¹EU Ciencias de la Salud, Universidad de Zaragoza, Zaragoza, Spain

²Departamento de Pediatría, Facultad de Medicina, Universidad de Zaragoza, Zaragoza, Spain

³Departamento de Medicina Preventiva, Facultad de Medicina, Universidad de Zaragoza, Zaragoza, Spain

⁴Area de Nutrición y Bromatología, Facultad de Veterinaria, Universidad de Zaragoza, Zaragoza, Spain

Correspondence to: L A Moreno, EU Ciencias de la Salud, Universidad de Zaragoza, Avda Domingo Miral s/n, 50009 Zaragoza, Spain. E-mail: lmoreno@posta.unizar.es

^aGuarantor: LA Moreno.

^bContributors: LAM was responsible for study design, analysis and writing; GR was responsible for analysis and writing; JG was responsible for analysis and discussion; MJR, JFL and AA were responsible for conception, design and comments.

Objective: To analyze the impact of choosing the left or the right side of the body on the anthropometric measurements and derived nutritional indices, in prepubertal children.

Design: Cross-sectional pilot nutrition survey.

Setting: General prepubertal school-age population.

Subjects: One-hundred and sixty-four children (97 boys and 67 girls) aged 7-9 y.

Interventions: None.

Results: The agreement between anthropometric measurements in both sides of the body showed that in males and in females, suprailiac skinfold thickness and arm circumference were significantly higher in the left than in the right side of the body. The agreement between body composition assessed by anthropometric measurements in both sides of the body showed that only in males was arm muscle area significantly higher in the left than in the right side, and arm fat percentage was higher in the right than in the left side of the body. Total body fat percentage calculated from skinfold thickness did not show statistically significant differences when skinfolds were obtained in the both sides of the body, either in boys and in girls.

Conclusions: Our results show that differences between the sides of the body were lower than the technical error of measurement of the anthropometric measurements obtained and seem not to be biologically significant in this age group. It is necessary to standardize the method of anthropometric assessment of the nutritional status in terms of body side.

Sponsorship: Universidad de Zaragoza (216-17).

European Journal of Clinical Nutrition (2002) 56, 1208-1215. doi:10.1038/sj.ejcn.1601493

anthropometry; handedness; child; nutrition surveys; skinfold thickness; circumferences

Introduction

Because of its importance to health, body composition is commonly investigated in epidemiological, clinical and population studies. Reliable methods for measurement of body fat and fat distribution are therefore of importance. During the past decade, investigators have emphasized the accuracy of newer techniques, such as dual-energy X-ray absorptiometry (DXA), magnetic resonance imaging and computerized tomography, for measuring body composition; nevertheless, anthropometry is still the most widely used method, and it has recently been used to estimate fat distribution (Goran et al, 1998; Moreno et al, 1997,2001a). The distinct advantages of anthropometry are that it is portable, noninvasive, inexpensive and useful in field studies, and there is also a substantial literature available (Wang et al, 2000).

In choosing the instrument to assess nutritional status, researchers often elect to measure only height and weight. These measures are quick, simple and require only limited training. More comprehensive measurement sets which include skinfolds and circumferences require more training and carry different degrees of error with them. Some skinfold thicknesses and circumferences can be taken on both sides of the body. There is no agreement about the side of the body from which anthropometric measurements must be obtained. In general, in Europe (Weststrate et al, 1989; Deurenberg et al, 1990,1999; Gurrici et al, 1998) and in children (Tanner & Whitehouse, 1962,1975; Gerver & de Brain, 1996; Moreno et al, 1998; Paul et al, 1998), we choose the left side of the body, and in sports medicine (Peters et al, 1994; Eliakim et al, 1997; Katzmarzyk et al, 1999) and in North America (Johnston et al, 1988; Must et al, 1991; Sangi & Mueller, 1991), the right side of the body is used. Some papers even show data from both sides (Lean et al, 1996) and others use the non-dominant side of each individual (Barker et al, 1997).

