Obesity is one of the major global health problems, affecting millions of individuals worldwide, with a particularly alarming prevalence among children and adolescents (1). Its primary causes include sedentary behavior, unhealthy dietary patterns, and insufficient physical activity (2). In paediatric populations, obesity is frequently associated with metabolic abnormalities such as insulin resistance, dyslipidaemia and hypertension, hallmark features of metabolic syndrome (MetS) (3). Cross-sectional studies have reported MetS prevalence rates ranging from 10 to 38% among children and adolescents with obesity (4, 5).

These metabolic disturbances are primarily driven by excessive fat mass (FM), particularly visceral adiposity (6). Visceral fat plays a key role in metabolic dysfunction through the secretion of adipokines involved in the pathogenesis of cardiometabolic diseases (7). Furthermore, increased FM promotes the accumulation of lipid intermediates, such as ceramides and diacylglycerols, within skeletal muscle (8). Combined with physical inactivity, this accumulation contributes to reduced mitochondrial density, enhanced adipogenesis, and decreased activity of key enzymes involved in aerobic energy production and fatty acid oxidation (9, 10). As a result, fat oxidation (FAT) at rest may be impaired (11, 12), and the ability to shift toward carbohydrate oxidation (CHO) in response to insulin stimulation is diminished (13). This impaired capacity to adjust substrate oxidation according to substrate availability, known as metabolic inflexibility, is typically characterised by an elevated respiratory exchange ratio (RER) at rest, reflecting a predominant reliance on carbohydrate metabolism (9). This phenomenon has been observed in individuals with metabolically unhealthy obesity (12). Indeed, resting substrate oxidation may provide additional predictive value for cardiometabolic risk, as it reflects both mitochondrial efficiency and the ability to utilise fat as an energy source in basal conditions (14). An elevated CHO oxidation rate at rest, may suggest impaired lipid oxidation and insulin resistance, both of which are implicated in the pathophysiology of MetS (15). Moreover, alterations in substrate utilisation may precede clinically evident metabolic dysfunction, offering an earlier window for risk identification (10). Nonetheless, findings remain inconsistent, particularly in paediatric populations with obesity, and this potentially valuable marker has been largely underexplored in paediatric risk models (16).

These metabolic alterations may also contribute to a reduction in fat-free mass (FFM) (16), leading to a lower basal metabolic rate (BMR), as FFM is the principal determinant of BMR (17). Over time, this unfavourable metabolic profile may increase susceptibility to the development of MetS (18). Despite these associations, to the best of our knowledge, no studies have specifically examined the relationship between resting substrate oxidation (i.e., CHO and FAT oxidation) and MetS risk in adolescents with obesity.

Given the increasing prevalence of MetS in this population, several indirect indexes have been proposed for its early identification in both clinical and epidemiological settings. Body mass index (BMI) remains the most used parameter (19); however, it does not differentiate between FM and FFM and provides no information on fat distribution (20). In paediatric populations, age- and sex-adjusted BMI z-scores are typically used, although their association with cardiometabolic risk is non-linear (21). Other indexes, such as waist circumference (WC), which indirectly reflect both the quantity and distribution of adipose tissue, have been used to assess body composition and cardio-metabolic risk factors (22). The waist-to-height ratio (WHR) has also emerged as a reliable screening tool to identify MetS in the paediatric population (23). More recently, the Visceral Adiposity Index (VAI), a sex-specific algorithm incorporating anthropometric measures (BMI and WC) and lipid profile parameters [triglycerides and high-density lipoprotein cholesterol (HDL-C)], has been proposed as a novel marker of cardiometabolic risk (24). VAI reflects visceral fat accumulation and dyslipidaemia, and has been linked to insulin resistance, impaired glucose regulation, and an increased cardiovascular risk. Its utility in identifying MetS has also been confirmed in paediatric populations (25). However, none of these indexes incorporates direct assessments of body composition, such as FM and FFM. Instead, they rely on indirect anthropometric parameters, such as BMI or WC. Moreover, to date, no studies have explored the potential role of resting substrate oxidation parameters in developing metabolic indexes for use in paediatric populations to facilitate the early identification of MetS.

Therefore, the aims of the present study were (i) to evaluate the differences in body composition, cardiometabolic parameters, and substrate oxidation at rest in a large cohort of adolescents with obesity, with and without MetS; (ii) to assess which of the parameters mentioned above act as protective or risk factors for the development of MetS; and (iii) to propose a novel MetS index that incorporates direct measures of body composition and substrate oxidation at rest.

