SciELO - Scientific Electronic Library Online

vol.92 issue1Hospitalizations and deaths due to pertussis in children from 1996 to 2013☆The mediator role of psychological morbidity on sleep and health behaviors in adolescents author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand




Related links


Jornal de Pediatria

Print version ISSN 0021-7557On-line version ISSN 1678-4782

J. Pediatr. (Rio J.) vol.92 no.1 Porto Alegre Jan./Feb. 2016 


Deuterium oxide dilution and body composition in overweight and obese schoolchildren aged 6-9 years

Wendell Costa Bila a   * 

André Everton de Freitas b  

Alexsandro Sobreira Galdino c  

Eduardo Ferriolli d  

Karina Pfrimer d  

Joel Alves Lamounier a  

aPost-Graduate Program in Health Sciences, Universidade Federal de São João del Rei (UFSJ), Divinópolis, MG, Brazil

bPost-Graduate Program in Child and Adolescent Health, Faculdade de Medicina, Universidade Federal de Minas Gerais (UFMG), Belo Horizonte, MG, Brazil

cPost-Graduate Program in Biotechnology, Universidade Federal de São João Del Rei (UFSJ), Divinópolis, MG, Brazil

dFaculdade de Medicina de Ribeirão Preto, Universidade de São Paulo (USP), Ribeirão Preto, SP, Brazil



To correlate different methods of body composition assessment in overweight or obese schoolchildren, using deuterium oxide (D2O) dilution as a reference.


Percentage of total body water (%TBW), fat free mass (%FFM), and body fat (%BF) were assessed by D2O and tetrapolar electrical bioimpedance analysis (BIA) in 54 obese and overweight students aged 6-9 years. Skinfold thickness (ST), body mass index (BMI), conicity index (CI), waist circumference (WC), waist-to-height ratio (WHtR), and waist-to-hip (WHR) ratio were also used.


Mean values for body composition were 38.4% ± 8.4% BF, 44.9% ± 6.1% TBW and 61.6% ± 8.4% FFM. There was no significant difference in body weight, body fat mass (FM), TBW, and FFM between genders. Regarding D2O, ST underestimated %BF, and overestimated %FFM in both genders (p < 0.05). BIA overestimated %TBW in the group as a whole and in males (p < 0.05). The only positive and strong correlations occurred in females regarding the WC (s = 0.679), CI (r = 0.634), and WHtR (r = 0.666).


In this sample of obese and overweight children, there were strong correlations between body composition measured by D2O and some indices and anthropometric indicators in females, but there was no positive and strong correlation of fat tissue with the indices/indicators at all ages and in both genders.

Keywords: Deuterium; Body composition; Obesity; Overweight; Child


High levels of body fat are associated with increased morbidity, and overweight and obesity definitions are associated with health risks.1 As childhood obesity has become an epidemic worldwide, studies performed in several regions of Brazil have demonstrated the marked increase of overweight and obesity in childhood and adolescence.2and3

Normally, body composition is used as an important nutritional indicator, allowing the division of body weight into components, specifically lean body mass or fat-free mass FFM-(muscle, bone, and water) and body fat mass (FM).4

To assess the body composition of an individual, the total body water (TBW) can be measured through the ingestion of a dose of labeled water and, subsequently, FFM and FM are calculated using hydration coefficients.5 Evidence demonstrates a growing interest in the use of isotopes in research activities aimed at nutritional and metabolic status, mainly in children.6

Isotopic dilution with deuterium oxide (D2O) is the standard technique for measuring TBW, and is most commonly used to measure TBW in children.7 Body composition characterization using D2O offers important advantages when compared to other methods, as it is innocuous to humans and can be used in pregnant women, children, and the elderly, with no clinical consequences. Very small amounts of material are required for the sample, which is particularly advantageous in pediatric use.8

The aim of this study was to correlate different methods usually employed in clinical practice for assessing body composition, specifically in schoolchildren aged between 6 and 9 years diagnosed with overweight and obesity, using deuterium oxide (D2O) dilution as the reference method. Among the methods used, body mass index (BMI), waist circumference (WC), waist-to-height ratio (WHtR), waist-to-hip ratio (WHR), conicity index (CI), skinfold thickness (ST), and tetrapolar bioelectrical impedance analysis (BIA) were analyzed.

