The Meaning and Measurement of Lean Body Mass Ronenn Roubenoff, M.D., M.H.S. and Joseph J. Kehayias, Ph.D. Methods of measuring lean body mass have evolved greatly in the past decade. Because all practical methods of measurement are indirect (i.e., not involving carcass analysis), appropriate models of body composition must be applied. Today’s improved precision is due both to better instrumentation and to improved models based on partition of the body into its constituent compartments. With this improved methodology has come the recognition that loss of lean body mass is associated with survival in starvation, acute illness, and aging. In addition, correlations of lean body mass with immune competence and functional status are beginning to emerge. It remains to be proven that changes in lean body mass in aging and disease independently determine outcome. Interventions that preserve or normalize lean body mass could be important for preservation of vigorous functional status.
Although the composition of the human body has been under study for at least a century, investigation of the interaction among lean body mass and functional status, immune competence, and survival is only now beginning to emerge. Body composition concepts have recently been reviewed in detail by Heymsfield;’ this article will focus on the connection between lean body mass (LBM) and function. Because the method of measurement of any body compartment largely defines that compartment, we first touch on current methods of LBM measurements and their limitations. Approaches to Composition of the Organism The body can be divided into a variable number of different compartments, depending in large part on who is doing the dividing and what instrument(s) they are Dr. Roubenoff is Research Associate in and Dr. Kehayias is Director of the Body Composition Laboratory, USDA Human Nutrition Research Center on Aging at Tufts University, 71 1 Washington Street, Boston, MA, 021 11. Supported by NIH Grant DK08443, by a Pew National Nutrition Program Clinical Fellowship. and by the USDA.
using to measure the various parts (Figure 1 and Table 1). The most intuitive division is fat mass (FM) from fat-free mass (FFM), a dichotomy based on the chemical analysis of human cadavers by dissolving away the lipid in organic ~ o l v e n t . FFM ~ . ~ is slightly different from LBM because the fat in cell membranes is excluded from FFM but is retained in LBM due to the anatomical proximity of this small amount of fat to lean tissue^.^ LBM, in turn, is made of body cell mass (BCM) and the nonfatty intercellular connective tissue, such as bone (excluding fatty marrow), tendons, ligaments, basement membranes, etc. It is within the BCM that over 99O/0 of the body’s metabolic processes take place.4 At the same time, another compartment that partially overlaps LBM, FFM, and BCM is total body water (TBW). TBW clearly excludes FM. However, it is not identical to any of the nonfat compartments discussed earlier because it includes both intracellular and extracellular water, but not the extracellular connective tissues or the intracellular nonwater ~ t r u c t u r e s . TBW ~’~ can be measured with deuterium, tritium, or antipyrene dilution, methods that have rbeen available for several decades. In addition, TBW can be divided into intracellular NUTRl’TtON RNEWSIVOL 49. NO (VJUNE 1991 163
100%
76%
60%
I
Ieolb
I/
26%
I1 /6AI
I
0%
Male
Female
=
PIaoma ECF 0Wator Fat-Fro. Solldo Fat Figure 1. Division of the body into compartments for normal males and females. Note that TBW = water + ECF + plasma = 60% of the body in men and 50% in women. LBM = TBW + fat-free solids = 82% of the body in men and 68% in women. After Moore (4).
and extracellular water using various tracers.'TBW also overlaps to some extent with muscle, an important compartment that is very difficult to measure because it contains water, nitrogen, and potassium. Yet muscle may be the most dynamic and rapidly influenced part of the LBM/FFM/BCM triad.* The previous two-compartment approach to body composition is unsatisfactory for several reason^.^ Most importantly, by lumping all nonfat tissues into one compartment, the distinct physiologic functions of the various tissues (i.e., bone vs. muscle, water vs. cell mass, etc.) are obscured. From a practical standpoint, the two-compartment model is not sufficiently precise to easily allow longitudinal measures of change in body composition under various dietary and pathologic conditions. Therefore, over the past decade, a series of multicompartment models have been developed." These models, taking advantage of new techniques, measure several parts of the nonfat tissue independently in order
164 NUTRITION RffIEWSNOL 49. NO WJUNE 1991
both to minimize measurement error and to obtain data on important components of body composition. The subdivision of the lean and fat compartments along elemental rather than compartmental lines is illustrated in Table 2. Elements that contribute exclusively to the nonfat tissue include nitrogen (present only in protein), calcium (bone), and potassium (BCM and LBM), as well as TBW. In contrast, carbon, hydrogen, and oxygen are present in both fat and lean tissue. If the amount of lean tissue can be estimated from one or more of its elemental contributors (i.e., potassium), then the amount of fat can be estimated directly from the amount of carbon in the body after subtracting the amount of carbon in lean tissue. This method affords much better precision than do any of the methods that only measure the lean compartment and total weight, and then subtract to estimate fat, leading to a propagated error that can typically be as large as 20% of the fat mass''*'2 (see subsequent discussion).
