Validation of Bioelectrical Impedance Analysis for the Measurement of Appendicular Lean Mass in Older Women

Segmental multi-frequency bioelectrical impedance analysis (SMF-BIA) has been shown to be a valid, more portable, and less expensive alternative to dual energy x-ray absorptiometry (DXA) for the measurement of appendicular lean mass (ALM) in several populations. However, few studies have examined the validity of SMF-BIA specifically among populations of older women classified as sarcopenic or dynapenic. The aim of this cross sectional study was to investigate the accuracy of SMF-BIA compared to the gold standard DXA for the measurement of ALM among sarcopenic and/or dynapenic older women. Physical function, ALM, strength, and anthropometric tests were performed to determine the presence of sarcopenia and/or dynapenia in a sample of 25 older (72.3 ± 4.6 years) women using established sarcopenia classification criteria. Estimation of ALM using SMF-BIA and DXA were performed under standardized testing conditions. Both techniques were administered within the same hour with participants adequately hydrated, fasted, following urine elimination, and while wearing hospital scrubs. A Pearson correlation coefficient was used to determine a relationship between the two methods for ALM and agreement between the two techniques was assessed using a Bland-Altman plot method. Statistical significance was set at p < 0.05. A significant and strong correlation was observed between the two techniques with a Pearson’s correlation coefficient of 0.88 (95% CI= 0.75 to 0.95; p < 0.001). The Bland-Altman plot showed a mean difference of 0.5 kg and an indication of overall agreement between techniques. However, SMFBIA overestimated ALM for one participant (-2.9 kg, 95% CI= -3.76 to -2.03) and underestimated ALM for another participant (1.8 kg, 95% CI= 0.98 to 2.71) compared to DXA. The findings indicate overall agreement between SMF-BIA and DXA for the estimation of ALM among older women with characteristics of sarcopenia, but SMFBIA may overestimate or underestimate ALM in some individuals. These data suggest that SMF-BIA may be an acceptable alternative to DXA for the assessment of ALM in older women with characteristics of sarcopenia.


INTRODUCTION
Sarcopenia is defined as the progressive loss of muscle mass, physical function, and strength (i.e. dynapenia) which accelerates with aging [1][2][3]. Sarcopenia is associated with several adverse health outcomes among older adults including frailty, weakness, functional decline, increased risk of falls and fractures, and loss of independence [1,2]. In the year 2000, estimated healthcare costs associated with sarcopenia were $12.6 billion for older men, and $29.5 billion for older women [4].
The economic burden associated with sarcopenia is likely far higher today particularly in women, as they have a greater life expectancy and risk for functional impairment compared to men [5]. Although a universally accepted definition for sarcopenia is currently lacking, several working groups have convened to establish operational criteria for classification including measures of physical function, strength, and muscle mass.
While a wide range of techniques are available for assessing muscle mass, dual energy X-ray absorptiometry (DXA) is considered as the gold standard in research and clinical practice to determine appendicular lean mass (ALM) despite its limitations including exposure to radiation, expense, and lack of portability [2,[6][7][8]. Thus, is it critical to find alternative methods for ALM assessment.
The use of segmental multi-frequency bioelectrical impedance analysis (SMF-BIA) for the measurement of muscle mass has been accepted as a valid, more portable, and less expensive alternative in several populations [9][10][11][12] including older women [13,14]. While one study has validated the use of SMF-BIA for the estimation of ALM among a general population of older women [13], the device has not yet been validated for the assessment of ALM and its application to sarcopenia identification based on several accepted definitions. Thus, the purpose of this investigation was to assess whether SMF-BIA would accurately quantify ALM compared to DXA among a population of older women with characteristics of sarcopenia and/or dynapenia. It was hypothesized that SMF-BIA would be in agreement with DXA for the estimation of ALM. A secondary aim was to examine if SMF-BIA similarly classifies older women as sarcopenic compared to DXA based on ALM compared to height or body mass index (BMI) using established cut-points.

