In developed countries cardiovascular diseases – including coronary heart disease with acute myocardial infarction and ischaemic heart failure, stroke and chronic cerebrovascular disease, aortic aneurysm, valve disease and peripheral arterial occlusive disease – are by far the main cause of death. Since first cardiovascular events are usually seen in mid-life and old age, atherosclerosis has long been regarded as an inevitable degenerative disorder associated with ageing. However, autopsy studies performed in children and adolescents who have died from accidental causes have revealed early atherosclerotic lesions such as fatty streaks and intimal thickening of aorta and coronary arteries even in youth.1,2 Atherosclerosis as the clinical sign of vascular damage that develops over decades begins silently in early childhood and is strongly associated with cardiovascular risk factors. Recently, non-invasive methods have been established to detect subclinical atherosclerosis, such as endothelial dysfunction, thickening of the arterial intima media layer and arterial stiffness.
Subclinical Measures of Vascular Damage
Endothelial Dysfunction
Endothelial cells cover the underlying connective tissue with a non-thrombotic and non-adherent surface and modulate the vascular tone by releasing nitric oxide (NO), prostacyclin and endothelin.3 Flow-mediated arterial dilatation (FMD) in response to suprasystolic blood flow occlusion can be measured ultrasonographically and is a measure of endothelial function. The finger plethysmographic assessment of peripheral arterial tonometry is a novel, operatorin-dependent method of FMD evaluation.4
In the 1990s, Ross described the dysfunction of the endothelial layer as a condition sine qua non for the pathogenesis of atherosclerosis.5 Endothelial dysfunction is caused by a reduced quantity and activity of endothelial NO synthase and inactivation of NO by reactive oxygen species,6 increased radial strain as a result of arterial hypertension and a decline in regeneration of diseased endothelium by circulating endothelial progenitor cells released by the bone marrow.7
Several studies evaluating the prognostic impact of endothelial dysfunction demonstrate the crucial role of the endothelium in the development of clinically relevant atherosclerosis. Schächinger et al. were able to show that the presence of endothelial dysfunction in angiographically normal coronary arteries preceeds future cardiovascular events.8 This was confirmed by Perticonne, who demonstrated an inverse relationship between endothelium-dependent forearm blood flow and the risk of cardiovascular events in hypertensive patients.9 The Cardiovascular Risk in Young Finns Study showed that dyslipidaemia in children three to 18 years of age could be shown to predict endothelial dysfunction in adulthood 21 years after the assessment of lipid profile.10
However, endothelial dysfunction is described not only in adults with cardiovascular risk factors but also in children with familial hypercholesterolaemia,11 familial combined hyperlipidaemia,12 type 1 diabetes13 and obesity.14 Recently, Hopkins et al. determined FMD in 129 children and found an inverse relationship between percentage body fat as well as physical activity and FMD. In this study, lack of high-intensity physical activity was predictive of impaired FMD.15
Intima-media Thickness
An increased intima-media thickness (IMT) of large- and medium-sized arteries, such as the aorta, carotid and iliac arteries, is considered to be the first anatomical change in atherosclerosis. It follows endothelial dysfunction via deposition of lipids, infiltration of inflammatory cells and proliferation of fibroblasts. IMT can be measured easily and reproducibly by B-mode high-resolution ultrasound of the common carotid artery. Similar to endothelial dysfunction, increased IMT is an independent risk factor for cardiovascular events.16
There have been a number of large cross-sectional studies in children in Europe and North America, including the above-mentioned Cardiovascular Risk in Young Finns Study, the Bogalusa Heart Study and the Muscatine trial. These studies evaluated the association between classic cardiovascular risk factors in children and the occurrence of subclinical and clinical manifestations of atherosclerosis in adulthood in roughly 10,000 individuals. They all found an increase in IMT in adults who had risk factors such as dyslipidaemia, elevated body mass index (BMI), increased blood pressure and cigarette smoking in childhood, independent of contemporaneous risk factors.17–19 Pathologically altered IMT is already evident in children with cardiovascular risk factors.13,14,20
Arterial Stiffness
Another player in the pathophysiology of the cardiovascular system is stiffness of the arterial wall leading to a reduced vascular cushioning (Windkessel) effect. The elasticity of large vessels during systole and the recoil of the vessel wall during diastole causes continuous blood flow in the capillaries, reduces left ventricular workload and enhances diastolic myocardial perfusion. Vascular remodelling is associated with increased pulse pressure, enhanced pulse-wave velocity and enhanced augmentation index, representing an early reflection of the pulse wave from the periphery during systole.21
There is no gold standard for evaluating arterial stiffness – different methods and devices are used. They are all based on pulse transit time, pulse contour analysis and direct measurement of arterial geometry.22 Nevertheless, pulse pressure and pulse-wave velocity have been found to be predictive for cardiovascular events in adults.23,24
The Cardiovascular Risk Factors in Young Finns Study and the Bogalusa Heart Study revealed an impaired vascular elasticity in adults who developed their risk factor profile in childhood.25,26 In children, obesity, body fat and physical activity, as well as exposure to tobacco smoke, have been shown to be associated with increased arterial stiffness.27,28
Risk Factors for Cardiovascular Disease in Children
During the last 50 years arterial hypertension, dyslipidaemia, diabetes, smoking, obesity, low levels of physical activity and low-grade systemic inflammation have been identified as independent risk factors for atherosclerotic diseases. All of these conditions can be observed in children and adolescents and have been found to be associated with earlier occurrence of the above-mentioned vascular abnormalities.