TABLE I Two-Compartment Models of Body Composition and Methods Used to Measure Them' Compartment Fat mass (FM)
Measurement Methods Cadaver-lipid Densitometry
extraction
Anthropometry Fat-free mass (FFM)
Cadaverdissection Bioelectrical impedance
TOBEC Lean body mass (LBM)
Anthropometry
Total body water (TBW)
Dilution (D,O, 'H,O)
Comment Not routinely available. Assume known density of fat and lean; need cooperative subjects; commonly available. Interobserver error large; can be done nearly anywhere, Edema limits use in some clinical situations. Not routinely available. Precise, but may not be accurate. Very sensitive to hydration status. Depends on population-specific equations. Portable, easy to use. Same as above, but more accurate; equipment more expensive, less portable. Interobserver error large; can be done nearly anywhere. Edema limits use in some clinical situations. Difficult to measure D 2 0 precisely; radiation exposure from 'H.
'
Methods are classified by the compartment they measure (directly or indirectly);the remaining compartment may be calculated by subtraction.
Methods of Measuring Lean Body Mass The two time-honored techniques of LBM assessment, anthropometric mea~ures'~ and densitometry by underwater weighing,14 are still the most commonly used today. Both are cumbersome and require a good deal of skill on the part of the investigator to achieve maximal precision. These disadvantages are often dwarfed by the sheer availability of these methods in many situations. Anthropometric measures are extremely sensitive to interobserver error, and training and cross-validation of investigators are crucial in order to maximize
preci~ion.'~ However, anthropometrics require little subject cooperation and can be carried out almost anywhere. Underwater weighing, on the other hand, is much more difficult to carry out in subjects who are ill, elderly, or afraid of water. Table 3 shows the precision and accuracy available from these and other methods. In contrast, bioelectrical impedance (BIA) requires little investigator training, practically no patient cooperation, and no elaborate eq~ipment.'~.''However, the technology is very sensitive to the hydration status of the subject, limiting its usefulness in clin-
TABLE 2 Elemental Analysis of Body Composition (Four-Compartment Models)' Element
Tissue Represented
Potassium
Body cell mass
Calcium
Bone ash
Carbon Nitrogen
Fat Protein
'
Measurement Technique
Comment
Passive counting of 40K NaJK, by ' H and "Na Dual photon absorptiometry Dual x-ray absorptiometry Delayed neutron activation analysis Neutron inelastic scattering Neutron capture
Radiation exposure Easily available May be more precise. Expensive, not easily available. Expensive, not easily available. Expensive, not easily available.
The element measured. the method of measurement. and the tissues represented bv the element are shown. Tissues represented may be overlapping (see text). TBW is usually included as one of the cohpartments measured for these models.
TABLE 3 Precision of Techniques Used in Compartmental and Elemental Analysis of Body Composition Precision (c.v.) Technique (Reference) Anthropometry ( 13,I5) Densitometry (14.15) Bioelectrical impedance ( 16) TOBEC ( 19) 4UK-counting(20.63) NVK, (21 ) ‘H or D,O dilution (5.63) DPA, DPX (22) Neutron capture (63) Neutron scattering ( I I , 12) Neutron activation (63)
Target Tissue FM. LBM FM
LBM LBM BCM, LBM BCM, LBM TBW Bone Protein (nitrogen) Fat (carbon) Calcium (total body)
ical situations.” BIA is now being used in the Third National Health and Nutrition Examination Survey (HANES 111) and in the Framingham Offspring Study, and has been validated in normal subjects by densitometry.16 A related method, utilizing electrical conductivity rather than capacitance, is the TOBEC, or total body electrical conductivity, approach.” TOBEC equipment is larger and more cumbersome than BIA, but the result may be more accurate and precise.’’ Another approach is to measure TBW by deuterium, tritium, or l80 dilution, and to use this as an estimate of FFM on the basis of the known average hydration of lean tissue (LBM = TBW/0.732).4 This method has the advantage of being relatively easy to carry out. However, although TBW is part of the FFM and correlates with LBM, it is by no means identical to it. Thus, while TBW measurements would be adequate to adjust for confounding effects of lean mass on certain outcomes, such techniques are probably not adequate in research in which the outcome of interest is LBM itself. There are two major limitations of the two-compartment approach as currently defined. First, the assumption of constant hydration of lean tissues may not hold for all persons, especially those at the extremes of body composition or health status. Second, only one of the two compartments is measured and the other is esti-
166 NUTRI”
RfflEWSNOL 49, NO &JUNE 19Sl
Accuracy (Subjective) 5-10 3
Although the composition of the human body has been under study for at least a century, investigation of the interaction among lean body mass and functional status, immune competence, and survival is only now beginning to emerge. Body composition concepts have recently been reviewed in detail by Heymsfield;’ this article will focus on the connection between lean body mass (LBM) and function. Because the method of measurement of any body compartment largely defines that compartment, we first touch on current methods of LBM measurements and their limitations. Approaches to Composition of the Organism The body can be divided into a variable number of different compartments, depending in large part on who is doing the dividing and what instrument(s) they are Dr. Roubenoff is Research Associate in and Dr. Kehayias is Director of the Body Composition Laboratory, USDA Human Nutrition Research Center on Aging at Tufts University, 71 1 Washington Street, Boston, MA, 021 11. Supported by NIH Grant DK08443, by a Pew National Nutrition Program Clinical Fellowship. and by the USDA.
