Estimation of body composition by bioelectrical impedance in HIV-infected children

Abstract

To the Editor:We read with interest the article by Arpadi et al. 1Arpadi SM Wang J Cuff PA Thornton J Horlick M Kotler DP et al.Application of bioimpedance analysis for estimating body composition in prepubertal children infected with human immunodeficiency virus type 1.J Pediatr. 1996; 129: 755-757Abstract Full Text Full Text PDF PubMed Scopus (21) Google Scholar and the related editorial 2Miller TL Nutritional assessment and its clinical application in children infected with HIV.J Pediatr. 1996; 129: 633-636Abstract Full Text Full Text PDF PubMed Scopus (51) Google Scholar on the use of bioelectrical impedance (BIA) in HIV-infected children. Both reports are suitably cautious in their conclusions with respect to the validity of BIA. The conclusion that it is necessary to develop BIA prediction equations that are specific to HIV-infected children is perhaps premature. BIA prediction equations are highly population-specific, 3Reilly JJ Wilson J McColl JH Carmichael M Durnin JVGA Ability of bioelectrical impedance to predict fat-free mass in prepubertal children.Pediatr Res. 1996; 39: 176-179Crossref PubMed Scopus (41) Google Scholar so the observation that several prediction equations systematically overestimate or underestimate fat-free mass in a sample of HIV-infected children is not in itself evidence that the problem is specific to HIV infection. We note that the direction and magnitude of errors that Arpadi et al. 1Arpadi SM Wang J Cuff PA Thornton J Horlick M Kotler DP et al.Application of bioimpedance analysis for estimating body composition in prepubertal children infected with human immunodeficiency virus type 1.J Pediatr. 1996; 129: 755-757Abstract Full Text Full Text PDF PubMed Scopus (21) Google Scholar observed when using some of the prediction equations are similar to our results when we applied the same equations to healthy prepubertal children in Scotland. 3Reilly JJ Wilson J McColl JH Carmichael M Durnin JVGA Ability of bioelectrical impedance to predict fat-free mass in prepubertal children.Pediatr Res. 1996; 39: 176-179Crossref PubMed Scopus (41) Google Scholar With the absence of control subjects, it is not possible to be confident that the problem is lack of applicability to HIV-infected children as distinct from a more general lack of “cross-validity” with BIA. 3Reilly JJ Wilson J McColl JH Carmichael M Durnin JVGA Ability of bioelectrical impedance to predict fat-free mass in prepubertal children.Pediatr Res. 1996; 39: 176-179Crossref PubMed Scopus (41) Google ScholarWe endorse the calls for more research in order to answer such questions. 1Arpadi SM Wang J Cuff PA Thornton J Horlick M Kotler DP et al.Application of bioimpedance analysis for estimating body composition in prepubertal children infected with human immunodeficiency virus type 1.J Pediatr. 1996; 129: 755-757Abstract Full Text Full Text PDF PubMed Scopus (21) Google Scholar, 2Miller TL Nutritional assessment and its clinical application in children infected with HIV.J Pediatr. 1996; 129: 633-636Abstract Full Text Full Text PDF PubMed Scopus (51) Google Scholar In assessing whether HIV-specific BIA equations are necessary, it would also be useful if greater consideration of the effect of clinical status, hydration status, and nutritional status of the children studied were presented. It is possible that errors in BIA prediction equations might vary across the range of nutritional status (e.g., from wasted to obese 3Reilly JJ Wilson J McColl JH Carmichael M Durnin JVGA Ability of bioelectrical impedance to predict fat-free mass in prepubertal children.Pediatr Res. 1996; 39: 176-179Crossref PubMed Scopus (41) Google Scholar). Information on the presence of clinical features of particular relevance (e.g., chronic diarrhea) would also be useful. Errors can also arise from differences in the reference method of body composition used between studies (e.g., dual-energy x-ray absorptiometry vs hydrodensitometry). Use of dual-energy x-ray absorptiometry as a “gold standard” method is premature. 4Jebb SA Measurement of soft tissue composition by dual energy x-ray absorptiometry.Br J Nutr. 1997; 77: 151-163Crossref PubMed Scopus (81) Google Scholar It should also be pointed out that the reference methods used to derive some of the BIA prediction equations tested by Arpadi et al. 1Arpadi SM Wang J Cuff PA Thornton J Horlick M Kotler DP et al.Application of bioimpedance analysis for estimating body composition in prepubertal children infected with human immunodeficiency virus type 1.J Pediatr. 