There are limited data comparing the effect of measuring skinfolds and circumferences in both sides of the body on the assessment of nutritional status (Burgert & Anderson, 1979). Therefore, the objective of this study was to analyze the impact of choosing the left or the right side of the body on the anthropometric measurements and derived nutritional indices, in prepubertal children.

Subjects and method

Subjects

One-hundred and sixty-four children (97 boys and 67 girls) aged 7-9 y were studied. The main characteristics of these children are shown in Table 1. For the recruitment of these children we selected three schools (one of medium-high socioeconomic status, one of medium socioeconomic status and one of medium-low socioeconomic status); this represents the childhood population of Zaragoza (Spain; Moreno et al, 1999,2001b). In each school, all the children aged 7-9 y were asked to participate in the pilot survey. Parents or children supervisors were informed by a letter about the nature and purpose of the study. After receiving their written consent, the children were considered for inclusion in the study. The children had no apparent disease and took no medication; none of them had a history of endocrine, nutritional, growth or renal problems. The participation rate was 84%.

Anthropometric method

All the anthropometric measurements were taken three times by the same person. We measured all the anthropometric variables in order, and then we repeated the same measurements a second and a third time. Measurements were recorded by another person, who did not collaborate in any other way with the measurer. Body mass index (BMI) was calculated as body weight (kg) without shoes and with light clothing, divided by height (m) squared. Body weight was measured to 0.05 kg using a standard beam balance. Height was measured to the nearest 1 mm using a Harpenden stadiometer. All the anthopometric material was calibrated every day.

Skinfold thicknesses were measured at the left and the right side of the body to the nearest 0.1 mm with a Holtain skinfold caliper, at the following sites: (1) triceps, halfway between the acromion process and the olecranon process; (2) biceps, at the same level as the triceps skinfold, directly above the centre of the cubital fossa; (3) subscapular, about 20 mm below the tip of the scapula, at an angle of 45° to the lateral side of the body; and (4) suprailiac, about 20 mm above the iliac crest and 20 mm towards the medial line.

Circumferences were also measured with a non-elastic tape on the left and the right sides of the body. For measuring the arm circumference, the subjects stood relaxed with his/her side to the observer, and the arm hanging freely at the side; the tape was passed around the arm at the level of the midpoint of the upper arm. Proximal thigh circumference was measured just below the gluteal fold and perpendicular to its long axis; the subject stood erect with the feet slightly apart and the body mass evenly distributed between both legs.

The technical error of measurement (TEM) was calculated with the following formula:

The coefficient of reliability (R) was calculated using the equation:

where s.d.² is the total inter-subject variance for the study, including measurement error. This coefficient reveals the proportion of the between-subject variance in a measured population that is free from measurement error.

Technical errors of measurement for skinfolds ranged from 0.48 to 0.56 mm, and for circumferences were 0.40 cm for arm circumference and 0.73 cm for thigh circumference. Reliability for skinfolds ranged from 97.13 to 99.39%, and for circumferences were 97.85% for arm circumference and 99.85 for thigh circumference.

Body density was calculated using the formulas described by Brook (1971):

Density was converted to an estimate of body fat (%) using Weststrate and Deurenberg's (1989) equation:

Body composition indices based on arm measurements were calculated using the equations describe by Jellife (1966):

Because it was known at the start of the study that the sample size would be too small to include sizable subsamples differing in the degree of left-handed preference, the subjects were questioned only to determine whether or not, in performing manual activities, they were right-handed. The two handedness subsamples were right (n=157) and nonright (n=7).

Statistical analysis

Statistical analysis was performed with SPSS software and agreement plots with Microsoft Excel software for Windows. The distribution of quantitative variables was tested for normality using the Kolmogorov-Smirnov test, with Lilliefors correction, in both males and females. Skinfold thickness and circumferences showed a non-gaussian distribution in both males and females. Body composition indices showed a gaussian distribution. Therefore, anthropometric measurements were described by median and interquartile range (IQR) and body composition indices with mean and standard deviation. Comparisons between males and females were done with the Wilcoxon test for the anthropometric measurements and with the paired t-test for the body composition indices. For judging statistical significance of the tests, a probability of 0.005 was chosen because of the large number of tests that were performed.

