Abstract
We analyzed the discriminatory capacity of anthropometric indicators for body fat in children and adolescents. This systematic review and meta-analysis included cross-sectional and clinical studies comprising children and adolescents aged 2–19 y that tested the discriminatory value for body fat measured by anthropometric methods or indexes generated by anthropometric variables compared with precision methods in the diagnosis of body fat [dual-energy X-ray absorptiometry (DXA), computed tomography, air displacement plethysmography (ADP), or MRI]. Five studies met the eligibility criteria and presented high methodologic quality. The anthropometric indicators that had high discriminatory power to identify high body fat were body mass index (BMI) in males [area under the curve (AUC): 0.975] and females (AUC: 0.947), waist circumference (WC) in males (AUC: 0.975) and females (AUC: 0.959), and the waist-to-height ratio (WTHR) in males (AUC: 0.897) and females (AUC: 0.914). BMI, WC, and WTHR can be used by health professionals to assess body fat in children and adolescents.
Keywords: body composition, anthropometry, children, adolescents, HIV
Introduction
Excess body fat is a public health problem and is considered an independent risk factor for several noncommunicable chronic diseases, such as insulin resistance, type II diabetes, high blood pressure, and metabolic syndrome (1). WHO (2) showed that 43 million children and adolescents worldwide in 2010 were obese, 35 million of whom lived in middle-income countries. The prevalence of obesity in adolescents increased by 2.5% in 20 y (from 4.2% in 1990 to 6.7% in 2010), and a prevalence of 9.1% is expected by 2020 (2).
Children and adolescents with excess body fat are more likely to develop structural anatomic changes (postural deviations), increased heart workload (hypertrophy and cardiac arrhythmia), changes in pulmonary functions (airway obstruction and apnoea), endocrine disorders (insulin resistance, increased cortisol, and reduced growth hormone), and immunologic disorders (increased production of cytokines) (3). In addition, they present psychosocial difficulties, including reduced quality of life, anxiety, depression, and increased risk for the development of eating disorders (4).
The early identification of excess body fat in the pediatric population is essential for preventing other chronic diseases in adult life (5). Techniques with high accuracy for estimating body fat such as DXA, air displacement plethysmography (ADP), computed tomography, and MRI are operationally costly and require expensive training (5, 6). Thus, alternative methods such as anthropometric indicators that discriminate body fat with low operational costs are necessary in clinical practice (5).
Systematic reviews have examined the discriminatory capacity of anthropometric indicators for body fat (6, 7). However, some limitations do not allow greater generalizations regarding the theme—for example, an analysis of the accuracy of ≤3 anthropometric indicators in such reviews [perimeter of neck, waist-to-height ratio (WTHR), and waist circumference (WC)]. This limitation does not allow the verification of other possible anthropometric indicators that may be used in clinical practice with high discriminatory power for body fat. Another limitation is that these reviews compared the anthropometric indicators with reference techniques for body fat that are less accurate than DXA, ADP, computed tomography, and MRI, such as skinfold equations for estimating fat percentage, BMI, and electrical bioimpedance (BIA) (6, 7). In this sense, systematic reviews with meta-analyses that include studies with accurate reference methods are necessary. In addition, to our knowledge, no meta-analyses addressing the discriminatory capacity for body fat of a range of anthropometric indicators in pediatric populations have been reported. Therefore, we analyzed the discriminatory capacity of anthropometric indicators for body fat in children and adolescents.
Methods
The method used in this systematic review and meta-analysis was consistent with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses statement (8).