1996; 129: 755-757Abstract Full Text Full Text PDF PubMed Scopus (21) Google Scholar were inappropriate in that adult values were used for hydration of fat-free mass. The prediction equations derived were therefore inappropriate for pediatric use, and the resulting systematic errors were inevitable. To the Editor:We read with interest the article by Arpadi et al. 1Arpadi SM Wang J Cuff PA Thornton J Horlick M Kotler DP et al.Application of bioimpedance analysis for estimating body composition in prepubertal children infected with human immunodeficiency virus type 1.J Pediatr. 1996; 129: 755-757Abstract Full Text Full Text PDF PubMed Scopus (21) Google Scholar and the related editorial 2Miller TL Nutritional assessment and its clinical application in children infected with HIV.J Pediatr. 1996; 129: 633-636Abstract Full Text Full Text PDF PubMed Scopus (51) Google Scholar on the use of bioelectrical impedance (BIA) in HIV-infected children. Both reports are suitably cautious in their conclusions with respect to the validity of BIA. The conclusion that it is necessary to develop BIA prediction equations that are specific to HIV-infected children is perhaps premature. BIA prediction equations are highly population-specific, 3Reilly JJ Wilson J McColl JH Carmichael M Durnin JVGA Ability of bioelectrical impedance to predict fat-free mass in prepubertal children.Pediatr Res. 1996; 39: 176-179Crossref PubMed Scopus (41) Google Scholar so the observation that several prediction equations systematically overestimate or underestimate fat-free mass in a sample of HIV-infected children is not in itself evidence that the problem is specific to HIV infection. We note that the direction and magnitude of errors that Arpadi et al. 1Arpadi SM Wang J Cuff PA Thornton J Horlick M Kotler DP et al.Application of bioimpedance analysis for estimating body composition in prepubertal children infected with human immunodeficiency virus type 1.J Pediatr. 1996; 129: 755-757Abstract Full Text Full Text PDF PubMed Scopus (21) Google Scholar observed when using some of the prediction equations are similar to our results when we applied the same equations to healthy prepubertal children in Scotland. 3Reilly JJ Wilson J McColl JH Carmichael M Durnin JVGA Ability of bioelectrical impedance to predict fat-free mass in prepubertal children.Pediatr Res. 1996; 39: 176-179Crossref PubMed Scopus (41) Google Scholar With the absence of control subjects, it is not possible to be confident that the problem is lack of applicability to HIV-infected children as distinct from a more general lack of “cross-validity” with BIA. 3Reilly JJ Wilson J McColl JH Carmichael M Durnin JVGA Ability of bioelectrical impedance to predict fat-free mass in prepubertal children.Pediatr Res. 1996; 39: 176-179Crossref PubMed Scopus (41) Google ScholarWe endorse the calls for more research in order to answer such questions. 1Arpadi SM Wang J Cuff PA Thornton J Horlick M Kotler DP et al.Application of bioimpedance analysis for estimating body composition in prepubertal children infected with human immunodeficiency virus type 1.J Pediatr. 1996; 129: 755-757Abstract Full Text Full Text PDF PubMed Scopus (21) Google Scholar, 2Miller TL Nutritional assessment and its clinical application in children infected with HIV.J Pediatr. 1996; 129: 633-636Abstract Full Text Full Text PDF PubMed Scopus (51) Google Scholar In assessing whether HIV-specific BIA equations are necessary, it would also be useful if greater consideration of the effect of clinical status, hydration status, and nutritional status of the children studied were presented. It is possible that errors in BIA prediction equations might vary across the range of nutritional status (e.g., from wasted to obese 3Reilly JJ Wilson J McColl JH Carmichael M Durnin JVGA Ability of bioelectrical impedance to predict fat-free mass in prepubertal children.Pediatr Res. 1996; 39: 176-179Crossref PubMed Scopus (41) Google Scholar). Information on the presence of clinical features of particular relevance (e.g., chronic diarrhea) would also be useful. Errors can also arise from differences in the reference method of body composition used between studies (e.g., dual-energy x-ray absorptiometry vs hydrodensitometry). Use of dual-energy x-ray absorptiometry as a “gold standard” method is premature. 4Jebb SA Measurement of soft tissue composition by dual energy x-ray absorptiometry.Br J Nutr. 1997; 77: 151-163Crossref PubMed Scopus (81) Google Scholar It should also be pointed out that the reference methods used to derive some of the BIA prediction equations tested by Arpadi et al. 1Arpadi SM Wang J Cuff PA Thornton J Horlick M Kotler DP et al.Application of bioimpedance analysis for estimating body composition in prepubertal children infected with human immunodeficiency virus type 1.J Pediatr. 