A plot of left-right differences by each anthropometric measurement and body composition index was made to assess agreement between both sides of the body, following the Bland-Altman method (Bland & Altman, 1986). With this graphical and numerical analysis, the lack of agreement between the both sides of the body was assessed by calculating the bias (mean difference, standard deviation of the differences and 95% confidence intervals for the bias).

Results

Ages were 7.67±0.6 in boys and 7.75±0.6 in girls. Weight was 30.5±6.1 and 30.3±6.0 kg; height 1.31±0.06 and 1.30±0.07 m; and BMI 17.7±2.6 and 17.8±2.6 kg/m², respectively.

The anthropometric measurements in both sides of the body in males and in females are described in Table 1. With a level of significance £0.005, we only observed significant results for subscapular skinfold in females; subscapular skinfold thickness was higher in the left than in the right side of the body. For body composition, we observed significant results only in males; the arm muscle area was higher in the left than in the right side of the body, and arm fat percentage was higher in the right than in the left side of the body (Table 2). Some P-values were low with identical medians in each group; in this case, interquartile ranges were also taken into account.

The agreement between anthropometric measurements in both sides of the body showed that, in males and in females, suprailiac skinfold thickness and arm circumference were significantly higher in the left than in the right side of the body (Table 3). The agreement between body composition assessed by anthropometric measurements in both sides of the body showed that only in males was arm muscle area significantly higher in the left side than in the right side, and arm fat percentage was higher in the right than in the left side of the body (Table 4). Total body fat percentage calculated from skinfold thickness measures taken on the right side of the body compared to the left side did not differ. In Figures 1 and 2 we show the Bland-Altman plots for different anthropometric measurements and indices, in males (Figure 1) and females (Figure 2); mean differences were close to 0, except in males for arm muscle area and arm fat percentage.

Discussion

Skinfold thickness and circumferences are accepted as body fatness predictors. Skinfold thickness represent about 40-60% of total body fat in the subcutaneous region of the body, and can be directly measured using a well-calibrated caliper. Some circumference measurements have also been used in equations for predicting body fatness. Circumferences measured at midarm, midthigh, waist and hip are used more frequently than others, because they indicate differences among people in major regions of the body. Recently, many studies have used circumferences for estimating skeletal muscle mass and fat distribution (Rolland-Cachera et al, 1997; Moreno et al, 1999).

Equations that predict body composition values provide a way of obtaining such data from variables that can be measured easily and accurately in large-scale epidemiological and population studies where sophisticated laboratory settings are impractical. The predicted values of body composition are less precise than those from measured laboratory procedures, but they are less expensive, and they are practical and easy to apply.

Asymmetry of paired dimensions has been recognized as a methodological problem in anthropometry (Schell et al, 1985). The conventional statistical tests seems not to be adequate to analyse this asymmetry. In this pilot survey, we have used the method described by Bland and Altman (1986) to assess the agreement between two methods of measurement. This is not the case of our study, but the method seems to be adequate to evaluate the measurements obtained with the same method in both sides of the human body.

For analysis of our data, we have used, first of all, conventional statistical tests. In the results we have given exact P-values, aiming to understand the extent of the differences between both sides of the body. In our opinion other types of analysis would be better aiming to assess these differences. With the method of Bland and Altman (1986) we did not observe significant differences in total body fat percentage, calculated either with left or right skinfold thickness; such a 95% confidence interval for total body fat percentage bias shows that, if we repeat the same experiment under similar conditions in other groups of prepubertal children, mean differences between total body fat percentage calculated from left and right measurements will vary (95% of the time) by 0.3% for males and 0.4% for females.

We also observed that in prepubertal males, arm muscle area was significantly higher in the left side than in the right side, and arm fat percentage was higher in the right side than in the left side of the body. However, we must be very cautious in the interpretation of these results because the observed differences are lower than the technical errors of measurement for each anthropometric measurement.