Materials and methods

Study site and population

This study was approved by CEPES/UFSJ-CCO (Comitê de Ética em Pesquisa/Universidade Federal de São João del Rei - Campus Centro Oeste), Edict No. 265,272, according to the current norms of the Post-Graduate Program in Health Sciences, UFSJ-CCO.

Data were obtained from a study of prevalence of overweight and obesity, registered under No. 5533p2 in the Brazilian Registry of Clinical Trials (Registro Brasileiro de Ensaios Clínicos [ReBEC]), in a sample of 1564 school children aged 6-9 years (full years) enrolled in public schools, from the first to the fifth year of elementary school. A total of 1019 children were evaluated; those who were in the BMI percentile range ≥ 85 or ≥95, according to gender and age, were classified as overweight or obese, respectively.9 The prevalence of overweight and obesity was 11.6%,10 which is equivalent to 119 children, characterizing the subset of data that comprised the initial study sample.

Exclusion criteria included children who refused to participate in the study, failure to submit the informed consent signed by parents and/or guardians, those with infectious diseases, balance disorders, and history of brain surgery, as the data collected were also part of a study that evaluated the results of a physical activity program on biochemical variables. On the collection day, students who did not follow the previous recommendations (such as those who fasted less than eight hours or who participated in strenuous physical activity in the previous 24 h) were also excluded.

Data collection

All assessments were performed in the morning in a private room where anthropometric and body composition parameters were obtained. Measurements were conducted by previously trained evaluators.

Body weight was measured using a digital electronic scale (BF-683 W; Tanita(r) - IL, USA) with a maximum capacity of 150 kg and precision of 100 g, according to the techniques recommended by Jellife in 1968.11 Height was measured using a mobile vertical anthropometer (ALTUREXATA(r) - BH, Brazil) graded in centimeters up to 2.13 m and with precision of 0.1 cm.

A two-meter long, flexible, inelastic anthropometric tape, divided into centimeters and subdivided into millimeters, was used to measure waist (WC) and hip (HC) circumference. The individual remained in the orthostatic position and all measures occurred in the horizontal plane, taking care not to compress soft tissues. WC was obtained during normal expiration, using the midpoint between the last rib and the iliac crest as reference, and the HC measurement was obtained using the widest point of the hip at the level of the greater trochanter of the femur as reference.

To carry out the measurement of triceps (TSF) and calf skinfold (CSF) that were used in the equations proposed by Slaughter et al.,12 a skinfold caliper was used (Lange Skinfold Caliper - MI, USA) with a scale of 0-60 mm and precision of ±1 mm.12 All measurements were performed on the right side of the body, in triplicate, considering the arithmetic mean value as representative of the region. The TSF was measured on the posterior part of the right arm (vertical direction), at midpoint between the scapula acromion and the ulna olecranon. The CSF was measured on the medial side of the right leg (vertical direction) at the point of maximum circumference of the calf, with the knee and hip flexed to 90°. The equation proposed by Slaughter et al.13 was used to calculate %BF. The equation FFM = body weight - FM proposed by Barbosa-Cortés et al.14 was used to calculate the absolute value of the FFM, and its result was multiplied by 100 and divided by the body weight value to obtain the %FFM.

BIA variables were obtained using the horizontal tetrapolar BIA equipment (Quadscan 4000; Bodystat - Isle of Man, British Isles).

The plateau protocol15 was used for body composition measurement through the isotope dilution of deuterium oxide (D2O). Standard doses of 2 g, 3 g, and 4 g of deuterium oxide were administered to children weighing 20-30 kg, 30-50 kg, and 50-70 kg, respectively, with doses distributed in individual vials and administered orally to every child. The basal saliva sample was collected after an eight-hour fast. Subsequently, the diluted D2O solution was offered. The second sample was collected three hours after the administration of the diluted dose (post-dose).

Based on the fundamental principle of dilution, concentration and volume of D2O present and measured in saliva are correlated before and after dose ingestion, disclosing the total body water volume, which provides a calculation of FFM, using the specific hydration coefficients.

All samples were stored at -20 °C and sent to the Mass Spectrometry Laboratory of the Department of Clinical Medicine at Faculdade de Medicina de Ribeirão Preto of Universidade de São Paulo, for laboratory analyses.

