Trunk-to-limb fat ratio and its relationship with cardiometabolic risk factors in Korean children and adolescents

Article information

Ann Pediatr Endocrinol Metab. 2025;30(6):320-329

Publication date (electronic) : 2025 December 31

doi : https://doi.org/10.6065/apem.2550020.010

Jong Han Kim¹, Jun Hong Park², Hae Sang Lee², Young Suk Shim^,2

¹Ikey Pediatric Clinic, Seoul, Korea

²Department of Pediatrics, Ajou University Hospital, Ajou University School of Medicine, Suwon, Korea

Address for correspondence: Young Suk Shim Department of Pediatrics, Ajou University School of Medicine, 164, World cup-ro, Yeongtong-gu, Suwon 16499, Korea Email: royjays@gmail.com

Received 2025 January 16; Revised 2025 February 5; Accepted 2025 February 20.

Abstract

Hightlights

· This study shows that a higher dual-energy x-ray absorptiometry-derived trunk-to-limb fat ratio is strongly associated with cardiometabolic risk factors and metabolic syndrome in Korean youth aged 10–18 years. Adolescents in the highest trunk-to-limb fat ratio group had markedly increased odds of metabolic syndrome, especially girls. The trunk-to-limb fat ratio has incremental value for metabolic risk beyond body mass index and waist circumference.

Introduction

Obesity in children and adolescents has become a serious health problem worldwide, and its prevalence has increased rapidly in recent decades [1]. This trend is particularly concerning because of the strong association between childhood obesity and various cardiometabolic risk factors, which can later lead to metabolic syndrome and other chronic diseases [2]. In Korea, where the prevalence of childhood obesity is increasing, there is thus a growing need to identify more accurate predictors of cardiometabolic risk in children and adolescents [3,4].

In response to this situation, interest in obesity management is growing, and many patients are having their obesity status evaluated through body measurements. Currently, the most widely used method to measure obesity is body mass index (BMI), which is calculated as body weight divided by the square of height (kg/m²) [5]. However, BMI is generally accepted as having limitations in accurately predicting complications that may occur in children and adolescents [6]. For this reason, many studies have sought to develop tools to supplement the predictive power of BMI, such as the tri-measurement index (TMI), waist circumference (WC), waistto-hip ratio, waist-to-height ratio, and skinfold thickness, individually or in combination [7-12].

Dual-energy x-ray absorptiometry (DXA) has emerged as a valuable tool for assessing body composition and providing detailed information on the amount and distribution of fat throughout the body [13]. Compared with conventional methods, this technique can distinguish between fat mass (FM) and muscle mass in various body regions. The trunk-to-limb fat ratio derived from DXA measurements provides a more nuanced approach for assessing fat distribution and its potential impact on metabolic health [14,15]. There have been many previous studies related to central obesity and BMI. However, few studies have used DXA in relation to metabolic diseases, and even fewer in children and adolescents. Therefore, we investigated the risk of developing metabolic diseases according to body fat distribution on DXA among children and adolescents and evaluated sex-based differences.

The aim of this study was to examine associations between the trunk-to-limb fat ratio, as measured by DXA, and various cardiometabolic risk factors in Korean children and adolescents aged 10–18 years. Data from the Korean National Health and Nutrition Examination Survey (KNHANES, 2009–2011) were analyzed. We aimed to assess the distribution of trunk-to-limb fat ratio by sex and age using the Lambda-Mu-Sigma (LMS) method and to evaluate the correlations between the trunk-to-limb fat ratio and factors associated with the risk of developing metabolic syndrome using the adjusted mean value and adjusted odds ratios (ORs).

Materials and methods

We analyzed data from the KNHANES (2009–2011) [16]. The KNHANES is a national surveillance system that assesses the health and nutritional status of the Korean population and is conducted by the Korea Centers for Disease Control and Prevention on the basis of the National Health Promotion Act. Moreover, it provides data on various health indicators, including oral health indicators, thereby playing a vital role in formulating policies to improve national health. From 2009– 2011, 17,720 subjects in the KNHANES dataset were excluded based on an age over 19 or under 10 years (n=157,620), and subjects without anthropometric data (n=6) or laboratory or DXA data were also excluded. Subjects without assessed variables (n=289) were excluded. Finally, subjects with triglyceride (TG) >400 mg/dL were also excluded, and a total of 1,661 subjects were ultimately included in the study. Study protocols were approved by the Institutional Review Board of Ajou University Hospital (AJOUIRB-EX-2025-025).

