Estimating body composition in children with ...

Estimating body composition in children with Duchenne muscular dystrophy: comparison of bioelectrical impedance analysis and skinfold-thickness measurement1–3 Elise Mok, Laurent Béghin, Pierre Gachon, Christel Daubrosse, Jean-Eudes Fontan, Jean-Marie Cuisset, Frédéric Gottrand, and Régis Hankard

KEY WORDS Obesity, Duchenne muscular dystrophy, bioelectrical impedance, body composition, isotope labeling, children, fat-free mass, percentage fat mass, nutritional assessment, handicap INTRODUCTION

As in many chronic diseases of childhood, nutritional support has become an important part of the care of children with Duchenne muscular dystrophy (DMD). DMD is a progressive genetic disease that is characterized by a dramatic loss of muscle mass and function. Obesity occurs early in the course of the disease and aggravates the burden of the weakened muscles. Malnutrition further exacerbates disease progression in the latter

stages of DMD, when handicap impairs oral intake and complications increase nutritional needs. In a large cohort of patients with DMD, 44% of patients were obese by the age of 12 y, and 44% of patients had malnutrition by the age of 18 y (1). It is possible that weight control may delay the onset of wheelchair dependency in DMD. Moreover, maintenance of fat-free mass (FFM) is a key issue in nutritional support, whether the goal is to prevent or limit increased adiposity or to maintain body weight. Methods for evaluating nutritional status can be difficult and often misleading in DMD. For instance, body composition can be estimated from measurements of skinfold thickness (ST), but the method can be uncomfortable, particularly in children. Moreover, ST measurement requires an experienced observer and has poor interobserver reliability. Bioelectrical impedance analysis (BIA) can serve as an alternative method for estimating body composition because it is easy to administer and does not cause discomfort. BIA is also less dependent on the observer’s skills. In contrast to normal subjects, BIA in patients with DMD provides lower estimates of FFM [ie, a higher percentage of fat mass (%FM)] than does ST measurement when the 2 methods are performed simultaneously (R Hankard, unpublished data, 2002). 1

From INSERM Centre D’Investigation Clinique 9202, Assistance Publique-Hôpitaux de Paris, Hôpital Robert Debré, Paris, France (EM, CD, and RH); Laboratoire Adaptation Physiologique aux Activités Physiques EA3813, Université de Poitiers, Poitiers, France (EM and RH); EA3925 INSERM Centre D’Investigation Clinique 9301, Centre Hospitalier Régional Universitaire de Lille, Lille, France (LB); EA3925 Clinique de Pédiatrie, Centre Hospitalier Régional Universitaire de Lille, Hôpital Jeanne de Flandre, Lille, France (FG and LB); Unité du métabolisme protéinoénergétique, UMR INRA 1019 Laboratoire de Nutrition Humaine, Clermont-Ferrand, France (PG); the Pharmacy, Assistance PubliqueHôpitaux de Paris, Hôpital Jean-Verdier, Paris, France (J-EF); and the Service de Neuropédiatrie, Centre Hospitalier Régional Universitaire Lille, Hôpital Roger-Salengro, Lille, France (J-MC). 2 Supported by grant no. RBM 98019 from the French National Institute for Health and Medical Research (INSERM) (to RH). E. Mok is supported by Le Prix de Nutrition de la Fédération Association Nationale pour les Traitements a Domicile, les Innovations et la Recherche, awarded by La Société Francophone de Nutrition Entérale et Parentérale, and by a doctoral fellowship from Les Fonds de la recherche en santé Québec. 3 Reprints not available. Address correspondence to R Hankard, Pédiatrie, Centre Hospitalier Universitaire de Poitiers, 2 rue de la Milétrie, 86021 Poitiers Cedex, France. E-mail: [email protected]. Received June 10, 2005. Accepted for publication September 26, 2005.