Statistical analysis

All data were coded and stored using SPSS statistical software (IBM Corp. Released 2011. IBM SPSS Statistics for Windows, version 20.0. NY, USA). Descriptive results were obtained through central trend (mean and median) and dispersion measures (standard deviation).

The analytical statistics was initially performed through univariate analysis, with a 95% confidence interval and a significance level of 5%. The Shapiro-Wilk test was used to verify the normal distribution of variables. Student's t and Mann-Whitney tests were used to compare, respectively, variables (FM, TBW, or FFM) with normal or non-normal distribution in the independent groups.

ANOVA was used to compare independent variables and Tukey's test was used to identify the variables of cause.

Pearson's correlation coefficient (r) and Spearman's rho (s) were used to evaluate the correlation, respectively, of variables with normal or non-normal distribution between %BF measured by deuterium method and anthropometric indexes/variables, qualitatively evaluated according to the criteria established by Callegari-Jacques. 16


Of the 79 children who agreed to participate and were initially included in the study, 54 individuals (28 boys and 26 girls) comprised the final sample, due to losses incurred during data collection and analysis. Both groups (pre- and post-loss) statistically maintained similar anthropometric characteristics (p = 0.682). Except for age, weight, BMI, and CI, all other individuals in this total sample showed normal distribution. Table 1 shows the data distribution. Median age and BMI in the group were eight years and 20.3 kg/m2, respectively.

Table 1 Description of the study variables of schoolchildren aged 6-9 years with overweight and obesity. 

Variable Male Female
n Measure of central tendency ~N (p-value)a n Measure of central tendency ~N (p-value)a
Age (years) 28 8.0 (7.0–9.0) <0.05b 26 7.5 (7.0–9.0) <0.05b
Weight (kg) 28 38.4 ± 8.5 0.313c 26 38.0 (33.5–42.0) <0.05b
Height (cm) 28 133.2 ± 8.2 1.000c 26 134.3 ± 6.8 0.991c
BMI (kg/m2) 28 20.5 ± 3.1 0.051c 26 20.7 (19.1–22.4) <0.05b
WC (cm) 25 70.9 ± 9.7 0.725c 25 75.0 (66–77.5) <0.05b
CI 25 1.2 ± 0.1 0.063c 25 1.3 ± 0.1 0.387c
WHtR 25 0.5 ± 0.1 0.531c 25 0.5 ± 0.1 0.234c
WHR 25 0.9 ± 0.1 0.807c 25 0.9 ± 0.1 0.597c
BF% (D2O) 27 37.8 ± 10.7 0.284c 26 39.1 ± 5.4 0.588c
BF% (BIA) 28 35.6 ± 8.8 0.159c 25 41.9 ± 6.3 0.305c
BF% (ST) 25 26.3 ± 7.6 0.857c 26 31.5 ± 6.0 0.322c
TBW% (D2O) 27 45.4 ± 7.8 0.287c 26 44.5 ± 3.9 0.591c
TBW% (BIA) 28 49.3 ± 6.8 0.197c 25 45.0 ± 4.8 0.251c
FFM% (D2O) 27 62.2 ± 10.7 0.284c 26 60.9 ± 5.4 0.588c
FFM% (BIA) 28 64.4 ± 8.8 0.159c 25 58.1 ± 6.3 0.305c
FFM% (ST) 25 73.7 ± 7.6 0.856c 26 68.5 ± 6.0 0.322c

n, sample size; ~N, normal distribution; BMI, body mass index; WC, waist circumference; CI, conicity index; WHtR, waist-to-height ratio; WHR, waist-to-hip ratio; BF%, body fat percentage; D2O, deuterium oxide; BIA, Bioimpedance; ST, skinfold thickness; TBW%, total body water percentage; FFM%, fat-free mass percentage.

aShapiro-Wilk test (confidence interval = 95%).

bDoes not have normal distribution.

cHas normal distribution.

The percentage of total body water (%TBW) obtained by D2O method and BIA were significantly different (ANOVA, a = 0.05). Differences in this variable were significant for the group as a whole (44.9% vs. 47.6%) and for the male gender (45.4% vs. 49.3%) but not for females.

The %FFM measured through D2O and ST also differed significantly (p < 0.05) in the total sample (61.6% vs. 71.1%), in males (62.2% vs. 73.7%), and in females (60.9% vs. 68.5). Significant differences were not verified between the means of %FFM using the D2O and BIA methods.