In 135 adolescents involved in organized school sports, Schell et al (1985) observed that arm measurements were significantly asymmetric in favor of the right side; subscapular skinfolds and leg measurements were not significantly asymmetric. Among the sample of right-handed subjects, upper arm circumference and biepicondylar breadth were significantly larger on the right side and, among the males of this subsample, triceps skinfold was as well. The nonright-handed subjects did not show statistically significant asymmetry. Asymmetry was negatively and weakly related to body composition.

Our results show that differences between both sides of the body were lower than the technical error of measurement of the anthropometric measurements obtained. The consequences of measuring skinfold thickness and circumferences in one or the other side of the body are limited and seemed not to be biologically significant in this age group, but the range of body fat composition was not very large. However, these differences could be different in adolescents and adults. In order to compare nutritional status assessed by anthropometry in different populations, more studies are necessary to standardize the method of anthropometric assessment in terms of body side. We must choose one side, perhaps the non-dominant one, as already used by some authors (Barker et al, 1997).

Acknowledgements

This work was supported by grant 216-17 from Universidad de Zaragoza, Zaragoza (Spain).

Barker M, Robinson S, Osmond C, Barker DJP. (1997). Birth weight and body fat distribution in adolescent girls. Arch. Dis. Child., 77: 381-383. PubMed

Bland JM, Altman DG. (1986). Statistical methods for assessing agreement between two methods of clinical measurement. Lancet, 8: 307-310.

Brook CGD. (1971). Determination of body composition of children from skinfold measurements. Arch. Dis. Child., 46: 182-184. PubMed

Burgert SL, Anderson CF. (1979). An evaluation of upper arm measurements used in nutritional assessment. Am. J. Clin. Nutr., 32: 2136-2142.

Deurenberg P, Pieters JJL, Hauvast JGAJ. (1990). The assessment of the body fat percentage by skinfold thickness measurements in childhood and young adolescence. Br. J. Nutr., 63: 293-303. PubMed

Deurenberg P, Deurenberg Yap M, Wang J, Lin FP, Schmidt G. (1999). The impact of body build on the relationship between body mass index and percent body fat. Int. J. Obes. Relat. Metab. Disord., 23: 537-542. Article PubMed

Eliakim A, Burke GS, Cooper DM. (1997). Fitness, fatness, and the effect of training assessed by magnetic resonance imaging and skinfold-thickness measurements in healthy adolescent females. Am. J. Clin. Nutr., 66: 223-231.

Gerver WJM, de Bruin R. (1996). Body composition in children based on anthropometric data. A presentation of normal values. Eur. J. Pediatr., 155: 870-876. PubMed

Goran MI, Gower BA, Treuth M, Nagy TR. (1998). Prediction of intra-abdominal and subcutaneous abdominal adipose tissue in healthy prepubertal children. Int. J. Obes. Relat. Metab. Disord., 22: 549-558. Article PubMed

Gurrici S, Hartriyanti Y, Hautvast JGAJ, Deurenberg P. (1998). Relationship between body fat and body mass index: differences between Indonesians and Dutch Caucasians. Eur. J. Clin. Nutr., 52: 779-783. Article PubMed

Jellife DB. (1966). The assessment of the nutritional status of the community. Geneva: World Health Organization.

Johnston JL, Leong MS, Checkland EG, Zuberbuhler PC, Conger PR, Quinney HA. (1988). Body fat assessed from body density and estimated from skinfold thickness in normal children and children with cystic fibrosis. Am. J. Clin. Nutr., 48: 1362-1366.

Katzmarzyk PT, Pérusse L, Malina RM, Bouchard C. (1999). Seven-year stability of indicators of obesity and adipose tissue distribution in the Canadian population. Am. J. Clin. Nutr., 69: 1123-1129. PubMed

Lean MEJ, Han TS, Deurenberg P. (1996). Predicting body composition by densitometry from simple anthropometric measurements. Am. J. Clin. Nutr., 63: 4-14. PubMed

Moreno LA, Fleta J, Mur L, Feja C, Sarría A, Bueno M. (1997). Indices of body fat distribution in Spanish children aged 4.0 to 14.9 y. J. Pediatr. Gastroenterol. Nutr., 25: 175-181. PubMed

Moreno LA, Fleta J, Mur L, Sarría A, Bueno M. (1998). Fat distribution in obese and nonobese children and adolescents. J. Pediatr. Gastroenterol. Nutr., 27: 176-180.