1996; 129: 755-757Abstract Full Text Full Text PDF PubMed Scopus (21) Google Scholar were inappropriate in that adult values were used for hydration of fat-free mass. The prediction equations derived were therefore inappropriate for pediatric use, and the resulting systematic errors were inevitable. We read with interest the article by Arpadi et al. 1Arpadi SM Wang J Cuff PA Thornton J Horlick M Kotler DP et al.Application of bioimpedance analysis for estimating body composition in prepubertal children infected with human immunodeficiency virus type 1.J Pediatr. 1996; 129: 755-757Abstract Full Text Full Text PDF PubMed Scopus (21) Google Scholar and the related editorial 2Miller TL Nutritional assessment and its clinical application in children infected with HIV.J Pediatr. 1996; 129: 633-636Abstract Full Text Full Text PDF PubMed Scopus (51) Google Scholar on the use of bioelectrical impedance (BIA) in HIV-infected children. Both reports are suitably cautious in their conclusions with respect to the validity of BIA. The conclusion that it is necessary to develop BIA prediction equations that are specific to HIV-infected children is perhaps premature. BIA prediction equations are highly population-specific, 3Reilly JJ Wilson J McColl JH Carmichael M Durnin JVGA Ability of bioelectrical impedance to predict fat-free mass in prepubertal children.Pediatr Res. 1996; 39: 176-179Crossref PubMed Scopus (41) Google Scholar so the observation that several prediction equations systematically overestimate or underestimate fat-free mass in a sample of HIV-infected children is not in itself evidence that the problem is specific to HIV infection. We note that the direction and magnitude of errors that Arpadi et al. 1Arpadi SM Wang J Cuff PA Thornton J Horlick M Kotler DP et al.Application of bioimpedance analysis for estimating body composition in prepubertal children infected with human immunodeficiency virus type 1.J Pediatr. 1996; 129: 755-757Abstract Full Text Full Text PDF PubMed Scopus (21) Google Scholar observed when using some of the prediction equations are similar to our results when we applied the same equations to healthy prepubertal children in Scotland. 3Reilly JJ Wilson J McColl JH Carmichael M Durnin JVGA Ability of bioelectrical impedance to predict fat-free mass in prepubertal children.Pediatr Res. 1996; 39: 176-179Crossref PubMed Scopus (41) Google Scholar With the absence of control subjects, it is not possible to be confident that the problem is lack of applicability to HIV-infected children as distinct from a more general lack of “cross-validity” with BIA. 3Reilly JJ Wilson J McColl JH Carmichael M Durnin JVGA Ability of bioelectrical impedance to predict fat-free mass in prepubertal children.Pediatr Res. 1996; 39: 176-179Crossref PubMed Scopus (41) Google Scholar We endorse the calls for more research in order to answer such questions. 1Arpadi SM Wang J Cuff PA Thornton J Horlick M Kotler DP et al.Application of bioimpedance analysis for estimating body composition in prepubertal children infected with human immunodeficiency virus type 1.J Pediatr. 1996; 129: 755-757Abstract Full Text Full Text PDF PubMed Scopus (21) Google Scholar, 2Miller TL Nutritional assessment and its clinical application in children infected with HIV.J Pediatr. 1996; 129: 633-636Abstract Full Text Full Text PDF PubMed Scopus (51) Google Scholar In assessing whether HIV-specific BIA equations are necessary, it would also be useful if greater consideration of the effect of clinical status, hydration status, and nutritional status of the children studied were presented. It is possible that errors in BIA prediction equations might vary across the range of nutritional status (e.g., from wasted to obese 3Reilly JJ Wilson J McColl JH Carmichael M Durnin JVGA Ability of bioelectrical impedance to predict fat-free mass in prepubertal children.Pediatr Res. 1996; 39: 176-179Crossref PubMed Scopus (41) Google Scholar). Information on the presence of clinical features of particular relevance (e.g., chronic diarrhea) would also be useful. Errors can also arise from differences in the reference method of body composition used between studies (e.g., dual-energy x-ray absorptiometry vs hydrodensitometry). Use of dual-energy x-ray absorptiometry as a “gold standard” method is premature. 4Jebb SA Measurement of soft tissue composition by dual energy x-ray absorptiometry.Br J Nutr. 1997; 77: 151-163Crossref PubMed Scopus (81) Google Scholar It should also be pointed out that the reference methods used to derive some of the BIA prediction equations tested by Arpadi et al. 1Arpadi SM Wang J Cuff PA Thornton J Horlick M Kotler DP et al.Application of bioimpedance analysis for estimating body composition in prepubertal children infected with human immunodeficiency virus type 1.J Pediatr. 1996; 129: 755-757Abstract Full Text Full Text PDF PubMed Scopus (21) Google Scholar were inappropriate in that adult values were used for hydration of fat-free mass. The prediction equations derived were therefore inappropriate for pediatric use, and the resulting systematic errors were inevitable.