November 27, 2004

Improvement of Nutritional Status As Assessed By Multifrequency BIA During 15 Weeks of Refeeding in Adolescent Girls With Anorexia Nervosa1


In patients with anorexia nervosa (AN), an assessment of changes in body composition and nutritional status is crucial for adequate nutritional management during refeeding therapies. Phase-sensitive multifrequency bioelectrical impedance analysis (BIA) is an inexpensive and noninvasive technique with which to determine nutritional status and body composition. We investigated 21 female adolescents with AN (initial BMI 15.5 1.1 kg/m^sup 2^) 4 times between wk 3 and 15 of inpatient refeeding and 19 normal-weight, age- matched female controls. From wk 3 to 15, BMI, fat mass, body cell mass (BCM), total body water (TBW), intracellular water (ICW) but not extracellular mass (ECM), and extracellular water (ECW) increased significantly. Reactance (Xc), phase angle (PhA), and the ECM/BCM index as parameters of nutritional status improved significantly in patients and no longer differed from controls in wk 15, although the BMI of patients was significantly lower than those of controls. Changes in the ECM/BCM index were due to accretion of BCM, which was associated with an increase of ICW. Multifrequency phase-sensitive BIA seems to be a promising tool for the assessment of changes in nutritional status and body composition in patients with AN. An individually determined and controlled hyperenergetic diet as part of a multidimensional, interdisciplinary treatment program for eating disorders seems to quickly improve the nutritional status of AN patients. J. Nutr. 134: 3026-3030, 2004.

KEY WORDS: * anorexia nervosa * eating disorders * body composition * refeeding * nutritional status

Anorexia nervosa (AN)3 is estimated to be the third most common chronic medical disease in girls aged 15-19 y, affecting 0.5-1% of adolescent girls in Western countries (1). It is characterized primarily by severe malnutrition as well as psychological disturbances. Malnutrition leads to both a loss of fat mass (FM) and lean body mass (LBM). Many patients with AN require hospitalization and refeeding to restore normal body weight.

From both a physiological and a psychological point of view, an assessment of the changes in body composition and in nutritional status is crucial for adequate nutritional management during refeeding. However, anthropometry-based equations or measurements of skin-fold thickness are often unreliable in determining body composition and nutritional status, particularly in children and adolescents (2-4).

Bioelectrical impedance analysis (BIA) is a quick, inexpensive, and noninvasive technique that enables determination of body composition based on the measurement of electrical characteristics of the human body. Most studies of AN used phase-sensitive single- frequency BIA measurement, which allows the impedance Z to be differentiated into its 2 components, resistance (R, water resistance) and reactance (Xc, cell resistance). With this technique, a simple 3-compartment model of body composition can be assessed by prediction equations. The 3-compartment model includes body cell mass (BCM), extracellular mass (ECM), and FM. The BCM is the sum of the oxygen-consuming, potassium-rich, glucose-oxidizing cells, whereas constituents of the ECM are the connective tissue structures and fluids consisting of plasma, interstitial water, and transcellular water (5). Therefore the ECM/BCM index is used to describe nutritional status (6). In healthy individuals, the BCM is always distinctly larger than the ECM, so that the index is

In this context, multifrequency BIA has important advantages. By using an additional determination of resistance at low frequency (e.g., 5 kHz), it is possible to achieve a subdivision of the total body water (TBW) into intracellular (ICW) and extracellular water (ECW) (12). Deviations in BCM and ECM can thus be estimated differentially with regard to anabolic/catabolic processes or shifts in water. Therefore multifrequency BIA is particularly indicated in patients with disturbances in water balance, e.g., in heart and kidney failure. To our knowledge, multifrequency BIA has not been used in AN subjects to date.

Another BIA parameter used for the assessment of nutritional status is the phase angle (PhA) (13-15), which can be assessed by phase-sensitive single- or multifrequency BIA. PhA is the difference between the voltage and the current and can be obtained from the measured values for resistance (R) and reactance (Xc) without using equation predictions (16). Therefore, it is not susceptible to errors arising from the use of imprecise prediction equations. PhA is interpreted as an indicator of water distribution between the extra- and intracellular spaces and therefore as an indicator of membrane integrity of the cells (17,18).

