Maeda 

et al. Advances in Rheumatology (2022) 62:11 Advances in Rheumatology

https://doi.org/10.1186/s42358-022-00240-9

POSITION ARTICLE AND GUIDELINES Open Access

Official Position of the Brazilian Association

of Bone Assessment and Metabolism (ABRASSO)
on the evaluation of body composition
by densitometry—part II (clinical aspects):
interpretation, reporting, and special situations
Sergio Setsuo Maeda1*  , Ben‑Hur Albergaria2, Vera Lúcia Szejnfeld3, Marise Lazaretti‑Castro1,
Henrique Pierotti Arantes4, Marcela Ushida3, Diogo Souza Domiciano5, Rosa Maria Rodrigues Pereira5,
Rosângela Villa Marin‑Mio1, Mônica Longo de Oliveira1, Laura Maria Carvalho de Mendonça6, Mirley do Prado7,
Guilherme Cardenaz de Souza8, Cecília Zanin Palchetti9, Roseli Oselka Saccardo Sarni10, Maria Teresa Terreri11,
Luiz Claudio Gonçalves de Castro12, Silvana Martinez Baraldi Artoni13, Lizandra Amoroso13,
Débora Emy Karcher13, Carla M. Prado16, Maria Cristina Gonzalez14,15 and Marcelo de Medeiros Pinheiro3 

Abstract 
Objective:  To present an updated and evidence-based guideline for the use of dual-energy x-ray absorptiometry
(DXA) to assess body composition in clinical practice.
Materials and methods:  This Official Position was developed by the Scientific Committee of the Brazilian Associa‑
tion of Bone Assessment and Metabolism (Associação Brasileira de Avaliação Óssea e Osteometabolismo, ABRASSO) and
experts in the field who were invited to contribute to the preparation of this document. The authors searched current
databases for relevant publications in the area of body composition assessment. In this second part of the Official
Position, the authors discuss the interpretation and reporting of body composition parameters assessed by DXA and
the use of DXA for body composition evaluation in special situations, including evaluation of children, persons with
HIV, and animals.
Conclusion:  This document offers recommendations for the use of DXA in body composition evaluation, including
indications, interpretation, and applications, to serve as a guiding tool in clinical practice and research for health care
professionals in Brazil.
Keywords:  Body composition, Absorptiometry, X-ray, Lean mass, Fat mass, Sarcopenia, Pediatrics, HIV

Background
In clinical practice, body composition may be assessed by
different methods such as air or water displacement, bio-
*Correspondence: [email protected] electrical impedance (BIA), computed tomography (CT),
1
Discipline of Endocrinology, Department of Medicine, Universidade
magnetic resonance imaging (MRI), and dual-energy
Federal de São Paulo (UNIFESP), Rua Estado de Israel, 639, São Paulo, SP
CEP: 04022‑001, Brazil x-ray absorptiometry (DXA). Anthropometric meas-
Full list of author information is available at the end of the article urements can also be used as a surrogate approach for

© The Author(s) 2022. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which
permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the
original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or
other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line
to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory
regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this
licence, visit http://​creat​iveco​mmons.​org/​licen​ses/​by/4.​0/.
Maeda et al. Advances in Rheumatology (2022) 62:11 Page 2 of 25

estimating body composition. Each of these methods has Materials and methods

strengths and limitations in terms of accessibility, accu- This document is a result of efforts by the Brazil-
racy, and comprehensiveness [1]. ian Association of Bone Assessment and Metabo-
Systems of BIA measure the resistance to a harmless lism (Associação Brasileira de Avaliação Óssea e
electric current flow passing through the body. The elec- Osteometabolismo/Brazilian Society on Bone and Osteo-
tricity conducted through the water in the body provides metabolism Evaluation, ABRASSO) for the development
an estimate of the total body water and predicts an indi- of recommendations based on the current evidence avail-
vidual’s fat-free mass (FFM) based on assumed constant able in the scientific literature regarding measurement
hydration, as used in the deuterium dilution method. of body composition using DXA. The ABRASSO Scien-
However, BIA is not a reference method to measure body tific Committee invited experts in the field to contribute
composition since it relies on specific assumptions, of to the preparation of this document. The authors were
which the most important is constant hydration [2]. invited by ABRASSO to provide scientific information
In contrast, MRI and CT offer more detailed measure- on body composition measurements. ABRASSO was
ments of specific tissues and small areas, e.g., fat infiltra- chosen as the official organization for the preparation of
tion and visceral adiposity. Critical limitations of both this document considering its national expression and
methods include radiation exposure related to CT scan- the fact that it congregates professionals from several
ning, and high cost and limited availability related to MRI medical areas related to bone and mineral metabolism
scanning [1]. (rheumatology, endocrinology, gynecology, orthopedics,
Ultrasound has inconsistent image reproducibility that geriatric and gerontology, physiatry, sports medicine and
may be caused by variations in the pressure applied by rehabilitation, nephrology, infectious diseases, pediatrics,
the transducer on the skin, measurement site, anisotropy, veterinary medicine) along with supporting health care
and protocol feasibility, limiting the diagnostic and moni- professionals (nutritionists, dietitians, biomedical sci-
toring application of ultrasound for assessment of nutri- entists, biologists, pharmacists, physical therapists, psy-
tional status [3]. According to the current best evidence, chologists, and basic researchers). The main criteria for
only anthropometric measurements correlate with car- inviting collaborators were their areas of expertise, con-
diovascular risk and metabolic syndrome [4]. tributions to the field, association with medical organi-
Originally developed to measure bone mineral density zations related to the topics covered in this document,
(BMD) and bone mineral content (BMC), DXA is also and publication of papers and practical management
used to measure lean and fat mass. DXA systems straddle on the covered topic, thus fulfilling the endorsement
the line between a three-compartment model and a two- by ABRASSO and other participating medical socie-
compartment model (FFM = BMC + lean mass) of the ties. The invited authors were divided into small groups
body [2]. DXA is preferable for body composition assess- (with 2 to 6 authors per group) according to their areas
ment, as it performs whole-body scanning in a short of expertise and questions to be addressed. Additionally,
time, emits low radiation, provides regional analyses, and all the authors composed the steering committee for the
is likely to be more accessible and affordable than CT or development of the study that resulted in the present
MRI. Compared with other methods, DXA has been rec- document and designed the protocol to address specific
ommended as the standard method for assessing body questions related to the applicability of body composi-
composition in most patient groups (Table 1) [1, 3, 5, 6]. tion measurements (including technical and practical

Table 1  Strengths and limitations of different methods of evaluation of body composition (adapted from References [1–5])
Measurement technology Time/cost/ Radiation-free Accuracy Precision Regional versus Muscle/fat Intratissue fat VAT
availability whole body

Anthropometry ++ ++ −/+ −/+ − − − −

Air/water displacement − ++ + ++ − − − −
BIA + ++ −/+ + + −/+ − −/+
Ultrasound + ++ + −/+ − + + +
CT + − ++ ++ − ++ ++ ++
MRI − ++ ++ ++ ++ ++ ++ ++
DXA + + + ++ + + − +
DXA dual-energy X-ray absorptiometry, BIA bioelectrical impedance, CT computed tomography, MRI magnetic resonance imaging, VAT visceral adipose tissue, + 
strength, with +  + defining strong evidence, − limitation
 Maeda et al. Advances in Rheumatology (2022) 62:11 Page 3 of 25

issues). All the authors wrote the manuscript with input then by the ABRASSO Steering Committee. Using elec-
from each other, critically reviewed the manuscript, tronic correspondence (email), the collaborators in each
and approved its final version for submission (fulfill- group discussed the articles based on their expertise until
ing the criteria for authorship). Only one of the authors they reached a consensus regarding the best and most
had a conflict of interest to disclose related to the topic current scientific evidence. The final questions presented
of body composition measurements (reported at the end in this second part of the Position Statement were chosen
of the paper), and all the authors participated actively in by the ABRASSO Steering Committee and by experts in
the discussions and are responsible for the information body composition assessment using DXA. These ques-
reported in this document. tions were based on the main questions and problems
The aim of this position statement is to answer routine encountered in clinical practice concerning the clinical
questions about body composition assessment and serve aspects of body composition assessment by DXA and are
as a guideline for clinicians and researchers in Brazil. The presented into the following sections: clinical aspects,
authors searched current databases for relevant publica- interpretation, reporting, and special situations. Finally,
tions and described their findings below using a narrative the authors and the ABRASSO Steering Committee pre-
review format. The search strategy was similar among all pared a statement answering each question based on
authors and was conducted by each group using the elec- current scientific evidence. Using a Likert scale, the final
tronic databases MEDLINE (via PubMed), Embase, and agreement level (from 0 to 100%) was reached through
SciELO. The expressions used included “adult and pediat- electronic voting among all collaborators for all 10 state-
ric normative data,” “lean mass measurements,” “fat mass ments (Table 2).
measurements,” “basic area and technical science,” “other
anthropometrical measurements,” “other non-DXA body Section I: Interpretation
composition measurements,” among others. The authors 1. What are the validated criteria for bone mass
also searched for other potential studies not retrieved by assessment?
the search strategies by consulting review articles, meta- The Official Position of the International Society for Clin-
analyses/systematic reviews, and guidelines issued by ical Densitometry (ISCD) recommends reports of DXA
specialty societies, particularly the International Society body composition in adults to include measurement of
for Clinical Densitometry (ISCD) Official Position. To whole-body (including head) BMD and BMC [7, 8]. How-
increase the search sensitivity, MeSH search terms were ever, these measurements are not used as isolated skel-
used for clinical conditions and therapeutic interven- etal health markers nor as diagnostic of osteoporosis or
tions but not for comparators or outcomes. Only stud- low bone mass in adults [7, 9]. Both the ISCD and our
ies published in Portuguese, English, and Spanish were guidelines establish that the diagnostic criteria of the
considered. The search was limited to studies published World Health Organization (WHO) for densitometric
between January 1st, 2000, and July 31st, 2021. The osteoporosis only apply to the skeletal sites of the proxi-
search in each electronic database included the follow- mal femur (femoral neck and total hip), lumbar spine
ing descriptors (key words): “body composition meas- (L1–L4), and 33% radius. [7–9] The only exception is the
urements,” “DXA,” “other measurements NO DXA,” use of total body less head (TBLH) bone mass Z-score
“skinfold,” “plethysmography,” “ultrasound,” “computed values as a diagnostic criterion of low bone mass in pedi-
tomography,” “magnetic resonance imaging,” “bioelectri- atric patients (5–19  years of age), with an adopted cut-
cal impedance analysis,” “absorptiometry,” “x-ray,” “meth- off value of − 2.0 standard deviations (SDs) of the mean
odology,” “artifacts,” “technical procedures,” “fat mass,” value obtained from individuals of the same age [7, 9].
“bone mass,” “lean mass,” “sarcopenia,” “DXA,” “clinical The 1999–2004 National Health and Nutrition Exami-
conditions,” “elderly,” “obesity,” “adiposity,” “children and nation Survey (NHANES) reference data for total body
adolescents,” “HIV,” “animals,” “physical parameters,” BMC should be adopted when DXA is used for body
“transgenders,” “Brazilian normality data,” and “clinical composition assessment in children. The data comprise
applicability.” Due to the extent of the position statement, calculated Z-scores and percentiles for children as young
it was divided into two parts. Part I was dedicated to a as 8 years of age and adults aged up to 85 years, men and
revision of methods for evaluation of body composition women, and Whites, Blacks, and Mexican Americans.
and their technical aspects, and Part II focused on the Still, NHANES reference data are not available for many
interpretation of results and clinical applications. ethnic minorities. Also, more country-specific reference
A total of 131 articles were reviewd for the preparation datasets should be developed [10].
of this second part of the Position Statement. All articles In summary, reports of body composition assessed by
were carefully analyzed first by the groups of authors and DXA should include the following:
Maeda et al. Advances in Rheumatology (2022) 62:11 Page 4 of 25

Table 2  Statements from the Official Position of the Brazilian Association of Bone Assessment and Metabolism (ABRASSO) regarding
the clinical application of body composition measurements using dual-energy x-ray absorptiometry (DXA), along with the levels of
agreement (interrater reliability) among the statement’s collaborators
Question Statement Level of
agreement
(%)

1. What are the validated criteria for bone mass assessment? The WHO criteria for densitometric osteoporosis only apply to 96.7
the skeletal sites of the proximal femur (femoral neck and total
hip), lumbar spine (L1–L4), and 33% radius. The only exception is
the use of total body less head (TBLH) bone mass Z-score values
as a diagnostic criterion of low bone mass in pediatric patients,
with an adopted cutoff value of − 2.0 standard deviations (SDs)
of the mean value obtained from individuals of the same age.
The NHANES III reference database for total body BMC should be
adopted when DXA is used for body composition assessment in
children
Body composition DXA reports regarding bone mass should
include:
BMC results (in grams);
BMD values (in g/cm2) and Z-scores (SDs) should be reported
for adults, but without establishing a diagnosis of osteopenia or
osteoporosis. For individuals with Z-score values below − 2.0 SD,
the sentence “low bone mass for age” may be reported;
TBLH and Z-scores should be reported in children and adolescents
2. What are the criteria for assessing fat mass? Recommendations for assessment of fat mass include the fat mass 95.7
index (FMI; in kg/m2), interpreted according to the NHANES III cut‑
off values, the estimated abdominal visceral adipose tissue (VAT; in
g/cm3 if assessed with a GE-Lunar device or g/cm2 if assessed with
a Hologic device), and the android-to-gynoid (A/G) fat ratio
3. What are the validated criteria for assessing lean mass? Several validated criteria are available for assessing appendicular 90.7
lean mass using DXA, including the Baumgartner criteria, the
Newman criteria, and the Foundation for the National Institutes of
Health (FNIH) Sarcopenia Project criteria
Lean mass can also be assessed using muscle strength param‑
eters, including dynamometry and indirect and dynamic physical
fitness and functional capacity tests evaluating static and dynamic
balance, mobility, and flexibility, such as the chair-stand test and
isokinetic chair
4. Which parameters should be included in the DXA body composi‑ For adults older than 20 years, report whole-body (including head) 98
tion report? values of:
Anthropometry: weight (kg), height (m), and BMI (kg/m2)
Bone mass compartment: BMD (g/cm2), BMC (g), and Z-score (in
SDs)
Fat mass compartment: total fat mass (in kilograms), percentage of
fat (in %), FMI (total fat mass/height2, in kg/m2), A/G ratio, and VAT
(in g and c­ m3)
Lean mass compartment: total lean mass (kg), ALM (kg), ALMI
(adjusted by height [ALM/height2] and adjusted by BMI in patients
over 65 years old [ALM/BMI])
5. What should be taken into account regarding quality control, The quality control program should adhere to the manufacturer’s 99.7
accuracy, and least significant change (LSC)? guidelines for system maintenance. The quality control should
include:
Periodic (at least once a week) phantom scans for any DXA system
as an independent assessment of system calibration
Plotting and reviewing of data from calibration and phantom scans
Establishment and enforcement of corrective action thresholds
that trigger a call for service
The precision error supplied by the manufacturer should not be
used
Each DXA device should have its own in vivo precision error deter‑
mined and LSC calculated for all body composition variables
 Maeda et al. Advances in Rheumatology (2022) 62:11 Page 5 of 25

