Novel approaches for the assessment of relative body weight and body fat in diagnosis and treatment of anorexia nervosa: A cross-sectional study

Open AccessPublished:January 10, 2019DOI:https://doi.org/10.1016/j.clnu.2018.12.031

      Summary

      Background & aims

      Anorexia nervosa (AN) is a severe psychosomatic disease that seriously affects nutritional status. Therapeutic approaches primarily aim for rapid weight restoration by high caloric diets and activity restriction. This often promotes abdominal body fat gain, which potentially negatively influences the patient's compliance and increases the risk of relapse. This study focused on the evaluation of body weight and subcutaneous adipose tissue (SAT) in AN patients by novel approaches.

      Methods

      The SAT of AN patients (n = 18, body mass index (BMI) 15.3 ± 1.3 kg/m2) was determined by a highly accurate and reliable ultrasound method. The sum of SAT thicknesses of eight sites (DINCL) was calculated. Individual metabolic profiles were analyzed. The mass index (MI), which considers body proportions, was used in addition to BMI. Additional to the standard laboratory diagnostics, dermal carotenoids measured by resonance Raman spectroscopy, leptin, and oxidative stress indicators were determined.

      Results

      The mean MI was 15.7 ± 1.4 kg/m2. The DINCL considerably differed between individuals with the same BMI. Half of the patients (Group 1) had low DINCL: 1.3–28.4 mm, and Group 2 showed values up to 58.2 mm (corresponding to approximately 6 kg SAT mass). The two group means differed by more than 300% (P < 0.001). Accordingly, leptin levels significantly differed (P < 0.001). Mean SAT thicknesses were significantly higher in Group 2 at all eight sites. The groups also significantly differed in two oxidative stress parameters: total antioxidative capacity, malondialdehyde-modified low density lipoprotein immunoglobulin M (MDA-LDL IgM), and in the carotenoid level.

      Conclusion

      Half of the patients had sufficiently high fat mass, despite very low BMI. Consequently, their muscle (and other organ) masses must have been extremely low. Diagnostic criteria and treatment protocols for AN should consider each patient's body composition. In addition to dietary treatments, muscle training at low energy turnover rates may be essential for avoiding unnecessary body fat gain, better treatment results, and long-term recovery.

