Population Reference Values by Age, Sex, and Body Mass Index

Phase Angle From Bioelectrical Impedance Analysis: Population Reference Values by Age, Sex, and Body Mass Index

Bosy-Westphal, Anja

ABSTRACT. Background: The use of bioelectrical impedance phase angle has been recommended as a prognostic tool in the clinical setting, but published reference data bases are discrepant and incomplete (eg, they do not consider body mass index [BMI], and data are lacking for children). Methods: Phase angle reference values stratified by age, sex, and BMI were generated in a large German data base of 15,605 children and adolescents and 214,732 adults, and the determinants of phase angle values were assessed. The reference values were applied to 3 groups of patients and compared with previously published reference values from the United States and Switzerland. Results: Gender and age were the main determinants of phase angle in adults, with men and younger subjects having higher phase angles. In children and adolescents, age and BMI were the main determinants of phase angle. In normal and overweight adults, phase angle increased with increasing BMI, but there was an inverse association at a BMI >40 kg/m^sup 2^. In cirrhosis, the prevalence of a low phase angle increased with the state of disease, whereas it was not different between patients with the metabolic syndrome and controls. There are considerable differences between phase angle reference values from different populations. These differences are not explained by age or BMI and may be due to differences between impedance analyzers. Conclusion: The determinants of phase angle differ between adults and children. In adults, the influence of BMI on phase angle depended on the BMI range. The prognostic value of phase angle may differ in different clinical settings. The use of population-specific and probably impedance-analyzer-specific reference values for phase angle is recommended. (Journal of Parenteral and Enterai Nutrition 30:309316, 2006)

The use of raw data from bioelectrical impedance analysis has gained popularity in nutrition assessment and monitoring of nutrition status in patients. The advantage of the use of pure electrical properties of tissues without equations or models for body composition analysis means that the main assumption of a constant tissue hydration, which is an unlikely condition in many clinical situations, is not required. Results are also not biased by the choice of regression equation, the accuracy of the criterion method, or the selection criteria of the reference population. Today, many publications focus on the use of resistance (R) and reactance (Xc) standardized by body height in a bivariate vector analysis (BIVA)1-9 and reference values from different populations stratified according to ethnic, age, and body mass index (BMI) groups are available.6,7,10

Clinically, the most established impedance parameter is, however, the phase angle for the diagnosis of malnutrition and clinical prognosis, both associated with changes in cellular membrane integrity and alterations in fluid balance.11 Phase angle expresses both changes in the amount as well as the quality of soft tissue mass (ie, cell membrane permeability and soft tissue hydration). It can be directly calculated from R and Xc as arc-tangent(Xc/R) × 180°/π. Therefore, the phase angle is on the one hand dependent on the capacitive behavior of tissues (Xc) associated with cellularity, cell size, and integrity of the cell membrane, and on the other hand on its pure resistive behavior (R), mainly dependent on tissue hydration.

There is now a large body of clinical trials that propose phase angle as a useful prognostic marker in clinical conditions like liver cirrhosis,12 dialysis,13-16 several types of cancer,17-19 HIV infection and AIDS,20,21 bacteremia and sepsis,22,23 and pulmonary disease.24

Phase angle reference values are mandatory for the assessment of individual deviations of a patient in relation to population average. However, to our knowledge these data are lacking in children and adolescents. In adults, sex- and age-specific phase angle reference values have been published by Dittmar25 in 2003 and Kyle et al26,27 in 2001 and 2004 and only recently by Barbosa-Silva et al.11 However, there are considerable differences between those reference values. In the Swiss population,”6 phase angles were lower (10.5% in men and 7.7% in women) than in the American study sample,11 and the lowest values were observed in the German population.25 The reason for these discrepancies remains unclear. Because BMI was shown to have an independent effect on impedance measures R, Xc, and consequently on phase angle,10,11,25 the differences might be due to differences in BMI between the reference populations. However, higher phase angles in the American compared with the Swiss population remained even after adjusting for BMI and percentage of fat mass.11 Thus, it was suggested that there may be real differences in phase angle between populations, and population-specific reference values may be required.