In a large cohort of nearly 280,000 Danish schoolchildren, a higher BMI was strongly associated with an increased risk of coronary heart disease in adulthood, and the risk increased across the entire BMI distribution.29 In 1987, Smoak et al. reported a clustering of cardiovascular risk factors in children and young adults with obesity. In 3,503 subjects of the Bogalusa Heart Study five to 24 years of age, the relative risk of dyslipidaemia, elevated blood pressure and fasting glucose was 3.1 in obese compared with 0.4 in lean subjects.30
In a large cross-sectional study of about 2,500 children, Reich et al. identified a mild but continuous increase in the prevalence of hypertensive blood pressure levels that increased sharply when children became overweight.31 In a large cohort of overweight and obese European children, 70% of children already had at least one additional cardiovascular risk factor, mostly arterial hypertension.32
In 1988 Reaven introduced the concept of the metabolic syndrome, characterised by the simultaneous occurrence of hyperinsulinaemia with several other cardiovascular risk factors in the same patient. This clustering resulted in a markedly higher cardiovascular morbidity.33 Twenty years later, in a joint statement of the American Diabetes Association and the European Association for the Study of Diabetes, the clarity and accuracy of the existing definition of the metabolic syndrome was questioned.34 Some criteria used are ambiguous or incomplete and it has not been proved that the predictive value of the ‘syndrome’ over that of single components themselves is actually higher. Studies evaluating the metabolic syndrome uniformly show that it is seen in the paediatric population, with a prevalence of up to 50% in obese and overweight children.35–38
The above data demonstrate a pivotal role of obesity in the pathogenesis of atherosclerosis in children.
Childhood Obesity
Definition
The nadir of the association curve between BMI and all-cause mortality in adults was found to be 22.0–24.9. Mortality increases with BMI.39 Therefore, in adults a BMI >30 is uniformly classified as obesity, whereas a BMI between 25 and 29.9 is considered to be overweight. By contrast, the definition of obesity in childhood is surprisingly difficult since the degree of body fat mass depends on ethnic background, gender, developmental stage and age. In children, increased BMI was also found to be associated with increased risk of mortality from cardiovascular disease.29 Since BMI is easy to obtain and correlates sufficiently with direct measures of fatness, in clinical practice BMI is adequate for the diagnosis of childhood obesity. However, waist and hip circumference, skin-fold thickness and direct measurements of body fat, e.g. hydrodensitometry, bioimpedance or dual-energy X-ray absortiometry (DEXA), are useful tools in scientific studies.40
BMI in children substantially changes with age, such that reference percentiles have to be used to estimate the weight of a child in comparison with the distribution in a population. Accordingly, a child with a BMI above the 97th percentile with regard to age and gender is considered to be obese, and a child with a BMI between the 90th and the 97th percentile would be considered to be overweight (European Task Force for Childhood Obesity).41 In order to establish an international standard definition of overweight and obesity in childhood, Cole et al. published BMI cut-off points for children from two to 18 years of age corresponding to BMI values of 25 and 30 at 18 years of age based on international data. These cut-off points allow international comparisons of the prevalence of obesity and overweight, despite different reference populations used locally for percentile curves.41
Prevalence
Obesity and overweight are among the most important health problems in the industrialised world and are becoming an ever more important issue in developing countries. The prevalence of childhood obesity has reached alarming levels and epidemic proportions. It is estimated that worldwide about 1 billion people are overweight (with 22 million of these under five years of age) and 300 million are obese. In parts of Europe, up to 35% of children were reported to be overweight, with increasing prevalence rates year on year.42,43 By 2010 it is estimated that 26 million children in EU countries will be overweight, including 6.4 million who will be obese. The highest prevalence of childhood obesity was found in the US, with more than 35% of children being overweight and 13% being obese in 2003–2004.44