      Keywords

      1. Introduction

      Anorexia nervosa (AN) is a severe psychiatric disease with a high mortality rate caused by severe malnutrition [
      • Smink F.R.
      • van Hoeken D.
      • Hoek H.W.
      Epidemiology of eating disorders: incidence, prevalence and mortality rates.
      ], a partly irreversible loss of body structure, and various other health consequences [
      • Yager J.
      • Andersen A.E.
      Clinical practice. Anorexia nervosa.
      ,
      • Achamrah N.
      • Coeffier M.
      • Dechelotte P.
      Physical activity in patients with anorexia nervosa.
      ]. The causes of AN are complex, which makes effective therapeutic approaches difficult [
      • Attia E.
      • Walsh B.T.
      Behavioral management for anorexia nervosa.
      ,
      • Wilson G.T.
      • Shafran R.
      Eating disorders guidelines from NICE.
      ,
      • Treasure J.
      • Claudino A.M.
      • Zucker N.
      Eating disorders.
      ,
      • Marzola E.
      • Nasser J.A.
      • Hashim S.A.
      • Shih P.A.
      • Kaye W.H.
      Nutritional rehabilitation in anorexia nervosa: review of the literature and implications for treatment.
      ,
      National Collaborating Centre for Mental Health (UK)
      Eating disorders: core interventions in the treatment and management of anorexia nervosa, bulimia nervosa and related eating disorders.
      ,
      American Psychiatric Association
      Practice guideline for the treatment of patients with eating disorders. 3rd ed.
      ]. Current treatment strategies focus primarily on rapid weight restoration [
      • Marzola E.
      • Nasser J.A.
      • Hashim S.A.
      • Shih P.A.
      • Kaye W.H.
      Nutritional rehabilitation in anorexia nervosa: review of the literature and implications for treatment.
      ,
      National Collaborating Centre for Mental Health (UK)
      Eating disorders: core interventions in the treatment and management of anorexia nervosa, bulimia nervosa and related eating disorders.
      ,
      American Psychiatric Association
      Practice guideline for the treatment of patients with eating disorders. 3rd ed.
      ], and are based on two key determinants that influence energy metabolism [
      • Achamrah N.
      • Coeffier M.
      • Dechelotte P.
      Physical activity in patients with anorexia nervosa.
      ]: increased energy intake taking into consideration any possible refeeding complications [
      • Marzola E.
      • Nasser J.A.
      • Hashim S.A.
      • Shih P.A.
      • Kaye W.H.
      Nutritional rehabilitation in anorexia nervosa: review of the literature and implications for treatment.
      ,
      • Agostino H.
      • Erdstein J.
      • Di Meglio G.
      Shifting paradigms: continuous nasogastric feeding with high caloric intakes in anorexia nervosa.
      ,
      • Garber A.K.
      • Mauldin K.
      • Michihata N.
      • Buckelew S.M.
      • Shafer M.A.
      • Moscicki A.B.
      Higher calorie diets increase rate of weight gain and shorten hospital stay in hospitalized adolescents with anorexia nervosa.
      ,
      • Mehanna H.M.
      • Moledina J.
      • Travis J.
      Refeeding syndrome: what it is, and how to prevent and treat it.
      ], including the administration of artificial nutrient-enriched dietary supplements to achieve the energy requirements for weight gain [
      • Yager J.
      • Andersen A.E.
      Clinical practice. Anorexia nervosa.
      ,
      • Marzola E.
      • Nasser J.A.
      • Hashim S.A.
      • Shih P.A.
      • Kaye W.H.
      Nutritional rehabilitation in anorexia nervosa: review of the literature and implications for treatment.
      ,
      American Psychiatric Association
      Practice guideline for the treatment of patients with eating disorders. 3rd ed.
      ]; restrict physical activity and exercise to reduce energy expenditure [
      • Yager J.
      • Andersen A.E.
      Clinical practice. Anorexia nervosa.
      ,
      • Marzola E.
      • Nasser J.A.
      • Hashim S.A.
      • Shih P.A.
      • Kaye W.H.
      Nutritional rehabilitation in anorexia nervosa: review of the literature and implications for treatment.
      ,
      National Collaborating Centre for Mental Health (UK)
      Eating disorders: core interventions in the treatment and management of anorexia nervosa, bulimia nervosa and related eating disorders.
      ,
      American Psychiatric Association
      Practice guideline for the treatment of patients with eating disorders. 3rd ed.
      ], and avoid over-exercise [
      • Cockfield A.
      • Philpot U.
      Feeding size 0: the challenges of anorexia nervosa. Managing anorexia from a dietitian's perspective.
      ], hyperactivity [
      • Rizk M.
      • Lalanne C.
      • Berthoz S.
      • Kern L.
      • Godart N.
      EVHAN Group
      Problematic exercise in anorexia nervosa: testing potential risk factors against different definitions.
      ], hypermetabolism [
      • Marzola E.
      • Nasser J.A.
      • Hashim S.A.
      • Shih P.A.
      • Kaye W.H.
      Nutritional rehabilitation in anorexia nervosa: review of the literature and implications for treatment.
      ], and compulsive and compensatory behavior that may lead to an increased risk of adverse physiological outcomes [
      • Cockfield A.
      • Philpot U.
      Feeding size 0: the challenges of anorexia nervosa. Managing anorexia from a dietitian's perspective.
      ,
      • Rizk M.
      • Lalanne C.
      • Berthoz S.
      • Kern L.
      • Godart N.
      EVHAN Group
      Problematic exercise in anorexia nervosa: testing potential risk factors against different definitions.
      ,
      • Gianini L.M.
      • Klein D.A.
      • Call C.
      • Walsh B.T.
      • Wang Y.
      • Wu P.
      • et al.
      Physical activity and post-treatment weight trajectory in anorexia nervosa.
      ,
      • Nicholls D.
      • Hudson L.
      • Mahomed F.
      Managing anorexia nervosa.
      ,
      • El Ghoch M.
      • Calugi S.
      • Pellegrini M.
      • Chignola E.
      • Dalle Grave R.
      Physical activity, body weight, and resumption of menses in anorexia nervosa.
      ]. However, some therapeutic approaches support supervised physical activity in the recovery process [
      • Achamrah N.
      • Coeffier M.
      • Dechelotte P.
      Physical activity in patients with anorexia nervosa.
      ,
      • Nicholls D.
      • Hudson L.
      • Mahomed F.
      Managing anorexia nervosa.
      ,
      • Sauchelli S.
      • Arcelus J.
      • Sanchez I.
      • Riesco N.
      • Jimenez-Murcia S.
      • Granero R.
      • et al.
      Physical activity in anorexia nervosa: how relevant is it to therapy response?.
      ], since beneficial effects on the individual's well-being and positive influences on body composition have been reported [
      • Achamrah N.
      • Coeffier M.
      • Dechelotte P.
      Physical activity in patients with anorexia nervosa.
      ,
      • Rizk M.
      • Lalanne C.
      • Berthoz S.
      • Kern L.
      • Godart N.
      EVHAN Group
      Problematic exercise in anorexia nervosa: testing potential risk factors against different definitions.
      ,
      • Nicholls D.
      • Hudson L.
      • Mahomed F.
      Managing anorexia nervosa.
      ,
      • Sauchelli S.
      • Arcelus J.
      • Sanchez I.
      • Riesco N.
      • Jimenez-Murcia S.
      • Granero R.
      • et al.
      Physical activity in anorexia nervosa: how relevant is it to therapy response?.
      ,
      • Keyes A.
      • Woerwag-Mehta S.
      • Bartholdy S.
      • Koskina A.
      • Middleton B.
      • Connan F.
      • et al.
      Physical activity and the drive to exercise in anorexia nervosa.
      ,
      • Fernandez-del-Valle M.
      • Larumbe-Zabala E.
      • Graell-Berna M.
      • Perez-Ruiz M.
      Anthropometric changes in adolescents with anorexia nervosa in response to resistance training.
      ]. This may contribute to more sustainable therapy [
      • Achamrah N.
      • Coeffier M.
      • Dechelotte P.
      Physical activity in patients with anorexia nervosa.
      ,
      • Cockfield A.
      • Philpot U.
      Feeding size 0: the challenges of anorexia nervosa. Managing anorexia from a dietitian's perspective.
      ,
      • Nicholls D.
      • Hudson L.
      • Mahomed F.
      Managing anorexia nervosa.
      ], and does not necessarily impair weight gain [
      • Achamrah N.
      • Coeffier M.
      • Dechelotte P.
      Physical activity in patients with anorexia nervosa.
      ,
      • Vancampfort D.
      • Vanderlinden J.
      • De Hert M.
      • Soundy A.
      • Adamkova M.
      • Skjaerven L.H.
      • et al.
      A systematic review of physical therapy interventions for patients with anorexia and bulimia nervosa.
      ,
      • Cook B.J.
      • Wonderlich S.A.
      • Mitchell J.E.
      • Thompson R.
      • Sherman R.
      • McCallum K.
      Exercise in eating disorders treatment: systematic review and proposal of guidelines.
      ,
      • Ng L.W.
      • Ng D.P.
      • Wong W.P.
      Is supervised exercise training safe in patients with anorexia nervosa? A meta-analysis.
      ]; on the contrary, it may reinforce it [
      • Achamrah N.
      • Coeffier M.
      • Dechelotte P.
      Physical activity in patients with anorexia nervosa.
      ].
      Most commonly, body mass index (BMI) and the speed of weight gain are used for assessing nutritional status and therapy progress [
      • Fernandez-del-Valle M.
      • Larumbe-Zabala E.
      • Graell-Berna M.
      • Perez-Ruiz M.
      Anthropometric changes in adolescents with anorexia nervosa in response to resistance training.
      ,
      • Herpertz-Dahlmann B.
      • Schwarte R.
      • Krei M.
      • Egberts K.
      • Warnke A.
      • Wewetzer C.
      • et al.
      Day-patient treatment after short inpatient care versus continued inpatient treatment in adolescents with anorexia nervosa (ANDI): a multicentre, randomised, open-label, non-inferiority trial.
      ,
      • Zipfel S.
      • Wild B.
      • Gross G.
      • Friederich H.C.
      • Teufel M.
      • Schellberg D.
      • et al.
      ANTOP study group
      Focal psychodynamic therapy, cognitive behaviour therapy, and optimised treatment as usual in outpatients with anorexia nervosa (ANTOP study): randomised controlled trial.
      ]. Thus, the patient's body composition remains unconsidered, which is a substantial health criterion. Despite expected decreased muscle and fat masses in AN patients [
      • Nicholls D.
      • Hudson L.
      • Mahomed F.
      Managing anorexia nervosa.
      ,
      • Yamashita S.
      • Kawai K.
      • Yamanaka T.
      • Inoo T.
      • Yokoyama H.
      • Morita C.
      • et al.
      BMI, body composition, and the energy requirement for body weight gain in patients with anorexia nervosa.
      ], an excessive gain in abdominal body fat is a known side effect of the current therapy strategies [
      • Achamrah N.
      • Coeffier M.
      • Dechelotte P.
      Physical activity in patients with anorexia nervosa.
      ,
      • El Ghoch M.
      • Calugi S.
      • Lamburghini S.
      • Dalle Grave R.
      Anorexia nervosa and body fat distribution: a systematic review.
      ,
      • El Ghoch M.
      • Milanese C.
      • Calugi S.
      • Pellegrini M.
      • Battistini N.C.
      • Dalle Grave R.
      Body composition, eating disorder psychopathology, and psychological distress in anorexia nervosa: a longitudinal study.
      ,
      • Mayer L.
      • Walsh B.T.
      • Pierson Jr., R.N.
      • Heymsfield S.B.
      • Gallagher D.
      • Wang J.
      • et al.
      Body fat redistribution after weight gain in women with anorexia nervosa.
      ,
      • Mayer L.E.
      • Klein D.A.
      • Black E.
      • Attia E.
      • Shen W.
      • Mao X.
      • et al.
      Adipose tissue distribution after weight restoration and weight maintenance in women with anorexia nervosa.
      ,
      • Achamrah N.
      • Nobis S.
      • Breton J.
      • Jesus P.
      • Belmonte L.
      • Maurer B.
      • et al.
      Maintaining physical activity during refeeding improves body composition, intestinal hyperpermeability and behavior in anorectic mice.
      ], and a major risk factor for relapse [
      • Marzola E.
      • Nasser J.A.
      • Hashim S.A.
      • Shih P.A.
      • Kaye W.H.
      Nutritional rehabilitation in anorexia nervosa: review of the literature and implications for treatment.
      ,
      • El Ghoch M.
      • Calugi S.
      • Lamburghini S.
      • Dalle Grave R.
      Anorexia nervosa and body fat distribution: a systematic review.
      ,
      • Mayer L.E.
      • Klein D.A.
      • Black E.
      • Attia E.
      • Shen W.
      • Mao X.
      • et al.
      Adipose tissue distribution after weight restoration and weight maintenance in women with anorexia nervosa.
      ] since it enhances body image disturbances and concerns about body shape [
      • Gutierrez E.
      • Carrera O.
      Anorexia nervosa and body-image disturbance.
      ,
      • Gailledrat L.
      • Rousselet M.
      • Venisse J.L.
      • Lambert S.
      • Rocher B.
      • Remaud M.
      • et al.
      Marked body shape concerns in female patients suffering from eating disorders: relevance of a clinical sub-group.
      ], which are diagnostic criteria for AN [
      American Psychiatric Association
      ,
      ]. There is still a lack of satisfactory assessment tools for the therapy progress and body composition determination in therapy guidelines and clinical practice [
      National Collaborating Centre for Mental Health (UK)
      Eating disorders: core interventions in the treatment and management of anorexia nervosa, bulimia nervosa and related eating disorders.
      ,
      American Psychiatric Association
      Practice guideline for the treatment of patients with eating disorders. 3rd ed.
      ,
      • Yager J.
      • Devlin M.J.
      • Halmi K.A.
      • Herzog D.B.
      • Mitchell J.E.
      • Powers P.
      • et al.
      Guideline watch (August 2012): practice guideline for the treatment of patients with eating disorders. 3rd ed.
      ,
      • Mayer L.E.
      • Roberto C.A.
      • Glasofer D.R.
      • Etu S.F.
      • Gallagher D.
      • Wang J.
      • et al.
      Does percent body fat predict outcome in anorexia nervosa?.
      ]. Widely used methods have known inherent problems [
      • Ackland T.R.
      • Lohman T.G.
      • Sundgot-Borgen J.
      • Maughan R.J.
      • Meyer N.L.
      • Stewart A.D.
      • et al.
      Current status of body composition assessment in sport: review and position statement on behalf of the ad hoc research working group on body composition health and performance, under the auspices of the I.O.C. Medical Commission.
      ]. Dual-energy X-ray absorptiometry (DXA) is considered to be the reference method; however, its application is not very feasible in clinical routine. Bioelectrical impedance analysis (BIA) is often used in the field [
      • Aguera Z.
      • Romero X.
      • Arcelus J.
      • Sanchez I.
      • Riesco N.
      • Jimenez-Murcia S.
      • et al.
      Changes in body composition in anorexia nervosa: predictors of recovery and treatment outcome.
      ,
      • Mika C.
      • Herpertz-Dahlmann B.
      • Heer M.
      • Holtkamp K.
      Improvement of nutritional status as assessed by multifrequency BIA during 15 weeks of refeeding in adolescent girls with anorexia nervosa.
      ,
      • Marra M.
      • Sammarco R.
      • De Filippo E.
      • Caldara A.
      • Speranza E.
      • Scalfi L.
      • et al.
      Prediction of body composition in anorexia nervosa: results from a retrospective study.
      ], but it has limited reproducibility [
      • Savegnago Mialich M.
      • Maria Faccioli Sicchieri J.
      • Afonso Jordao Junior A.
      Analysis of body composition: a critical review of the use of bioelectrical impedance analysis.
      ] and accuracy [
      • Ackland T.R.
      • Lohman T.G.
      • Sundgot-Borgen J.
      • Maughan R.J.
      • Meyer N.L.
      • Stewart A.D.
      • et al.
      Current status of body composition assessment in sport: review and position statement on behalf of the ad hoc research working group on body composition health and performance, under the auspices of the I.O.C. Medical Commission.
      ], and low validity [
      • Talma H.
      • Chinapaw M.J.
      • Bakker B.
      • HiraSing R.A.
      • Terwee C.B.
      • Altenburg T.M.
      Bioelectrical impedance analysis to estimate body composition in children and adolescents: a systematic review and evidence appraisal of validity, responsiveness, reliability and measurement error.
      ], especially in AN patients [
      • Marra M.
      • Sammarco R.
      • De Filippo E.
      • Caldara A.
      • Speranza E.
      • Scalfi L.
      • et al.
      Prediction of body composition in anorexia nervosa: results from a retrospective study.
      ,
      • Kyle U.G.
      • Bosaeus I.
      • De Lorenzo A.D.
      • Deurenberg P.
      • Elia M.
      • Gomez J.M.
      • et al.
      Composition of the ESPEN Working Group. Bioelectrical impedance analysis-part I: review of principles and methods.
      ,
      • Kyle U.G.
      • Bosaeus I.
      • De Lorenzo A.D.
      • Deurenberg P.
      • Elia M.
      • Manuel Gomez J.
      • et al.
      ESPEN
      Bioelectrical impedance analysis-part II: utilization in clinical practice.
      ].
      The aim of the present study was to assess subcutaneous adipose tissue (SAT) in AN patients by a novel ultrasound method which allows fat thickness layer measurements with high accuracy and precision to gain detailed information on fat patterning in these patients [
      • Störchle P.
      • Müller W.
      • Sengeis M.
      • Ahammer H.
      • Fürhapter-Rieger A.
      • Bachl N.
      • et al.
      Standardized ultrasound measurement of subcutaneous fat patterning: high reliability and accuracy in groups ranging from lean to obese.
      ]. Ultrasound results were compared with bioimpedance analysis. Additionally, a new measure for relative body weight, the mass index (MI) [
      • Müller W.
      Determinants of ski-jump performance and implications for health, safety and fairness.
      ,
      • Müller W.
      Towards research-based approaches for solving body composition problems in sports: ski jumping as a heuristic example.
      ], which considers individual body proportions, was applied in addition to BMI. It was hypothesized that AN patients have very low SAT values, and that both measures for relative body weight do not sufficiently predict body fat.

      2. Materials and methods

      2.1 Participants

      2.1.1 Recruitment and selection criteria

      Female patients with AN (n = 18) according to International Classification of Diseases (ICD-10) criteria [
      ], aged between 18 and 40 years, were recruited from three psychiatric clinics in Graz, Austria. Exclusion criteria were: acute or chronic diseases or infection, alcohol or drug abuse, major cognitive deficits, life-threatening conditions during AN, history of digestive diseases (e.g. inflammatory bowel diseases and irritable bowel syndrome), history of gastrointestinal surgery, treatment with antibiotics and intake of pre- or probiotics within the previous 2 months, pregnancy, or breastfeeding. The study population was a subgroup of a larger cross-sectional study (five groups of different energy status n = 107). The study was conducted according to the Helsinki Declaration and approved by the ethics committee of the Medical University of Graz (MUG-26-383ex13/14). All participants gave their written informed consent for anonymous use of their data.

      2.1.2 Additional information on the study population

      Dieticians provided dietary advices. The nutritional treatment was based on high-caloric diets and recommended reduction of physical activity. Additional information on physical activity (International Physical Activity Questionnaire IPAQ Score) [
      • Craig C.L.
      • Marshall A.L.
      • Sjostrom M.
      • Bauman A.E.
      • Booth M.L.
      • Ainsworth B.E.
      • et al.
      International physical activity questionnaire: 12-country reliability and validity.
      ], nutritional intake (repeated 24-h recalls analyzed by a national specific software) [
      • Denkwerkzeuge dato
      Software: nut.s science - nutritional software, v1.32.61; Vienna.
      ], history of weight cycling [
      • Wallner S.J.
      • Luschnigg N.
      • Schnedl W.J.
      • Lahousen T.
      • Sudi K.
      • Crailsheim K.
      • et al.
      Body fat distribution of overweight females with a history of weight cycling.
      ], family history, and depression status were evaluated [
      • Beck A.T.
      • Ward C.H.
      • Mendelson M.
      • Mock J.
      • Erbaugh J.
      An inventory for measuring depression.
      ,
      • Hamilton M.
      A rating scale for depression.
      ]. Demographic and clinical data (education, marital status, medication, smoking status) were collected [
      • Fagerstrom K.O.
      Measuring degree of physical dependence to tobacco smoking with reference to individualization of treatment.
      ]. Information on the disease and treatment history of the patients are provided in the appendix. Patient numbers were ordered according to their DINCL values (P1 P18). DINCL represents the sum of SAT thicknesses of eight standardized body sites measured by ultrasound. The patients were assigned to two groups: Group 1 (lower SAT) with DINCL values below and Group 2 (higher SAT) above the DINCL median.