Anthropometric measurements were performed by trained experts. Height was measured to the nearest 0.1 cm using a Seca 225 (Seca, Germany). Weight was measured to the nearest 0.1 kg using a GL-6000-20 (G-tech, Korea). WC was measured to the nearest 0.1 cm at the midpoint between the lower rib margin and the iliac crest. Standard deviation scores (SDSs) were used for height, weight, BMI, TMI, and WC because of the variable distributions of these parameters between individuals according to age and sex. The height SDS, weight SDS, BMI SDS, TMI SDS, and WC SDS were determined through LMS methods using the 2017 Korean reference [17].

Blood pressure was measured 3 times on the right upper arm using a calibrated sphygmomanometer (Baumanometer Desk model 0320, Baum, USA). BP was measured at 2-minute intervals, and the mean of the last 2 BP measurements was used for analysis. Blood samples were collected after at least 8 hours of fasting. Serum glucose, total cholesterol, HDL cholesterol, and TGs were measured using a Hitachi 7600 automatic analyzer (Hitachi, Japan). LDL cholesterol was calculated using the Friedewald’s equation. Body composition was measured by DXA using a Hologic DISCOVERY-W scanner. This provided data on lean mass, FM, and percent body fat for the whole body and trunk regions.

In this study, information on lifestyle-related variables and socioeconomic status was collected through a questionnaire. Lifestyle variables included smoking, drinking, and physical activity. Smokers were defined as those who smoked more than 5 packs of cigarettes in their lifetime, with all others defined as nonsmokers, and drinkers were defined as those who drank more than twice a month in the past year, and all others were considered nondrinkers. Physical activity was defined as meeting at least one of the following: vigorous activity for 20 minutes 3 or more days a week, moderate activity for 30 minutes 5 or more days a week, or walking for 30 minutes 5 or more days a week. The participants were classified into exercise and nonexercise groups. Socioeconomic status variables included household income and residence, with household income reported in 4 quartiles and divided into the lowest quartile and the second quartile or higher, and residence was classified into urban and rural areas on the basis of location.

On the basis of the 2017 Korean growth chart, elevated WC was defined as WC ≥ 90th percentile for sex and age, and elevated BP was defined as systolic or diastolic blood pressure ≥90th percentile for sex, age, and height or use of antihypertensive medication. Elevated glucose was considered a fasting blood sugar ≥ 110 mg/dL or a history of type 2 diabetes, and type 2 diabetes was diagnosed when one of the following criteria were met: selfreported, drug treatment, or fasting blood sugar ≥ 126 mg/dL. Elevated TGs were defined as TGs ≥ 110 mg/dL or when taking medication for dyslipidemia, and reduced high-density lipoprotein cholesterol (HDL-C) was defined as HDL cholesterol < 40 mg/dL. When 3 or more of these 5 criteria are met, metabolic syndrome can be diagnosed according to the modified criteria of the National Cholesterol Education Program Adult Treatment Panel III [18].

We divided participants by age and sex to study the correlation between metabolic syndrome and the trunkto-limb fat ratio in subgroups. In addition, we divided the patients into 3 groups according to trunk-to-limb fat ratio FM percentile to evaluate correlations with the risk of developing metabolic syndrome: group 1, <85th percentile; group 2, ≥85th percentile and <95th percentile; and group 3, ≥95th percentile.

Statistical analysis was performed using R 3.5.1 (The R Foundation for Statistical Computing, Austria). Continuous variables with a normal distribution are presented as the mean±standard deviation. Differences between sexes were analyzed using the independent t-test for continuous variables with a normal distribution and the chi-square test for categorical variables. Sex- and age-specific reference data for the trunk-tolimb fat ratio were obtained using the LMS model, which fits smoothed L (asymmetry), M (median), and S (coefficient of variation) curves. We extracted the LMS value using the Box-Cox t distribution using the gamlss package of the R program. Age- and sex-adjusted Pearson correlation analyses were conducted to assess the relationships among the trunk-to-limb fat ratio SDS, BMI SDS, TMI SDS, and clinical parameters. The adjusted mean values of cardiometabolic risk factors were compared among the 3 groups using analysis of covariance and Bonferroni post hoc correction after adjusting for sex, age, and BMI. Adjusted ORs and 95% confidence intervals between trunk-to-limb fat ratio groups were calculated using multiple logistic regression analysis, with the group with a trunk-to-limb fat ratio <85th percentile used as the reference group and adjustment for age, sex, FM index, alcohol consumption status, smoking status, physical activity status, rural residence, and household income, and diagnosis of T2DM, hypertension, or dyslipidemia.