Study Design
The study utilized a cross-sectional design to compare differences between SMF-BIA and DXA for the estimation of ALM among 25 older women pre-screened for the presence of sarcopenia and/or dynapenia (reduced strength). The participants were pre-screened for entry into a 12-week University of Rhode Island Institutional Review Board approved, Resistance Exercise Study to Reclaim Lean Muscle and Strength (URI RESTORE ME Project: IRB # HU1415-168). The two techniques being compared for this study were performed at baseline of the URI RESTORE ME project.

Participants
Community dwelling women (n = 25) aged 65 to 84 years were recruited from Rhode Island via flyers, newspaper ads, press releases, and by word of mouth within local senior centers. All women initially eligible based on information provided during a telephone screening were invited to an onsite orientation and pre-screening assessment. During pre-screening, all women provided written informed consent and were evaluated for the presence of sarcopenia and/or dynapenia determined from the assessment of lean mass, height, weight, grip strength, single chair stand, and gait speed based on the European Working Group on Sarcopenia in Older People (EWGSOP) and the Foundation for the National Institute of Health Sarcopenia Project (FNIHSP) criteria. The pre-screening visit determined that 38 participants were eligible based on the sarcopenia criteria. Although eligible, 13 participants were subsequently eliminated from the final cohort due to questions of time commitment or orthopedic concerns that would limit full participation in the intervention phase of the project. Figure 1 summarizes each phase of study recruitment.

Outcome Measures
All testing sessions were standardized with participants arriving in the morning in a fasted state. Participants were instructed to arrive well hydrated and having not exercised. Both techniques for measuring body composition were performed on the same day, within the same hour, and following urine elimination. All participants were dressed in hospital scrubs for both methods.
Appendicular Lean Mass: ALM was estimated using fan-beam technology (GE Lunar iDXA, Waukesha, WI). The DXA scans were performed using a standardized procedure for patient positioning by a licensed radiology technician. The DXA scans provided estimations of ALM defined as the total skeletal muscle mass from the right arm, left arm, right leg, and left leg measured in kilograms.
In addition to DXA, a SMF-BIA was used to estimate segmental muscle mass whereby ALM could then be calculated. The InBody 570 Biospace device (Biospace Co, Ltd, Korea) was used according to the manufacturer's guidelines during both the pre-screening assessment (to determine sarcopenia status) and baseline testing to assess ALM. Body composition is estimated via SMF-BIA using the difference of conductivity of various tissues due to the differences in their biological characteristics.
The risk of adverse events is low using SMF-BIA and its use has been performed safely on older adults in other studies [15]. As a precaution, women with a cardiac pacemaker or internal defibrillator were not tested.
Sarcopenia Status: Prior to the collection of baseline testing data, participants completed pre-screening to determine their eligibility as presenting with sarcopenia and/or dynapenia. An algorithm for determining eligibility was developed based on a combination of both the EWGSOP and FNIHSP working definitions for sarcopenia.
Initial ALM estimates using SMF-BIA in addition to outcomes of physical function using the gait speed, single chair stand, and grip strength tests were included in the algorithm. The cut-points established for each criteria included an ALM < 5.67 kg/m 2 or ALM/BMI < 0.512, a grip strength < 20 kg or inability to complete a single chair stand, and a gait speed < 0.8 m/s [2,16]. Women selected were classified as having low lean mass (LM), low strength, low function (slow gait), or all three characteristics.
Physical Function: As part of the short physical performance battery (SPPB) of tests, usual gait speed was measured during pre-screening and during baseline to assess physical function and as a fundamental criterion for defining sarcopenia status.
Gait speed has been suggested as being included in routine evaluations as the sixth vital sign [7] as it is a strong predictor of major health outcomes and survival in older adults [6]. The EWGSOP, International Working Group (IWG), and FNIHSP are all in agreement that a gait speed < 0.8 m/s should be used as a cut-point for identifying sarcopenia in older adults [2,16,18]. Standardized protocols were utilized for administration of the gait speed test which required participants to walk 4 meters at a usual pace [19]. The test was repeated twice with the faster of the two trials recorded in seconds.
Muscle Strength: Grip strength is reliable, valid, and simple to administer in both clinical and community settings and is a predictor of mobility impairment [20,21]. Grip strength was tested during pre-screening and during baseline using a handgrip dynamometer (Jaymar Hydraulic Dynamometer, J.A. Preston, Corp., Jackson, MS). The test was performed with the participant seated in a chair with the elbow bent to 90 degrees using a standardized protocol [22]. The dynamometer was set accordingly based on each individual's hand size so that the second knuckle of all four fingers rested flat on the handle. Two trials were performed on each hand and the highest score was recorded in kilograms without accounting for hand dominance.
Grip strength values < 20 kg were also used as criteria for determining the presence of sarcopenia and/or dynapenia. Participants also completed a single chair stand test to assess lower body strength using a standardized protocol [23]. The test required that the participant rise from a chair with the arms crossed over the chest and with both feet flat on the floor. The ability or inability to rise from the chair was recorded and used to classify individuals as weak (i.e. dynapenia).
Anthropometrics: Each participant's height and weight were measured twice and the average of the two readings were used. Height and weight were measured with each participant barefoot and wearing hospital scrubs. Height was measured using a wall mounted stadiometer and weight was recorded using the InBody 570 SMF-BIA device. From these measurements, BMI was calculated by dividing the participants weight in kilograms by their height in meters squared.

