Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

From BMI to functional body composition

BMI (body mass index, weight relative to height, expressed as kg/m2. This index is minimally correlated with height and, thus, mainly relates on weight.) is formally used to categorize obesity and underweight. BMI cut-offs (that is, >30 kg/m2 for obesity and <18.5 kg/m2 for underweight, respectively) are based on epidemiological data, that is, the statistical associations between decreasing and increasing BMI and excess mortality.1 Although BMI is very popular and has value in daily clinical practice, it is not a uniform measure. BMI is a continously distributed variable, thus, any cut-off is arbitrary. Although BMI is a calculated rather than a biological property, it is frequently and uncritically used as a reference for genes, metabolism, nutrition and mechanisms of disease.2

The present evidence suggests that the impact of the BMI is variable and limited. Although there is an association between BMI and percentage fat mass in overweight subjects, BMI cannot be considered as a good proxy of fat mass.2 In adults with a BMI of 30 kg/m2, adiposity can range between 23 and 41% in males, and 30 and 51% in females, respectively.2 By contrast, there is no association between BMI and total body protein content in patients with anorexia nervosa questioning its value characterizing underweight.3 Regarding population studies on the genetic basis of body weight, genes explained up to 10% of the variance in BMI only.2, 4 In addition, BMI contributed to less than 35% of variance in insulin resistance in normal and overweight subjects.5 All these data point out to possible shortcomings of the BMI, which cannot reflect either sarcopenia (that is, the loss of muscle mass) at underweight or adequately characterize adiposity, because of its inability to discriminate between (i) an excess in body fat and increased lean mass and (ii) different types of body fat distribution. Alltogether, BMI is a weak mean to characterize the nutritional and the clinical phenotype. Thus, the clinical practice to use BMI should not be transferred into research.

BMI has limitations for comparing individuals as well as populations because of differences in body composition. To overcome the shortcomings of the BMI and to get into more details of the phenotype, body composition analysis (BCA) is necessary. Phenotypic differences in fat distribution are related to phenotypic differences in metabolism, for example, insulin secretion and insulin resistance are related to liver fat, which again is related to visceral adipose tissue.5 Body composition also gives evidence that the BMI-based categories of over- and underweight cannot be treated as entities. For example, 30% of the obese population are considered as metabolically healthy. When compared with the ‘healthy obese’ the ‘metabolically unhealthy obese’ is characterized by increased liver fat.5, 6, 7 It is likely that individualizing the phenotype may improve research and presumably future treatment on nutritional and metabolic issues. Today, BCA is within the center of integrative physiology on understanding the body responses to internal and external factors at all biological levels, that is, applying concepts of cellular/molecular physiology, biochemistry and experimental approaches to understand the function at the level of whole body or its individual organs.

BCA subdivides the crude BMI phenotype into more sharply divided phenotypes. These are characterized by differences in fat mass, regional fat depots, muscle and organ masses or bone mineral content. There are up to 30–40 individual body components, which can be combined at different levels.8 About 70 years ago, BCA started with the classical Behnke’s two component model; this is still the most widely applied model dividing body weight into fat mass and fat-free mass (FFM, that is, the actively metabolizing body component.8, 9 Present models of body composition refer to five different levels, that is, ‘atomic’ (including the 11 major elements, H, O, N, C, Na, K, Cl, P, Ca, Mg, S), ‘molecular’ (including 6 components, lipid, water, protein, carbohydrates, bone minerals, soft tissue minerals), ‘cellular’ (that is, 3 or 4 components, cell mass, extracellular fluids, extracellular solids, where cell mass can be divided into fat and actively metabolizing body cell mass) and ‘tissue-organ levels’ (=4 major tissues, adipose tissue, skeletal muscle, visceral organs, bone with further organ-level components such as brain, liver, kidneys, heart, spleen) resulting in many body components or so-called multicomponent models.8 Finally, at the ‘whole body level’ the body can be divided into body regions, that is, brain, trunk, upper and lower limbs. All component models rely on certain assumptions (=constants), which are considered as stable or fixed (for example, a 73.2% water content of FFM or a body temperature of 36 or 37 °C.9 It is assumed that an individual component is homogenous in composition.

