CAM 60140

Whole Body Dual X-ray Absorptiometry (DEXA) to Determine Body Composition

Category:Radiology   Last Reviewed:February 2020
Department(s):Medical Affairs   Next Review:February 2021
Original Date:February 2012    

Description:
Using low-dose x-rays of two different energy levels, whole-body dual-energy x-ray absorptiometry (DXA) measures lean tissue mass, total and regional body fat, as well as bone density. DXA scans have become a tool for research on body composition (eg, as a more convenient replacement for underwater weighing). This evidence review addresses potential applications in clinical care rather than research use of the technology.

For individuals who have a clinical condition associated with abnormal body composition who receive DXA body composition studies, the evidence includes systematic reviews and several cross-sectional studies comparing DXA with other techniques. The relevant outcomes are symptoms and change in disease status. The available studies were primarily conducted in research settings and often used DXA body composition studies as a reference standard; these studies do not permit conclusions about the accuracy of DXA for measuring body composition. A systematic review exploring the clinical validity of DXA against reference methods for the quantification of intra-abdominal adipose tissue raised concerns regarding precision and reliability. More importantly, no studies were identified in which DXA body composition measurements were actively used in patient management. The evidence is insufficient to determine the effects of the technology on health outcomes.

For individuals who have a clinical condition managed by monitoring changes in body composition over time who receive serial DXA body composition studies, the evidence includes several prospective studies monitoring patients over time. The relevant outcomes are symptoms and change in disease status. The studies used DXA as a tool to measure body composition and were not designed to assess the accuracy of DXA. None of the studies used DXA findings to make patient management decisions or addressed how serial body composition assessment might improve health outcomes. The evidence is insufficient to determine the effects of the technology on health outcomes.

Background
Body Composition Measurement
Body composition measurements can be used to quantify and assess the relative proportions of specific body compartments such as fat and lean mass (eg, bones, tissues, organs, muscles). These measurements may be more useful in informing diagnosis, prognosis, or therapy than standard assessments (eg, body weight, body mass index) that do not identify the contributions of individual body compartments or their particular relationships with health and disease. While these body composition measurements have been most frequently utilized for research purposes, they may be useful in clinical settings to:

  • Evaluate the health status of undernourished patients, those impacted by certain disease states (eg, anorexia nervosa, cachexia), or those undergoing certain treatments (eg, antiretroviral therapy, bariatric surgery).
  • Evaluate the risk of heart disease or diabetes by measuring visceral fat vs total body fat.
  • Assess body composition changes related to growth and development (eg, infancy, childhood), aging (eg, sarcopenia), and in certain disease states (eg, HIV, diabetes).
  • Evaluate patients in situations where body mass index is suspected to be discordant with total fat mass (eg, body-building, edema).

A variety of techniques has been researched, including most commonly, anthropomorphic measures, bioelectrical impedance, and dual-energy x-ray absorptiometry (DXA). All of these techniques are based in part on assumptions about the distribution of different body compartments and their density, and all rely on formulas to convert the measured parameter into an estimate of body composition. Therefore, all techniques will introduce variation based on how the underlying assumptions and formulas apply to different populations of subjects (ie, different age groups, ethnicities, or underlying conditions). Techniques using anthropomorphics, bioelectrical impedance, underwater weighing, and DXA are briefly reviewed below. 

Anthropomorphic Techniques
Anthropomorphic techniques for the estimation of body composition include measurements of skinfold thickness at various sites, bone dimensions, and limb circumference. These measurements are used in various equations to predict body density and body fat. Due to its ease of use, measurement of skinfold thickness is one of the most common techniques. The technique is based on the assumption that the subcutaneous adipose layer reflects total body fat but this association may vary with age and sex.

Bioelectrical Impedance
Bioelectrical impedance analysis is based on the relation among the volume of the conductor (ie, human body), the conductor's length (ie, height), the components of the conductor (ie, fat and fat-free mass), and its impedance. Estimates of body composition are based on the assumption that the overall conductivity of the human body is closely related to lean tissue. The impedance value is then combined with anthropomorphic data to give body compartment measures. The technique involves attaching surface electrodes to various locations on the arm and foot. Alternatively, the patient can stand on the pad electrodes.

Underwater Weighing
Underwater weighing requires the use of a specially constructed tank in which the subject is seated on a suspended chair. The subject is then submerged in the water while exhaling. While valued as a research tool, weighing people underwater is typically not suitable for routine clinical use. This technique is based on the assumption the body can be divided into two compartments with constant densities: adipose tissue, with a density of 0.9 g/cm3, and lean body mass (ie, muscle and bone), with a density of 1.1 g/cm3. One limitation of the underlying assumption is the variability in density between muscle and bone; eg, bone has a higher density than muscle, and bone mineral density varies with age and other conditions. Also, the density of body fat may vary, depending on the relative components of its constituents (eg, glycerides, sterols, glycolipids).

Dual-energy X-ray Absorptiometry
While the cited techniques assume two body compartments, DXA can estimate three body compartments consisting of fat mass, lean body mass, and bone mass. DXA systems use a source that generates x-rays at two energies. The differential attenuation of the two energies is used to estimate the bone mineral content and soft tissue composition. When two x-ray energies are used, only two tissue compartments can be measured; therefore, soft tissue measurements (ie, fat and lean body mass) can only be measured in areas in which no bone is present. DXA can also determine body composition in defined regions (ie, the arms, legs, and trunk). DXA measurements are based in part on the assumption that the hydration of fat-free mass remains constant at 73%. Hydration, however, can vary from 67% to 85% and can vary by disease state. Other assumptions used to derive body composition estimates are considered proprietary by DXA manufacturers. The use of DXA for bone mineral density assessment in patients diagnosed with or at risk of osteoporosis is addressed separately in evidence review 60101. Vertebral fracture assessment with densitometry by DXA is addressed separately in evidence review 60144.

