Journal Logo

SPECIAL COMMUNICATIONS: Methodological Advances

Accuracy and Reliability of Assessing Lateral Compartmental Leg Composition Using Dual-Energy X-ray Absorptiometry


Author Information
Medicine & Science in Sports & Exercise: April 2017 - Volume 49 - Issue 4 - p 833-839
doi: 10.1249/MSS.0000000000001168
  • Free


Body composition assessment has been used clinically to examine bone disease (e.g., osteoporosis), skeletal muscle loss (e.g., sarcopenia), visceral adiposity, and the effectiveness of physical rehabilitation efforts, among other purposes (3,13,16,22). Multiple body composition methods have been developed, including anthropometry, hydrodensitometry, bioelectrical impedance analysis, computed tomography (CT), magnetic resonance imaging, and dual-energy x-ray absorptiometry (DXA). DXA is currently considered the “gold standard” for the quantitative assessment of total and regional body composition and bone mineral density in both adults and children (11,19,20). Current literature reports valid and reliable comparisons of total and regional body composition values using DXA in the standard frontal plane view in various populations (3,4,6,8,10,14,15) against the criterion 4 component model (bone mineral, fat mass [FM], lean mass [LM], and total-body water) and other body composition assessment methods (5,7,13,22,23).

Recent DXA software developments allow for both automatically and manually generated regions of interest (ROIs) to quantify regional (e.g., arms, legs, and trunk) body composition measures, including contralateral (right vs left) body comparisons and visceral adiposity measurements (10,13,14,16). Recent studies (5,13) using DXA have assessed regional quantification of bone mass, FM, and LM in the upper (thigh) and lower (shank) legs in the frontal view, but DXA has not been used to measure specific ipsilateral (same leg) compartments (e.g., anterior vs posterior) within these respective regions. Notably, anterior and posterior compartments in the upper leg refer to pre- and postfemoral soft tissue regions, respectively, as seen in the lateral view. Lateral subject positioning using DXA would provide a more in-depth body composition analysis—in addition to the standard total-body frontal scanning method—regarding ipsilateral symmetry, injury cause or prevention (via identifying opposing compartmental composition imbalances), athletic performance, and disease and aging processes, among others.

To the best of our knowledge, no studies have examined segmented measures of FM and LM in the lateral view using DXA technology. Therefore, in the present study, we sought 1) to examine the accuracy of measures gathered using this novel “segmented lateral” scanning method compared with the same measures obtained using the total-body frontal DXA scanning method, 2) to quantify contralateral and ipsilateral tissue composition, and 3) to demonstrate the inter- and intrarater reliability of using this novel method to assess compartmental (i.e., anterior/posterior) composition. We hypothesized that FM and LM measured by DXA in the traditional frontal position and novel lateral position would not significantly differ within individuals. If we observed no difference between the two scanning methods, we planned to measure the inter- and intrarater reliability of this segmented lateral scanning method using custom ROIs.


Study participants

Twenty-one (10 males/11 females) participants 18–23 yr old (mean age 20.3 ± 1.3 yr) were recruited from the University of Minnesota–Twin Cities campus. Participants were healthy and had a body mass index (BMI) >18 kg·m−2. All participants wore minimal, light clothing free of metallic material. The study protocol was approved by the University of Minnesota Institutional Review Board, and written informed consent was obtained from all participants.

Scan procedures

All testing was performed at the Clinical and Translational Science Institute located within the Delaware Clinical Research Unit on the University of Minnesota campus between 8:00 a.m. and 12:00 p.m. Each participant's height and weight were measured using an electronic scale and wall-mounted stadiometer. BMI was calculated as the body weight in kilograms divided by height in meters squared. Study participant characteristics are presented in Table 1. All female participants were screened for a negative pregnancy test before undergoing DXA scans. After height and weight measurements, total-body composition was measured using standard procedures (9) in the supine position (Fig. 1A) on a GE Lunar iDXA system (iDXA; General Electric Medical Systems, Madison, WI) and scans were analyzed using enCoreTM software (platform version 16.0, General Electric Medical Systems). After the full-body scan, participants underwent two DXA leg scans (right and left), using the full-body scan mode, to quantify FM and LM in the lateral view. The leg scans were completed after the scan had reached the shoulder, thereby excluding the head. Participants were repositioned for these segmented lateral scans by lying on their side (i.e., right side to scan the left leg, left side to scan the right leg) with their feet at the start of the scanning area (Fig. 1B). The scanned leg was kept straight along the centerline of the DXA table, whereas the other leg was bent and moved out of the scanning field. The scanned leg was elevated using foam pads at the ankle, and a quarter length foam roller was used at the widest portion of the upper leg to keep the leg straight, to maintain the shape of the muscle, and to ensure participant comfort. The pad used to support the thigh was scanned before the study by DXA to ensure it would not be recognized by DXA and influence the analyses. To create reliable landmarks, before scanning each leg, metallic markers were placed on each participant's lateral epicondyle and greater trochanter to identify ROIs for analyses.