Search strategy
The systematic search was performed in the Lilacs, Embase, Google Scholar, Pubmed, Web of Science, Scopus, and SPORTDiscus databases with the use of Boolean operators AND, and OR, parentheses, quotation marks, and asterisks. The term AND was used for the purpose of aggregating ≥1 word from each group. The term OR was used to relate ≥1 word from each block. Parentheses were used to match search terms by outcome, exposure, and population categories. Quotation marks were used to search for exact terms or expressions. Asterisks were used to search for all words derived from the same prefix. Descriptors came from Health Sciences Descriptors, Medical Subject Headings, and words related to the subject. The groups of descriptors of the search strategy were the following: outcome of interest (fats OR fatty OR “fat body” OR “fat mass” OR adiposity OR “body composition” OR “body fat” OR “lean mass” OR “free-fat mass” OR “free fat mass” OR “muscle mass”); reference analysis methods for quantifying body fat (DXA OR “dual X-ray absorptiometry” OR “absorptiometry, photon” OR plethysmography OR “magnetic resonance” OR tomography); anthropometric indicators for body fat (diagnostic method to be tested) (anthropometry OR “skinfold thickness” OR skinfold OR “skinfold calf” OR circumference OR perimeter OR weight OR “conicity index” OR “body adiposity index” OR “body mass index” OR BMI OR “Quetelet index” OR “waist-height-ratio”); and population analyzed (child * OR adolescent OR adolescence OR youth OR teen OR teenager OR scholar). The search did not utilize a year limit or filters, and articles published from 30 July to 31 August 2016 were searched. The search results in each database were exported to the reference EndNote version 7 (Thomson Reuters).
Selection criteria
Studies that had unclear titles or lacked abstracts were read in their entirety. Eligibility criteria were that the article 1) be original, 2) have a cross-sectional and/or clinical design, 3) include children and adolescents aged 2–19 y, and 4) have tested the discriminatory value for body fat measured by anthropometry methods or indexes generated with the use of anthropometric variables compared with precision methods in the diagnosis of body fat (DXA or computed tomography or ADP or MRI). Duplicate articles, review articles, dissertations, abstracts, book chapters, viewpoints and expert opinions, monographs, theses, abstracts, chapters, articles in which the population and sample evaluated were composed only of individuals with some morbidity, articles that did not address anthropometric indicators such as body fat discriminators, and studies with athletes were excluded. The reference list of included articles was read.
Data extraction
Data from eligible articles were extracted independently by 2 reviewers. In the case of divergence, a third reviewer was consulted. Data extracted were the following: author names and year published; methodologic quality scores; locations; ages; populations and samples; test and reference methods; discriminatory capacities; accuracies of discriminations; AUCs, 95% CIs, and SEs; sensitivities and specificities; and suggested cutoffs.
Quality assessment
The quality assessment of studies was carried out with the use of a tool that assesses the quality of diagnostic accuracy studies on a scale from 0 to 14 points. The assessment takes into consideration factors that determine the validity of studies, specifically the quality assessment of diagnostic accuracy tests (9). Studies classified as high quality were those with scores ≥10, and studies classified as low quality were those with scores <10 (9). Based on the instrument used (9), the following information was used to indicate the methodologic quality of the studies: standardization of the measurement of the anthropometric indicators; use of reference methods for body fat; description of the time between the anthropometric measures and between the measurement of percentage of fat; and detailed instructions provided to subjects on diet and exercise before body composition measurements (9).
Statistical analysis
For the meta-analysis, studies with AUCs, 95% CIs, and SEs were included. Because of the heterogeneous nature of studies regarding the different categories of classification of anthropometric indicators for body fat (10–12) and the variations in the cutoffs adopted, each anthropometric indicator was grouped in a single measurement for each study for later inclusion in the meta-analysis. For this grouping, AUCs or SEs were used. For studies that did not present SEs, it was necessary to calculate this measure (SE = upper limit of 95% CI − AUC/1.96) (13). In addition, the random-effects model was used because of the heterogeneity of studies (13). For the classification of anthropometric indicators in relation to the discriminatory power by the AUC curve, the values proposed by Swets (14) were used, with ≤0.5 considered without discriminatory power, >0.5 and ≤0.7 with low discriminatory power, >0.7 and ≤0.9 with excellent discriminatory power, and 1 a perfect test. Medcalc version 15.2 was used for the statistical analyses. P < 0.05 was considered significant.
Results
Selection of studies
A total of 6211 articles were found. Of these, 2425 were duplicates, resulting in 3786 articles. After reading the titles and abstracts, 3728 studies were excluded, and 58 studies were read in full. Four studies were included because they presented eligibility criteria, and after reading the references 1 additional study was included, resulting in 5 articles. The selection process of articles is presented in Figure 1.
