Sage Journals: Discover world-class research

Abstract

Background

Metabolic syndrome (MetS) is defined as a cluster of risk factors for predicting type 2 diabetes mellitus and cardiovascular disease.

Objectives

This cross-sectional study aimed to develop a cut-off value for fat versus lean mass ratio (FLMR) in predicting MetS and to investigate the association between this indicator with MetS and its components.

Methods

Subjects comprised 238 Malay adolescents (79% female) aged 18–19 years old. Anthropometric assessment comprised weight, height and waist circumference (WC). Body composition was measured using bioelectrical impedance analysis techniques while blood pressure was measured using a blood pressure monitor. Fasting blood glucose, total cholesterol , triglycerides, high-density lipoprotein cholesterol (HDL-c), and low-density lipoprotein cholesterol were determined from an overnight fasting blood sample. MetS was determined based on International Diabetes Federation (2007) definition for adolescents aged 16-year-old and above.

Results

The prevalence of MetS was 2.1%. Receiver Operating Characteristics curve analysis revealed that the optimal cut-off value for FLMR was 0.441 with an Area Under the Curve of 0.874 (95% CI: 0.825, 0.913); with sensitivity of 80.0% and specificity of 71.0%. FLMR cut-off of 0.441 was associated with high WC (p < .001), low HDL-c (p < .001) and MetS (p < .05). Binary Logistic Regression analysis revealed that adolescents with high WC, low HDL-c and MetS had higher odds of developing increased FLMR than the cut-off value with an odds ratio (OR) of 43.4 (95% CI: 9.7,193.9), 4.7 (95% CI: 2.3,9.8) and 13.3 (95% CI: 1.5,121.2), respectively.

Conclusion

FLMR possesses fair discriminatory ability in identifying MetS among adolescents and significant association exists between FLMR and MetS and some of its components.

Keywords

Metabolic syndrome anthropometry body composition adolescent

Introduction

Metabolic syndrome (MetS) is defined as a cluster of risk factors that includes central obesity, raised blood pressure, raised fasting plasma glucose, increased triglycerides, and decreased high-density lipoprotein-cholesterol (HDL-c) for predicting type 2 diabetes mellitus (T2DM) and cardiovascular disease.¹ Among all the risk factors, central obesity is positioned as the necessary parameter to diagnose MetS.^1,2 In Malaysia, current literature revealed that 15–17% and 12–14.9% of adolescents were overweight and obese respectively; while the prevalence of MetS among adolescents was reported as 2.6%.^3–5 Nevertheless, Wee and colleagues found no significant association between age and gender with the MetS among children living in metropolitan Kuala Lumpur.⁶ In recent years, data from several studies have proven that a higher rate of MetS was found among overweight and obese Malaysian adolescents compared to their normal-weight counterparts.^4,6,7

As obesity have a close relationship with MetS, it is important to deeply investigate the relationship between body composition; which divides our body into fat mass (FM) and lean body mass (LBM) with MetS. This is because although body mass index (BMI) provides a useful index for measuring body weight status, it does not estimate the proportion of FM and lean LBM.⁸ Since the FM compartment is composed of adipose tissue while the LBM compartment is composed of water, organs and muscles, it is important to understand the roles of body composition in regulating energy metabolism, glucose and lipid metabolism which are closely related to metabolic disorders such as T2DM, hypertension and dyslipidemia.^9,10 This was supported by a systematic review which concluded that the risk of cardiovascular risk factors increases simultaneously with the increase in central body fat deposition.¹¹ As such, body composition indices are known to be the best tools to define the risk of having MetS. Several international studies had established that body composition indices such as waist circumference (WC), waist-height ratio, the sum of four skinfolds, fat mass index (FMI) and fat free mass index were shown to have the high discriminatory ability (area under the curve (AUC) >0.7) in predicting MetS among adolescents.^12–18 Nevertheless, there are some shortfalls behind these indicators as they only focused on the effect on fat and non-fat component independently and do not consider the antagonistic effects of FM and LBM on MetS.

In this study, we aim to integrate these two variables into a single anthropometric tool by expressing them into a ratio which is the fat-to-lean mass ratio (FLMR). Existing research recognised the role of FLMR in predicting MetS among adults.^19,20 This indicator had also been employed in the clinical setting where higher FLMR was significantly associated with metabolic dysfunction among women with polycystic ovary syndrome, suggesting the clinically relevant utility of this indicator.²¹ Although some research had been carried out on FLMR among young adults and adults, there have been no attempts to develop a cut-off value for FLMR among Malaysian adolescents. To address the literature gap, this paper aims to develop a cut-off value for FLMR in predicting the risk of MetS and to study the association between FLMR with MetS and its components among Malay adolescents in Terengganu.

Methods

Study design and sample selection

A cross-sectional study was conducted from 1st January to 31st May 2019 at a local university in Terengganu. Malay adolescents aged 18–19-year-old from the foundation centre and all the diploma programmes were recruited as participants as this study was conducted in a university setting with pre-university programme. Adolescents who were aged above 19-year-old, non-Malay ethnic and were diagnosed with chronic diseases related to MetS including hypertension, diabetes mellitus, heart disease, hypothyroidism or hyperthyroidism or endocrine disorder as well as cannot undergo fasting before blood sample withdrawal were excluded.