Using more advanced technologies and concepts, the composition as well as heterogeneity of the individual body components become evident. For example, at the ‘tissue-organ level’, adipose tissue is distributed within the body in numerous depots, that is, intraabdominal, subcutaneous abdominal and peripheral fat, fat tissue at the trunk or around blood vessels etc. Fat is also located within organs (for example, as liver fat) and different fat depots differ in function (for example, white vs brown adipose tissue, BAT). In addition, at the ‘organ level’, the systematic use of imaging techniques in BCA resulted in huge databases on organ and tissue masses, which again allowed to create a new ‘Reference Man’.10, 11, 12, 13 These data also give rise to the idea to combine individual body components across the different levels (for example, liver fat combines the ‘molecular’ and the ‘tissue-organ level’.8 Presently, multicomponent models have reached the highest level of body composition research. These models have a sound theoretical and methodological basis, but the data are merely descriptive. They do not take into account tissue-organ function and, thus, different metabolic properties of the individual body components.

Individual body components do not stand alone, they are interrelated, both, in mass (for example, in a weight loosing subject fat mass and FFM are lost at a fixed relation;14 and function (for example, muscle mass and adipose tissue are interrelated with respect to metabolism and inflammation) mainly brought about by hormonal regulation and organ-tissue cross-talks.15, 16, 17, 18, 19, 20 Gaining adipose tissue with overfeeding is an important source of signaling and proinflammatory molecules thereby communicating with skeletal muscle, the liver and vascular endothelial cells.16 Vice versa, with caloric restriction, the loss of fat mass and its secretory activity have been related to metabolic adaptation (that is, adaptive thermogenesis).21, 22 With physical activity, skeletal muscle produces and releases myokines exerting metabolic and anti-inflammatory effects on the muscle itself, the liver, white adipose tissue, BAT, cells of the immune system, and pancreatic islets with positive effects on energy expenditure and insulin-induced glucose disposal.20

These data led to the idea of functional body composition.23 Functional body composition integrates body components into regulatory systems, that is, it relates body composition to its corresponding in vivo function and metabolic processes. Suitable applications of BCA are (i) interpretation of body functions (for example, energy expenditure, insulin sensitivity) and their disturbances in the context of body components and vice versa and (ii) interpretation of the meaning of individual body components in the context of their functional consequences (for example, energy expenditure). As different body functions and metabolic processes are differently related to individual body components, functional body composition extends the traditional view, thus, crossing as well as combining the different body composition levels.

Figure 1 compares two traditional models (a two- or three-component model and a detailed ‘tissue-organ level’ model) with more advanced models related to different functional aspects of metabolism. In the case of energy expenditure, FFM is its major determinant, but functional body components relate to low and high metabolic rate organs and tissues within and outside FFM. When compared with a two-component model, white adipose tissue, extracellular mass and bone are considered as low metabolic rate organs. By contrast, muscle, visceral organs, brain and BAT have high metabolic rates. Thus, the functional body composition model mixes the classical body components, that is, individual components of FFM relate to high versus low metabolic rate components with BAT included into the list of the former group.

Figure 1
figure 1

Proposed framework of functional body composition. Individual body components are grouped according to conventional body composition models (left) as well as according to different body functions, that is, energy expenditure, glucose turnover, lipid and protein metabolism (right). AT, adipose tissue; BAT, brown adipose tissue; BCM, body cell mass; ECM, extracellular mass; Gut, gastrointestinal tract; TBW, total body water; VAT, visceral adipose tissue.

The concept of functional body composition differs between different body functions (Figure 1). When compared to energy expenditure, body components related to glucose turnover and insulin sensitivity are divided into insulin-sensitive and non-insulin-sensitive body components. The former comprise muscle, gut, heart and different fat depots, including BAT, whereas the latter group combines, brain, kidneys, bone and extracellular mass. Again the insulin-sensitive component of the body mixes up fat and FFM components of the descriptive two-component model. In the case of lipid metabolism, three functional body components can be characterized. First, a lipid storage (or non-oxidative lipid metabolism) component which includes the different adipose tissue depots (plus plasma volume as far as plasma triglycerides are concerned). Second, organs related to lipid transfer and oxidation (for example, muscle, white adipose tissue, liver and BAT). Third, organs, tissues and cells with no oxidative lipid metabolism (for example, brain). Finally, body components related to protein metabolism can be grouped according to low, high and no protein turnover rate. White adipose tissue and bone have a low protein turnover, whereas visceral organs and skeletal muscle are considered as high protein turnover organs.

Functional body composition extends the traditional models of body composition. The present ideas (Figure 1) confer to the ‘cellular’ and the ‘tissue-organ level’. It is possible to extend this view to the ‘atomic’ and ‘molecular level’ (for example, defining body components contributing to protein turnover according to their protein or nitrogen content). Alternatively, functionally defined body components can be grouped according to regulation, for example, taking into account cross-talks between the organs. Then, an ‘adipose tissue–skeletal muscle component’ can be related to inflammatory activity seen in obese patients. Accordingly, a ‘skeletal muscle–heart–liver component’ may give rise to a better understanding of metabolic adaptation to exercise. Finally, a ‘brain–BAT–skeletal muscle component’ may add to understand metabolic adaptations to cold exposure.

