The study presents new age- and sex-related reference values and charts for various body composition parameters in a large, randomly selected population of ethnically homogeneous, normal-weight adolescents aged 10-18 years from Central Europe. To ensure a clear reference point for body composition parameters, individuals who were underweight, overweight or obese were deliberately excluded. Underweight participants may have lower lean mass and bone mineral density, indicating potential deficiencies in muscle and bone development [36]. Overweight or obese individuals may have unique growth patterns and body composition characteristics that may be influenced by excess adiposity and potential hormonal fluctuations [37]. Using normal weight adolescents as a reference group provides insight into typical growth and body composition patterns and identifies deviations from this established norm. Furthermore, our study includes additional measures of body composition such as BCM, TBW, and MM, which is in contrast to many studies in the field that focus solely on FFM and FM. This methodology improves the understanding of adolescent body composition, improving screening and clinical applications related to adolescent nutritional health.
The results of the study indicate that the metrics of body composition generally increase with age during adolescence, reflecting the overall growth in body size leading up to adulthood, particularly during the significant growth spurt of puberty. However, a significant deviation from this pattern is observed in boys' FM%, which decreases throughout adolescence due to the slower increase in fat mass and faster increase in fat-free mass. On average, boys tend to have higher cellular mass, muscle tissue, overall fat-free body mass, and a higher proportion of water compared to girls. In contrast, boys consistently show significantly lower values for FM, FMI, and FM%, indicating a lower proportion of body fat compared to girls. This trend is consistent with previous research that has shown a gradual increase in body composition measures with age, as well as a significant impact of puberty on sex-related differences in these measures [24, 25]. It is well established that the observed differences in body composition between boys and girls are attributed to a number of factors, including genetics, hormones, and lifestyle. These factors are associated with pubertal growth spurts and sexual maturation [38]. During puberty, boys experience a surge in testosterone production, which stimulates muscle growth and reduces body fat. Similarly, girls experience an increase in estrogen levels, which can lead to fat accumulation, particularly in the hips and thighs [39]. It is crucial to recognise that boys may require greater caloric intake to facilitate muscle growth and energy demands, while girls may have different dietary needs. Although there are generally recognised growth patterns during adolescence, the timing and intensity of growth spurts can differ between individuals. Some individuals may experience growth spurts earlier or later, with the rate of growth varying, as well as the timing of sexual maturation stages. Consequently, there is a notable degree of individual variation in the parameters of body composition, both within and between sexes. It is therefore imperative to consider these variations when interpreting findings related to growth in adolescence, as they emphasise the distinctive nature of each individual's development during this transitional phase.
A comparative analysis was conducted between the reference centiles for FFMI and FMI among Polish normal weight adolescents and population data from adolescents of European origin in Austria (LEAD study) [24], Britain [40], Germany (MoMo study population data) [25] and Tyrol (EVA4YOU study) [41]. The raw centile curves for the 3rd (2nd in Britain cohort), 50th, and 97th (98th in Britain cohort) are presented in Figure 4.
The overall pattern is similar, indicating a rise in values for these two metrics. One of the most notable distinctions is the significantly lower centile values of the 97th (98th in Britain cohort) centile values observed in our data set compared to the others. This is understandable because our sample consists only of normal weight individuals, while the comparative samples represent the total population. A similar comparison was made between our study and the normal weight subsample of German adolescents from the MoMo study [25]. The raw centile curves are presented in Figure 5.
The growth trajectories of the two compared samples were similar, although there were more noticeable disparities in the data for boys than for girls, particularly regarding the 97th percentile compared to others. In boys in our sample, the FMI values were slightly lower, showing a tendency toward an increase in the discrepancy by the age of 18. On the contrary, for girls, the differences in FFMI and FMI between the comparative samples were negligible, except among those aged 18 years. Overall, our data align closely with those of a cohort of German adolescents of normal weight. These findings support our suggestion that normal weight adolescents can be used as a benchmark to evaluate body composition parameters.
It is unclear whether the method used to measure body composition makes a difference. LEAD and Britain cohort studies used DXA, while MoMo and EVA4YOU studies used BIA, a method similar to our own approach. Achamrah colleagues [33] examined the agreement between these two methods in a sample of more than 3,000 adult patients and found only a slight discrepancy between them, particularly among patients with a BMI between 16 and 18. This indicates that these two methods can be used interchangeably at the population level. However, they observed significant discrepancies at the individual level, regardless of BMI. It appears that the choice of measurement method does not significantly affect the results of our comparison conducted at the population level. Furthermore, studies conducted to validate BIA against dilution techniques for measuring TBW have revealed a satisfactory general agreement between these two methods in healthy children and adolescents [42].
With the establishment of new age- and sex-specific reference data for body composition in Polish adolescents, a question arises: Should these local reference data be preferred over international benchmarks? This decision carries considerable weight. The existing literature consistently supports the use of local growth references to monitor changes in physical growth during childhood and adolescence [43-45]. For example, the study by Kułaga and colleagues highlights the superiority of local references in capturing population-specific variations in physical growth among Polish children and adolescents compared to international benchmarks [46]. International reference values are undoubtedly essential to compare global growth patterns and ensure consistency in research. They provide a standardised framework and facilitate cross-population analysis, making them indispensable in clinical practice. They also provide clinicians with guidelines and recommendations for nutrition, health, and child development in diverse populations around the world, facilitating the evaluation and monitoring of growth and the diagnosis of growth-related disorders. On the contrary, local references take into account cultural, ethnic, and environmental differences that are specific to certain regions or populations. By providing accurate benchmarks tailored to local demographics, they enable accurate growth assessments and guide interventions to address the unique needs of these populations. It is important to note that neither set of reference data should be prioritised over the other. Instead, the choice should depend on the objective of the study. A collaborative approach is essential to ensure a comprehensive understanding and effective monitoring of body composition in adolescence.
The assessment of body composition involves the establishment of cutoff points for recognising normal or pathological body composition, which would be clinically and epidemiologically useful for identifying individuals at risk of various health outcomes.
However, there is no consensus on these cut-off points in the literature. McCarthy and colleagues recommend using fat mass cut-off points in the second, 85th and 95th centiles to define the underfat, normal, overfat and obese categories, respectively [47]. Although there is broad agreement on the definitions of overweight and obese, it is recommended to use the 5th centile rather than the 2nd for the underfat category [48]. It is generally preferable to establish cut-off points based on the risk of various diseases rather than arbitrary choices based only on centile thresholds [41, 49]. As each parameter of body composition has different implications for health, our future research will aim to establish cut-off points for centile values of these parameters. This will contribute to a better understanding of healthy and pathological body composition.
Study limitations
This study provides valuable information on body composition during adolescence. However, it is important to be aware of some limitations that may affect the interpretation of the results. First, due to the cross-sectional design, body composition measurements were taken only once, making it difficult to follow individual variations over time and potentially biasing the results. Furthermore, the lack of information on the stage of sexual maturation of subjects makes it difficult to interpret the data, particularly with regard to changes in muscle mass and fat accumulation associated with puberty. The sample does not represent the entire cohort of Polish adolescents, but our previous studies showed that the difference between the nationwide sample [26] and ours can be negligible [29]. Despite these limitations, the study has several notable strengths. First, it is based on a population-based sample that includes healthy adolescents of all types of school and different social groups, ensuring representativeness. Second, the large sample size in each age category, covering all of adolescence (10 to 18 years) makes the results more reliable and valid. To ensure that body composition measurements were as accurate as possible, the study was conducted according to the relevant standards [50]. In addition, a range of body composition parameters gives a more complete picture.