Am J Clin Nutr 2006;83:65–9. Printed in USA. © 2006 American Society for Nutrition

65

Downloaded from ajcn.nutrition.org by guest on May 29, 2013

ABSTRACT Background: Duchenne muscular dystrophy (DMD) is often associated with obesity, which worsens the handicap early in the course of the disease. Nutritional assessment, however, can be difficult and often misleading in DMD. Objective: Two methods of estimating body composition in DMD, skinfold-thickness (ST) measurement and bioelectrical impedance analysis (BIA), were compared with a reference method, labeled water dilution (WD). Design: Body composition was estimated by using ST measurements and BIA (50 kHz, 800 mAmp), as well as the WD method (1 mL H218O/kg) in 11 DMD patients with a mean (앐SD) age of 10.0 앐 2.5 y. Results: When compared with the WD method, ST measurement significantly (P 쏝 0.01) overestimated fat-free mass (FFM) (x៮ 앐 SD ST: 24.5 앐 5.9 kg; x៮ 앐 SD WD: 18.2 앐 2.5 kg), which led to an underestimation of the percentage of fat mass (%FM) (ST: 23.3 앐 10.4%; WD: 40.1 앐 17.1%; P 쏝 0.05). In contrast, estimates obtained with BIA (FFM: 21.5 앐 4.5 kg; %FM: 31.3 앐 13.9%) did not differ from those obtained with WD. The difference from the reference method was less for BIA (x៮ : 3.3 kg; 95% CI: 0.8, 4.9 kg) than for ST (6.3 kg; 2.2, 8.6 kg). WD and BIA defined 73% and 55%, respectively, of the children as obese (%FM associated with body mass index cutoffs for obesity), whereas ST measurements defined 9% as obese (P 쏝 0.01). Conclusions: Body-composition estimates by BIA are closer to those by WD than are those by ST measurement. Early detection of fat accumulation and longitudinal monitoring of nutritional care are 2 relevant applications of BIA to prevent obesity and hence lessen the burden of DMD. Am J Clin Nutr 2006;83:65–9.

66

MOK ET AL

However, neither of these 2 methods has been validated in patients with DMD. We hypothesize that the BIA method will provide body-composition estimates closer to those of the reference method than will ST measurements in children with DMD. Hence, the aim of the study was to compare body-composition estimates from BIA and ST measurement with those from a reference method (WD method) in children with DMD.

in taking ST measurements. In the first group of children studied at the site in Lille, measurements were carried out by both observers on the same subject and skinfold site, to ensure reproducibility. Body density (BD) was calculated by using Brook’s equation for boys aged 쏝 12 y (7):

BD ⫽ 1.1533 ⫺ 0.0643 · Log (⌺4ST)

and the equation of Durnin and Rahaman (8) for boys aged 쏜 12 y:

SUBJECTS AND METHODS

BD ⫽ 1.1690 ⫺ 0.0788 · Log (⌺4ST)

Subjects

Methods z Scores (weight-for-age, weight-for-height, and height-forage) were calculated by using reference values from French growth charts (2). Body mass index (BMI; in kg/m2) was expressed in z scores by using reference data for a French population (3). Obesity cutoffs were BMI 쏜 30, according to the growth charts of Cole et al (4)—ie, an age-specific BMI equivalent to a BMI 쏜 30 in an 18-y-old person, as established by the International Obesity Task Force (5)—and %FM associated with BMI cutoffs for obesity—ie, age-specific BMI equivalent to a BMI of 30 in an 18-y-old (5). BIA measurements were performed using a monofrequency (50 kHz, 800 mAmp) unit (101Q; RJL Systems, Clinton Township, MI). FFM was calculated from the resistance index (RI) by the calculation RI ҃ H2/R, where H ҃ height in cm and R ҃ resistance in ⍀, and by using the equation of Houtkooper et al (6):

(1)

where X ҃ reactance in ⍀. ST measurements were performed at 4 sites (prebicipital, retrotricipital, suprailiac, and subscapular) by using a Harpenden Caliper (John Bull; British Indicators Ltd, St Albans, UK). ST measurements were carried out by 2 different observers. At the Paris site, measurements were taken by the principal investigator, and, at the Lille site, they were made by a researcher (trained by the principal investigator). Both observers were experienced

(3)

in both of which ⌺4ST ҃ the sum of the 4 ST measurements. Fat mass expressed as a percentage of body weight (%FM) was obtained from Siri’s equation (9):

%FM ⫽ (4.95/BD ⫺ 4.5) · 100

(4)

Thus, the method for estimating body composition provides FFM estimates as in BIA or %FM estimates as in ST measurement, with the use of the following equations:

Body weight (BW) ⫽ FFM ⫹ FM

(5)

%FM ⫽ (FM/BW) · 100

(6)

or

Labeled water (H218O 2%; Euriso-top, CEA Group, SaintAubin, France) was prepared according to good pharmaceutical practices in the pharmacy of each hospital. The dose given was 1 mL/kg. 18O enrichments in the urine were measured by using an isotope ratio–mass spectrometer (␮Gas-Optima; MICROMASS, Manchester, United Kingdom) according to the H2OCO2 equilibration method described by Vaché et al (10). FFM was calculated from total body water (TBW) by assuming a 75% water content of FFM at this age (ie, x៮ 10 y) (11, 12). Expected TBW was calculated as 64% of total body weight (11, 12). Muscle mass was estimated from creatinine excretion by assuming that the excretion of 1 g urinary creatinine/d represents 20 kg muscle mass (13). Statistical analysis BIA and ST measurement methods were compared with the reference method as described by Bland and Altman (14). Limits of agreement were defined as the mean (앐 2 SDs) for the difference between methods. 95% CIs were defined as the mean 앐 2.23 SEMs for the difference between methods with t equal to 2.23 for 10 df and a significance level of 0.05. The effect of method was tested for significance by using a one-way analysis of variance, followed by post hoc comparisons of means by using Tukey’s test. Differences in proportions were tested for significance by using the chi-square test. Data are presented as means 앐 SD and as proportions (percentages) for continuous and categorical variables, respectively. Statistical analysis was performed by using Statview software (version 5.0; Abacus Concepts, Berkeley, CA). RESULTS

Eleven children with DMD were included in the study (n ҃ 2 at Robert Debré Hospital and n ҃ 9 at Lille University Hospital). The children had a mean age of 10.0 앐 2.5 y. The anthropometric and body-composition data of the children are shown in Table 1.


Patients were recruited among those followed in multidisciplinary outpatient facilities from university hospitals in Paris and Lille, France. All patients were diagnosed with DMD by muscle biopsy, molecular biology, or both. Children were admitted the night before the study to the Clinical Investigation Centre of Robert Debré Hospital in Paris and of Lille University Hospital in Lille. The sequence of events for the experiment was as follows: After voiding and while in the postabsorptive state, patients were asked to drink labeled water. Physical exam including BIA measurement was then performed. Urine collection was done 3– 4 h after ingestion of labeled water. Urine samples collected for labeled water technique were analyzed in a single laboratory, eliminating any potential site differences. Routine urinanalysis was conducted at the individual hospitals. Both laboratories adhered to the French national standards for quality control for biological and medical analyses. All families gave written informed consent after the investigator thoroughly explained the study protocol to both the parents and the children. The protocol was approved by the Paris-Bichat Ethics Committee (Comité Consultatif pour la Protection de la Personne dans la Recherche Biomédicale) and conducted according to the principles of the Declaration of Helsinki.

FFM (kg) ⫽ 0.61 RI ⫹ 0.25 X ⫹ 1.31

(2)

BIA IN CHILDREN WITH DUCHENNE MD

67

TABLE 1 Anthropometric and body-composition measures in children with Duchenne muscular dystrophy (DMD)1 Children with DMD (n ҃ 11) Weight (kg) Weight-for-age z score Height (cm) Weight-for-height z score Height-for-age z score BMI (kg/m2) z Score Muscle mass (kg) Total body water (kg) Expected total body water (kg) 1

33.3 앐 12.3 0.2 앐 1.6 130.6 앐 11.2 1.8 앐 3.0 Ҁ0.5 앐 1.3 19.1 앐 5.1 1.3 앐 2.8 4.5 앐 0.9 13.6 앐 1.9 20.6 앐 7.5

All values are x៮ 앐 SD.

