JCF-01011; No of Pages 4
Journal of Cystic Fibrosis xx (2014) xxx – xxx www.elsevier.com/locate/jcf
Short Communication
Agreement of bioelectric impedance analysis and dual-energy X-ray absorptiometry for body composition evaluation in adults with cystic fibrosis S. Ziai a,b , A. Coriati a,b , K. Chabot a,b , M. Mailhot c , M.V. Richter d , R. Rabasa-Lhoret a,b,c,e,⁎ a
c
Nutrition Department, Faculty of Medicine, Université de Montréal, Montréal, Québec, Canada b Institut de Recherches Cliniques de Montréal (IRCM), Montréal, Québec, Canada Cystic Fibrosis Clinic, Centre Hospitalier de l'Université de Montréal (CHUM) & CHUM Research Center (CR-CHUM), Montréal, Québec, Canada d Department of Medicine, Faculty of Medicine and Health Sciences, Université de Sherbrooke, Sherbrooke, Québec, Canada e Montreal Diabetes Research Centre (MDRC), Montréal, Québec, Canada Received 18 November 2013; revised 7 January 2014; accepted 17 January 2014
Abstract Malnutrition in cystic fibrosis (CF) is associated with increased mortality and can lead to fat-free (FFM) and fat mass (FM) loss. Dual-energy X-ray absorptiometry (DXA) is used and validated to measure FFM and FM. DXA's high cost has led to the utilization of less costly techniques such as bioelectrical impedance analysis (BIA). The aim of this study was to determine the agreement of FFM, FM and %FM measurements taken with DXA and BIA in adults with CF. We measured FFM, FM and %FM in 34 adults with CF with a leg-to-leg BIA and an iDXA and determined agreement using Bland–Altman analysis. While DXA and BIA measurements were well correlated (r N 0.8), mean biases between both methods were between 8 and 11%. BIA underestimated FM and %FM and overestimated FFM. In a clinical research setting where these measurements are used to phenotype patients, BIA cannot replace DXA. © 2014 Elsevier B.V. All rights reserved. on behalf of European Cystic Fibrosis Society. Keywords: Body composition; Bioelectrical impedance analysis; Dual-energy X-ray absorptiometry; Bland–Altman analysis
1. Introduction Cystic fibrosis (CF) is the most common autosomal genetic disease among Caucasians [1] and is associated with exocrine pancreatic insufficiency [2] and malnutrition [3]. Malnutrition can lead to fat-free (FFM; bones, muscles and organ mass) and fat mass (FM) loss [4]. Studies have suggested that loss of FFM is associated with lung disease and disease severity [3,5,6], and that body mass index (BMI) is not sensitive enough to detect its depletion [5,6]. It is important to accurately detect malnutrition by measuring FFM and FM losses as they are linked with decreased lung functions and increased mortality in CF [3]. Dual-energy X-ray absorptiometry (DXA) was first developed to evaluate bone mass but is also widely used and validated to ⁎ Corresponding author at: IRCM, 110, avenue des Pins Ouest, Montréal, Québec H2W 1R7, Canada. Tel.: +1 514 987 5762; fax: +1 514 987 5670. E-mail address: [email protected] (R. Rabasa-Lhoret).
