The Reference Values of Body Composition for Adult Females Who are Classified as Normal Weight, Overweight or Obese Accoding to Body Mass Index
.........................................................................................................................
Aliye Ozenoglu, PhD
Department of Psychiatry, Cerrahpasa Medical Faculty, University of Istanbul, Istanbul, Turkey
Serdal Ugurlu, MD
Department of Medicine, Medical Faculty, University of Cumhuriyet, Sivas, Turkey
Gunay Can, MD
Department of Public Health, Cerrahpasa Medical Faculty, University of Istanbul, Istanbul, Turkey
Hüsrev Hatemi, MD
Professor of Medicine, Division of Endocrinology-Metabolism and Diabetes and Department of Medicine, Cerrahpasa Medical Faculty, University of Istanbul, Istanbul, Turkey

Corresspondence to:
Aliye Özenoglu, PhD
Cerrahpasa Medical Faculty, Psychiatry Department
Istanbul University,
34303 Cerrahpasa, Istanbul, TURKEY
Tel:+90 (212) 414 31 30
Fax:+90 (212) 414 31 30

Obesity. defined as the increase of fat tissue in the body, is an important public health problem which leads to increased morbidity and mortality of some diseases, and has negative influences on the duration and quality of life. As well as physiological aspects, it also has social and psychological aspects^(1-5). In parallel with technological advances, the ease in obtaining, buying and consuming various kinds of food has increased and physical inactivity, consumption of more cigarettes and alcohol, stress and weakness of the mechanisms for coping with stress are other environmental factors which make it easier for obesity to develop^(1,4-7). Presently, obesity,the prevalence of which is increasing in all age groups in many countries, is regarded as a disease which must be treated^(4,5). In order to define someone as obese, a person's body weight, body composition and fat distribution, should be correctly determined. It should not be forgotten that when evaluated according to body weight there are subjects who are accepted to be obese as they have a lot of muscle mass; but there are also others who have normal body weight but accepted to be obese based on body fat composition and other metabolic parameters^(4,8). So, it is quite important to diagnose obesity correctly in order to prevent the organic, metabolic, and psychosocial problems it might cause. On the other hand, the measurement of body composition is highly helpful in understanding whether the individual has need for any nutritional support.

Until now, many methods have been developed to detect body composition^(2,3,9). Although direct methods to evaluate body composition in humans do exist, they are not easily applicable in routine clinical practice. So, the applicable methods are indirect. Among these Bioelectrical Impedance Analysis (BIA), the reliability of which was confirmed in many studies, is a practical method which makes use of the conductivity of the body, namely the tissues^(9-15). The body composition might vary according to age, sex, ethnic background, nutritional status, exercise, climate, the presence of some illnesses and the administration of some drugs; so there need to be standards for different conditions. However, reference values available for this purpose are quite few^{(10,13,16-18)}. In literature, although there are reference values of body composition for different age groups obtained by BIA; there are no reference values of body composition in healthy adults matched for age, sex, and body mass index.

In this study, we aimed to determine reference values of body composition in adult females who were classified as normal, overweight, obese, and morbidly obese according to body mass index (BMI); and who had no endocrine-metabolic disturbances except exogenous obesity and also no history of any drug usage.

A total of 327 female subjects (all >18 years of age) were admitted to our department between 1999 and 2003 with various complaints evaluated retrospectively. Patients who had no endocrine-metabolic disease or denied usage of drugs affecting metabolism, were included in this study. Mean age of subjects was 39.18±12.02 years. As a part of the nutritional status assesments, patients' height, weights, waist and hip circumferences were determined and the body compositions measurements were performed by BIA. BMI was calculated by adjusting the known formula as weight (kg)/ height2 (m2). Subjects with BMI=18.5-24.9 kg/m² were accepted to be normal, those with BMI=25.0-29.9 kg/m2 overweight, those with BMI=30.0-39.9 kg/m² to be obese, and those with BMI³40 kg/m² to be morbidly obese^(1-5,19,20). Waist/hip ratio (WHR) was obtained by dividing the waist circumference into the hip circumference.