The main findings of the present study were: (i) adolescents with MetS exhibited higher values of WC and HC, as well as elevated BMI, FM, triglycerides, fasting insulin, and HOMA-IR compared to their peers without MetS; (ii) FFM (%) and high-density lipoprotein cholesterol (HDL-C) were protective factors for MetS; (iii) higher BMI z-scores, HOMA-IR, and resting CHO were identified as significant risk factors for the development of MetS. Moreover, this paper proposed a novel predictive index for MetS that incorporates direct measures of body composition and resting substrate oxidation, which demonstrated good discriminative ability in identifying the presence of MetS.

Our findings revealed that 202 participants (28%) in our sample met the diagnostic criteria for MetS. This prevalence aligns with previous cross-sectional studies in paediatric populations with obesity, where rates of MetS range from 10 to 38% (4, 5). WC was significantly higher in adolescents with MetS, reinforcing its role as an indirect marker of visceral adipose tissue (VAT) accumulation (20). Consistent with prior research (39, 44, 45), adolescents with MetS exhibited a higher prevalence of metabolic abnormalities, including elevated triglycerides, hyperinsulinemia, insulin resistance, and reduced levels of HDL-C. These data suggest that increased central adiposity, reflected by greater WC, may contribute to the development of insulin resistance and hyperinsulinemia through several well-established mechanisms (46). VAT is highly metabolically active and exhibits increased lipolytic activity, leading to elevated circulating free fatty acids (47). Free fatty acids, in turn, impair insulin-mediated glucose uptake in peripheral tissues partially by the secretion of pro-inflammatory cytokines, which interfere with insulin signalling pathways (47, 48). All together, these alterations contribute to impaired insulin action and may promote early metabolic dysfunction, although this mechanistic pathway was not directly assessed in the present study.

The second key finding of our study was the identification of several significant risk and protective factors associated with the development of MetS in a large cohort of adolescents with obesity. Notably, our analysis highlighted two main protective factors. The first is FFM (%). FFM plays a crucial protective role in the development of MetS in adolescents with obesity. Indeed, higher FFM is associated with improved insulin sensitivity, enhanced glucose uptake, and increased resting energy expenditure, all of which contribute to a more favourable cardiometabolic profile. This is partially supported by our findings, which revealed significant differences in body composition parameters between adolescents with and without MetS. Moreover, skeletal muscle, the main component of FFM, serves as a primary site for glucose disposal and lipid oxidation, playing a central role in maintaining metabolic homeostasis (49, 50). Regular physical activity, particularly resistance and aerobic training, promotes muscle hypertrophy and mitochondrial adaptations, thereby increasing FFM (10, 51). These physiological changes enhance substrate utilisation efficiency and may substantially lower the risk of developing MetS later in life (52). The second protective factor identified is HDL-C. Higher levels of HDL-C were associated with a significantly reduced risk of MetS (OR < 1), underscoring the well-established protective role of HDL in metabolic health. Beyond its role in reverse cholesterol transport, HDL also exerts anti-inflammatory and antioxidant effects that may enhance insulin sensitivity and lower cardiometabolic risk (53). These findings underscore the importance of comprehensive metabolic profiling in adolescents with obesity to detect early metabolic alterations. Identifying both risk and protective factors can inform personalised prevention strategies aimed at improving long-term metabolic health and reducing the progression of MetS into adulthood. Among the most relevant risk factors, a higher rate of resting CHO emerged as a novel and independent predictor of MetS. This observation suggests that adolescents with obesity who predominantly rely on carbohydrates as an energy source at rest may display impaired metabolic flexibility, potentially indicating early alterations in mitochondrial efficiency or insulin signalling (10, 14). From a physiological perspective, elevated resting CHO oxidation may reflect a reduced capacity for lipid oxidation, possibly linked to diminished mitochondrial oxidative function or a blunted response to insulin (10, 14). Such metabolic shifts may favour glycolytic pathways even in the absence of acute energy demand. Moreover, several studies showed that resting metabolic inflexibility and elevated resting CHO are not features of obesity per se, but rather distinctive traits of youth with metabolically unhealthy obesity (13, 54). As previously suggested, these individuals also exhibit poorer insulin sensitivity compared to their peers with metabolically healthy obesity, findings that are partially supported by our results. In our adolescent sample, those with MetS displayed a worse glycaemic profile than those without MetS, despite both groups being affected by obesity. Nonetheless, several confounding factors may influence substrate utilisation at rest, including habitual diet, recent food intake, cardiorespiratory fitness, and physical activity levels (17, 55). These variables can affect insulin sensitivity and substrate availability, thereby modulating the balance between fat and carbohydrate oxidation. While some of these aspects were not directly accounted for in the present analysis, they warrant consideration in future investigations to clarify the underlying mechanisms and strengthen the interpretability of resting CHO oxidation as a marker of cardiometabolic risk. However, the strength of our model lies in having adjusted the odds ratio for age, sex, and the presence of MetS.