Results

The clinical characteristics of the 1,661 participants (891 boys, 770 girls) are presented in Table 1. No significant between-sex differences were found in age, height SDS, weight SDS, WC SDS, or BMI SDS. Boys had higher SBP, DBP, and serum glucose levels, whereas girls had higher total cholesterol (T-C), HDL-C, TG, and low-density lipoprotein cholesterol (LDL-C) levels. DXA results revealed that boys had greater total mass and lean mass, whereas girls had higher FM and FM percentage, both for the whole body and specific regions (trunk, arms, legs). In terms of the trunk-to-limb fat ratio, boys had greater FM and FM percentage than girls.

Table 1.

Clinical characteristics of study population (n=1,661)

LMS values and specific percentile limits for trunk-tolimb fat ratio FM percentile, categorized by age and sex (n=1,661 individuals), are presented in Table 2. The median trunk-to-limb fat ratio (50th percentile) in males tended to increase with age, from 0.787 in 10-year-old males to 0.940 in 18-year-old males. Additionally, the 3rd percentile also increased with age from 0.611 in 10-yearold males to 0.724 in 18-year-old males, as did the 97th percentile, which increased from 0.996 in 10-year-old males to 1.200 in 18-year-old males. Median trunk-tolimb fat ratio (50th percentile) did not change as much in women as in men. There was no significant difference, with a median of 0.774 for 10-year-old females and 0.771 for 18-year-old females. The 3rd percentile showed a decreasing trend, at 0.610 for 10-year-old females and 0.597 for 18-year-old females; however, the 97th percentile did not differ significantly; it was 0.971 for 10-yearold females and 0.973 for 18-year-old females (Fig. 1).

Table 2.

Lambda-Mu-Sigma values and specific percentile limits for trunk-to-limb ratio of fat mass percent according to age and sex (n=1,661)

Fig. 1.

Trunk-to-limb ratio fat mass percentile according to age and sex (n=1,661).

Correlations among trunk-to-limb fat ratio SDS and cardiovascular metabolic risk factors are shown in Table 3. Analysis was performed through 3 models. All risk factors except for DBP and serum glucose were significantly correlated with the trunk-to-limb fat ratio SDS in both models 1 and 2 (all P-values <0.001). The WC SDS showed a relatively strong positive correlation, with r values of 0.506 in model 1 and 0.509 in model 2. The BMI SDS and TMI SDS were also positively correlated with the SDS for the trunk-to-limb fat ratio in models 1 and 2. In terms of lipid profiles, T-C, TG and LDL-C were positively correlated in all 3 models. HDL-C showed a negative correlation in models 1 and 2, but the statistical significance was lost in model 3 (P=0.052). DBP and glucose were not significantly correlated in any of the 3 models.

Table 3.

Unadjusted and adjusted correlation between standard deviation score for trunk-to-limb ratio of fat mass percent and cardiometabolic risk factors in study population (n=1661)

The adjusted means of cardiometabolic risk factors for the 3 trunk-to-limb fat ratio subgroups are shown in Table 4. In boys and men, the WC SDS, BMI SDS, and TMI SDS were significantly higher in group 3 than in group 1, and the result persisted after adjusting for age. The levels of T-C, TG, and LDL-C significantly increased from group 1 to group 2 and from group 2 to group 3 and remained significant after adjusting for age and FM index. As with men, in girls and women, the WC SDS, BMI SDS, and TMI SDS were significantly higher in group 3 than in group 1. In addition, TG tended to increase significantly and HDL-C tended to decrease significantly by consecutive groups in girls and women.

Table 4.

Adjusted means of cardiometabolic risk factors according to 3 group of age- and sex-trunk-to-limb ratio of fat mass percent (n=1,661)

The adjusted ORs (aORs) were calculated to assess associations among trunk-to-limb fat ratio percentile groups (groups 2 and 3), the presence of metabolic syndrome, and cardiometabolic risk factors. Group 1 served as the reference group (Table 5). Among all the participants, those in group 3 had greater risk of elevated WC and TG (aORs, 3.99 and 2.45, respectively) and greater risk of decreased HDL-C (aOR, 2.81) than did those in group 1. In addition, metabolic syndrome was more common in group 3 (aOR, 3.63). In boys and men, the risk of elevated WC (aOR, 3.36) and metabolic syndrome (aOR, 2.69) was greater in group 3 than in group 1. Similar patterns were observed in girls and women, with high risk of elevated WC and metabolic syndrome, at ORs of 4.94 and 8.66, respectively.

Table 5.