using to measure the various parts (Figure 1 and Table 1). The most intuitive division is fat mass (FM) from fat-free mass (FFM), a dichotomy based on the chemical analysis of human cadavers by dissolving away the lipid in organic ~ o l v e n t . FFM ~ . ~ is slightly different from LBM because the fat in cell membranes is excluded from FFM but is retained in LBM due to the anatomical proximity of this small amount of fat to lean tissue^.^ LBM, in turn, is made of body cell mass (BCM) and the nonfatty intercellular connective tissue, such as bone (excluding fatty marrow), tendons, ligaments, basement membranes, etc. It is within the BCM that over 99O/0 of the body’s metabolic processes take place.4 At the same time, another compartment that partially overlaps LBM, FFM, and BCM is total body water (TBW). TBW clearly excludes FM. However, it is not identical to any of the nonfat compartments discussed earlier because it includes both intracellular and extracellular water, but not the extracellular connective tissues or the intracellular nonwater ~ t r u c t u r e s . TBW ~’~ can be measured with deuterium, tritium, or antipyrene dilution, methods that have rbeen available for several decades. In addition, TBW can be divided into intracellular NUTRl’TtON RNEWSIVOL 49. NO (VJUNE 1991 163
100%
76%
60%
I
Ieolb
I/
26%
I1 /6AI
I
0%
Male
Female
=
PIaoma ECF 0Wator Fat-Fro. Solldo Fat Figure 1. Division of the body into compartments for normal males and females. Note that TBW = water + ECF + plasma = 60% of the body in men and 50% in women. LBM = TBW + fat-free solids = 82% of the body in men and 68% in women. After Moore (4).
and extracellular water using various tracers.'TBW also overlaps to some extent with muscle, an important compartment that is very difficult to measure because it contains water, nitrogen, and potassium. Yet muscle may be the most dynamic and rapidly influenced part of the LBM/FFM/BCM triad.* The previous two-compartment approach to body composition is unsatisfactory for several reason^.^ Most importantly, by lumping all nonfat tissues into one compartment, the distinct physiologic functions of the various tissues (i.e., bone vs. muscle, water vs. cell mass, etc.) are obscured. From a practical standpoint, the two-compartment model is not sufficiently precise to easily allow longitudinal measures of change in body composition under various dietary and pathologic conditions. Therefore, over the past decade, a series of multicompartment models have been developed." These models, taking advantage of new techniques, measure several parts of the nonfat tissue independently in order
164 NUTRITION RffIEWSNOL 49. NO WJUNE 1991
both to minimize measurement error and to obtain data on important components of body composition. The subdivision of the lean and fat compartments along elemental rather than compartmental lines is illustrated in Table 2. Elements that contribute exclusively to the nonfat tissue include nitrogen (present only in protein), calcium (bone), and potassium (BCM and LBM), as well as TBW. In contrast, carbon, hydrogen, and oxygen are present in both fat and lean tissue. If the amount of lean tissue can be estimated from one or more of its elemental contributors (i.e., potassium), then the amount of fat can be estimated directly from the amount of carbon in the body after subtracting the amount of carbon in lean tissue. This method affords much better precision than do any of the methods that only measure the lean compartment and total weight, and then subtract to estimate fat, leading to a propagated error that can typically be as large as 20% of the fat mass''*'2 (see subsequent discussion).