BIA depends on the principle that fat is a bad conductor to the applied current; however, lean body mass is a good conductor depending on its content of water and electrolytes. In order to determine the body impedance, two tetrapolar electrodes are placed on the lateral surfaces of both the right hand and the right foot while the subject is lying down in supine position. There is a low current between the electrodes. Depending on the amplitude of the current, whole body water; or, in low frequencies only extracellular water might be determined^(2,3,9,21). As the method mainly measures body water, the hydration of the subject should be normal. Errors in measurement might occur due to the presence of metals on the bed or the subject, daily variations in body weight and composition due to food intake and exercise, intake of drugs, menstrual cycle in females, and placing the electrodes incorrectly^(2,9). In this study, optimum care has been taken to minimize errors in measurement and all measurements were performed by the same observer in morning hours.

Temperature of the test room was maintained between 18-20°C. The resistance and reactance values obtained by BIA, together with the age, sex, height and weight of the subject were entered into a computer programme to calculate body water, fat and lean body mass and basal metabolism. Percentage values for body composition, and basal metabolism as kcal. which was obtained automatically via a computer programme, were mainly used in this study. Body fat and lean weight as grams and water as liters were not included in this study. As the body composition of females and males differ, only female subjects were taken into the study. ANOVA test was used in comparing the anthropometric values and body compositions in percentages between groups who were formed according to BMI.

The mean BMI, waist and hip circumferences, and WHR of females according to BMI are given in Table 1.

As the degree of obesity increases, BMI, waist and hip circumferences also increased significantly. WHR showed a parallel increase with BMI significantly (p<0.0001).
The percentages of body water, fat mass (FM) and lean body mass (LBM), and basal metabolism (BM) obtained by BIA for females are shown in Table 2.

In parallel with the increase in BMI, the percentage of body fat and BM increases; however, the percentages of body water and lean body mass decrease significantly (p<0.0001).
The correlations of the various measurements made in the study are shown in Table 3.

A highly significant positive correlation was found between BMI and waist and hip circumferences and between the percentage of body fat and BM; whereas a negative correlation was present between BMI and lean body mass, and the percentage of body water. In addition, percentage of body fat had significant correlation with waist and hip circumferences and WHR.

In order to know the nutritional status, it is important to determine the body composition. Body composition might vary according to the stages of growth and development, age, sex, ethnicity, genetic and environmental factors, nutritional and exercise habits, various diseases and different therapies^{(2,9,10,20-22)}. Today, body composition is evaluated at the anatomic, molecular, cellular, tissue-system and the whole body level^{(1,3,4,20,24-29)}. As direct measurements in humans cannot be performed under in vivo conditions, body composition might be determined by indirect methods. Direct measurements might be performed only on cadavers. Methods of indirect measurement which can be performed in humans are anthropometric measurements, isotope or chemical dilution method, determination of body density, conductivity measurements, imaging methods, whole body neutron activation analysis and dual-energy X-ray absorbiometry (DEXA)^{(1-3,9,20-24)}. Today, BIA which is a method dependant on conductivity is one of the most preferred methods because it is easily performed, portable, has no danger, is more economic when compared to other methods and the results are reliable^{(2,3,9,20,21)}.

To predict the nutritional status of a patient and plan their nutritional treatment, it is helpfull to follow up the changes in body composition. Determination of the body fat composition is an important criterion especially in understanding the risk for obesity and related diseases. In spite of having normal body weight, there might be subjects with insulin resistance, hypertension, dyslipidemia, and above normal body fat mass (normal weight, metabolically obese); on the contrary it must not be forgotten that some others might be obese when body weight is regarded, but their metabolic parameters might be normal in contrast with what is expected (obese, metabolically normal)^(4,8).