Characteristics of study participants.
Comparison of body composition measures between the total-body scan in the frontal view (A) and the segmented body scan in the lateral view (B) using ROI boxes. (Note: Proximal and distal ROI borders were placed 60% of the length from the lateral epicondyle to the greater trochanter and lateral malleolus, respectively.)

Segmentation quantification

Upon scan completion, a two-dimensional image was automatically produced for postscan analysis. To examine the accuracy of using this segmented lateral scanning method as compared with the total-body frontal scanning method, custom ROIs of equal area were created on the frontal and lateral scans. The proximal border of each custom ROI was placed 60% of the length from the lateral epicondyle to the greater trochanter, and the distal border was placed 60% of the length from the lateral epicondyle to the lateral malleolus (Fig. 1A and B). Lateral and medial borders of the ROIs were placed outside the leg circumference, ensuring inclusion of the entire leg.

To examine the reliability of using the segmented lateral scanning method to assess compartmental composition, ROI boxes were manually created, each encompassing the respective anterior and posterior upper leg compartments. The anterior ROI borders were placed at the lateral epicondyle (distal), 80% of the length between the lateral epicondyle and the greater trochanter (proximal), down the shaft of the femur (medial), and outside of the leg area (lateral). The posterior ROI borders were placed similarly, thereby mirroring the borders of the anterior ROI box (Fig. 2). The proximal border was placed at 80% to ensure there was no tissue overlap with any other portion of the body (e.g., contralateral leg) while participants were lying on their side. To assess the reliability of using this segmented lateral scanning method, three investigators analyzed each participant's scans twice to facilitate evaluation of inter- and intrarater reliability of total mass, FM, and LM.

Left leg DXA scan image with a custom ROI box displaying lateral subject positioning and corresponding body composition measurements for anterior and posterior segmented upper leg compartments.

Statistical analyses

To examine the accuracy of this segmented lateral scanning method, paired t-tests were used to compare the composition from the lateral DXA scan to the standard total-body frontal DXA scan. Intraclass correlation coefficients (ICC) and coefficients of variation (CV) assessed the inter- and intrarater reliability of the segmented lateral scan measures. All data were analyzed using statistical analysis software (version 23.0; SPSS Inc., Chicago, IL), with α set at 0.05 for paired t-tests. The strength of reliability for ICC was classified in accordance with Hopkins (15), with CV values less than 5% considered highly reliable.


Descriptive participant characteristics and baseline measurements are summarized in Table 1. Right and left leg measures of total tissue mass, FM, and LM between the total frontal and the segmented lateral DXA scans—used to assess the accuracy of this lateral scanning method—are presented in Table 2. Notably, comparisons between the frontal and the lateral view scans were all nonsignificant (P > 0.05; P value range = 0.15–0.91). The mean ± SD differences between the two scan views of the right leg total, fat, and LM measures averaged over all participants were 8.42 ± 195.57, 61.26 ± 215.66, and −103.00 ± 302.54 g, respectively, and of the left leg total, fat, and LM measures were 19.47 ± 131.80, −5.89 ± 239.97, and −27.58 ± 288.14 g, respectively. Gender differences were examined by comparing mean differences of each mass measure in the right and left legs of males and females. No significant differences were found except for left leg FM (males, 118.20 g; females, −118.89 g).

Frontal and lateral view total leg measures of total, fat, and lean tissue masses, averaged over all study participants.

Interrater reliability

Quantified measures of total tissue mass, FM, and LM in the anterior/posterior compartments—used to assess reliability—are shown in Table 3. Segmented lateral compartmental analysis resulted in high interrater reliability for left and right leg measures of LM, FM, and total mass in the anterior and posterior compartments (Table 4). Interrater CV values also demonstrated high reliability and precision for segmented lateral compartmental LM, FM, and total mass in the anterior and posterior compartments (Table 4). Segmented anterior and posterior leg measurements resulted in greater reliability and smaller variation than did segmented total (i.e., sum of anterior and posterior compartments) upper leg measurements across LM, FM, and total tissue mass. All interrater CV values for segmented total upper leg composition were ≤4.80% and for compartmental (i.e., anterior/posterior) composition were ≤3.55% (Table 4).