and Altman, the limits of agreement for the difference in FFM between BIA and WD was 3.3 앐 6.0 kg—ie, Ҁ2.7–9.3 kg—and the 95% CI was 0.79, 4.93 kg (Figure 2). This difference in FFM was larger when ST measurements were compared with WD measurements (limits of agreement for the difference in FFM between ST and WD: 6.3 앐 9.9 kg—ie, Ҁ3.6 –16.2 kg and 95% CI: 2.16, 8.62 kg) than when BIA was compared with WD. A similar trend was observed when the methods were compared with respect to obesity diagnosis. Specifically, obesity defined by %FM associated with BMI cutoffs for obesity (eg, %FM 쏜 35% for a 10-y-old) was present in 8 (73%) and 6 (55%) of 11 chilldren for WD and BIA methods, respectively, compared with 1 (9%) of 11 children for ST measurements (P 쏝 0.01). DISCUSSION

FIGURE 1. Mean (앐SD) fat-free mass (FFM) and percentage fat mass (%FM) estimates calculated by using skinfold-thickness (ST) measurements, bioelectrical impedance analysis (BIA), and the water-dilution (WD) reference method in 11 children with Duchenne muscular dystrophy. *, **Significantly different from WD reference method (Tukey’s test after ANOVA on 3 groups): *P 쏝 0.01, **P 쏝 0.05.

The BIA method provided estimates of body composition closer to those of the reference method than to those of ST measurement. Specifically, ST measurement overestimated FFM and underestimated %FM. Furthermore, BIA differed from the reference method less than did ST. Although clinicians must be warned of its limitations, the BIA method provides a more appropriate tool for assessing body composition than does ST measurement, particularly for early detection of fat accumulation and for longitudinal monitoring in DMD patients. TBW estimates were 66% of expected values obtained from reference data in boys (11, 12) and consistent with those from other studies (15, 16). Previous results showed that muscle mass estimations accounted for a 25% decrease from normal values (ie, an 8.5-kg muscle deficit; 17). Although reduced muscle mass may contribute to decreased TBW (75% muscle water content), the correlation between these 2 variables was reported to be weak (15, 16). In fact, TBW is made of intracellular and extracellular water. Muscle mass correlates with intracellular water and total body potassium and declines at a rate of 4%/y, which reflects the progression of the disease (18 –20). At variance, exchangeable sodium and extracellular water grow with age, which suggests the replacement of muscle by connective tissue, as observed by histologic experiments in DMD (21). In clinical practice, BIA is


Obesity was present in 2 (18%) of 11 children using BMI 쏜 30 (International Obesity Task Force) calculated according to Cole’s reference curves. The effect of the method used on estimates of both FFM and %FM was significant. Compared with the reference method, ST measurement provided significantly higher estimates of FFM (ST: 24.5 앐 5.9; WD: 18.2 앐 2.5 kg) (Figure 1) and lower estimates of %FM (ST: 23.3 앐 10.4; WD: 40.1 앐 17.1%) (Figure 1). In contrast, FFM and %FM estimates from BIA method (FFM: 21.5 앐 4.5 kg; %FM: 31.3 앐 13.9%) did not significantly differ from those from the reference method (Figure 1). In addition, fat mass was 8.9 앐 6.6, 11.8 앐 8.2, and 15.1 앐 10.8 kg, respectively, for ST, BIA, and WD methods. According to Bland

FIGURE 2. Bland-Altman plot of the mean difference between fat-free mass (FFM) calculated by using bioelectrical impedance analysis (BIA) and FFM obtained with the labeled water-dilution (WD) reference method in 11 children with Duchenne muscular dystrophy (the data points for 2 children nearly completely overlap at mean FFM-BIA and FFM-WD of 앒22 kg). Lower- and upper-limit 95% CIs show the highest risk of dispersion (mean 앐 2 SD).

68

MOK ET AL

dietary counseling as well as monitor progress and perhaps reduce excess adiposity in DMD patients. Moreover, in DMD, as in other situations it is usually preferable to prevent rather than correct obesity. Finally, the BIA method may serve to evaluate the effect of new treatments on body composition (26). For instance, in an effort to preserve muscle mass and reduce overweight in obese DMD patients, an intervention study could test the effect of a specific diet therapy on body composition. BIA is therefore an easy and simple tool that provides useful indicators in nutritional follow-up of DMD patients as well as in clinical research. Furthermore, it would be worthwhile to test its use as a method of assessing body composition in other diseases associated with increased muscle protein catabolism, eg, cystic fibrosis (27), wherein body compartments may differ from normal. We thank the children who participated in the study and their families and the personnel of both clinical investigation units and pharmacies in Paris and Lille. EM contributed to the analysis of the data and the writing of the manuscript. LB, PG, CD, J-EF, J-MC, and FG contributed to the collection of the data. RH contributed to the design of the experiment, the collection and analysis of the data, and the writing of the manuscript. None of the authors had a personal or financial conflict of interest.