The present study was conducted to establish sex-, age-, and BMI-specific reference values for phase angle in a large healthy German population ( 15,605 children and adolescents, and 214,732 adults) with a wide range of age (6-102 years) and BMI (13.2-60 kg/m^sup 2^ in adults). The results are compared with previously published American and Swiss reference databases. The new reference values were finally applied to 3 independent databases to compare the prevalence of a low phase angle in patients with liver cirrhosis, anorexia nervosa, and the metabolic syndrome, respectively.

SUBJECTS AND METHODS

Data from a previously published dataset were used.11 Adult subjects were recruited from commercial weight-management facilities (Precon-centers) in Germany. Data from children were taken from the Kiel Obesity Prevention Study (KOPS).28 Informed consent was obtained from all volunteers before participation. In the case of children, parents provided informed written consent. The study was approved by the ethical committee of the Christian-Albrechts-University of Kiel. Data from 10,127 girls (11.5 ± 3.9 years, 6-17 years), 6110 boys (9.5 ± 3.2 years, 6-17 years), 183,982 women (42.5 ± 13.2 years, 18-102 years), and 30,750 men (44.6 ± 13.5 years, 18-100 years) were collected over a period of 14 years, from June 1990 to August 2003. Data from 420 girls and 212 boys were omitted due to BMI exceeding the highest age-dependent BMI category. Thus, the final study population consisted of 15,605 children and adolescents (age range, 6-17 years) and 214,732 adults (age range, 18-102 years) who were examined by a total of 530 trained observers. Training of the investigators followed the same manual. All subjects were Caucasians, nonpregnant and nonlactating, and healthy (defined as absence of a clinical condition that potentially influences fluid balance, ie, renal, endocrine, or myocardial disease, ascertained by participant questionnaire). Measurements were performed between 7 and 12 AM. In the case of adults, all subjects were measured after an overnight fast. Fasting was no precondition for study participation in children. Body weight and standing height were measured in underwear and without shoes to the nearest 0.1 kg and 0.5 cm, using an anthropometer and a calibrated electronic scale, respectively. BMI was calculated as the ratio between weight and height squared (kg/m2). A single tetrapolar BIA measurement of resistance (R) and reactance (Xc) was taken at a fixed frequency of 50 kHz at the right side of the subject between the right wrist and ankle while in a supine position on a nonconductive surface with a body impedance analyzer (BIA 2000-S, Data Input, Frankfurt, Germany), which applies an 800-µA current. Adhesive gel electrodes (Bianostic MG, Data Input) were placed at denned anatomical sites on the dorsal surfaces of the hand, wrist, ankle, and foot according to manufacturer’s instructions as follows: the proximal edge of the first electrode was attached from an imaginary line at styloid process of the ulna and the distal edge of the finger electrode on an imaginary line from the middle of the metacarpophalangeal joints of the index and middle fingers. The distal edge of the toe electrode was placed from an imaginary line through the middle of the metatarsophalangeal joints of the second and third toes. The proximal edge of the ankle joint electrode was attached along a line through the highest points of the outer and inner ankle bones. see Kushner29 for a detailed description of the measurement procedure. Reliability for within-day and between-day measurements by the same observer were

Phase angle (degrees) = arctan(Xc/R) × (180/π)

All data are given as means ± SD. The statistical analyses were performed by using SPSS for Windows 13.0 (SPSS Inc, Chicago, IL). Pearson’s correlation coefficients were calculated for relationships between phase angle and other variables. ANOVA was used to analyze differences between sexes. Stepwise multivariate regression analysis was performed to analyze the variance in phase angles. Potential predictors of phase angle considered were age, sex, and BMI. All tests were 2-tailed, and a p value

The application of the newly developed reference values for phase angle to different groups of patients was performed by using previously published datasets of our group.30-32 Impedance analyzers used were BIA 2000-M Data Input,32 body composition analyzer model TVI-IO; Danninger Medical Technology Inc, Detroit, MI,31 and BIA 101 RJL Systems, Detroit, MI.30 All analyzers operated at 800 µA and 50 kHz. Impedance measurements in each study followed the same standardized protocol as mentioned above.