measure FFM and FM [7]. The DXA uses X-rays with two different energy levels. Then, the energy of the attenuated rays is used in equations and the DXA can determine if the matter scanned is either fat-free mass, fat mass or bone [8]. DXA's high cost has led to the utilization of less costly techniques such as bioelectric impedance analysis (BIA) for body composition analysis. Leg-to-leg BIA uses an electric current that runs from one foot to the other and uses the resistance of mass to determine if it is either FM or FFM [9]. Three studies in patients with CF have compared these two techniques and reported discordant results [10,12]. While Pichard et al. reported good agreement of FFM between BIA and DXA [10], King et al. showed that BIA incorrectly estimated FFM in adults with CF [11]. Furthermore, Beaumesnil et al. found significant differences of FFM and FM between both methods in children and adults with CF [12]. None of these studies evaluated the accuracy of BIA to measure FFM, FM and percentage FM
1569-1993/$ -see front matter © 2014 Elsevier B.V. All rights reserved. on behalf of European Cystic Fibrosis Society. http://dx.doi.org/10.1016/j.jcf.2014.01.006 Please cite this article as: Ziai S, et al, Agreement of bioelectric impedance analysis and dual-energy X-ray absorptiometry for body composition evaluation in adults with cystic fibrosis, J Cyst Fibros (2014), http://dx.doi.org/10.1016/j.jcf.2014.01.006
2
S. Ziai et al. / Journal of Cystic Fibrosis xx (2014) xxx–xxx
(%FM) in a group of adults with CF. Therefore, the aim of this study was to determine the agreement between measurements of FFM, FM and %FM taken with BIA and DXA in adults with CF. 2. Material and methods This cross sectional study included 34 adults with CF (15 men and 19 women) and is a sub-analysis of a larger project studying CF-related diabetes. All CF subjects were recruited from the Centre Hospitalier de l'Université de Montréal (CHUM) and tested at the Institut de Recherches Cliniques de Montréal (IRCM). Participants were excluded if they were pregnant or having a pulmonary exacerbation diagnosed by a CF pulmonologist of the CHUM and defined by a change in sputum production (volume, colour, consistency), new or increased haemoptysis, increased cough, increased dyspnoea, fatigue or lethargy, fever N 38 °C, anorexia, sinus pain, a N 10% decrease in FEV1 compared to previously recorded values, intravenous antibiotic treatment and change in chest sounds [13]. The protocol was approved by the Research Ethics Committees of the CHUM and IRCM. Study participants did not fast prior to the tests but did not ingest caffeine or do any type of physical exercise 8 h before. Tests were done while participants were wearing light clothing and no metal objects. After measuring weight and height, all subjects performed a barefoot leg-to-leg BIA using a Tanita Body Composition Analyzer TBF-310 (Tanita Corporation of America, Arlington Heights, IL, USA) and a DXA using a Lunar iDXA (GE HealthCare, Mississauga, ON, Canada) to estimate FFM, FM and %FM. For the DXA, all subjects were lying supine on a flat couch for 15–30 min. The DXA was calibrated with a phantom every morning before scans. 2.1. Statistics Data are expressed in either mean ± standard deviation (SD) (for age, BMI) or median ± interquartile range (IQR) (for FFM, FM, %FM) depending on whether they are normally distributed. Statistical analysis was done with R (R 2.13.0). We compared measures of FFM, FM and %FM taken with DXA and BIA
between men and women using Wilcoxon signed-rank tests. We corrected for multiple comparisons and considered that a p ≤ 0.0083 was significant (p = 0.05/6 comparisons = 0.0083). We also associated FFM, FM and %FM measured by DXA and BIA using Spearman correlations. Then, Bland–Altman (BA) analysis was used to evaluate agreement of both methods [14]. Bias was defined as the percentage difference between both methods ((BIA − DXA) / DXA ∗ 100) for measures of FFM, FM and %FM and limits of agreement were ± 2 standard deviations (SDs) of the bias. 3. Results Characteristics of 34 adults with CF included in the study are shown in Table 1. Study participants were 30 ± 9 years old and had a wide range of BMIs varying from 17.8 to 27.9 kg/m2 with an average of 22.0 ± 2.56 kg/m2. Average FFM, FM and %FM are also shown in Table 1 and they were statistically different between men and women (p ≤ 0.0083). Fig. 1 illustrates the associations between measures of FFM (A), FM (B) and %FM (C) obtained with DXA (D) and BIA (B). The associations between FFM, FM and %FM measurements were strong with correlation coefficients of 0.915, 0.914 and 0.833 respectively. The mean bias for FFM was − 8.04%, 10.2% for FM and 9.79% for %FM. BA analysis plots showing the distribution of the biases for FFM (A), FM (B) and %FM (C) are presented in Fig. 2. The biases of FFM are heterogeneously distributed. There is a trend for BIA overestimating FFM for people with b 40 kg of FFM and underestimating for the others. The distributions of FM and %FM biases are also heterogeneous. BIA seems to underestimate FM and %FM in individuals with b 20 kg (~ 15%) of FM. Results of mean biases for FM, %FM and FFM were similar between men and women (results not shown). 4. Discussion Although measures taken with both techniques were highly correlated, mean biases between both methods were between 8 and 11% for all three measurements using BA analysis.