Sampling method

Systematic random sampling was used as the sampling method where the participants were chosen based on a fixed interval according to their number in the name list. The sample size was calculated using EpiInfo software by using a prevalence of 3.8% and an odds ratio (OR) of 4.93 which was adapted from a recent cross-sectional study in Malaysia.⁷ By considering a 20% drop-out rate, the final sample size calculated was 254.

Data collection

Firstly, students who were interested to participate in this study were asked to fill in the consent form and also the socio-demographic form prior to the day of data collection. After all the socio-demographic forms were collected, the students were screened based on their personal information to ensure they were free of diseases related to MetS as mentioned in the exclusion criteria above and fulfilled the age criteria.

Measurement of variables

Socio-demographic information

The socio-demographic questionnaire consists of three sections: personal information, and paternal and maternal information. Personal information such as age, history of disease related to MetS including hypertension, diabetes mellitus, heart disease, hypothyroidism or hyperthyroidism, paternal and maternal medical history as well as household income were obtained. For household income, categorization was done according to the latest definition released by the Department of Statistics Malaysia (2017): less than or equal to RM3000 considered as B40 (low), less than or equal to RM6275 considered as M40 (moderate), more than RM6275 considered as T20 (high).

Anthropometric measurement

The height and weight of the participants were measured to the nearest 0.1 using SECA Stadiometer 217 (SECA, Germany) and weighing scale SECA Clara 803 (SECA, Germany) respectively. All of them were informed to wear light clothes before the measurement and removed all the things such as keys, loose change, jewellery and jackets which will affect the final results.²² The height and weight were measured three times and the average reading was recorded. The BMI was calculated and the body weight was categorized based on WHO classification²³: (1) <-3SD: severe thinness; (2) <-3SD - <-2SD: thinness; (3) ≥-2SD - <1SD: normal; (4) 1SD-2SD: overweight; (5) >2SD: obesity.

Waist circumference was measured using an anthropometric tape (Rosscraft Anthrotape, Washington, USA) to the nearest 0.1 cm. The tape was passed around the body at the point midway between the iliac crest and the costal margin. The measurement was taken at the end of a normal expiration^22,24 and was taken three times and the average reading was calculated.

Body composition measurement

The participants’ FM and LBM were measured using the bioelectrical impedance analysis (BIA) technique. The instrument used was Bodystat Quadscan 4000 (Isle of Man, United Kingdom) which is a type of direct segmental multifrequency bioelectrical impedance analysis (DSM-BIA). Prior to the measurement, the participants were asked to have overnight fasting, not to drink water and not to carry out any exercise. Besides that, the participants were asked to wear light clothing and empty their bladders prior to the assessment. During the measurement, the participants were asked to lie down in a supine position with both of their legs and arm spreading out. The skin sides were rubbed with alcohol before placement of the electrode to ensure proper electrical conduction.²⁵

Blood pressure measurement

Blood pressure was measured using a blood pressure monitor (Omron HEM 7203, Kyoto, Japan). Before the measurement was taken, the participant was asked not to talk during the measurement process, not to drink caffeine 30 min before the measurement, not to carry out exercises and was advised to stay calm to ensure accurate measurement. Blood pressure was taken twice and the mean blood pressure was calculated.²⁶

Biochemical measurement

All participants were reminded to undergo a 10–12 h of fasting before the blood sample is collected. A venous blood sample was withdrawn by a trained phlebotomist to determine the fasting glucose level and lipid profiles. The blood was centrifuged and the serum was assayed for triglycerides, and plasma glucose was assayed using a colorimetric test using Cobas 8000 instrument (Roche Diagnostics, Selangor, Malaysia) by a private laboratory with ISO qualification.

Diagnostic criteria

Metabolic syndrome is diagnosed based on the criteria published by International Diabetes Federation (IDF 2007). The participant is diagnosed when there is the presence of central obesity (WC ≥90 cm for male; ≥80 cm for female) and the other two out of four following criteria: (1) increased triglycerides (TC): ≥1.7 mmol/L or specific treatment for high triglycerides; (2) decreased HDL-c level: <1.03 mmol/L in males and <1.29 mmol/L in females; (3) ≥130/85 mmHg or treatment of previously diagnosed hypertension; (4) hyperglycaemia: fasting plasma glucose ≥5.6 mmol/L².

Statistical analysis

Data obtained from this study were analysed using IBM SPSS for Windows, Version 22.0 (IBM Corp, 2013, Armonk, NY) and MedCalc for Windows Version 15. (MedCalc Software, Ostend, Belgium). The parametric test was used to analyse normally distributed data while the non-parametric test was used to analyse skewed data. Mann-Whitney U Test or independent t-test was used to compare the differences between values of anthropometric and metabolic parameters between gender. The significant level was set at p < .05. The cut-off value for FLMR and the reliability of this cut-off value was determined by performing Receiver Operating Characteristics (ROC) curve analysis. Binary logistic regression was used to analyse the raw data as metabolic syndrome will be classified as “MetS” or “Non-MetS” while FLMR will be classified as “higher than the threshold” or “lower than the threshold”. The threshold value was obtained after the ROC was carried out.