The aim of this study was to assess nutritional status and body composition during an inpatient dietary treatment in adolescent AN patients by using phase-sensitive multifrequency BIA, in comparison with a healthy control group matched for age and sex.


Subjects. A consecutive sample of 21 female AN patients (age: 14.4 1.5 y, height: 1.65 0.08 m) admitted to inpatient refeeding at the Department of Child and Adolescent Psychiatry and Psychotherapy of the University of Aachen (Germany) and 19 healthy, normal-weight controls matched for age (15.1 2.2 y, P = 0.214), height (1.64 0.06 m, P = 0.665), and education were investigated. All patients met the criteria for anorexia nervosa according to the Diagnostic and Statistical Manual of the American Psychiatric Association (19). The Structured Interview of Anorexia and Bulimia nervosa (20) was used to establish the diagnosis of patients (including AN subtype) and to exclude eating disorders in controls. Four patients were classified as binge eating/purging type, and 17 were classified as restricting subjects. One AN patient had to be excluded from the study because of a lack of weight gain. At baseline, patients weighed 42.3 3.9 kg and had a BMI of 15.5 1.1 kg/m^sup 2^ (all below the 1st age-percentile). Controls, who were recruited from a local comprehensive secondary school after attending a lecture on anorexia nervosa, weighed 56.3 7.7 kg and had a BMI of 20.8 2.0 kg/m^sup 2^. One control subject had an initial BMI of 17.0 kg/m^sup 2^, corresponding to the 13th BMI age- percentile. In children and adolescents, a BMI at or below the 10th BMI age-percentile was defined as the weight cutoff for diagnosis of the Guidelines of the German Society of Child and Adolescent Psychiatry and Psychotherapy (21). At initial assessment, 4 AN patients had primary amenorrhoea and 17 AN patients had secondary amenorrhoea (mean duration: 9.5 5.7 mo). All controls had a normal menstrual cycle. Two AN patients and 3 control subjects were taking oral contraceptives. No patients or controls suffered from any gastrointestinal, systemic, or metabolic disease or received other medication known to affect body composition. The study was approved by the Ethics Committee of the University of Aachen. Patients, controls, and their parents gave written informed consent for participation.

Methods. BIA measurements (R and Xc at 50 kHz) were performed during wk 3, 7, 11, and 15 of inpatient treatment. BIA measurements started in wk 3 of inpatient treatment for reasons of standardization and to avoid measurement errors due to extreme dehydration or overhydration as well as due to acrocyanosis and low body temperatute. By wk 3 of inpatient treatment, all patients usually have a comparable fluid and energy intake that allows steady weight gain. Moreover, acrocyanosis and low body tempetature that may have been present at admission would have been resolved by wk 3. In controls, only 3 measurements were performed (wk 3,7, and 15). BlA was carried out between 0900 and 1030 h. To avoid acute fluid shifts, patients and controls were instructed to refrain from strenuous exercise and alcohol consumption for 12 h before the measurement. Room temperature was kept between 20 and 24C to prevent unclesired effects on cutaneous blood flow or compartmental changes in water (22). Height was measured with a wall stadiometer to the nearest cm (with the subjects standing straight and without shoes), and body weight was measured to the nearest 0.1 kg on an electronic scale. BIA of the whole body was performed with the subjects supine and their arms and legs abducted. The 4 electrodes were placed on the proximal and dorsal surface of the right hand and of the \right foot. We used the BIA 2000 device (Data Input GmbH) whose reliability was recently evaluated by Dittmar (12). Evaluation of the data obtained from the measurements was calculated using the manufacturer's software (Nutri4, Data Input GmbH; see Table 1). To subdivide TBW into ICW and ECW, an additional determination of resistance at low frequency (R^sub 5^;5 kHz) was performed.

During refeeding, AN patients consumed a diet that was precisely adjusted to their individual requirements, taking into account the patient's energy intake before admission, the somatic state, and the energy required for weight gain. At the beginning of refeeding, energy intake was 7.8 0.6 MJ/d and increased continuously to 9.3 0.7 MJ/d in wk 15. The intake of dietary protein, fat, and carbohydrate was calculated according to the recommendations of the Food and Nutrition Board of the National Academies (23): 15-20% of the daily energy intake consisted of protein, 30% was fat, and 50- 55% was carbohydrate. All but one of the patients were tube fed during the first 2-4 wk of inpatient treatment followed by stepwise oral food replacement.