      2.2 Laboratory assessments

      Standard blood values were determined and dermal carotenoids were assessed at the palm (resonance Raman spectroscopy) [
      • Zidichouski J.A.
      • Mastaloudis A.
      • Poole S.J.
      • Reading J.C.
      • Smidt C.R.
      Clinical validation of a noninvasive, Raman spectroscopic method to assess carotenoid nutritional status in humans.
      ,
      • Mayne S.T.
      • Cartmel B.
      • Scarmo S.
      • Lin H.
      • Leffell D.J.
      • Welch E.
      • et al.
      Noninvasive assessment of dermal carotenoids as a biomarker of fruit and vegetable intake.
      ]. Oxidative stress parameters were determined in serum: total peroxides (TOC®) [
      • Tatzber F.
      • Griebenow S.
      • Wonisch W.
      • Winkler R.
      Dual method for the determination of peroxidase activity and total peroxides-iodide leads to a significant increase of peroxidase activity in human sera.
      ], endogenous peroxidase-activity (EPA®), total antioxidative capacity (TAC®) (Labor Diagnostic Nord, Nordhorn, Germany). Titers of autoantibodies against oxidized LDL (oLAb®) (Biomedica, Vienna, Austria) and malondialdehyde-modified LDL (MDA-LDL IgM®) (Omnignostica GmbH, Höflein/Klosterneuburg, Austria) were measured by ELISA [
      • Tatzber F.
      • Esterbauer H.
      Autoantibodies to oxidized low density lipoprotein.
      ,
      • Resch U.
      • Tatzber F.
      • Budinsky A.
      • Sinzinger H.
      Reduction of oxidative stress and modulation of autoantibodies against modified low-density lipoprotein after rosuvastatin therapy.
      ]. Plasma leptin levels were measured by ELISA (BioVendor, Brno, Czech Republic).

      2.3 Anthropometry

      Measurements of body mass (m), body height (h), and sitting height (s) were performed in accordance with the International Society for the Advancement of Kinanthropometry (ISAK) [
      • Stewart A.
      • Marfell-Jones M.
      • Olds T.
      • Deridder H.
      International standards for anthropometric assessment.
      ]. The BMI (m/h2) and MI = 0.53 m/(h s) were calculated [
      • Müller W.
      Determinants of ski-jump performance and implications for health, safety and fairness.
      ,
      • Müller W.
      Towards research-based approaches for solving body composition problems in sports: ski jumping as a heuristic example.
      ,
      • Müller W.
      • Gröschl W.
      • Müller R.
      • Sudi K.
      Underweight in ski jumping: the solution of the problem.
      ]. The MI considered individual sitting height for assessing relative body weight: in individuals with long legs, the MI was higher than the BMI and vice versa. The SAT mass (mSAT) was calculated according to: mSAT (kg) = 0.65 d8 S ρ, with d8 being the mean of DINCL, S the body surface area (according to Du Bois: S = 0.20247 h0.725 m0.425) [
      • Du Bois D.
      • Du Bois E.F.
      A formula to estimate the approximate surface area if height and weight be known.
      ], and ρ the density of fat (0.92 kg/m3) [
      • Herman I.P.
      Physics of the human body.
      ]. The calibration factor of 0.65 resulted from comparative measurements at 216 randomly distributed sites in a set of test persons [
      • Storchle P.
      • Muller W.
      • Sengeis M.
      • Lackner S.
      • Holasek S.
      • Furhapter-Rieger A.
      Measurement of mean subcutaneous fat thickness: eight standardised ultrasound sites compared to 216 randomly selected sites.
      ]. The SAT thicknesses were accurately measured values that should primarily be used, whereas the calculated SAT mass included model assumptions.

      2.4 Ultrasound imaging technique

      Brightness mode ultrasound (US) accurately measures SAT thicknesses and is reproducible in all groups from extremely lean to obese people [
      • Ackland T.R.
      • Lohman T.G.
      • Sundgot-Borgen J.
      • Maughan R.J.
      • Meyer N.L.
      • Stewart A.D.
      • et al.
      Current status of body composition assessment in sport: review and position statement on behalf of the ad hoc research working group on body composition health and performance, under the auspices of the I.O.C. Medical Commission.
      ,
      • Störchle P.
      • Müller W.
      • Sengeis M.
      • Ahammer H.
      • Fürhapter-Rieger A.
      • Bachl N.
      • et al.
      Standardized ultrasound measurement of subcutaneous fat patterning: high reliability and accuracy in groups ranging from lean to obese.
      ,
      • Müller W.
      • Maughan R.J.
      The need for a novel approach to measure body composition: is ultrasound an answer?.
      ,
      • Müller W.
      • Lohman T.G.
      • Stewart A.D.
      • Maughan R.J.
      • Meyer N.L.
      • Sardinha L.B.
      • et al.
      Subcutaneous fat patterning in athletes: selection of appropriate sites and standardisation of a novel ultrasound measurement technique: ad hoc working group on body composition, health and performance, under the auspices of the IOC Medical Commission.
      ]. The standardized eight sites (upper abdomen (UA), lower abdomen (LA), erector spinae (ES), distal triceps (DT), brachioradialis (BR), lateral thigh (LT), front thigh (FT), and medial calf (MC)) were measured [
      • Müller W.
      • Lohman T.G.
      • Stewart A.D.
      • Maughan R.J.
      • Meyer N.L.
      • Sardinha L.B.
      • et al.
      Subcutaneous fat patterning in athletes: selection of appropriate sites and standardisation of a novel ultrasound measurement technique: ad hoc working group on body composition, health and performance, under the auspices of the IOC Medical Commission.
      ]. Lateral thigh was replaced by external oblique (EO) because LT was defined after the study data were collected [
      • Müller W.
      • Horn M.
      • Fürhapter-Rieger A.
      • Kainz P.
      • Kröpfl J.M.
      • Ackland T.R.
      • et al.
      Body composition in sport: interobserver reliability of a novel ultrasound measure of subcutaneous fat tissue.
      ,
      • Müller W.
      • Horn M.
      • Fürhapter-Rieger A.
      • Kainz P.
      • Kröpfl J.M.
      • Maughan R.J.
      • et al.
      Body composition in sport: a comparison of a novel ultrasound imaging technique to measure subcutaneous fat tissue compared with skinfold measurement.
      ]. These body sites represent the body parts, trunk, arms, and legs. Sites are defined with respect to the person's body height, which ensures interpersonal comparability. A thick gel layer between the probe and the skin was used to prevent SAT compression. Figure. 1 shows a typical US image with clearly visible SAT layer borders. For US imaging, a conventional US system (GE Logiq-e, General Electric) with a linear probe (L8-18i RS) operated at 8–16 MHz was used. The US images were evaluated with semiautomatic evaluation software (Rotosport, Stattegg, Austria). The software provides information on DINCL, DEXCL (SAT thicknesses without embedded tissues, e.g. fibrous structures) and calculates embedded structures (DES = DINCLDEXCL). The US measurement was performed by two certified investigators (iasms.org, 2-day course and supervised post-course training).
      Fig. 1
      Fig. 1Example of a typical ultrasound image. Ultrasound site: brachioradialis; subcutaneous adipose tissue (SAT) layers can be detected uncompressed due to the application of a thick gel layer between the probe and the skin. The upper and lower borders of SAT (dermis and muscle fascia, respectively) are clearly visible. The area in-between represents the SAT layer. Embedded fibrous structures are also visible. SAT thicknesses were calculated via the semiautomatic image segmentation software. Speed of sound in fat: c = 1450 m/s. SAT, subcutaneous adipose tissue.

      2.5 Bioelectrical impedance analysis

      Single-frequency bioelectrical impedance analysis (BIA) (BIA 101, Akern) was conducted according to recommended procedures [
      • Kyle U.G.
      • Bosaeus I.
      • De Lorenzo A.D.
      • Deurenberg P.
      • Elia M.
      • Gomez J.M.
      • et al.
      Composition of the ESPEN Working Group. Bioelectrical impedance analysis-part I: review of principles and methods.
      ,
      • Kyle U.G.
      • Bosaeus I.
      • De Lorenzo A.D.
      • Deurenberg P.
      • Elia M.
      • Manuel Gomez J.
      • et al.
      ESPEN
      Bioelectrical impedance analysis-part II: utilization in clinical practice.
      ], and analyzed with the BodyComposition–Professional software v9.0.14325, which uses equations from Sun et al. for calculating fat free mass (FFM) and total body water [
      • Sun S.S.
      • Chumlea W.C.
      • Heymsfield S.B.
      • Lukaski H.C.
      • Schoeller D.
      • Friedl K.
      • et al.
      Development of bioelectrical impedance analysis prediction equations for body composition with the use of a multicomponent model for use in epidemiologic surveys.
      ], and Sergie et al. for extracellular water calculation [
      • Sergi G.
      • Bussolotto M.
      • Perini P.
      • Calliari I.
      • Giantin V.
      • Ceccon A.
      • et al.
      Accuracy of bioelectrical impedance analysis in estimation of extracellular space in healthy subjects and in fluid retention states.
      ]. Total body fat (TBF) was calculated by subtracting FFM from body mass (m): TBF = m – FFM. Resistance (R), reactance (Xc), and phase angle (PA) were measured at 50 kHz. This software was used for the comparison with the US measurement because it is commercially available and therefore widely used in clinical practice. Additionally, other BIA equations that are supposed to be more appropriate in AN patients were applied. For example, the equations from Deurenberg and Kushner were used [
      • Deurenberg P.
      • van der Kooy K.
      • Leenen R.
      • Weststrate J.A.
      • Seidell J.C.
      Sex and age specific prediction formulas for estimating body composition from bioelectrical impedance: a cross-validation study.
      ,
      • Kushner R.F.
      • Schoeller D.A.
      Estimation of total body water by bioelectrical impedance analysis.
      ], which Matter et al. suggested for AN patients [
      • Mattar L.
      • Godart N.
      • Melchior J.C.
      • Falissard B.
      • Kolta S.
      • Ringuenet D.
      • et al.
      Underweight patients with anorexia nervosa: comparison of bioelectrical impedance analysis using five equations to dual X-ray absorptiometry.
      ].