Table 2  (continued)
Question Statement Level of
agreement
(%)

6. What are the differences between normative data for the Brazil‑ Based on Brazilian normative database studies in adult men and 98
ian compared with other populations? women, the Brazilian population, compared with other popula‑
tions, has some significant differences in body composition
parameters, particularly regarding appendicular lean mass adjusted
for height. Cutoff values of 7.77 kg/m2 and 5.62 kg/m2 (− 1 SD) are
suggested for men and women. The combination of calf circumfer‑
ence (≤ 34 cm for males and ≤ 33 cm for females) and SARC-F into
a modified index significantly improves the performance of SARC-F
for screening sarcopenia
7. What is the application of body composition assessment in Numerous conditions may potentially interfere with body compart‑ 95.3
pediatrics? ment distribution (lean, fat, and bone mass), including exogenous
and endogenous overweight and obesity, environmental and
disease-related undernutrition, anorexia, chronic drug therapy (e.g.,
corticosteroids, chemotherapy), and chronic diseases (e.g., systemic
inflammatory disorders, inborn errors of metabolism, muscular
dystrophies, and endocrine, gastrointestinal, heart, and pulmonary
diseases). The most frequently used parameter for estimating body
composition in routine practice in pediatrics is the BMI
Interpretation of pediatric DXA data may be challenging due
to physiological changes in body composition during growth,
particularly in the absence of Brazilian normative reference data for
children and adolescents. Thus, the adoption of the US normative
database (NHANES III) is recommended for pediatric assessments
in Brazil
8. What is the clinical application of body composition assessment Body composition assessment is recommended in patients 97.3
in patients infected with HIV? infected with HIV for monitoring of body composition changes
related to the disease and adverse effects associated with antiret‑
roviral therapy, particularly abnormal body fat redistribution in the
HIV-associated lipodystrophy spectrum
The following parameters may be useful for assessing the presence
of lipodystrophy in HIV-infected patients: limb-to-trunk fat ratio,
trunk/leg fat ratio, and fat mass ratio
9. How should body composition be assessed in transgender Consistent data on body composition assessment in transgen‑ 94.7
individuals? der individuals are currently unavailable. Until studies with more
consistent data are published, we recommended the calculation of
T-scores using a uniform Caucasian (non-race adjusted) female nor‑
mative database for all transgender individuals of all ethnic groups
and all transgender individuals aged 50 years or older, regardless of
hormonal status. Z-scores should be calculated using the norma‑
tive database that matches the gender identity of the individual (or
based on both male and female databases, if requested by the phy‑
sician). In gender-nonbinary individuals, the normative database
that matches the sex recorded at birth should be used
10. What is the role of DXA in veterinary medicine and zootechnics? DXA can be used in veterinary medicine and animal sciences for 98
measurement of whole-body composition in pigs, broilers, cats,
dogs, and sheep, among others. Although normative data in these
animals are scarce, this technique has a great potential in accu‑
rately evaluating the effectiveness of feeding interventions on the
amount of lean and fat mass

• BMC results (in grams), which should always be • The sentence “The diagnostic criteria proposed by
reported. the WHO are not applicable to whole-body DXA
• BMD values (in grams/cm2) and Z-scores (SDs) analyses.”
should be reported for adults but without establish-
ing a diagnosis of osteopenia or osteoporosis.
• TBLH and Z-scores should be reported in children Statement 1
and adolescents. The WHO criteria for densitometric osteoporosis
only apply to the skeletal sites of the proximal femur
Maeda et al. Advances in Rheumatology (2022) 62:11 Page 6 of 25

(femoral neck and total hip), lumbar spine (L1–L4), and Table 3  Classification of body fat based on the fat mass index
33% radius. The only exception is the use of total body (FMI, kg/m2) for men and women [10, 13]
less head (TBLH) bone mass Z-score values as a diag- Category Men Women
nostic criterion of low bone mass in pediatric patients,
with an adopted cutoff value of − 2.0 standard devia- Severe fat deficit < 2 < 3.5
tions (SDs) of the mean value obtained from individuals Moderate fat deficit 2 to < 2.3 3.5 to < 4
of the same age. The NHANES III reference database Mild fat deficit 2.3 to < 3 4 to < 5
for total body BMC should be adopted when DXA is Normal 3–6 5–9
used for body composition assessment in children. Excess fat > 6 to 9 > 9 to 13
Body composition DXA reports regarding bone mass Obese class I > 9 to 12 > 13 to 17
should include: Obese class II > 12 to 15 > 17 to 21
Obese class III > 15 > 21
• BMC results (in grams);
• BMD values (in g/cm2) and Z-scores (SDs) should
be reported for adults, but without establishing a
bolic syndrome [14] and is a predictor of coronary
diagnosis of osteopenia or osteoporosis. For indi-
heart disease, according to a prospective European
viduals with Z-score values below − 2.0 SD, the
study [15]. Compared with subcutaneous fat, VAT
sentence “low bone mass for age” may be reported;
is a better predictor of mortality [16]. Only a few
• TBLH and Z-scores should be reported in children
DXA systems are able to measure VAT. This meas-
and adolescents.
urement is performed in an area 5-cm thick located
1 cm above the iliac crest, approximately at the level
of the L4 vertebra, and is calculated as android fat
minus subcutaneous fat. VAT measured by DXA
2. What are the criteria for assessing fat mass?
correlates highly with visceral fat measured by CT
Body mass index (BMI) is a measure of weight adjusted
[17, 18], is associated with lower radiation exposure,
for height or length. In some circumstances, BMI val-
and is highly accurate. In a cross-sectional study
ues may reveal an individual’s excess weight, which is
including healthy women, VAT correlated posi-
not always reflective of excessive fat, for example, in the
tively with glycemia levels and with the homeostatic
case of heavily muscled persons. For this reason, DXA
model assessment insulin resistance index (HOMA;
body composition has been recommended for assessing
a method for assessing insulin resistance) and nega-
parameters including fat mass index (FMI), abdominal
tively with HDL-cholesterol [19]. Other studies have
visceral adipose tissue (VAT), and android-to-gynoid
analyzed the association between VAT measured by
(A/G) fat ratio.
DXA and cardiovascular risk parameters [20, 21]. A
cross-sectional study including 2,317 US adults aged
• FMI This index evaluates the ratio of total body fat
18–74  years determined, based on receiver operat-
(in kilograms) to squared height (in square meters)
ing characteristic (ROC) curves, a VAT threshold of
[11]. FMI is advantageous over BMI since BMI may
126  ­cm2 (76% sensitivity, 68% specificity) to identify
correlate poorly with body fat in some populations
individuals with two or more cardiometabolic risk
[12]. In this sense, patients with increased lean
factors [22]. Indeed, VAT may replace the A/G ratio
mass may be falsely classified as overweight based
in assessing cardiometabolic risk factors. Although
on BMI but not on FMI. For calculation of FMI,
population studies have established reference values
1999–2004 NHANES data are used [13]. The crite-
for VAT [23–25], the values still need to be stand-
ria for classifying body fat based on FMI are shown
ardized, especially with prospective studies assessing
in Table 3, with different cutoff values for men and
cardiometabolic risk factors and cardiovascular out-
women. Clinical applications for FMI have not been
comes [10].
clearly established yet, and prospective studies are
• A/G fat ratio this ratio is analogous to the waist/
still needed to better evaluate and validate this
hip ratio and correlates with dyslipidemia in men
method. Of note, consensuses on cutoff values for
and women, mortality in women, and risk of myo-
defining obesity categories using FMI are currently
cardial infarction [26, 27]. The A/G fat ratio may be
lacking [10].
assessed with DXA. The android region comprises
• Estimated abdominal VAT visceral fat is associated
the area located between the ribs and the pelvis, with
with cardiometabolic risk factors including diabe-
an upper demarcation at 20% of the distance between
tes mellitus, dyslipidemia, hypertension, and meta-
 Maeda et al. Advances in Rheumatology (2022) 62:11 Page 7 of 25

the iliac crest and the neck and a lower demarca- m2 in women. Further analysis of the elderly population
tion at the top of the pelvis. The gynoid region com- included in the cross-sectional study by Baumgartner
prises the area located between the hip and the upper et al. [35] found that the prevalence of low muscle mass/
thighs, with an upper demarcation below the top of sarcopenia increased with age and was associated with
the iliac crest at a distance of 1.5 times that of the higher degrees of self-reported physical disability. Since
android height. The total height of the gynoid region reference values for low ALM in young Black and White
is two times the height of the android region [28]. adults have not been established, recent consensuses
A study analyzing 2005–2006 NHANES data has recommend the definition of low ALM as the finding
shown that the correlation with cardiometabolic risk of lean mass value below the 20th percentile of distri-
factors was much stronger for the A/G percent fat bution of values for healthy young adults, with some
ratio compared with the android percent fat, gynoid authors considering the cutoff values of ≤ 7.23  kg/m2
percent fat, or BMI [29]. In contrast, another study in men and ≤ 5.67  kg/m2 in women [36, 37]. Also, stud-
showed that DXA-measured abdominal fat and A/G ies have shown significant differences in the prevalence
fat ratio were not superior to waist circumference or of low lean mass in Asian compared with White popu-
CT-measured intra-abdominal fat areas in detecting lations when the same definition was used, highlighting
metabolic risk factors in obese women [30]. The A/G the importance of reference values based on population-
ratio varies among individuals of different ethnicities specific data [10].
[31].
(b) Newman criteria
The Baumgartner criteria appear to underestimate the
Statement 2 prevalence of sarcopenia in several elderly populations
Recommendations for assessment of fat mass include the [36–40], including community-dwelling older women
fat mass index (FMI; in kg/m2), interpreted according to in Brazil [40]. This is because the formula proposed by
the NHANES III cutoff values, the estimated abdominal Baumgartner et  al. (ALM/squared height) is unable to
visceral adipose tissue (VAT; in g/cm3 if assessed with adequately identify sarcopenia in overweight and obese
a GE-Lunar device or g/cm2 if assessed with a Hologic individuals. Lean mass is generally greater in obese
device), and the android-to-gynoid (A/G) fat ratio. compared with lean individuals because both lean mass
and fat mass increase with weight gain, usually at an
3. What are the validated criteria for assessing lean mass? approximate proportion of 1:4 [41]. Due to the interre-
Several validated criteria are available for assessing lean lation between lean mass and fat mass, obese individu-
mass using DXA, e.g., the Baumgartner criteria, the als may not be considered sarcopenic, but their muscle
Newman criteria, and the Foundation for the National mass may be inadequate for their body size and physi-
Institutes of Health (FNIH) Sarcopenia Project crite- cal performance [36]. Therefore, Newman et  al. pro-
ria. However, the accuracy of DXA-measured lean mass posed ALM to be adjusted for both height and total fat
is still questionable [32, 33]. Appendicular lean mass mass. The authors’ methodology includes a statistical
(ALM), a sum of lean mass values in the upper and lower linear regression model to characterize the association
limbs, has shown a consistent correlation with muscle between ALM and height (in meters) and fat mass (in
mass throughout life [34]. The sections below include a kilograms). Linear regression residuals are used to iden-
discussion about the criteria defining sarcopenia. tify individuals whose ALM value (obtained with DXA)
is below the expected value (obtained from the equation
(a) Baumgartner criteria resulting from the model) for a given amount of fat mass.
Baumgartner et  al. [35] were pioneers in proposing a The ALM value resulting from the equation (expected
method to identify low ALM using DXA. Based on the muscle mass) must be subtracted from the ALM value
fact that absolute lean mass correlates strongly with obtained by DXA (actual muscle mass), resulting in the
height, the authors calculated the relative muscle mass ALM residual. Thus, positive residuals indicate adequate
(or ALM index [ALMI]) as ALM (in kg) divided by the muscle mass, while negative residuals indicate relative
squared height (in square meters, m­ 2), similarly to using sarcopenia [36]. Most studies adopting this classification
BMI to estimate weight excess. The authors defined low used the 20th percentile of the residuals distribution as a
muscle mass/sarcopenia as an ALMI ≥ 2 SDs below the cutoff value for sarcopenia, since the mean values of lean
sex-specific mean values from the Rosetta Study, a ref- mass and fat mass in younger populations are not consid-
erence population of 18–40-year-old adults. Therefore, ered as reference values [40, 42]. This approach correlates
according to Baumgartner et al. [35], the cutoff values for with functional limitation and inflammation markers in
sarcopenia were ALMI < 7.26 kg/m2 in men and < 5.45 kg/ older individuals [43].
Maeda et al. Advances in Rheumatology (2022) 62:11 Page 8 of 25