All patients were participating in an inpatient-specific eating disorder program for adolescents including weight and nutritional management, dietician therapy, psychoeducation, group therapy, individual cognitive-behavioral psychotherapy, and family counseling. Additionally, at least one of the patients' parents took part in a psychoeducative group for parents of adolescents with AN. During inpatient refeeding, a mean weight gain of 500-1000 g/wk up to achievement of target weight (25th BMI percentile) was intended using reinforcement strategies and sometimes nasogastral tube- feeding in case of insufficient weight gain. This treatment program is in line with the Guidelines of the German Society of Child and Adolescent Psychiatry and Psychotherapy (21).

Statistical methods. Statistical analyses were performed using SPSS 11.0 software. All variables were normally distributed (analysis of skewness and kurtosis). Repeated-measure ANOVA (group vs. time, with time as repeated measure) including data of week 3, 7, and 15 was performed for BMI and BIA parameters. Post-hoc tests were performed when there was a significant main effect of group or a significant group time interaction. Independent-sample t tests (2- sided) were used to compare groups. Paired-samples t tests (2- sided) were used to compare values of AN subjects at wk 3 and 15. Data are presented as means SD. Differences were considered significant at P


Equations for the prediction of body composition


BIA parameters of AN patients during 15 wk of dietary treatment compared with healthy controls1


All subjects of the AN group and controls completed the study. The ECM/BCM index decreased significantly during treatment in the AN group (P


Absolute and relative water distribution of AN patients during 15 wk of dietary treatment compared with healthy controls1

Compared with controls, resistance (R) was higher in the AN group at the beginning of the study (P

At the beginning, LBM was significantly lower in the AN group than in controls (P

FIGURE 1 Change in the phase angle (PhA) vs. reactance (Xc) at 50 kHz in 21 AN patients before (wk 3, upper panel) and at the end of refeeding (wk 15, lower panel) compared with 19 normal-weight controls.


To our knowledge, this is the first study using multifrequency BIA to investigate the development of nutritional status and body composition in adolescent anorectics in the course of inpatient weight rehabilitation. Our BIA results indicate that a 15-wk dietetic treatment greatly improves the nutritional status of adolescent AN patients despite continuing low LBM, FM, and BMI.

Graphical evaluation (Fig. 1) of the raw data for reactance (Xc) and phase angle (PhA) as proposed by Dietmar (12) demonstrates an increase in both parameters in the patient group from wk 3 to 15, leading to a rightward shift of the graphical plot (Fig. 1). This can be interpreted as an improvement in nutritional status (12,24). Xc describes the capacitive impedance of cell membranes, which is correlated with the amount of BCM. Our data indicate that low values of Xc in AN patients before refeeding reflect a decrease in BCM and an increase of this parameter during refeeding. Because PhA reflects the ratio of intracellular to extracellular water (12,24), low PhA in AN subjects in wk 3 indicates that semistarvation leads to a loss of ICW that is normalized during nutritional rehabilitation. Previously, it was demonstrated that PhA is. a good indicator of cellular health and proper nutrition (12,10).

Our results are consistent with similar findings recently reported by Scalfi and co-workers (10). They found that the PhA of 13 adult AN patients was normalized after 16 mo of psychiatric treatment and a counseling dietary program, compared with healthy controls. Nevertheless, our results of a normalization of PhA within only 15 wk demonstrate the short-term effectiveness of an individually determined and controlled hyperenergetic dietary treatment that was part of a multidimensional, interdisciplinary inpatient treatment program for AN in a specialized eating disorder unit.

Another parameter of nutritional status derived from BIA measurements is the ECM/BCM index (6). We observed a decline of the ECM/BCM index during 1.5 wk of refeeding. In wk 15, this index no longer differed between groups, indicating a normalization of nutritional status in the patients. The decrease of the ECM/BCM index was due to an increase of BCM, whereas ECM including ECW remained unchanged. Thus, the changes in the ECM/BCM index in our patients indicate a real improvement in anabolic processes because of an accretion of metabolic active body cell mass.