      2.6 Statistics

      SPSS Statistics v23 software (IBM, Armonk, NY, USA) was used for statistical analysis. Shapiro-Wilk tests revealed that not all distributions were normal. Descriptive parameters were presented as mean ± standard deviation (SD) when data were normally distributed, otherwise the median and interquartile ranges (IQR) were used. For group comparisons Student's t-test and Mann-Whitney U test were used, respectively. Chi-squared test was used for qualitative variables. The Pearson correlation coefficient (r) was used for data in Fig. 5 and Supplemental Fig. 1 (appendix). The significance level was set at P < 0.05. All analyses were explanatorily interpreted.

      3. Results

      3.1 Body mass index and body fat

      The study population had a mean BMI of 15.3 ± 1.3 kg/m2 and a mean DINCL of 29.13 ± 18.69 mm. Some patients had almost the same BMI, but DINCL (and DEXCL) differed substantially (Fig. 2, Table 1).
      Fig. 2
      Fig. 2Comparison of fat patterning in two anorexia nervosa patients with the same BMI (four selected sites). A series of subcutaneous adipose tissue (SAT) ultrasound measurements at the four selected sites upper abdomen (UA), lower abdomen (LA), front thigh (FT) and distal triceps (DT) of patients P5 and P14 who had the same BMI is shown here. Additional anthropometric data and information on the measurement and evaluation procedure are provided in the appendix. Mean SAT thicknesses (in mm) of P5 are: UA: 1.94, LA: 3.65, FT: 0.28, DT: 0.17. The sum of the eight standardized site collection (not all are shown here) DINCL = 10.15 mm. Mean SAT thicknesses (in mm) of P14 are: UA: 6.54, LA: 11.89, FT: 6.51, DT: 4.50, DINCL = 43.88 mm. SAT mass resulted in 1.2 kg for P5 and 4.5 kg for P14. Similar cases of great variations of SAT layer thickness at every single site and DINCL were also found in several other patients with almost the same BMI. MI, body mass index; DINCL, the sum of SAT layers at the eight standardized ultrasound sites; DT, distal triceps; FT, front thigh; LA, lower abdomen; P, patient; SAT, subcutaneous adipose tissue; UA, upper abdomen.
      Table 1Anthropometric and body composition data, and striking blood values of the study population (Group 1 and Group 2) that was divided into the subgroups Group 1 below and Group 2 above the median of DINCL (the sum of subcutaneous adipose tissue (SAT) of the eight body sites measured by ultrasound). When data were normally distributed, they were presented as means and standard deviation (SD), otherwise they were reported as median and interquartile range (IQR) in italic font.
      NGroup 1 lower SATGroup 2 higher SATGroup 1 and Group 2P
      9918
      Mean MedianSD IQRMean MedianSD IQRMean MedianSD IQR
      Age, years22.73.622.23.022.43.20.778
      Anthropometry
      Height, m1.680.081.650.061.660.070.340
      Sitting height, m0.870.040.850.030.860.040.216
      Body mass, kg40.915.4143.704.0042.314.840.232
      BMI, kg/m214.51.116.10.915.31.30.003
      MI, kg/m214.81.216.60.815.71.40.002
      Ultrasound measurement of subcutaneous adipose tissue
      DINCL, mm13.669.5544.5910.6229.1318.69<0.001
      DEXCL, mm11.768.6240.5710.4426.1717.49<0.001
      DES, mm2.111.753.750.963.081.75<0.001
      DES, %14%9%10%
      Fat patterning measured by ultrasound
      UA, mm2.322.075.471.923.902.450.004
      LA, mm2.632.179.912.986.274.38<0.001
      ES, mm2.011.694.441.723.232.020.008
      DT, mm1.691.806.141.653.922.76<0.001
      BR, mm0.380.632.581.781.102.28<0.001
      EO, mm0.450.673.122.471.682.73<0.001
      FT, mm2.351.897.632.154.993.26<0.001
      MC, mm1.001.935.244.332.914.470.002
      Single frequency bioimpedance analysis
      R, ohm690.060.3767.255.2728.668.70.012
      Xc, ohm56.19.772.810.464.413.00.003
      PA, °4.650.685.410.605.030.730.022
      RI, cm2/ohm41.14.435.53.938.34.90.012
      Plasma values – selection of striking parameters
      Leptin, ng/ml1.000.153.704.391.552.78<0.001
      Oxidative stress
      TAC, mmol/l0.690.311.210.180.950.370.001
      MDA-LDL IgM, mU/ml87.87116.60302.85548.23137.79407.080.024
      Liver
      ALT, U/l26.116.5819.117.3422.617.660.049
      Heart
      LDH, U/l180.5620.09157.6720.18169.1122.810.028
      Nutritive aspects
      Carotenoid, Counts43,88912,21126,33312,79735,11115,1270.009
      ALT, alanine aminotransferase; BMI, body mass index; BR, brachioradialis; DES, embedded fibrous structures; DEXCL, the sum of subcutaneous adipose tissue (SAT) layer thicknesses without embedded fibrous structures (DES = DINCL – DEXCL); DINCL, the sum of SAT layers at the eight standardized ultrasound sites; DT, distal triceps; EO, external oblique; ES, erector spinae; FT, front thigh; LA, lower abdomen; LDH, lactate dehydrogenase; MC, medial calf; MDA-LDL IgM, malondialdehyde-modified low density lipoprotein immunoglobulin M; MI, mass index (0.53 × body mass/(body height × sitting height)); PA, phase angle; R, resistance; RI, resistance index (height2/resistance); SAT, subcutaneous adipose tissue; TAC, total antioxidative capacity; UA, upper abdomen; Xc, reactance.
      In some individuals, SAT thickness was quite high, although the BMI was extremely low (Fig. 3A, appendix). The thickness of the fat layers varied by several hundred percent at a given BMI. For instance, two patients with BMIs of 13.2 kg/m2 and 13.3 kg/m2 had DINCL values of 1.3 mm and 24.4 mm, respectively. Another example: BMIs of 15.2 kg/m2 and 15.1 kg/m2 and according DINCL values of 10.2 mm and 43.9 mm, respectively. The correlation of DINCL and BMI was r (18) = 0.741, P < 0.001.
      Fig. 3
      Fig. 3A. Body mass index versus DINCL. The patients (P1 P18) are ordered according to increasing BMI. Within this group of anorexia nervosa patients, the sums of eight subcutaneous adipose tissue (SAT) thicknesses (DINCL) ranged from 1.3 to 58.2 mm. The patient numbers P1 P18 were assigned according to increasing DINCL values. The two patients described in are highlighted in this figure to point out the discrepancy for observed SAT and BMI classification. BMI, body mass index; DINCL, the sum of SAT layers at the eight standardized ultrasound sites; P, patient; SAT, subcutaneous adipose tissue.
      B. Body mass index and mass index. Patients P1 P18 are ordered according to increasing subcutaneous adipose tissue (SAT) thicknesses. MI = 0.53m/(hs); BMI = m/h2. A BMI ≤17.5 kg/m2 is one criterion of ICD-10 anorexia nervosa diagnosis, and the WHO underweight border ≤18.5 kg/m2 are both highlighted with horizontal dashed lines. The vertical dashed line symbolizes the median of DINCL. It divides the patients into subgroup Group 1 with low SAT layers and Group 2 with higher SAT amount. BMI, body mass index; DINCL, the sum of SAT layers at the eight standardized ultrasound sites; ICD-10, International Classification of Diseases 10th revision; MI, mass index; P, patient; SAT, subcutaneous adipose tissue; WHO, World Health Organization.

      3.2 Measures for relative body weight: body mass index vs mass index

      When using mass index (MI) instead of BMI, patient P17 would not be classified as an AN patient (BMI ≤ 17.5 kg/m2, ICD-10). In 10 patients the MI differed by ≥0.5 kg/m2 to the BMI. Figure. 3B shows the participants’ relative body weights in terms of BMI and MI [
      • Müller W.
      Determinants of ski-jump performance and implications for health, safety and fairness.
      ,
      • Müller W.
      Towards research-based approaches for solving body composition problems in sports: ski jumping as a heuristic example.
      ]. In this group of AN patients, the mean MI (15.7 kg/m2) was larger than the mean BMI (15.3 kg/m2) [
      • Ackland T.R.
      • Lohman T.G.
      • Sundgot-Borgen J.
      • Maughan R.J.
      • Meyer N.L.
      • Stewart A.D.
      • et al.
      Current status of body composition assessment in sport: review and position statement on behalf of the ad hoc research working group on body composition health and performance, under the auspices of the I.O.C. Medical Commission.
      ,
      • Müller W.
      • Gröschl W.
      • Müller R.
      • Sudi K.
      Underweight in ski jumping: the solution of the problem.
      ].

      3.3 Body mass index and subcutaneous fat thickness

      The group of investigated AN patients showed a large range of DINCL from 1.3 mm to 58.2 mm. This indicates enormously divergent body compositions between patients. A median DINCL = 30.17 mm was used to divide the group of AN patients into two subgroups. Detailed descriptive data of the groups are shown in Table 1. One subgroup of patients had extremely low SAT values (M = 13.7 mm, SD = 9.6), whereas the other subgroup had surprisingly high amounts of SAT (M = 44.6 mm, SD = 10.6), despite extremely low BMIs. Comparisons of DINCL and DEXCL showed significant differences (t(16) = −6.498, P < 0.001) and t(16) = −6.384, P < 0.001), respectively). The relative amount of embedded structures (DES = DINCLDEXCL) decreased with increasing SAT thickness. The mean percentage of DES of Group 2 (higher SAT) was 9%, whereas the mean percentage in Group 1 (lower SAT) was 14%.

      3.4 Fat patterning in anorexia nervosa patients

      Fat patterning of the subgroups was compared. SAT layers significantly differed between the two groups at every single site (P < 0.01) (Fig. 4, Table 1). The SAT layers were extremely low at all sites in Group 1 (lower SAT): medians ranged from 0.4 to 2.1 mm. In Group 2 (higher SAT), the median SAT thickness was >5 mm at four sites: UA, LA, FT, MC (appendix).
      Fig. 4
      Fig. 4Comparison of fat patterning between the two anorexia nervosa groups. Medians of dINCL (SAT thickness at a single measurement site) of Group 1 with lower SAT (n = 9) and Group 2 with higher SAT (n = 9) are compared with each other. Group 2 has higher SAT thickness at every measured body site. BR, brachioradialis; dINCL, SAT thickness at a single measurement site; DT, distal triceps; EO, external oblique; ES, erector spinae; FT, front thigh; LA, lower abdomen; MC, medial calf; SAT, subcutaneous adipose tissue; UA, upper abdomen.

      3.5 Bioelectrical impedance analysis measurement of body fat

      Figure. 5 depicts contradictory results obtained with the US method compared to BIA (TBF). For example, BIA assessed 1 kg of TBF for four patients, but DINCL varied from 1.3 to 24.4 mm. Six patients with a mean TBF of 3.9 ± 0.3 kg according to BIA had DINCL values ranging from 10.2 to 43.9 mm. The correlation of DINCL and BIA TBF (kg) was r(18) = 0.759, P < 0.001.
      Fig. 5
      Fig. 5Individual differences of DINCL and total body fat with bioelectrical impedance analysis. Several patients with the same or similar total body fat (TBF) determined by bioelectrical impedance analysis (BIA) had totally different body fat measured by ultrasound. The arrows point out the great discrepancy between the DINCL values and TBF determined by BIA. For example, TBF of all patients that are highlighted as squares resulted in 1 kg according to BIA, but ultrasound measurement revealed DINCL values ranging from 1.3 to 24.4 mm (which amounts to about 0.1–2.6 kg SAT). Patients with similar TBF measured by BIA are labelled with the same symbols. Also, for patients with similar DINCL values BIA results differed extremely. For example, the two encircled patients P8 and P12 had DINCL values of 24.4 and 33.4 mm (2.6 kg and 3.7 kg SAT) and BIA measurement resulted in 1 kg and 8.2 kg TBF, respectively. BIA, bioelectrical impedance analysis; DINCL, the sum of SAT layers at the eight standardized ultrasound sites; P, patient; SAT, subcutaneous adipose tissue; TBF, total body fat.