On the other hand, correct determination of the body compositions of patients who need nutritional support for various reasons is of utmost importance. Follow up of the changes in body composition are useful for determination of the nutritional status and planning therapy and to evaluate the efficiency of administered therapies in disorders of hormones affecting metabolism, or diseases which make it obligatory to use drugs affecting the metabolism; inborn disorders of metabolism in which special diets have vital importance; inflammatory bowel diseases (IBD) which cause deterioration of the patients' nutritional status; chronic renal failure; neurologic disorders; diseases like cancer the presence of severe disease states or traumas. Body composition is helpful in evaluating the efficacy of diet therapy especially in muscle-type glycogenoses and other muscle diseases in which an increase in the amount of protein in the diet is needed for preservation of the body muscle masses^(30,31,32).

Correct determination of the nutritional status in patients with renal failure is important as it is closely related with prognosis. In these patients, total body water (TBW), hypertension and cardiac morbidity are accepted to be independant prognostic markers⁽¹²⁾. In one study which was planned to assess TBW and nutritional status in end-stage renal failure patients, it was found out that TBW varied greatly depending on the method of calculation⁽¹²⁾.

IBD patients are frequently faced with malnutrition as a result of malabsorption and decreased food intake due to gastrointestinal symptoms. Many studies conducted in IBD patients in order to assess the nutritional status revealed that body fat mass was significantly decreased when compared to controls, however lipid oxidation rate was increased^(33,34,35). It was thought that increased lipid oxidation and insufficient energy intake could explain the decrement in fat mass and it was put forward that enteral diets relatively rich in fat might be useful to sustain the nutritional status of these patients.

Determination of body composition is important for some occupations. It is desirable for athletes, artists, ballet-dancers, and people occupied in military and legal jobs to keep a certain body fat standard. Malina et al.⁽³⁶⁾ studied the percentage and the distribution of fat of the athletes who took place in the Olympic Games in Montreal in 1976. They reported that percentage of body fat was affected mostly from sports and exercise; however the distribution of fat was dependant on biologic factors. Studies revealed that the distribution of body fat differs between whites and blacks; and that blacks store more fat on the upper part of the trunk when compared to whites⁽¹⁰⁾. In whites, the ratio of extremity skin-fold thickness to trunk skin-fold thickness was found to be higher than in blacks⁽³⁶⁾. Because of these reasons, when determining body composition age, sex, ethnic background, concurrent diseases and therapies, nutritional habits, activity level, socioeconomic and environmental factors should be carefully investigated.

In order to be able to interpret body composition measurements, reference values formed under various conditions are needed. However, the number of studies about reference values formed under various conditions is limited. In this study, we aimed to determine mean body composition values by grouping adult females according to BMI; and, we found out that body composition changed significantly in accordance with BMI (Tables 2). This finding shows that while interpretting the body composition of one individual, BMI must be taken into consideration.

The first study in which fat and lean-body masses according to sex and age were determined in healthy subjects was performed by Pichard et al.⁽¹³⁾. In this study, they wanted to determine fat and lean-body masses in different decades by BIA, to detect changes in these values with advancing age, and to develop percentile values for these parameters in a population composed of healthy whites (1838 males and 1555 females) between 15-64 years of age. It was demonstrated that mean fat mass and percentage of body fat in males increased progressively; whereas, in females this increase occurred after 45 years of age. It was reported that the data in that study could be used as a reference to evaluate whether body compositions of healthy and sick subjects at certain ages were normal.

In another study⁽¹⁸⁾, 25th-75th percentiles of fat and lean-body masses were formed in 4566 healthy females and males between 20-79 years of age. In this study, subjects were grouped into 20-39, 40-59, 60-79 age ranges and measurements were performed with BIA. The authors concluded that lean-body mass decreased and fat mass increased with age; and that these reference percentile values made it possible to interpret the results of BIA and to
determine subjects with abnormal muscle and fat mass.