Other Measures
Physical Activity: To assess each participant's physical activity level, the Yale Physical Activity Survey (YPAS) was administered during baseline testing. The YPAS quantifies the type and intensity of physical activity among older adults estimating total energy expenditure (kcals/week) and a total activity summary index.
The YPAS has been shown to be a valid measurement for the assessment of physical activity among older adults [24]. The YPAS was used to describe physical activity patterns of the participant population.
Dietary Intake: Participants completed a Dietary Screening Tool (DST) developed for older adults at baseline to assess dietary patterns [25]. The DST identifies 3 levels of nutritional risk: at risk (< 60), at possible risk (60 to 75), and not at risk (> 75) [26]. Nutritional risk assessed with the DST was used to describe the participant population.

Statistical Analysis
Estimated between-group differences were calculated for the primary outcomes, and a paired t-test was used to compare differences in estimated ALM by SMF-BIA and DXA. A previous study comparing SMF-BIA and DXA among frail women 75 years or older demonstrated an overall mean difference for ALM between the two techniques as 0.08 ± 0.05 kg [13]. Based on the findings of this study, a sample size of 16 were required to provide sufficient power (0.80) and alpha at 0.05.
Clinical and demographic characteristics of the participants are expressed as mean ± standard deviation. A Shapiro-Wilk test was performed to verify that data for the primary outcome variables were within normal distribution. To determine the strength of the relationship between the two variables, a Pearson moment correlation coefficient was performed. To test the primary hypothesis, a Bland-Altman method was employed to compare agreement between DXA and SMF-BIA whereby the differences in techniques are plotted against the averages of the two techniques [27].
To test the secondary hypothesis, a Fisher's exact test was used to compare sarcopenia status based on ALM between the two techniques. Statistical analyses were performed using SAS version 9.2 (SAS Institute Inc., Cary, NC, USA).

RESULTS
A total of 25 female participants with a mean age of 72.3 ± 4.6 years were enrolled in this study. Demographic characteristics, anthropometric measures, indictors of physical functioning, physical activity, and dietary habits of the study population are summarized in Table 1. A significant difference (p = 0.0145) was observed between DXA and SMF-BIA for the measurement of ALM by a mean of 0.5 kg for each participant.
Additionally, a 4.3% error of ALM by SMF-BIA was observed compared to DXA as the gold standard (p = 0.0162). A significant and strong positive correlation was observed between SMF-BIA and DXA as indicated by a Pearson's correlation coefficient of 0.88 (95% CI= 0.75 to 0.95, p < 0.001) highlighted in Figure 2. In addition, agreement between the two methods was assessed using a Bland-Altman plot When comparing sarcopenia status based on ALM measured by SMF-BIA and DXA, the Fisher's exact test indicates a significant difference (p < 0.001) between the two techniques when using both the EWGSOP and FNIHSP criteria. Of the 25 women found to demonstrate characteristics of sarcopenia and/or dynapenia based on EWGSOP criteria during pre-screening, SMF-BIA identified less participants (n = 7) as sarcopenic compared to DXA (n = 11). While SMF-BIA and DXA similarly identified the normal and pre-sarcopenic participants, DXA was able to identify more participants as having sarcopenia based on current EWGSOP criteria (Table 2).
Similarly, DXA was able to identify more participants as having low lean mass (n = 9) compared to SMF-BIA (n = 6) based on FNIHSP criteria (Table 3).