      3.6 Additional information

      3.6.1 Energy and nutrient intake

      Although the patients received similar dietary treatment, the two groups reported significant differences in energy intake: Group 1 (lower SAT): M = 10,100 kJ, SD = 3100/day (2400 ± 700 kcal; 59 ± 14 kcal/kg body mass); Group 2 (higher SAT): M = 6200 kJ, SD = 3800/day (1500 ± 900 kcal; 35 ± 22 kcal/kg body mass). Besides the higher energy intake, Group 1 (lower SAT) also reported a higher protein and fat intake than Group 2 (higher SAT). The intake of other macronutrients such as carbohydrates, monosaccharides, fiber, saturated fatty acids and water showed no significant differences. Group 1 (lower SAT) reported having more frequently consumed high energy supplements than Group 2 (higher SAT) (not significant).

      3.6.2 Lifestyle information

      Physical activity level, smoking habits and other assessed patients’ data (appendix) did not differ significantly between the two groups. SAT thicknesses were not correlated with physical activity as determined by International Physical Activity Questionnaire.

      3.6.3 Laboratory assessments

      Laboratory chemistry revealed no deviation from the reference blood values, including parameters of lipid and carbohydrate metabolism, kidney, liver and thyroid function. However, Alanin-Aminotransferase (ALT) (t(16) = 2.130, P = 0.049) and lactate-dehydrogenase (LDH) (t(16) = 2.411, P = 0.028) were significantly higher in the lower SAT group (Group 1) (Table 1).
      Plasma leptin levels ranged from 1.0–7.7 ng/ml. Group 1 (lower SAT) had mean leptin values of 1.1 ± 0.3 ng/ml and Group 2 (higher SAT) a mean of 4.2 ± 2.3 ng/ml. Leptin differed significantly between the two groups (U = 1, P < 0.001).
      Regarding oxidative stress, the TAC of AN patients (M = 0.95 mmol/l, SD = 0.37) was low (reference: > 1.3 mmol/l). In addition, TAC (t(12.9) = −4.341, P = 0.01) and MDA-LDL IgM (U = 15, P = 0.024) were significantly lower in the lower SAT Group 1 (Table 1). Although in Group 2 (higher SAT) the antioxidative levels with respect to TAC, TOC, EPA, oLAB and MDA-LDL IgM were lower compared to reference values, their concentrations were still higher compared to the levels of Group 1 (lower SAT). According to these indicators, Group 1 (lower SAT) had higher oxidative stress than Group 2 (higher SAT).

      3.6.4 Dermal carotenoid level

      The level of accumulated carotenoids in the skin at the palm measured by resonance Raman spectroscopy differed significantly between the groups (t(16) = 2.978, P = 0.009, Table 1).

      4. Discussion

      This study applied, for the first time, novel approaches for determining both relative body weight and body fat in AN patients.

      4.1 Measures for relative body weight

      According to ICD-10 (F.50.0) and DSM-V, a BMI cut-off is among the three main criteria for diagnosis of AN. Regarding BMI, the World Health Organization (WHO) points out that ‘problems arise, however, in adults whose shape differs from the norm, particularly those whose legs are shorter or longer than might be expected for their height.’ [
      WHO Expert Committee
      Physical status: the use and interpretation of anthropometry.
      ] Based on this remark, the MI, which considers individual leg lengths, was introduced [
      • Müller W.
      Determinants of ski-jump performance and implications for health, safety and fairness.
      ,
      • Müller W.
      Towards research-based approaches for solving body composition problems in sports: ski jumping as a heuristic example.
      ,
      • Müller W.
      • Gröschl W.
      • Müller R.
      • Sudi K.
      Underweight in ski jumping: the solution of the problem.
      ]. The MI was larger than the BMI in 13 cases and lower in two cases (Fig. 3B) in the current study. Using the MI would shift some of the individuals' relative body weight beyond the AN criterion of <17.5 kg/m2. In individuals with relatively long or short legs, the difference between BMI and MI can be a full unit or more (e.g. P8: BMI = 13.3 kg/m2, MI = 14.4 kg/m2). However, relative body weight without accurate and reliable assessment of body fat is a weak criterion for healthy weight [
      • Nicholls D.
      • Hudson L.
      • Mahomed F.
      Managing anorexia nervosa.
      ,
      WHO Expert Committee
      Physical status: the use and interpretation of anthropometry.
      ,
      • Fernandez-del-Valle M.
      • Larumbe-Zabala E.
      • Morande-Lavin G.
      • Perez Ruiz M.
      Muscle function and body composition profile in adolescents with restrictive anorexia nervosa: does resistance training help?.
      ].

      4.2 Body composition assessment

      The US method revealed great variability of SAT in AN patients [
      • Störchle P.
      • Müller W.
      • Sengeis M.
      • Ahammer H.
      • Fürhapter-Rieger A.
      • Bachl N.
      • et al.
      Standardized ultrasound measurement of subcutaneous fat patterning: high reliability and accuracy in groups ranging from lean to obese.
      ,
      • Müller W.
      • Lohman T.G.
      • Stewart A.D.
      • Maughan R.J.
      • Meyer N.L.
      • Sardinha L.B.
      • et al.
      Subcutaneous fat patterning in athletes: selection of appropriate sites and standardisation of a novel ultrasound measurement technique: ad hoc working group on body composition, health and performance, under the auspices of the IOC Medical Commission.
      ]. Half of the patients (Group 2, higher SAT) had SAT amounts comparable with healthy normal weight women (Fig. 2, Fig. 3A) [
      • Ackland T.R.
      • Müller W.
      Imaging method: ultrasound.
      ]. According to preliminary reference values for DINCL [
      • Ackland T.R.
      • Müller W.
      Imaging method: ultrasound.
      ], values from 35 to 50 mm are considered as ‘desirable range’ for athletes, and 35–80 mm for all other women. The median of the current group (30.2 mm) was close to the lower border of the ‘desirable range’. This was found, although EO was used instead of LT. In women SAT thickness is substantially higher in LT compared to EO; using LT would shift the median towards even higher values. The two subgroups' BMI medians differed by 12%, whereas DINCL values differed by 330%, indicating that the large difference in SAT (and thus also of TBF [
      • Ibrahim M.M.
      Subcutaneous and visceral adipose tissue: structural and functional differences.
      ]) cannot be captured by the BMI or the MI. Plasma leptin was expectantly low in all AN patients; however, it also significantly differed between the groups, underpinning the observed differences in body composition. Additionally, the two groups showed great differences in the fat patterning (Fig. 4). Several individuals had almost the same BMI although their SAT amount differed enormously (Fig. 2, Fig. 3A, B).
      Yager and Anderson mentioned a loss of SAT as a common sign and symptom of AN [
      • Yager J.
      • Andersen A.E.
      Clinical practice. Anorexia nervosa.
      ]. However, this study found that the reduced weight did not indicate low SAT in all patients. In Group 2 (higher SAT, DINCL > 30 mm) the patients' extremely low weight must have resulted from losses of other body structures, particularly muscle, organ and bone mass. Four patients had DINCL values > 50 mm, which is considered as ‘ballast fat’ in competitive female sports [
      • Ackland T.R.
      • Müller W.
      Imaging method: ultrasound.
      ].
      Group 1 (lower SAT) reported significantly higher energy intake than Group 2 (higher SAT). However, AN patients may overestimate portion sizes according to their altered perception. Furthermore, the lower SAT Group 1 (n = 7) received energy dense sip food more often than the higher SAT Group 2 (n = 3). The generally increased oxidative stress status indicates the body's challenge in severe catabolic metabolism [
      • Solmi M.
      • Veronese N.
      • Manzato E.
      • Sergi G.
      • Favaro A.
      • Santonastaso P.
      • et al.
      Oxidative stress and antioxidant levels in patients with anorexia nervosa: a systematic review and exploratory meta-analysis.
      ]. Hypercarotenemia and altered lipid metabolism occur in some AN patients [
      • Solmi M.
      • Veronese N.
      • Manzato E.
      • Sergi G.
      • Favaro A.
      • Santonastaso P.
      • et al.
      Oxidative stress and antioxidant levels in patients with anorexia nervosa: a systematic review and exploratory meta-analysis.
      ,
      • Boland B.
      • Beguin C.
      • Zech F.
      • Desager J.P.
      • Lambert M.
      Serum beta-carotene in anorexia nervosa patients: a case-control study.
      ]. The observed differences in oxidative stress and carotenoid concentration (Table 1) may indicate various metabolic disturbances due to body fat content.
      However, there are currently no established threshold values available for minimum fat [
      • Sundgot-Borgen J.
      • Meyer N.L.
      • Lohman T.G.
      • Ackland T.R.
      • Maughan R.J.
      • Stewart A.D.
      • et al.
      How to minimise the health risks to athletes who compete in weight-sensitive sports review and position statement on behalf of the Ad Hoc Research Working Group on Body Composition, Health and Performance, under the auspices of the IOC Medical Commission.
      ]. This is partly the case because sufficiently accurate measurement methods were missing, and the acceptable minimum of fat may be genetically predisposed [
      • Sundgot-Borgen J.
      • Meyer N.L.
      • Lohman T.G.
      • Ackland T.R.
      • Maughan R.J.
      • Stewart A.D.
      • et al.
      How to minimise the health risks to athletes who compete in weight-sensitive sports review and position statement on behalf of the Ad Hoc Research Working Group on Body Composition, Health and Performance, under the auspices of the IOC Medical Commission.
      ]. Also in very lean people, US provides a reliable and accurate tool to measure SAT [
      • Müller W.
      • Lohman T.G.
      • Stewart A.D.
      • Maughan R.J.
      • Meyer N.L.
      • Sardinha L.B.
      • et al.
      Subcutaneous fat patterning in athletes: selection of appropriate sites and standardisation of a novel ultrasound measurement technique: ad hoc working group on body composition, health and performance, under the auspices of the IOC Medical Commission.
      ], whereas widely used methods like DXA [
      • Ackland T.R.
      • Lohman T.G.
      • Sundgot-Borgen J.
      • Maughan R.J.
      • Meyer N.L.
      • Stewart A.D.
      • et al.
      Current status of body composition assessment in sport: review and position statement on behalf of the ad hoc research working group on body composition health and performance, under the auspices of the I.O.C. Medical Commission.
      ], magnetic resonance imaging (MRI) scans, skin folds or BIA do not reach the necessary accuracy (see appendix) [
      • Ackland T.R.
      • Lohman T.G.
      • Sundgot-Borgen J.
      • Maughan R.J.
      • Meyer N.L.
      • Stewart A.D.
      • et al.
      Current status of body composition assessment in sport: review and position statement on behalf of the ad hoc research working group on body composition health and performance, under the auspices of the I.O.C. Medical Commission.
      ,
      • Müller W.
      • Horn M.
      • Fürhapter-Rieger A.
      • Kainz P.
      • Kröpfl J.M.
      • Maughan R.J.
      • et al.
      Body composition in sport: a comparison of a novel ultrasound imaging technique to measure subcutaneous fat tissue compared with skinfold measurement.
      ]. The current study compared fat mass assessed by BIA and US results (DINCL). The TBF determined by BIA revealed inaccurate results in individuals (Fig. 5), although the correlation coefficient between DINCL and BIA TBF was high. Also, the results obtained from formulas, suggested for AN patients [
      • Mattar L.
      • Godart N.
      • Melchior J.C.
      • Falissard B.
      • Kolta S.
      • Ringuenet D.
      • et al.
      Underweight patients with anorexia nervosa: comparison of bioelectrical impedance analysis using five equations to dual X-ray absorptiometry.
      ], deviated substantially from DINCL. The BIA equations often rely on comparisons with DXA with its known shortcomings (particularly in lean persons), instead of comparing it to multi-component methods [
      • Wang Z.
      • Shen W.
      • Withers R.T.
      • Heymsfield S.B.
      Multicomponent molecular-level models of body composition anlysis.
      ]. However, no algorithm can compensate the basic shortcomings of the BIA method (appendix) [
      • Kerr A.
      • Hume P.A.
      Non-imaging method: bioelectrical impedance analysis.
      ].
      Differences in body composition in AN patients at treatment baseline have previously been observed and a decrease in skeletal muscle and internal organ mass was associated with decreased fat mass [
      • Yamashita S.
      • Kawai K.
      • Yamanaka T.
      • Inoo T.
      • Yokoyama H.
      • Morita C.
      • et al.
      BMI, body composition, and the energy requirement for body weight gain in patients with anorexia nervosa.
      ,
      • Mayer L.E.
      • Roberto C.A.
      • Glasofer D.R.
      • Etu S.F.
      • Gallagher D.
      • Wang J.
      • et al.
      Does percent body fat predict outcome in anorexia nervosa?.
      ,
      • Bodell L.P.
      • Mayer L.E.
      Percent body fat is a risk factor for relapse in anorexia nervosa: a replication study.
      ]. Since the current study observed large differences in SAT at the same BMI, muscle and other organ masses must have independently decreased from fat mass. For example, participant P14 (BMI = 15.1 kg/m2, m = 40.0 kg; Fig. 2, Fig. 3A) had a DINCL of 43.9 mm. There was no need to increase the fat level of this patient [
      • Ackland T.R.
      • Müller W.
      Imaging method: ultrasound.
      ]. Her SAT mass amounted to 11.4% of her body mass; this percentage did not contain the visceral fat and fat embedded in other organs. In contrast to P14, patient P5 (BMI = 15.2 kg/m2, m = 45.5 kg) had the extremely low SAT mass of 2.5% of her body weight; in this case, interventions focusing on increasing fat mass are obviously important. In Group 1 (lower SAT), percentage of fibrous structures was higher, which further reduced the pure amount of fat (Table 1, appendix).