FIGURE 1.
Search, selection, and exclusion of articles.
Characteristics of studies
Five studies were included in this review (Table 1) (10–12, 15, 16). Studies were published between 1999 and 2011, 2 of which were carried out in Europe (11, 12), 2 in Oceania (15, 16), and 1 in Asia (10). The children and adolescents were aged 3–19 y. The number of subjects surveyed in the studies ranged from 328 to 778. In all studies, the information was stratified by sex. Studies tested the accuracy of the anthropometric indicators tricipital skinfold (TSF), conicity index, BMI, perimeter of the relaxed arm (PRA), WC, WTHR, and waist-to-hip ratio (WHR) as discriminators of high body fat without a diagnosis of diseases. All studies reported AUCs and 95% CIs. Four studies presented sensitivity and specificity values (10, 11, 15, 16), and only one study did not present such values (12).
TABLE 1.
Studies on anthropometric indicators with discriminatory capacity of body fat in children and adolescents1
| References | Country | Quality score | Age, y | n | Test method | Reference method | Test to verify discriminatory capacity | Accuracy of discrimination [AUC (95% CI)] | Sensibility | Specificity | Suggested cutoffs |
| Fujita et al. (10) | Japan | 12 | 9–11 | 226 males; 196 females | BMI (in kg/m2) | DXA | AUC | BMI♂ 85th percentile[0.980 (0.964, 0.997)] | 0.94 | 0.93 | 18.6 |
| BMI♂ 90th percentile[0.982 (0.966, 0.998)] | 1.00 | 0.90 | 18.7 | ||||||||
| BMI♂ 95th percentile[0.983 (0.968, 0.999)] | 1.00 | 0.96 | 20.8 | ||||||||
| BMI♀ 85th percentile[0.924 (0.877, 0.971)] | 0.97 | 0.73 | 17.0 | ||||||||
| BMI♀ 90th percentile[0.959 (0.927, 0.991)] | 0.95 | 0.85 | 17.8 | ||||||||
| BMI♀ 95th percentile[0.981 (0.961, 1.001)] | 1.00 | 0.92 | 19.6 | ||||||||
| WC | DXA | AUC | WC♂ 85th percentile[0.980 (0.964, 0.996)] | 0.91 | 0.95 | 68.5 cm | |||||
| WC♂ 90th percentile[0.980 (0.964, 0.997)] | 1.00 | 0.91 | 68.5 cm | ||||||||
| WC♂ 95th percentile[0.987 (0.974, 1.001)] | 1.00 | 0.97 | 76.5 cm | ||||||||
| WC♀ 85th percentile[0.932 (0.882, 0.982)] | 0.80 | 0.93 | 68.0 cm | ||||||||
| WC♀ 90th percentile[0.976 (0.952, 1.000)] | 0.95 | 0.93 | 69.0 cm | ||||||||
| WC♀ 95th percentile[0.986 (0.971, 1.001)] | 1.00 | 0.92 | 73.0 cm | ||||||||
| WTHR | DXA | AUC | WTHR♂ 85th percentile[0.987 (0.976, 0.999)] | 1.00 | 0.90 | 0.467 | |||||
| WTHR♂ 90th percentile[0.986 (0.973, 0.999)] | 1.00 | 0.90 | 0.488 | ||||||||
| WTHR♂ 95th percentile[0.981 (0.964, 0.998)] | 1.00 | 0.95 | 0.519 | ||||||||
| WTHR♀ 85th percentile[0.951 (0.905, 0.995)] | 0.93 | 0.89 | 0.460 | ||||||||
| WTHR♀ 90th percentile[0.981 (0.963, 0.998)] | 1.00 | 0.87 | 0.460 | ||||||||
| WTHR♀ 95th percentile[0.992 (0.981, 1.004)] | 1.00 | 0.95 | 0.499 | ||||||||