Quantified lateral segmented body measures of total, fat, and lean tissue masses in the separate anterior/posterior upper leg compartments, averaged over all study participants.
Interrater and intrarater reliability CV for the anterior and posterior compartments of the upper leg and the right and left segmented total upper legs for FM, LM, and total mass.

Intrarater reliability

Segmented lateral compartmental analysis resulted in strong intrarater reliability for left and right leg measures of LM, FM, and total mass in the anterior and posterior compartments (Table 4). Intrarater CV values also demonstrated high reliability and precision for LM, FM, and total mass in the anterior and posterior compartments (Table 4). Segmented total (i.e., sum of anterior and posterior compartments) upper leg measurements resulted in slightly lower reliability, on average, and larger CV values than those obtained for compartmental (i.e., anterior/posterior) measurements across LM, FM, and total mass. All intrarater CV values for segmented total upper leg composition were ≤3.85% and for compartmental composition were ≤2.69% (Table 4).


This is the first study to assess lateral subject positioning using DXA to analyze ipsilateral upper leg compartmental (i.e., anterior/posterior) composition. The current study demonstrates that the segmented lateral DXA scanning method developed by our research group is both accurate and reliable in assessing LM, FM, and total tissue mass compared with traditional scanning methods. This study observed no significant differences between the total-body frontal and the segmented lateral DXA scan body composition measures in either leg of each participant. Inter- and intrarater reliability was high for the quantification and assessment of segmented lateral upper leg compartment-specific LM, FM, and total tissue mass.

To understand the context of our findings, current literature only reports the high reliability of contralateral regional comparisons of the upper leg (thigh) and lower leg (calf muscles) in the traditional total-body frontal view using manually generated ROIs and supine and prone subject positioning (5,13,18,23). As such, current literature does not report the accuracy nor the reliability of a lateral view DXA scan—a scanning method that would provide a more in-depth body composition analysis of ipsilateral symmetry and regional differences within a leg in athletic and clinical populations. The findings of this study therefore add to the current body of literature by providing initial accuracy and reliability support for using lateral subject positioning as an additional method to assess body composition, including LM, FM, and total tissue mass, using DXA—not only for contralateral comparisons, as previous research (5,13,23) has examined, but also for ipsilateral compartmental (anterior vs posterior) comparisons. Although the preceding measurements may be possible using CT and magnetic resonance imaging, the limitations associated with these two methods (i.e., cost, feasibility, accessibility, and CT radiation) make their use unlikely for assessing compartmental composition (1). In comparison, DXA’s increased feasibility and accessibility, lower cost, quick scan time, and minimal radiation make DXA a practical method to assess compartmental body composition.

Among athletes, current literature notes the utilization of DXA as a practical method to assess an athlete’s body composition in the total-body frontal plane as it relates to performance and nutritional intervention (1). Nana et al. (21) note the use of this assessment to describe athletes’ physical characteristics across sports or within the same sport (26,27), to examine an athlete’s suitability for a weight class in a weight division sport (e.g., wrestling) (6), and to examine athletes’ contralateral leg asymmetries—in the standard total-body frontal plane—for FM and LM measures (17). Therefore, a lateral scan analysis would not only provide information about athletes’ physical characteristics and contralateral symmetry but also ipsilateral symmetry in opposing upper leg compartments. Further, examination of opposing ipsilateral, compartmental differences in the lower limbs may be more beneficial in assessing injury risk, causes of injury, and the rehabilitation process, as changes in body composition affect elite athletes’ competitive performance (24). In fact, segmented lateral DXA scans would allow for longitudinal tracking of compositional changes (e.g., baseline/preinjury, postinjury, before returning to play, etc., or during multiple seasons) in opposing ipsilateral body compartments of athletes, as previous studies have made this assessment using the total-body frontal scanning method (2,25).