Results

A total of 290 participants were recruited from the sampling frame to participate in this study. All of the screened participants met the inclusion criteria and hence were invited to participate in this study. Fifty-two respondents withdrew due to personal reasons such as being unwell, needing to attend other activities, or having family matters hence the final sample achieved was 238 people which was equal to an 82.8% response rate.

Out of 238 subjects, 21.0% were male while 79.0% were female. The majority of them were in the low (39.1%) and high household income (38.2%) groups while the remaining belonged to the medium household income group (22.7%). The majority of the subject’s fathers (59.2%) and mothers (62.4%) do not have any significant family history of obesity, hypertension, diabetes mellitus, cardiovascular disease and thyroid diseases (Table 1).

Table 1.

Socio-demographic characteristics of the participants.

	Male (n = 50)n (%)	Female (n = 188)n (%)	Total (n = 238)
Gender	50 (21.0%)	188 (79.0%)
Total household income (n = 238)
Low (≤ RM 3000)	25 (50.0%)	68 (36.2%)	93 (39.1%)
Medium (≤ RM 6725)	7 (14.0%)	47 (25.0%)	54 (22.7%)
High (> RM 6725)	18 (36.0%)	73 (38.8%)	91 (38.2%)
Paternal education level (n = 228)
Primary level	3 (6.4%)	13 (7.2%)	16 (7.0%)
Secondary level	24 (51.1%)	72 (39.8%)	96 (42.1%)
Tertiary level	20 (42.6%)	96 (53.0%)	116 (50.9%)
Maternal education level (n = 237)
Primary level	2 (4.0%)	9 (4.8%)	11 (4.6%)
Secondary level	25 (50.0%)	96 (51.3%)	121 (51.1%)
Tertiary level	23 (46.0%)	82 (43.9%)	105 (44.3%)
Paternal medical history (n = 228)
Yes	17 (36.2%)	76 (42.0%)	93 (40.8%)
No	30 (63.8%)	105 (58.0%)	135 (59.2%)
Maternal medical history (n = 237)
Yes	17 (34.0%)	72 (61.5%)	89 (37.6%)
No	33 (66.0%)	115 (38.5%)	148 (62.4%)

Anthropometric measurements

Anthropometric measures and the bodyweight status of the participants are displayed in Table 2 and Figure 1. The mean BMI-for-age z-score of the participants was reported as 0.082 ± 1.299. For all the measurements, there was a significant difference (p < .05) between males and females. Males had significantly higher weight, height, BMI-for-age and waist circumference than females. On the other hand, females had significantly (p < .001) higher fat mass, body fat percentage and FLMR than males while males had significantly (p < .001) higher lean mass than females. As shown in Figure 1, most of the subjects (66.8%) had normal BMI followed by overweight (18.9%), obese (8.0%) and thinness (6.3%). The prevalence of obesity (9.0%) among females was higher than among males (4.0%). On the other hand, the prevalence of thinness among males (14.0%) was higher than females (4.3%).

Table 2.

Anthropometric characteristics of the participants.

Anthropometric/body composition parameters	Male(n = 50)Mean ± SD/Median (IQR)	Female(n = 188)Mean ± SD/Median (IQR)	Total(n = 238)Mean ± SD/Median (IQR)	p-value
Weight (kg)	61.10 (17.80)	53.40 (14.50)	54.50 (15.80)	.001^a
Height (cm)	167.60 (7.30)	156.05 (6.80)	157.50 (9.60)	<.001^a
BMI-for-age (z-score)	−0.256 ± 1.476	0.171 ± 1.237	0.082 ± 1.299	.039^b
WC (cm)	70.35 (13.72)	67.50 (12.40)	67.90 (11.48)	.006^a
FM (kg)	8.6 (4.4)	14.2 (7.7)	13.0 (7.3)	<.001^a
LBM (kg)	53.7 (12.6)	36.6 (7.5)	40.5 (9.3)	<.001^a
BF %	14.49 (5.72)	27.43 (7.22)	25.81 (9.09)	<.001^a
FLMR	0.186 ± 0.081	0.402 ± 0.107	0.357 ± 0.135	<.001^b

BMI: body mass index; WC: waist circumference; FM: fat mass; LBM: lean body mass; BF: body fat percentage; FLMR: fat-to-lean mass ratio; SD: standard deviation; IQR: inter-quartile range.

^aMann-Whitney U-test was applied. Median and IQR were reported.

^bIndependent t-test was applied. Mean and SD were reported.

Figure 1.

Body weight status of the participants.