The %BF obtained by D2O in the group as a whole was 38.4% ± 8.4% (37.8% ± 10.7% and 39.1% ± 5.4% in males and females, respectively). When the %BF measured by D2O, BIA and ST methods were compared, there were significant differences (p < 0.05) between the means in D2O and ST, either in the group as a whole (38.4% vs. 29.3%), in the males (37.8% vs. 26.3%), and in females (39.1% vs. 31.5%). There was no significant difference in %BF measured by the BIA and D2O methods.

Table 2 shows the absolute values of body weight, height, TBW, FFM, and FM measured by D2O, considering gender.

Table 2 Distribution of absolute values of body composition variables obtained by D2O method, according to gender. 

Variable (kg) Male Female
n ±s n ±s p-valuea
Body weight 28 38.4 ± 8.5 38.1 26 38.9 ± 8.4 38.0 0.924b
TBW 27 17.4 ± 3.7 18.0 26 17.1 ± 3.1 16.3 0.777c
FFM 27 23.8 ± 5.0 24.6 26 23.5 ± 4.3 22.4 0.775c
FM 27 15.0 ± 5.9 15.1 26 15.4 ± 5.0 14.3 0.972b

n , sample size; , sample mean; s , standard deviation; , Median; TBW, total body water; FFM, fat-free mass; FM, fat mass.

aConfidence interval = 95%.

bMann-Whitney test.

cStudent's t-test.

There was no statistically significant difference between body weight and FM between the genders (p = 0.924 and p = 0.972, respectively). TBW and FFM did not significantly differ, either (p = 0.777 and p = 0.775, respectively).

Table 3 shows the correlation of %BF by D2O, with the anthropometric indices and measurements. Fair but significant correlations (p < 0.05) were observed in the group as a whole between the %BF, body weight (s = 0.385), BMI (s = 0.445), and CI (s = 0.549). Similar results were found between %BF, WC (r = 0.440), and WHtR (r = 0.463).

Table 3 Comparisons between body fat percentage measured by deuterium oxide dilution with the anthropometric indices and measurements in schoolchildren aged 6-9 years with overweight and obesity. 

Index/anthropometric measure n Measure of central tendency Correlation coefficient Level of correlationc p-value
Body weight 28 38.4 ± 8.5 0.374a Fair 0.055
BMI 28 20.5 ± 3.1 0.389a Fair 0.045d
WC 25 70.9 ± 9.7 0.371a Fair 0.074
CI 25 1.2 ± 0.1 0.346a Fair 0.097
WHtR 25 0.5 ± 0.1 0.384a Fair 0.064
WHR 25 0.9 ± 0.1 0.158a Weak 0.461

Body weight 26 38.0 (33.5–42.0) 0.385b Fair 0.052
BMI 26 20.7 (19.1–22.4) 0.513b Fair 0.007d
WC 25 75.0 (66–77.5) 0.679b Strong 0.000d
CI 25 1.3 ± 0.1 0.634a Strong 0.001d
WHtR 25 0.5 ± 0.1 0.666a Strong 0.000d
WHR 25 0.9 ± 0.1 0.481a Fair 0.015d

BMI, body mass index; WC, waist circumference; CI, conicity index; WHtR, waist-to-height ratio; WHR, waist-to-hip ratio.

aPearson's correlation (normal distribution).

bSpearman's rho (nonparametric distribution).

cAccording to Callegari-Jacques.16

dSignificant correlation (a = 0.05).

A fair, but significant correlation was observed between %BF and BMI (r = 0.389) among boys.

Fair correlations between %BF by D2O, BMI (r = 0.513), and WHR (r = 0.481) were found in the female gender. Moreover, strong and significant correlations (p < 0.05) were observed in this group between %BF, WC (s = 0.679), CI (r = 0.634), and WHtR (r = 0.666).


The %BF obtained by D2O in this study was greater than that described by Deurenberg et al., cited by Filho17 in a study using skinfolds of 378 children and adolescents to define excess of body fat (20% and 25% for boys and girls, respectively). Therefore, this sample of children aged 6-9 years classified as obese or overweight had excess body fat. None of the anthropometric indices/measurements used in this study positively and strongly correlated with the amount of adipose tissue at all ages and in both genders, which calls for precautions regarding their exclusive and definitive use for the diagnosis of excess body fat.

