Background
The prevalence of childhood obesity has dramatically increased over the last decades [
1,
2]. This has prompted the need for obesity stratification to identify those at increased cardiometabolic risk. Substantial data have accumulated on the association of childhood obesity with cardiometabolic risk, though obesity was usually considered as a single entity [
3‐
8]. Extreme values of body mass index (BMI), categorized as severe obesity, were once a phenomenon of adults. However, the problem is escalating in the pediatric population. According to a recent report, in many of 21 European countries, 1 in 4 children with obesity at school age have severe obesity [
9]. Furthermore, 2.1% of children between ages 2 to 5 years from the US National Health and Nutrition Examination Surveys (NHANES) [
10] and 1.3% of Canadians of this age group [
11]; and 1.5% of toddlers (2–3 year-old) from the US National Institute (NIH) of Health Environmental Influences on Child Health Outcomes (ECHO) program [
12] and 0.3% of a cohort of Canadian children aged 17 to 24 months [
13] reportedly had severe obesity. Since severe obesity in childhood has become common only in the last decades, the relative contribution of this phenomenon to disease burden has only recently been approached. Identifying those with substantial risk for severe obesity and its deleterious cardiovascular sequelae is of great clinical and public health importance. This review aims to summarize (i) the association between severe obesity and the prevalence of cardiovascular risk factors, diabetes and cancer in cross-sectional studies; (ii) the relationship between severe obesity and incident cardiometabolic, cancer and mortality in longitudinal studies. We aimed to highlight gaps in evidence in order to direct future research.
Discussion
During the last two decades, the prevalence of severe obesity in children has increased by twofold both in the US [
93‐
95], and in other countries [
2,
96]. In this systematic review of the association of childhood severe obesity with immediate cardiometabolic risk factors, we focused on the risk of severe compared to mild obesity. The data show that children with severe obesity are at greater risk for dyslipidemia, hypertension, type 2 diabetes and fatty liver disease than children with mild obesity. However, the long-term risk and actual point estimates in adulthood are lacking.
Several studies have reported associations of severe obesity in children with increased prevalence of cardiovascular risk factors. Data of children and adolescents from the US NHANES showed greater prevalence of abnormal TGs, diastolic blood pressure and glycated haemoglobin levels in class 3 than class 2 obesity [
28]. In addition, significantly higher prevalences of abnormal HDL, systolic blood pressure and glycated hemoglobin levels were observed in class 2 than in mild obesity. The magnitude of associations of severe obesity with cardiovascular risk factors in children differs by the specific risk factor [
38,
47,
48,
51]. Notably, the latent period for developing significant cardiometabolic morbidity may require several decades [
24,
25]. Therefore, the current data evidently does not reflect the actual morbidity, given the sharp increase in severe adolescent obesity in some Western countries over the recent period [
2,
93,
94].
Type 2 diabetes risk is predominantly affected by obesity class. Higher BMI has been found to be associated with younger age at diabetes diagnosis [
97,
98]. Young onset diabetes has a greater productivity burden and was associated with higher rates of cardiovascular morbidity and mortality [
99‐
101]. Thus, the relative contribution of childhood severe obesity to type 2 diabetes is of high clinical and public health importance. A 1.7-fold (95% CI 1.3–2.2) increased risk for dysglycemia (in the prediabetes or diabetes range) was reported among children with class 2 obesity compared to those with mild obesity [
28]. Numerous studies have reported a strong relation between obesity in youth and subsequent type 2 diabetes in adulthood, but none compared outcomes among those with mild and severe obesity [
31,
102‐
110].
