in increased plasma levels of homocysteine. ¹⁰⁶ Finally, hereditary plays a role in homocysteine levels for about 10% of the population. ¹⁰⁷

Homocysteine ≥15 µM/L has been associated with increased incidence of thrombosis and CVD. ¹⁰⁸ Homocysteine seems to interfere with the formation of collagen, elastin, and the proteoglycans found in arteries. Supplementation of the vitamins folic acid, B₆, and B₁₂ can reduce homocysteine levels and, in the long term, significantly reduce strokes by approximately 25%. ¹⁰⁹ One hypothesis is that a lack of folic acid may directly cause an increased build-up of arterial plaque, independent of its homocysteine-lowering effects. ¹¹⁰ Since studies in patients with preexisting CVD found no significant benefit or damage by supplementing these vitamins, ¹¹¹ the benefit could be related to prevention.^109,112,113 Alternatively, folic acid and vitamin B₁₂ may cause an overall change in gene methylation levels in vascular cells, which may also promote plaque growth. Finally, altering methylation activity in cells might increase methylation of l-arginine, which can increase the risk of vascular disease. Thus, alternative homocysteine-lowering therapies may yet be developed that show greater effects on the development and progression of CVD.

There is little information on homocysteine in children and no studies were found on PA, or exercise, and homocysteine in children. This void may be related to the difficulty of accurately assessing PA as mentioned earlier in this article. However, with regard to body mass, Pedrosa and colleagues found that homocystiene was higher in obese children. ¹⁰² Conversely, Martos et al found that not all obese children had higher homocysteine levels, only those with hyperinsulinemia. ⁹⁰ If homocysteine is related to hyperinsulinemia and exercise improves the glycemic state, then it follows that exercise should improve homocysteine levels. However, this is purely speculation and in need of confirmation. Any studies of homocysteine are complicated as folate and vitamins B₆ and B₁₂ status need to be taken into consideration.

Summary and Conclusions

The relationship between PA and the CVD risk factors in children is specific to the risk factor. The majority of evidence suggests that greater habitual levels of PA are related to lower blood pressure, improved glycemic control (risk of type 2 diabetes), and hemostatic factors, independent of obesity. The relationships between PA and blood lipids or some of the newer risk factors (eg, CRP, cytokines, adipokines) are not as strong but still in the expected direction. Failure to find positive results appears to be related to (a) the use of a healthy population such that an intervention would have minimal impact, (b) the intensity or duration of the PA in the intervention may not have been sufficient to cause a change, or (c) observational studies of cross-sectional populations in which the effects are only evident in the most extreme levels of PA—most sedentary or most active. In addition, the majority of the intervention studies that increase PA over and above habitual levels result in some improvement in the cardiometabolic risk factors. The key factor appears to be that PA levels must increase beyond habitual levels; and for those children who are overweight, a program that induces weight loss can provide further benefits over exercise alone.

Implications

Our present state of knowledge strongly suggests that PA can modify the CVD risk profile of children. ¹¹⁴ However, the question remains as to how much PA is needed to affect CVD risk factors in children. Recommendations by the US Centers for Disease Control and Prevention suggest that 30 to 60 minutes of daily moderate-to-vigorous PA are needed to improve health. ¹¹⁵ The Canadian guidelines are more specific and broken down by age ¹¹⁶ : infants (<1 year) should be active several times daily, through interactive floor-based play; toddlers (1-2 years) and preschoolers (3-4 years) should accumulate at least 180 minutes of PA throughout the day and include a variety of activities in different environments, activities that develop movement skills, and progression toward at least 60 minutes of energetic play by 5 years of age, similar to the guideline for older children. ¹¹⁷ These recommendations are based on the assumption that more daily PA provides greater benefits. The majority of the guidelines are somewhat theoretical, “best practice,” and not truly developed in relationship to a given health issue. Pahkala et al, using a PA questionnaire, recommended 10 hours of moderate intensity PA per week to improve endothelial function as estimated from maximum flow mediated dilatation. ¹¹⁸ Studies, like the Pahkala study, on the relationship between health issues and the amounts of PA, are needed to provide a definitive answer.

These general recommendations may work for some individuals, but there are numerous individuals whose habitual activity levels meet these guidelines, but still have significant CVD risk. ¹¹⁹ Thus, we would recommend that individuals follow the World Health organization 2010 Guidelines, ¹²⁰ which recommend 60 minutes of moderate-to-vigorous PA on top of or above habitual levels, in order to reduce CVD risk factors. Also, in children there appears to be a stronger relationship between aerobic power (Vo_2max) and CVD risk factors than simply amount of PA.^9,12 Trevino et al have shown that to induce positive changes in fitness in normal pediatric populations, the amount and frequency of exercise are important factors to consider. ¹²¹ Thus, the exercise should be of sufficient intensity and duration to increase aerobic power, which may require differing goals depending on the present levels of PA. Taken together, we advocate the World Health Organization guideline of increasing PA above habitual levels if risk factors are present, or if the child has a sedentary lifestyle.