Defining an individual phenotype depends on a controlled experimental condition (for example, controlled under- or overfeeding). Most of our present knowledge is based on cross-sectional or long-term longitudinal data (for example, before and after weight loss in obese patients). These clinical protocols cannot be considered as strictly controlled and the phenotypes characterized so far reflect both, regulation (for example, in response to positive or negative energy balance) and adaptation of metabolism (for example, to over- or undernutrition). This methodological shortcoming may add to the high inter-individual variance of data and also to the frequently mentioned complexity of metabolic regulation (which may not only reflect true compexity but some confusion due to our limited approaches). Thus, for future studies on functional body composition there is a need to define (and then to follow) suitable and controlled protocols.24, 25

To summarize, methodological advances in BCA, and the worldwide proliferation of suitable devices (for example, MRI, DXA, Bod Pod and BIA) into scientific, clinical and consumer communities, provide good and challenging opportunities of interfacing body composition with studies on nutrition and metabolism, and, preferably, with clinical practice. About 70 years ago, body composition research had started with the hydrodensitometry method, based on a two-component model that has now been extended to various multicomponent models at different body composition levels. Application of these methods allowed a detailed description of body composition. Within the past 20 years, BCA has been widely applied to clinical research (for example, in studies on diabetes and obesity) to show the effects of diseases on body composition. It is only recently that we begin to understand the role of different body components and their interactions in the pathogenesis of that diseases. Functional body composition provides a conceptual framework to enter the next era of body composition research.

References

  1. Prospective Studies Collaboration. Body-mass index and cause-specific mortality in 900 000 adults: collaborative analyses of 57 prospective studies. Lancet 2009; 373: 1083–1096.

    Article  Google Scholar 

  2. Müller MJ, Bosy-Westphal A, Krawczak M . Genetic studies of common types of obesity: a critique of the current use of phenotypes. Obes Rev 2010; 11: 612–618.

    Article  Google Scholar 

  3. Haas VK, Kohn MR, Clarke SD, Allen JR, Madden S, Müller MJ et al. Body composition changes in female adolescents with anorexia nervosa. Am J Clin Nutr 2009; 89: 1005–1010.

    CAS  Article  Google Scholar 

  4. Walley AJ, Asher JE, Froguel P . The genetic contribution to non-syndromic human obesity. Nat Rev Genet 2009; 10: 431–442.

    CAS  Article  Google Scholar 

  5. Müller MJ, Lagerpusch M, Enderle J, Schautz B, Heller M, Bosy-Westphal A . Beyond the body mass index: tracking body composition in the pathogenesis of obesity and the metabolic syndrome. Obes Rev 2012; 13 (Suppl.2), 6–13.

    Article  Google Scholar 

  6. Stefan N, Kantartzis K, Machann J, Schick F, Thamer C, Rittig K et al. Identification and characterization of metabolically benign obesity in humans. Arch Intern Med 2008; 168: 1609–1616.

    Article  Google Scholar 

  7. Wildman RP, Muntner P, Reynolds K, McGinn AP, Rajpathak S, Wylie-Rosett J et al. The obese without cardiometabolic risk factor clustering and the normal weight with cardiometabolic risk factor clustering: prevalence and correlates of 2 phenotypes among the US population (NHANES 1999–2004). Arch Intern Med 2008; 168: 1617–1624.

    Article  Google Scholar 

  8. Shen W, St-Onge M-P, Wang Z, Heymsfield SB . Study of body composition: an overview. In Human Body Composition 2nd edn. Heymsfield SB, Lohman TG, Wang Z, Going SB Eds., Human Kinetics: Champaign, IL.. pp 3–14, 2005.

    Google Scholar 

  9. Wang Z, Shen W, Withers RT, Heymsfield SB . Multicomponent molecular-level models of body composition analysis. In: Human Body Composition 2nd edn. Heymsfield SB, Lohman TG, Wang Z, Going SB Eds., Human Kinetics: Champaign, IL.. pp 163–176, 2005.

    Google Scholar 

  10. Müller MJ, Bosy-Westphal A, Kutzner D, Heller M . Metabolically active components of fat-free mass and resting energy expenditure in humans: recent lessons from imaging technologies. Obes Rev 2002; 3: 113–122.