REFERENCES 1. Willig TN, Carlier L, Legrand M, Riviere H, Navarro J. Nutritional assessment in Duchenne muscular dystrophy. Dev Med Child Neurol 1993;35:1074 – 82. 2. Sempe M, Pédron G, Roy-Pernot MP. Auxologie: méthode et séquences. (Auxology: methods and sequences.) Paris: Laboratoire Théraplix, 1979 (in French). 3. Rolland-Cachera MF, Sempe M, Guilloud-Bataille M, Patois E, Pequignot-Guggenbuhl F, Fautrad V. Adiposity indices in children. Am J Clin Nutr 1982;36:178 – 84. 4. Cole TJ, Bellizzi MC, Flegal KM, Dietz WH. Establishing a standard definition for child overweight and obesity worldwide: international survey. BMJ 2000;320:1240 –3. 5. Taylor RW, Jones IE, Williams SM, Goulding A. Body fat percentages measured by dual-energy X-ray absorptiometry corresponding to recently recommended body mass index cutoffs for overweight and obesity in children and adolescents aged 3–18 y. Am J Clin Nutr 2002;76: 1416 –21. 6. Houtkooper LB, Lohman TG, Going SB, Hall MC. Validity of bioelectric impedance for body composition assessment in children. J Appl Physiol 1989;66:814 –21. 7. Brook CG. Determination of body composition of children from skinfold measurements. Arch Dis Child 1971;46:182– 4. 8. Durnin JV, Rahaman MM. The assessment of the amount of fat in the human body from measurements of skinfold thickness. Br J Nutr 1967; 21:681–9. 9. Siri WE. Body composition from fluid spaces and density: analysis of methods. Nutrition 1993;9:480 –91. 10. Vache C, Gachon P, Ferry M, Beaufrere B, Ritz P. Low-cost measurement of body composition with 18O-enriched water. Diabet Metab 1995; 21:281– 4. 11. Fomon SJ, Haschke F, Ziegler EE, Nelson SE. Body composition of reference children from birth to age 10 years. Am J Clin Nutr 1982;35: 1169 –75. 12. Haschke F. Body composition of adolescent males. Part I. Total body water in normal adolescent males. Part II. Body composition of the male reference adolescent Acta Paediatr Scand Suppl 1983;307:1–23. 13. Griggs RC, Forbes G, Moxley RT, Herr BE. The assessment of muscle mass in progressive neuromuscular disease. Neurology 1983;33:158 – 65. 14. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986;1:307–10. 15. Blahd WH, Lederer M, Cassen B. The significance of decreased body