The prevalence of low lean mass/sarcopenia is gener- trees [CART]), the project defined specific cutoff values
ally higher when assessed using the Newman criteria for manual handgrip strength, which is associated with
compared with the Baumgartner criteria, especially in lower functional capacity defined by gait speed < 0.8 m/s,
overweight/obese populations. For example, in commu- and for ALM, measured by DXA, which in turn is asso-
nity-dwelling women over the age of 65  years in Brazil, ciated with handgrip strength. Since sensitivity analyses
the prevalence of low lean mass was 3% according to the also indicated that obesity influences the relationship
Baumgartner criteria and 19% according to the Newman between lean mass and muscle strength, the cutoff values
criteria, in which lean mass is adjusted for body fat [40]. were further adjusted for BMI.
The Newman criteria are relevant in epidemiology but Thus, clinically relevant cutoff values for low ALM
have limited applicability to individual cases in clinical adjusted for BMI (ALM [in kg] divided by BMI in [kg/
practice since they rely on the construction of a linear m2]) for elderly individuals according to sex are: < 0.512
regression model specific to each population analyzed. for women and < 0.789 for men [46]. These are the first
Thus, the equation for calculating the expected ALM criteria based on a clinically relevant outcome (gait
obtained from the regression model and, consequently, speed), which is directly related to muscle dysfunc-
the residuals (cutoff values for definition of sarcopenia) tion, and the most comprehensive criteria in terms of
vary according to the study population. For example, in populations, since they derive from multiple cohorts of
the São Paulo Aging Health (SPAH) study, the equations community-dwelling elderly individuals from different
obtained for calculating the expected ALM for elderly populations worldwide [44–48].
individuals in Brazil according to sex were as follows: [40]
  (d) Appendicular lean mass adjusted for fat mass index
Women : ALM kg = −14.51 + 17.27 Since adjusting lean mass values for adiposity appears to
× height (m) + 0.20 improve the association of lean mass with physical func-
 
× fat mass kg tion, the NHANES developed a method to define adipos-
  ity-adjusted low lean mass, i.e., the adipose mass index,
Men : ALM kg = −20.673 + 22.478
defined as total body fat mass (in kilograms) divided by
× height (m) + 0.177 squared height (in square meters) [49]. Data from whole-
body DXA scanning of 14,850 adults (20–85 years) com-
 
× fat mass kg
prised a database that generated specific SDs for ALMI
The residuals (cutoff values below which an individual (kg/m2) and FMI (kg/m2) according to sex and race
is considered sarcopenic) are − 1.45 in women and − 2.06 (Z-score) and relative to age (T-score relative to the age
in men [40, 42]. of 25  years). Sarcopenia was defined as a T-score below
− 2.0 and low lean mass for age as a Z-score below − 1.0.
(c) Foundation for the National Institutes of Health (NIH) The results of this study reinforce the importance of
criteria adjusting lean mass for body fat in addition to height, as
Both the Newman and Baumgartner criteria are based approximately twice more individuals are categorized as
exclusively on the statistical distribution of lean mass sarcopenic based on FMI-adjusted lean mass compared
within a single population [35, 36]. However, the cor- with the unadjusted criteria.
relation of lean mass with muscle strength and func-
tion, which are directly linked to physical performance, (e) Physical parameters
remains unclear. Additionally, none of these criteria have Considering that the currently validated criteria for
been validated on their ability to predict relevant clinical defining low lean mass (Baumgartner, FNIH, and New-
outcomes, such as immobility, fractures, and mortality. man) only identify an individual as having low lean mass
The FNIH has compiled and analyzed longitudinal or lean mass within normal values and are limited in ana-
data from different studies to develop definitions of low lyzing sarcopenia or cachexia, we provide a brief com-
lean mass and muscle weakness [44–48]. The project ment regarding the main physical tests used to assess
was based on the clinical paradigm of differential diag- muscle strength and performance in clinical practice.
nosis among physically limited individuals because of Lean mass can also be assessed using muscle strength
weakness and individuals who are weak from having parameters. Muscle mass and strength are known
low muscle mass. The project sought to determine the to decline with aging [50], especially after the age of
degree of muscle mass that contributes to muscle weak- 40  years, with a 5% decline at each decade after that.
ness (clinically relevant low lean mass). Gathering data After 40  years, the lean mass assessment becomes even
from eight longitudinal cohort studies and six clinical tri- more important, considering that low lean mass is an
als using a regression model (classification and regression important risk factor for sarcopenia and bone mass loss
 Maeda et al. Advances in Rheumatology (2022) 62:11 Page 9 of 25

[51]. The most affected muscles are the knee extensors disease, cognitive impairment, loss of quality of life and
and hip flexors, which are fundamental for walking and independence, and death [62, 63]. Physical performance
balance maintenance. A decline in these muscles’ func- evaluation and qualitative assessment are required for a
tion is accompanied by an increased rate of falls and complete diagnosis of sarcopenia. In the absence of this
their consequences (e.g., fractures), especially in the qualitative evaluation, the DXA report should be limited
elderly population [52, 53]. Additionally, muscle strength to mentioning the occurrence of “low muscle mass” with-
assessed by the handgrip test correlates strongly with out establishing a diagnosis of sarcopenia [10].
DXA-assessed BMD [54]. The most widely accepted consensus on the definition
Muscle strength in the trunk and upper and lower of sarcopenia is by the European Working Group on Sar-
limbs can be measured by dynamometry. Dynamometers copenia in Older People (EWGSOP), first published in
are instruments that measure isometric muscle strength 2010 and updated in 2019 (EWGSOP2) [62, 63]. Muscle
in the upper limbs (handgrip), trunk (trunk extensors), strength is the most reliable measure of muscle func-
and lower limbs (hip flexors and knee extensors). Muscle tion, and the EWGSOP2 adopts low muscle strength as
strength can also be measured by indirect and dynamic the primary parameter of sarcopenia. The occurrence of
physical fitness and functional capacity tests evaluat- sarcopenia is likely when low muscle strength is detected,
ing static and dynamic balance, mobility, and flexibility and the diagnosis is confirmed in the presence of low
[55–57]. These tests identify indirect parameters of mus- muscle quality or quantity. When low muscle strength,
cle strength, such as the chair-stand test, which evaluates low muscle quality or quantity, and low physical perfor-
lower limb strength and dynamic balance. mance are all detected, sarcopenia is considered severe
The Cybex (or isokinetic chair) can also be used to [62]. Of note, muscle strength is not dependent on mus-
assess lean mass. This method measures muscle strength cle mass alone, and the relationship between strength
and variables such as muscle power and resistance, which and muscle mass is not linear [64]. Thus, the definition
can be analyzed for each muscle group alone with fixed of sarcopenia comprises several factors, including hor-
speed and variable resistance. Disadvantages of the isoki- monal, inflammatory, biochemical, nutritional, and func-
netic chair include high cost and limited availability in tional parameters considered in terms of muscle mass
clinical practice. Thus, dynamometers are more advan- quantity and quality.
tageous than Cybex since they are portable, reliable, less The first step of the EWGSOP2 algorithm for case find-
expensive, and easy to access and handle [58]. ing is the 5-item SARC-F questionnaire, a self-reported
The long and short versions of the International Physi- screening test for sarcopenia. The SARC-F is a conveni-
cal Activity Questionnaire (IPAQ) (simplified version) is ent and inexpensive method for screening sarcopenia
widely used in Brazil and other countries for assessing and has a low-to-moderate sensitivity and a very high
physical activity level, measured as a report of the time specificity to predict low muscle strength (Table  5) [62,
spent on activities of moderate and vigorous intensity 65, 66]. The second step of the algorithm is to assess mus-
for at least 10 continuous minutes in one usual week. cle strength through handgrip strength and compare the
The questionnaire considers physical activities related to results with data from reference populations [62, 67].
work, household chores, transportation, exercise, sports, If the handgrip strength assessment is not feasible due
leisure, family care, and time spent seated [59]. to hand disability, the chair-stand test can be used as a
Table  4 shows the main physical fitness tests that can proxy to assess leg (quadriceps) muscle strength [62].
be performed, including the muscle groups evaluated and The next step is to confirm the diagnosis of sarcopenia
the tests’ objectives, methodology, and results [60, 61]. by measuring muscle mass, preferably by DXA, the rec-
These tests and questionnaires can draw a physical fit- ommended method in clinical practice. According to the
ness profile of the patients in terms of muscle strength, EWGSOP2, cutoff values for low muscle mass assessed
functional mobility, and static and dynamic balance. They by CT and MRI have not been well defined yet. The cut-
also identify the patients as sedentary or physically active off values for DXA-assessed ALMI adjusted for squared
and assess their cardiovascular fitness and risk of falls for height are < 7.0 kg/m2 in men and < 5.5 kg/m2 in women
planning physical training and treatment strategies. [62].
Assessment of sarcopenia severity should include phys-
3.1. Sarcopenia ical performance evaluation using the timed up and go
Sarcopenia, a syndrome closely related to aging, is char- (TUG) test or, preferably, the gait speed test (speed val-
acterized by progressive and generalized loss of skel- ues ≤ 0.8 m/s identify a compromised gait speed). Other
etal muscle mass and strength that increases the risk of tests that can also assess sarcopenia severity are the
adverse outcomes, including falls and fractures, physi- 400-m walk test and the Short Physical Performance Bat-
cal disability and mobility disorders, cardiorespiratory tery (SPPB) [62].
Table 4  Tests for evaluation of muscle and physical performance
Test name: assessments Device Execution Results (units of measure) Images

Handgrip test handgrip strength [54] Takei dynamometer (GRIP-THE-Takei Physical Arms parallel to the body. While standing, Best measured strength value from each
Fitness Test T.K.K.5001, Japan) the subject holds the dynamometer on one hand (in kilograms)
hand, flexing the joints of the proximal pha‑
langes of the second and fifth metacarpals.
Upon the command of “Attention!! Go!!”
from the observer, the individual applies
force against the bar of the dynamometer
with the fingers, hand palms, and thumb
Maeda et al. Advances in Rheumatology

base, trying to pull the bar up with maxi‑

mum isometric strength for 5 s. Two move‑
ments are recorded for each hand
Trunk extension trunk muscle strength Lafayette Manual Muscle Test System, Model Subject in the prone position on an elevated Peak strength value 5 s after the subject
[57–59] 01163. Lafayette Instrument, IN, USA examination table, chin beyond the limit of pushes against the device. The result
the table to avoid improper cervical spine obtained is the average of the 3 attempts (in
(2022) 62:11

positioning, arms extended alongside the kilograms)

body, and palms of the hands parallel to the
hips. The observer places the dynamom‑
eter in the midline of the eighth thoracic
vertebra between the inferior angle of both
scapulae. Upon command, the subject per‑
forms maximum trunk extension (attempt‑
ing to raise the chest from the table) for 5 s,
trying to "push" the observer’s hand. The
observer should not offer resistance to the
subject’s movement. The test is performed
3 times
Hip flexor strength pelvitrochanteric muscle Lafayette Manual Muscle Test System, Model Subject seating on a chair with lumbar Peak strength value 5 s after the subject
strength [57–59] 01163. Lafayette Instrument, IN, USA support, hip and knees flexed at 90°, and pushes against the device. The final result
both hands crossed in front of the body. The is an average of the 3 measurements (in
observer positions the device perpendicular kilograms)
to the thigh, 5 cm above the patella’s base,
while an assistant stabilizes the subject’s
shoulders to ensure stable movement and
avoid lateral or anteroposterior compensa‑
tions. After a command to start, the subject
performs hip flexion by raising the knee
from the chair. The test is performed 3 times
Page 10 of 25
Table 4  (continued)
Test name: assessments Device Execution Results (units of measure) Images

Knee extensor strength quadriceps strength Lafayette Manual Muscle Test System, Model Subject seating on a chair with lumbar sup‑ Peak strength value 5 s after the subject
[57–59] 01163. Lafayette Instrument, IN, USA port, hip and knees flexed at 90°, and arms pushes against the device. The final result
crossed in front of the chest to avoid touch‑ is an average of the 3 measurements (in
ing the side of the chair. The dynamometer kilograms)
is positioned 5 cm above a medial point
between the medial and lateral ankle
malleoli. The observer voices a command to
start the test, and the subject extends the
Maeda et al. Advances in Rheumatology

knee at maximum effort against resistance

by the dynamometer for 5 s. The test is
performed 3 times
Elbow flexion strength upper limb strength Stopwatch and dumbbell (2 kg for women Subject seated on a chair, with back straight Total number of movements performed in
[54, 55] and 4 kg for men) against the back of the chair and feet flat on 30 s (in number of maximum repetitions)
the floor, holds the dumbbell on his or her
(2022) 62:11

side with the hand closed. The test begins

with the subject’s arm extended downwards
alongside the chair and perpendicular to
the floor. At the command of "Attention!!
Now!!," the subject rotates the palm of the
hand upwards while flexing the arm until
completing the entire range of movement,
then returns to the initial position with the
elbow fully extended. Upon returning to the
initial position, the subject must hold the
dumbbell with a closed hand
Chair to stand lower limb strength, dynamic Stopwatch and armless chair Subject seated on a chair, back straight Total number of movements performed in
balance [54, 55] up, feet flat on the floor, and arms crossed 30 s (in number of maximum repetitions)
in front of the chest. Upon command, the
subject stands up fully and then returns to
the initial seated position
Stationary march gait, cardiorespiratory Stopwatch and measuring tape While the subject is standing, the observer Greatest number of times in 2 min that the
endurance, general motor coordination, measures with a measuring tape the dis‑ foot touches the floor after the knee eleva‑
dynamic balance, and lower limb strength tance between the subject’s anterosuperior tion (in number of maximum repetitions)
[54, 55] iliac crest and patella’s base. The midpoint
of this measurement is marked (on a wall,
with the observer holding up his/her hand,
or with a rope tied across two chairs) and
the subject is instructed to elevate the
knees to the marked height while walking
in place. The observer times the executions
with a stopwatch, counting the number of
times that the subject’s right foot touches
the floor
Page 11 of 25
Table 4  (continued)
Test name: assessments Device Execution Results (units of measure) Images

Timed up and go functional mobility [60] Stopwatch and armless chair Subject seated on a chair with back sup‑ Shortest duration of two executions (in
ported against the chair. Upon the com‑ seconds)
mand of "Attention!! Go!!! from the observer,
the subject stands up and walks 3 m up to
a predetermined spot marked on the floor.
The subject turns around, walks back to the
chair, and sits again with the back supported
against the chair. The subject should walk
Maeda et al. Advances in Rheumatology

at maximum speed but without running.