| Neovius et al. (11) | Sweden | 11 | 17 | 202 males; 279 females | BMI | ADP | AUC | BMI♂ for overweight[0.97 (0.94, 1.00)] | 0.92 | 0.93 | 23.9 |
| BMI♂ for obesity[0.97 (0.94, 1.00)] | 0.90 | 0.90 | 24.6 | ||||||||
| BMI♀ for overweight[0.84 (0.80, 0.89)] | 0.77 | 0.77 | 21.3 | ||||||||
| BMI♀ for obesity[0.99 (0.97, 1.00)] | 0.93 | 0.95 | 25.1 | ||||||||
| WC | ADP | AUC | WC♂ for overweight[0.94 (0.89, 1.00)] | 0.84 | 0.88 | 79.5 cm | |||||
| WC♂ for obesity[0.96 (0.92, 1.00)] | 0.93 | 0.95 | 83.5 cm | ||||||||
| WC♀ for overweight[0.94 (0.89, 1.00)] | 0.84 | 0.88 | 70.5 cm | ||||||||
| WC♀ for obesity[0.95 (0.92, 0.99)] | 0.93 | 0.89 | 78.0 cm | ||||||||
| WHR | ADP | AUC | WHR♂ for overweight[0.79 (0.69, 0.89)] | 0.72 | 0.72 | 0.82 | |||||
| WHR♂ for obesity[0.80 (0.66, 0.94)] | 0.80 | 0.71 | 0.83 | ||||||||
| WHR♀ for overweight[0.63 (0.56, 0.70)] | 0.56 | 0.56 | 0.77 | ||||||||
| WHR♀ for obesity [0.65 (0.49, 0.80)] | 0.64 | 0.61 | 0.78 | ||||||||
| Sardinha et al. (12) | Portugal | 11 | 10–15 | 165 males; 163 females | BMI | DXA | AUC | BMI♂ 10–11 y[0.97 (0.88, 0.99)] | 0.96 | 0.14 | 19.0 |
| BMI♂ 12–13 y[0.86 (0.73, 0.94)] | 0.86 | 0.24 | 19.4 | ||||||||
| BMI♂ 14–15 y[0.61 (0.47, 0.72)] | 0.50 | 0.08 | 24.0 | ||||||||
| BMI♀ 10–11 y[0.95 (0.32, 0.99)] | 0.83 | 0.13 | 19.6 | ||||||||
| BMI♀ 12–13 y[0.89 (0.79, 0.96)] | 0.67 | 0.03 | 21.2 | ||||||||
| BMI♀ 14–15 y[0.90 (0.78, 0.98)] | 0.77 | 0.11 | 21.9 | ||||||||
| TSF | DXA | AUC | TSF♂ 10–11 y[0.98 (0.91, 0.99)] | 1.00 | 0.10 | 17.0 mm | |||||
| TSF♂ 12–13 y[0.94 (0.84, 0.98)] | 0.86 | 0.05 | 19.0 mm | ||||||||
| TSF♂ 14–15 y[0.95 (0.84, 0.99)] | 0.63 | 0.26 | 19.0 mm | ||||||||
| TSF♀ 10–11 y[0.96 (0.87, 0.99)] | 0.79 | 0.00 | 21.0 mm | ||||||||
| TSF♀ 12–13 y[0.94 (0.85, 0.98)] | 0.78 | 0.03 | 21.0 mm | ||||||||
| TSF♀ 14–15 y[0.95 (0.84, 0.99)] | 0.86 | 0.07 | 24.0 mm | ||||||||
| PRA | DXA | AUC | PRA♂ 10–11 y[0.97 (0.88, 0.99)] | 0.96 | 0.14 | 22.8 cm | |||||
| PRA♂ 12–13 y[0.85 (0.72, 0.93)] | 0.71 | 0.18 | 25.8 cm | ||||||||
| PRA♂ 14–15 y[0.62 (0.48, 0.74)] | 0.50 | 0.24 | 28.2 cm | ||||||||
| PRA♀ 10–11 y[0.95 (0.86, 0.99)] | 0.83 | 0.03 | 24.2 cm | ||||||||
| PRA♀ 12–13 y[0.94 (0.84, 0.98)] | 0.74 | 0.00 | 25.8 cm | ||||||||
| PRA♀ 14–15 y[0.62 (0.48, 0.74)] | 0.50 | 0.24 | 28.2 cm | ||||||||
| Taylor et al. (15) | New Zealand | 11 | 3–19 | 302 males; 278 females | Conicity index | DXA | AUC | Conicity index♂ for trunk fat[0.81 (0.74, 0.88)] | 0.61 | 0.88 | Conicity index, WC, and WHR varied according to sex and age group |
| Conicity index♀ for trunk fat[0.80 (0.72, 0.86)] | 0.57 | 0.88 | |||||||||
| WC | DXA | AUC | WC♂ for trunk fat[0.97 (0.95, 0.99)] | 0.87 | 0.92 | ||||||
| WC♀ for trunk fat[0.97 (0.95, 0.99)] | 0.89 | 0.94 | |||||||||
| WHR | DXA | AUC | WHR♂ for trunk fat[0.71 (0.62, 0.80)] | 0.46 | 0.85 | ||||||
