A DXA Whole Body Composition Cross-Calibration Experience: Evaluation With Humans, Spine, and Whole Body Phantoms

Cross-Calibration of iDXA and pQCT Scanners at Rural and Urban Research Sites in The Gambia, West Africa

Between-scanner differences in measures of bone and body composition can obscure or exaggerate physiological differences in multi-site studies or the magnitude of changes in longitudinal studies. We conducted a cross-calibration study at two bone imaging centres in The Gambia, West Africa where DXA (dual-energy X-ray absorptiometry) and pQCT (peripheral Quantitative-Computed Tomography) are routinely used. Repeat scans were obtained from 64 Gambian adults (58% Male) aged Mean(SD) 30.9 (13.5) years with Mean(SD) body mass index (BMI) 21.7 (4.0) kg/m², using DXA (GE Lunar iDXA, whole body [WB], total hip [TH], lumbar spine [LS]) and pQCT (Stratec XCT2000L/XCT2000, tibia 4%, 50% sites). Between-scanner differences were tested using paired t tests (p < 0.05). Between-scanner correlation was explored with linear regression, and cross-calibration equations derived. Bland-Altman analysis investigated machine trend/bias. When differences were detected (p < 0.05), cross-calibration equations were applied to urban values, with t tests and Bland Altman analysis repeated. Between-scanner differences exceeded the predefined level of statistical significance (p < 0.05) for WB aBMD and BA; all pQCT measures vBMD, BMC, cortical cross-sectional area (CSA) and stress-strain index (SSI). Between-scanner correlation was high (R²:0.92-0.99), except pQCT Mu.Den (R² = 0.51). Bland Altman plots indicated bias increased with increasing BMD. Cross-calibration equations attenuated all between-scanner differences and systematic bias. Cross-calibration, particularly of pQCT scanners, is an important consideration in multi-site studies particularly where between population comparisons are intended. Our experiences and findings may be generalisable to other resource-limited settings where the logistics of sourcing parts and in-country repair may result in lengthy scanner downtime.

Quantifying misdiagnosis rates from cross-calibration biases and precision errors in dual-energy X-ray absorptiometry of the femoral neck

BACKGROUND

Dual-energy X-ray absorptiometry (DXA) is an exam that measures areal bone mineral density (aBMD) and is regularly used to diagnose and monitor osteoporosis. Except for exam quality issues such as operator error, the quantitative results of an exam are not modified by a radiologist or other physician. DXA cross-calibration errors can shift diagnoses, conceivably leading to alternate intervention decisions and patient outcomes.

PURPOSE

After identifying and correcting a cross-calibration bias of 3.8% in our two DXA scanners' aBMD measurements, we investigated misdiagnosis rates for given cross-calibration errors in a single patient cohort to determine the impact on patient care and the value of cross-calibration quality control.

METHODS

The studied cohort was 8012 patients of all ages and sexes with femoral neck exams that were scanned on a single DXA unit from October 1, 2018 to March 31, 2021. There were six subcohorts delineated by age and sex, three female groups and three male groups. Data reporting focused on the highest risk subcohort of 2840 females aged 65 or older. The DXA unit had no calibration changes during that time. Only one femoral neck-left or right-was randomly chosen for analysis. Patients with multiple qualifying exams within the time interval had one exam randomly chosen. The proof-of-principle simulation shifted the aBMD values within a range of ±10%, ±8%, ±6%, ±4%, ±3.5%, ±3%, ±2.5%, ±2%, ±1.5%, ±1%, ±0.5%, and 0 (no shift); the cross-calibration shifts were informed by published results and institutional experience. Measurement precision was modeled by randomly sampling a Gaussian distribution characterized by the worst acceptable least significant change (LSC) of 6.9%, with 100 000 samplings for each patient. T-scores were recalculated from the shifted aBMD values, followed by reassigned diagnoses from the World Health Organization's T-score-based scheme.

RESULTS

The unshifted original subcohort of women aged 65 and older had 599 normal diagnoses (21.1% of the cohort), 1784 osteopenia diagnoses (62.8%), and 455 osteoporosis diagnoses (16.1%). Osteoporosis diagnosis rates were highly sensitive to aBMD shifts. At the extrema, a -10% aBMD shift led to +161% osteoporosis cases, and a +10% aBMD shift led to -64.5% osteoporosis cases. Within the more plausible ±4% aBMD error range, the osteoporosis diagnosis rate changed -10.5% per +1% aBMD shift as indicated by linear regression (R²  = 0.98). Except for the men aged 49 years and younger subcohort, the total cohort and five subcohorts had fit line slopes ranging between -9.7% and -12.1% with R² ≥ 0.98. Cross-calibration bias had greater influence for diagnosis count rates compared to measurement precision, that is, LSC.

CONCLUSIONS

These results quantify the degree of misdiagnosis that can occur in a clinically relevant cohort due to cross-calibration bias. In medical practices where patients may be scanned on more than one DXA unit, ensuring cross-calibration quality is a critical and high-value quality control task with direct impact on patient diagnosis and treatment course. The clinical impact and incidence of poor DXA quality control practices, and cross-calibration in particular, should be studied further.

