A New Paradigm in Cardiovascular Disease Risk Reduction: Childhood Metabolic Syndrome

The last decade has seen a rapid growth in interest in pediatric atherosclerosis prevention. Recent data have shown increased pediatric prescribing of antihypertensives, antihyperlipidemics, and anti-diabetes drugs in the United States.[1] This heightened awareness has provided the impetus to define and study the metabolic syndrome (MetS) in children. The MetS describes a cluster of cardiovascular risk factors, including insulin resistance/glucose intolerance, obesity (notably abdominal), dyslipidemia, and hypertension, as well as inflammation. While this syndrome has been clearly recognized in adults, its significance in children is less obvious and its definition, prognosis, and management are controversial. With the burgeoning epidemic of childhood obesity in this country and its repercussions, including an alarmingly increased incidence of type 2 diabetes mellitus (T2DM) in young patients, the various components that comprise the MetS are more evident and of greater concern. However, the limited longitudinal data on tracking from childhood through adulthood make it difficult to know who is at long-term risk. What relation this cluster of risk factors in children has to future arteriosclerotic cardiovascular disease (ASCVD) is still to be determined, but the possibility of synergism is an important testable hypothesis.

The scientific statement on the progress and challenges in the MetS in children and adolescents by Steinberger and colleagues is a comprehensive review of the literature surrounding the syndrome in this patient population.[2] Additionally, we are provided with a critical evaluation of the construct of the MetS in pediatrics, as opposed to a mere breakdown into its components with their attendant cardiovascular and metabolic risk over time. Steinberg et al. provide an important and valuable update and synthesis on the advances in our understanding since this topic was last reviewed.[3]

Utilizing studies to date, Steinberger et al. report that the prevalence of the MetS in youth is quite variable, ranging from about 4% to upwards of 30%, and even higher in severely obese populations. Part of this variability is due to the many obstacles of defining the MetS in the pediatric population. Notably, while for adults one can use fixed cut-off points to define abnormal, in children, risk factor distributions vary with age, gender, and physiologic development, and percentile graphs or tables are used as reference data. Furthermore, it is not clear whether we should be using the 90th (or 10th) percentile as opposed to the 75th (or 25th) percentile for the individual risk factors. Obviously, the latter would detect more cases, which might be beneficial from a clinical standpoint, as more individuals would be eligible for preventive efforts. However, this could be excessive, as well as potentially stigmatizing. Additionally, lipid threshold values have not been modified by the National Cholesterol Education Program (NCEP) since 1991. Another issue is the lack of consensus on measures of waist circumference (WC). While central adiposity is coming to be accepted as the main anthropometric indicator of cardiometabolic risk, and WC is more specific than body mass index (BMI) percentile, only recently did one study devise a set of WC percentiles for children of various ethnicities.[4] These have not been applied in epidemiologic studies or clinical trials of the MetS to date.

It remains debatable whether insulin resistance or obesity is the principal underlying cause of the MetS in adults. Obesity seems to be strongly associated, although not absolutely necessary, for the development of insulin resistance, and risk for insulin resistance is heightened with increasing degrees of obesity. Conversely, hyperinsulinemia is known to contribute to weight gain in specific situations (e.g., infants of diabetic mothers, insulin therapy initiation in diabetics) and it generally seems to exacerbate visceral adiposity. Several studies in children, notably the Cardiovascular Risk in Young Finns Study [5], as well as those done in the Pima Indian population [6], indicate that hyperinsulinemia precedes the development of other risk factors, including obesity, hypertension, and hypertriglyceridemia. There are some impediments to the study of insulin resistance in youth, namely lack of consensus for serum insulin norms and the well-documented physiologic insulin resistance of puberty. Despite ongoing controversy in this area, one recent study has shown that insulin resistance as measured by fasting insulin is associated with failure to respond to therapeutic lifestyle change (TLC) treatment in obese adolescents.[7] Recent data from the NHANES III showed a 7% prevalence of impaired fasting glucose, which was higher among the overweight and Mexican-Americans and was associated with significantly higher fasting insulin as well as dyslipidemia and hypertension.[8] As more studies of this phenomenon emerge, we will be able to more rationally diagnose insulin resistance and/or the MetS during this developmental stage.

It is widely accepted that there is a strong correlation between obesity and insulin resistance, T2DM, and ASCVD. However, obesity does not universally lead to insulin resistance and/or the MetS. As was already mentioned, prevalence of the MetS in an overweight population of children was approximately 30%. What about the other 70%? Some are clearly not at risk, while others are: what is the proportion and how should we identify “pre-MetS”? These are questions that have yet to be investigated. Insulin resistance may be the mediator by which adiposity negatively impacts serum lipids and blood pressure. Studies seem to suggest that the distribution of body fat (i.e., visceral vs. peripheral fat) may be crucial in manifestation of risk, even in childhood. Despite this, WC is not currently being widely used as a clinical or research measure of adiposity.