Metabolic syndrome components

Table 3 reveals the overall biochemical profile of the participants. The statistical test revealed that females had significantly (p < .05) higher total cholesterol, HDL-c and LDL-c than males. On the opposite side, males had significantly (p < .05) higher fasting blood glucose and systolic blood pressure than females. No significant difference was observed in total triglycerides level and diastolic blood pressure between gender (p > .05). Overall, the prevalence of MetS was 2.1% among Malay adolescents. In terms of gender breakdown, surprisingly none of the male subjects had MetS whilst the prevalence of MetS among females was 2.7%.

Table 3.

Metabolic parameters of the participants.

Variables	Male (n = 50)Mean ± SD/Median (IQR)	Female (n = 188)Mean ± SD/Median (IQR)	Total (n = 238)Mean ± SD/Median (IQR)	p-value*
Biochemical markers
TC (mmol/L)	4.30 (0.95)	4.61 (0.87)	4.57 (0.88)	.006^a
HDL-c (mmol/L)	1.43 + 0.263	1.57 + 0.297	1.54 + 0.295	0.003^b
LDL-c (mmol/L)	2.48 + 0.594	2.71 + 0.662	2.67 + 0.654	.025^b
TG (mmol/L)	0.78 (0.36)	0.72 (0.31)	0.74 (0.31)	.267^a
FBG (mmol/L)	4.60 (0.33)	4.50 (0.77)	4.50 (0.50)	.005^a
Blood pressure measurement
SBP (mmHg)	120.00 (15.00)	112.00 (13.00)	114.00 (15.00)	<.001^a
DBP (mmHg)	72.00 (7.00)	72.00 (9.00)	72.00 (9.00)	.550^a
MetS prevalence, n (%)
Presence	0	5	5
Absence	50	183	233

TC: total cholesterol; HDL-c: high-density lipoprotein-cholesterol; LDL-c: low-density lipoprotein cholesterol; TG: total triglycerides; FBG: fasting blood glucose; SBP: systolic blood pressure; DBP: diastolic blood pressure; MetS: Metabolic syndrome.

^aMann-Whitney U-test was applied. Median and IQR were reported.

^bIndependent t-test was applied. Mean and SD were reported.

Fat to lean mass ratio as a screening tool for metabolic syndrome using receiver operating characteristics curve

When considering the full sample size (n = 238), the best FLMR cut-off point for identifying MetS according to IDF 2007 criteria was 0.441 (Figure 2) with the AUC of 0.874, the sensitivity of 80.0% and specificity of 76.8% (Table 4). Since no male subject was diagnosed with MetS in this study, a subgroup analysis of the FLMR cut-off of females was performed and was reported as 0.443 which is almost similar to the value among all samples (Table 5).

Table 4.

Predictive properties of fat-to-lean mass ratio in identifying high risk of metabolic syndrome (n = 238).

Parameter	Value
Area under curve	0.874
95% confidence interval (95% CI)	0.825, 0.913
Cut-off value	>0.441
Sensitivity (95% CI)	80.0
Specificity (95% CI)	76.8
p-value	<.001

Figure 2.

Receiver operating characteristics curve analysis of fat-to-lean mass ratio in predicting metabolic syndrome (n = 238).

Table 5.

Predictive properties of fat-to-lean mass ratio in identifying high risk of metabolic syndrome among female (n = 188).

Parameter	Value
Area under curve	0.843
95% confidence interval (95% CI)	0.690, 0.995
Cut-off value	>0.443
Sensitivity (95% CI)	80.0
Specificity (95% CI)	71.0
p-value	.009

Association between fat to lean mass ratio with metabolic syndrome and its components

The cut-off value obtained from the ROC curve analysis was used to categorize the subjects into two groups which were the “low risk” and “high risk” groups. From Table 6, it was observed that a bigger proportion of participants with FLMR of more than 0.441 had MetS (6.9%) and its metabolic components including high systolic blood pressure (10.3%), high diastolic blood pressure (12.1%), hypertension (15.5%), high WC (32.8%), high fasting blood glucose (3.4%), high triglycerides (5.2%) and low HDL-c level (34.5%) compared to their counterparts with FLMR lower than 0.441. However, only MetS (p = .022), high WC and low HDL-c revealed a significant association with a high risk of FLMR (p < .001). Further analysis revealed that adolescents with high WC, low HDL-c and MetS had higher odds of developing increased FLMR than the cut-off value with an odds ratio (OR) of 43.4 (95% CI: 9.7,193.9), 4.7 (95% CI: 2.3,9.8) and 13.3 (95% CI: 1.5,121.2), respectively.

Table 6.

Association between fat versus lean mass ratio (FLMR) with MetS and its components.