REGULATORY STATUS
Body composition software for several bone densitometer systems have been approved by the U.S. Food and Drug Administration through the premarket approval process. They include the Lunar iDXA systems (GE Healthcare, Madison, WI), Hologic DXA systems (Hologic, Bedford, MA), and Norland DXA systems (Norland, at Swissray, Fort Atkinson, WI).

Policy:
Dual X-ray absorptiometry (DEXA) body composition studies is investigational and/or unproven and is therefore considered NOT MEDICALLY NECESSARY.

Policy Guidelines
This service should now be coded using the unlisted CPT code 76499.

Benefit Application
BlueCard®/National Account Issues
State or federal mandates (e.g., FEP) may dictate that all devices approved by the U.S. Food and Drug Administration (FDA) may not be considered investigational and, thus, these devices may be assessed only on the basis of their medical necessity.

Rationale 
Evidence reviews assess whether a medical test is clinically useful. A useful test provides information to make a clinical management decision that improves the net health outcome. That is, the balance of benefits and harms is better when the test is used to manage the condition than when another test or no test is used to manage the condition.

The first step in assessing a medical test is to formulate the clinical context and purpose of the test. The test must be technically reliable, clinically valid, and clinically useful for that purpose. Evidence reviews assess the evidence on whether a test is clinically valid and clinically useful. Technical reliability is outside the scope of these reviews, and credible information on technical reliability is available from other sources.

Dual-energy X-ray Absorptiometry as a Test to Detect Abnormal Body Composition
Clinical Context and Test Purpose
The purpose of DXA body composition studies is to improve the diagnosis and managementof patients who have a clinical condition associated with abnormal body composition.

The question addressed in this evidence review is: Does the use of DXA improve the net health outcome in patients with clinical conditions associated with abnormal body composition?

The following PICOs were used to select literature to inform this review.

Patients
The relevant population of interest are individuals with clinical conditions. 

Interventions
The test being considered is DXA body composition studies administered in an outpatient setting.

Comparators
The following practices are currently being used to make decisions in this patient group: standard of care without DXA or an alternative method of body composition analysis.

Outcomes
The general outcomes of interest include symptom management and change in disease status. For patients at risk of osteoporosis, outcomes of interest would include fracture incidence. For patients with HIV who are treated with antiretroviral therapy, outcomes of interest would include lipodystrophy.

Technically Reliable
Assessment of technical reliability focuses on specific tests and operators and requires a review of unpublished and often proprietary information. Review of specific tests, operators, and unpublished data are outside the scope of this evidence review and alternative sources exist. This evidence review focuses on the clinical validity and clinical utility.

Clinically Valid
A test must detect the presence or absence of a condition, the risk of developing a condition in the future, or treatment response (beneficial or adverse).

Systematic Reviews
A systematic review and meta-analysis comparing the accuracy of alternative comparators vs reference standard computed tomography (CT) and magnetic resonance imaging (MRI) methods for the quantification of intra-abdominal adipose tissue (IAAT) was published by Murphy et al (2019).1, This systematic review assessed the performance of DXA for IAAT volume quantification and compared the performance of both DXA and bioelectric impedance analysis (BIA) approaches for IAAT area quantification. The American Society for Parenteral and Enteral Nutrition (ASPEN) also conducted a systematic review to evaluate the validity of relevant body composition methods in various clinical populations.2, The use of DXA, ultrasound, and BIA for body composition analysis was investigated. Fifteen studies featuring comparisons of DXA to reference standard methods (eg, MRI and CT) were identified. Nine studies using CT or MRI to validate DXA measures of abdominal fat mass (FM) or total body FM were used for pooled analyses. Characteristics and results of studies included for meta-analysis are summarized in Tables 1 and 2.

Table 1. SR & M-A Characteristics

Study; Subgroup Dates Trials Participants1 N (Range) Design Duration

Murphy et al (2019)1,

1995-2018

23

Studies:

  • With IAAT quantified in humans by CT or MRI reference methods and one of DXA, ultrasound, BIA, or air displacement plethysmography
  • With reference and comparator methods that quantify IAAT at the same anatomical location in the same unit of measurement
  • With reported or quantifiable mean differences and SDs of IAAT quantity

6116 (29-2689)

Cross-sectional, diagnostic test accuracy studies

Retrospective studies

NR

IAAT Area

DXA

2012-2014

3

Included population groups:

  • Elderly adult men and women evaluated by DXA and CT at L4-L5
  • Premenopausal women evaluated by DXA and CT at L4-L5
  • Premenopausal women evaluated by DXA and CT at L4

381 (115-135)

Cross-sectional, diagnostic test accuracy studies

Retrospective studies

NR

BIA

2008-2018

9*

Included population groups:

  • Elderly Caucasian men and women evaluated by BIA and CT at L3-L4
  • Elderly Korean adult men and women evaluated by BIA and CT at umbilicus
  • Elderly Korean adult men and women evaluated by BIA and CT at L4-L5
  • Japanese outpatients with obesity evaluated by BIA and CT at umbilicus
  • Elderly, middle-aged, and adult Chinese men and women evaluated by BIA and CT at L4-L5
  • Elderly adult men and women evaluated by BIA and MRI at L4-L5
  • Elderly, middle-aged, adult, and young men and women evaluated by BIA and CT at L4-L5

2139 (100-1006)

Cross-sectional, diagnostic test accuracy studies

Retrospective studies

NR

IAAT Volume

DXA

2012-2018

7**

Included population groups:

  • Adult men and women evaluated by DXA and CT from S1 to head region
  • Elderly adult men and women evaluated by DXA and CT from S1 to head region
  • Women with PCOS evaluated by DXA and MRI at L3
  • Middle-Eastern adult men and women evaluated by DXA and MRI at android region
  • Adult men and women evaluated by DXA and MRI at L2-L3 with conversion to L1-L5