DISCUSSION
The results of this study indicate overall agreement between SMF-BIA and DXA for the estimation of ALM among older women with characteristics of sarcopenia and/or dynapenia. A significant and strong correlation (r = 0.88) and the lack of significant difference between the two techniques indicates acceptability of SMF-BIA compared to DXA. The Bland Altman plot provided a visual judgement of agreement between techniques with a mean overestimation of ALM by SMF-BIA of 0.5 kg. While most participants fell within the 95% limits of agreement, two outliers were identified. The plot indicates that SMF-BIA overestimated ALM but was not statistically different from DXA. The results of the present study are in line with previous studies which support that SMF-BIA is a valid, more portable, and affordable alternative to DXA for the estimation of ALM [7,9,10,13,14].
While the study indicates that SMF-BIA is a valid alternative to DXA, DXA appears to be more reliable for assessing ALM for diagnosing sarcopenia. Criteria for sarcopenia classification required that the participants ALM be < 5.67 kg/m 2 or ALM/BMI < 0.512 in combination with weak grip strength (< 20 kg) and/or slow gait speed (< 0.8 m/s). These findings suggest that SMF-BIA may overestimate ALM just enough to allow some participants to be above the sarcopenia ALM cut-off value.
While the grip strength and gait speed functional outcomes stand alone to identify individuals at risk for sarcopenia, DXA is the most reliable method for identifying sarcopenic females based on ALM. Although overall agreement between SMF-BIA and DXA for the estimation of ALM was observed in this study, investigators should use caution when using some impedance models for diagnosing sarcopenia.
In agreement with these findings, previous studies have demonstrated that SMF-BIA is a valid alternative to DXA for the estimation of ALM among healthy adult [9,10] and older female participants [13,14]. Andersen et al. [10] determined that similar SMF-BIA devices, namely the InBody 520 and 720, were valid for the estimation of ALM among healthy young women between the ages of 18 and 49 years.
While both devices were strongly correlated to DXA, (r = 0.62 -0.87) the conclusions were that SMF-BIA underestimated ALM by approximately 1.0 kg in the 25 female participants. Similar results were observed by Leahy et al. [9] with BIA underestimating arm and leg fat free mass by 0.2 kg (2.4 %) and 0.8 kg (3.4 %), respectively, in men and 0.2 kg (4.5 %) and 0.7 kg (4.4 %), respectively, in women (p < 0.001). Another similar study identified SMF-BIA as an appropriate alternative to DXA for the estimation of fat mass, percent body fat, and total lean mass, but not for the evaluation of ALM among a healthy adult population between 18 and 85 years [12]. Although the present study identifies the InBody 570 SMF-BIA device to be a valid alternative to DXA, the tendency for SMF-BIA to overestimate or underestimate ALM in some participants did occur as consistent with previous study findings.
Conflicting results regarding the agreement between SMF-BIA and DXA specifically for the measurement of ALM among elderly women have also been reported in other similar studies [7,13,14]. Buckinx et al. [7] observed low agreement between the two methods among approximately 25 women in the subcategory of participants over the age of 65 (n = 48) using a Bland-Altman plot method. The study reported a tendency for SMF-BIA to overestimate ALM by approximately 1.75 kg [7].
Other studies investigating older women report acceptable accuracy of SMF-BIA compared to DXA for the estimation of ALM. Among an older female Japanese population (n = 330) aged 65 to 87 years old, SMF-BIA tended to underestimate ALM by 1.59 kg (95% CI= 1.49 to 1.68) [14]. Segmental multi-frequency BIA also provided acceptable accuracy for the estimation of ALM among frail women 75 years and older with a tendency to underestimate ALM (r = -1.42, p < 0.01) [13]. In the present study, despite significant differences in sarcopenia definitions between SMF-BIA and DXA, the Bland-Altman plot indicated overall agreement between the two methods for the estimation of ALM with a mean difference of 0.5 kg among the older female participants.
This study has both strengths and weaknesses. The main strength was the use of standardized testing conditions. All participants were instructed to arrive for testing fasted, well hydrated, having not exercised, and following urine elimination. These confounding variables were well-controlled in addition to SMF-BIA and DXA being performed within the same hour. Hydration status upon arrival for testing was assessed based on self-report in addition to a normal ratio of extracellular water to total body water content (ECW/TBW) estimated by SMF-BIA. All participants ECW/TBW ratios were within the normal range of 0.36 to 0.39 as defined by the InBody manufacturer. The use of DXA as the gold standard method for body composition analysis by which SMF-BIA could be compared was an additional strength of the present study.
Potential limitations include the small sample size and homogeneity of the participant population as the results may not be generalizable to other populations. A larger sample size of older women including those from different racial, ethnic, and socioeconomic backgrounds could have demonstrated more similarity between SMF-BIA and DXA using Fisher's exact analysis. The present study was limited to investigating the validity of only one of many SMF-BIA devices, the InBody 570.
Additionally, the study was cross sectional therefore, future studies should examine changes in ALM longitudinally and following interventions.