Adipocytokines, inflammatory mediators, signs of oxidative stress, and alterations in cortisol are becoming active areas of investigation in children and adolescents as summarized in the current statement. They may soon give us important clues useful in understanding the etiology of the MetS, as well as better targets for screening and treatment. Improved measures of vascular structure and function as well as in vivo measures of atherosclerosis extent, progression, and regression will also be important to advance the field.[9]

Increases in blood pressure appear to be related more to the constellation of risk factors, rather than to each individual component when considered separately. Abnormal lipid profiles, notably high triglycerides, low-density lipoprotein (LDL)-cholesterol, and total cholesterol, along with low levels of high-density lipoprotein (HDL)-cholesterol, have been linked to obesity as well as insulin resistance in children and adolescents. However, it is unknown whether the relationship is causative, or whether dyslipidemia and insulin resistance are connected by a third factor, such as obesity.

The obesity epidemic that we are witnessing in this country’s youth has brought with it a younger generation of patients being diagnosed with T2DM as well. However, it must be noted that not all children with glucose intolerance progress to frank diabetes, nor do all overweight children and adolescents develop impaired glucose tolerance. Nonetheless, the prevalence of T2DM is more than two times that of type 1 diabetes mellitus (T1DM) in adolescents, and most of those diagnosed are overweight. In patients for whom disease commences early, there is an even greater risk for accelerated atherosclerosis than in those who acquire T2DM as adults. The MetS is also associated with other diseases, including hyperandrogenism in females [polycystic ovary syndrome (PCOS)]. The insulin resistance seen in this syndrome may indeed be caused by the same cellular defect that causes the increased ovarian and adrenal androgen production.

Why do some people develop the MetS and others do not? Genetics and environment both play a role. A family history of parental ASCVD is known to increase cardiovascular (CV) risk in children and is already used for screening and prevention. Family studies have demonstrated clustering of MetS components. Obesity and T2DM are also familial in nature. Furthermore, racial/ethnic differences are seen with respect to the various individual criteria. African-Americans have less dyslipidemia than other racial/ethnic groups, but a higher prevalence of hypertension, irrespective of the degree of obesity. Insulin resistance seems to be more common in children of African-American and Hispanic background than in Caucasian children. Individuals of Asian descent seem to be at increased CV risk at lower BMI and WC levels. These differences also make it harder to use one set of norms to define the MetS, and it has been suggested to develop and use criteria that are ethnically specific in assessment. Lifestyle behaviors, such as television watching habits, physical activity, and dietary intake, are associated with overweight, insulin resistance, and inflammatory mediators. While these are environmental in nature, families may also share these characteristics, making it somewhat difficult to separate genetics and environment.

Although the certainty of the MetS construct is still in question, it is apparent that the various risk factors start clustering in childhood and that individually they are becoming more and more prevalent in youth. TLC remains the primary treatment for the MetS and its components, in both adults and children. TLC focuses on healthy eating, reduction in sedentary activities, and increase in regular physical activity. Healthy eating may involve discovering the optimum energy balance, which can vary depending on the stage of growth and development. Additionally, specific macronutrients and types of diet may prove to be more effective (e.g., low glycemic index diet, Mediterranean-style diet, high-fiber diet). Although weight loss may be the primary goal, TLC alone can improve the CV risk profile, as there may be changes in body composition, as well as lipid profile, blood pressure, and carbohydrate metabolism, irrespective of degree of weight loss.[10] We found that in a sample of 16 adolescents with PCOS, modest weight loss (6% on average) following dietary and exercise counseling was associated with a significant decrease in WC and a rapid and significant improvement in menstrual function. These changes are best explained by a reduction in insulin resistance.[11]

The statement by Steinberger et al. provides a stimulus to pay more attention clinically to children who may be at risk for the various components of the MetS, both by family history as well as clinical findings. This statement also proposes areas for future research, including large-scale outcome studies, molecular etiologies, genetic/prenatal disposition, and racial/ethnic diversity and disparities in mechanism of clustering. Additionally, it seems important to determine what the role of medical management (e.g., use of metformin vs. TLC) is in the regression or progression of the individual risk factors as well as identification of the syndrome as a whole. From some of our own preliminary data in a clinical sample of adolescents, we know that whether we used stringent criteria vs. more flexible criteria to define the MetS, one-third improved to the point of not meeting criteria over a 9-month treatment period with medical-nutrition therapy.

The importance of the current statement is that it points out the major questions to be asked by researchers and clinicians, as well as policy makers. What key definitions lack consensus? What is the significance of the clustering of risk factors? What interventions are most effective and to whom should they be applied? Should we only be concentrating on the “worst of the worst” or, rather, target a larger population of subclinical cases? A more fundamental question is does the MetS actually exist as an aggregation in this younger population or should we only be focusing our prevention and treatment efforts on the individual risk factors (e.g., obesity, insulin resistance)? Does classifying children by the presence or absence of this syndrome change our approach to CV risk prevention and research? If one risk factor is present, it would be practical to screen for all risk factors, as we know they may cluster together, even at an earlier age. The accumulation of additional risk factors in an individual or population should serve as a red flag, as the risk factors are likely synergistic. Perhaps the MetS could function as shorthand for coding and tracking, in both the clinical and research arenas. Regardless, the MetS certainly cannot be ignored as a predictor of CV risk.

While we await the needed consensus based on better data, it seems prudent to develop preventive policies that ensure all children have opportunities for regular physical activity of a variety of types, combined with healthy choices of nutrient-dense foods based on the best current scientific evidence. Improved and extended availability of physical education for all children, decreased exposure to television and food advertisements for children, limited availability of sugar-sweetened beverages, local sourcing of fruits and vegetables, and calorie-labeling of restaurant foods are all small steps that deserve study and implementation.