In response to the increasing prevalence of childhood obesity, the percentile curves used to quantify obesity in childhood have had to be readjusted two times within the last two decades. Comparing BMI percentiles 15 years apart, there was a clear rise in the 97th percentile over time, whereas the third and 50th percentiles remained stable over the same period. This indicates not only that more children have become obese, but, in addition, that the degree of obesity has increased.45
Causes
Twin studies suggest that approximately 50% of the tendency towards obesity is inherited.46 However, a rapid change in human genetics over recent decades causing the obesity epidemic can be excluded. In fact, the reasons for the increase in childhood obesity include environmental factors, lifestyle preferences and also cultural background. Major predictors of childhood obesity are paternal obesity and a low socioeconomic status.31,47 An increase in calorie and fat intake in our affluent society as one of the major causes for developing overweight and obesity is currently controversial.
Briefel and Johnson in an analysis of the dietary intake of the US population between 1971 and 2000 demonstrated that the mean daily energy intake remained unchanged in children and male adolescents over time but increased in female adolescents. During the same period of time the percentage of overweight children and adolescents steadily went up irrespective of gender.46 This phenomenon can only be explained by a mismatch of the net energy balance due to a marked decline in physical activity and energy expenditure.
Kimm et al. used a validated questionnaire to measure leisure-time physical activity on the basis of metabolic equivalents (METs) for reported activities and their frequency in MET times per week to evaluate the physical activity of 2,300 black and white girls. By year 10 of the study they found a 100% decline in leisure-time physical activity for black girls and a 64% decline for white girls. The decline was associated with an increase in BMI.48
Instead of total amount of physical activity, high levels of vigorous physical activity seem to be beneficial in the prevention of obesity. Ortega et al. reported that a low level of vigorous physical activity (>6 MET) is independently associated with increased waist circumference in a large number of Swedish children and adolescents. In this study, children and adolescents in the lowest tertile of vigorous activity had four-fold higher odds of being overweight. This observation remained significant after controlling for sedentary activities.49 Dencker et al. reported that vigorous physical activity, but not moderate activity, is a significant correlate of abdominal adiposity measured by DEXA scan.50
Adipokines
Adipose tissue can be regarded as an endocrine organ that produces a variety of messenger molecules, called adipokines. For example, leptin interacts with hypothalamic receptors that control body temperature and energy expenditure, hunger and eating behaviour. In peripheral tissues leptin enhances glucose metabolism. In overweight individuals plasma leptin concentrations are elevated, but a relative leptin deficiency induced by leptin resistance paradoxically produces orexigenic adiposity signals that engender a state of hyperphagia and decreased energy expenditure. Hyperleptinaemia contributes to ectopic lipid deposition, a proinflammatory milieu, endothelial dysfunction and insulin resistance. By contrast, plasma levels of adiponectin that is derived from adipose tissue are reduced in obesity. Adiponectin appears to prevent atherogenesis in humans via endothelial production of NO and has anti-inflammatory as well as insulin-sensitising effects.40,51 Overweight children show both elevated leptin and reduced adiponectin plasma levels.52
Co-morbidities
Obesity is a key player in the development of metabolic syndrome and leads to elevated triglyceride and low-density lipoprotein (LDL) cholesterol levels, decreased high-density lipoprotein (HDL) cholesterol levels, insulin resistance and type 2 diabetes, hypertension, non-alcoholic fatty liver disease and sleep apnoea. Additionally, obesity is associated with orthopaedic disorders and back pain, leading to a reduced physical activity. This vicious circle is amplified by social marginalisation and eating disorders.
Strategies of Risk Reduction
Secondary Prevention
In obesity intervention programmes with calorie and fat restriction, health education, behaviour interventions and physical activity are used to achieve a reduction in bodyweight over time. Most of these studies used a combination of family, school and community-based interventions with education, diet and exercise. Overall there is strong evidence for a positive effect of intervention programmes on BMI in the obese.