The clinical usefulness of single-frequency BIA in AN was criticized because of the inability to assess accurately the distribution of water between intracellular and extracellular compartments (11). To our knowledge, we are first to use multifrequency BIA to differentiate between ECW and ICW in AN patients during refeeding. We found that the baseline measurement of TBW as a percentage of body weight was nearly 10% higher and absolute ICW was significantly decreased in AN patients compared with controls, whereas absolute ECW did not differ between groups. During refeeding, we observed an increase in absolute TBW, ICW, and BCM but no change in absolute ECW and ECM. Thusj the rise in TBW was due to an increase in ICW.and was accompanied by an enhanced BCM. Therefore, in contrast to: single-frequency BIA, multifrequency BIA may be an appropriate tool for monitoring nutritional status during refeeding in AN.

From wk 3 to 15, the body FM of patients increased by 70.7% and the percentage of body fat rose from 13.3 to 20.6%. Our results are consistent with those of previous studies demonstrating a higher increase of FM compared with the rise in LBM in both adolescent (9) and adult anorexic patients (25-29) during weight rehabilitation. In our patients, most of the fat gain occurred between wk 3 and 7, whereas only a minor gain in LBM was observed during this time. Thereafter, both LBM and FM increased in a parallel fashion until wk 15. The reason for the initial rapid increase in FM compared with LBM remains to be elucidated. Dulloo et al. (30) reviewed the data of the Minnesota Semistarvation Study. They found that the deposition of lean and fat tissue during weight recovery in healthy humans was determined mainly by individual variations in the control of energy-partitioning, for which the percentage of body fat before weight loss was the most important predictor of gain in FM during refeeding. The same may apply for AN patients. Nevertheless, it seems that there are also differences in anabolic processes during refeeding between weight-reduced healthy subjects and acutely ill AN patients. In the former, a reduction in thermogenesis seems to specifically acceler\ate the replenishment of fat stores (30), whereas in the latter, an increased diet-induced thermogenesis in the early phase of refeeding was recently described (31,32). Thus, the processes of energy-partitioning in the anorexic body must be investigated further.

Some limitations of our study should be considered. Multifrequency BIA measurement uses standardized formulas for the calculation of LBM with the assumption that the water content of LBM is 73%. In AN, the hydration status of LBM is said to be slightly higher (11). In this case, a calculation of LBM on the basis of 73% water content would lead to an overestimation of LBM and an underestimation of FM, which is calculated as the difference between body weight and LBM. A more appropriate method with which to measure body FM is dual-energy X-ray absorptiometry, which is not suitable for research in children and adolescents because of radiation exposure. Nevertheless, a validation of our BIA results against a reference method would have further enhanced the quality of our sttidy.

In sum, multifrequency phase-sensitive BIA seems to be a promising tool for the assessment of changes in nutritional status and of body composition in AN patients during refeeding. This is of particular clinical importance in the planning of adequate energy and nutrient intake to support restoration of weight and proportionate body composition. Furthermore, in acute AN, an individually determined and controlled hyperenergetic diet as part of a multidimensional, interdisciplinary treatment program for eating disorders seems to quickly improve the nutritional status as assessed by multifrequency BIA.

0022-3166/04 $8.00 2004 American Society for Nutritional Sciences.

Manuscript received 14 June 2004. Initial review completed 19 July 2004. Revision accepted 31 August 2004.

1 Supported by the Christina Barz-Stiftung (Essen, Germany), the Space Flight Program of the German Aerospace Center (DLR, Cologne, Germany) and the START Program of the Technical University of Aachen (Aachen, Germany).

3 Abbreviations used: BCM, body cell mass; BIA, bioelectrical impedance analysis; BW, body weight; ECM, extracellular mass; ECW, extracellular water; FM, fat mass; ICM, intracellular mass; ICW, intracellular water; LBM, lean body mass; PhA, phase angle; R, resistance; TBW, total body water; Xc, reactance.


1. Fairburn, C. G. & Harrison, P. J. (2003) Eating disorders. Lancet 361: 407-416.

2. Holt, T. L., Cui, C., Thomas, B. J., Ward, L. C., Quirk, P. C., Crawford, D. & Shepherd, R. W. (1994) Clinical applicability of bioelectric impedance to measure body composition in health and disease. Nutrition 10: 221-224.