Multicomponent density models for body composition: Review of the dual energy X-ray absorptiometry volume approach

Accurate and precise body composition estimates, notably of total body adiposity, are a vital component of in vivo physiology and metabolic studies. The reference against which other body composition approaches are usually validated or calibrated is the family of methods referred to as multicomponent "body density" models. These models quantify three to six components by combining measurements of body mass, body volume, total body water, and osseous mineral mass. Body mass is measured with calibrated scales, volume with underwater weighing or air-displacement plethysmography, total body water with isotope dilution, and osseous mineral mass by dual-energy X-ray absorptiometry. Body density is then calculated for use in model as body mass/volume. Studies over the past decade introduced a new approach to quantifying body volume that relies on dual-energy X-ray absorptiometry measurements, an advance that simplifies multicomponent density model development by eliminating the need for underwater weighing or air-displacement plethysmography systems when these technologies are unavailable and makes these methods more accessible to research and clinical programs. This review critically examines these new dual-energy X-ray approaches for quantifying body volume and density, explores their shortcomings, suggests alternative derivation approaches, and introduces ideas for potential future research studies.

Explaining Discrepancies Between Total and Segmental DXA and BIA Body Composition Estimates Using Bayesian Regression

INTRODUCTION/BACKGROUND

Few investigations have sought to explain discrepancies between dual-energy X-ray absorptiometry (DXA) and bioelectrical impedance analysis (BIA) body composition estimates. The purpose of this analysis was to explore physiological and anthropometric predictors of discrepancies between DXA and BIA total and segmental body composition estimates.

METHODOLOGY

Assessments via DXA (GE Lunar Prodigy) and single-frequency BIA (RJL Systems Quantum V) were performed in 179 adults (103 F, 76 M, age: 33.6 ± 15.3 yr; BMI: 24.9 ± 4.3 kg/m²). Potential predictor variables for differences between DXA and BIA total and segmental fat mass (FM) and lean soft tissue (LST) estimates were obtained from demographics and laboratory techniques, including DXA, BIA, bioimpedance spectroscopy, air displacement plethysmography, and 3-dimensional optical scanning. To determine meaningful predictors, Bayesian robust regression models were fit using a t-distribution and regularized hierarchical shrinkage "horseshoe" prior. Standardized model coefficients (β) were generated, and leave-one-out cross validation was used to assess model predictive performance.

RESULTS

LST hydration (i.e., total body water:LST) was a predictor of discrepancies in all FM and LST variables (|β|: 0.20-0.82). Additionally, extracellular fluid percentage was a predictor for nearly all outcomes (|β|: 0.19-0.40). Height influenced the agreement between whole-body estimates (|β|: 0.74-0.77), while the mass, length, and composition of body segments were predictors for segmental LST estimates (|β|: 0.23-3.04). Predictors of segmental FM errors were less consistent. Select sex-, race-, or age-based differences between methods were observed. The accuracy of whole-body models was superior to segmental models (leave-one-out cross-validation-adjusted R² of 0.83-0.85 for FM_TOTAL and LST_TOTAL vs. 0.20-0.76 for segmental estimates). For segmental models, predictive performance decreased in the order of: appendicular lean soft tissue, LST_LEGS, LST_TRUNK and FM_LEGS, FM_ARMS, FM_TRUNK, and LST_ARMS.

CONCLUSIONS

These findings indicate the importance of LST hydration, extracellular fluid content, and height for explaining discrepancies between DXA and BIA body composition estimates. These general findings and quantitative interpretation based on the presented data allow for a better understanding of sources of error between 2 popular segmental body composition techniques and facilitate interpretation of estimates from these technologies.

DXA body composition corrective factors between Hologic Discovery models to conduct multicenter studies

BACKGROUND

Dual X-ray absorptiometry body composition measurements are widely used for clinical and research settings. It is well known that measurements vary across instruments, needing caution for longitudinal monitoring or multicentric studies. This study was to quantify intra- and inter-center variability of bone mineral content, bone mineral density, fat and lean body composition measurements between Hologic Discovery models in order to calculate the corrective factors to be applied for multicenter research projects.

MATERIALS AND METHODS

A whole body phantom composed of materials representing the thickness and percentage of bone, lean and fat mass in the human physiological range was analyzed ten times in three different centers using dual energy X-ray absorptiometry scanners (Two Hologic Discovery QDR A and one QDR W). In addition, we used a morphometric vertebral phantom to monitor stability and a three steps block phantom to check accuracy.

RESULTS

We found a good long-term stability and accuracy for the three devices. Intra-center coefficients of variation were within the range of the manufacturer acceptable values (bone mineral density: 1.40%, bone mineral content: 1%, area: 1.50%, fat mass: 0.89%, lean mass: 0.76%, total mass: 0.12%). Whereas the inter-center coefficient of variation exceeded 8% (bone mineral density: 8.18%, bone mineral content: 3.03%, area: 8.63%: fat mass: 3,92%, lean mass: 7.89%, total mass: 2.85%).