Clinically, precise and accurate body composition measurement is important in assessing certain medical conditions (e.g., sarcopenia), the aging process, and evaluating interventions (12). In addition, segmented DXA body composition assessment methods may be useful in populations where standard DXA positioning remains a significant challenge (e.g., musculoskeletal disorders). This segmented lateral scanning method could be used to assess and longitudinally monitor body composition changes in elderly, diseased, and disabled populations affected by muscle wasting; this is in addition to monitoring LM improvements (and prevention of LM loss) over time in response to individualized therapeutic interventions (e.g., pretherapy, midpoint, posttherapy). Therefore, the ability of this novel segmented lateral scanning method to evaluate opposing ipsilateral and contralateral regions of the body would provide greater insight into the injury, aging, and disease processes affecting LM and FM measures.

Major strengths of the current study include the study population's wide body composition variability (BMI range = 19.0–32.0 kg·m−2) and the type of statistical analyses performed (paired t-tests and ICC). The former allowed for reliability assessment across a broad BMI value range. A study population with a narrower BMI range may have resulted in lower variation in the standard deviation of the mean differences (i.e., greater accuracy) and smaller CV values (i.e., greater reliability)—biased results which, while desired, limit generalizability. Second, the within-subject statistical analyses chosen—for both accuracy and reliability examinations—controlled for differences between subjects.

Limitations of the current pilot study include its small sample size, the potential systematic measurement errors in the manual generation of ROIs, and the limited custom ROI box size due to the ankle foam pad. Specifically, the small sample size may have contributed to higher variation in the standard deviation of the mean differences comparing the two scanning methods and between and within raters reflected in higher CV values. Yet despite its small sample size, this study provides initial evidence for this segmented lateral scanning method’s accuracy and reliability. Second, although specific instructions detailed the manual production of ROI borders, slight differences in ROI box measurements may have occurred between the frontal and the lateral scans and for compartmental tissue quantification between and within raters. Third, the ankle foam pad limited the custom ROI box size in comparing the two scanning methods, as frontal and lateral scan ROI borders were drawn to avoid pad inclusion. Further, the postscan analysis of this segmented lateral scanning method may not be capable of fully separating muscle compartments—a limitation of the DXA scanner. Despite these limitations, no significant differences in body composition measures were observed between frontal and lateral scanning methods, and ICC and CV values were reliable between and within raters. A larger sample size and greater precision of custom ROI borders may demonstrate higher reliability.

In conclusion, we made the novel observation that lateral subject positioning using DXA is an accurate and reliable method to assess compartmental composition, allowing for the assessment of LM, FM, and total tissue mass in the anterior and posterior upper leg compartments. Future research should examine reliability measures using a larger sample size and a diverse array of populations across age, body size, body fatness, athletic status, and musculoskeletal development. Future studies may also evaluate the feasibility of examining upper extremity compartmental composition using this novel lateral DXA scanning method to assess potential imbalances. Limitations of examining the upper limbs, however, may include difficulty in upper extremity lateral positioning on the DXA scanner due to shoulder and forearm rotational differences across individuals, difficulty in delineating the upper arm’s anatomically smaller anterior/posterior compartments, and difficulty in obtaining a maximal visible area of soft tissue without head interference. Nonetheless, the ability to assess upper and lower extremity compartmental composition using this study's lateral DXA scanning method may provide a more in-depth analysis for rehabilitative, clinical, and performance purposes.

This study was funded by the Clinical and Translational Science Institute (CTSI) at the University of Minnesota (CTSA: NIH UL1TR000114).

There were no conflicts of interest in the current study. The results of the present study do not constitute endorsement by the American College of Sports Medicine. The results of this study are presented clearly, honestly, and without fabrication, falsification, or inappropriate data manipulation.