      4.3 Consequences for the therapeutic approach

      In common treatment approaches, high caloric diet and limited physical activity is suggested to rapidly increase body weight [
      National Collaborating Centre for Mental Health (UK)
      Eating disorders: core interventions in the treatment and management of anorexia nervosa, bulimia nervosa and related eating disorders.
      ,
      American Psychiatric Association
      Practice guideline for the treatment of patients with eating disorders. 3rd ed.
      ,
      • Yager J.
      • Devlin M.J.
      • Halmi K.A.
      • Herzog D.B.
      • Mitchell J.E.
      • Powers P.
      • et al.
      Guideline watch (August 2012): practice guideline for the treatment of patients with eating disorders. 3rd ed.
      ]. Practice guidelines suggest that ‘for severely underweight patient, exercise should be restricted and always carefully supervised and monitored’ [
      American Psychiatric Association
      Practice guideline for the treatment of patients with eating disorders. 3rd ed.
      ]; however, these authors state that further research is needed [
      • Yager J.
      • Devlin M.J.
      • Halmi K.A.
      • Herzog D.B.
      • Mitchell J.E.
      • Powers P.
      • et al.
      Guideline watch (August 2012): practice guideline for the treatment of patients with eating disorders. 3rd ed.
      ]. Rapid weight restoration is essential [
      • Attia E.
      • Walsh B.T.
      Behavioral management for anorexia nervosa.
      ,
      • Wilson G.T.
      • Shafran R.
      Eating disorders guidelines from NICE.
      ,
      • Treasure J.
      • Claudino A.M.
      • Zucker N.
      Eating disorders.
      ,
      • Marzola E.
      • Nasser J.A.
      • Hashim S.A.
      • Shih P.A.
      • Kaye W.H.
      Nutritional rehabilitation in anorexia nervosa: review of the literature and implications for treatment.
      ,
      National Collaborating Centre for Mental Health (UK)
      Eating disorders: core interventions in the treatment and management of anorexia nervosa, bulimia nervosa and related eating disorders.
      ,
      American Psychiatric Association
      Practice guideline for the treatment of patients with eating disorders. 3rd ed.
      ]; however, too quick weight gain at the beginning of treatment is considered to be unfavorable for later weight maintenance and long-term recovery [
      • Marzola E.
      • Nasser J.A.
      • Hashim S.A.
      • Shih P.A.
      • Kaye W.H.
      Nutritional rehabilitation in anorexia nervosa: review of the literature and implications for treatment.
      ,
      • Cockfield A.
      • Philpot U.
      Feeding size 0: the challenges of anorexia nervosa. Managing anorexia from a dietitian's perspective.
      ,
      • Fernandez-del-Valle M.
      • Larumbe-Zabala E.
      • Graell-Berna M.
      • Perez-Ruiz M.
      Anthropometric changes in adolescents with anorexia nervosa in response to resistance training.
      ]. Fast weight gain is often associated with abdominal fat accumulation [
      • Achamrah N.
      • Coeffier M.
      • Dechelotte P.
      Physical activity in patients with anorexia nervosa.
      ,
      • El Ghoch M.
      • Calugi S.
      • Lamburghini S.
      • Dalle Grave R.
      Anorexia nervosa and body fat distribution: a systematic review.
      ,
      • El Ghoch M.
      • Milanese C.
      • Calugi S.
      • Pellegrini M.
      • Battistini N.C.
      • Dalle Grave R.
      Body composition, eating disorder psychopathology, and psychological distress in anorexia nervosa: a longitudinal study.
      ,
      • Achamrah N.
      • Nobis S.
      • Breton J.
      • Jesus P.
      • Belmonte L.
      • Maurer B.
      • et al.
      Maintaining physical activity during refeeding improves body composition, intestinal hyperpermeability and behavior in anorectic mice.
      ], which can have negative metabolic effects [
      • El Ghoch M.
      • Calugi S.
      • Lamburghini S.
      • Dalle Grave R.
      Anorexia nervosa and body fat distribution: a systematic review.
      ]. AN is associated with body image disturbance [
      • Gutierrez E.
      • Carrera O.
      Anorexia nervosa and body-image disturbance.
      ], and body shape concerns [
      • Gailledrat L.
      • Rousselet M.
      • Venisse J.L.
      • Lambert S.
      • Rocher B.
      • Remaud M.
      • et al.
      Marked body shape concerns in female patients suffering from eating disorders: relevance of a clinical sub-group.
      ], therefore, an inadequate gain of body fat can be expected to negatively influence compliance. However, El Gouch et al. did not find a connection between body fat gain and long-term outcome [
      • El Ghoch M.
      • Calugi S.
      • Chignola E.
      • Bazzani P.V.
      • Dalle Grave R.
      Body mass index, body fat and risk factor of relapse in anorexia nervosa.
      ]. Mayer et al. observed that the accumulated body fat disappeared after long-term weight restoration [
      • Mayer L.
      • Walsh B.T.
      • Pierson Jr., R.N.
      • Heymsfield S.B.
      • Gallagher D.
      • Wang J.
      • et al.
      Body fat redistribution after weight gain in women with anorexia nervosa.
      ,
      • Mayer L.E.
      • Klein D.A.
      • Black E.
      • Attia E.
      • Shen W.
      • Mao X.
      • et al.
      Adipose tissue distribution after weight restoration and weight maintenance in women with anorexia nervosa.
      ]. Nevertheless, the relapse rate in AN patients is high [
      • Smink F.R.
      • van Hoeken D.
      • Hoek H.W.
      Epidemiology of eating disorders: incidence, prevalence and mortality rates.
      ], and thus long-term weight restored patients are rare. Strategies to increase muscle mass may enhance therapeutic success. It is indicated to hypothesize that strength training (few repetitions, thus low energy turnover) may be advantageous to increase muscle mass and avoid excessive gain of fat mass [
      • Brooks G.A.
      • Fahey T.D.
      • Baldwin K.M.
      Exercise physiology: human bioenergetics and its applications.
      ]. The restoration of lean body mass is a key determinant of outcome and quality of life [
      • Achamrah N.
      • Coeffier M.
      • Dechelotte P.
      Physical activity in patients with anorexia nervosa.
      ]. Accurate body composition testing should be routinely implemented in the standard care of AN patients [
      • Mayer L.E.
      • Roberto C.A.
      • Glasofer D.R.
      • Etu S.F.
      • Gallagher D.
      • Wang J.
      • et al.
      Does percent body fat predict outcome in anorexia nervosa?.
      ]. The potential therapeutic benefits have also been pointed out by Yamashita et al. [
      • Yamashita S.
      • Kawai K.
      • Yamanaka T.
      • Inoo T.
      • Yokoyama H.
      • Morita C.
      • et al.
      BMI, body composition, and the energy requirement for body weight gain in patients with anorexia nervosa.
      ].

      4.4 Limitations of the study

      In this study, 18 AN patients who had already received treatment were included. Further studies should include larger groups. Longitudinal studies will be necessary to test the suggested modifications of treatment practices. Information on nutrient intake and activity levels were based on self-assessment of the patients. The IPAQ could not map possible differences in physical activity habits because most participants were inpatients.

      5. Conclusion

      Ultrasound measurement enabled accurate monitoring of body fat. This method revealed enormous differences in SAT among AN patients with similar BMIs, and biochemical parameters (leptin, ALT, LDH, oxidative stress indicators, and carotenoid levels) corresponded with this finding. Although all patients were characterized by very low BMI, SAT thicknesses varied from extremely low to normal ranges. Half of the patients had sufficiently high fat mass. Body mass index is not an adequate criterion for classification of body fat mass in AN patients. The current data suggest a rethink of current treatment practice. For those with extremely low muscle mass, activity is recommended to increase muscle mass at a low energy turnover rate, which may also improve compliance and thus therapeutic outcome and long-term recovery.

      Funding statement

      This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. All researchers act completely independently from funding.