| WHR♀ for trunk fat[0.73 (0.64, 0.81)] | 0.47 | 0.85 | |||||||||
| Swets (14) | New Zealand | 11 | 5–18 | 382 males; 396 females | WTHR | DXA | AUC | WTHR♂2[0.89 (0.85, 0.94)] | 0.70 | 0.95 | NR |
| WTHR♀2[0.91 (0.87, 0.94)] | 0.65 | 0.92 | |||||||||
| WTHR♂3[0.72 (0.64, 0.79)] | 0.33 | 0.96 | |||||||||
| WTHtR♀3[0.73 (0.66, 0.80)] | 0.47 | 0.78 | |||||||||
| WTHR♂4[0.72 (0.62, 0.83)] | 0.58 | 0.82 | |||||||||
| WTHR♀4[0.80 (0.73, 0.88)] | 0.46 | 0.89 |
The 90th and 95th percentiles were used to define the limits of excess body fat in both sexes (10). ADP, air displacement plethsmography; NR, not reported; PRA, perimeter of relaxed arm; TSF, tricipital skinfold; WC, waist circumference; WHR, waist-to-hip ratio; WTHR, waist-to-height ratio; ♂, male; ♀, female.
Default value of 1 from Taylor et al. (15).
Age- and sex-specific constants were used by Tybor et al. (16) for males and females aged 5–18 y.
Age- and sex-specific constants were used by Nambiar et al. (16) for males and females aged 5–18 y.
The discriminatory capacity of TSF for body fat was presented in one study (12), in which good capacity to discriminate high body fat in children and adolescents aged 10–15 y in both sexes was found, with specific cutoffs by age and sex. The reference method for body fat used was DXA (12). The conicity index was a good body fat discriminator, as presented in one study (15). Cutoffs varied according to sex and age, and the reference method for fat was DXA. The PRA was evaluated as a body fat discriminator in one study (12) and showed a good capacity to discriminate fat in children and adolescents of both sexes without a diagnosis of diseases. The cutoffs were specific for age and sex, with DXA being the reference method used. It was not possible to perform a meta-analysis of TSF, the conicity index, or PRA because only one study evaluated the discriminatory capacity for body fat.
Three studies evaluated the discriminatory capacity of BMI for body fat, and all articles showed that this indicator had good capacity to discriminate body fat in children and adolescents (10–12). Two studies used DXA as a reference method for body fat (10–12), and another study evaluated fat by ADP (11). The cutoffs of studies were distinct and varied according to the age range of the population and the form of classification (percentage distributions) (10–12). The meta-analysis of the studies that analyzed BMI as a body fat discriminator demonstrated that this indicator presents excellent discriminatory power for males (Figure 2, Table 2) and females (Figure 3, Table 2).
FIGURE 2.
Discriminatory power of BMI, WC, WTHR, and WHR for body fat through the AUC in males. WC, waist circumference; WHR, waist-to-hip ratio; WTHR, waist-to-height ratio.