3410 (40-2689)

Cross-sectional, diagnostic test accuracy studies

Retrospective studies

NR

IAAT Thickness

US

2010-2014

4

Included population groups:

  • Obese women with infertility evaluated by US and CT at L4-L5
  • Middle-aged men and women evaluated by US and CT at L2-L3
  • Elderly and adult men and women evaluated by US and MRI at L2-L3
  • Elderly men and women evaluated by US and MRI at L4

186 (29-74)

Cross-sectional, diagnostic test accuracy studies

Retrospective studies

NR

Sheean et al (2019)

2001-2013

9

Studies:

  • With body compositions assessed in clinical populations via DXA and a reference standard method (eg, MRI or CT)
  • With correlation analyses

1660 (39-625)

Cross-sectional, diagnostic accuracy studies

Retrospective studies

NR

Study; Subgroup

Dates

Trials

Participants1

N (Range)

Design

Duration

Sheean et al (2019)2, (ASPEN)

2001-2013

9

Studies:

  • With body compositions assessed in clinical populations via DXA and a reference standard method (eg, MRI or CT)
  • With correlation analyses

1660 (39-625)

Cross-sectional, diagnostic accuracy studies

Retrospective studies

NR

Abdominal FM in any disease via DXA

2004-2013

4

Included population groups:

  • Urban Asian Indians with type 2 diabetes
  • Premenopausal women with anorexia nervosa
  • Middle-aged Indian men with CVD
  • Multiethnic cohort of men and women with HIV

874 (39-625)

Cross-sectional, diagnostic accuracy studies

Retrospective studies

NR

Total FM in any disease via DXA

2001-2013

7

Included population groups:

  • Women with CVD
  • Postmenopausal women with CVD
  • Men and women with CVD sub 96
  • Middle-aged Indian men with CVD
  • Individuals with myosteatosis
  • Multiethnic cohort of men and women with HIV

1473 (66-625)

Cross-sectional, diagnostic accuracy studies

Retrospective studies

NR

Total FM in CVD via DXA

2001-2013

5

Included population groups:

  • Men and women with CVD sub 96 (103), 92
  • Postmenopausal women with CVD 132, 66
  • Middle-aged Indian men with CVD 128

521 (66-132)

Cross-sectional, diagnostic accuracy studies

Retrospective studies

NR

ASPEN: American Society for Parenteral and Enteral Nutrition; BIA: bioelectrical impedance analysis; CT: computed tomography; CVD: cardiovascular disease; DXA: dual-energy x-ray absorptiometry; FM: fat mass; HIV: human immunodeficiency virus; IAAT: intra-abdominal adipose tissue; M-A: meta-analysis; MRI: magnetic resonance imaging; NR: not reported; SD: standard deviation; SR: systematic review; US: ultrasound.
1 Key study eligibility criteria and demographics of included subgroup participants.
* 3 of 9 trials were sampled twice for a total of 12 result sets due to use of multiple techniques for IAAT quantification via BIA.
** one of eight trials was categorized as an outlier and excluded from pooled analysis.

Table 2. SR & M-A Results

Study

Mean Difference in IAAT Volume

Mean Difference in IAAT Area

Mean Difference in IAAT Thickness

Murphy et al (2019)1,

DXA*

DXA

BIA

US

Total N

3410

381

2139

186

Pooled mean difference (95% LoA)

-10 (-280, 300) (cm3)

8.09 (-98.88, 115.07) (cm2)

-11.63 (-43.12, 19.85) (cm2)

-0.32 (-3.82, 3.17) (cm)

Significance of mean difference (p)

p = 0.808

p = 0.061

p = 0.004

p = 0.400

I2 (p)

99 (<0.001)

98 (<0.001)

94 (<0.001)

93 (<0.001)

Q

Q(6) = 458

Q(2) = 31

Q(11) = 544

Q(3) = 41

Range of N

40-2689

115-135

100-1006

29-74

Range of pooled mean differences

(-451, 262) (cm3)

(3.78, 16.70) (cm2)

(-57.20, 10.96) (cm2)

(-1.10, 0.40) (cm)

DXA Subgroup Analysis

Mean Difference in IAAT Volume by DXA and Gender

Mean Difference in IAAT Volume by DXA and Reference Method

Subgroup

Men

Women

CT

MRI

Subgroup N (Total N)

1483 (3287)

1804 (3287)

377 (3410)

3033 (3410)

Pooled mean difference (95% LoA) (cm3)

144.04 (-512.29, 800.38)

59.96 (-381.08, 492.99)

-41.15 (-881.96, 930.25)

49.52 (-498.42, 586.23)

Significance for subgroup comparison (p)

p = 0.042

p = 0.311

I2

95

90

100

90

Range of Subgroup N

20-1212

20-1477

109-145

40-2689

Range of pooled mean differences (cm3)

(-43, 379)

(4, 143)

(-451, 262)

(4, 104)

Sheean et al (2019)2, (ASPEN)

DXA-derived Abdominal FM

DXA-derived Total FM

DXA vs CT-derived VAT in any disease

DXA vs CT/MRI-derived VAT in any disease

DXA vs CT/MRI-derived VAT in CVD

Total N

874

1473

521

Pooled random effects correlation (95% CI)

0.74 (0.52-0.86)

0.71 (0.45-0.86)

0.71 (0.45-0.84)

I2 (p)

87 (<0.01)

98 (<0.01)

95 (<0.01)

Range of N

39-625

66-625

66-132

Range of individual correlations

(0.52-0.86)

(0.49-0.80)

(0.49-0.87)

ASPEN: American Society for Parenteral and Enteral Nutrition; BIA: bioelectrical impedance analysis; CI: confidence interval; CT: computed tomography; CVD: cardiovascular disease; DXA: dual-energy x-ray absorptiometry; FM: fat mass; IAAT: intra-abdominal adipose tissue; LoA: limits of agreement; M-A: meta-analysis; MRI: magnetic resonance imaging; SR: systematic review; US: ultrasound; VAT: visceral adipose tissue.
* Results following the removal of a study due to identification as an outlier.