Table 1 Characteristics of subjects with cystic fibrosis (CF).
Age (years) BMI (kg/m2) FFM (kg) a FM (kg) a %FM a FFM (kg) b FM (kg) b %FM b
Total (n = 34)
Men (n = 15)
Women (n = 19)
30 ± 9 [20–48] 22.0 ± 2.56 [17.8–27.9] 43.4 ± 10.3 [29.8–66.4] 14.2 ± 8.27 [4.82–27.6] 23.5 ± 12.38 [8.8–38.8] 46.6 ± 8.85 [37.8–61.2] 12.9 ± 8.65 [3.8–25.4] 21.2 ± 13.3 [7.6–36.3]
29.1 22.0 47.7 8.91 14.8 50.4 8.20 13.9
30.8 22.0 39.2 16.4 28.7 41.6 14.6 26.7
± ± ± ± ± ± ± ±
8.72 [20–45] 2.98 [18.1–27.9] 7.51 [40.7–66.4] 6.85 [4.82–27.6] 6.45 [8.80–31.7] 8.00 [42.8–69.0] 8.70 [3.8–25.4] 10.0 [7.6–29.4]
± ± ± ± ± ± ± ±
8.97 [20–54] 2.25 [17.8–26.1] 6.56 [29.8–50.7] # 6.47 [9.00–25.3] # 8.90 [17.1–38.8] # 4.10 [37.8–47.8] # 8.90 [7.60–25.2] # 9.00 [16.3–36.3] #
BMI: body mass index, FM: fat mass, FFM: fat free mass. a Measured with dual-energy X-ray absorptiometry. b Measured with bioimpedance analysis. # p ≤ 0.0083 vs men. Please cite this article as: Ziai S, et al, Agreement of bioelectric impedance analysis and dual-energy X-ray absorptiometry for body composition evaluation in adults with cystic fibrosis, J Cyst Fibros (2014), http://dx.doi.org/10.1016/j.jcf.2014.01.006
S. Ziai et al. / Journal of Cystic Fibrosis xx (2014) xxx–xxx
A)
3
A) r=0.915 + 2 SD
Mean Bias =-8.04%
-2 SD
B)
B) r=0.914
+ 2 SD
Mean Bias =10.2%
-2 SD
C)
C) r=0.833
+ 2 SD
Mean Bias =9.79%
-2 SD
Fig. 1. Spearman correlations of body composition measurements ((A) fat-free mass (FFM), (B) fat mass (FM), (C) percentage fat mass (%FM)) taken by bioelectric impedance analysis (B) and dual-energy X-ray absorptiometry (D) in adults with cystic fibrosis.
Fig. 2. Bland–Altman analysis plots of the percentage difference between bioelectric impedance analysis (BIA) and dual-energy X-ray absorptiometry (DXA) in measuring (A) fat-free mass (FFM), (B) fat mass (FM), and (C) percentage fat mass (%FM)) in adults with cystic fibrosis.
BIA underestimated FM and %FM and overestimated FFM. Assuming that BIA error is constant over time, these differences might not have a significant impact if clinicians are evaluating the differences of body composition measures from one visit to the other. However, in a research setting, where precise FFM, FM and %FM values are used to phenotype patients, BIA may not replace DXA. Our study is the first to compare FFM, FM and %FM measurements taken with DXA and BIA in adults with CF. The aim of this study was to determine the agreement of body composition measurements taken with two widely used non-invasive methods. If both techniques were comparable,
we could replace the DXA with minimal X-ray exposure in a population already highly exposed [15] with the leg-to-leg BIA. However, in line with results reported by King et al. [11], the biases found are relevant in a research setting making it inappropriate to replace DXA with the BIA. These biases could be explained, by altered sodium content in the sweat of people with CF [9]. To overcome this problem, the use of specific equations has been suggested [9], however most commercial devices do not allow using specific formulas for a sub-group of patients. As bias could be affected by the amount of FFM and FM, future studies should include patients with low, normal and high FFM and FM.