TABLE 2.
Discriminatory properties of anthropometric indicators for body fat1
| Males |
Females |
|||||
| Variables | AUC (95% CI) | Default error | P value | AUC (95% CI) | Default error | P value |
| BMI, kg/m2 | 0.975 (0.960, 0.980) | 0.0007 | <0.001 | 0.947 (0.914, 0.979) | 0.0164 | <0.001 |
| WC, cm | 0.975 (0.960, 0.980) | 0.0007 | <0.001 | 0.959 (0.942, 0.977) | 0.0089 | <0.001 |
| WTHR | 0.897 (0.697, 0.761) | 0.0697 | <0.001 | 0.914 (0.804, 1.000) | 0.0563 | <0.001 |
| WHR | 0.754 (0.673, 0.835) | 0.0414 | <0.001 | 0.675 (0.610, 0.741) | 0.0333 | <0.001 |
WC, waist circumference; WHR, waist-to-hip ratio; WTHR, waist-to-height ratio.
FIGURE 3.
Discriminatory power of BMI, WC, WTHR, and WHR for body fat through the AUC in females. WC, waist circumference; WHR, waist-to-hip ratio; WTHR, waist-to-height ratio.
Three studies evaluated the discriminatory power of WC for body fat (10, 11, 15). Two studies used DXA as a reference method for body fat (10, 15), and 1 study evaluated body fat by ADP (11). WC showed good discriminatory capacity in both sexes. The cutoffs of studies varied according to the age group and the form of classification (percentage distributions) (10–12, 15). The meta-analysis of the studies that analyzed WC as a body fat discriminator demonstrated that this indicator presents excellent discriminatory power for males (Figure 2, Table 2) and females (Figure 3, Table 2).
Two studies evaluated WTHR as a body fat discriminator (10, 16). The studies demonstrated that WTHR was a good fat discriminator in both sexes and presented DXA as the reference criterion for body fat. However, only 1 study listed cutoffs, which varied according to the distribution in percentiles of the population (10). The meta-analysis of studies that analyzed WTHR as a body fat discriminator showed that this indicator presents excellent discriminatory power for males (Figure 2, Table 2) and females (Figure 3, Table 2).
WHR presented moderate and low discriminatory capacity in males and females, respectively, for total (11) and trunk body fat (15). Cutoffs varied according to the overweight and obesity classification (11) and sex (11, 15). The reference criteria for body fat differed between studies: one used DXA (16), whereas the other used ADP (11). The meta-analysis of this anthropometric indicator was not performed because only one study for males and one for females evaluated the discriminatory capacity of WHR.
Quality of studies
The 5 studies presented high methodologic quality (10–12, 15, 16). However, some studies did not accurately report the inclusion criteria of the sample (10, 12); some studies did not describe the procedures for using the reference method. In addition, the studies did not report whether the measurement of anthropometric indicators and fat assessment by reference methods were performed on the same day (11, 12, 15, 16).
Discussion
BMI, WC, and WTHR showed excellent discriminatory power for body fat in children and adolescents of both sexes without a diagnosis of diseases. WHR showed moderate discriminatory power for body fat in males and low discriminatory power in females.
BMI is widely used in studies to classify individuals for obesity because it uses variables that are easy to measure such as body mass and height (17). BMI is highly correlated with body fat. In addition, as body mass (fat mass and fat-free mass) increases, BMI increases (17). This study demonstrated that such an indicator can be used for this purpose. However, the use of BMI should be undertaken with caution in athletes and in individuals with infectious diseases that can change the redistribution of body fat (10–12).
WC is used to verify fat distribution in the central region (7). This region presents high correlations with body fat (18). The studies herein that analyzed WC took the measurement at the midpoint between the last rib and the iliac crest (10, 11, 15). Although there is no consensus as to where such measures should be taken, this study demonstrates that measuring WC as described previously demonstrates an excellent capacity to discriminate body fat.