TABLE I Two-Compartment Models of Body Composition and Methods Used to Measure Them' Compartment Fat mass (FM)
Measurement Methods Cadaver-lipid Densitometry
extraction
Anthropometry Fat-free mass (FFM)
Cadaverdissection Bioelectrical impedance
TOBEC Lean body mass (LBM)
Anthropometry
Total body water (TBW)
Dilution (D,O, 'H,O)
Comment Not routinely available. Assume known density of fat and lean; need cooperative subjects; commonly available. Interobserver error large; can be done nearly anywhere, Edema limits use in some clinical situations. Not routinely available. Precise, but may not be accurate. Very sensitive to hydration status. Depends on population-specific equations. Portable, easy to use. Same as above, but more accurate; equipment more expensive, less portable. Interobserver error large; can be done nearly anywhere. Edema limits use in some clinical situations. Difficult to measure D 2 0 precisely; radiation exposure from 'H.
'
Methods are classified by the compartment they measure (directly or indirectly);the remaining compartment may be calculated by subtraction.
Methods of Measuring Lean Body Mass The two time-honored techniques of LBM assessment, anthropometric mea~ures'~ and densitometry by underwater weighing,14 are still the most commonly used today. Both are cumbersome and require a good deal of skill on the part of the investigator to achieve maximal precision. These disadvantages are often dwarfed by the sheer availability of these methods in many situations. Anthropometric measures are extremely sensitive to interobserver error, and training and cross-validation of investigators are crucial in order to maximize
preci~ion.'~ However, anthropometrics require little subject cooperation and can be carried out almost anywhere. Underwater weighing, on the other hand, is much more difficult to carry out in subjects who are ill, elderly, or afraid of water. Table 3 shows the precision and accuracy available from these and other methods. In contrast, bioelectrical impedance (BIA) requires little investigator training, practically no patient cooperation, and no elaborate eq~ipment.'~.''However, the technology is very sensitive to the hydration status of the subject, limiting its usefulness in clin-
TABLE 2 Elemental Analysis of Body Composition (Four-Compartment Models)' Element
Tissue Represented
Potassium
Body cell mass
Calcium
Bone ash
Carbon Nitrogen
Fat Protein
'
Measurement Technique
Comment
Passive counting of 40K NaJK, by ' H and "Na Dual photon absorptiometry Dual x-ray absorptiometry Delayed neutron activation analysis Neutron inelastic scattering Neutron capture
Radiation exposure Easily available May be more precise. Expensive, not easily available. Expensive, not easily available. Expensive, not easily available.
The element measured. the method of measurement. and the tissues represented bv the element are shown. Tissues represented may be overlapping (see text). TBW is usually included as one of the cohpartments measured for these models.
TABLE 3 Precision of Techniques Used in Compartmental and Elemental Analysis of Body Composition Precision (c.v.) Technique (Reference) Anthropometry ( 13,I5) Densitometry (14.15) Bioelectrical impedance ( 16) TOBEC ( 19) 4UK-counting(20.63) NVK, (21 ) ‘H or D,O dilution (5.63) DPA, DPX (22) Neutron capture (63) Neutron scattering ( I I , 12) Neutron activation (63)
Target Tissue FM. LBM FM
LBM LBM BCM, LBM BCM, LBM TBW Bone Protein (nitrogen) Fat (carbon) Calcium (total body)
ical situations.” BIA is now being used in the Third National Health and Nutrition Examination Survey (HANES 111) and in the Framingham Offspring Study, and has been validated in normal subjects by densitometry.16 A related method, utilizing electrical conductivity rather than capacitance, is the TOBEC, or total body electrical conductivity, approach.” TOBEC equipment is larger and more cumbersome than BIA, but the result may be more accurate and precise.’’ Another approach is to measure TBW by deuterium, tritium, or l80 dilution, and to use this as an estimate of FFM on the basis of the known average hydration of lean tissue (LBM = TBW/0.732).4 This method has the advantage of being relatively easy to carry out. However, although TBW is part of the FFM and correlates with LBM, it is by no means identical to it. Thus, while TBW measurements would be adequate to adjust for confounding effects of lean mass on certain outcomes, such techniques are probably not adequate in research in which the outcome of interest is LBM itself. There are two major limitations of the two-compartment approach as currently defined. First, the assumption of constant hydration of lean tissues may not hold for all persons, especially those at the extremes of body composition or health status. Second, only one of the two compartments is measured and the other is esti-
166 NUTRI”
RfflEWSNOL 49, NO &JUNE 19Sl
Accuracy (Subjective) 5-10 3