Please cite this article as: Ziai S, et al, Agreement of bioelectric impedance analysis and dual-energy X-ray absorptiometry for body composition evaluation in adults with cystic fibrosis, J Cyst Fibros (2014), http://dx.doi.org/10.1016/j.jcf.2014.01.006
4
S. Ziai et al. / Journal of Cystic Fibrosis xx (2014) xxx–xxx
In conclusion, we suggest that DXA cannot be replaced by BIA to measure FFM, FM and %FM in adults with CF in a research setting. Conflicts of interest No conflicts of interest to report. Acknowledgement S.Z. has a Banting and Best Doctoral Scholarship from the Canadian Institutes of Health Research and A.C. is supported by the Jacques Gauthier scholarship of the Institut de Recherches Cliniques de Montréal (IRCM). R.R.-L. is a senior scholar of the FRQ-S (Fonds de Recherches du Québec en santé) and holds the J-A De Sève Research Chair. M.R. is a Junior 2 Scholar of the FRQ-S. This work was funded by a Cystic Fibrosis Canada team grant. References [1] Davis PB. Cystic fibrosis since 1938. Am J Respir Crit Care Med 2006 Mar 1;173(5):475–82. [2] Quinton PM. Cystic fibrosis: lessons from the sweat gland. Physiology (Bethesda) 2007 Jun;22:212–25. [3] Pencharz PB, Durie PR. Pathogenesis of malnutrition in cystic fibrosis, and its treatment. Clin Nutr 2000 Dec;19(6):387–94. [4] Hickson M. Malnutrition and ageing. Postgrad Med J 2006;82:2–8.
[5] King SJ, Nyulasi IB, Strauss BJG, Kotsimbos T, Bailey M, Wilson JW. Fatfree mass depletion in cystic fibrosis: associated with lung disease severity but poorly detected by body mass index. Nutrition 2010;26:753–9. [6] Ionescu AA, Evans WD, Pettit RJ, Nixon LS, Stone MD, Shale DJ. Hidden depletion of fat-free mass and bone mineral density in adults with cystic fibrosis. Chest 2003;124:2220–8. [7] Norcross J, Van Loan MD. Validation of fan beam dual energy X-ray absorptiometry for body composition assessment in adults aged 18–45 years. Br J Sports Med 2004;38:472–6. [8] Bertin E, Ruiz JC, Mourot J, Peiniau P, Portha B. Evaluation of dualenergy X-ray absorptiometry for body composition assessment in rats. J Nutr 1998;126:1550–4. [9] Azcue M, Fried M, Pencharz PB. Use of bioelectrical impedance analysis to measure total body water in patients with cystic fibrosis. J Pediatr Gastroenterol Nutr 1993;16:440–5. [10] Pichard C, Kyle UG, Slosman DO. Fat-free mass in chronic illness: comparison of bioelectrical impedance and dual-energy X-ray absorptiometry in 480 chronically ill and healthy subjects. Nutrition 1999;15:668–76. [11] King S, Wilson J, Kotsimbos T, Bailey M, Nyulasi I. Body composition assessment in adults with cystic fibrosis: comparison of dual-energy X-ray absorptiometry with skinfolds and bioelectrical impedance analysis. Nutrition 2005;21:1087–94. [12] Beaumesnil M, Chaillou E, Wagner AC, Rouquette A, Audran M, Giniès JL. Body composition analysis in patients with cystic fibrosis. Comparison of 3 methods: dual-energy X-ray absorptiometry, bioelectrical impedance analysis, and skinfolds measurements. Arch Pediatr 2011;18:370–5. [13] Costa M, Potvin S, Hammana I, Malet A, Berthiaume Y, Jeanneret A, et al. Increased glucose excursions in cystic fibrosis and its association with a worse clinical status. J Cyst Fibros 2007;6:376–83. [14] Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986;8:307–10. [15] O'Connell OJ, McWilliams S, McGarrigle A, O'Connor OJ, Shanahan F, Mullane D, et al. Cumulative effective dose and changing trends over 2 decades. Chest 2012;141:1575–83.