CONCLUSION
While previous studies have investigated the validity of SMF-BIA for estimation of ALM compared to DXA among older women, this is the first study to validate SMF-BIA in a population of older women who present with sarcopenia and/ or dynapenia. In conclusion, the present study confirmed overall agreement between SMF-BIA and DXA for the measurement of ALM among older women who present with characteristics of sarcopenia and/or dynapenia based on current definitions. It is possible however, that SMF-BIA may overestimate or underestimate ALM in some participants. While SMF-BIA is an acceptable, less expensive, and more portable alternative to DXA for measuring ALM, future studies are required using other devices and larger samples to justify that SMF-BIA is more realistic for clinical and epidemiological identification of sarcopenia based on current definitions.

Introduction
Sarcopenia is defined as the progressive loss of muscle mass, physical function, and strength (i.e. dynapenia) which accelerates with advancing age. contributing to the loss of strength at an even greater rate. 1,5,6 Muscle strength has been reported to decline 2 to 5 times faster than muscle mass with age contributing to poor physical function, mobility limitation, and disbaility. [7][8][9] Sarcopenia is associated with several adverse health outcomes among older adults including frailty, weakness, functional decline, increased risk of falls and fractures, and loss of independence. 1,2 Moreover, in the year 2000, estimated healthcare costs associated with sarcopenia were $12.6 billion for older men, and $29.5 billion for older women. 10 The economic burden associated with sarcopenia is likely far higher today particularly in women, as they have a greater life expectancy and risk for functional impairment compared to men. 11,12 Although there is currently no universal definintion for identifying sarcopenic While a wide range of techniques are available for assessing muscle mass, DXA is considered as the gold standard in research and clinical practice to determine ALM despite its limitations including exposure to radiation, expense, and lack of portability. 2,14,15,4 Thus, is it critical to find alternative methods for ALM assessment.
The use of segmental multi-frequency bioelectrical impedance analysis (SMF-BIA) for the measurement of muscle mass has been accepted as a valid, more portable, and less expensive alternative in several populations [16][17][18][19] including older women. 20,21 While some studies have validated the use of SMF-BIA for the estimation of ALM among older women, 20 the device has not yet been validated for the assessment of sarcopenia based on current international definitions.