While this analysis was primarily focused on the utilization of the different body composition methods for the management of obesity, direct effects on key health outcomes were not explored and patient populations included for analysis displayed extensive heterogeneity and largely featured healthy populations. Measurements of IAAT volume were deemed comparable to the reference methods, however, 95% limits of agreement (LoA) were wide and these results were not seen until the removal of an outlying study. Rationale for identifying the study as an outlier and removing it from the meta-analysis was limited. Prior to the removal of the outlier, the pooled mean difference was significant compared to the reference methods at -124 cm(95% LoA: [-479, 230]; p = 0.013; I2 = 99 [p <0.001]; Q(7) = 773). Performance of DXA for the measurement of IAAT volume also varied significantly between male and female subgroups. Furthermore, included studies did not pre-determine clinically meaningful LoA. The authors' further caution that DXA measurement of IAAT volume has the capacity to differ from reference methods by more than 100%, however, the clinical significance of these margins of error are uncertain in individuals with obesity. While IAAT area cutoff points have been described for the determination of metabolic risk and visceral obesity based on single-slice CT , the authors do not recommend utilization of DXA IAAT area measurements for this purpose due to wide LoA. The clinical utility of existing IAAT area cutpoints is also uncertain as these parameters were found to have applicability for women and cannot necessarily be extrapolated to mixed populations.

ASPEN recommends the use of DXA for the assessment of FM in patients with a specific disease or clinical outcome with a strong recommendation rating based on their analysis. Due to the lack of studies reporting on the validity of DXA for lean mass measurements, no recommendations could be made for assessments of this body compartment. The systematic review acknowledges that while the quality of the included evidence was low, the strong recommendation rating was applied with the rationale that the net benefits of FM assessment via DXA outweigh potential harms. However, the use of DXA findings to make patient management decisions and reporting of adverse events was not featured in the included studies.

Calella et al (2019) performed a systematic review exploring various methods for body composition analysis in patients with cystic fibrosis (CF).3, A previous systematic review by Calella et al (2018) presented on differences in body composition between patients with CF and healthy controls evaluated by DXA and other methods.4, DXA was most frequently used to measure lean body or fat-free mass which was significantly reduced in CF patients. While several included studies showed a correlation between lower fat-free mass and impaired pulmonary function, application, and use of this measure in patient management and its impact on health outcomes was not explored and requires further clarification. As these reviews featured qualitative analyses, data on clinical validity could not be extracted.

A systematic review by Bundred et al (2019) evaluated body composition assessment and sarcopenia in patients with pancreatic ductal adenocarcinoma.5, Meta-analyses revealed that sarcopenia was associated with lower overall survival in both operable (harms ratio: 1.95; 95% confidence interval: 1.35-2.81; p<0.001) and unresectable patients (harms ratio: 2.49; 95% confidence interval : 1.38-4.48; p = 0.002). However, of the 42 included studies, only 1 utilized measurements obtained by DXA, limiting the relevance of the overall findings to this technology and preventing extraction of pertinent clinical validity data. Furthermore, the authors caution that many studies failed to account for variation introduced by gender, race, tumor stage, and other factors. Additionally, clear criteria for the diagnosis of sarcopenia or cachexia via body composition assessments with DXA are lacking.

Cross-Sectional Studies
Most of the literature on DXA as a diagnostic test to detect abnormal body composition involves the use of the technology in the research setting, often as a reference test; studies have been conducted in different populations of patients and underlying disorders.6,7,8,9,10,11,12,13,14,15, In some cases, studies have compared other techniques with DXA to identify simpler methods of determining body composition. In general, these studies have shown that DXA is highly correlated to various methods of body composition assessment. For example, a study by Alves et al (2014) compared 2 bioelectrical impedance devices with DXA for the evaluation of body composition in heart failure.6, Ziai et al (2014) compared bioelectric impedance analysis with DXA for evaluating body composition in adults with CF .7, Whether or not a DXA scan is considered the reference standard, the key consideration regarding its routine clinical use is whether the results of the scan can be used to manage the patients and improve health outcomes.

Case-Control Studies
As a single diagnostic measure, it is important to establish diagnostic cutoff points for normal and abnormal values. This is problematic because normal values will require the development of normative databases for the different components of body composition (ie, bone, fat, lean mass) for different populations of patients at different ages. Regarding measuring bone mineral density (BMD), normative databases have largely focused on postmenopausal white women, and these values cannot necessarily be extrapolated to men or to different races. DXA determinations of BMD are primarily used for fracture risk assessment in postmenopausal women and to select candidates for various pharmacologic therapies to reduce fracture risk. In an example regarding lean mass, Reina et al (2019) conducted a case-control study to assess the correlation of body mass index (BMI) or serum albumin levels to DXA-derived parameters of nutritional status and sarcopenia in women (n=89) with rheumatoid arthritis.16, While 44% of cases met diagnostic criteria for sarcopenia based on quantification of the skeletal muscle index, a reference technique was not clearly identified in this study. Skeletal muscle index is calculated by dividing appendicular skeletal muscle mass by the square of the patient's height. A previously identified threshold of ≤5.75 kg/m2 in women was applied, however, this metric was established through the use of BIA in a slightly older patient population. Given that DXA provides measures of lean mass which may be influenced by body compartments other than skeletal muscle, the relevance of this diagnostic cutoff point is uncertain. Furthermore, the study utilized a control group composed of patients affected by non-inflammatory rheumatic disorders as opposed to healthy controls, further limiting the relevance of applied cutoff points. In addition to the aforementioned uncertainties of establishing and applying normal values for components of body composition, it also is unclear how a single measure of body composition would be used in patient management. Studies discussing appropriate use and determination of DXA-derived lean mass cutoffs for sarcopenia in various populations of patients and underlying disorders continue to be featured in the literature.17,18,

Clinically Useful
A test is clinically useful if the use of the results informs management decisions that improve the net health outcome of care. The net health outcome can be improved if patients receive correct therapy, or more effective therapy, or avoid unnecessary therapy, or avoid unnecessary testing.