The time is recorded using a stopwatch
and an interval of 1 min is given between
executions
Single-leg stance (Flamingo) static balance Stopwatch Subject standing upright, with hands on the Longer duration or completed test (in
[54, 55] hips, is instructed to look at a fixed point at seconds)
(2022) 62:11

a wall (2 m ahead) and stand on one leg by

flexing back the opposite leg to the level
of the knee, and attempting to maintain
the position for at least 30 s. The observer,
on the subject’s side, starts the stopwatch
when the subject’s foot is raised from the
floor and stops it as soon as the foot touches
the floor again, even if before 30 s. The test is
performed 3 times
Five times chair to stand reaction time, lower Stopwatch, armless chair Subject seated on a chair, back straight Shortest duration of the two executions (in
limb strength, dynamic balance, risk of falls against the back of the chair, and arms seconds)
[61] crossed in front of the chest. Upon the com‑
mand "Attention!! Now!!," the subject must
stand up and sit 5 times as fast as possible.
The test is performed 2 times
Gait speed functional mobility, dynamic bal‑ Stopwatch, ground demarcation of a Subject standing, at command, must walk Shortest duration of the three executions (in
ance, strength [55] 3.33-m × 33.3-cm rectangle on the demarcated rectangle. In the first seconds)
part of the test, must walk at a normal
speed, and in the second part, must walk
quickly. Each test is performed 3 times at
each stage
Cone test functional mobility, muscle Stopwatch, chair, ground demarcation Subject seated on the chair with supported Shortest duration of the two executions (in
strength, dynamic balance [55] with a 6-m × 4-m rectangle, two cones back, upon command from the observer the seconds)
positioned on a straight line 3 m in front of subject stands up and walks swiftly, going
a chair around the first cone, and returns to sit on
the chair with feet on the floor. The subject
stands up again and walks in a straight line
to go around the other cone, and returns to
the chair. The execution is done twice with
a 1-min break
Page 12 of 25
Table 4  (continued)
Maeda et al. Advances in Rheumatology

Test name: assessments Device Execution Results (units of measure) Images

Short Physical Performance Battery (SPPB) Stopwatch, armless chair The balance test is evaluated in 3 positions Each test receives its own score. For the bal‑
timed test evaluating balance, walking for 10 s: two feet together, one foot slightly ance test, the individual receives a score of 1
speed, and lower limb strength (5-time chair in front of the other (semi-tandem), and if able to stay in position for 10 s and a score
stand test) [62] one foot in front of the other (tandem). In all of 0 if unable to stay in position. For the test
(2022) 62:11

positions, the subject is standing with eyes of the third position of the feet (tandem),
open. For the walking test, the subject walks the individual receives a score of 1 if able to
400 m (marked on the floor with a tape) at stay in position for 3–9.9 s, a score of 2 if able
maximum speed. For the chair stand test, to stay in position for 10 s, and 0 if unable
the subject sits on a chair without back sup‑ to stay in position. For the gait speed test,
port and stands up 5 times with the arms the subject receives a score between 1 and
crossed in front of the chest 4 depending on the reached speed: the
greater the speed, the higher the score. For
the chair stand test, the subject receives a
score between 1 and 4 depending on the
time taken to stand up: the lower the time,
the greater the score; the score is 0 if the
subject is unable to perform the test
400-m walk gait, cardiovascular endurance Space for a 400-m walk Subject standing, must complete 20 laps of Total walking duration and maximum ­VO2
(maximum ­VO2 consumption), mortality 20 m each, walking as fast as possible. Rest‑ consumption (calculated using tables and
prediction [57] ing is allowed twice during the test formulas) in the last 10 s. Unit of measure:
min/sec
6-min walk gait, cardiovascular endurance Space of about 50 m Subject standing, must walk as fast as Distance walked (in meters)
[55] possible (without running) during 6 min a
45.72-m rectangular course demarcated at
every 4.57 m
Page 13 of 25
Maeda et al. Advances in Rheumatology (2022) 62:11 Page 14 of 25

Table 5  SARC-F questionnaire

Component Question Scoring

Strength How much difficulty do you have in lifting and carrying 4.5 kg? None = 0
Some = 1
A lot or unable = 2
Assistance in walking How much difficulty do you have walking across a room? None = 0
Some = 1
A lot, use aids, or unable = 2
Rise from a chair How much difficulty do you have transferring from a chair or bed? None = 0
Some = 1
A lot or unable without help = 2
Climb stairs How much difficulty do you have climbing 10 flights of stairs? None = 0
Some = 1
A lot or unable = 2
Falls How many times have you fallen in the past year? None = 0
1–3 falls = 1
≥ 4 falls = 2
A score ≥ 4 is suggestive of sarcopenia

Table 6 Normative cutoff values for handgrip strength after excluding some conditions such as starvation, mal-
according to the FNIH Sarcopenia Project absorption, major depression, lipoatrophy, hyperthyroid-
Handgrip strength Normal Intermediate Weak ism, and age-related muscle loss, cachexia is diagnosed in
(kg) adults by weight loss ≥ 5% (corrected for fluid retention)
over ≤ 12  months (or BMI < 20  kg/m2, if undocumented
Men ≥ 32 26–31.9 < 26
weight loss) in the presence of underlying illness, plus
Women ≥ 20 16–19.9 < 16
three of the following criteria: [69].

More recently, the FNIH Sarcopenia Project defined • Decreased muscle strength (lowest tertile).
normative cutoff values for handgrip strength in men and • Fatigue.
women older than 65 years (Table 6) [44, 45]. • Anorexia (total caloric intake < 20  kcal/kg of body
According to the Sarcopenia Definition and Outcomes weight/day, < 70% of the usual food intake), or lack of
Consortium, DXA-assessed lean mass measurements appetite.
should not be included in the definition of sarcopenia • Low FFM index.
as they are not good predictors of self-reported mobil- • Abnormal biochemical tests: anemia (< 12 g/dL), low
ity limitations and other health-related outcomes such as serum albumin (< 3.2 g/dL), and increased inflamma-
falls, hip fracture, and mortality [68]. tory markers (C-reactive protein > 5.0 mg/L or inter-
Still lacking a clear definition, sarcopenic obesity is leukin-6 > 4.0 pg/mL).
a distinct condition marked by fat infiltration of mus-
cle and associated with decreased physical function and A cachexia score (CASCO) has been designed for can-
increased risk of mortality and physical disability [62]. cer patients, but this subject is outside of the scope of the
present document [70].
3.2. Cachexia
Cachexia is a complex metabolic syndrome associated Statement 3
with an underlying disease and characterized by loss of Several validated criteria are available for assessing
muscle with or without loss of fat mass. Cachexia is asso- appendicular lean mass using DXA, including the Baum-
ciated with cancer, congestive cardiomyopathy, end-stage gartner criteria, the Newman criteria, and the Foun-
renal disease, and other diseases, and is often associated dation for the National Institutes of Health (FNIH)
with inflammation, insulin resistance, anorexia, and mus- Sarcopenia Project criteria.
cle proteolysis [69]. Thus, most individuals with cachexia Lean mass can also be assessed using muscle strength
also have sarcopenia, while most individuals with sarco- parameters, including dynamometry and indirect and
penia may not have cachexia. Sarcopenia is one of the dynamic physical fitness and functional capacity tests
elements of the definition proposed for cachexia [69]. evaluating static and dynamic balance, mobility, and flex-
According to the Cachexia Consensus Working Group, ibility, such as the chair-stand test and isokinetic chair.
 Maeda et al. Advances in Rheumatology (2022) 62:11 Page 15 of 25

Section II: Reporting • The precision error supplied by the manufacturer

4. Which parameters should be included in the DXA body should not be used.
composition report? • Each DXA device should have its own precision error
For adults older than 20  years, report whole-body determined and LSC calculated. The LSC is calcu-
(including head) values of: lated the same way as done for BMD sites accord-
ing to ISCD protocol: 15 patients three times, or 30
• Anthropometry weight (kg), height (m), and BMI (kg/ patients two times, repositioning the patient after
m2). each scan, calculate the root mean square standard
• Bone mass compartment BMD (g/cm2), BMC (g), deviation (RMS-SD) for the group and finally calcu-
and Z-score (in SDs). late the LSC with 95% confidence intervals for the
• Fat mass compartment total fat mass (in kilograms), group. A change is considered significant when the
percentage of fat (in %), FMI (total fat mass/height2, difference between the previous and the new values
in kg/m2), A/G ratio, and VAT (in g and c­ m3). of total fat mass and total lean mass is above the LSC.
• Lean mass compartment total lean mass (kg), ALM • The technician should perform an in  vivo precision
(kg), ALMI (adjusted by height [ALM/height2] and assessment for all body composition values of inter-
adjusted by BMI in patients over 65 years old [ALM/ est using patients representative of the population
BMI]); should be cited in all reports [10, 46]. of patients from the clinic. This procedure should
be repeated when the technician’s level changes. The
Figure 1 shows a DXA report proposed by ABRASSO. minimum acceptable precision values for an individ-
ual technician are 3%, 2%, and 2% for total fat mass,
Statement 4 total lean mass, and percent fat mass, respectively. If
For adults older than 20  years, report whole-body more than one technician works on the same DXA
(including head) values of: device, an average precision error combining data
from the entire team should be used to establish pre-
• Anthropometry weight (kg), height (m), and BMI (kg/ cision error and LSC for the facility, providing the
m2). precision error for each technician is within a pre-
• Bone mass compartment BMD (g/cm2), BMC (g), established range of acceptable performance.
and Z-score (in SDs). • The precision assessment should be repeated when
• Fat mass compartment total fat mass (in kilograms), a new DXA system is installed, and cross-calibration
percentage of fat (in %), FMI (total fat mass/height2, should be done by a technician by performing 10
in kg/m2), A/G ratio, and VAT (in g and c­ m3). phantom scans with repositioning, before and after
• Lean mass compartment total lean mass (kg), ALM a hardware change. If a difference greater than 2% in
(kg), ALMI (adjusted by height [ALM/height2] and the mean fat mass, percent fat mass, or lean mass is
adjusted by BMI in patients over 65 years old [ALM/ observed, the manufacturer should be contacted for
BMI]). service [71].

The imprecision of body composition measurements,

5. What should be taken into account regarding quality especially in subregions, can be much larger and more
control, accuracy, and least significant change (LSC)? variable than for regional BMD scans. Precision may
The quality control program should adhere to the manu- vary according to device and scan mode, subregion and
facturer’s guidelines for system maintenance. According compartment, body habitus, and age. Caution is advised
to the ISCD recommendations, quality control should when considering soft tissue results from subregions of
include: [71]. whole-body scans. In general, lean mass precision is bet-
ter than fat mass precision. A trend has been observed
• Periodic (at least once a week) phantom scans for any for greater precision for recent models of Hologic and
DXA system as an independent assessment of system GE-Lunar systems [28].
calibration. Although body composition phantoms are part of any
• Plotting and reviewing of data from calibration and body composition teaching course, they were not com-
phantom scans. pletely addressed in this manuscript because they are
• Establishment and enforcement of corrective action unavailable in most centers. The procedure of quality
thresholds that trigger a call for service. control in bone mass studies follows a schedule set by
the equipment manufacturer and uses either phantoms
About precision, ISCD recommends that:
Maeda et al. Advances in Rheumatology (2022) 62:11 Page 16 of 25

Patient´s Full Name:____________________________ Date of Birth:______________ Scan Date:___________

Body Composition by DXA – 3-Compartment Model

Scan performed in a ________ dual x-ray absorptiometry (DXA) device.

1. Anthropometric data:

Weight (kg): ______ Height (m): ________ BMI (kg/m2): ________

Conclusion: According to the WHO criteria, the patient’s BMI is compatible with (underweight / eutrophy / overweight / obesity I, II
or III).

2. Bone Mass Compartment

Bone Mass Result (Z-score) Reference

Bone mineral density (g/cm2)
Bone mineral content (g)

The whole-body bone mineral content should not be used as a stand-alone marker of bone health. For comparison with BMD
reference values and fracture risk, the lumbar spine, hip, or forearm skeletal site should be evaluated.

3. Fat Mass Compartment

Fat Mass Result Reference

Total fat mass (kg)
Total body fat mass (%)
Fat mass index – FMI (kg/m2) 3-6 kg/m2 (men)
5-9 kg/m2 (women)
A/G ratio < 1.0
VAT (g and cm3)

Conclusion: The fat mass index is compatible with (severe, moderate or mild fat deficit / normal / overweight / obesity I, II, or III)
according to sex, with (android / gynoid) predominance.

4. Lean Mass Compartment

Lean Mass Result Reference

Total lean mass (kg)

Appendicular lean mass (kg)
Appendicular lean mass index
Adjusted by squared height (kg/m2) ≥ 7.26 kg/m2 (men) e ≥ 5.45 kg/m2 (women)
Adjusted by BMI (patients ≥ 65 years of age) ≥ 0.789 (men) e ≥ 0.512 (women)

Conclusion: Lean mass is (below / adequate for) the expected values according to sex.