A reduction in BMI is associated with a decline in systolic and diastolic blood pressure, reduction of triglyceride and LDL cholesterol levels, increase in HDL cholesterol levels and reduction of fasting glucose and insulin resistance. It is also associated with a reduction of leptin and C-reactive protein levels53–56 and an increase in adiponectin levels.52 Weight reduction in overweight and obese children via diet and exercise programmes can reverse early atherosclerotic changes and lead to an improvement of FMD and a reduction of IMT and arterial stiffness57,58 (see Figure 1).
Pharmacological treatment has to be considered in obese adolescents if intensive lifestyle modification has failed and in overweight adolescents with severe co-morbidities despite lifestyle modification. Three drugs have been tested in randomised, placebo-controlled studies in children: orlistat, sibutramine and metformin. Orlistat is a pancreatic lipase blocker that affects absorbtion from the intestine. In 539 American adolescents, orlistat in addition to a hypocaloric diet plus exercise and behavioural therapy resulted in a significant decrease in BMI compared with placebo. Gastrointestinal side effects, such as fatty stools, were common with treatment.59 Sibutramine, a noradrenalin and serotonin re-uptake inhibitor that promotes satiety and enhances energy expenditure, was shown to significantly reduce bodyweight in combination with behaviour therapy compared with placebo.60 Adverse events reported included tachycardia and, less frequently, headache, dry mouth, constipation, dizziness, insomnia and elevated blood pressure. A four-month intervention with metformin, an insulin sensitiser, in 28 obese adolescents with normal glucose tolerance resulted in a decreased BMI and a reduction in subcutaneous fat and serum leptin levels compared with placebo.61 In children with familial hypercholesterolaemia, several studies documented that statins can be used to effectively reduce LDL cholesterol levels. This is associated with a restoration of FMD and a regression of IMT.62,63 In children and adolescents with arterial hypertension, a reduction of systolic and diastolic blood pressure can be achieved with angiotensin-converting enzyme (ACE) inhibitors. However, black children do not respond as well as white children to ACE inhibitors.64 Overall, randomised controlled trials evaluating the efficacy and safety of pharmacotherapy to reduce cardiovascular risk factors in children are rare, especially in long-term interventions.
Bariatric surgery as a treatment option for extremely obese adults is well established, since a significant reduction in bodyweight without substantial rebound effect can be achieved. This is associated with an improvement in co-morbidities. Data in the paediatric population are scarce. Case reports and small series document that bariatric surgery can be helpful in weight reduction in adolescents. This is associated with improvements in lipid and glucose metabolism after surgery.65 Based on expert opinion, the Endocrine Society’s Clinical Practice Guidelines suggest bariatric surgery for adolescents with a BMI above 50 or a BMI above 40 in patients with severe co-morbidities in whom lifestyle modification and/or pharmacotherapy has failed.66 Nonetheless, non-surgical options should be tried before the use of weight loss surgery in adolescents, and the risk–benefit profile should be considered seriously.67
Primary Prevention
Since being overweight is the key player in the development of a cardiovascular risk profile in childhood, over 100 studies have been conducted to prevent obesity in children and adolescents. In some of them, educational, dietary or physical activity interventions alone were used to affect the net energy balance. Most trials intervene with a multicomponent approach, combining health education of children and their parents, dietary recommendations and additional physical activity programmes at school or in the community. The results are summarised in recent review articles and a meta-analysis. However, these results are conflicting and are described below.
In 2001 Steinbeck reported encouraging results of the first school-based intervention programmes and called for the support and involvement of community sectors other than health in order to prevent obesity by increasing physical activity.68
In 2005 a Cochrane analysis by Summerbell and co-workers reviewed 22 randomised controlled trials with a minimum duration of 12 weeks that addressed dietary education, physical activity and BMI. Nearly all of the 12 short-term (<12 months) and 10 long-term (>12 months) interventions resulted in some improvement in diet or physical activity. However, none of the interventions combining physical activity and diet significantly affected BMI.