Direct Evidence
Direct evidence of clinical utility is provided by studies that have compared health outcomes for patients managed with and without the test. Because these are intervention studies, the preferred evidence would be from randomized controlled trials (RCTs).

No RCTs were identified to support the utility of DXA for this indication.

Chain of Evidence
Indirect evidence on clinical utility rests on clinical validity. If the evidence is insufficient to demonstrate test performance, no inferences can be made about clinical utility.

Because the clinical validity of DXA for this population is limited, a chain of evidence cannot be constructed.

Section Summary: DXA as a Test to Detect Abnormal Body Composition
The available evidence was generated primarily in research settings and often used DXA body composition studies as a reference standard; these studies do not permit conclusions about the accuracy of DXA for measuring body composition. A systematic review exploring the clinical validity of DXA measurements against reference methods for the quantification of intra-abdominal adipose tissue raised concerns regarding precision and reliability. Additionally, no studies were identified in which DXA body composition measurements were actively used in patient management.

DXA as a Test to Monitor Changes in Body Composition
Clinical Context and Test Purpose
The purpose of serial DXA body composition studies in patients who have a clinical condition managed by monitoring body composition changes over time is to improve disease management.

The question addressed in this evidence review is: Does serial DXA improve the net health outcome in patients with clinical conditions managed by monitoring body composition changes over time?

The following PICOs were used to select literature to inform this review.

Patients
The relevant population of interest are individuals with clinical conditions managed by monitoring body composition changes over time.

Interventions
The test being considered is serial DXA body composition studies.

Comparators
The following practices are currently being used to make decisions in this patient group: standard of care without DXA or an alternative method of body composition analysis.

Outcomes
The general outcomes of interest include symptom management and change in disease status. For patients with anorexia nervosa, outcomes of interest would include disease-related morbidity, disease-related mortality, and rate of remission.

Technically Reliable
Assessment of technical reliability focuses on specific tests and operators and requires a review of unpublished and often proprietary information. Review of specific tests, operators, and unpublished data are outside the scope of this evidence review and alternative sources exist. This evidence review focuses on the clinical validity and clinical utility.

Clinically Valid
A test must detect the presence or absence of a condition, the risk of developing a condition in the future, or treatment response (beneficial or adverse).

The ability to detect a change in body composition over time is related in part to the precision of the technique, defined as the degree to which repeated measurements of the same variable give the same value. For example, DXA measurements of bone mass are thought to have a precision error of 1% to 3% and, given the slow rate of change in BMD in postmenopausal women treated for osteoporosis, it is likely that DXA scans would only be able to detect a significant change in BMD in the typical patient after two years of therapy. Of course, changes in body composition are anticipated to be larger and more rapid than changes in BMD in postmenopausal women; therefore, precision errors in DXA scans become less critical in interpreting results. However, precision errors for other body compartments such as lean and fat mass may differ and impact clinical validity. Coefficients of variation as high as 42.2% have been reported for fat mass.19,

Prospective Studies
Several studies have reported on DXA measurement of body composition changes over time in clinical populations; none of these studies used DXA findings to make patient management decisions and few addressed how serial body composition assessment might improve health outcomes.20,21,19,22,A long-term prospective study assessing the association between body fat and breast cancer risk in postmenopausal women with a normal BMI was published by Iyenagar et al (2019), featuring the ad hoc secondary analysis of results from the Women's Health Initiative randomized clinical trial and observational study cohorts.22, Women (n=3460) were assessed at baseline and during years 1, 3, 6, and 9 for BMI and via DXA. Multivariable-adjusted hazard ratios for the association of various body fat measures with the risk of developing invasive or estrogen receptor positive breast cancer were reported. Median follow-up duration was 16.9 years. Characteristics and results of clinical validity for breast cancer risk assessment are summarized in Tables 3 and 4.

Table 3. Study Characteristics of Clinical Validity of Risk Assessment

Study

Study Population

Designa

Reference Standard

Timing of Reference and Index Tests

Blinding of Assessors

Commentb

Iyengar et al (2019)22,

Postmenopausal women aged 50 to 79 years enrolled in the Women's Health Initiative (WHI) RCT or observational study were considered for study. Women from 3 WHI trial centers were assessed longitudinally for body fat composition. Data from women with normal BMIs were assessed for correlations with breast cancer outcomes.

Prospective, sample selection NR.

Clinical outcomes were confirmed via questionnaires. Breast cancer cases were confirmed via review of medical records and pathology reports.

NR

NR

Risk outcomes for women in the RCT and observational cohorts were not analyzed separately. Given that treatments utilized in the RCT group may have had an impact on breast cancer risk and outcomes, the relevance and utility of this study is uncertain.

BMI: body mass index; NR: not reported; RCT: randomized controlled trial. 
Note 2 aspects of design: prospective, retrospective or nonconcurrent prospective and sample selection random or consecutive
Note other characteristics that could cause bias or limit relevance such as timeframe or practice setting.