3. Segal, K. R., Burastero, S., Chun, A., Coronel, P., Pierson, R. N., Jr. & Wang, J. (1991) Estimation of extracellular and total body water by multiple-frequency bioelectrical-impedance measurement. Am. J. Clin. Nutr. 54: 26-29.

4. Ellis, K. J. (2001) Selected body composition methods can be used in field studies. J. Nutr. 131: 1589S-1595S.

5. Moore, F. D., Olesen, K. H., McMurray, J. D., Parker, H. V., Ball, M. R. & Boyden, C. M. (1963) The body cell mass and its supporting environment. In: Body Composition in Health and Disease. W.B. Saunders, Philadelphia, PA.

6. Talluri, T., Lietdke, R. J., Evangelisti, A., Talluri, J. & Maggia, G. (1999) Fat-free mass qualitative assessment with bioelectric impedance analysis (BIA). Ann. N.Y. Acad. Sci. 873: 94- 98.

7. Mushnick, R., Fein, P. A., Mittman, N., Goel, N., Chattopadhyay, J. & Avram, M. M. (2003) Relationship of bioelectrical impedance parameters to nutrition and survival in peritoneal dialysis patients. Kidney Int. 64 (suppl.): S53-S56.

8. Bedogni, G., Marra, M., Bianchi, L., Malavolti, M., Nicolai, E., De Filippo, E. & Scalfi, L. (2003) Comparison of bioelectrical impedance analysis and dual-energy X-ray absorptiometry for the assessment of appendicular body composition in anorexic women. Eur. J. Clin. Nutr. 57: 1068-1072.

9. Schmidt, M. H. & Klein, M. (1997) [Determination of body feat in treatment of anorexia nervosa], Z. Kinder Jugendpsychiatr. Psychother. 25: 27-34.

10. Scalfi, L., Marra, M., Caldara, A., Silvestri, E. & Contaldo, F. (1999) Changes in bioimpedance analysis after stable refeeding of undernourished anorexic patients. Int. J. Obes. Relat. Metab. Disord. 23: 133-137.

11. Birmingham, C. L., Jones, P. J., Orphanidou, C., Bakan, R., Cleator, I. G., Goldner, E. M. & Phang, P. T. (1996) The reliability of bioelectrical impedance analysis for measuring changes in the body composition of patients with anorexia nervosa. Int. J. Eating Disord. 19: 311-315.

12. Dittmar, M. (2003) Reliability and variability of bioimpedance measures in normal adults: effects of age, gender, and body mass. Am. J. Phys. Anthropol. 122: 361-370.

13. Johansen, K. L., Kaysen, G. A., Young, B. S., Hung, A. M., da Silva, M. & Chertow, G. M. (2003) Longitudinal study of nutritional status, body composition, and physical function in hemodialysis patients. Am. J. Clin. Nutr. 77: 842-846.

14. Nagano, M., Suita, S. & Yamanouchi, T. (2000) The validity of bioelectrical impedance phase angle for nutritional assessment in children. J. Pediatr. Surg. 35: 1035-1039.

15. Pupim, L. B., Kent, P., Caglar, K., Shyr, Y., Hakim, R. M. & Ikizler, T. A. (2002) Improvement in nutritional parameters after initiation of chronic hemodialysis. Am. J. Kidney Dis. 40: 143-151.

16. Ellis, K. J. (2000) Human body composition: in vivo methods. Physiol. Rev. 80: 649-680.

17. Goovaerts, H. G., Faes, T. J., de Valk-de Roo, G. W., ten Bolscher, M., Netelenbosch, J. C., van der Vijgh, W. J. & Heethaar, R. M. (1998) Extracellular volume estimation by electrical impedance- phase measurement or curve fitting: a comparative study. Physiol. Meas. 19: 517-526.

18. Ackmann, J. J. & Seitz, M. A. (1984) Methods of complex impedance measurements in biologic tissue. Crit. Rev. Biomed. Eng. 11: 281-311.

19. American Psychiatric Association (1997) Diagnostic and Statistical Manual of Mental Disorders, DSM IV. American Psychiatric Association, Washington, DC.