RESULTS

Basal characteristics for adults and separately for children and adolescents are given in Table I. When compared with women, men were significantly older and had a higher weight, height, and BMI. By contrast, in children and adolescents, girls were significantly older than boys and also had higher values for weight, height, and BMI, respectively. Although the differences in age and BMI between sexes were low (2.5 years and 1.3 kg/m^sup 2^ in adults and 2.1 years and 2.7 kg/m^sup 2^ in children and adolescents) they were statistically significant due to the high number of subjects in both generations.

Study groups were divided into sex, BMI, and age classes as follows: 15,605 children and adolescents weight 16- 25-30 kg/m^sup 2^, and obese >300 -35, >35-40, >400-50, >50 kg/m^sup 2^) were stratified into 7 age groups (18-19, 20-29, 30-39, 40-49, 50-59, 60-69, and >70 years). Due to the small numbers in age categories, the underweight group (454 women and 56 men, BMI 50 kg/m^sup 2^) were excluded from sex- and age-specific analysis of phase angle reference values.

In Table II means (± SD) and the respective 10th (10.P) and 5th percentiles (5.P) for phase angle are shown by sex, BMI, and age categories for adults. Phase angle was significantly greater in men than in women in all categories except for subjects >70 years and a BMI > 18.5-25 or >35 kg/m^sup 2^. In each BMI group, phase angle decreased with increasing age. Within each sex and age group, phase angle tended to increase with increasing BMI up to a value of 35 kg/m^sup 2^. By contrast, a decrease in phase angle was observed in higher BMI groups.

Reference values for phase angle in children and adolescents (means ± SD and the 10th and fifth percentiles) are presented in Table III. In contrast to adults, there were no sex differences in phase angle for the younger age groups ( 19 kg/m^sup 2^, mean phase angle was higher in males when compared with females. There was an increase in phase angle with age in both sexes and with increasing BMI in each age category. In children and adolescents, phase angle showed a positive correlation with BMI (r = 0.31, p 40 kg/m^sup 2^, the correlations between BMI and phase angle were inverse for both sexes.

Multivariate regression analysis taking into account age, gender, and BMI and interactions between these parameters explained 12% and 15% of the variance in phase angle in children and adolescents and in adults, respectively. In children and adolescents, the mean predictors of phase angle were (i) age accounting for 10.8% of the variance and (ii) BMI explaining additional 1%. Although sex and the interaction term sex × age remained additional predictors of phase angle in children and adolescents, their influence was low: altogether these variables explained only 0.5% of the interindividual variance in phase angle. In contrast to children and adolescents, the main predictors of phase angle in adults were sex and age, each explaining 7% of its variance. BMI and the interaction of BMI or sex with age had additional minor contributions to phase angle in adults, together explaining 1.5% of its variance.

Figure 1 shows the dependency of mean phase angle on gender, age, and BMI in adults. For comparative reasons, the population averages for each age group are also shown for the American database by Barbosa-Silva et al11 and the Swiss population by Kyle et al.26 When compared with the American reference, the German values for phase angle of all BMI classes were lower in each gender and age category. This was also true for the Swiss reference values, except for the age group >70 years, where mean phase angle was similar to German values.

In Table V, the prevalence of phase angle below the respective age-, sex-, and BMI-specific reference values of Table II is given for 3 groups of patients. In patients with liver cirrhosis, with increasing severity of liver disease (indicated by Child-Pugh classification) there was a concomitant increase in the prevalence of low phase angles. Nearly half of the patients with Child C had phase angles below the fifth percentile of the healthy population. There was also a high prevalence of low values for phase angle in patients with anorexia nervosa, reaching 39% below the fifth reference percentile (mean BMI on admission 15.2 ± 1.5 kg/m^sup 2^). Nineteen patients were reinvestigated during weight gain 43 days and 84 days after baseline. During weight gain from time point (T)0 to T2 ( +9.0 ± 3.3 kg), phase angles improved considerably, leaving 14% of low-phase-angle values at T2. However, this prevalence is still nearly 3-fold increased when compared with healthy females of the same age group. By contrast, the prevalence of low phase angles in patients with the metabolic syndrome was not increased when compared with healthy controls. However, in this database we found a significant negative correlation between systolic blood pressure and phase angle (r = -0.18; p

