Global, regional, and country estimates of metabolic syndrome burden in children and adolescents in 2020: a systematic review and modelling analysis

Background: Halting the rise in cardiometabolic risk factors in children and adolescents is crucial to curb the global burden of cardiovascular diseases. We aim to provide global, regional, and national estimates of the prevalence of metabolic syndrome in children and adolescents to support the development of evidence-based prevention strategies. Methods: In this systematic review with modelling analysis, we searched PubMed, Embase, Africa Journal Online, and Global Index Medicus from database inception to Jan 30, 2021, with no restriction on language or geographical location. We included community-based and school-based cross-sectional studies and cross-sectional analysis of cohort studies that reported prevalence of metabolic syndrome in the general population of children (6–12 years) and adolescents (13–18 years). Only studies with a low risk of bias were considered. Eligible studies included at least 200 participants and used probabilistic-based sampling. Diagnosis of metabolic syndrome had to meet at least three of the following criteria: high systolic or diastolic blood pressure (≥90th percentile for age, sex, and height); waist circumference in at least the 90th percentile for age, sex, and ethnic group; fasting plasma glucose 5·6 mmol/L or greater; fasting plasma triglycerides 1·24 mmol/L or greater; and fasting plasma high density lipoprotein cholesterol 1·03 mmol/L or less. Independent investigators selected eligible studies and extracted relevant data. The primary outcome was a crude estimate of metabolic syndrome prevalence, assessed using a Bayesian hierarchical model. Findings: Our search yielded 6808 items, of which 169 studies were eligible for analysis, including 306 prevalence datapoints, with 550 405 children and adolescents from 44 countries in 13 regions. The between-study variance (τ²) was 0·52 (95% CI 0·42–0·67), which could reflect the measurement of each component of the metabolic syndrome and covariates as sources of between-study heterogeneity. We estimated the global prevalence of metabolic syndrome in 2020 at 2·8% (95% uncertainty interval [UI] 1·4–6·7) for children and 4·8% (2·9–8·5) for adolescents, equating to around 25·8 (12·6–61·0) million children and 35·5 (21·3–63·0) million adolescents living with metabolic syndrome. In children, the prevalence of metabolic syndrome was 2·2% (95% UI 1·4–3·6) in high-income countries, 3·1% (2·5–4·3) in upper-middle-income countries, 2·6% (0·9–8·3) in lower-middle-income countries, and 3·5% (1·0–8·0) in low-income countries. In adolescents, the prevalence of metabolic syndrome was 5·5% (4·1–8·4) in high-income countries, 3·9% (3·1–5·4) in upper-middle-income countries, 4·5% (2·6–8·4) in lower-middle-income countries, and 7·0% (2·4–15·7) in low-income countries. Prevalence in children varied from 1·4% (0·6–3·1) in northwestern Europe to 8·2% (6·9–10·1) in Central Latin America. Prevalence for adolescents ranged from 2·9% (95% UI 2·6–3·3) in east Asia to 6·7% (5·9–8·3) in high-income English-speaking countries. The three countries with the highest prevalence estimates in children were Nicaragua (5·2%, 2·8–10·4), Iran (8·8%, 8·0–9·6), and Mexico (12·3%, 11·0–13·7); and the three countries with the highest prevalence estimates in adolescents were Iran (9·0%, 8·4–9·7), United Arab Emirates (9·8%, 8·5–10·3), and Spain (9·9%, 9·1–10·8). Interpretation: In 2020, about 3% of children and 5% of adolescents had metabolic syndrome, with some variation across countries and regions. The prevalence of metabolic syndrome was not consistently higher with increasing level of development, suggesting that the problem is not mainly driven by country wealth. The high number of children and adolescents living with metabolic syndrome globally highlights the urgent need for multisectoral interventions to reduce the global burden of metabolic syndrome and the conditions that lead to it, including childhood overweight and obesity. Funding: None.