Adjusted odds ratio and 95% confidence interval of metabolic syndrome and its components according to 3 group of age- and sex-specific trunk-to-limb ratio of fat mass percent (n=1,661)

Discussion

In this study, we analyzed the association between DXA-measured trunk-to-limb fat ratio and metabolic syndrome, as well as its key components: WC, BP, fasting glucose, TG, and HDL-C. The results revealed notable differences in trunk-to-limb fat ratio between males and females during adolescence. In males, an increasing trend was observed with age, whereas in females, the ratio remained relatively stable. This difference seems to reflect disparate patterns of adipose distribution during puberty, with males exhibiting a tendency to accumulate more central fat [19]. These findings highlight the importance of considering sex-specific patterns when assessing body composition in adolescents [20] (Table 2).

Our study also revealed that trunk-to-limb fat ratio SDS was strongly correlated with multiple cardiometabolic risk factors. A similar previous study was conducted in Korea [21]. In this study, higher FM in the legs than in the arms, higher fat-free mass (FFM) in the arms than in the legs, and higher FM or FFM in the limbs than in the trunk were associated with lower prevalence of cardiovascular risk factors. Our study also revealed a similar trend, and the robust positive correlations of trunk-to-limb fat ratio with the WC SDS, BMI SDS, and TMI SDS suggest that it is closely associated with overall obesity and abdominal obesity. Notably, the results revealed significant differences between males and females. In males, T-C, TG, and LDL cholesterol were more strongly correlated with trunk-to-limb fat ratio, whereas in females, TG and HDL cholesterol-C were more strongly correlated. These sex-based differences likely reflect sex differences in fat distribution and metabolic function, suggesting the need for sex-specific, tailored metabolic risk assessments in the future [22]. One remarkable finding was that the risk of developing metabolic syndrome was significantly increased in the group with a high trunk-to-limb fat ratio (≥95th percentile). This relationship was notably more pronounced in girls (aOR, 8.66). These findings suggest that the trunkto-limb fat ratio has potential value as a predictor of metabolic syndrome and may be particularly useful in females.

DXA has emerged as a powerful tool for assessing body composition, offering several advantages over traditional anthropometric measures [23]. DXA offers a more comprehensive body composition analysis, including the ability to measure regional fat distribution, such as in the trunk and limbs [13]. The ability to distinguish between trunk fat and limb fat enables a more accurate assessment of fat distribution, which is crucial in determining metabolic risk. While BMI remains a useful initial screening tool because of its simplicity and low cost, it has limitations in predicting complications and distinguishing between FM and lean mass. Therefore, the trunk-to-limb fat ratio derived from DXA measurements appears to be a more sensitive marker of cardiometabolic risk than conventional measures such as BMI or WC.

Our study revealed significant correlations among DXA-derived trunk-to-limb fat ratio and various cardiometabolic risk factors, including WC, BMI, TMI, and lipid profiles. Notably, higher trunk-to-limb fat ratio was associated with increased odds of metabolic syndrome and its components, particularly in the highest percentile group (Table 5). Several similar studies have investigated the correlation with body FM using DXA, which may support the results of this study. Samsell et al. [24] reported that visceral adipose tissue was strongly associated with cardiometabolic risk factors when body composition was measured via DXA. Weber et al. reported that trunk-to-limb fat ratio strongly predicts insulin resistance in children and adolescents [25]. Teixeira et al. [26] used DXA-derived waist-to-hip ratio to assess metabolic risk in adolescents and reported that it was a useful indicator of cardiometabolic health. Choi et al. [27] investigated the relationship between trunk-to-limb fat ratio measured by DXA and metabolic syndrome in Korean adults and reported that a higher trunk-to-limb fat ratio was associated with a greater risk of developing diabetes.

This study has several notable limitations. Because of its cross-sectional design, the study is unable to examine longitudinal trends and causal relationships between identified significant factors. In addition, the study used data exclusively from the KNHANES, which limits the generalizability of the results to other racial or ethnic groups. The study population consisted of children and adolescents aged 10–18 years. Children younger than 10 years were excluded, which limits the applicability of the results to all children and adolescents. Moreover, individual differences in the onset and progression of puberty within the age group (10–18 years) of these study populations were not considered, which may have influenced the observed changes in trunk-to-limb fat ratio.

In conclusion, this study demonstrated that trunk-tolimb fat ratio is significantly associated with cardiometabolic risk factors and metabolic syndrome in Korean children and adolescents. This measure appears to provide valuable information on metabolic health beyond traditional anthropometric measurements. The observed sex-specific patterns highlight the need for a tailored approach to assess body composition and metabolic risk in male and female adolescents. These findings have important implications for clinical practice and public health strategies to prevent and manage metabolic disorders in young populations.

Article information Continued