      Statement of authorship

      SL, SM, WM, and SH designed the project and the manuscript. WM introduced the mass index MI and together with AF supported the accurate application of the ultrasound fat measurement. Patients were recruited by SM, CB, OA and ML. The investigations were conducted by SL, SM, AO, CB and OA. Oxidative stress parameters were measured by WW, MM suggested to include dermal carotenoid status assessment, and chemical laboratory diagnostics were conducted by HM and SZ. All authors contributed to the final form of the manuscript.

      Conflicts of interest

      WM and AF contributed to developing the commercially available image evaluation software used here and participate in the returns. Except for this, all authors have declared that no competing interests exist.

      Acknowledgements

      We would like to thank all the participants for their patience and their interest in research, Josef Smolle who was the Initiator of the ESAN-Study IC3490, Medical University Graz, and the Doctoral School of the Medical University of Graz for partially providing the open access fees.

      Appendix A. Supplementary data

      figS1
      figS2
      figS3

      References

        • Smink F.R.
        • van Hoeken D.
        • Hoek H.W.
        Epidemiology of eating disorders: incidence, prevalence and mortality rates.
        Curr Psychiatr Rep. 2012; 14: 406-414
        • Yager J.
        • Andersen A.E.
        Clinical practice. Anorexia nervosa.
        N Engl J Med. 2005; 353: 1481-1488
        • Achamrah N.
        • Coeffier M.
        • Dechelotte P.
        Physical activity in patients with anorexia nervosa.
        Nutr Rev. 2016; 74: 301-311
        • Attia E.
        • Walsh B.T.
        Behavioral management for anorexia nervosa.
        N Engl J Med. 2009; 360: 500-506
        • Wilson G.T.
        • Shafran R.
        Eating disorders guidelines from NICE.
        Lancet. 2005; 365: 79-81
        • Treasure J.
        • Claudino A.M.
        • Zucker N.
        Eating disorders.
        Lancet. 2010; 375: 583-593
        • Marzola E.
        • Nasser J.A.
        • Hashim S.A.
        • Shih P.A.
        • Kaye W.H.
        Nutritional rehabilitation in anorexia nervosa: review of the literature and implications for treatment.
        BMC Psychiatry. 2013; 13 (290-29244X-13-290)
        • National Collaborating Centre for Mental Health (UK)
        Eating disorders: core interventions in the treatment and management of anorexia nervosa, bulimia nervosa and related eating disorders.
        The British Psychological Society & The Royal College of Psychiatrists, 2004
        • American Psychiatric Association
        Practice guideline for the treatment of patients with eating disorders. 3rd ed.
        Am J Psychiatry. 2006; 163: 1-54
        • Agostino H.
        • Erdstein J.
        • Di Meglio G.
        Shifting paradigms: continuous nasogastric feeding with high caloric intakes in anorexia nervosa.
        J Adolesc Health. 2013; 53: 590-594
        • Garber A.K.
        • Mauldin K.
        • Michihata N.
        • Buckelew S.M.
        • Shafer M.A.
        • Moscicki A.B.
        Higher calorie diets increase rate of weight gain and shorten hospital stay in hospitalized adolescents with anorexia nervosa.
        J Adolesc Health. 2013; 53: 579-584
        • Mehanna H.M.
        • Moledina J.
        • Travis J.
        Refeeding syndrome: what it is, and how to prevent and treat it.
        BMJ. 2008; 336: 1495-1498
        • Cockfield A.
        • Philpot U.
        Feeding size 0: the challenges of anorexia nervosa. Managing anorexia from a dietitian's perspective.
        Proc Nutr Soc. 2009; 68: 281-288
        • Rizk M.
        • Lalanne C.
        • Berthoz S.
        • Kern L.
        • Godart N.
        • EVHAN Group
        Problematic exercise in anorexia nervosa: testing potential risk factors against different definitions.
        PLoS One. 2015; 10e0143352
        • Gianini L.M.
        • Klein D.A.
        • Call C.
        • Walsh B.T.
        • Wang Y.
        • Wu P.
        • et al.
        Physical activity and post-treatment weight trajectory in anorexia nervosa.
        Int J Eat Disord. 2016; 49: 482-489
        • Nicholls D.
        • Hudson L.
        • Mahomed F.
        Managing anorexia nervosa.
        Arch Dis Child. 2011; 96: 977-982
        • El Ghoch M.
        • Calugi S.
        • Pellegrini M.
        • Chignola E.
        • Dalle Grave R.
        Physical activity, body weight, and resumption of menses in anorexia nervosa.
        Psychiatr Res. 2016; 246: 507-511
        • Sauchelli S.
        • Arcelus J.
        • Sanchez I.
        • Riesco N.
        • Jimenez-Murcia S.
        • Granero R.
        • et al.
        Physical activity in anorexia nervosa: how relevant is it to therapy response?.
        Eur Psychiatry. 2015; 30: 924-931
        • Keyes A.
        • Woerwag-Mehta S.
        • Bartholdy S.
        • Koskina A.
        • Middleton B.
        • Connan F.
        • et al.
        Physical activity and the drive to exercise in anorexia nervosa.
        Int J Eat Disord. 2015; 48: 46-54
        • Fernandez-del-Valle M.
        • Larumbe-Zabala E.
        • Graell-Berna M.
        • Perez-Ruiz M.
        Anthropometric changes in adolescents with anorexia nervosa in response to resistance training.
        Eat Weight Disord. 2015; 20: 311-317
        • Vancampfort D.
        • Vanderlinden J.
        • De Hert M.
        • Soundy A.
        • Adamkova M.
        • Skjaerven L.H.
        • et al.
        A systematic review of physical therapy interventions for patients with anorexia and bulimia nervosa.
        Disabil Rehabil. 2014; 36: 628-634
        • Cook B.J.
        • Wonderlich S.A.
        • Mitchell J.E.
        • Thompson R.
        • Sherman R.
        • McCallum K.
        Exercise in eating disorders treatment: systematic review and proposal of guidelines.
        Med Sci Sports Exerc. 2016; 48: 1408-1414
        • Ng L.W.
        • Ng D.P.
        • Wong W.P.
        Is supervised exercise training safe in patients with anorexia nervosa? A meta-analysis.
        Physiotherapy. 2013; 99: 1-11
        • Herpertz-Dahlmann B.
        • Schwarte R.
        • Krei M.
        • Egberts K.
        • Warnke A.
        • Wewetzer C.
        • et al.
        Day-patient treatment after short inpatient care versus continued inpatient treatment in adolescents with anorexia nervosa (ANDI): a multicentre, randomised, open-label, non-inferiority trial.
        Lancet. 2014; 383: 1222-1229
        • Zipfel S.
        • Wild B.
        • Gross G.
        • Friederich H.C.
        • Teufel M.
        • Schellberg D.
        • et al.
        • ANTOP study group
        Focal psychodynamic therapy, cognitive behaviour therapy, and optimised treatment as usual in outpatients with anorexia nervosa (ANTOP study): randomised controlled trial.
        Lancet. 2014; 383: 127-137
        • Yamashita S.
        • Kawai K.
        • Yamanaka T.
        • Inoo T.
        • Yokoyama H.
        • Morita C.
        • et al.
        BMI, body composition, and the energy requirement for body weight gain in patients with anorexia nervosa.
        Int J Eat Disord. 2010; 43: 365-371
        • El Ghoch M.
        • Calugi S.
        • Lamburghini S.
        • Dalle Grave R.
        Anorexia nervosa and body fat distribution: a systematic review.
        Nutrients. 2014; 6: 3895-3912
        • El Ghoch M.
        • Milanese C.
        • Calugi S.
        • Pellegrini M.
        • Battistini N.C.
        • Dalle Grave R.
        Body composition, eating disorder psychopathology, and psychological distress in anorexia nervosa: a longitudinal study.
        Am J Clin Nutr. 2014; 99: 771-778
        • Mayer L.
        • Walsh B.T.
        • Pierson Jr., R.N.
        • Heymsfield S.B.
        • Gallagher D.
        • Wang J.
        • et al.
        Body fat redistribution after weight gain in women with anorexia nervosa.
        Am J Clin Nutr. 2005; 81: 1286-1291
        • Mayer L.E.
        • Klein D.A.
        • Black E.
        • Attia E.
        • Shen W.
        • Mao X.
        • et al.
        Adipose tissue distribution after weight restoration and weight maintenance in women with anorexia nervosa.
        Am J Clin Nutr. 2009; 90: 1132-1137
        • Achamrah N.
        • Nobis S.
        • Breton J.
        • Jesus P.
        • Belmonte L.
        • Maurer B.
        • et al.
        Maintaining physical activity during refeeding improves body composition, intestinal hyperpermeability and behavior in anorectic mice.
        Sci Rep. 2016; 6: 21887
        • Gutierrez E.
        • Carrera O.
        Anorexia nervosa and body-image disturbance.
        Lancet Psychiatry. 2016; 3: e9-e10
        • Gailledrat L.
        • Rousselet M.
        • Venisse J.L.
        • Lambert S.
        • Rocher B.
        • Remaud M.
        • et al.
        Marked body shape concerns in female patients suffering from eating disorders: relevance of a clinical sub-group.
        PLoS One. 2016; 11e0165232
        • American Psychiatric Association
        Diagnostic and statistical manual of mental health disorders: DSM-5. 5th ed. American Psychiatric Publishing, Washington DC2013
      1. World Health Organization International statistical classification of diseases and related health problems: 10th revision (ICD-10). 5th ed. WHO Library Cataloguing-in-Publication Data, Geneva2016
        • Yager J.
        • Devlin M.J.
        • Halmi K.A.
        • Herzog D.B.
        • Mitchell J.E.
        • Powers P.
        • et al.
        Guideline watch (August 2012): practice guideline for the treatment of patients with eating disorders. 3rd ed.
        Focus. 2014; 12: 416-431
        • Mayer L.E.
        • Roberto C.A.
        • Glasofer D.R.
        • Etu S.F.
        • Gallagher D.
        • Wang J.
        • et al.
        Does percent body fat predict outcome in anorexia nervosa?.
        Am J Psychiatry. 2007; 164: 970-972
        • Ackland T.R.
        • Lohman T.G.
        • Sundgot-Borgen J.
        • Maughan R.J.
        • Meyer N.L.
        • Stewart A.D.
        • et al.
        Current status of body composition assessment in sport: review and position statement on behalf of the ad hoc research working group on body composition health and performance, under the auspices of the I.O.C. Medical Commission.
        Sports Med. 2012; 42: 227-249
        • Aguera Z.
        • Romero X.
        • Arcelus J.
        • Sanchez I.
        • Riesco N.
        • Jimenez-Murcia S.
        • et al.
        Changes in body composition in anorexia nervosa: predictors of recovery and treatment outcome.
        PLoS One. 2015; 10e0143012
        • Mika C.
        • Herpertz-Dahlmann B.
        • Heer M.
        • Holtkamp K.
        Improvement of nutritional status as assessed by multifrequency BIA during 15 weeks of refeeding in adolescent girls with anorexia nervosa.
        J Nutr. 2004; 134: 3026-3030
        • Marra M.
        • Sammarco R.
        • De Filippo E.
        • Caldara A.
        • Speranza E.
        • Scalfi L.
        • et al.