    Article  Google Scholar 

  11. Gallagher D, Elia M . Body composition, organ mass, and resting energy expenditure. In Human Body Composition 2nd edn. Heymsfield SB, Lohman TG, Wang Z, Going SB Eds., Human Kinetics: Champaign, IL.. pp 219–240, 2005.

    Google Scholar 

  12. Later W, Bosy-Westphal A, Kossel E, Glüer CC, Heller M, Müller MJ . Is the 1975 Reference Man still a suitable reference? Eur J Clin Nutr 2010; 64: 1035–1042.

    CAS  Article  Google Scholar 

  13. Bosy-Westphal A, Braun W, Schautz B, Müller MJ . Issues in characterizing resting energy expenditure in obesity and after weight loss. Front Physiol 2013; 4: 47.

    Article  Google Scholar 

  14. Heymsfield SB, Thomas D, Nguyen AM, Peng JZ, Martin C, Shen W et al. Voluntary weight loss: systematic review of early phase body composition changes. Obes Rev 2011; 12: e348–e361.

    CAS  Article  Google Scholar 

  15. Argilés JM, López-Soriano J, Almendro V, Busquets S, López-Soriano FJ . Cross-talk between skeletal muscle and adipose tissue: a link with obesity? Medical Res Rev 2005; 25: 49–65.

    Article  Google Scholar 

  16. Lee D-E, Kehlenbrink S, Lee H, Hawkins M, Yudkin JS . Getting the message across: mechanisms of physiological cross talk by adipose tissue. Am J Physiol Endocrin Metab 2009; 296: E1210–E1229.

    CAS  Article  Google Scholar 

  17. Havekes B, Sauerwein HP . Adipocyte-myocyte crosstalk in skeletal muscle insulin resistance; is there a role for thyroid hormone? Curr Opin Clin Nutr Metabol Care 2010; 13: 641–646.

    CAS  Article  Google Scholar 

  18. Trayhurn P, Drevon CA, Eckel J . Secreted proteins from adipose tissue and skeletal muscle—adipokines, myokines and adipose/muscle cross-talk. Arch Physiol Biochem 2011; 117: 47–56.

    CAS  Article  Google Scholar 

  19. Taube A, Schlich R, Sell H, Eckardt K, Eckel J . Inflammation and metabolic dysfunction: links to cardiovascular diseases. Am J Physiol Heart Circ Physiol 2012; 302: H2148–H2165.

    CAS  Article  Google Scholar 

  20. Pedersen B, Febbraio MA . Muscles, exercise and obesity: skeletal muscle as a secretory organ. Nat Rev Endocrinol 2012; 8: 457–465.

    CAS  Article  Google Scholar 

  21. Dulloo AG, Jacquet J, Montani J-P, Schutz Y . Adaptive thermogenesis in human body weight regulation: more a concept than a measurable entity? Obes Rev 2012; 13 (Suppl.2), 105–121.

    Article  Google Scholar 

  22. Müller MJ, Bosy-Westphal A . Adaptive thermogenesis with weight loss in humans. Obesity (Silver Spring) 2013; 21: 218–228.

    Article  Google Scholar 

  23. Müller MJ, Bosy-Westphal A, Later W, Haas V, Heller M . Functional body composition: insights into the regulation of energy metabolism and some clinical applications. Eur J Clin Nutr 2009; 63: 1045–1056.

    Article  Google Scholar 

  24. Lagerpusch M, Bosy-Westphal A, Kehden B, Peters A, Müller MJ . Effects of brief perturbations in energy balance on indices of glucose homeostasis in healthy lean men. Int J Obes Relat Metab Disord 2012; 36: 1094–1101.

    CAS  Article  Google Scholar 

  25. Lagerpusch M, Enderle J, Later W, Eggeling B, Pape D, Müller MJ et al. Impact of glycaemic index and dietary fibre on insulin sensitivity during the refeeding phase of a weight cycle in young healthy men. Br J Nutr 2013; 109: 1606–1616.

    CAS  Article  Google Scholar 

Download references

Acknowledgements

Body composition research performed at my institution is funded by a grant from the German Ministry of Education and Research (BMBF 0315681), BMBF Competent Network of Obesity (CNO) and the German Research Foundation (DFG Bo 3296/1-1 and DFG Mü 714/ 8-3).

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to M J Müller.

Ethics declarations

Competing interests

The author is a consultant of seca Germany.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Müller, M. From BMI to functional body composition. Eur J Clin Nutr 67, 1119–1121 (2013). https://doi.org/10.1038/ejcn.2013.174

Download citation

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/ejcn.2013.174

Further reading

Search

Quick links