1. Ackland TR, Lohman TG, Sundgot-Borgen J, Maughan RJ, Meyer NL, Stewart AD, et al. Current status of body composition assessment in sport: review and position statement on behalf of the ad hoc research working group on body composition health and performance, under the auspices of the I.O.C. Medical Commission. Sports Med. 2012;42(3):227–49.
2. Bilsborough JC, Greenway K, Opar D, Livingstone S, Cordy J, Coutts AJ. The accuracy and precision of DXA for assessing body composition in team sport athletes. J Sports Sci. 2014;32(19):1821–8.
3. Blake GM, Fogelman I. Technical principles of dual energy x-ray absorptiometry. Semin Nucl Med. 1997;27(3):210–28.
4. Bracco D, Thiébaud D, Chioléro RL, Landry M, Burckhardt P, Schutz Y. Segmental body composition by bioelectric impedance and DEXA in humans. J Appl Physiol (1985). 1996;81:2580–7.
5. Burkhart TA, Arthurs KL, Andrews DM. Manual segmentation of DXA scan images results in reliable upper and lower extremity soft and rigid tissue mass estimates. J Biomech. 2009;42(8):1138–42.
6. Clark RR, Sullivan JC, Bartok CJ, Carrel AL. DXA provides a valid minimum weight in wrestlers. Med Sci Sports Exerc. 2007;39:2069–75.
7. Fields DA, Goran MI. Body composition techniques and the four-compartment model in children. J Appl Physiol (1985). 2000;89:613–20.
8. Fuller NJ, Laskey MA, Elia M. Assessment of the composition of major body regions by dual-energy x-ray absorptiometry (DEXA), with special reference to limb muscle mass. Clin Physiol. 1992;12:253–66.
9. GE Healthcare Lunar. enCORE-based x-ray Bone Densitometer: User Manual. Madison (WI): GE Healthcare; 2013. p. 59. Available from:
10. Glickman SG, Marn CS, Supiano MA, Dengel DR. Validity and reliability of dual-energy x-ray absorptiometry for the assessment of abdominal adiposity. J Appl Physiol (1985). 2004;97:509–14.
11. Haarbo J, Gotfredsen A, Hassager C, Christiansen C. Validation of body composition by dual energy x-ray absorptiometry (DEXA). Clin Physiol. 1991;11(4):331–41.
12. Hairi NN, Cumming RG, Naganathan V, et al. Loss of muscle strength, mass (sarcopenia), and quality (specific force) and its relationship with functional limitation and physical disability: the Concord Health and Ageing in Men Project. J Am Geriatr Soc. 2010;58(11):2055–62.
13. Hart NH, Nimphius S, Spiteri T, Cochrane JL, Newton RU. Segmental musculoskeletal examinations using dual-energy x-ray absorptiometry (DXA): positioning and analysis considerations. J Sports Sci Med. 2015;14:620–6.
14. Heymsfield SB, Smith R, Aulet M, et al. Appendicular skeletal muscle mass: measurement by dual-photon absorptiometry. Am J Clin Nutr. 1990;52:214–8.
15. Hopkins W. A scale of magnitudes for effect statistics. 2002 [cited 2016 March 15]. Available from:
16. Kaul S, Rothney MP, Peters DM, et al. Dual-energy x-ray absorptiometry for quantification of visceral fat. Obesity (Silver Spring). 2012;20(6):1313–8.
17. Krzykała M, Leszczyński P. Asymmetry in body composition in female hockey players. Homo. 2015;66(4):379–86.
18. Lambrinoudaki I, Georgiou E, Douskas G, Tsekes G, Kyriakidis M, Proukakis C. Body composition assessment by dual-energy x-ray absorptiometry: comparison of prone and supine measurements. Metabolism. 1998;47(11):1379–82.
19. Lee SY, Gallagher D. Assessment methods in human body composition. Curr Opin Clin Nutr Metab Care. 2008;11:566–72.
20. Mazess RB, Barden HS, Bisek JP, Hanson J. Dual-energy x-ray absorptiometry for total-body and regional bone-mineral and soft-tissue composition. Am J Clin Nutr. 1990;51:1106–12.
21. Nana A, Slater GJ, Stewart AD, Burke LM. Methodology review: using dual-energy x-ray absorptiometry (DXA) for the assessment of body composition in athletes and active people. Int J Sport Nutr Exerc Metab. 2014;24:198–215.
22. Peiffer JJ, Galvão DA, Gibbs Z, et al. Strength and functional characteristics of men and women 65 years and older. Rejuvenation Res. 2010;13(1):75–82.
23. Rothney MP, Martin FP, Xia Y, et al. Precision of GE Lunar iDXA for the measurement of total and regional body composition in nonobese adults. J Clin Densitom. 2012;15(4):399–404.
24. Stanforth PR, Crim BN, Stanforth D, Stults-Kolehmainen MA. Body composition changes among female NCAA division 1 athletes across the competitive season and over a multiyear time frame. J Strength Cond Res. 2014;28(2):300–7.
25. Stewart AD. Assessing body composition in athletes. Nutrition. 2001;17(7–8):694–5.
26. Sutton L, Scott M, Wallace J, Reilly T. Body composition of English Premier League soccer players: influence of playing position, international status, and ethnicity. J Sports Sci. 2009;27:1019–26.
27. Wittich A, Oliveri MB, Rotemberg E, Mautalen C. Body composition of professional football (soccer) players determined by dual x-ray absorptiometry. J Clin Densitom. 2001;4:51–5.


© 2017 American College of Sports Medicine