	FLMR> 0.441* (n=58)n (%)	FLMR≤ 0.441(n=180)n (%)	B	S.E.	OR (95% CI)	p-value^a
MetS Presence Absence	4 (6.9 %)54 (93.1 %)	1 (0.6 %)179 (99.4 %)	2.585	1.129	13.259 (1.451,121.150)	0.022
High Systolic Blood Pressure Present Absence	6 (10.3 %)52 (89.7 %)	11 (6.1 %)169 (93.9 %)	0.573	0.532	1.773 (0.625, 5.026)	0.282
High Diastolic Blood Pressure Present Absence	7 (12.1 %)51 (87.9 %)	12 (6.7 %)168 (93.9%)	1.986	0.403	1.922 (0.719, 5.137)	0.193
Hypertension Present Absence	9 (15.5%)49 (84.5%)	18 (10.0%)162 (90.0%)	0.503	0.440	1.653 (0.698, 3.913)	0.253
High Waist Circumference Present Absence	19 (32.8%)39 (67.2%)	2 (1.1%)178 (98.9%)	3.770	0.764	43.359 (9.697,193.868)	< 0.001
High Fasting Blood Glucose Present Absence	2 (3.4%)56 (96.6%)	1 (0.6 %)179 (99.4 %)	1.855	1.234	6.393 (0.569, 71.831)	0.133
High Triglycerides Present Absence	3 (5.2 %)55 (94.8 %)	3 (1.7 %)177 (98.3%)	1.169	0.831	3.218 (0.631, 16.403)	0.157
Low HDL-cholesterol Present Absence	20 (34.5%)38 (65.5%)	18 (10.0%)162 (90.0%)	1.555	0.372	4.737 (2.287, 9.812)	< 0.001

^aBinary logistic regression was applied using Enter method.

^*value was obtained from ROC curve analysis.

^**FLMR as independent variable, metabolic syndrome components as dependent variables.

B: Beta Coefficient; S.E: Standard Error; CI: Confidence Interval; OR: Odd Ratio.

Discussion

This study showed that 2.1% of the adolescents had MetS while 97.9% of them did not have MetS. This study showed a lower prevalence of MetS compared with data from Asia and Western countries which ranges from 2.4–10.1%.^16,27–34 This inconsistency may be due to the difference in the definition used in diagnosing MetS as some of the studies applied the NCEP-ATP III³⁵ definition while some of them applied the IDF criteria which are similar to our study. Another possible explanation could be due to the age difference. In this study, subjects aged between 18 and 19-year-old where most of them are within the late puberty or post-puberty stage while the subjects in previous studies had a wider age range ranging from 7 to 19 years old. In addition, the youngest age of students enrolled in the university programme is at age 18, as such, the researchers were not able to reach the younger population which possibly skewed the final results. Several studies have demonstrated that some of the prevalence of cardiometabolic risk factors increased significantly during pre-puberty and puberty stage^36–38 as adolescents entering puberty stage is associated with drastic changes in sex hormones secretion and a rapid increase in total body fat as well as BMI.^37,39 Hence, it can interfere as previous studies involved subjects of younger age that are more prone to have MetS compared to this study. In addition to that, the finding of the current study is contrary to the previous research as none of the male subjects had MetS while all the MetS cases come from female subjects as our sample size for males is relatively low compared to the previous studies.

It was hypothesized that FLMR can be employed as an anthropometric tool in predicting the risk of MetS among Malaysian adolescents. This hypothesis was confirmed in our current study where the ROC curve analyses demonstrated that FLMR possessed fair discriminatory capacity as shown by the AUC value (>0.8). In addition to that, our study also confirmed the hypothesis that there was a significant association between FLMR with MetS and some of its components. The present finding is in agreement with the work of other studies. These studies revealed that FLMR was highly predictive of MetS which can be employed as useful predictors of metabolic risk in early adulthood.^19,20,40,41 Interestingly, when compared with other international studies, it was found that the FLMR value obtained from the current study was higher than the cut-off value among males but lower than the value among females.^19,20,41 The discrepancy could be attributed to the cut-off value developed in this study representing the whole sample size regardless of gender. As such, it is urged that study involving a larger sample size to be carried out in the future in order to develop a gender-specific cut-off value as female typically presents with higher body fat compared to male.⁴² Besides that, due to the recruitment issues, our study only managed to include adolescents aged 18–19-year-old; which resulted in a higher cut-off value compared to the previous studies with adolescents aged 10–15-year-old.⁴³ It has been demonstrated that FLMR varies across age groups which might be due to the progressive increase of FM as age increases.⁴⁴

One of the more significant findings to emerge from this study is that adolescents who have FLMR higher than the threshold will have 13.26 times of odds of having MetS. This statistic raised an important theoretical issue about the mechanism of FLMR that have a bearing on MetS among adolescents. FM and LBM are considered two contrasting traits in maintaining homeostasis of the body. FM is characterized as “metabolic load” since it is influenced by behavioural changes such as dietary intake, sedentary behaviour and high level of adiposity which will burden or disrupt the body’s natural homeostasis mechanism. On the other hand, LBM which is made up of organs, tissues and muscles is resembled as “metabolic capacity” as all of these components are responsible for maintaining normal homeostasis processes such as regulating glucose and lipid metabolism in the body.^10,45,46 Our study has successfully translated this load-capacity model into a specific predictive model for MetS by developing a specific cut-off value for FLMR. This cut-off value acts as an indicator in categorizing individuals into different body phenotypes in the way that adolescents with higher FLMR are considered to have a higher risk of having MetS. A high FLMR means the individual has a high percentage of body fat while less amount of LBM. Muscle mass or known as bone-free lean tissue is the largest component of LBM. It plays a vital role in regulating resting energy expenditure (REE) and also muscle glucose uptake.⁴⁶ In another word, an individual with low muscle mass will have lower REE which increases the risk of having obesity since REE is the largest component of energy expenditure.^47,48 At the same time, a decrease in muscle mass will decrease the normal rate of glucose uptake due to insulin resistance and leads to hyperglycaemia.^46,48 Since obesity and insulin resistance are the main culprits in the pathophysiology pathway of MetS,⁴⁹ it is not surprising to find that an individual with higher FLMR will have a higher risk of having MetS and its components.