WTHR is another anthropometric indicator with excellent discriminatory power for body fat in children and adolescents of both sexes. This indicator is used to discriminate total and central fat and cardiovascular risk factors associated with obesity (16, 19). A study carried out from New Zealand showed that the division of WC by height correctly discriminated children and adolescents with high and low concentrations of total and central fat ≥90% of the time (16). One possible justification refers to the period of childhood and adolescence during which there is a greater secretion of growth hormone, which causes an increase in fat mass, fat-free mass, and height (20). The WTHR indicator shows height as one of the variables, which may explain its excellent discriminatory capacity (7).
WHR showed moderate and low discriminatory power in males and females, respectively. WHR is used to identify peripheral and total body fat in children, adolescents, and adults. However, in children and adolescents who have not yet reached the postpubertal stage of sexual maturation, this measure is compromised because of the natural growth and development of young people (21). Because studies included in this review did not distinguish children and adolescents by maturational stage, it could be inferred that the sample heterogeneity does not make WHR a good fat indicator in children and adolescents.
Studies in previous systematic reviews that verified the discriminatory power of anthropometric indicators for body fat showed reference methods considered to be inaccurate for assessing body fat compared with reference methods adopted in the studies analyzed herein. Previous reviews have used BMI and skinfold equations as reference methods for estimating fat percentage and BIA (6, 7). Although there is no consensus on which method would be the most accurate method for identifying high body fat in children and adolescents (11, 15), DXA and ADP techniques are considered more accurate for discriminating body fat than anthropometric indicators and BIA (5).
The small number of studies included in this systematic review and meta-analysis is a limitation of this study. However, some positive aspects are notable: the search strategy for studies employed Portuguese, English, and Spanish; the search was performed in 8 different databases, which allowed for greater comprehensiveness in the systematic search; and the systematic review and meta-analysis did not limit the number of anthropometric indicators analyzed and included only studies that presented accurate reference methods in body fat assessment. In addition, the systematic review and meta-analysis allowed the identification of anthropometric indicators that best discriminated body fat in children and adolescents.
BMI, WC, and WTHR are excellent body fat discriminators in both sexes. These indicators can be used by health professionals to quickly assess body fat in children and adolescents with low operational costs.
Acknowledgments
All authors read and approved the final manuscript.
Footnotes
Abbreviations used: ADP, air displacement plethsmography; BIA, electrical bioimpedance; PRA, perimeter of the relaxed arm; TSF, tricipital skinfold; WC, waist circumference; WHR, waist-to-hip ratio; WTHR, waist-to-height ratio.
References
- 1.He F, Rodriguez-Colon S, Fernandez-Mendoza J, Vgontzas AN, Bixler EO, Berg A, Liao D. Abdominal obesity and metabolic syndrome burden in adolescents—Penn State Children Cohort study. J Clin Densitom 2015;18:30–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.WHO. Non communicable diseases country profiles 2011. Geneva (Switzerland): WHO; 2011. [Google Scholar]
- 3.Eckel N, Meidtner K, Kalle-Uhlmann T, Stefan N, Schulze MB. Metabolically healthy obesity and cardiovascular events: a systematic review and meta-analysis. Eur J Prev Cardiol 2016;23:956–66. [DOI] [PubMed] [Google Scholar]
- 4.Wilson AL, Goldfield GS. Overweight or obese young people are not at increased risk of depression, but young people with depression are at increased risk of obesity. Evid Based Nurs 2014;17:112. [DOI] [PubMed] [Google Scholar]