DISCUSSION

By using a huge database, we were able to investigate a large range of age and BMI values. Thus, it was possible to analyze the impact of BMI and childhood or adolescence on phase angle in addition to sex and aging. We provided reference values for phase angle stratified by age groups and considering the influence of gender and BMI. A low phase angle is an established parameter for assessment of malnutrition and prognosis; by contrast, the value of a high phase angle remains unknown. Thus, we confined our analysis on the estimation of population averages with SDs and respective 10th and fifth reference percentiles. The population average and SD may be used to standardize phase angle by transforming an individual value into a z score (by dividing mean sex-, age-, and BMI-specific phase angles by their SD). This will allow us to obtain continuous data of patients instead of dichotomy (eg, below or above the fifth reference percentile).

Determinants of Phase Angle in the Healthy Population

When compared with adults, the main predictors of phase angle were different in children and adolescents. In adults, mean phase angle was higher in men than in women and gender explained 7% of interindividual variance in this parameter. By contrast, there were no significant sex differences in phase angle between boys and girls until the age group of 14-17 years. In children and adolescents, age was the main predictor of phase angle and explained nearly 11% of its variance. Growing up was associated with increasing phase angles, a relationship that is likely due to an increase in cell mass with age. Age also contributed to interindividual variance in phase angle of adults, but in contrast to children and adolescents, the regression coefficient was negative, implying a decrease in phase angle with aging. Because an age-dependent decline in Xc values was also observed after adjusting for stature and body circumferences, it was suggested that in addition to quantitative changes (decline in tissue mass) the electric properties of tissues may be altered with age.3 Gender differences in phase angle and a negative correlation with age were also found in previous studies.3,11,25,26

In adults as well as in children and adolescents, BMI was associated with phase angle independently from age and sex, and explained 1% of its variance. In both generations, phase angle showed a significant positive correlation with BMI. However, this relationship was moderate only in children and adolescents (r = 0.31, p 300-40 kg/m^sup 2^, the relationship was lost, and at a very high BMI range, >40 kg/m^sup 2^, there was even an inverse relationship between phase angle and BMI. There might be 2 underlying mechanisms for this finding. First, a tendency to a decrease in phase angle with increasing BMI in the morbidly obese range may derive from an increase in tissue hydration (i) due to a physiologically higher extracellular to intracellular water ratio of adipose tissue'” or (ii) due to a pathophysiological fluid overload in severe obesity (eg, edema).34 Fluid overload in morbid obesity was also suggested by others who found a prevalence of a low phase angle of 19% in women with a BMI between 40 and 64 kg/m^sup 2^ and only 5% in the BMI group 30-35 kg/m^sup 2^.5 Second, a loss in functional status of cell membranes might have contributed to a lower phase angle in severe obesity. Such a loss in functional status of cell membranes with increasing BMI might be due to an increasing secretion of inflammatory cytokines of adipose tissue in severe obesity.35 Consistent with this hypothesis, in a population with a high prevalence of obesity and the metabolic syndrome, subjects with a low phase angle (

Because of a minor impact of the parameter, previous publications did not differentiate phase angle reference values for BMI.11,27 However, patients often exceed the normal BMI range, and a differentiation between phase angle as an indicator of body composition (cell mass and hydration) on the one hand and cellular function altered by a disease (cell membrane integrity) on the other may require BMI-specific reference values to exclude an influence of body composition.

Comparison to Previously Published Reference Values of Phase Angle

Mean sex- and age-specific reference values for the fifth percentile of phase angle in the American population studied by Barbosa-Silva et al11 were higher than in the present investigation. This difference was irrespective of differences in BMI between the populations (Figure 1). When compared with the American population, phase angle values in the Swiss population27 were 10.5% lower in men and 7.7% lower in women, respectively,11 and a difference between the populations remained after adjusting for differences in BMI. In the present study, mean phase angle values controlled for age group and BMI were 7.5% and 12.5% lower in women and 8.4% and 16.0% lower in men when compared with the Swiss and American populations, respectively. Because differences in age or BMI do not explain the discrepant findings, there may also be a certain dependency of impedance raw data on the type of analyzer. Impedance results have been shown to differ between impedance devices from several manufacturers.36-38 In the American population, an RJL instrument was used (model 101; RJL Systems, MI). However, several BIA instruments were used in the Swiss population, and no significant difference was found for R at 50 kHz between Xitron (Xitron Technologies, Inc, CA), Bio-Z (Bio-Z2, Spengler, France) and RJL 101 devices (RJL Systems, Inc) in 29 subjects.26