Table 4. Clinical Validity of Breast Cancer Risk Assessment with DXA

Study; Subgroup; Body Fat DXA Measurement (Cutoff)

Initial N

Final N

Cases/Person-Years

Excluded Samples

Prevalence of Condition

Clinical Validity Outcome: Multivariable Adjusted HR (95% CI)
Baseline Body Fat Measures Serial Body Fat Measures

Iyengar et al (2019)22,

Invasive Breast Cancer

3464*

3460

4*

182

Highest Quartile

P-value for trend

Per 5-unit increase

Cutoff

Time-Dependent

Whole-body fat mass, kg (>25.1)

NR

NR

NR

57

1.89 (1.21-2.95)

0.004

1.28 (1.10-1.49)

≥ 22.1

1.43 (1.06-1.93)

Whole-body fat, % (>41.3)

NR

NR

NR

52

1.79 (1.14-2.83)

0.03

1.19 (1.03-1.37)

≥ 38.0

1.45 (1.07-1.95)

Fat mass of trunk, kg (>11.4)

NR

NR

NR

50

1.88 (1.18-2.98)

0.002

1.46 (1.14-1.87)

≥ 9.4

1.50 (1.12-2.03)

Ratio of trunk fat mass to mean of legs (>2.6)

NR

NR

NR

43

1.30 (0.83-2.02)

0.10

NR

NR

NR

Iyengar et al (2019)22,

ER+ Breast Cancer

3464

3460

4*

146

Highest Quartile

P-value for trend

Per 5-unit increase

Cutoff

Time-Dependent

Whole-body fat mass, kg (>25.1)

NR

NR

NR

48

2.21 (1.23-3.67)

0.002

1.35 (1.14-1.60)

≥ 22.1

1.41 (1.01-1.97)

Whole-body fat, % (>41.3)

NR

NR

NR

44

2.17 (1.29-3.66)

0.01

1.27 (1.08-1.48)

≥ 38.0

1.50 (1.07-2.10)

Fat mass of trunk, kg (>11.4)

NR

NR

NR

41

1.98 (1.18-3.31)

0.003

1.56 (1.18-2.06)

≥ 9.4

1.46 (1.05-2.04)

Ratio of trunk fat mass to mean of legs (>2.6)

NR

NR

NR

34

1.28 (0.78-2.10)

0.13

NR

NR

NR

CI: confidence interval; DXA: dual-energy x-ray absorptiometry; ER+: estrogen receptor-positive; HR: hazard ratio; NR: not reported.
* Excluded cases were lost to follow-up with ER+ status not reported.

These results suggest that standard BMI categorization may be inadequate for the risk assessment of invasive breast cancers in postmenopausal women. However, the clinical utility of DXA findings on patient management protocols and health outcomes requires further study. Relevance and study design and conduct limitations are summarized in Tables 5 and 6.

Table 5. Relevance Limitations

Study Populationa Interventionb Comparatorc Outcomesd

Duration of Follow-Upe

Iyengar et al (2019)22,

1, 4. Study population is unclear; study population not representative of intended use.

2. Version used unclear regarding both DXA and patient participation in RCT treatment or observational groups.

3. Not compared to other tests used for same purpose.

3, 5. Key clinical validity outcomes not reported; adverse events of the test not described.

 

DXA: dual-energy x-ray absorptiometry; RCT: randomized controlled trial.
Population key: 1. Intended use population unclear; 2. Clinical context is unclear; 3. Study population is unclear; 4. Study population not representative of intended use.
Intervention key: 1. Classification thresholds not defined; 2. Version used unclear; 3. Not intervention of interest.
c Comparator key: 1. Classification thresholds not defined; 2. Not compared to credible reference standard; 3. Not compared to other tests in use for same purpose.
d Outcomes key: 1. Study does not directly assess a key health outcome; 2. Evidence chain or decision model not explicated; 3. Key clinical validity outcomes not reported (sensitivity, specificity, and predictive values); 4. Reclassification of diagnostic or risk categories not reported; 5. Adverse events of the test not described (excluding minor discomforts and inconvenience of venipuncture or noninvasive tests).
e Follow-Up key: 1. Follow-up duration not sufficient with respect to natural history of disease (true-positives, true-negatives, false-positives, false-negatives cannot be determined).

Table 6. Study Design and Conduct Limitations

Study

Selectiona

Blindingb

Delivery of Testc

Selective Reportingd

Data Completenesse

Statisticalf

Iyengar et al (2019)22,

1. Selection not described.

1. Blinding not described.

1, 4. Timing of delivery of index or reference tests not clear; expertise of evaluators not described.

2. Evidence of selective reporting (covariates did not have to be pre-specified).

1. Inadequate description of indeterminate and missing samples.

2. Comparison with other tests not reported.

Selection key: 1. Selection not described; 2. Selection not random or consecutive (ie, convenience).
bBlinding key: 1. Not blinded to results of reference or other comparator tests.
cTest Delivery key: 1. Timing of delivery of index or reference test not described; 2. Timing of index and comparator tests not same; 3. Procedure for interpreting tests not described; 4. Expertise of evaluators not described.
d Selective Reporting key: 1. Not registered; 2. Evidence of selective reporting; 3. Evidence of selective publication.
e Data Completeness key: 1. Inadequate description of indeterminate and missing samples; 2. High number of samples excluded; 3. High loss to follow-up or missing data.
f Statistical key: 1. Confidence intervals and/or p values not reported; 2. Comparison with other tests not reported.

Clinically Useful
A test is clinically useful if the use of the results informs management decisions that improve the net health outcome of care. The net health outcome can be improved if patients receive correct therapy, or more effective therapy, or avoid unnecessary therapy, or avoid unnecessary testing.

Direct Evidence
Direct evidence of clinical utility is provided by studies that have compared health outcomes for patients managed with and without the test. Because these are intervention studies, the preferred evidence would be from RCTs.

No RCTs were identified to support the utility of DXA for this indication.

Chain of Evidence
Indirect evidence on clinical utility rests on clinical validity. If the evidence is insufficient to demonstrate test performance, no inferences can be made about clinical utility.

Because the clinical validity of DXA for this population cannot be established, a chain of evidence cannot be constructed.

Section Summary: DXA as a Test to Monitor Changes in Body Composition
Studies assessing serial DXA used it as a tool to measure body composition and were not designed to assess the accuracy of DXA. None of the studies used DXA findings to make patient management decisions or addressed how serial body composition assessment might improve health outcomes.