There are different methods to assess the energy needed for basal metabolism^(3,37,38). One of these is indirect calorimetry which depends on the principle of measuring oxygen used in biologic oxidations and it is possible to obtain correct and reliable results with it; however, its use is not widespread because of the difficulties in practice. In daily life, to calculate the energy requirements, methods which can be applied more practically and which depend indirectly on measurement and calculation are used. In our study, it was found that energy for basal metabolism detected by BIA increased significantly in parallel with the increase in BMI (Tables 2).

Frequently used equations for determination of energy needed for basal metabolism are Harris Benedict, Schofield and WHO equations^(3,17,37,38). Some of the studies in which energy for basal metabolism determined by measurement and calculation were compared found differences in values obtained by two different methods; but, some other studies could not find any difference^(17,39,40).

Barot et al.⁽³⁹⁾ compared the measured resting energy expenditure (REE) in 12 IBD patients -9 of whom had Crohn's disease- with a calculated formula according to Harris Benedict equation. They found that there was no significant difference between these two values in patients having >90% of ideal weight; but, patients <90% of ideal weight were hypermetabolic when compared to controls. Stokes and Hill⁽⁴⁰⁾ reported that resting metabolic rate detected by measurement in 13 active Crohn's patients was 14% higher than that found by calculation.

Recently, we compared BM measured by BIA and calculated by adjusting Harris Benedict equation in female patients with active IBD and women with normal weights according to BMI⁽⁴¹⁾, and saw that the value obtained by measurement was significantly higher than that of calculated (respectively, 1325.75±122.92 kcal vs. 1272.82±102.67 kcal, p=0.02). In this study in which healthy women matched for age and BMI served as controls, BM in the control group was significantly higher than the calculated value (respectively, 1451.88±83.5 kcal vs. 1323.27±74.65 kcal, p<0.0001).

In another study⁽⁴²⁾, we compared BM in normal weight, healthy adult females and males by two different methods, and found that BM measured by BIA was significantly higher than this calculated by Harris Benedict formula in both sexes (p<0.0001). In a different study⁽⁴³⁾, we compared BM found by two different methods in females who were of normal weight, overweight or obese according to BMI, had no endocrine-metabolic disorder and no history of any drug usage; the results revealed that BM measured by BIA was significantly higher than that calculated with Harris Benedict equation (p<0.0001). Depending on the results that we obtained in these studies, we concluded that -not only in healthy subjects; but, also in the presence of diseases affecting metabolism- methods to determine body composition and basal metabolism, which based on measurement might be more reliable in predicting energy needs and planning nutritional therapy.

Obesity which is a major health problem in many developing and developed populations, is an independant risk factor for the development of coronary artery disease (CAD).Also, obesity increases the risk of CAD by its relation with insulin resistance, hypertenion and dyslipidemia^{(1,4,5,7,8,19,22)}. Determination of body fat is important in defining obesity which might impair quality of life and cause increased morbidity and mortality. Although techniques like computed tomography (CT), magnetic resonance (MR) make it possible to evaluate regional body fat distribution and abdominal fat depositions, their use for clinical and epidemiological purposes is limited by their high expenses. The determinants of body fat distribution, like WHR, waist diameter, sagittal diameter, are important alternatives in defining individuals with high risk^(1,4,5,8,20).

WHR is a simple, useful and sensitive index of body fat distribution. By using this anthropometric index Bray and Gray⁽⁴⁴⁾ put forward that in males a WHR>0.95 and in females a WHR>0.85 might be useful in the detection of high-risk subjects. Similarly, Pauliot et al.⁽⁴⁵⁾ reported that in females a WHR>0.85 and in males a WHR>1.0 might be related with changes in glucose-insulin homeostasis and lipid-lipoprotein metabolism. It was said that waist circumference determined with techniques like CT and MR was a better indicator of visceral fat than WHR (4, 8). It is accepted that independant of BMI and WHR, a waist circumference >88 cm in females and >102 cm in males increase the risk for complications of obesity and mortality^(4,9,22). Under the light of this knowledge, females in our study who were in obese groups (BMI>30.0 kg/m2) according to BMI are accepted to have risk for obesity-related diseases (Table 1). As waist circumference and BMI have a very strong relation with percentage of body fat and WHR has a weak relation, this confirms that waist circumference is a more sensitive indicator (Table 3).