Please cite this article as: Ziai S, et al, Agreement of bioelectric impedance analysis and dual-energy X-ray absorptiometry for body composition evaluation in adults with cystic fibrosis, J Cyst Fibros (2014), http://dx.doi.org/10.1016/j.jcf.2014.01.006
Journal of Cystic Fibrosis xx (2014) xxx – xxx www.elsevier.com/locate/jcf
Short Communication
Agreement of bioelectric impedance analysis and dual-energy X-ray absorptiometry for body composition evaluation in adults with cystic fibrosis S. Ziai a,b , A. Coriati a,b , K. Chabot a,b , M. Mailhot c , M.V. Richter d , R. Rabasa-Lhoret a,b,c,e,⁎ a
c
Nutrition Department, Faculty of Medicine, Université de Montréal, Montréal, Québec, Canada b Institut de Recherches Cliniques de Montréal (IRCM), Montréal, Québec, Canada Cystic Fibrosis Clinic, Centre Hospitalier de l'Université de Montréal (CHUM) & CHUM Research Center (CR-CHUM), Montréal, Québec, Canada d Department of Medicine, Faculty of Medicine and Health Sciences, Université de Sherbrooke, Sherbrooke, Québec, Canada e Montreal Diabetes Research Centre (MDRC), Montréal, Québec, Canada Received 18 November 2013; revised 7 January 2014; accepted 17 January 2014
Abstract Malnutrition in cystic fibrosis (CF) is associated with increased mortality and can lead to fat-free (FFM) and fat mass (FM) loss. Dual-energy X-ray absorptiometry (DXA) is used and validated to measure FFM and FM. DXA's high cost has led to the utilization of less costly techniques such as bioelectrical impedance analysis (BIA). The aim of this study was to determine the agreement of FFM, FM and %FM measurements taken with DXA and BIA in adults with CF. We measured FFM, FM and %FM in 34 adults with CF with a leg-to-leg BIA and an iDXA and determined agreement using Bland–Altman analysis. While DXA and BIA measurements were well correlated (r N 0.8), mean biases between both methods were between 8 and 11%. BIA underestimated FM and %FM and overestimated FFM. In a clinical research setting where these measurements are used to phenotype patients, BIA cannot replace DXA. © 2014 Elsevier B.V. All rights reserved. on behalf of European Cystic Fibrosis Society. Keywords: Body composition; Bioelectrical impedance analysis; Dual-energy X-ray absorptiometry; Bland–Altman analysis
1. Introduction Cystic fibrosis (CF) is the most common autosomal genetic disease among Caucasians [1] and is associated with exocrine pancreatic insufficiency [2] and malnutrition [3]. Malnutrition can lead to fat-free (FFM; bones, muscles and organ mass) and fat mass (FM) loss [4]. Studies have suggested that loss of FFM is associated with lung disease and disease severity [3,5,6], and that body mass index (BMI) is not sensitive enough to detect its depletion [5,6]. It is important to accurately detect malnutrition by measuring FFM and FM losses as they are linked with decreased lung functions and increased mortality in CF [3]. Dual-energy X-ray absorptiometry (DXA) was first developed to evaluate bone mass but is also widely used and validated to ⁎ Corresponding author at: IRCM, 110, avenue des Pins Ouest, Montréal, Québec H2W 1R7, Canada. Tel.: +1 514 987 5762; fax: +1 514 987 5670. E-mail address: [email protected] (R. Rabasa-Lhoret).