In three trials using physical activity exclusively, a small but positive impact on BMI was achieved. The heterogeneity in terms of study design, quality, target population and outcome measures made a direct comparison of study findings difficult. Taken together there is no convincing evidence that the school-based intervention programmes analysed helped to prevent childhood obesity. The authors recommended that the design, duration and intensity of interventions to prevent obesity in childhood be reconsidered.69
Inconsistently, Doak et al. found that 17 of 25 school-based overweight prevention programmes were ‘effective’ based on a statistically significant reduction in BMI or skin-fold thickness for the intervention group. Four study interventions were found to be effective in reducing both BMI and skin-fold measures. Of these, two targeted reductions in television viewing. The remaining two studies targeted direct physical activity intervention through a physical education programme combined with nutrition education.70
A large review article was published in 2007 by van Slujis et al. focusing on the effectiveness of interventions to promote physical activity separately in children under 12 years of age and adolescents 12–18 years of age. There was strong evidence that school-based interventions with the involvement of family or community and multicomponent interventions can increase physical activity in adolescents. Studies on children alone (33 of 57 studies) found no evidence for a positive effect of intervention programmes on physical activity.71
Recently, Harris et al. performed a meta-analysis of 18 randomised controlled studies involving 18,000 children to evaluate the effect of school-based physical activity interventions on BMI. No improvement was observed with physical activity intervention on BMI or other body composition measures. Harris concluded that increased physical activity in schools is unlikely to have a significant impact on the prevalence of childhood obesity.72
Most of the studies reviewed are limited by indirect measures of physical activity and cardiopulmonary fitness using questionnaires and self-reports, which are often not validated. Furthermore, in several studies individual adherence to the intervention programme was low or not reported. Assuming that overweight children especially would benefit most from such interventions, and conversely children with higher baseline BMI may have lower levels of adherence, consequently they do not benefit.
In a predominantly healthy population it is difficult to achieve significant changes in physical activity or BMI via low-grade intervention programmes. Recently, Walther et al. randomised seven sixth-grade classes (182 children) to an intervention group (four classes with 109 students) with daily school exercise lessons for one year and a control group (three classes with 73 students) with regular school sports twice weekly. This study design warranted nearly 100% adherence to study intervention. They failed to show a significant reduction in BMI in the overall population of predominantly lean children but found an increase in cardiopulmonary fitness measured by cardiopulmonary exercise testing and an increase in the amount of circulating progenitor cells in the intervention group. Thereby they demonstrated that additional exercise lessons at school reduce cardiovascular risk even if BMI remains unchanged.73
The ongoing STRIP (Special Turku Coronary Risk Factor Intervention Project) is evaluating the effects of biannual dietary counselling given to healthy Finnish children, starting at seven months of age, and their parents Participants were randomly assigned to an intervention or a control group. Families of the intervention group received individualised dietary counselling, given by a team consisting of a physician and a dietitian who suggested appropriate dietary changes for a healthy low-saturated fat, low-cholesterol diet. Children in the control group did not receive detailed dietary counselling. After 10 years the prevalence of overweight in girls but not in boys was lower in the intervention compared with the control group.74 Boys but not girls from the intervention group had significantly better FMD.75 Insulin sensitivity in the intervention group children was better than in the control group.76
Conclusion
Subclinical vascular damage can already be found in children and is clearly associated with an accumulation of risk factors. Childhood obesity is a key player in the pathogenesis of atherosclerosis, since obesity causes metabolic, immunological and haemodynamic alterations. Pre-eminent factors for the increasing prevalence of overweight and obesity to alarming levels can be found in the net energy balance. A calorie intake that is too high combined with too little energy expenditure causes weight gain. Therapeutic approaches to correct the energy balance by dietary restriction and/or exercise are effective in overweight and obese children, and lead to weight loss and an improvement in metabolic and haemodynamic status.
Pharmacotherapy and bariatric surgery remain treatment options reserved for severely obese adolescents with co-morbidities and/or after failure of less invasive treatment strategies. So far, routine use of drugs or even surgical procedures to reduce cardiovascular risk in children and adolescents cannot be recommended. Data on long-term results, risks and safety profiles in large randomised controlled trials are lacking.
The evidence for dietary recommendations and exercise in primary prevention is not as strong as in secondary prevention. Although intervention trials report some positive effects on risk factors, overall the results of primary preventive interventions are conflicting. It is extremely challenging to demonstrate the effects of primary prevention programmes on cardiovascular risk factors in already healthy individuals within a few years. Therefore, large long-term trials and registries are needed to document the efficacy of primary prevention.
Medical societies and companies make a huge effort to treat patients with acute cardiovascular events. With regard to children’s health and quality of life, at least the same attention needs to be paid to preventative strategies to avoid the formation of a risk profile and subsequent atherosclerotic disease.