Coefficients of variation (CV) for lean, fat, and bone mass are shown below, as an example:

Variable TFC (kg) TFC (%) TLM (kg) BMD (g/cm2) BMC (g)
CV 1.62% 1.53% 1.14% 0.67% 1.72%
TFC – total fat content (% and kg), TLM – total lean mass, BMD – bone mineral density, BMC – bone mineral content

Fig. 1  Proposed body composition report by ABRASSO

 Maeda et al. Advances in Rheumatology (2022) 62:11 Page 17 of 25

with density similar to lean and fat mass (GE-Lunar) or of body composition assessed by DXA (GE-Lunar Sys-
an internal calibration system through lean and fat tissue tem) and adjusted to BMI differed between the study
equivalents (Hologic). population and other ethnic groups [77]. Compared
with the NHANES III data assessed with DXA Hol-
Statement 5 ogic [13], fat and lean mass index tended to be lower
The quality control program should adhere to the manu- in men in Brazil compared with those in North Amer-
facturer’s guidelines for system maintenance. The quality ica. When the authors used NHANES III DXA data
control should include: converted from Hologic to Lunar-GE [78], the values
between the populations remained different for fat
• Periodic (at least once a week) phantom scans for any mass, especially in older age groups, but became simi-
DXA system as an independent assessment of system lar regarding lean mass [77]. The authors reported FMI
calibration. values similar to those for the US population (3–6  kg/
• Plotting and reviewing of data from calibration and m2) as normative for men in Brazil [77]. However, the
phantom scans. cutoff value for ALMI recommended by Baumgartner
• Establishment and enforcement of corrective action et al. [35] was higher than that reported in the Brazilian
thresholds that trigger a call for service. study (7.26 kg/m2 for individuals aged 20–29 years and
• The precision error supplied by the manufacturer 6.6 kg/m2 for normal BMI).
should not be used. Carvalho et  al. evaluated 689 adults aged 20–59  years
• Each DXA device should have its own in  vivo pre- in Brazil using DXA (GE-Lunar) to establish percentile
cision error determined and LSC calculated for all curves for measures and indices of body composition by
body composition variables. age and sex. The cutoff values for ALMI, derived from
204 men and 221 women aged 20–39  years and consid-
ering an SD of − 2.0, were 6.34  kg/m2 and 4.45  kg/m2,
6. What are the differences between normative data
respectively. [79].
for the Brazilian compared with other populations?
Barbosa-Silva et  al., in turn, adopted a − 1 SD cutoff
There is currently a lack of normative data regard- value relative to a reference young population to deter-
ing body composition assessed by DXA for the Brazil- mine low ALMI in elderly individuals, different than the
ian population or comparing such data with those from EWGSOP recommendation. The mean ± SD values of
other populations. Data from a few Brazilian studies ALMI in young adults obtained by DXA in a follow-up
(Brazilian Osteoporosis Study [BRAZOS] and from the study conducted in 2012 (when the participants were
Brazilian Institute of Geographic and Statistics [Insti- 30  years old) were 8.76 ± 0.99 and 6.44 ± 0.82  kg/m2 for
tuto Brasileiro de Geografia e Estatística, IBGE]) [72, 73] men and women, respectively. Based on that, cutoff val-
and international studies [74, 75] have reported a simi- ues of 7.77  kg/m2 and 5.62  kg/m2 (− 1 SD) were deter-
lar increment of overweight and obesity rates defined by mined for men and women, respectively. Barbosa-Silva
BMI categories. et al. also proposed a combination of calf circumference
A Brazilian study conducted by Sousa et  al. evaluated (≤ 34 cm for males and ≤ 33 cm for females) and SARC-
body composition data of 500 women older than 20 years F into a modified index that significantly improved the
(245 of whom were premenopausal) with BMI between performance of SARC-F for screening sarcopenia [66, 80,
18.5 and 34.9 kg/m2 using DXA (GE-Lunar) and reported 81].
a bimodal variation for body fat, with increasing values Machado et  al. followed up 433 community-dwelling
until the age of 50–59  years followed by a reduction to women (mean age 72.8 ± 4.7 years) and found 28 incident
the lowest levels at the age of 80 and above. The authors nonspinal osteoporotic fractures during a mean period
also reported that FMI was consistently higher in Afri- of 4.3 ± 0.8 years. After adjustments for age, race, previ-
can American and Hispanic American women compared ous fractures, and BMD, the authors found a significant
with Brazilian women, and that ALMI (ALM/height2, association between the participants’ VAT (mass, area,
Baumgartner criteria) was consistently lower in women volume) and the incidence of nonspinal fractures among
in Brazil compared with those in the US, regardless of age nonobese elderly women, suggesting a potential negative
or ethnicity. Lean body mass showed minor deteriora- effect of visceral adiposity on bone health in this particu-
tion and decreased from the ages of 50–59 years onward, lar group [82].
reaching the lowest values in older (≥ 80  years) women Additional population studies including functional
[76]. analyses are still needed to define cutoff thresholds for
In another Brazilian study, Ushida et al. assessed 403 sarcopenia and low lean mass in the Brazilian population.
men older than 20 years and found that the distribution
Maeda et al. Advances in Rheumatology (2022) 62:11 Page 18 of 25

Statement 6 in previous sessions also apply to the pediatric popula-

Based on Brazilian normative database studies in adult tion [86, 87].
men and women, the Brazilian population, compared DXA is considered the method of choice for quantita-
with other populations, has some significant differences tive assessment of body composition in pediatrics due to
in body composition parameters, particularly regarding precision, accuracy, reproducibility, low radiation doses
appendicular lean mass adjusted for height. Cutoff values (1–5  µSv), accessibility, cost-effectiveness, and practica-
of 7.77 kg/m2 and 5.62 kg/m2 (− 1 SD) are suggested for bility. However, this method has considerable limitations,
men and women. The combination of calf circumference including age (the child must be older than 2–5  years,
(≤ 34 cm for males and ≤ 33 cm for females) and SARC-F depending on the densitometer software), lack of integra-
into a modified index significantly improves the perfor- tive sex/age/pubertal stage and population-specific nor-
mance of SARC-F for screening sarcopenia. mative data, and lower precision and accuracy for VAT
assessment [1].
Section III: Special situations Interpretation of pediatric DXA data may be challeng-
7. What is the application of body composition assessment ing due to physiological changes in body composition
in pediatrics? during growth. This is especially critical during adoles-
In pediatrics, body composition assessment is important cence, when each pubertal stage has different patterns
in clinical practice and research settings both as a routine of lean, fat, and bone mass distribution and acquisition,
follow-up and in specific diseases. Numerous conditions associated not only with chronological age, but also
may potentially interfere with body compartment distri- dependent on bone age and hormonal and metabolic sta-
bution (lean, fat, and bone mass), including exogenous tus [88, 89].
and endogenous overweight and obesity, environmental Establishing reference data for specific populations is
and disease-related undernutrition, anorexia, chronic fundamental for an accurate analysis of DXA-acquired
drug therapy (e.g., corticosteroids, chemotherapy), and body composition, considering the impact of ethnicity,
chronic diseases (e.g., systemic inflammatory disorders, diet, sex, and pubertal stage on body compartment pro-
inborn errors of metabolism, muscular dystrophies, file and distribution [90, 91]. However, the number of
and endocrine, gastrointestinal, heart, and pulmonary available regional/country reference data on the quan-
diseases). tification of pediatric body composition by DXA is lim-
The most frequently used parameter for estimating ited worldwide; among the few data available for children
body composition in routine practice in pediatrics is BMI. and adolescents are those from the US [84], Chile [92],
However, this index has some limitations, for example, Argentina [93], UK [94], China [95], and India [96].
it is unable to identify the percentage of distribution of The most robust pediatric DXA database currently
each body compartment. In some clinical conditions, it is available is from the 1999–2004 NHANES, compiling
desirable to differentiate and quantify the different body data on BMC, areal mineral density, and lean and fat
compartments for diagnostic purposes, define therapeu- mass of 412 boys and 931 girls aged 8–19 years in the US
tic interventions, or evaluate the impact of a procedure [13]. Specifically in Brazil, a recent study reported age-
on the overall health of children and adolescents [83, 84]. and sex-specific DXA-acquired reference data for lean
For instance, it is important to evaluate the extension of and fat mass based on the evaluation of 541 adolescents
fat mass loss in lipodystrophy syndromes. On the other (aged 12–17 years, 170 girls) in the state of Parana [97].
side, greater BMI may signal excessive weight that may The Brazilian population has unique features compared
not necessarily result from excess fat, e.g., a heavily mus- with other populations. Brazil is a large country with spe-
cular adolescent. In other scenarios, differentiating sub- cific genetic background and phenotypic patterns clus-
cutaneous from VAT in a pediatric patient with obesity tered in some regions, and an overall population with
leads to a more emphatic approach to prevent the onset substantial ethnic miscegenation, all of which affect body
of cardiometabolic disorders. composition. Based on that, a nationwide reference data-
Other anthropometric measurements, including skin- base of DXA-assessed body composition representative
folds, waist-hip ratio, and abdominal, arm and neck cir- of Brazilian children and adolescents would be desirable
cumferences, are technically easy to perform but are [98].
highly observer-dependent and may also present limita- In pediatrics, assessment of body composition using
tions in interpretation [85]. DXA should follow the same ISCD standards as those
Body composition assessment in pediatrics can also be used for adult scanning: fasting, adequate hydration,
performed by other methods, including DXA, CT, MRI, empty bladder, clothing, and body positioning [28].
plethysmography, and BIA, as done in other stages of life. Research studies have contributed to a better
The advantages and limitations of each method described understanding of the physiological changes in body
 Maeda et al. Advances in Rheumatology (2022) 62:11 Page 19 of 25

composition during growth. However, body composi- 8. What is the clinical application of body composition
tion assessment using DXA should be performed judi- assessment in patients infected with HIV?
ciously in children and adolescents, considering that no Adequate nutritional status is essential for patients
consensually established references are currently avail- infected with HIV since compromised nutrition in this
able for precise and accurate quantification of different population has been negatively associated with immune
body compartments in this population. Still, the cur- system dysregulation, disease progression, morbidity,
rently available data may be applicable in clinical practice and mortality. Due to complex and unclear mechanisms,
to evaluate the impact of diseases on body composition, patients infected with HIV may present body com-
offer parameters to define specific interventions on nutri- position changes even without weight loss [104, 105].
tional health, and evaluate the impact of clinical proce- Therefore, noninvasive methods for body composition
dures on global health and growth [86, 99, 100]. assessment are useful to monitor and identify possible
Increasingly more studies are providing new data on changes in this population [105].
the assessment of body composition by DXA in ado- HIV-associated lipodystrophy is a condition charac-
lescent athletes, elucidating some of the mechanisms terized by abnormal body fat redistribution. Subtypes of
driving the impact of diet and physical activity on body this condition included lipoatrophy (peripheral fat wast-
compartments. Results from such studies can optimize ing, with subcutaneous fat loss in the face, arms, legs, and
the guidance for sports performance and recognize situa- buttocks), lipohyperthophy (abdominal visceral fat accu-
tions that could potentially trigger health risks [101–103]. mulation, neck enlargement, gynecomastia, and develop-
Currently available pediatric data still do not support ment of dorsocervical fat pad or “buffalo hump”), and a
the analysis of body composition by DXA for population phenotype of mixed (combined) lipodystrophy, with the
screening or comprehensive monitoring of clinical condi- clinical presentation of both lipoatrophy and lipohyper-
tions involving the risk of metabolic and nutritional dis- trophy [104–106]. Although first described in adults,
orders. The use of the method for assessing nutritional HIV-associated lipodystrophy can also occur early in life
disorders, if carried out, should be judicious, observing [107, 108].
the clinical context of each patient [86]. Some studies have reported early body composition
In summary, understanding the advantages and limi- changes detectable by DXA in pediatric patients infected
tations of body composition analysis by DXA and other with HIV, even in those without typical clinically visual
methods in pediatrics and the changing nature of body signs of lipodystrophy [109, 110]. A study following HIV-
composition during childhood and adolescence are infected children into adolescence reported progressive
essential steps for choosing the best measurement tech- subcutaneous fat loss and greater accumulation of vis-
nique for each individual, population, or clinical issue in ceral adiposity in those with lipodystrophy [111]. BIA is a
research settings, as well as the correct interpretation of cost-effective method to predict lean body mass and total
the obtained data. body fat in HIV-infected children but requires specific
prediction equations [105, 112], is unable to assess body
Statement 7 fat redistribution, and may be imprecise in patients with
Numerous conditions may potentially interfere with body lipodystrophy [104]. A Brazilian study comparing body
compartment distribution (lean, fat, and bone mass), composition assessment with BIA versus DXA in pre-
including exogenous and endogenous overweight and pubertal HIV-infected children showed a high homo-
obesity, environmental and disease-related undernutri- geneity between both methods for total body fat but no
tion, anorexia, chronic drug therapy (e.g., corticoster- concordance regarding FFM [112].
oids, chemotherapy), and chronic diseases (e.g., systemic Anthropometric measurements are also useful in
inflammatory disorders, inborn errors of metabolism, assessing body composition in HIV-infected persons
muscular dystrophies, and endocrine, gastrointestinal, [105]. The trunk-to-arm skinfold ratio (the sum of the
heart, and pulmonary diseases). The most frequently subscapular and suprailiac skinfolds divided by the
used parameter for estimating body composition in rou- sum of the biceps and triceps skinfolds) may be a useful
tine practice in pediatrics is the BMI. parameter of body fat redistribution [107] and correlates
Interpretation of pediatric DXA data may be challeng- inversely with the limb-to-trunk fat ratio (the sum of the
ing due to physiological changes in body composition fat mass in the arms and legs divided by the fat mass in
during growth, particularly in the absence of Brazilian the trunk) obtained by DXA [108]. Of note, ratios such
normative reference data for children and adolescents. as trunk/limb fat, trunk/leg fat, fat mass ratio, and even
Thus, the NHANES III database is also recommended for trunk or limb fat as a percent of total fat are unable to
use in pediatric patients (ages 5–19 years) in Brazil. fully differentiate between peripheral fat loss and central
Maeda et al. Advances in Rheumatology (2022) 62:11 Page 20 of 25