measure FFM and FM [7]. The DXA uses X-rays with two different energy levels. Then, the energy of the attenuated rays is used in equations and the DXA can determine if the matter scanned is either fat-free mass, fat mass or bone [8]. DXA's high cost has led to the utilization of less costly techniques such as bioelectric impedance analysis (BIA) for body composition analysis. Leg-to-leg BIA uses an electric current that runs from one foot to the other and uses the resistance of mass to determine if it is either FM or FFM [9]. Three studies in patients with CF have compared these two techniques and reported discordant results [10,12]. While Pichard et al. reported good agreement of FFM between BIA and DXA [10], King et al. showed that BIA incorrectly estimated FFM in adults with CF [11]. Furthermore, Beaumesnil et al. found significant differences of FFM and FM between both methods in children and adults with CF [12]. None of these studies evaluated the accuracy of BIA to measure FFM, FM and percentage FM
1569-1993/$ -see front matter © 2014 Elsevier B.V. All rights reserved. on behalf of European Cystic Fibrosis Society. http://dx.doi.org/10.1016/j.jcf.2014.01.006 Please cite this article as: Ziai S, et al, Agreement of bioelectric impedance analysis and dual-energy X-ray absorptiometry for body composition evaluation in adults with cystic fibrosis, J Cyst Fibros (2014), http://dx.doi.org/10.1016/j.jcf.2014.01.006
2
S. Ziai et al. / Journal of Cystic Fibrosis xx (2014) xxx–xxx
(%FM) in a group of adults with CF. Therefore, the aim of this study was to determine the agreement between measurements of FFM, FM and %FM taken with BIA and DXA in adults with CF. 2. Material and methods This cross sectional study included 34 adults with CF (15 men and 19 women) and is a sub-analysis of a larger project studying CF-related diabetes. All CF subjects were recruited from the Centre Hospitalier de l'Université de Montréal (CHUM) and tested at the Institut de Recherches Cliniques de Montréal (IRCM). Participants were excluded if they were pregnant or having a pulmonary exacerbation diagnosed by a CF pulmonologist of the CHUM and defined by a change in sputum production (volume, colour, consistency), new or increased haemoptysis, increased cough, increased dyspnoea, fatigue or lethargy, fever N 38 °C, anorexia, sinus pain, a N 10% decrease in FEV1 compared to previously recorded values, intravenous antibiotic treatment and change in chest sounds [13]. The protocol was approved by the Research Ethics Committees of the CHUM and IRCM. Study participants did not fast prior to the tests but did not ingest caffeine or do any type of physical exercise 8 h before. Tests were done while participants were wearing light clothing and no metal objects. After measuring weight and height, all subjects performed a barefoot leg-to-leg BIA using a Tanita Body Composition Analyzer TBF-310 (Tanita Corporation of America, Arlington Heights, IL, USA) and a DXA using a Lunar iDXA (GE HealthCare, Mississauga, ON, Canada) to estimate FFM, FM and %FM. For the DXA, all subjects were lying supine on a flat couch for 15–30 min. The DXA was calibrated with a phantom every morning before scans. 2.1. Statistics Data are expressed in either mean ± standard deviation (SD) (for age, BMI) or median ± interquartile range (IQR) (for FFM, FM, %FM) depending on whether they are normally distributed. Statistical analysis was done with R (R 2.13.0). We compared measures of FFM, FM and %FM taken with DXA and BIA
between men and women using Wilcoxon signed-rank tests. We corrected for multiple comparisons and considered that a p ≤ 0.0083 was significant (p = 0.05/6 comparisons = 0.0083). We also associated FFM, FM and %FM measured by DXA and BIA using Spearman correlations. Then, Bland–Altman (BA) analysis was used to evaluate agreement of both methods [14]. Bias was defined as the percentage difference between both methods ((BIA − DXA) / DXA ∗ 100) for measures of FFM, FM and %FM and limits of agreement were ± 2 standard deviations (SDs) of the bias. 3. Results Characteristics of 34 adults with CF included in the study are shown in Table 1. Study participants were 30 ± 9 years old and had a wide range of BMIs varying from 17.8 to 27.9 kg/m2 with an average of 22.0 ± 2.56 kg/m2. Average FFM, FM and %FM are also shown in Table 1 and they were statistically different between men and women (p ≤ 0.0083). Fig. 1 illustrates the associations between measures of FFM (A), FM (B) and %FM (C) obtained with DXA (D) and BIA (B). The associations between FFM, FM and %FM measurements were strong with correlation coefficients of 0.915, 0.914 and 0.833 respectively. The mean bias for FFM was − 8.04%, 10.2% for FM and 9.79% for %FM. BA analysis plots showing the distribution of the biases for FFM (A), FM (B) and %FM (C) are presented in Fig. 2. The biases of FFM are heterogeneously distributed. There is a trend for BIA overestimating FFM for people with b 40 kg of FFM and underestimating for the others. The distributions of FM and %FM biases are also heterogeneous. BIA seems to underestimate FM and %FM in individuals with b 20 kg (~ 15%) of FM. Results of mean biases for FM, %FM and FFM were similar between men and women (results not shown). 4. Discussion Although measures taken with both techniques were highly correlated, mean biases between both methods were between 8 and 11% for all three measurements using BA analysis.