Sociodemographic variables are potential confounders of the association between degree of childhood obesity and cardiometabolic risk. Although some analyses were adjusted for sex and ethnicity, most studies did not adjust the observed risk for socioeconomic confounders that have been shown to be closely associated with both severe obesity and cardiometabolic morbidity [
111]. However, in a cross-sectional study of pre-recruitment adolescents, the odds ratios for type 2 diabetes and for hypertension among individuals with mild compared to severe obesity were not materially affected by adjustment for residential socioeconomic status based on locality of residence and for education level assessed by years of formal schooling [
2]. A recent Australian study exemplifies the complexity of sociodemographic adjustment [
112]. Among children with severe obesity, average neighborhood education/occupation and family education level were negatively associated with BMI, waist circumference and body fat percentage, but not with cardiometabolic risk factors. Neighborhood walkability in that study was related to lower waist circumference. However, better access to basic shopping facilities including playgrounds and parks was also related to higher prevalences of dyslipidemia and fatty liver. A study from New Zealand revealed that severe obesity was more common among adolescents living in areas of high deprivation [
113]. Household dysfunction has been associated with severe obesity. Furthermore, traumatic life experiences, such as physical and sexual abuse during childhood and adolescence, have been far more common in adults with severe obesity [
114] and also in children [
115].
Children with early-onset severe obesity display more pronounced obesogenic behaviours than do their peers with overweight or mild obesity [
116]. Among the factors that have been found to be associated with severe obesity among young children are: lower consumption of fruits and vegetables, higher consumption of fast food, less outdoor play, shorter sleep duration, lack of bedtime rules, increased screen time and less involvement in team sports [
117,
118]. Thirty percent of adolescents with severe obesity were current cigarette users, compared with 14% among those with healthy weight students [
113]. These data on modifiable cardiometabolic risk factors were not considered by most of the studies included in the current review. In 2016, the US NIH initiated the ECHO program in an attempt to address environmental origins of childhood obesity [
12]. Data generated from this initiative should assist in setting the stage for intervention studies aimed to lessen the burden of childhood obesity. Clinicians should promote healthier diet, physical activity, and avoidance of smoking, particularly among adolescents with obesity [
119], to mitigate cardiometabolic risk.
The above data suggest that cardiovascular disease and risk factors are intensified in severe compared to mild obesity, and appear at earlier ages. The existence of morbidity related to severe obesity already in childhood or youth is especially detrimental since it affects this young population in the most productive years of adulthood. Two pieces of data should be considered in this instance. First, severe obesity has increased among US children, even as young as 2–5 years old [
1]. Second, youth and young adults may be less aware of cardiometabolic morbidities such as type 2 diabetes than older adults [
120]. The upshot is delayed diagnosis and worse clinical course. This is exemplified by the existence of complications already at diagnosis and the harmful course of the disease [
78]. Notably, mortality among individuals with diabetes was shown to decline in all age groups except young adults (ages 20 to 44 years) [
121]. Estimated diabetes costs in the US in 2017 were $327 billion, which includes $237 billion in direct medical costs and $90 billion in lost productivity. Annual per capita health care expenditure is 2.3 times higher for people with than without diabetes. A large portion of medical costs associated with diabetes costs is for comorbidities [
122], which were much more frequently reported in young individuals with diabetes [
123]. Two recent trends are of particular concern: the decreasing age for the onset of severe obesity among preschool children and the decreasing age for the onset of type 2 diabetes in youth. These, together with their deleterious clinical courses, likely portend a pronounced increase in economic burden over the ensuing years.
This systematic review has a number of limitations. As noted above, studies that have addressed the risk for cardiometabolic complications in children and adolescents have used highly variable definitions of severe obesity; less than half of them used BMI > 120% of the 95th percentile, which is recommended. Since this definition has a discriminatory advantage in identifying children with severe obesity who are at increased cardiometabolic risk, point estimates may be underestimated [
22]. Furthermore, point estimates are susceptible to age and sex bias, although this may be mitigated by adjustment for these variables. This caveat should be particularly emphasized in longitudinal studies that used definitions that were shown to have weaker correlations with sequential obesity measurements [
21]. Second, our systematic search was based on an explicit definition of severe or morbid obesity. Therefore, studies that did not use this terminology but included data regarding higher classes of obesity may have been missed. We expect that the number of such studies is low, and we acknowledged some of them in this review. Third, the assessment of sex-based differences is limited, given the low disproportionate representation of females and the minority of the studies that stratified the analyses by sex. Despite the above limitations, important observations are evident regarding the burden of comorbidities in children with severe obesity at an early age.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.