Moreno LA, Fleta J, Mur L, Rodríguez G, Sarría A, Bueno M. (1999). Waist circumference values in Spanish children¾gender-related differences. Eur. J. Clin. Nutr., 53: 429-433. Article PubMed

Moreno LA, Fleta J, Sarría A, Rodríguez G, Gil C, Bueno M. (2001a). Secular changes in body fat patterning in children and adolescents of Zaragoza (Spain), 1980-1995. Int. J. Obes. Relat. Metab. Disord., 125: 1656-1660.

Moreno LA, Fleta J, Sarría A, Rodriguez G, Bueno M. (2001b). Secular increases in body fat percentage in male children of Zaragoza (Spain), 1980-1995. Prev. Med., 33: 357-363.

Must A, Dallal GE, Dietz WH. (1991). Reference data for obesity: 85th and 95th percentiles of body mass index (wt/ht²) and triceps skinfold thickness. Am. J. Clin. Nutr., 53: 839-846. PubMed

Paul AA, Cole TJ, Ahmed EA, Whitehead RG. (1998). The need for revised standards for skinfold thickness in infancy. Arch. Dis. Child., 78: 354-358. PubMed

Peters D, Fox K, Armstrong N, Sharpe P, Bell M. (1994). Estimation of body fat and body fat distribution in 11-year-old children using magnetic resonance imaging and hydrostatic weighing, skinfolds, and anthropometry. Am. J. Hum. Biol., 6: 237-243.

Rolland-Cachera MF, Brambilla P, Manzoni P, Akrout M, Sironi S, Del Maschio A, Chiumello G. (1997). Body composition assessed on the basis of arm circumference and triceps skinfold thickness: a new index validated in children by magnetic resonance imaging. Am. J. Clin. Nutr., 65: 1709-1713.

Sangi H, Mueller WH. (1991). Which measure of body fat distribution is best for epidemiologic research among adolescents? Am. J. Epidemiol., 133: 870-883.

Schell LM, Johnston FE, Smith DR, Paolone AM. (1985). Directional asymmetry of body dimensions among white adolescents. Am. J. Phys. Anthropol., 67: 317-322.

Tanner JM, Whitehouse RH. (1962). Standards for subcutaneous fat in British children. Percentiles for thickness of skinfolds over triceps and below scapula. Br. Med. J., 1: 446-450.

Tanner JM, Whitehouse RH. (1975). Revised standards for triceps and subscapular skinfolds in British children. Arch. Dis. Child., 50: 142-145. PubMed

Wang J, Thornton JC, Kolesnik S, Pierson RN. (2000). Anthropometry in body composition. An overview. Ann. NY Acad. Sci., 904: 317-326. PubMed

Weststrate JA, Deurenberg P. (1989). Body composition in children: proposal for a method for calculating body fat percentage from total body density or skinfold-thickness measurements. Am. J. Clin. Nutr., 50: 1104-1115. PubMed

Weststrate JA, Deurenberg P, van Tinteren H. (1989). Indices of body fat distribution and adiposity in Dutch children from birth to 18 y of age. Int. J. Obes., 13: 465-477.

Figure 1 Bland-Altman plots for anthropometric indices in males. (A) total body fat percentage; (B) arm muscle area; and (C) arm fat percentage.

Figure 2 Bland-Altman plots for anthropometric indices in females. (A) total body fat percentage; (B) arm muscle area; and (C) arm fat percentage.

Table 1 Anthropometric measurements in both sides of the body in prepubertal children

Table 2 Body composition assessed by anthropometric measurements in both sides of the body in prepubertal children

Table 3 Agreement between anthropometric measurements in both sides of the body in prepubertal children

Table 4 Agreement between body composition assessed by anthropometric measurements in both sides of the body in prepubertal children