fat gain; however, they may be useful and must be inter- gender identity of the individual (both male and female
preted with caution [10]. databases if requested); (4) in gender-nonbinary indi-
Bone metabolism in HIV-infected persons can be viduals, the normative database that matches the sex
affected by several factors, including antiretroviral drugs recorded at birth should be used [71].
and the infection itself. Low BMD for chronologic age is Several factors can interfere with bone, lean, and fat
reported in HIV-infected children and adolescents and mass in transgender individuals, e.g., the time elapsed
may result in suboptimal peak bone mass in adulthood since gonadectomy and beginning of hormone therapy,
[113, 114]. Additionally, adults with HIV have a high risk use of GnRH analogs, adherence or use of inadequate
of osteopenia, osteoporosis, and low BMD [115, 116]. CSHT doses, presence of other risk factors for bone loss,
According to the American Dietetic Association (ADA) associated diseases, and medications (e.g., corticoster-
[105], there is plenty of evidence for assessing body oids) [71, 119]. However, no consistent data about body
composition in HIV-infected children, adolescents, and composition in transgender individuals are available at
adults. The Adult Official Positions of the ISCD recom- this time.
mend DXA total body composition with regional analy-
sis to evaluate fat distribution in patients with HIV using
antiretroviral drugs associated with a risk of lipoatrophy Statement 9
(currently stavudine and zidovudine) [71]. The Osteo Consistent data on body composition assessment in
Renal Exchange Program (OREP), which addresses bone transgender individuals are currently unavailable. Until
disease in HIV-infected patients, recommends DXA to studies with more consistent data are published, we rec-
be performed in the following adults with HIV infec- ommended the calculation of T-scores using a uniform
tion: men aged ≥ 50  years, postmenopausal women, and Caucasian (non-race adjusted) female normative data-
patients at high risk of falls, with a history of fragility base for all transgender individuals of all ethnic groups
fracture, or receiving chronic corticosteroid treatment and all transgender individuals aged 50  years or older,
[117]. regardless of hormonal status. Z-scores should be cal-
culated using the normative database that matches the
gender identity of the individual (or based on both male
Statement 8
and female databases, if requested by the physician). In
Body composition assessment is recommended in
gender-nonbinary individuals, the normative database
patients infected with HIV for monitoring of body com-
that matches the sex recorded at birth should be used.
position changes related to the disease and adverse
effects associated with antiretroviral therapy, particularly
abnormal body fat redistribution in the HIV-associated Perspective
lipodystrophy spectrum. 10. What is the role of DXA in veterinary medicine
The following parameters may be useful for assessing and zootechnics?
the presence of lipodystrophy in HIV-infected patients: The topic of DXA use in animal studies has not been
limb-to-trunk fat ratio, trunk/leg fat ratio, and fat mass explored much in the literature and brings an interest-
ratio. ing perspective regarding other innovative applications of
this technique.
9. How should body composition be assessed The relatively recent introduction of DXA in veteri-
in transgender individuals? nary medicine and animal sciences demonstrates the vast
A systematic review has evaluated the bone mass effects potential of applicability of the method in these areas.
of long-term cross-sex hormone therapy (CSHT) in Historically, Kronacher and Hogreve were pioneers in
transgender individuals. However, the conclusions had using noninvasive diagnostic methods in animals, using
moderate- to low-quality evidence due to studies with an x-ray to analyze the pelvis in pigs [120]. In  vivo body
observational design, small sample sizes, and variations composition measurement by DXA has been obtained
in hormone therapy protocols [118]. from porks [121], broilers [122], and sheep [123]. There
According to the ISCD statement: (1) gender data are also examples of the application of DXA as a reliable
should be obtained on the intake questionnaire; (2) technique and alternative to traditional methods in the
T-scores should be calculated using a uniform Cauca- evaluation of body composition in ovine carcasses [124],
sian (non-race adjusted) female normative database for pigs [125, 126], broilers [127], and beef carcass sides and
all transgender individuals of all ethnic groups and be primal cuts [128].
used in all transgender individuals age 50 years or older, Mawby et  al. [129] used DXA to analyze the body
regardless of hormonal status; (3) Z-scores should be cal- composition of dogs with obesity due to malnutrition.
culated using the normative database that matches the
 Maeda et al. Advances in Rheumatology (2022) 62:11 Page 21 of 25

German et  al. [130] evaluated the body condition score measurement has some limitations and is not included
(BCE) and DXA scanning in dogs to estimate changes in in the definition of sarcopenia issued by some medical
weight and body composition and found that the animals societies.
had increased lean mass, lower fat mass, and decreased Special care is recommended regarding quality control
weight and BCE. Reference values for body composition and LSC calculation to allow for accurate and reproduc-
and age and gender differences can be obtained from ible measurements and longitudinal control when the
healthy adult cats using DXA scanning; these values patient’s condition requires follow-up assessments.
allow for monitoring of nutritional status, assessment of Data interpretation in pediatric patients is challeng-
skeletal muscle development, and investigation of meta- ing, partially due to continuous physiological changes in
bolic and endocrine disorders [131]. The reference values body composition during linear growth, especially during
also have the potential to evaluate the effectiveness of adolescence. DXA-assessed body composition in pediat-
feeding interventions on the amount of lean and fat mass, rics is particularly interesting in chronic diseases, mainly
for example, with the commercial purpose of selling ani- those involving nutritional disorders and muscle mass.
mals with less fat and greater lean content or vice versa. Trunk/limb fat ratio, trunk/leg fat ratio, and fat mass
Of note, the recent use of DXA in veterinary medicine ratio assessed by DXA may be useful in HIV-infected
has proven to be valid, reliable [123, 125, 126, 129], and patients to assess the presence of lipodystrophy.
reproducible, confirming that DXA is an excellent poten- No consistent data on body composition assessment in
tial instrument for applications in animal health and pro- transgender individuals are available currently.
duction. However, reference values at different animal In veterinary medicine, DXA has been proven valid,
ages are still required to monitor body changes during reproducible, and a potential tool for assessing animal
lactation, analysis of data after use of nutritional addi- health.
tives, monitoring of dietary regimes, or even experimen-
Acknowledgements
tally- or naturally-induced obesity. Brazilian Association of Bone Assessment and Osteometabolism (ABRASSO;
Brazilian Society of Bone Metabolism), and Milena Braga-Basaria, MD (Voxmed
Medical Communications) for critical review and suggestions of improve‑
Statement 10 ments to the manuscript.
DXA can be used in veterinary medicine and animal
sciences for measurement of whole-body composition Authors’ contributions
B-HA, VLS, ML-C, HPA, MU, DSD, RMRP, RVM-M, ML de O, LMC de M, M do P, GC
in pigs, broilers, cats, dogs, and sheep, among others. de S, CZP, ROSS, MTT, LCG de C, SMBA, LA, DEK, CMM do P, MCG contributed
Although normative data in these animals are scarce, this with each of the items. MCG and CMM do P contributed with a critical analy‑
technique has a great potential in accurately evaluating sis. SSM, M de MP were the concievers of the manuscript and made a final
revision. All authors read and approved the final manuscript.
the effectiveness of feeding interventions on the amount
of lean and fat mass. Funding
Brazilian Association of Bone Assessment and Osteometabolism (ABRASSO; Brazil‑
ian Society of Bone Metabolism).
Conclusion
Of all current technologies for body composition assess- Availability of data and materials
Not applicable.
ment, DXA should be the preferred method since it per-
forms whole-body analyses in a shorter time and with
Declarations
less radiation exposure, providing a particularly accurate
analysis of fat parameters. In general, BMD and total Ethics approval and consent to participate
body BMC (including head) should not be used as iso- Not applicable.
lated skeletal health markers or to diagnose osteoporosis Consent for publication
and low bone mass in adults. BMI may be a measure of Not applicable.
weight gain, but not necessarily of excess fat.
Competing interests
The following results should be included in DXA- The authors declare no competing interests.
assessed body composition reports: anthropometric data,
total fat mass, percentage of fat mass, FMI, VAT, A/G Author details
1
 Discipline of Endocrinology, Department of Medicine, Universidade
ratio, ALMI (Baumgartner criteria), and BMI-adjusted Federal de São Paulo (UNIFESP), Rua Estado de Israel, 639, São Paulo, SP
lean mass index (FNIH for over 65-year-old individuals). CEP: 04022‑001, Brazil. 2 Department of Epidemiology, Universidade Federal
The diagnosis of sarcopenia is based on low muscle do Espírito Santo (UFES), Vitória, ES, Brazil. 3 Discipline of Rheumatology,
Department of Medicine, Universidade Federal de São Paulo (UNIFESP), São
mass associated with low muscle strength or perfor- Paulo, SP, Brazil. 4 School of Medicine, Instituto Master de Ensino Presidente
mance; these parameters can be evaluated by handgrip Antônio Carlos (IMEPAC), Uberlândia, MG, Brazil. 5 Discipline of Rheumatology,
strength and gait speed, which are the tests mostly Department of Medicine, Universidade de São Paulo (USP), São Paulo, SP, Brazil.
6
 Discipline of Rheumatology, Department of Medicine, Universidade Federal
used for this purpose in clinical practice. Lean mass
Maeda et al. Advances in Rheumatology (2022) 62:11 Page 22 of 25