Besides that, the values of FM and LBM can be easily and accurately obtained from the BIA machine, suggesting that FLMR can be used as a screening tool for MetS without performing a full physical examination.²⁰ The BIA machine used in this study is a segmental multi-frequency BIA (SMF-BIA) which measures impedance at five segments of the body. The accuracy of this tool in assessing body composition has been validated against dual-energy X-ray absorptiometry which is well known as a gold standard in measuring body composition.⁵⁰ Our study suggested that this safe, non-invasive and simple method can be used by healthcare professionals to determine the nutritional status of an individual.⁵¹

The main limitation of this study is we are unable to develop a gender-specific cut-off value for FLMR among adolescents. As the prevalence of the MetS is 0% among male subjects, the ROC analysis cannot be carried out. In addition to that, our subjects are university students and have a narrowed age range. As such, the cut-off value might not be generalized and represented to the whole adolescents population. Thirdly, there are some very high odds ratios (OR) and 95% Confidence Interval (CI). This is because, following the principles of sampling theory, the width of the CI increases when the sample size decreases. The decrease in precision (high CI) is due to the variance of a statistic which is expressed by its standard error.⁵² Besides that, the occurrence of outliers further increases the width of the CI. Hence, it is very difficult to draw meaningful inferences from the results of this study alone due to the width CI.

Nevertheless, this study has its strength. This is the first study conducted among Malaysian adolescents to propose a cut-off value for FLMR in predicting MetS. The cut-off value is of great significance in providing public health professionals in predicting MetS among adolescents. It acts as an indicator in categorizing individuals into different body phenotypes. Furthermore, this is also the first study to investigate the association between FLMR with MetS among Malaysian adolescents. Another significance of this study is it has identified the odds posed by adolescents with metabolic abnormalities such as abdominal obesity, dyslipidaemia and hypertension in developing FLMR higher than threshold values.

In summary, the findings from this cross-sectional study demonstrated that FLMR had a good discriminatory ability in identifying MetS among adolescents and can be used as a predictive tool in the future. This study also highlighted the high prevalence of MetS components among the adolescents with FLMR higher than the threshold value. Hence, intervention strategies such as nutrition educational talks, health screening and also booth exhibitions should be carried out in universities frequently to provide health information to the students thus preventing the progression of the risk factors among adolescents.

Footnotes

Acknowledgements

The authors would like to thank all subjects involved for their participation and cooperation during the course of this study. The effort of the researchers, data collection team, and all those involved in this project are also acknowledged.

Author contributions

J.Y.L was involved in data collection, data analysis and drafting the manuscript; B.S.W chief investigator of the research grant, development of main study design and research protocol, supervision of data collection, interpretation of results and editing of manuscript; A.M.M.J, M.R.S, M.M and P.L.L involved in manuscript editing. All authors involved in manuscript revision and have accepted responsibility for the entire content of this manuscript and approved its submission.

Data availability

The data sets generated and/or analysed during the current study are available from the corresponding author.

Ethical approval

Ethics approval was obtained from UniSZA Human Research and Ethics Committee (UHREC) with a study protocol UniSZA/UHREC/2018/64.

Informed consent

Written informed consent was obtained from all subjects before the study.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was funded by UniSZA Research Project Grant (UniSZA/2017/DPU43, Project Code: R0018 – R347).

ORCID iDs

Jing Ying Lim

Bee Suan Wee

References

Alberti

KGMM

Eckel

Grundy

, et al. Harmonizing the metabolic syndrome: a joint interim statement of the international diabetes federation task force on epidemiology and prevention; national heart, lung, and blood institute; American heart association; world heart federation; international atherosclerosis society; and international association for the study of obesity. Circulation 2009; 120: 1640–1645.

Zimmet

Alberti

KGM

Kaufman

, et al. The metabolic syndrome in children and adolescents ? an IDF consensus report. Pediatr Diabetes 2007; 8: 299–306.

Ahmad

Zulaily

Abdul Manan

, et al. Body weight status of school adolescents in Terengganu, Malaysia: a population baseline study. BMC Public Health 2017; 17: 1–12.

Fadzlina

Harun

Nurul Haniza

, et al. Metabolic syndrome among 13 year old adolescents: prevalence and risk factors. BMC Public Health 2014; 14: S7.

Lai

Mohd Sidik

Lekhraj

, et al. Prevalence and predictors of overweight and obesity among adolescents in Seremban, Negeri Sembilan, Malaysia. Cureus 2022; 14: e21795.