Sarcopenia in Older Women
Women represent the largest proportion of the elderly, and their patterns of functional impairment related to declines in muscle mass and strength differ from men. 12 Women are at a greater risk for developing decreased muscle strength (dynapenia) and sarcopenia with aging which also increases their risk for mortality compared to men as they have more fat, lower absolute muscle mass, and less strength. 1,22 The earlier development of muscle weakness observed in women compared to men can be partially explained by menopause as a reduction in the sexhormone estrogen is associated with loss of bone mass and strength. 11,23 Changes in the sex hormones estrogen and progesterone begin a decade or more prior to menopause, which may accelerate the morphological changes in skeletal muscle characteristic of sarcopenia. Consequently, post-menopausal women present with half the concentration of estrogen observed before menopause thus increasing their incidence of sarcopenia. 24 Women also experience earlier declines in muscle power at a rate of 3 to 4% faster than strength compared to men. 24 Early declines in muscle power, defined as the ability to perform muscular work per unit of time could explain the variance in poorer physical function and subsequent disability. 24

Working Definitions for Sarcopenia
Screening individuals for clinically significant sarcopenia has presented as a major challenge for the justification of intervention strategies that could reduce disability among older adults without a universally accepted definition. As a result, several working groups have convened to establish criteria to identify individuals at risk in order to prevent adverse events associated with reduced lean mass (LM) and physical function. 13,26 Muscle mass can be quantified using DXA, SMF-BIA, and anthropometry while muscle strength can be measured most effectively by testing handgrip strength using a handheld dynamometer. 27 In clinical practice, physical performance is measured using the short physical performance battery ( The EWGSOP algorithm requires that low muscle mass accompanied by either low grip strength, slow gait speed, or a combination of all three be present for classification of pre-sarcopenia, sarcopenia, or severe sarcopenia. 2,13 The EWGSOP defines low muscle mass as an ALM in kilograms relative to height squared (kg/m 2 ) less than 7.23 kg/m 2 in men and less than 5.67 kg/m 2 in women. Their cutoff values for low grip strength are set at less than 30 kg in men and less than 20 kg in women.
In order to be considered slow, and individual must present with a gait speed less than 0.8 meters per second (m/s). Based on the results from screening of the above mentioned criteria, the severity of sarcopenia may be determined. Pre-Sarcopenia is defined by low muscle mass without low muscle strength whereas severe sarcopenia is defined by demonstrating all three criteria. Individuals with sarcopenia present with low muscle mass in combination with either low grip strength or slow gait speed.
The International Working Group defines sarcopenia as the age related loss of skeletal muscle mass and function caused by several factors including disuse, altered endocrine function, chronic diseases, inflammation, insulin resistance, and nutritional deficiencies which contribute to disability, hospitalization, and death. 28 The group considers that all older patients who present with observed decrements in physical function, strength, and overall health be evaluated. Their diagnosis of sarcopenia is based on the presence of low whole body or ALM in combination with poor physical functioning. Similar to the EWGSOP criteria, the IWG confirms diagnosis of sarcopenia based on the same ALM criteria however, their cut-points are less conservative with slow gait speed defined as less than 1.0 m/s. 28 Rather than define sarcopenia, the FNIH have proposed a two-step clinical paradigm to identify individuals with low muscle strength and low muscle mass by first screening for mobility impairment and low muscle strength measured by gait speed and grip strength, respectively. 13,29 The FNIHSP recommends that individuals be identified using either "weak with low LM" or "weak and slow with low LM" rather than use the term sarcopenia. 30 The group is in agreement with the EWGSOP and IWG on the use of a cut-point for slow gait speed defined as less than 0.8 m/s, as it is easy to measure in most settings and is highly predictive of survival and major health outcomes. 31 The cut-points established to identify weak individuals based on a grip strength are set as less than 26 kg in men and less than 16 kg in women compared to the EWGSOP criteria (< 30 kg in men and < 20 kg in women). 32 Their criteria for low LM is defined as an ALM < 19.75 kg for men and < 15.02 kg for women and the criteria for low LM based on the ratio of ALM relative to body mass index (BMI) is defined as < 0.789 in men and < 0.512 in women. 30,33 The prevalence of sarcopenia in adults over 50 varies greatly depending on the definition being used therefore, adoption of an operational definition is required for effective prevention and treatment strategies to be implemented. 2 A universally accepted sarcopenia definition would allow effective interventions to be delivered to at risk individuals identified using a variety of screening techniques for low ALM and low physical function.