do Rio de Janeiro (UFRJ), Rio de Janeiro, RJ, Brazil. 7 Clínica Janice Lamas 16. Kang SM, Yoon JW, Ahn HY, et al. Android fat depot is more closely asso‑
Radiologia, Brasília, DF, Brazil. 8 Bone D Consultoria e Treinamento, São Paulo, ciated with metabolic syndrome than abdominal visceral fat in elderly
SP, Brazil. 9 Department of Nutrition, School of Public Health, Universidade de people. PLoS ONE. 2011;6:e27694.
São Paulo (USP), São Paulo, SP, Brazil. 10 Discipline of Allergy, Clinical Immunol‑ 17. Kaul S, Rothney MP, Peters DM, et al. Dual-energy X-ray absorptiometry
ogy, and Rheumatology, Department of Pediatrics, Universidade Federal de for quantification of visceral fat. Obesity (Silver Spring). 2012;20:1313–8.
São Paulo (UNIFESP), São Paulo, SP, Brazil. 11 Section of Pediatric Rheumatology, 18. Micklesfield LK, Goedecke JH, Punyanitya M, Wilson KE, Kelly TL. Dual-
Department of Pediatrics, Universidade Federal de São Paulo (UNIFESP), São energy X-ray performs as well as clinical computed tomography for the
Paulo, SP, Brazil. 12 Endocrine Pediatric Unit, Department of Pediatrics, School measurement of visceral fat. Obesity (Silver Spring). 2012;20:1109–14.
of Medicine, Universidade de Brasília (FM-UnB), Brasília, DF, Brazil. 13 School 19. Miazgowski T, Krzyzanowska-Swiniarska B, Dziwura-Ogonowska J,
of Agricultural and Veterinary Sciences, Universidade Estadual de São Paulo Widecka K. The associations between cardiometabolic risk factors and
(UNESP), Jaboticabal, SP, Brazil. 14 Postgraduate Program in Health and Behav‑ visceral fat measured by a new dual-energy X-ray absorptiometry-
ior, Universidade Católica de Pelotas (UCPel), Pelotas, RS, Brazil. 15 Postgradu‑ derived method in lean healthy Caucasian women. Endocrine.
ate Program in Nutrition and Food, Universidade Federal de Pelotas (UFPel), 2014;47:500–5.
Pelotas, RS, Brazil. 16 Department of Agricultural, Food and Nutritional Science, 20. Bi X, Seabolt L, Shibao C, et al. DXA-measured visceral adipose tis‑
Division of Human Nutrition, University of Alberta, Edmonton, Canada. sue predicts impaired glucose tolerance and metabolic syndrome
in obese Caucasian and African-American women. Eur J Clin Nutr.
Received: 21 May 2021 Accepted: 4 March 2022 2015;69:329–36.
21. Schousboe JT, Langsetmo L, Schwartz AV, et al. Comparison of associa‑
tions of DXA and CT visceral adipose tissue measures with insulin
resistance, lipid levels, and inflammatory markers. J Clin Densitom.
2017;20:256–64.
References 22. Katzmarzyk PT, Greenway FL, Heymsfield SB, Bouchard C. Clinical utility
1. Kendler DL, Borges JL, Fielding RA, et al. The official positions of and reproducibility of visceral adipose tissue measurements derived
the International Society for Clinical Densitometry: indications of from dual-energy X-ray absorptiometry in White and African American
use and reporting of DXA for body composition. J Clin Densitom. adults. Obesity (Silver Spring). 2013;21:2221–4.
2013;16:496–507. 23. Hirsch KR, Blue MNM, Trexler ET, Smith-Ryan AE. Visceral adipose tissue
2. Ward LC. Human body composition: yesterday, today, and tomorrow. normative values in adults from the United States using GE Lunar iDXA.
Eur J Clin Nutr. 2018;72:1201–7. Clin Physiol Funct Imaging. 2019;39:407–14.
3. Teigen LM, Kuchnia AJ, Mourtzakis M, Earthman CP. The use of technol‑ 24. Miazgowski T, Kucharski R, Soltysiak M, Taszarek A, Miazgowski B,
ogy for estimating body composition: strengths and weaknesses of Widecka K. Visceral fat reference values derived from healthy European
common modalities in a clinical setting [formula: see text]. Nutr Clin men and women aged 20–30 years using GE Healthcare dual-energy
Pract. 2017;32:20–9. x-ray absorptiometry. PLoS ONE. 2017;12:e0180614.
4. Prospective Studies C, Whitlock G, Lewington S, et al. Body-mass index 25. Ofenheimer A, Breyer-Kohansal R, Hartl S, et al. Reference values of
and cause-specific mortality in 900 000 adults: collaborative analyses of body composition parameters and visceral adipose tissue (VAT) by DXA
57 prospective studies. Lancet. 2009;373:1083–96. in adults aged 18–81 years-results from the LEAD cohort. Eur J Clin Nutr.
5. Lemos T, Gallagher D. Current body composition measurement tech‑ 2020;74:1181–91.
niques. Curr Opin Endocrinol Diabetes Obes. 2017;24:310–4. 26. Wiklund P, Toss F, Weinehall L, et al. Abdominal and gynoid fat mass are
6. Sinha J, Duffull SB, Al-Sallami HS. A review of the methods and associ‑ associated with cardiovascular risk factors in men and women. J Clin
ated mathematical models used in the measurement of fat-free mass. Endocrinol Metab. 2008;93:4360–6.
Clin Pharmacokinet. 2018;57:781–95. 27. Jacobs EJ, Newton CC, Wang Y, et al. Waist circumference and all-cause
7. Schousboe JT, Shepherd JA, Bilezikian JP, Baim S. Executive summary mortality in a large US cohort. Arch Intern Med. 2010;170:1293–301.
of the 2013 international society for clinical densitometry posi‑ 28. Hangartner TN, Warner S, Braillon P, Jankowski L, Shepherd J. The
tion development conference bone densitometry. J Clin Densitom. Official Positions of the International Society for Clinical Densitometry:
2013;16:455–66. acquisition of dual-energy X-ray absorptiometry body composition and
8. Shuhart CR, Yeap SS, Anderson PA, et al. Executive summary of the 2019 considerations regarding analysis and repeatability of measures. J Clin
ISCD position development conference on monitoring treatment, DXA Densitom. 2013;16:520–36.
cross-calibration and least significant change, spinal cord injury, peri- 29. Okosun IS, Seale JP, Lyn R. Commingling effect of gynoid and android
prosthetic and orthopedic bone health, transgender medicine, and fat patterns on cardiometabolic dysregulation in normal weight Ameri‑
pediatrics. J Clin Densitom. 2019;22:453–71. can adults. Nutr Diabetes. 2015;5:e155.
9. Brandao CM, Camargos BM, Zerbini CA, et al. 2008 official positions of 30. Lee K, Lee S, Kim YJ, Kim YJ. Waist circumference, dual-energy X-ray
the Brazilian Society for Clinical Densitometry–SBDens. Arq Bras Endo‑ absortiometrically measured abdominal adiposity, and computed
crinol Metabol. 2009;53:107–12. tomographically derived intra-abdominal fat area on detecting meta‑
10. Petak S, Barbu CG, Yu EW, et al. The Official Positions of the International bolic risk factors in obese women. Nutrition. 2008;24:625–31.
Society for Clinical Densitometry: body composition analysis reporting. 31. Novotny R, Going S, Teegarden D, et al. Hispanic and Asian pubertal
J Clin Densitom. 2013;16:508–19. girls have higher android/gynoid fat ratio than whites. Obesity (Silver
11. VanItallie TB, Yang MU, Heymsfield SB, Funk RC, Boileau RA. Height- Spring). 2007;15:1565–70.
normalized indices of the body’s fat-free mass and fat mass: potentially 32. Scafoglieri A, Clarys JP. Dual energy X-ray absorptiometry: gold standard
useful indicators of nutritional status. Am J Clin Nutr. 1990;52:953–9. for muscle mass? J Cachexia Sarcopenia Muscle. 2018;9:786–7.
12. Gallagher D, Heymsfield SB, Heo M, Jebb SA, Murgatroyd PR, Sakamoto 33. Sheean P, Gonzalez MC, Prado CM, McKeever L, Hall AM, Braunsch‑
Y. Healthy percentage body fat ranges: an approach for developing weig CA. American society for parenteral and enteral nutrition clinical
guidelines based on body mass index. Am J Clin Nutr. 2000;72:694–701. guidelines: the validity of body composition assessment in clinical
13. Kelly TL, Wilson KE, Heymsfield SB. Dual energy X-Ray absorptiom‑ populations. JPEN J Parenter Enteral Nutr. 2020;44:12–43.
etry body composition reference values from NHANES. PLoS ONE. 34. Hansen RD, Raja C, Aslani A, Smith RC, Allen BJ. Determination of
2009;4:e7038. skeletal muscle and fat-free mass by nuclear and dual-energy x-ray
14. Rothney MP, Catapano AL, Xia J, et al. Abdominal visceral fat measure‑ absorptiometry methods in men and women aged 51–84 y (1–3). Am J
ment using dual-energy X-ray: association with cardiometabolic risk Clin Nutr. 1999;70:228–33.
factors. Obesity (Silver Spring). 2013;21:1798–802. 35. Baumgartner RN, Koehler KM, Gallagher D, et al. Epidemiology of
15. Canoy D, Boekholdt SM, Wareham N, et al. Body fat distribution and sarcopenia among the elderly in New Mexico. Am J Epidemiol.
risk of coronary heart disease in men and women in the European 1998;147:755–63.
Prospective Investigation Into Cancer and Nutrition in Norfolk cohort: a
population-based prospective study. Circulation. 2007;116:2933–43.
 Maeda et al. Advances in Rheumatology (2022) 62:11 Page 23 of 25

36. Newman AB, Kupelian V, Visser M, et al. Sarcopenia: alternative defini‑ 57. Vestergaard S, Patel KV, Bandinelli S, Ferrucci L, Guralnik JM. Character‑
tions and associations with lower extremity function. J Am Geriatr Soc. istics of 400-meter walk test performance and subsequent mortality in
2003;51:1602–9. older adults. Rejuvenation Res. 2009;12:177–84.
37. Delmonico MJ, Harris TB, Lee JS, et al. Alternative definitions of sarco‑ 58. Martin HJ, Yule V, Syddall HE, Dennison EM, Cooper C, Aihie SA. Is
penia, lower extremity performance, and functional impairment with hand-held dynamometry useful for the measurement of quadriceps
aging in older men and women. J Am Geriatr Soc. 2007;55:769–74. strength in older people? A comparison with the gold standard Bodex
38. Zoico E, Di Francesco V, Guralnik JM, et al. Physical disability and mus‑ dynamometry. Gerontology. 2006;52:154–9.
cular strength in relation to obesity and different body composition 59. Lee PH, Macfarlane DJ, Lam TH, Stewart SM. Validity of the international
indexes in a sample of healthy elderly women. Int J Obes Relat Metab physical activity questionnaire short form (IPAQ-SF): a systematic
Disord. 2004;28:234–41. review. Int J Behav Nutr Phys Act. 2011;8:115.
39. Lau EM, Lynn HS, Woo JW, Kwok TC, Melton LJ 3rd. Prevalence of and 60. Podsiadlo D, Richardson S. The timed “Up & Go”: a test of basic func‑
risk factors for sarcopenia in elderly Chinese men and women. J Geron‑ tional mobility for frail elderly persons. J Am Geriatr Soc. 1991;39:142–8.
tol A Biol Sci Med Sci. 2005;60:213–6. 61. Wallmann HW, Evans NS, Day C, Neelly KR. Interrater reliability of
40. Domiciano DS, Figueiredo CP, Lopes JB, et al. Discriminating sarco‑ the five-times-sit-to-stand test. Home Health Care Manag Pract.
penia in community-dwelling older women with high frequency of 2013;25:13–7.
overweight/obesity: the Sao Paulo Ageing & Health Study (SPAH). 62. Cruz-Jentoft AJ, Bahat G, Bauer J, et al. Sarcopenia: revised European
Osteoporos Int. 2013;24:595–603. consensus on definition and diagnosis. Age Ageing. 2019;48:16–31.
41. Heymsfield SB, Wang Z, Baumgartner RN, Ross R. Human body 63. Cruz-Jentoft AJ, Baeyens JP, Bauer JM, et al. Sarcopenia: European
composition: advances in models and methods. Annu Rev Nutr. consensus on definition and diagnosis: report of the European Working
1997;17:527–58. Group on Sarcopenia in Older People. Age Ageing. 2010;39:412–23.
42. Figueiredo CP, Domiciano DS, Lopes JB, et al. Prevalence of sarcopenia 64. Delmonico MJ, Harris TB, Visser M, et al. Longitudinal study of muscle
and associated risk factors by two diagnostic criteria in community- strength, quality, and adipose tissue infiltration. Am J Clin Nutr.
dwelling older men: the Sao Paulo Ageing and Health Study (SPAH). 2009;90:1579–85.
Osteoporos Int. 2014;25:589–96. 65. Malmstrom TK, Miller DK, Simonsick EM, Ferrucci L, Morley JE. SARC-F:
43. Cesari M, Kritchevsky SB, Baumgartner RN, et al. Sarcopenia, obesity, a symptom score to predict persons with sarcopenia at risk for poor
and inflammation–results from the Trial of Angiotensin Converting functional outcomes. J Cachexia Sarcopenia Muscle. 2016;7:28–36.
Enzyme Inhibition and Novel Cardiovascular Risk Factors study. Am J 66. Barbosa-Silva TG, Menezes AM, Bielemann RM, Malmstrom TK, Gonzalez
Clin Nutr. 2005;82:428–34. MC, Grupo de Estudos em Composicao Corporal e N. Enhancing SARC-
44. Studenski SA, Peters KW, Alley DE, et al. The FNIH sarcopenia project: F: improving sarcopenia screening in the clinical practice. J Am Med Dir
rationale, study description, conference recommendations, and final Assoc. 2016;17:1136–41.
estimates. J Gerontol A Biol Sci Med Sci. 2014;69:547–58. 67. Roberts HC, Denison HJ, Martin HJ, et al. A review of the measurement
45. Alley DE, Shardell MD, Peters KW, et al. Grip strength cutpoints for the of grip strength in clinical and epidemiological studies: towards a
identification of clinically relevant weakness. J Gerontol A Biol Sci Med standardised approach. Age Ageing. 2011;40:423–9.
Sci. 2014;69:559–66. 68. Bhasin S, Travison TG, Manini TM, et al. Sarcopenia definition: the posi‑
46. Cawthon PM, Peters KW, Shardell MD, et al. Cutpoints for low appen‑ tion statements of the sarcopenia definition and outcomes consortium.
dicular lean mass that identify older adults with clinically significant J Am Geriatr Soc. 2020;68:1410–8.
weakness. J Gerontol A Biol Sci Med Sci. 2014;69:567–75. 69. Evans WJ, Morley JE, Argiles J, et al. Cachexia: a new definition. Clin Nutr.
47. McLean RR, Shardell MD, Alley DE, et al. Criteria for clinically relevant 2008;27:793–9.
weakness and low lean mass and their longitudinal association with 70. Argiles JM, Lopez-Soriano FJ, Toledo M, Betancourt A, Serpe R, Busquets
incident mobility impairment and mortality: the foundation for the S. The cachexia score (CASCO): a new tool for staging cachectic cancer
National Institutes of Health (FNIH) sarcopenia project. J Gerontol A Biol patients. J Cachexia Sarcopenia Muscle. 2011;2:87–93.
Sci Med Sci. 2014;69:576–83. 71. 2019 ISCD official position—adults. https://​www.​iscd.​org/​offic​ial-​posit​
48. Dam TT, Peters KW, Fragala M, et al. An evidence-based comparison of ions/​2019-​iscd-​offic​ial-​posit​ions-​adult/. Accessed 13 April 2020.
operational criteria for the presence of sarcopenia. J Gerontol A Biol Sci 72. Pinheiro MM, Ciconelli RM, Jacques Nde O, Genaro PS, Martini LA, Ferraz
Med Sci. 2014;69:584–90. MB. The burden of osteoporosis in Brazil: regional data from fractures in
49. Proctor DN, O’Brien PC, Atkinson EJ, Nair KS. Comparison of techniques adult men and women–the Brazilian Osteoporosis Study (BRAZOS). Rev
to estimate total body skeletal muscle mass in people of different age Bras Reumatol. 2010;50:113–27.
groups. Am J Physiol. 1999;277:E489-495. 73. Instituto Brasileiro de Geografi a e Estatística (IBGE). Antropometria e
50. Janssen I, Baumgartner RN, Ross R, Rosenberg IH, Roubenoff R. Skeletal Estado Nutricional de Crianças, Adolescentes e Adultos no Brasil. 2010.
muscle cutpoints associated with elevated physical disability risk in 74. World Health Organization. Global status report on noncommunicable
older men and women. Am J Epidemiol. 2004;159:413–21. diseases 2010—description of the global burden of NCDs, their risk
51. Mitchell WK, Williams J, Atherton P, Larvin M, Lund J, Narici M. factors and determinants. In: 2010.
Sarcopenia, dynapenia, and the impact of advancing age on human 75. Nielson CM, Marshall LM, Adams AL, et al. BMI and fracture risk in older
skeletal muscle size and strength; a quantitative review. Front Physiol. men: the osteoporotic fractures in men study (MrOS). J Bone Miner Res.
2012;3:260. 2011;26:496–502.
52. Edwards MH, Dennison EM, Aihie Sayer A, Fielding R, Cooper C. Osteo‑ 76. Sousa M, Pinheiro MM, Szejnfeld VL, Castro CHM. Body composi‑
porosis and sarcopenia in older age. Bone. 2015;80:126–30. tion parameters in healthy Brazilian women differ from white, black,
53. Moreira LD, Oliveira ML, Lirani-Galvao AP, Marin-Mio RV, Santos RN, and Hispanic American women reference range. J Clin Densitom.
Lazaretti-Castro M. Physical exercise and osteoporosis: effects of differ‑ 2013;16:360–7.
ent types of exercises on bone and physical function of postmenopau‑ 77. Ushida M, de Medeiros PM, de Moura Castro CH, Szejnfeld VL. Body
sal women. Arq Bras Endocrinol Metabol. 2014;58:514–22. composition analysis by DXA (dual X-ray absorptiometry) in Brazilian
54. Marin RV, Pedrosa MAC, Moreira-Pfrimer LDF, Matsudo SMM, Lazaretti- men: normative data. J Bone Miner Metab. 2017;35:554–61.
Castro M. Association between lean mass and handgrip strength with 78. Fan B, Shepherd JA, Levine MA, et al. National Health and Nutrition
bone mineral density in physically active postmenopausal women. J Examination Survey whole-body dual-energy X-ray absorptiometry
Clin Densitom. 2010;13:96–101. reference data for GE Lunar systems. J Clin Densitom. 2014;17:344–77.
55. Rikli RE, Jones CJ. Development and validation of a functional fit‑ 79. Carvalho CJ, Longo GZ, Juvanhol LL, et al. Body composition indices in
ness test for community-residing older adults. J Aging Phys Act. Brazilian adults: age-specific and sex-specific percentile curves. Arch
1999;7:129–61. Endocrinol Metab. 2019;63:358–68.
56. Pavasini R, Guralnik J, Brown JC, et al. Short physical performance bat‑ 80. Victora CG, Barros FC. Cohort profile: the 1982 Pelotas (Brazil) birth
tery and all-cause mortality: systematic review and meta-analysis. BMC cohort study. Int J Epidemiol. 2006;35:237–42.
Med. 2016;14:215.
Maeda et al. Advances in Rheumatology (2022) 62:11 Page 24 of 25