However, current findings in paediatric populations remain inconsistent, partly due to the limited consideration of confounding factors such as cardiorespiratory fitness and substrate utilisation during aerobic exercise (56). Elevated BMI z-scores and higher values of the HOMA-IR were also significantly associated with the presence of MetS, in line with previous literature emphasising the central role of excess adiposity and insulin resistance in the pathogenesis of metabolic dysfunction (57, 58). The ORs for these variables indicated a markedly increased likelihood of developing MetS, reinforcing their relevance in early clinical risk stratification.

The last important finding of our study was the development of a novel predictive index for MetS, which integrates WHR, FFM, resting FAT (%), age, BMI, BMR, and sex. This composite index yielded an AUC of 0.73, indicating moderate to good discriminative ability for identifying individuals at risk of MetS. The inclusion of BMR, FFM, and resting FAT (%) represents a key innovation of this model, as these parameters reflect essential aspects of metabolic function that are typically overlooked in standard clinical assessments. Both BMR and FFM are major determinants of total energy expenditure (17) while resting FAT (%) is closely related to metabolic flexibility (12). Moreover, reduced resting FAT (%) has been associated with an increased risk of developing metabolic dysregulation and insulin resistance later in life (13). However, in our cohort, we did not observe significant differences in resting FAT (%) between adolescents with and without MetS. In comparison, commonly used indexes such as BMFI, VAI, WtHR, CMI, and the MetS z-score have demonstrated AUC values ranging from 0.55 to 0.77 in paediatric populations (39, 45). While these indexes provide practical screening tools, they do not account for metabolic and bioenergetic resting parameters. In contrast, our model incorporates both structural components (e.g., FFM, BMI, WHR) and resting energetics parameters (e.g., BMR, resting FAT oxidation), offering a more integrated approach to metabolic risk assessment. Although the predictive performance of our index is comparable to that of existing models, its inclusion of physiologically relevant variables may enhance early risk stratification when used in conjunction with traditional markers. Further validation in larger and more diverse cohorts is warranted to optimise its predictive value and explore its clinical applicability in preventive care. Notably, while the model demonstrated good sensitivity, its specificity was low, suggesting that further investigation and external validation are necessary to improve its ability to accurately exclude individuals without metabolic syndrome and enhance its overall clinical performance. Moreover, external replication studies are necessary to confirm the model’s stability, particularly to investigate the wide confidence intervals observed for some predictors (55).

Our study has several limitations that should be acknowledged. First, longitudinal research is necessary to evaluate the ability of our index to predict the progression of MetS over time in adolescents with obesity. Second, we did not employ dual-energy X-ray absorptiometry (DEXA), the gold standard for body composition analysis, due to the large sample size, cost constraints, and concerns related to radiation exposure in paediatric populations. Instead, we used WC as a surrogate marker of central obesity. Although WC is widely adopted in clinical research, it may be subject to measurement variability. To mitigate this, all anthropometric assessments were conducted by trained and experienced healthcare professionals, enhancing data consistency and reliability.

Moreover, although BIA is less accurate than DEXA (which represents the gold standard), it offers a feasible, non-invasive, and scalable method for estimating body composition, particularly in large cohorts and standard clinical settings. Its affordability and ease of use make it especially valuable in paediatric obesity management, where access to advanced imaging modalities is often constrained. Similarly, although indirect calorimetry is not the gold standard for assessing substrate oxidation, it remains a validated and widely adopted method. Together, these tools offer a pragmatic yet scientifically sound approach, striking an important balance between methodological rigour and practical applicability. Emphasising this balance strengthens the translational relevance of the study, supporting its implementation in real-world settings where resources and time are often limited.

In conclusion, our study identified FFM and HDL-C as significant protective factors against the development of MetS in adolescents with obesity. In contrast, increased WC, BMI, HOMA-IR, and resting RER emerged as key risk factors. These findings emphasise the critical role of body composition, both as a protective and risk-related component, in the pathogenesis of MetS. Moreover, they underscore the importance of promoting regular aerobic and resistance exercise in this population as a targeted strategy to modify risk factors and reduce the likelihood of developing MetS. Importantly, we propose a novel predictive index that incorporates WHR, FFM, resting FAT (%), BMI, BMR, and sex, which demonstrates good discriminatory power and provides a more comprehensive assessment of metabolic risk compared to traditional anthropometric-based indexes. Future longitudinal studies are, however, necessary to understand how changes in this novel predictive index over time might be usefully employed in guiding personalised therapeutic interventions for this clinical condition in a more effective manner.