We did not compare the coefficient of variance between devices from different manufacturers used in the different studies (given in Table V). However, we compared Data Input BIA 2000S devices to Xitron 4000B analyzers (Xitron Technologies) in 165 healthy subjects (75 men and 90 women, mean age 41.4 ± 16.0 and BMI 26.1 ± 4.4 kg/m^sup 2^; unpublished data). The differences in R values at 50 kHz between the 2 devices were small (516.3 ± 84.7 Ohm for Data Input vs 524.2 ± 86.2 Ohm for Xitron, respectively; p

Application of Phase Angle Reference Values in Patients

Phase angle has been proven to be a prognostic tool in clinical practice (12-24; see above). However, different cutoff values were used in all studies. In general, the median or the lower quartile of phase angle are used or the cutoff is derived from the comparison with a healthy control group. Thus, it is currently not possible to compare different studies and also to compare the diagnostic value of phase angle in different groups of patients. This may be an interesting idea because the pathophysiology of disease may differ with respect to the effects on cell mass, cell membrane integrity, and cellular hydration. Thus, the prognostic value of phase angle may also differ between groups of patients with different clinical conditions. There may, for instance, be a close correlation between phase angle and the state of liver disease, whereas there is no difference in the prevalence of a low phase angle between patients with the metabolic syndrome and healthy controls (as shown in Table V). A limitation of our study is the use of different impedance analyzers in the analysis of different groups of patients. Thus, taking into account the possibility of differences between impedance devices, we may only compare the results within but not between the different studies shown in Table V. Using the same reference values throughout different studies facilitates an evaluation of the effect of different diseases on phase angle.

In summary, we have shown that the determinants of phase angle differ between adults and the group of children and adolescents. In adults, the influence of BMI on phase angle depended on the BMI range. In normal-weight and overweight subjects, phase angle increases with increasing BMI, whereas in the severely obese (BMI >40 kg/m^sup 2^), there was an inverse association between BMI and phase angle. There are considerable differences between phase angle reference values from different populations. These differences are not explained by age or BMI. It is therefore recommended to use population specific and probably impedance-analyzer-specific reference values for impedance phase angle.

ACKNOWLEDGMENTS

The work was supported by a grant from Precon, Darmstadt, Germany. R.-P. Dörhöfer, as an author responsible for data collection between 1990 and 2003, was an employee of Data Input Company, which in 2001 merged with Precon, Darmstadt.

REFERENCES

1. Savino F, Cresi F, Grasso G, Oggero R, Silvestro L. The Diagram vector: a graphical relation between reactance and phase angle measured by bioelectrical analysis in infants. Ann Nutr Metab. 2004;48:84-89.

2. Toso S, Piccoli A, Gusella M, et al. Bioimpedance vector pattern in cancer patients without disease versus locally advanced or disseminated disease. Nutrition. 2003;19:510-514.

3. Buffa R, Floris G, Marini E. Migration of the bioelectrical impedance vector in healthy elderly subjects. Nutrition. 2003;19:917-921.

4. Guida B, Belfiore A Angrisani L, et al. Laparoscopic gastric banding and body composition in morbid obesity. Nutr Metab Cardiovasc Dis. 2005;15:198-203.

5. Guida B, Trio R, Pecoraro P, et al. Impedance vector distribution by body mass index and conventional bioelectrical impedance analysis in obese women. Nutr Metab Cardiovasc Dis. 2003;13: 72-79.

6. Piccoli A, Pillon L, Dumler F. Impedance vector distribution by sex, race, body mass index, and age in the United States: standard reference intervals as bivariate Z scores. Nutrition. 2002; 18:153-167.