        Prediction of body composition in anorexia nervosa: results from a retrospective study.
        Clin Nutr. 2018; 37: 1670-1674
        • Savegnago Mialich M.
        • Maria Faccioli Sicchieri J.
        • Afonso Jordao Junior A.
        Analysis of body composition: a critical review of the use of bioelectrical impedance analysis.
        International J Clin Nutr. 2014; 2: 1-10
        • Talma H.
        • Chinapaw M.J.
        • Bakker B.
        • HiraSing R.A.
        • Terwee C.B.
        • Altenburg T.M.
        Bioelectrical impedance analysis to estimate body composition in children and adolescents: a systematic review and evidence appraisal of validity, responsiveness, reliability and measurement error.
        Obes Rev. 2013; 14: 895-905
        • Kyle U.G.
        • Bosaeus I.
        • De Lorenzo A.D.
        • Deurenberg P.
        • Elia M.
        • Gomez J.M.
        • et al.
        Composition of the ESPEN Working Group. Bioelectrical impedance analysis-part I: review of principles and methods.
        Clin Nutr. 2004; 23: 1226-1243
        • Kyle U.G.
        • Bosaeus I.
        • De Lorenzo A.D.
        • Deurenberg P.
        • Elia M.
        • Manuel Gomez J.
        • et al.
        • ESPEN
        Bioelectrical impedance analysis-part II: utilization in clinical practice.
        Clin Nutr. 2004; 23: 1430-1453
        • Störchle P.
        • Müller W.
        • Sengeis M.
        • Ahammer H.
        • Fürhapter-Rieger A.
        • Bachl N.
        • et al.
        Standardized ultrasound measurement of subcutaneous fat patterning: high reliability and accuracy in groups ranging from lean to obese.
        Ultrasound Med Biol. 2017; 43: 427-438
        • Müller W.
        Determinants of ski-jump performance and implications for health, safety and fairness.
        Sports Med. 2009; 39: 85-106
        • Müller W.
        Towards research-based approaches for solving body composition problems in sports: ski jumping as a heuristic example.
        Br J Sports Med. 2009; 43: 1013-1019
        • Craig C.L.
        • Marshall A.L.
        • Sjostrom M.
        • Bauman A.E.
        • Booth M.L.
        • Ainsworth B.E.
        • et al.
        International physical activity questionnaire: 12-country reliability and validity.
        Med Sci Sports Exerc. 2003; 35: 1381-1395
        • Denkwerkzeuge dato
        Software: nut.s science - nutritional software, v1.32.61; Vienna.
        2018
        http://www.nutritional-software
        Date accessed: October 3, 2018
        • Wallner S.J.
        • Luschnigg N.
        • Schnedl W.J.
        • Lahousen T.
        • Sudi K.
        • Crailsheim K.
        • et al.
        Body fat distribution of overweight females with a history of weight cycling.
        Int J Obes Relat Metab Disord. 2004; 28: 1143-1148
        • Beck A.T.
        • Ward C.H.
        • Mendelson M.
        • Mock J.
        • Erbaugh J.
        An inventory for measuring depression.
        Arch Gen Psychiatr. 1961; 4: 561-571
        • Hamilton M.
        A rating scale for depression.
        J Neurol Neurosurg Psychiatry. 1960; 23: 56-62
        • Fagerstrom K.O.
        Measuring degree of physical dependence to tobacco smoking with reference to individualization of treatment.
        Addict Behav. 1978; 3: 235-241
        • Zidichouski J.A.
        • Mastaloudis A.
        • Poole S.J.
        • Reading J.C.
        • Smidt C.R.
        Clinical validation of a noninvasive, Raman spectroscopic method to assess carotenoid nutritional status in humans.
        J Am Coll Nutr. 2009; 28: 687-693
        • Mayne S.T.
        • Cartmel B.
        • Scarmo S.
        • Lin H.
        • Leffell D.J.
        • Welch E.
        • et al.
        Noninvasive assessment of dermal carotenoids as a biomarker of fruit and vegetable intake.
        Am J Clin Nutr. 2010; 92: 794-800
        • Tatzber F.
        • Griebenow S.
        • Wonisch W.
        • Winkler R.
        Dual method for the determination of peroxidase activity and total peroxides-iodide leads to a significant increase of peroxidase activity in human sera.
        Anal Biochem. 2003; 316: 147-153
        • Tatzber F.
        • Esterbauer H.
        Autoantibodies to oxidized low density lipoprotein.
        in: Bellomo G. Finardi G. Maggi E. Rice-Evans C. Free radicals IX. Richelieu Press, London1995: 245-267
        • Resch U.
        • Tatzber F.
        • Budinsky A.
        • Sinzinger H.
        Reduction of oxidative stress and modulation of autoantibodies against modified low-density lipoprotein after rosuvastatin therapy.
        Br J Clin Pharmacol. 2006; 61: 262-274
        • Stewart A.
        • Marfell-Jones M.
        • Olds T.
        • Deridder H.
        International standards for anthropometric assessment.
        2011
        • Müller W.
        • Gröschl W.
        • Müller R.
        • Sudi K.
        Underweight in ski jumping: the solution of the problem.
        Int J Sports Med. 2006; 27: 926-934
        • Du Bois D.
        • Du Bois E.F.
        A formula to estimate the approximate surface area if height and weight be known.
        Nutrition 1989. 1916; 5 (discussion 312-3): 303-311
        • Herman I.P.
        Physics of the human body.
        Springer, Berlin, Heidelberg, New York2007
        • Storchle P.
        • Muller W.
        • Sengeis M.
        • Lackner S.
        • Holasek S.
        • Furhapter-Rieger A.
        Measurement of mean subcutaneous fat thickness: eight standardised ultrasound sites compared to 216 randomly selected sites.
        Sci Rep. 2018; 8 (16268-018-34213-0)
        • Müller W.
        • Maughan R.J.
        The need for a novel approach to measure body composition: is ultrasound an answer?.
        Br J Sports Med. 2013; 47: 1001-1002
        • Müller W.
        • Lohman T.G.
        • Stewart A.D.
        • Maughan R.J.
        • Meyer N.L.
        • Sardinha L.B.
        • et al.
        Subcutaneous fat patterning in athletes: selection of appropriate sites and standardisation of a novel ultrasound measurement technique: ad hoc working group on body composition, health and performance, under the auspices of the IOC Medical Commission.
        Br J Sports Med. 2016; 50: 45-54
        • Müller W.
        • Horn M.
        • Fürhapter-Rieger A.
        • Kainz P.
        • Kröpfl J.M.
        • Ackland T.R.
        • et al.
        Body composition in sport: interobserver reliability of a novel ultrasound measure of subcutaneous fat tissue.
        Br J Sports Med. 2013; 47: 1036-1043
        • Müller W.
        • Horn M.
        • Fürhapter-Rieger A.
        • Kainz P.
        • Kröpfl J.M.
        • Maughan R.J.
        • et al.
        Body composition in sport: a comparison of a novel ultrasound imaging technique to measure subcutaneous fat tissue compared with skinfold measurement.
        Br J Sports Med. 2013; 47: 1028-1035
        • Sun S.S.
        • Chumlea W.C.
        • Heymsfield S.B.
        • Lukaski H.C.
        • Schoeller D.
        • Friedl K.
        • et al.
        Development of bioelectrical impedance analysis 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
        • Sergi G.
        • Bussolotto M.
        • Perini P.
        • Calliari I.
        • Giantin V.
        • Ceccon A.
        • et al.
        Accuracy of bioelectrical impedance analysis in estimation of extracellular space in healthy subjects and in fluid retention states.
        Ann Nutr Metab. 1994; 38: 158-165
        • Deurenberg P.
        • van der Kooy K.
        • Leenen R.
        • Weststrate J.A.
        • Seidell J.C.
        Sex and age specific prediction formulas for estimating body composition from bioelectrical impedance: a cross-validation study.
        Int J Obes. 1991; 15: 17-25
        • Kushner R.F.
        • Schoeller D.A.
        Estimation of total body water by bioelectrical impedance analysis.
        Am J Clin Nutr. 1986; 44: 417-424
        • Mattar L.
        • Godart N.
        • Melchior J.C.
        • Falissard B.
        • Kolta S.
        • Ringuenet D.
        • et al.
        Underweight patients with anorexia nervosa: comparison of bioelectrical impedance analysis using five equations to dual X-ray absorptiometry.
        Clin Nutr. 2011; 30: 746-752
        • WHO Expert Committee
        Physical status: the use and interpretation of anthropometry.
        (Technical report series 854)1995
        • Fernandez-del-Valle M.
        • Larumbe-Zabala E.
        • Morande-Lavin G.
        • Perez Ruiz M.
        Muscle function and body composition profile in adolescents with restrictive anorexia nervosa: does resistance training help?.
        Disabil Rehabil. 2016; 38: 346-353
        • Ackland T.R.
        • Müller W.
        Imaging method: ultrasound.
        in: Hume P.A. Kerr D.A. Ackland T.R. Best practice protocols for physique assessment in sport. 1st ed. Springer, Singapore2018: 131-141
        • Ibrahim M.M.
        Subcutaneous and visceral adipose tissue: structural and functional differences.
        Obes Rev. 2010; 11: 11-18
        • Solmi M.
        • Veronese N.
        • Manzato E.
        • Sergi G.
        • Favaro A.
        • Santonastaso P.
        • et al.
        Oxidative stress and antioxidant levels in patients with anorexia nervosa: a systematic review and exploratory meta-analysis.
        Int J Eat Disord. 2015; 48: 826-841
        • Boland B.
        • Beguin C.
        • Zech F.
        • Desager J.P.
        • Lambert M.
        Serum beta-carotene in anorexia nervosa patients: a case-control study.
        Int J Eat Disord. 2001; 30: 299-305
        • Sundgot-Borgen J.
        • Meyer N.L.
        • Lohman T.G.
        • Ackland T.R.
        • Maughan R.J.
        • Stewart A.D.
        • et al.
        How to minimise the health risks to athletes who compete in weight-sensitive sports review and position statement on behalf of the Ad Hoc Research Working Group on Body Composition, Health and Performance, under the auspices of the IOC Medical Commission.
        Br J Sports Med. 2013; 47: 1012-1022
        • Wang Z.
        • Shen W.
        • Withers R.T.
        • Heymsfield S.B.
        Multicomponent molecular-level models of body composition anlysis.
        in: Heymsfield S.B. Lohman T.G. Wang Z. Going S.B. Human body composition. 2nd ed. Human Kinetics, Champaign2005: 163-176
        • Kerr A.
        • Hume P.A.
        Non-imaging method: bioelectrical impedance analysis.
        in: Hume P.A. Kerr D.A. Ackland T.R. Best practice protocols for physique assessment in sport. 1st ed. Springer, Singapore2018: 101-116
        • Bodell L.P.
        • Mayer L.E.
        Percent body fat is a risk factor for relapse in anorexia nervosa: a replication study.
        Int J Eat Disord. 2011; 44: 118-123
        • El Ghoch M.
        • Calugi S.
        • Chignola E.
        • Bazzani P.V.
        • Dalle Grave R.
        Body mass index, body fat and risk factor of relapse in anorexia nervosa.
        Eur J Clin Nutr. 2016; 70: 194-198
        • Brooks G.A.
        • Fahey T.D.
        • Baldwin K.M.
        Exercise physiology: human bioenergetics and its applications.
        McGraw-Hill Higher Education, New York2005