Summary of Evidence
For individuals who have a clinical condition associated with abnormal body composition who receive DXA body composition studies, the evidence includes systematic reviews and several cross-sectional studies comparing DXA with other techniques. The relevant outcomes are symptoms and change in disease status. The available studies were primarily conducted in research settings and often used DXA body composition studies as a reference standard; these studies do not permit conclusions about the accuracy of DXA for measuring body composition. A systematic review exploring the clinical validity of DXA measurements against reference methods for the quantification of intra-abdominal adipose tissue raised concerns for precision and reliability. More importantly, no studies were identified in which DXA body composition measurements were actively used in patient management. The evidence is insufficient to determine the effects of the technology on health outcomes.

For individuals who have a clinical condition managed by monitoring changes in body composition over time who receive serial DXA body composition studies, the evidence includes several prospective studies monitoring patients over time. The relevant outcomes are symptoms and change in disease status. The studies used DXA as a tool to measure body composition and were not designed to assess the accuracy of DXA. None of the studies used DXA findings to make patient management decisions or addressed how serial body composition assessment might improve health outcomes. The evidence is insufficient to determine the effects of the technology on health outcomes.

Practice Guidelines and Position Statements
International Society for Clinical Densitometry
The International Society for Clinical Densitometry (2015) updated its statements on the use of dual x-ray absorptiometry (DXA) for body composition.23, The following statements were made on the use of DXA total body composition with regional analysis:

To assess fat distribution in patients with HIV who are using antiretroviral agents known to increase the risk of lipoatrophy.

To assess fat and lean mass changes in obese patients undergoing bariatric surgery (or medical, diet, or weight loss regimens with anticipated large weight loss) when weight loss exceeds approximately 10%. The statement noted that the impact of DXA studies on clinical outcomes in these patients is uncertain.

To assess fat and lean mass in patients with muscle weakness and poor physical functioning. The impact on clinical outcomes is uncertain.

Of note, pregnancy is a contraindication to use of DXA to measure body composition. The statement also adds that the clinical utility of DXA measurements of adiposity and lean mass (eg, visceral adipose tissue, lean mass index, fat mass index) is uncertain. Furthermore, while the use of DXA adiposity measures such as fat mass index may be useful in risk-stratifying patients for cardio-metabolic outcomes, specific thresholds to define obesity have not been established.

International Conference on Sarcopenia and Frailty Research Task Force
Evidence-based clinical practice guidelines for the screening, diagnosis, and management of sarcopenia were developed by the International Conference on Sarcopenia and Frailty Research task force in 2018.24, The following recommendations were made:

  • Screening for sarcopenia can be performed using gait speed analysis or SARC-F questionnaire.
  • Individuals screened as positive for sarcopenia should be referred for further assessment to confirm the presence of the disease.
  • DXA imaging should be used to determine low levels of lean body mass when diagnosing sarcopenia.

The recommendation regarding the diagnostic use of DXA received a conditional (weak) recommendation. The certainty of the evidence for DXA assessment was ranked low due to:

  • DXA studies featuring populations from low-middle income countries are lacking.
  • DXA measurement of lean body mass rather than muscle mass may potentially misclassify body composition in certain individuals.
  • Incorporation of DXA measurements of lean body mass may have limited additional benefit for the prediction of relevant health outcomes (eg, falls, fractures, lowered physical performance, mobility).

U.S. Preventive Services Task Force Recommendations
No U.S. Preventive Services Task Force recommendations for whole-body DXA have been identified.

Ongoing and Unpublished Clinical Trials
A search of ClinicalTrials.govin July 2019 did not identify any ongoing or unpublished trials that would likely influence this review. 

References:

  1. Murphy J, Bacon SL, Morais JA et al. Intra-Abdominal Adipose Tissue Quantification by Alternative Versus Reference Methods: A Systematic Review and Meta-Analysis. Obesity (Silver Spring), 2019 Jul;27(7). PMID 31131996
  2. Sheean P, Gonzalez MC, Prado CM et al. American Society for Parenteral and Enteral Nutrition Clinical Guidelines: The Validity of Body Composition Assessment in Clinical Populations. JPEN J Parenter Enteral Nutr, 2019 Jun 20. PMID 31216070
  3. Calella P, Valerio G, Brodlie M et al. Tools and Methods Used for the Assessment of Body Composition in Patients With Cystic Fibrosis: A Systematic Review. Nutr Clin Pract, 2019 Feb 7. PMID 30729571
  4. Calella P, Valerio G, Brodlie M et al. Cystic fibrosis, body composition, and health outcomes: a systematic review. Nutrition, 2018 Nov;55-56:131-139. PMID 29981489
  5. Bundred J, Kamarajah SK, Roberts KJ. Body composition assessment and sarcopenia in patients with pancreatic cancer: a systematic review and meta-analysis. HPB (Oxford), 2019 May 27. PMID 31266698
  6. Alves FD, Souza GC, Biolo A, et al. Comparison of two bioelectrical impedance devices and dual-energy X-ray absorptiometry to evaluate body composition in heart failure. J Hum Nutr Diet. Dec 2014;27(6):632-638. PMID 24684316
  7. Ziai S, Coriati A, Chabot K, et al. Agreement of bioelectric impedance analysis and dual-energy X-ray absorptiometry for body composition evaluation in adults with cystic fibrosis. J Cyst Fibros. Sep 2014;13(5):585- 588. PMID 24522087
  8. Elkan AC, Engvall IL, Tengstrand B, et al. Malnutrition in women with rheumatoid arthritis is not revealed by clinical anthropometrical measurements or nutritional evaluation tools. Eur J Clin Nutr. Oct 2008;62(10):1239- 1247. PMID 17637600
  9. Jensky-Squires NE, Dieli-Conwright CM, Rossuello A, et al. Validity and reliability of body composition analysers in children and adults. Br J Nutr. Oct 2008;100(4):859-865. PMID 18346304
  10. Kullberg J, Brandberg J, Angelhed JE, et al. Whole-body adipose tissue analysis: comparison of MRI, CT and dual-energy X-ray absorptiometry. Br J Radiol. Feb 2009;82(974):123-130. PMID 19168691
  11. Liem ET, De Lucia Rolfe E, L'Abee C, et al. Measuring abdominal adiposity in 6 to 7-year-old children. Eur J Clin Nutr. Jul 2009;63(7):835-841. PMID 19127281
  12. Bedogni G, Agosti F, De Col A, et al. Comparison of dual-energy X-ray absorptiometry, air displacement plethysmography and bioelectrical impedance analysis for the assessment of body composition in morbidly obese women. Eur J Clin Nutr. Nov 2013;67(11):1129-1132. PMID 24022260
  13. Monteiro PA, Antunes Bde M, Silveira LS, et al. Body composition variables as predictors of NAFLD by ultrasound in obese children and adolescents. BMC Pediatr. Jan 29 2014;14:25. PMID 24476029
  14. Tompuri TT, Lakka TA, Hakulinen M, et al. Assessment of body composition by dual-energy X-ray absorptiometry, bioimpedance analysis and anthropometrics in children: the Physical Activity and Nutrition in Children study. Clin Physiol Funct Imaging. Jan 2015;35(1):21-33. PMID 24325400
  15. Alves Junior CAS, de Lima LRA, de Souza MC et al. Anthropometric measures associated with fat mass estimation in children and adolescents with HIV. Appl Physiol Nutr Metab, 2018 Oct 5;44(5). PMID 30286302
  16. Reina D, Gómez-Vaquero C, Díaz-Torné C et al. Assessment of nutritional status by dual X-Ray absorptiometry in women with rheumatoid arthritis: A case-control study. Medicine (Baltimore), 2019 Feb;98(6):e14361. PMID 30732168
  17. Sinclair M, Hoermann R, Peterson A et al. Use of Dual X-ray Absorptiometry in men with advanced cirrhosis to predict sarcopenia-associated mortality risk. Liver Int., 2019 Feb 13;39(6). PMID 30746903
  18. Lindqvist C, Brismar TB, Majeed A et al. Assessment of muscle mass depletion in chronic liver disease: Dual-energy x-ray absorptiometry compared with computed tomography. Nutrition, 2019 Feb 1;61:93-98. PMID 30703575
  19. Dordevic AL, Bonham M, Ghasem-Zadeh A et al. Reliability of Compartmental Body Composition Measures in Weight-Stable Adults Using GE iDXA: Implications for Research and Practice. Nutrients, 2018 Oct 12;10(10). PMID 30321991
  20. Bazzocchi A, Ponti F, Cariani S, et al. Visceral fat and body composition changes in a female population after RYGBP: a two-year follow-up by DXA. Obes Surg. Mar 2015;25(3):443-451. PMID 25218013
  21. Franzoni E, Ciccarese F, Di Pietro E, et al. Follow-up of bone mineral density and body composition in adolescents with restrictive anorexia nervosa: role of dual-energy X-ray absorptiometry. Eur J Clin Nutr. Feb 2014;68(2):247-252. PMID 24346474
  22. Iyengar NM, Arthur R, Manson JE et al. Association of Body Fat and Risk of Breast Cancer in Postmenopausal Women With Normal Body Mass Index: A Secondary Analysis of a Randomized Clinical Trial and Observational Study. JAMA Oncol, 2019 Feb;5(2):155-163. PMID 30520976
  23. International Society for Clinical Densitometry. 2015 ISCD Official Positions - Adult. 2015; https://www.iscd.org/official-positions/2015-iscd-official-positions-adult/. Accessed July 22, 2019.
  24. Dent E, Morley JE, Cruz-Jentoft AJ et al. International Clinical Practice Guidelines for Sarcopenia (ICFSR): Screening, Diagnosis and Management. J Nutr Health Aging, 2018 Dec 1;22(10). PMID 30498820

Coding Section

Codes Number Description
CPT 76499 Unlisted diagnostic radiographic procedure
ICD-9 Diagnosis   Investigational for all diagnoses
ICD-10-CM (effective 10/01/15)   Investigational for all diagnoses
ICD-10-PCS (effective 10/01/15)   ICD-10-PCS codes are only used for inpatient services. There is no specific ICD-10-PCS codes for this imaging
  BW0KZZZ, BW0LZZZ Imaging, anatomical regions, plain radiography, codes for whole body or whole skeleton
Type of Service    
Place of Service    

Procedure and diagnosis codes on Medical Policy documents are included only as a general reference tool for each policy. They may not be all-inclusive.

This medical policy was developed through consideration of peer-reviewed medical literature generally recognized by the relevant medical community, U.S. FDA approval status, nationally accepted standards of medical practice and accepted standards of medical practice in this community, Blue Cross and Blue Shield Association technology assessment program (TEC) and other non-affiliated technology evaluation centers, reference to federal regulations, other plan medical policies, and accredited national guidelines.

"Current Procedural Terminology © American Medical Association.  All Rights Reserved" 

History From 2013 Forward     

02/11/2020 

Annual review, no change to policy intent. Updating description, background, rationale and references. 

02/13/2019 

Annual review, no change to policy intent. Updating rationale and references. 

02/21/2018 

Annual review, no change to policy intent. Updating background, description, regulatory status, coding in guidelines and references. 

02/06/2017

Annual review, no change to policy intent.

02/15/2016 

Annual review, no change to policy intent. Updating background, description, rationale and references. 

02/24/2015 

Annual review, no change to policy intent. Updated background, description, rationale, references. Added guidelines, regulatory status and coding.

02/5/2014

Updated rationale and references. No change to policy intent.


Go Back