Table 1 Characteristics of subjects with cystic fibrosis (CF).
Age (years) BMI (kg/m2) FFM (kg) a FM (kg) a %FM a FFM (kg) b FM (kg) b %FM b
Total (n = 34)
Men (n = 15)
Women (n = 19)
30 ± 9 [20–48] 22.0 ± 2.56 [17.8–27.9] 43.4 ± 10.3 [29.8–66.4] 14.2 ± 8.27 [4.82–27.6] 23.5 ± 12.38 [8.8–38.8] 46.6 ± 8.85 [37.8–61.2] 12.9 ± 8.65 [3.8–25.4] 21.2 ± 13.3 [7.6–36.3]
29.1 22.0 47.7 8.91 14.8 50.4 8.20 13.9
30.8 22.0 39.2 16.4 28.7 41.6 14.6 26.7
± ± ± ± ± ± ± ±
8.72 [20–45] 2.98 [18.1–27.9] 7.51 [40.7–66.4] 6.85 [4.82–27.6] 6.45 [8.80–31.7] 8.00 [42.8–69.0] 8.70 [3.8–25.4] 10.0 [7.6–29.4]
± ± ± ± ± ± ± ±
8.97 [20–54] 2.25 [17.8–26.1] 6.56 [29.8–50.7] # 6.47 [9.00–25.3] # 8.90 [17.1–38.8] # 4.10 [37.8–47.8] # 8.90 [7.60–25.2] # 9.00 [16.3–36.3] #
BMI: body mass index, FM: fat mass, FFM: fat free mass. a Measured with dual-energy X-ray absorptiometry. b Measured with bioimpedance analysis. # p ≤ 0.0083 vs men. Please cite this article as: Ziai S, et al, Agreement of bioelectric impedance analysis and dual-energy X-ray absorptiometry for body composition evaluation in adults with cystic fibrosis, J Cyst Fibros (2014), http://dx.doi.org/10.1016/j.jcf.2014.01.006
S. Ziai et al. / Journal of Cystic Fibrosis xx (2014) xxx–xxx
A)
3
A) r=0.915 + 2 SD
Mean Bias =-8.04%
-2 SD
B)
B) r=0.914
+ 2 SD
Mean Bias =10.2%
-2 SD
C)
C) r=0.833
+ 2 SD
Mean Bias =9.79%
-2 SD
Fig. 1. Spearman correlations of body composition measurements ((A) fat-free mass (FFM), (B) fat mass (FM), (C) percentage fat mass (%FM)) taken by bioelectric impedance analysis (B) and dual-energy X-ray absorptiometry (D) in adults with cystic fibrosis.
Fig. 2. Bland–Altman analysis plots of the percentage difference between bioelectric impedance analysis (BIA) and dual-energy X-ray absorptiometry (DXA) in measuring (A) fat-free mass (FFM), (B) fat mass (FM), and (C) percentage fat mass (%FM)) in adults with cystic fibrosis.
BIA underestimated FM and %FM and overestimated FFM. Assuming that BIA error is constant over time, these differences might not have a significant impact if clinicians are evaluating the differences of body composition measures from one visit to the other. However, in a research setting, where precise FFM, FM and %FM values are used to phenotype patients, BIA may not replace DXA. Our study is the first to compare FFM, FM and %FM measurements taken with DXA and BIA in adults with CF. The aim of this study was to determine the agreement of body composition measurements taken with two widely used non-invasive methods. If both techniques were comparable,
we could replace the DXA with minimal X-ray exposure in a population already highly exposed [15] with the leg-to-leg BIA. However, in line with results reported by King et al. [11], the biases found are relevant in a research setting making it inappropriate to replace DXA with the BIA. These biases could be explained, by altered sodium content in the sweat of people with CF [9]. To overcome this problem, the use of specific equations has been suggested [9], however most commercial devices do not allow using specific formulas for a sub-group of patients. As bias could be affected by the amount of FFM and FM, future studies should include patients with low, normal and high FFM and FM.