TBW was significantly different when measured by BIA and D2O, which potentiates misdiagnosis regarding body hydration. Resende et al.18 also found significant differences in the %TBW in obese adolescents when measured by D2O and tetrapolar BIA, with an overestimation of the values obtained by BIA.

While in this study no significant differences in %FFM were found between D2O and BIA, a study by Resende et al.,19 with obese adolescents and mean age of 11.2 ± 0.9 years, also using tetrapolar bioimpedance, found significantly higher (overestimation) FFM values measured by BIA when compared to D2O.

The ST method with two skinfold measures and equations proposed by Slaughter et al.13 underestimated the measurement of %BF, which would lead to the misinterpretation of reduced health risks associated with excess body fat in these individuals. However, according to Himes,20 skinfold measurements usually show better correlation with total body fat than BMI.

BIA was efficient to measure the %BF and %FFM, but not to measure the %TBW. In the study by Lazzer et al., which assessed overweight and obese adolescents, cited by Himes,20 bipedal BIA underestimated the overall %BF by 2-3%. Camarneiro et al.,21 in their study with obese adolescents, showed a poor correlation between FFM and TBW obtained by tetrapolar BIA compared to D2O, but a strong correlation with FM.

BIA prediction equations vary depending on the reference population from which they are derived and different equations may also yield different body fat estimates. Consequently, although the measurements are performed with good reliability, biases resulting from prediction equations and their adaptations to the studied population are a concern regarding BIA.20

Body weight and BMI showed a significant positive correlation, but only a fair one with %BF measured at D2O in the group as a whole. Escott-Stump22 stated that BMI is not a useful indicator in childhood to be used as a screening tool, as it does not reflect body composition. Freedman and Sherry23 stated that changes in weight and height that occur during the growth process (ages between 5 and 18 years) result in substantial increases (50%) in BMI, which further complicates the interpretation of this index in children and adolescents.

The WHR was not reliable when compared to %BF measured by the reference method, but the WC, CI, and WHtR showed strong and positive correlations in the female gender, demonstrating they are good indicators of high body fat in this group. In multiple regression models, WC was also found to be better than BMI to predict insulin resistance, blood pressure, serum cholesterol and triglyceride levels, especially in adolescents, according to a study published by Himes.20

According to Oliveira et al.,24 WC has been used in outpatient care to assess fat deposition in the abdominal area of children and adolescents, as this single variable has shown a better association with metabolic changes than the waist-hip circumference ratio (WHR).

In this study, the mean value corresponding to the WC in the group as a whole showed to be higher than the cutoff of 71 cm, recommended as a limit for the occurrence of health risks.25 However, 24 individuals (44.4%) had a lower WC value than the cutoff, 16 (29.6%) with %BF > 33%, an important cutoff value for cardiovascular risk recommended by ABESO (Associação Brasileira para o Estudo da Obesidade e Síndrome Metabólica)25 measured by DEXA (Dual Energy X-ray Absorptiometry) in the study by Higgins et al.26 with children aged 4-11 years. Therefore, the exclusive use of the WC could lead to a false-negative risk assessment.

Additionally, six individuals (11.1%) had WC below the proposed cutoff points for detection of high fat mass in the trunk (adjusted for age/gender),27 but with %BF higher than the cutoff point used in this study, also potentially characterizing false negative results.

According to CI cutoff points used for excess fat identification recommended by Sant'Anna et al.28 in their study with children aged 6-9 years old, four individuals (7.4%) had acceptable values, but with %BF higher than that recommended in this study as borderline, adjusted by gender/age, which can also characterize false negative results.

The mean values of WHtR in the group as a whole showed to be slightly higher than the cutoff of 0.50 cm, which is the threshold for health risk in individuals of both genders after the age of six, according to a study carried out by McCarthy and Ashwell (2006) that evaluated 8135 children and adolescents.29 However, 15 children (27.8%) showed a lower WHtR than this value and a higher %BF than the reference used in this study, which could result in false negative results if it were the only variable assessed.

Freedman and Sherry23 affirm that the skinfold and circumference measurements require additional training and may be difficult to standardize. The information provided by these measures in children with excess weight is actually inexact.