It is known that fat tissue forms 20-25% of body weight in an adult female and 12-15% of that in a male (1,5,19,20). Obesity is accepted to be present when body fat exceeds 30% in females and 20% in males. The results of this study are compatible with the available literature both for body composition and body circumference measurements.

Correct determination of the nutritional status of an individual is important for planning therapy under various conditions and for evaluating efficiency of administered therapies. Measurements which are dependant on only height and weight are usually misleading and insufficient in evaluation of the nutritional status of one individual. However, as it is unfeasible to determine body composition in the clinical practice, there is need to develop the standards of methods of indirect measurement.Body composition might show variations according to age, sex, ethnic features, nutritional status, genetic and environmental factors, level of exercise, and even the presence of various diseases and administered therapies. For this reason, whenever the body composition of one individual is to be interpretted it will be useful to use reference values formed under similar conditions. In addition; whenever there is no possibility of determining the body composition of the individual, the usage of references in order to predict his nutritional status approximately will lead to more reliable results. To achieve this, it is obvious that standardized indirect measurement must be developed for both healthy individuals and states of disease. We assume that the results of this study which was the first to evaluate body composition by BIA in groups matched for age, sex, and BMI- might be used as reference values. We also think that diet and exercise programmes should be used for the treatment of obesity which is defined as excess fat tissue in the body; and the efficacies of diet and medications should be followed up with measurements of body composition.

1. Tüzün M. The general features of obesity. In: Yilmaz C (ed). Obesity and its treatment, Izmir, 1999; p. 11-29 (Book in Turkish).
   
2. Pekcan G. The definition and detection of obesity.In: Proceedings of the Third National Congress on Nutrition and Dietetics, Ankara, 2000; 93-104 (Abstract in Turkish).
   
3. Arslan P. (ed). Obesity, its influence on various diseases and scientific applications in dietary treatment. 4 th publication of the Turkish Dietetic Associations, Ankara, 1992 (Book in Turkish).
   
4. Korugan Ü, Damci T, Özbey N, Özer E. Clinical Obesity. Publication of the Obesity Working Group. Istanbul, 2000; 2-14 (Book in Turkish).
   
5. Yilmaz C (ed). Obesity. Nobel Medical Publications, Izmir, 1995 (Book in Turkish).
   
6. Bjorntorp P. Visceral obesity: a "civilization syndrome". Obes Res 1993; 1: 206-22.
   
7. Siervogel RM, Wisemandle W, Maynard LM, Guo SS, Chumlea WC, Towne B. Lifetime overweight status in relation to serial changes in body composition and risk factors for cardiovascular disease: The feels longitudinal study. Obes Res 2000; 8 : 422-30.
   
8. Brochu M, Poehlman ET, Ades PA. Obesity, body fat distribution and coronary artery disease. J Cardiopulm Rehabil 2000; 20: 96-108.
   
9. Ellis KJ. Human body composition: In vivo methods. Physiological Reviews 2000; 80: 649-80.
   
10. Wagner DR, Heyward VH. Measures of body composition in blacks and whites: a comparative review. Am J Clin Nutr 2000; 71: 1392-402.
    
11. Rosenbaum K, Wang J, Pierson RN, Kotler DP. Time dependent variation in weight and body composition in healthy adults. JPEN 2000; 24 (2): 52-55.
    
12. Cooper BA, Aslani A, Ryan M, Zhu FY, Ibels LS, Allen BJ, Pollock CA. Comparing different methods of assessing body composition in end stage renal failure. Kidney
    Int 2000; 58 : 408-16.
    