Please cite this article as: Ziai S, et al, Agreement of bioelectric impedance analysis and dual-energy X-ray absorptiometry for body composition evaluation in adults with cystic fibrosis, J Cyst Fibros (2014), http://dx.doi.org/10.1016/j.jcf.2014.01.006
4
S. Ziai et al. / Journal of Cystic Fibrosis xx (2014) xxx–xxx
In conclusion, we suggest that DXA cannot be replaced by BIA to measure FFM, FM and %FM in adults with CF in a research setting. Conflicts of interest No conflicts of interest to report. Acknowledgement S.Z. has a Banting and Best Doctoral Scholarship from the Canadian Institutes of Health Research and A.C. is supported by the Jacques Gauthier scholarship of the Institut de Recherches Cliniques de Montréal (IRCM). R.R.-L. is a senior scholar of the FRQ-S (Fonds de Recherches du Québec en santé) and holds the J-A De Sève Research Chair. M.R. is a Junior 2 Scholar of the FRQ-S. This work was funded by a Cystic Fibrosis Canada team grant. References [1] Davis PB. Cystic fibrosis since 1938. Am J Respir Crit Care Med 2006 Mar 1;173(5):475–82. [2] Quinton PM. Cystic fibrosis: lessons from the sweat gland. Physiology (Bethesda) 2007 Jun;22:212–25. [3] Pencharz PB, Durie PR. Pathogenesis of malnutrition in cystic fibrosis, and its treatment. Clin Nutr 2000 Dec;19(6):387–94. [4] Hickson M. Malnutrition and ageing. Postgrad Med J 2006;82:2–8.
[5] King SJ, Nyulasi IB, Strauss BJG, Kotsimbos T, Bailey M, Wilson JW. Fatfree mass depletion in cystic fibrosis: associated with lung disease severity but poorly detected by body mass index. Nutrition 2010;26:753–9. [6] Ionescu AA, Evans WD, Pettit RJ, Nixon LS, Stone MD, Shale DJ. Hidden depletion of fat-free mass and bone mineral density in adults with cystic fibrosis. Chest 2003;124:2220–8. [7] Norcross J, Van Loan MD. Validation of fan beam dual energy X-ray absorptiometry for body composition assessment in adults aged 18–45 years. Br J Sports Med 2004;38:472–6. [8] Bertin E, Ruiz JC, Mourot J, Peiniau P, Portha B. Evaluation of dualenergy X-ray absorptiometry for body composition assessment in rats. J Nutr 1998;126:1550–4. [9] Azcue M, Fried M, Pencharz PB. Use of bioelectrical impedance analysis to measure total body water in patients with cystic fibrosis. J Pediatr Gastroenterol Nutr 1993;16:440–5. [10] Pichard C, Kyle UG, Slosman DO. Fat-free mass in chronic illness: comparison of bioelectrical impedance and dual-energy X-ray absorptiometry in 480 chronically ill and healthy subjects. Nutrition 1999;15:668–76. [11] King S, Wilson J, Kotsimbos T, Bailey M, Nyulasi I. Body composition assessment in adults with cystic fibrosis: comparison of dual-energy X-ray absorptiometry with skinfolds and bioelectrical impedance analysis. Nutrition 2005;21:1087–94. [12] Beaumesnil M, Chaillou E, Wagner AC, Rouquette A, Audran M, Giniès JL. Body composition analysis in patients with cystic fibrosis. Comparison of 3 methods: dual-energy X-ray absorptiometry, bioelectrical impedance analysis, and skinfolds measurements. Arch Pediatr 2011;18:370–5. [13] Costa M, Potvin S, Hammana I, Malet A, Berthiaume Y, Jeanneret A, et al. Increased glucose excursions in cystic fibrosis and its association with a worse clinical status. J Cyst Fibros 2007;6:376–83. [14] Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986;8:307–10. [15] O'Connell OJ, McWilliams S, McGarrigle A, O'Connor OJ, Shanahan F, Mullane D, et al. Cumulative effective dose and changing trends over 2 decades. Chest 2012;141:1575–83.
Please cite this article as: Ziai S, et al, Agreement of bioelectric impedance analysis and dual-energy X-ray absorptiometry for body composition evaluation in adults with cystic fibrosis, J Cyst Fibros (2014), http://dx.doi.org/10.1016/j.jcf.2014.01.006