All individuals showed a %TBW below 50% of body weight. Overweight or obese children potentially have less TBW, probably justified by higher FM and %BF. Kliegman et al.30 reported that the greater amount of fat in overweight children causes a reduction in the TBW as a percentage of body weight.

Study limitations include the analysis of a subset of data from a study with cross-sectional design, which does not translate into any conclusions about causal associations and does not allow comparisons with other studies with identical design, methodology, and population.

The results of in this study suggest considering the use of indexes and anthropometric measurements, skinfold thickness, and bioimpedance as tools for the definitive diagnosis of overweight and obesity in children, as there was no absolute agreement of any of the techniques at all age ranges in both genders.

More studies are clearly necessary in this age range of the population, especially on the viability of greater use of the D2O method for body composition assessment, considered a reference for the measurement of TBW in children, as it requires smaller sample quantity and is innocuous to humans, with no risks or clinical consequences.


The authors would like to thank CAPES (Coordenação de Aperfeiçoamento de Pessoal de Nível Superior) for financial support. They also thank the Post-Graduate Program in Health Sciences of Universidade Federal de São João del Rei, for the financial and institutional support, through its professionals and teachers, particularly to Professor Gilberto Fontes, who contributed with important suggestions for the article.


1. Speiser PW, Rudolf MC, Anhalt H, Camacho-Hubner C, Chiarelli F, Eliakim A, et al. Consensus statement: childhood obesity. J Clin Endocrinol Metab. 2005;90:1871-87. [ Links ]

2. Abrantes MM, Lamounier JA, Colosimo E. Prevalência de sobrepeso e obesidade nas regiões nordeste e sudeste do Brasil. J Pediatr (Rio J). 2002;78:335-40. [ Links ]

3. Silveira JA, Colugnati FA, Cocetti M, Taddei JA. Secular trends and factors associated with overweight among Brazilian preschool children: PNSN-1989, PNDS-1996, and 2006/07. J Pediatr (Rio J). 2014;90:258-66. [ Links ]

4. Ferriolli E, Cruz BM, Pfrimer K. Uso de isótopos leves em ciências nutricionais. In: Oliveira JE, Marcrhini JS, editors. Ciências nutricionais: aprendendo a aprender. São Paulo (SP): Sarvier; 2008. p. 443-65. [ Links ]

5. IAEA. Targeting malnutrition: isotopic tools for evaluating nutrition worldwide; 1996. Available from: http://www.iaea. org/Publications/Booklets/Malnutrition/seven.html [cited 30.05.13]. [ Links ]

6. Wells JC, Chomtho S, Fewtrell MS. Programming of body composition by early growth and nutrition. Proc Nutr Soc. 2007;66:423-34. [ Links ]

7. Alemán-Mateo H, Esparza Romero J, Macias Morales N, Salazer G, Hernández Trian M, Valencia ME. Body composition by three- compartment model and relative validity of some methods to assess percentage body fat in Mexican healthy elderly subjects. Gerontology. 2004;50:366-72. [ Links ]

8. Koletzko B, Demmelmaira H, Hartlb W, Kindermann A, Koletzko S, Sauerwald T, et al. The use of stable isotope techniques for nutritional and metabolic research in paediatrics. Early Hum Dev. 1998;53:S77-97. [ Links ]

9. De Onis M, Onyango AW, Borgui E, Siyam A, Nishida C, Siekmann J. Development of a WHO growth reference for school-aged children and adolescents. WHO (Genebra). 2007;85:660-7. [ Links ]

10. Freitas AE [Thesis] Avaliação do impacto de um programa de atividade física na concentração plasmática de adipocitocinas e grelina em crianças obesas e com sobrepeso, de escolas públicas da zona urbana de Ouro Preto, MG. Belo Horizonte, MG: Universidade Federal de Minas Gerais; 2010. [ Links ]

11. Jelliffe DB. Evolución del estado de nutrición de la comunidad. Ginebra: Organización Mundial de la Salud; 1968. [ Links ]

12. Heyward VH, Stolarczyl LM. Avaliação da composição corporal. São Paulo, SP: Manole; 2000. [ Links ]

13. Slaughter MH, Lohman TG, Boileau RA, Horswill CA, Stillman RJ, Loan MD, et al. Skinfold equations for estimations of body fatness in children and youth. Hum Biol. 1988;60:709-23. [ Links ]