Evolution of Body Composition Evaluation Techniques
Since the birth of early methods for assessing body composition, several advancements in the techniques for providing accurate regional and total body estimates of skeletal muscle mass have become widely available. Within a very short time span in the early 1970s, all three contemporary clinical and reference methods, computerized axial tomography (CT), magnetic resonance imaging (MRI), and DXA came into existence. 34 Early methods (CT and MRI) have more recently developed into refined methods that have the ability to quantify regional and total body skeletal muscle mass including bioelectrical impedance analysis (BIA) and DXA, all of which have advantages and disadvantages. 14 Three dimensional imaging techniques including CT and MRI have the ability to quantify all major tissues in the human body. Computerized axial tomography consists of a rotating x-ray tube and detector, which move in a perpendicular plane to the subject. The difference in the attenuation or weakening of the x-rays as they penetrate body tissues determines the density of the underlying tissues. The attenuated x-ray beams can then generate a computer image of the scanned area separating bone, adipose tissue, and lean tissue. An advantage of CT is its ability to construct highquality images providing a measure of tissue composition and quality however, the degree of radiation exposure and relatively high cost limits its use in research and clinical settings. 14 Unlike CT scanning, MRI does not use radiation; rather it creates images from radio frequency signals emitted by hydrogen nuclei. The fundamental MRI concept is based on interaction between nuclei of abundant hydrogen atoms and magnetic fields produced and controlled by the system's instrumentation. Application of an external magnetic field followed by a pulsed radio frequency across a body part causes the hydrogen nuclei to line up and absorb energy. When the radio wave is turned off, the nuclei emit the energy previously absorbed thus, emitting a signal that can then be used to create an image. While CT and MRI are accepted as accurate measures of whole body composition, neither are available or affordable methods for conducting sarcopenia trials. 34,35 Although CT and MRI provide the most accurate assessment of body composition, these reference methods have several limitations including expense, equipment availability, and excessive radiation exposure with CT, 36 therefore DXA has been used more frequently due to its relatively low cost and minimal radiation exposure. 34,37 DXA uses an x-ray tube to evaluate bone mineral density and soft tissue composition. The attenuation of the x-rays with high and low photon energies is measurable and dependent on the thickness, density, and chemical composition of the underlying tissue. 35 The scan has the ability to evaluate fat, lean soft tissue, and bone separately for the extremities, trunk, and other selected body regions. This segmental analysis can then allow estimation of ALM for assessing skeletal muscle loss with aging. 34 The availability of DXA, its modest scan cost, low radiation exposure, short scan time, and the extensive information provided from each scan has ultimately led to this approach being most widely used in sarcopenia research and clinical practice today. 14 While DXA is the most widely used method for assessing body composition, the use of BIA has grown rapidly in the past two decades as a noninvasive, portable, quick, and inexpensive alternative. Bioelectrical impedance analysis is based on the relation between the volume of a conductor and its electrical resistance, therefore skeletal muscle is a dominant conductor because it is an electrolyte rich tissue with low resistance. 10 The devices deliver an alternating current at one or more frequencies via electrodes, and impedance to electrical flow is detected. Electrolyte-rich fluids such as body water pose the least impedance to electrical flow, while lipids and bone minerals provide the most. The traditional whole body, single frequency models have evolved into segmental multi-frequency models which can measure resistance for each body segment at a wide range of frequencies. 38 The use of SMF-BIA has become more common as segmental analysis of the arm and leg can be empirically calibrated to DXA appendicular lean mass however, stable subject conditions are required for accurate results. 10,14 Comparison between BIA and DXA Although several techniques for measuring body composition are available in research and clinical practice, DXA is considered to be the gold standard due to its precision for distinguishing fat, bone, and muscle tissue. 2,39 As an alternative to CT scans and MRI, DXA became available in 1987 as a more practical method with limited radiation exposure. 34 The ability to isolate body regions during DXA analysis allowed many investigators the opportunity to evaluate fat, lean soft tissue, and bone mineral content separately for the extremities, trunk, and other selected body regions.
With the introduction of portable SMF-BIA devices, the same segmental body composition analysis could be estimated with no exposure to radiation and at a very low cost. 2