Wee

Poh

Bulgiba

, et al. Risk of metabolic syndrome among children living in metropolitan Kuala Lumpur : a case control study. BMC Public Health 2011; 11: 0–6.

Tung

SEH

Mohd Nasir

Chin

, et al. Insulin resistance, inflammation and metabolic syndrome in normal weight and overweight/obese primary school children in Kuala Lumpur. Malays J Nutr 2018; 24: 153–161.

Wells

JCK

Fewtrell

. Measuring body composition. Arch Dis Child 2006; 91: 612–617.

Muller

Bosy-Westphal

Later

, et al. Functional body composition: insights into the regulation of energy metabolism and some clinical applications. Eur J Clin Nutr 2009; 63: 1045–1056.

10.

Morales

Bucarey

Espinosa

. Muscle lipid metabolism: role of lipid droplets and perilipins. J Diabetes Res 2017; 2017: 1–10.

11.

Kelishadi

Mirmoghtadaee

Najafi

, et al. Systematic review on the association of abdominal obesity in children and adolescents with cardio-metabolic risk factors. J Res Med Sci 2015; 20: 294–307.

12.

Gracia-Marco

Moreno

Ruiz

, et al. Body composition indices and single and clustered cardiovascular disease risk factors in adolescents: providing clinical-based cut-points. Prog Cardiovasc Dis 2016; 58: 555–564.

13.

Liu

Lou

, et al. The utility of fat mass index vs. body mass index and percentage of body fat in the screening of metabolic syndrome. BMC Public Health 2013; 13: 629.

14.

Sardinha

Santos

Silva

, et al. A comparison between BMI, waist circumference, and waist-to-height ratio for identifying cardio-metabolic risk in children and adolescents. PLoS One 2016; 11: e0149351.

15.

Wee

Poh

Bulgiba

, et al. Use of ROC curves in identifying suitable obesity indices for prediction of metabolic syndrome risk clustering among young adolescents in Kuala Lumpur. Obes Rev 2014; 15: 53–97.

16.

Weber

Leonard

Shults

, et al. A comparison of fat and lean body mass index to BMI for the identification of metabolic syndrome in children and adolescents. J Clin Endocrinol Metab 2014; 99: 3208–3216.

17.

Liu

, et al. Relation between metabolic syndrome and body compositions among Chinese adolescents and adults from a large-scale population survey. BMC Public Health 2017; 17: 337.

18.

Chung

Park

, et al. Waist-to-height ratio as an index for cardiometabolic risk in adolescents : results from the 1998 − 2008 KNHANES. Yonsei Med J 2016; 57: 658–663.

19.

Juan

ZHUH

Shi

, et al. Fat-to-muscle ratio: a new anthropometric indicator for predicting metabolic syndrome in the Han and Bouyei populations from Guizhou Province, China. Biomed Environ Sci 2018; 31: 261–271.

20.

Ramírez-Vélez

Carrillo

Correa-Bautista

, et al. Fat-to-muscle ratio: a new anthropometric indicator as a screening tool for metabolic syndrome in young colombian people. Nutrients 2018; 8: 1027.

21.

Ezeh

Pall

Mathur

, et al. Association of fat to lean mass ratio with metabolic dysfunction in women with polycystic ovary syndrome. Hum Reprod 2014; 29: 1508–1517.

22.

Lee

Nieman

. Nutritional assessment. 6th ed. New York, NY: McGraw-Hill US Higher ED, 2012.

23.

de Onis

Onyango

Borghi

, et al. Development of a WHO growth reference for school-aged children and adolescents. Bull World Health Organ 2007; 85(9): 660–667.

24.

Centres for Disease Control and Prevention . National health and nutrition examinatory survey: anthropometry procedures manual. Atlanta, GA: CDC, 2007.

25.

National Institute of Health Research . NIHR Southampton biomedical research centre procedure for measuring bioelectrical impedance using the Bodystat 1500. Southampton, UK: NIHR Southampton Biomedical Research Centre, 2016.

26.

Pickering

Stevens

. How to measure and record blood pressure. Community Eye Heal J 2013; 26: 76.

27.

Song

Chang

, et al. Prevalence and correlates of metabolic syndrome in nutrition survey: the China health and nutrition survey. Nutrients 2017; 9: 79.

28.

Wee

Poh

Bulgiba

, et al. Adiposity and insulin resistance among young Malaysian adolescents. Obes Rev 2016: 132.

29.

Lee

Park

. Relation between sleep duration, overweight, and metabolic syndrome in Korean adolescents. Nutr Metab Cardiovasc Dis 2014; 24: 65–71.

30.

Lee

Gurka

Deboer

. Trends in metabolic syndrome severity and lifestyle factors among adolescents. Pediatrics 2016; 137: e20153177.

31.

Miller

Kaylor

Johannsson

, et al. Prevalence of metabolic syndrome and individual criterion in US adolescents. Metab Syndr Relat Disord 2014; 12: 527–532.

32.

Ahmadi

Sadr

Mohammadi

, et al. Prevalence of abdominal obesity and metabolic syndrome in children and adolescents: a community based cross-sectional study. Iran J Public Health 2020; 49: 360–368.