- 5.Silva DR, Ribeiro AS, Pavão FH, Ronque ER, Avelar A, Silva AM, Cyrino ES. Validade dos métodos para avaliação da gordura corporal em crianças e adolescentes por meio de modelos multicompartimentais: uma revisão sistemática. [Validity of the methods to assess body fat in children and adolescents using multi-compartment models as the reference method: a systematic review]. Rev Assoc Med Bras (1992) 2013;59:475–86 (in Portuguese). [DOI] [PubMed] [Google Scholar]
- 6.dos Santos Cavalcanti CB, do Egito Carvalho SCB, de Barros MVG. Anthropometric indicators of abdominal obesity: review of the papers indexed on SciELO electronic library. Braz J Kinathrop Hum Perform 2009;11:217–5. [Google Scholar]
- 7.Magalhães EIDS, Sant’Ana LFDR, Priore SE, Franceschini SDCC. Waist circumference, waist/height ratio, and neck circumference as parameters of central obesity assessment in children. Rev Paul Pediatr 2014;32:273–81. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Ann Intern Med 2009;151:264–9. [DOI] [PubMed] [Google Scholar]
- 9.Whiting P, Rutjes AW, Reitsma JB, Bossuyt PM, Kleijnen J. The development of QUADAS: a tool for the quality assessment of studies of diagnostic accuracy included in systematic reviews. BMC Med Res Methodol 2003;3:25. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Fujita Y, Kouda K, Nakamura H, Iki M. Cut-off values of body mass index, waist circumference, and waist-to-height ratio to identify excess abdominal fat: population-based screening of Japanese schoolchildren. J Epidemiol 2011;21:191–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Neovius M, Linné Y, Rossner S. BMI, waist-circumference and waist-hip-ratio as diagnostic tests for fatness in adolescents. Int J Obes (Lond) 2005;29:163–9. [DOI] [PubMed] [Google Scholar]
- 12.Sardinha LB, Going SB, Teixeira PJ, Lohman TG. Receiver operating characteristic analysis of body mass index, triceps skinfold thickness, and arm girth for obesity screening in children and adolescents. Am J Clin Nutr 1999;70:1090–5. [DOI] [PubMed] [Google Scholar]
- 13.de Sousa MR, Ribeiro ALP. Revisão sistemática e meta-análise de estudos de diagnóstico e prognóstico: um tutorial. [Systematic review and meta-analysis of diagnostic and prognostic studies: a tutorial]. Arq Bras Cardiol 2009;92:241–51 (in Portuguese). [DOI] [PubMed] [Google Scholar]
- 14.Swets JA. Measuring the accuracy of diagnostic systems. Science 1988;240:1285–93. [DOI] [PubMed] [Google Scholar]
- 15.Taylor RW, Jones IE, Williams SM, Goulding A. Evaluation of waist circumference, waist-to-hip ratio, and the conicity index as screening tools for high trunk fat mass, as measured by dual-energy X-ray absorptiometry, in children aged 3–19 y. Am J Clin Nutr 2000;72:490–5. [DOI] [PubMed] [Google Scholar]
- 16.Taylor RW, Williams SM, Grant AM, Taylor BJ, Goulding A. Predictive ability of waist‐to‐height in relation to adiposity in children is not improved with age and sex‐specific values. Obesity (Silver Spring) 2011;19:1062–8. [DOI] [PubMed] [Google Scholar]
- 17.Javed A, Jumean M, Murad M, Okorodudu D, Kumar S, Somers VK, Lopez‐Jimenez F. Diagnostic performance of body mass index to identify obesity as defined by body adiposity in children and adolescents: a systematic review and meta‐analysis. Pediatr Obes 2015;10:234–44. [DOI] [PubMed] [Google Scholar]
- 18.Gao Y, Xie X, Cianflone K, Lapointe M, Guan J, Bu-jiaer GWB, Ma YT. Ethnic differences in acylation stimulating protein (ASP) in Xinjiang Uygur autonomous region, China. Int J Clin Exp Med 2015;8:2823–30. [PMC free article] [PubMed] [Google Scholar]
- 19.Brannsether B, Roelants M, Bjerknes R, Júlíusson PB. Waist circumference and waist-to-height ratio in Norwegian children 4–18 years of age: reference values and cut-off levels. Acta Paediatr 2011;100:1576–82. [DOI] [PubMed] [Google Scholar]
- 20.Shulman DI. Metabolic effects of growth hormone in the child and adolescent. Curr Opin Pediatr 2002;14:432–6. [DOI] [PubMed] [Google Scholar]
- 21.Malina RM, Bouchard C. Crescimento, maturação e atividade física. [Growth, maturation and physical activity.] São Paulo (Brazil): Phorte; 2009. [Google Scholar]