7. Piccoli A, Fanos V, Peruzzi L, et al. Reference values of the bioelectrical impedance vector in neonates in the first week after birth. Nutrition. 2002;18:383-387.

8. Piccoli A, Italian CAPD-BIA Study Group. Bioelectric impedance vector distribution in peritoneal dialysis patients with different hydration status. Kidney Int. 2004;65:1050-1063.

9. Piccoli A, Codognotto M, Pascoli L, Boffo G, Caregaro L. Body mass index and agreement between bioimpedance and anthropometry estimates of body compartments in anorexia nervosa. JPEN J Parenter Enterai Nutr. 2005;29:148-156.

10. Bosy-Westphal A, Danielzik S, Dörhöfer RP, Piccoli A, Müller MJ. Patterns of bioelectrical impedance vector distribution by body mass index and age: implications for body-composition analysis. Am J Clin Nutr. 2005;82:60-68. Erratum in: Am J Clin Nutr. 2005;82:1358.

11. Barbosa-Silva MCG, Barros AJD, Wang J, Heymsfield SB, Pierson RN Jr. Bioelectrical impedance analysis: population reference values for phase angle by age and sex. Am J Clin Nutr. 2005;82:49-52.

12. Selberg O, Selberg D. Norms and correlates of bioimpedance phase angle in healthy human subjects, hospitalized patients, and patients with liver cirrhosis. Eur J Appl Physiol. 2002;86: 509-516.

13. Nescolerade L, Piccoli A, Roman A, et al. Bioelectrical impedance vector analysis in haemodialysis patients: relation between oedema and mortality. Physiol Mean. 2004;25:1271-1280.

14. Maggiore Q, Nigrelli S, Ciccarelli C, Grimaldi C, Rossi GA, Michelassi C. Nutritional and prognostic correlates of bioimpedance indexes in hemodialysis patients. Kidney Int. 1996;50: 2103-2108.

15. Puim LB, Kent P, Ikizler TA. Bioelectrical impedance analysis in dialysis patients. Miner Electrolyte Metab. 1999;25:400-406.

16. Mushnick R, Fein PA, Mittman N, Goel N, Chattopadhyay J, Avram MM. Relationship of bioelectrical impedance parameters to nutrition and survival in peritoneal dialysis patients. Kidney Int Suppl. 2003;87:S53-S56.

17. Gupta D, Lammersfeld CA, Burrows JL, et al. Bioelectrical impedance phase angle in clinical practice: implications for prognosis in advanced colorectal cancer. Am J Clin Nutr. 2004;80: 1634-1638.

18. Gupta D, Lis CG, Dahlk SL, Vashi PG, Grutsch JF, Lammersfeld CA. Bioelectrical impedance phase angle as a prognostic indicator in advanced pancreatic cancer. Br J Nutr. 2004;92:957-962.

19. Toso S, Piccoli A, Gusella M, et al. Altered tissue electric properties in lung cancer patients as detected by bioelectric impedance vector analysis. Nutrition. 2000;16:120-124.

20. Schwenk A, Beisenherz A, Römer K, Kremer G, Salzberer B, Elia M. Phase angle from bioelectrical impedance analysis remains an independent predictive marker in HIV infected patients in the era of highly active antiretroviral treatment. Am J Clin Nutr. 2000;72:496-501.

21. Ott M, Fischer H, Polat H, et al. Bioelectrical impedance analysis as a predictor of survival in patients with human immunodeficiency virus infection. J Aquir Immune Defic Syndr Hum Retrovirol. 1995;9:20-25.

22. Schwenk A, Ward LC, Elia M, Scott GM. Bioelectrical impedance analysis predicts outcome in patients with suspected bacteremia. Infection. 1998;26:277-282.

23. GIBI Brazilian Group for Bioimpedance Study. Total body bioelectrical impedance measurement as a progressive outcome prediction and therapeutic index in the comparison between septic and non septic patients: a multicenter Brazilian study. R Metab Nutr. 1995;2:159-170.

24. Faisy C, Rabbat A, Kouchakji B, Laaban JP. Bioelectrical impedance analysis in estimating nutritional status and outcome of patients with chronic obstructive pulmonary disease and acute respiratory failure. Intensive Care Med. 2000;26:518-525.