81. Barbosa-Silva TG, Bielemann RM, Gonzalez MC, Menezes AM. Preva‑ health and performance, under the auspices of the I.O.C. Medical Com‑
lence of sarcopenia among community-dwelling elderly of a medium- mission. Sports Med. 2012;42:227–49.
sized South American city: results of the COMO VAI? study. J Cachexia 104. Polsky B, Kotler D, Steinhart C. HIV-associated wasting in the HAART era:
Sarcopenia Muscle. 2016;7:136–43. guidelines for assessment, diagnosis, and treatment. AIDS Patient Care
82. Machado LG, Domiciano DS, Figueiredo CP, et al. Visceral fat meas‑ STDS. 2001;15:411–23.
ured by DXA is associated with increased risk of non-spine fractures 105. Fields-Gardner C, Campa A, American DA. Position of the American
in nonobese elderly women: a population-based prospective cohort Dietetic Association: nutrition intervention and human immunodefi‑
analysis from the Sao Paulo Ageing & Health (SPAH) Study. Osteoporos ciency virus infection. J Am Diet Assoc. 2010;110:1105–19.
Int. 2016;27:3525–33. 106. Carr A, Samaras K, Burton S, et al. A syndrome of peripheral lipodystro‑
83. Wells JC. Toward body composition reference data for infants, children, phy, hyperlipidaemia and insulin resistance in patients receiving HIV
and adolescents. Adv Nutr. 2014;5:320S-329S. protease inhibitors. AIDS. 1998;12:F51-58.
84. Weber DR, Moore RH, Leonard MB, Zemel BS. Fat and lean BMI refer‑ 107. Jaquet D, Levine M, Ortega-Rodriguez E, et al. Clinical and metabolic
ence curves in children and adolescents and their utility in identifying presentation of the lipodystrophic syndrome in HIV-infected children.
excess adiposity compared with BMI and percentage body fat. Am J AIDS. 2000;14:2123–8.
Clin Nutr. 2013;98:49–56. 108. Palchetti CZ, Patin RV, Gouvea Ade F, Szejnfeld VL, Succi RC, Oliveira
85. Lohman TG, Hingle M, Going SB. Body composition in children. Pediatr FL. Body composition and lipodystrophy in prepubertal HIV-infected
Exerc Sci. 2013;25:573–90. children. Braz J Infect Dis. 2013;17:1–6.
86. Weber DR, Leonard MB, Zemel BS. Body composition analysis in the 109. Brambilla P, Bricalli D, Sala N, et al. Highly active antiretroviral-treated
pediatric population. Pediatr Endocrinol Rev. 2012;10:130–9. HIV-infected children show fat distribution changes even in absence of
87. Orsso CE, Silva MIB, Gonzalez MC, et al. Assessment of body composi‑ lipodystrophy. AIDS. 2001;15:2415–22.
tion in pediatric overweight and obesity: a systematic review of the reli‑ 110. Vigano A, Mora S, Testolin C, et al. Increased lipodystrophy is associated
ability and validity of common techniques. Obes Rev. 2020;21:e13041. with increased exposure to highly active antiretroviral therapy in HIV-
88. Staiano AE, Broyles ST, Gupta AK, Katzmarzyk PT. Ethnic and sex dif‑ infected children. J Acquir Immune Defic Syndr. 2003;32:482–9.
ferences in visceral, subcutaneous, and total body fat in children and 111. Palchetti CZ, Succi RCM, Szejnfeld VL, et al. Body composition changes
adolescents. Obesity (Silver Spring). 2013;21:1251–5. and predictors of lipodystrophy in a cohort of pre-pubertal HIV-infected
89. Loomba-Albrecht LA, Styne DM. Effect of puberty on body composi‑ children. Br J Med Med Res. 2015;8:601–11.
tion. Curr Opin Endocrinol Diabetes Obes. 2009;16:10–5. 112. Palchetti CZ, Patin RV, Machado DM, Szejnfeld VL, Succi RC, Oliveira FL.
90. Gordon CM, Leonard MB, Zemel BS, International Society for Clinical D. Body composition in prepubertal, HIV-infected children: a comparison
2013 pediatric position development conference: executive summary of bioelectrical impedance analysis and dual-energy X-ray absorptiom‑
and reflections. J Clin Densitom. 2014;17:219–24. etry. Nutr Clin Pract. 2013;28:247–52.
91. 2019 ISCD Official Positions—Pediatric. https://​iscd.​app.​box.​com/s/​ 113. DiMeglio LA, Wang J, Siberry GK, et al. Bone mineral density in children
ae9gu​sunsr​6e0fm​xkqxo​saczm​nuhuj​x4. Accessed 13 April 2020. and adolescents with perinatal HIV infection. AIDS. 2013;27:211–20.
92. Cossio Bolanos MA, Andruske CL, de Arruda M, et al. Muscle mass in 114. Palchetti CZ, Szejnfeld VL, de Menezes Succi RC, et al. Impaired bone
children and adolescents: proposed equations and reference values for mineral accrual in prepubertal HIV-infected children: a cohort study.
assessment. Front Endocrinol (Lausanne). 2019;10:583. Braz J Infect Dis. 2015;19:623–30.
93. Lifshitz F, Hecht JP, Bermudez EF, et al. Body composition analysis by 115. Bernardino JI, Mocroft A, Mallon PW, et al. Bone mineral density and
dual-energy X-ray absorptiometry in young preschool children. Eur J inflammatory and bone biomarkers after darunavir-ritonavir combined
Clin Nutr. 2016;70:1203–9. with either raltegravir or tenofovir-emtricitabine in antiretroviral-naive
94. Ward KA, Ashby RL, Roberts SA, Adams JE, Zulf MM. UK reference data adults with HIV-1: a substudy of the NEAT001/ANRS143 randomised
for the Hologic QDR Discovery dual-energy x ray absorptiometry trial. Lancet HIV. 2015;2:e464-473.
scanner in healthy children and young adults aged 6–17 years. Arch Dis 116. Escota GV, Mondy K, Bush T, et al. High prevalence of low bone mineral
Child. 2007;92:53–9. density and substantial bone loss over 4 years among HIV-infected
95. Guo B, Wu Q, Gong J, et al. Relationships between the lean mass index persons in the era of modern antiretroviral therapy. AIDS Res Hum
and bone mass and reference values of muscular status in healthy Chi‑ Retroviruses. 2016;32:59–67.
nese children and adolescents. J Bone Miner Metab. 2016;34:703–13. 117. Brown TT, Hoy J, Borderi M, et al. Recommendations for evaluation and
96. Marwaha RK, Garg MK, Bhadra K, Mahalle N, Mithal A, Tandon N. Assess‑ management of bone disease in HIV. Clin Infect Dis. 2015;60:1242–51.
ment and relation of total and regional fat mass with bone mineral 118. Fighera TM, Ziegelmann PK, Rasia da Silva T, Spritzer PM. Bone mass
content among Indian urban adolescents. J Pediatr Endocrinol Metab. effects of cross-sex hormone therapy in transgender people: updated
2015;28:1085–93. systematic review and meta-analysis. J Endocr Soc. 2019;3:943–64.
97. Ripka WL, Orsso CE, Haqq AM, Luz TG, Prado CM, Ulbricht L. Lean mass 119. Hembree WC, Cohen-Kettenis PT, Gooren L, et al. Endocrine treat‑
reference curves in adolescents using dual-energy x-ray absorptiom‑ ment of gender-dysphoric/gender-incongruent persons: an
etry (DXA). PLoS ONE. 2020;15:e0228646. endocrine society clinical practice guideline. J Clin Endocrinol Metab.
98. Moura RR, Coelho AV, Balbino Vde Q, Crovella S, Brandao LA. Meta- 2017;102:3869–903.
analysis of Brazilian genetic admixture and comparison with other Latin 120. Kronacher C HF. Investigations on the causes of rump shape in different
America countries. Am J Hum Biol. 2015;27:674–80. breeds of pigs from X-rays studies. Z Zücht. 1936;35.
99. Muller MJ, Lagerpusch M, Enderle J, Schautz B, Heller M, Bosy-Westphal 121. Mitchell AD, Scholz AM, Pursel VG. Prediction of pork carcass composi‑
A. Beyond the body mass index: tracking body composition in the tion based on cross-sectional region analysis of dual energy X-ray
pathogenesis of obesity and the metabolic syndrome. Obes Rev. absorptiometry (DXA) scans. Meat Sci. 2003;63:265–71.
2012;13(Suppl 2):6–13. 122. Gonçalves CA, Sakomura NK, Pereira da Silva E, Artoni SMB, Suzuki RM,
100. Helba M, Binkovitz LA. Pediatric body composition analysis with dual- Gous RM. Dual energy X-ray absorptiometry is a valid tool for assessing
energy X-ray absorptiometry. Pediatr Radiol. 2009;39:647–56. in vivo body composition of broilers. Anim Prod Sci. 2019;59:993–1000.
101. Nana A, Slater GJ, Stewart AD, Burke LM. Methodology review: using 123. Miller DW, Bennett EJ, Harrison JL, Anderson F, Adam CL. Dual-energy
dual-energy X-ray absorptiometry (DXA) for the assessment of body X-ray absorptiometry scans accurately predict differing body fat con‑
composition in athletes and active people. Int J Sport Nutr Exerc Metab. tent in live sheep. J Anim Sci Biotechnol. 2018;9:80.
2015;25:198–215. 124. Clarke RDKA, Bartle CM, Dobbie PM. Application of dual-energy X-ray
102. Bilsborough JC, Greenway K, Opar D, Livingstone S, Cordy J, Coutts AJ. absorptiometry for ovine carcass evaluation. Proc N Z Soc Anim Prod.
The accuracy and precision of DXA for assessing body composition in 1999;59:272–4.
team sport athletes. J Sports Sci. 2014;32:1821–8. 125. Kipper M, Marcoux M, Andretta I, Pomar C. Assessing the accuracy of
103. Ackland TR, Lohman TG, Sundgot-Borgen J, et al. Current status of body measurements obtained by dual-energy X-ray absorptiometry on pig
composition assessment in sport: review and position statement on carcasses and primal cuts. Meat Sci. 2019;148:79–87.
behalf of the ad hoc research working group on body composition
 Maeda et al. Advances in Rheumatology (2022) 62:11 Page 25 of 25

126. Soladoye OP, Lopez Campos O, Aalhus JL, Gariepy C, Shand P, Juarez
M. Accuracy of dual energy X-ray absorptiometry (DXA) in assess‑
ing carcass composition from different pig populations. Meat Sci.
2016;121:310–6.
127. Salas CER, England J, Cerrate S, Noon CN. Determination of chicken
body composition measured by dual energy X-ray absorptiometry. Int J
Poult Sci. 2012;11:462–8.
128. Lopez-Campos O, Roberts JC, Larsen IL, et al. Rapid and non-destructive
determination of lean fat and bone content in beef using dual energy
X-ray absorptiometry. Meat Sci. 2018;146:140–6.
129. Mawby DI, Bartges JW, d’Avignon A, Laflamme DP, Moyers TD, Cottrell T.
Comparison of various methods for estimating body fat in dogs. J Am
Anim Hosp Assoc. 2004;40:109–14.
130. German AJ, Holden SL, Bissot T, Morris PJ, Biourge V. Use of starting
condition score to estimate changes in body weight and composition
during weight loss in obese dogs. Res Vet Sci. 2009;87:249–54.
131. Lauten SD, Cox NR, Baker GH, Painter DJ, Morrison NE, Baker HJ. Body
composition of growing and adult cats as measured by use of dual
energy X-ray absorptiometry. Comp Med. 2000;50:175–83.

Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in pub‑
lished maps and institutional affiliations.

Ready to submit your research ? Choose BMC and benefit from:

• fast, convenient online submission

• thorough peer review by experienced researchers in your field
• rapid publication on acceptance
• support for research data, including large and complex data types
• gold Open Access which fosters wider collaboration and increased citations
• maximum visibility for your research: over 100M website views per year

At BMC, research is always in progress.

Learn more biomedcentral.com/submissions