33.

Mahajan

Kshatriya

. Prevalence of metabolic syndrome and associated risk factors among tribal adolescents of Gujarat. Diabetes Metab Syndr 2020; 14: 995–999.

34.

Dejavitte

RAS

Enes

Nucci

. Prevalence of metabolic syndrome and its associated factors in overweight and obese adolescents. J Pediatr Endocrinol Metab 2020; 33: 233–239.

35.

National Cholesterol education program (NCEP) expert panel on detection, evaluation, and treatment of high blood cholesterol in adults (adult treatment panel III) . Third report of the national cholesterol education program (NCEP) expert panel on detection. Circulation 2002; 106: 3143–3421.

36.

Ramírez-Vélez

García-Hermoso

Agostinis-Sobrinho

, et al. Pubertal stage, body mass index, and cardiometabolic risk in children and adolescents in Bogotá, Colombia: the cross-sectional fuprecol study. Nutrients 2017; 9: 644.

37.

Chan

NPT

Choi

Nelson

EAS

, et al. Associations of pubertal stage and body mass index with cardiometabolic risk in Hong Kong Chinese children: a cross-sectional study. BMC Pediatr 2015; 15: 136.

38.

Reinehr

. Metabolic syndrome in children and adolescents: a critical approach considering the interaction between pubertal stage and insulin resistance. Curr Diab Rep 2016; 16: 8.

39.

Siervogel

Demerath

Schubert

, et al. Puberty and body composition. Horm Res 2003; 60: 36–45.

40.

Park

Kim

. Validity of muscle-to-fat ratio as a predictor of adult metabolic syndrome. J Phys Ther Sci 2016; 28: 1036–1045.

41.

Seo

Song

. Fat-to-muscle ratio as a predictor of insulin resistance and metabolic syndrome in Korean adults. J Cachexia Sarcopenia Muscle 2020; 11: 710–725.

42.

Karastergiou

Smith

Greenberg

, et al. Sex differences in human adipose tissues - the biology of pear shape. Biol Sex Differ 2012; 3: 13.

43.

Gamboa-Gómez

Guerrero-Romero

Aradillas-García

, et al. The fat-to-lean mass ratio is associated with hyperinsulinemia in healthy Mexican adolescents. J Am Coll Nutr 2021; 40: 219–223.

44.

Xiao

Purcell

Prado

, et al. Fat mass to fat-free mass ratio reference values from NHANES III using bioelectrical impedance analysis. Clin Nutr 2018; 37: 2284–2287.

45.

Siervo

Prado

Mire

, et al. Body composition indices of a load–capacity model: gender- and BMI-specific reference curves. Public Health Nutr 2014; 18: 1245–1254.

46.

Evans

Mcmillin

Weyrauch

, et al. Regulation of skeletal muscle glucose transport and glucose metabolism by exercise training. Nutrients 2019; 1–24: 2432.

47.

Fonseca

Sala

Ferrerira

BAM

, et al. Body weight control and energy expenditure. Clin Nutr Exp 2018; 20: 55–59.

48.

Wolfe

. The underappreciated role of muscle in health and disease. Amerian J Clin Nutr 2006; 84: 475–482.

49.

Jung

Choi

M-S

. Obesity and its metabolic complications: the role of adipokines and the relationship between obesity, inflammation, insulin resistance, dyslipidemia and nonalcoholic fatty liver disease. Int J Mol Sci 2014; 15: 6184–6223.

50.

Ling

CHY

de Craen

AJM

Slagboom

, et al. Accuracy of direct segmental multi-frequency bioimpedance analysis in the assessment of total body and segmental body composition in middle-aged adult population Accuracy of direct segmental multi-frequency bioimpedance analysis in the assessment of total. Clin Nutr 2011; 30: 610–615.

51.

Kyle

Earthman

Pichard

, et al. Body composition during growth in children: limitations and perspectives of bioelectrical impedance analysis. Eur J Clin Nutr 2015; 69: 1298–1305.

52.

Sim

Reid

. Statistical inference by confidence intervals: issues of interpretation and utilization. Phys Ther 1999; 79: 186–195.

Fat-to-lean mass ratio as a new anthropometric indicator in predicting metabolic syndrome among Malay adolescents in Terengganu,Malaysia

Abstract

Background

Objectives

Methods

Results

Conclusion

Keywords

Introduction

Methods

Study design and sample selection

Sampling method

Data collection

Measurement of variables

Socio-demographic information

Anthropometric measurement

Body composition measurement

Blood pressure measurement

Biochemical measurement

Diagnostic criteria

Statistical analysis

Results

Anthropometric measurements

Metabolic syndrome components

Fat to lean mass ratio as a screening tool for metabolic syndrome using receiver operating characteristics curve

Association between fat to lean mass ratio with metabolic syndrome and its components

Discussion

Footnotes

Acknowledgements

Author contributions

Data availability

Ethical approval

Informed consent

Declaration of conflicting interests

Funding

ORCID iDs

References