13. Pichard C, Kyle UG, Bracco D, Slosman DO, Morabia A, Schutz Y. Reference values of fat-free and fat masses by bioelectrical impedance analysis in 3393 healthy subjects. Nutrition 2000; 16 : 245-54.
    
14. Lukaski HC. Assessing regional mass with segmental measurements of bioelectrical impedance in obese women during weight loss. Ann NY Acad Sci 2000; 904: 154-8.
    
15. Kyle UG, Genton L, Karsegard L, Slosman DO, Pichard C. Validation of a bioelectrical impedance analysis (BIA) equation for the swiss population. Clin Nutr 2000; 19 (Suppl 1): S 6.
    
16. Atkin LM, Davies PS. Diet composition and body composition in preschool children. Am J Clin Nutr 2000; 72 : 15-21.
    
17. Avitzur Y, Singer P, Dagan O, Dinari G, Shamir R. Measured vs calculated resting energy expenditure before and after open heart surgery in children with congenital heart disease. Clin Nutr 2000; 19 (Suppl 1): S 2.
    
18. Kyle UG, Genton L, Karsegard L, Slosman DO, Pichard C. 25-75 th percentiles for fat-free and fat masses in 4566 healthy adults aged 20-79 years determined by BIA. Clin Nutr 2000; 19 (Suppl 1): S 6.
    
19. PI-Sunyer FX. Obesity. In: Shils ME, Olson JA, Shike M, Ross AC (eds). Modern Nutrition in Health and Disease. 9th ed. Philadelphia, Lippincott Williams &Wilkins.1999;1395-414.
    
20. DeHoog S. The assessment of nutritional status. In: Mahan LK, Escott-Stump S (eds). Krause's Food, Nutrition and Diet Therapy. 9th ed. Philadelphia. W.B. Saunders Company, 1996; 369-74.
    
21. Jebb SA. Measuring body composition: from the laboratory to the clinic. In: Kopelman PG, Stock M. eds. Clinical Obesity. Oxford. Blackwell Science, 1998; 18-50.
    
22. Weight management and eating disorders. In: Mahan LK, Escott-Stump S (eds). Krause's Food, Nutrition and Diet Therapy. 9th ed. Philadelphia. W.B. Saunders Company, 1996; 451-63.
    
23. Brodie D, Moscrip V, Hutcheon R. Body composition measurements . A review of hydrodensitometry, antropometry and impedance methods. Nutrition 1998; 14 : 296.
    
24. Wang ZM, Pierson RN, Hemsfield SB. The five-level model: a new approach to organising body composition research. Am J Clin Nutr 1992; 56: 19.
    
25. Fuller N, Jebb SA, Laskey M, Cowerd M, Elia M. Four compartment model for the assesment of body composition in humans: Comparision with alternative methods and evaluation of the dansity and hydration of fat free mass. Clinical Science 1992; 82: 687.
    
26. Heymsfield SB, Wang ZM, Baumgartner RN, Ross R. Human body composition. Advances in models and methods. Ann Rev Nutr 1997; 17: 527.
    
27. Heymsfield SB, Tighe A, Wang ZM. Nutritional assesment by anthropometric and biochemical methods. In: Shils ME, Olson JA, Shike M. eds. Modern Nutrition in Health and Disease. 8 th ed. Philadelphia. Lea and Febiger, 1994; 812.
    
28. WHO. Physical status: The use and interpretation of anthropometry. WHO Tech Rep Ser WHO, Genova. 1995; 854.
    
29. Nunez C, Kovera AJ, Pietrobelli A, Heshka S, Horlick M, Kehayias JJ, et al. Body composition in children and adults by air displacement plethysmography. Eur J Clin Nutr 1999; 53: 382.
    