14. Barbosa-Cortéz L, Tapia- Rojas M, López -Aguilar E, Mejía- Aranguré JM, Rivera -Márquez H. Body composition by dilution of deuterium oxide in Mexican children with lymphoma and solid tumors. Nutrition. 2007;23:739-44. [ Links ]

15. Azócar MP, Cano FS, Marín VB, Díaz EB, Salazar GR, Vásquez LF. Estimación del água corporal total por deutério em diálisis perit- oneal pediátrica. Rev Chil Pedriatr (Santiago). 2003;74:504-10. [ Links ]

16. Callegari-Jacques SM. Bioestatística. Princípios e aplicações. São Paulo, SP: Artmed; 2007. [ Links ]

17. Filho JF. A prática da avaliação física. 2nd ed. Rio de Janeiro, RJ: Shape; 2003. [ Links ]

18. Resende CM, Júnior JS, Vieira MN, Perdoná GS, Ferriolli E, Pfrimer K, et al. Body composition in obese adolescents: deuterium oxide dilution method, bioelectrical impedance and predictive equations. Curr Nutr Food Sci (Sharjah). 2013;9:73-81. [ Links ]

19. Resen CM, Júnior JS, Vieira MN, Ferriolli E, Pfrimer K, Perdoná GS, et al. Body composition measures of obese adolescents by the deuterium oxide dilution method and by bioelectrical impedance. Braz J Med Biol Res. 2011;44:1164-70. [ Links ]

20. Himes JH. Challenges of accurately measuring and using BMI and other indicators of obesity in children. Pediatrics. 2009;124:S3-22. [ Links ]

21. Camarneiro JM, Junior JSC, Ciampo LAD, Navarro AM, Antonucci GA, Monteiro JP. Body composition estimatives by anthropometry, bioelectrical impedance and deuterium oxide dilution in obese adolescents. Food Nutr Sci (Irvine). 2013;4:9-17. [ Links ]

22. Escott-Stump S. Nutrição relacionada ao diagnóstico e tratamento. 5 ed. Barueri, SP: Manole; 2007. [ Links ]

23. Freedman DS, Sherry B. The validity of BMI as an indi- cator of body fatness and risk among children. Pediatrics. 2009;124:S23-34. [ Links ]

24. Oliveira CL, De Mello MT, Cintra IP, Fisberg M. Obesidade e síndrome metabólica na infância e adolescência. Rev Nutr (Campinas). 2004;17:237-45. [ Links ]

25. ABESO, Associação Brasileira para o Estudo da Obesidade e Síndrome Metabólica. Diretrizes brasileiras da obesidade; 2009. Available from: brasileirasobesidade200920101.pdf [cited 06.01.14]. [ Links ]

26. Higgins PB, Gower BA, Hunter GR, Goran MI. Defining health-related obesity in prepubertal children. Obes Res. 2001;9(4):233-40. [ Links ]

27. Taylor RW, Jones IE, Williams SM, Gouldling 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. [ Links ]

28. Sant&apos;anna MS, Tinôco AL, Rosado LE, Sant&apos;ana LF, Brito IS, Araújo LF, et al. Eficácia do índice de conicidade e da relação cintura/estatura em predizer o percentual de gordura corporal em crianças. Nutrire. 2010;35:67-80. [ Links ]

29. Mccarthy HD, Ashwell M. A study of central fatness using waist-to-height ratios in UK children and adolescents over two decades supports the simple message - &apos;keep your waist cir- cumference to less than half your height&apos;. Int J Obes (Lond). 2006;30:988-92. [ Links ]

30. Kliegman R, Behrman RE, Jenson HB, Stanton BF. Nelson, tratado de pediatria. 18th ed. Rio de Janeiro, RJ: Elsevier; 2009. [ Links ]

Please cite this article as: Bila WC, de Freitas AE, Galdino AS, Ferriolli E, Pfrimer K, Lamounier JA. Deuterium oxide dilution and body composition in overweight and obese schoolchildren aged 6-9 years. J Pediatr (Rio J). 2016;92:46-52.

Received: January 26, 2015; Accepted: March 24, 2015

Corresponding author. (W.C. Bila).

Conflicts of interest The authors declare no conflicts of interest.

Creative Commons License This is an open-access article distributed under the terms of the Creative Commons Attribution License