a better method than is ST measurement to estimate FFM (TBW) and %FM, but it cannot provide estimates of muscle mass. Because children with DMD had less TBW, the apparent normal weight observed in these children can be misleading, as it reflects excess body fat. Whichever method used, %FM exceeded normal values, which average 16% at this age. The physical exam is hence misleading in patients with DMD. To some extent, muscle mass loss masks FM accumulation. Moreover, in this population, fat accumulation cannot be detected by using nutritional indexes, such as BMI-for-age. This latter point was highlighted by Griffiths and Edwards (22), who adapted weightfor-age charts from those of Tanner and Whitehouse (23) and took into account muscle mass loss in DMD. The prevalence of obesity in patients with DMD is 쏜 50% by the age of 13 y (1). And, as do other obese populations, obese boys with DMD show a centralized body fat distribution (1). Furthermore, the higher total FM observed in boys with DMD than in healthy boys is mostly due to increased intramuscular fat deposition in both the central and the peripheral regions (24). A simple method for estimating body composition is therefore mandatory to estimate body FM and to monitor nutritional care in patients with DMD. Our study was not designed to validate a specific equation for children with DMD. Such studies require a large population, which is unrealistic with respect to rare diseases. In the current study, we compared the 2 most frequently used methods—ie, ST measurement and BIA—with a reference method and observed that ST measurements overestimated FFM, which led to an underestimation of %FM in these children. Similar findings were reported using magnetic resonance imaging as the reference method. Specifically, anthropometric measurements such as ST measurement underestimated the body fat percentage of children with DMD in comparison to measures by magnetic resonance imaging (24, 25). The ST method estimated the size of the subcutaneous fat depot to predict body composition as a whole with the use of equations validated against underwater densitometry as the reference technique (7, 8). The assumption that the fat distribution between peripheral and central regions is normal may not be borne out in DMD patients (24). As a consequence, FM estimated from ST measurement does not take into account fatty infiltration of muscles (intramuscular fat deposition) that occurs in DMD, and, hence, total-body FM is underestimated and, in turn, FFM is overestimated. This discrepancy highlights the importance of population-specific equations, particularly when body compartments are known to differ from normal, which could also occur in other diseases associated with increased muscle mass loss. At variance, BIA estimates body composition from a variable that encompasses the whole body—ie, TBW. BIA also has its own limitations, namely, the shape of the electrical model, cell membrane characteristics, and fraction of the current entering the intracellular space at different frequencies. However, these limitations do not prevent longitudinal comparisons, because errors cancel one another. Obesity aggravates handicap and complicates surgery in DMD children. Weight control is therefore crucial to limit the burden of the disease. Clinical evaluation of FM can, however, be misleading in patients with DMD. BIA is a simple and reliable technique, which can provide estimates of FFM and %FM closer to those obtained with the reference method. Hence, the BIA method can enable early detection of FM accumulation, begin

BIA IN CHILDREN WITH DUCHENNE MD

16. 17. 18. 19. 20. 21.

potassium concentrations in patients with muscular dystrophy and nondystrophic relatives. N Engl J Med 1967;276:1349 –52. Edmonds CJ, Smith T, Griffiths RD, Mackenzie J, Edwards RH. Total body potassium and water, and exchangeable sodium, in muscular dystrophy. Clin Sci (Lond) 1985;68:379 – 85. Hankard R, Gottrand F, Turck D, Carpentier A, Romon M, Farriaux JP. Resting energy expenditure and energy substrate utilization in children with Duchenne muscular dystrophy. Pediatr Res 1996;40:29 –33. Scott OM, Hyde SA, Goddard C, Dubowitz V. Quantitation of muscle function in children: a prospective study in Duchenne muscular dystrophy. Muscle Nerve 1982;5:291–301. Brooke MH, Fenichel GM, Griggs RC, et al. Clinical investigation in Duchenne dystrophy: 2. Determination of the “power” of therapeutic trials based on the natural history. Muscle Nerve 1983;6:91–103. Hosking JP, Bhat US, Dubowitz V, Edwards RH. Measurements of muscle strength and performance in children with normal and diseased muscle. Arch Dis Child 1976;51:957– 63. Jones DA, Round JM, Edwards RH, Grindwood SR, Tofts PS. Size and composition of the calf and quadriceps muscles in Duchenne muscular

22. 23. 24.

25. 26. 27.

69

dystrophy: a tomographic and histochemical study. J Neurol Sci 1983; 60:307–22. Griffiths RD, Edwards RH. A new chart for weight control in Duchenne muscular dystrophy. Arch Dis Child 1988;63:1256 – 8. Tanner JM, Whitehouse RH. Clinical longitudinal standards for height, weight, height velocity, weight velocity, and stages of puberty. Arch Dis Child 1976;51:170 –9. Leroy-Willig A, Willig TN, Henry-Feugeas MC, et al. Body composition determined with MR in patients with Duchenne muscular dystrophy, spinal muscular atrophy, and normal subjects. Magn Reson Imaging 1997;15:737– 44. Pichiecchio A, Uggetti C, Egitto MG, et al. Quantitative MR evaluation of body composition in patients with Duchenne muscular dystrophy. Eur Radiol 2002;12:2704 –9. Hankard RG, Hammond D, Haymond MW, Darmaun D. Oral glutamine slows down whole body protein breakdown in Duchenne muscular dystrophy. Pediatr Res 1998;43:222– 6. Roulet M. Protein-energy malnutrition in cystic fibrosis patients. Acta Paediatr Suppl 1994;83:43– 8.