30. Fernandes J, Chen YT. Glycogen storage disease. In: Fernandes J, Saudubray JM, Berghe GV (eds). Inborn Metabolic Disease. 2nd ed. Berlin. Springer-Verlag, 1995; 71-84
    
31. Dixon M. Disorders of carbohydrate metabolism. In: Shaw V, Lawson M (eds). Clinical Paediatric Dietetics. London. Blackwell Science Ltd. 1995; 210-16.
    
32. Forbes GB. Body composition: Influence of nutrition, physical activity, growth and aging. In: Shils ME, Olson JA, Shike M, Ross AC (eds). Modern Nutrition in Health and
    Disease. 9th ed. Philadelphia. Lippincott Williams & Wilkins, 1999;789-805.
    
33. Capristo E, Addolorato G, Mingrone G, Greco AV, Gasbarrini G. Effect of disease localization on the anthropometric and metabolic features of Crohn's Disease. Am J Gastroenterol 1998; 93: 2411-19.
    
34. Mingrone G, Greco AV, Benedetti G, Capristo E, Semeraro R, Zoli G, et al. Increased resting lipid oxidation in Crohn's Disease. Digestive Disease and Science 1996; 41 (1): 72-76.
    
35. Mingrone G, Capristo E, Greco AV, Benedetti G, De Gaetano A, Tataranni PA, et al. Elevated diet-induced thermogenesis and lipid oxidation rate in Crohn's Disease. Am J Clin Nutr 1999; 69 (2): 325-30.
    
36. Malina RM, Mueller WH, Bouchard C, Shoup RF, Lariviere G. Fatness and fat patterning among athletes at the Montreal Olympic Games. Med Sci Sports Exerc 1982; 14: 445-52.
    
37. Arslan P. The calculation of energy and nutritional element needs in enteral-parenteral nutrition. In: Basoglu S, Karaagaoglu N, Erbas N, Ünlü A (eds). Enteral-Parenteral Nutrition. Turkish Dietetics Associations, publication number: 8, Ankara, 1995;39-49 (Book in Turkish).
    
38. The calculation of energy needs. In: Özkarabulut A, Özenoglu A, Turanli F, Boneval H. (eds.), Formulas and enteral products in nutrition. Turkish Dietetics Associations, publication number: 1. Istanbul, 2000; 10-11 (Book in Turkish).
    
39. Barot LR, Rombeau JL, Feurer ID, Mullen JL. Caloric requirements in patients with inflammatory bowel disease. Ann Surg 1982; 195: 214-19.
    
40. Stokes MA, Hill GL. Total energy expenditure in patients with Crohn's Disease: Measurement by the combined body scan technique. JPEN 1993; 17: 3-7.
    
41. Özenoglu A, Pamuk GE, Pamuk ÖN, Hatemi HH. The comparison of basal metabolism found by two different methods in normal-weight females with inflammatory bowel disease with healthy controls. Goztepe Medical Journal 2004; 19: 15-18. (Journal in Turkish).
    
42. -Özenoglu A, Pamuk ÖN, Pamuk GE, et al. The comparison of body compositions and basal metabolism found by two different methods between normal-weight healthy females and males. J Endocr Invest. 2001; 10 (2): 21-4. (Journal in Turkish).
    
43. Özenoglu A, Pamuk GE, Pamuk ÖN, et al. The comparison of basal metabolism found by two different methods amongst females classified as normal, overweight or obese according to body mass index. In: Proceedings of the Third Congress of Clinical Enteral-Parenteral Nutrition. Istanbul, 2000; 182 (Abstract in Turkish).
    
44. Bray GA, Gray DS. Treatment of obesity. An overview. Diabetes Metab Rev 1998; 4: 653-79.
    
45. Pouliot MC, Despres JP, Lemieux S, et al. Waist circumference and abdominal sagittal diameter: Best simple anthropometric indexes of abdominal visceral adipose tissue accumulation and related cardiovascular risk in men and women. Am J Cardiol 1994; 73; 460-68.