20. Fichter, M. M., Herpertz, S., Quadflieg, N. & Herpertz- Dahlmann, B. (1998) Structured Interview for Anorexic and Bulimic disorders for DSM-IV and ICD-10: updated (third) revision. Int. J. Eating Disord. 24: 227-249.

21. German Scientific Society for Child and Adolescent Psychiatry and Psychotherapy (2003) Leitlinien zur Diagnostik und Therapie von psychischen Strungen im Suglings-, Kindes- und Jugendalter. Deutscher rzte Verlag, Kln, Germany.

22. Vienna, A. & Hauser, G. (1999) A qualitative approach to assessing body compartments using bioelectrical variables. Coll. Antropol. 23: 461-472.

23. Institute of Medicine (2002) Recommendations of the Food and Nutrition Board of the National Academies. file.asp?id=7300. [accessed August 17, 2004].

24. Foster, K. R. & Lukaski, H. C. (1996) Whole-body impedance- what does it measure? Am. J. Clin. Nutr. 64: 388S-396S.

25. Polito, A., Cuzzolaro, M., Raguzzini, A., Censi, L. & Ferro- Luzzi, A. (1998) Body composition changes in anorexia nervosa. Eur. J. Clin. Nutr. 52: 655-662.

26. Orphanidou, C. I., McCargar, L. J., Birmingham, C. L. & Belzberg, A. S. (1997) Changes in body composition and fat distribution after short-term weight gain in patients with anorexia nervosa. Am. J. Clin. Nutr. 65: 1034-1041.

27. Iketani, T., Kiriike, N., Nagata, T. & Yamagami, S. (1999) Altered body fat distribution after recovery of weight in patients with anorexia nervosa. Int. J. Eating Disord. 26: 275-282.

28. Scalfi, L., Polito, A., Bianchi, L., Marra, M., Caldara, A., Nicolai, E. & Contaldo, F. (2002) Body composition changes in patients with anorexia nervosa after complete weight recovery. Eur. J. Clin. Nutr. 56: 15-20.

29. Probst, M., Goris, M., Vandereycken, W. & Van Coppenolle, H. (2001) Body composition of anorexia nervosa patients assessed by underwater weighing and skinfold-thickness measurements before and after weight gain. Am. J. Clin. Nutr. 73: 190-197.

30. Dulloo, A. G., Jacquet, J. & Girardier, L. (1996) Autoregulation of body composition during weight recovery in human: the Minnesota Experiment revisited. Int. J. Obes. Relat. Metab. Disord. 20: 393-405.

31. Moukaddem, M., Boulier, ?., Apfelbaum, M. & Rigaud, D. (1997) Increase in diet-induced thermogenesis at the start of refeeding in severely malnourished anorexia nervosa patients. Am. J. Clin. Nutr. 66: 133-140.

32. Russell, J., Baur, L. A., Beumont, P. J., Byrnes, S., Gross, G., Touyz, S., Abraham, S. & Zipfel, S. (2001) Altered energy metabolism in anorexia nervosa. Psychoneuroendocrinology 26: 51-63.

33. Fischer, H. (1997) Bioelektrische Impedanz-Analyse (BIA): Grundlagen, Einsatz und Wertigkeit beim AIDS-Wasting Syndrom. In: Wasting und AIDS: die Behandlung der HIV-bedingten Kachexie (Jger, H., ed.), pp. 50-60. Thieme, Stuttgart, Germany.

34. Kushner, R. F. & Schoeller, D. A. (1986) Estimation of total body water by bioelectrical impedance analysis. Am. J. Clin. Nutr. 44: 417-424.

35. Leweling, H. (1995) Zusammensetzung des Krpers. In: Ernhrungsmedizin (Biesalski, H. K., ed.), pp. 3-12. Georg Thieme Verlag, Stuttgart, Germany.

Claudia Mika,*[dagger]2 Beate Herpertz-Dahlmann,* Martina Heer,** and Kristian Holtkamp*

* Department of Child and Adolescent Psychiatry and Psychotherapy, and [dagger] Institute of Aerospace Medicine, Technical University of Aachen, Aachen, Germany and ** DLR- Institute of Aerospace Medicine, Space Physiology, Cologne, Germany

2 To whom correspondence should be addressed. E-mail: [email protected]

Copyright American Institute of Nutrition Nov 2004