25. Dittmar M. Reliability and variability of bioimpedance measures in normal adults: effects of age, gender and body mass index. Am J Physiol Anthropol. 2003;122:361-370.

26. Kyle UG, Genton L, Slosman DO, Pichard C. Fat-free and fat mass percentiles in 5225 healthy subjects aged 15 to 98 years. Nutrition. 2001;17:534-541.

27. Kyle UG, Genon L, Karsegard VL, Raguso CA, Dupertuis YM, Pichard C. Percentiles (10, 25, 75 and 90th) for phase angle (PhA), determined by bioelectrical impedance analysis (BIA) in 2740 healthy adults aged 20-75 yr [abstract]. Clin Nutr. 2004; 23:758.

28. Danielzik S, Pust S, Landsberg B, Müller MJ. First lessons from the Kiel Obesity Prevention Study (KOPS). hit J Obes Relat Metab Disord. 2005;29(suppl):S78-S83.

29. Kushner RF. Bioelectrical impedance analysis: a review of principals and applications. Am Coll Nutr. 1992;11:199-209.

30. Müller MJ, Böttcher J, Selberg O, et al. Hypermetabolism in clinically stable patients with liver cirrhosis. Am J Clin Nutr. 1999;69:1194-1201.

31. Haas V, Onur S, Paul T, et al. Leptin and body weight regulation in patients with anorexia nervosa before and during weight recovery. Am J Clin Nutr. 2005;81:889-896.

32. Bosy-Westphal A, Geisler C, Onur S, Korth O, Schrezenmeir J, Millier MJ. Value of body fat mass vs anthropometric obesity indices in the assessment of metabolic risk factors. Int J Obes. 2006;30:475-483.

33. Waki M, Kral JG, Mazariegos M, Wang J, Pierson RN Jr, Heymsfleld SB. Relative expansion of extracellular fluid in obese vs. nonobese women. Am J Phys. 1991;261:E199-E203.

34. Adelman RD. Obesity and renal disease. Curr Opin Nephrol Hypertens. 2002;11:331-335.

35. Trayhurn P, Wood IS. Signalling role of adipose tissue: adipokines and inflammation in obesity. Biochem Soc Traim. 2005;33: 1078-1081.

36. Woodrow G, Oldroyd B, Turney JH, Davies PSW, Day JME, Smith MA. Measurement of total body water by bioelectrical impedance in chronic renal failure. Eur J Clin Nutr. 1996;50: 676-681.

37. Smye SW, Sutcliffe J, Pitt E. A comparison of four commercial systems used to measure whole-body electrical impedance. Physiol Meas. 1994;14:473-478.

38. Warner JT, Evans WD, Webb DKH, Gregory JW. Pitfalls in the assessment of body composition in survivors of acute lymphoblastic leukemia. Arch Dis Child. 2004;89:64-68.

39. Sun SS, Chumlea WC, Heymsfield SB, et al. Development of bioelectrical impedance prediction equations for body composition with the use of a multicomponent model for use in epidemiologic surveys. Am J Clin Nutr. 2003;77:331-340.

Anja Bosy-Westphal, PhD*; Sandra Danielzik, PhD*; Ralf-Peter Dörhöfer, MSc[dagger]; Wiebke Later, MSc*; Sonja Wiese, BSc*; and Manfred J. Millier, MD*

From the *Institut für Humanernährung und Lebensmittelkunde, Christian-Albrechts-University Kiel, Kiel, Germany; and the [dagger]Data Input Company, Darmstadt, Germany

Received for publication January 19, 2006.

Accepted for publication April 4, 2006.

Correspondence: Prof. Dr. med. M. J. Millier, Institut für Humanernährung und Lebensmittelkunde, Christian-Albrechts-Universität zu Kiel, Düsternbroker Weg 17, D-24105 Kiel, Germany. Electronic mail may be sent to mmueller@nutrfoodsc.uni-kiel.de.

Copyright American Society for Parenteral and Enteral Nutrition Jul/Aug 2006

Provided by ProQuest Information and Learning Company. All rights Reserved