CONCLUSION

Our study showed that the discrepancies across centers remain a major concern, particularly with regard to body composition results. Our study highlight the need of cross calibration between densitometers and proposes corrective factors evaluated from a whole body phantom to lead multicentric studies adjustment.

The Association between First Fractures Sustained during Childhood and Adulthood and Bone Measures in Young Adulthood

OBJECTIVE

To describe the association between fractures sustained at different stages of growth and bone measures in early adulthood.

STUDY DESIGN

Participants (n = 201) in southern Tasmania were at birth at a higher risk of sudden infant death syndrome; they were followed to age 25. Outcomes were areal bone mineral density at the spine, hip, and total body (by dual-energy x-ray absorptiometry) and trabecular and cortical bone measures at the radius and tibia (by high-resolution peripheral quantitative computed tomography). Fractures were self-reported and confirmed by radiographs at 8, 16, and 25 years of age. Multivariable linear regression was used to analyze the association of the occurrence of prepubertal (<9 years of age), pubertal (9-16 years of age), and postpubertal (17-25 years of age) fractures with all bone measures.

RESULTS

Over 25 years, 99 participants had at least 1 fracture. For high-resolution peripheral quantitative computed tomography measures at age 25, prepubertal fractures were negatively associated with cortical and trabecular volumetric bone mineral density and most microarchitecture measures at both the tibia and radius. Prepubertal fractures had a significant association with smaller increase of areal bone mineral density from age 8 to 16 years and at 25 years of age compared with participants with no fractures. Pubertal fractures had no association with any bone measures and postpubertal fractures were only associated with a lower trabecular number at the tibia.

CONCLUSIONS

Prepubertal fractures are negatively associated with areal bone mineral density increases during growth and high-resolution peripheral quantitative computed tomography bone measures in young adulthood. There is little evidence that fractures occurring from age 8 years onward with bone measures in young adulthood, implying that prepubertal fractures may be associated with bone deficits later in life.

The Association of Vitamin D in Youth and Early Adulthood with Bone Mineral Density and Microarchitecture in Early Adulthood

This study aimed to describe the association of vitamin D status at different stages of growth with bone measures in adolescence and early adulthood. There were 415 participants followed from age 8 to 16, and 201 further followed to age 25. Areal bone mineral density (BMD) at the spine, hip and total body was measured by dual-energy X-ray absorptiometry at ages 16 and 25, and tibial and radial trabecular and cortical bone microarchitecture by high resolution peripheral quantitative computerised tomography at age 25. Serum 25-hydroxyvitamin D (25OHD) concentrations were measured at ages 8, 16 and 25. Multivariable linear regression was used to analyse the association of 25OHD concentrations at three timepoints with bone measures at ages 16 and 25. The proportion of participants with vitamin D deficiency (< 50 nmol/L) was 11%, 43% and 41% at three timepoints, respectively. Serum 25OHD concentrations at age 8 were not significantly associated with any bone measures at age 16 or 25. Serum 25OHD concentrations at age 16 had a significant association with higher BMD at nearly all sites at ages 16 and 25 as well as lower radial porosity and more compact trabecular microarchitecture (higher density, increased number and reduced separation) at both the radius and tibia at age 25. Serum 25OHD concentrations at age 25 were only associated with hip BMD. Higher vitamin D concentrations in adolescence, to a lesser extent at age 25, have beneficial associations with BMD and bone microarchitecture in early adulthood. Optimising vitamin D status particularly during adolescence should be a priority.

A Cross-Sectional Analysis of Body Composition Among Healthy Elderly From the European NU-AGE Study: Sex and Country Specific Features

Body composition (BC) is an emerging important factor for the characterization of metabolic status. The assessment of BC has been studied in various populations and diseases such as obesity, diabetes, endocrine diseases as well as physiological and paraphysiological conditions such as growth and aging processes, and physical training. A gold standard technique for the assessment of human BC at molecular level is represented by dual-energy X-ray absorptiometry (DXA), which is able to precisely assess the body mass (and areal bone mineral density-aBMD) on a regional and whole-body basis. For the first time, within the framework of the NU-AGE project, BC has been assessed by means of a whole-body DXA scan in 1121 sex-balanced free-living, apparently healthy older adults aged 65–79 years enrolled in 5 European countries (Italy, France, United Kingdom, Netherlands, and Poland). The aim of this analysis is to provide a complete profile of BC in healthy elderly participants from five European countries and to investigate country- and sex-related differences by state-of-the-art DXA technology. To compare BC data collected in different centers, specific indexes and ratios have been used. Non-parametric statistical tests showed sex-specific significant differences in certain BC parameters. In particular, women have higher fat mass (FM) (Fat/Lean mass ratio: by 67%, p < 2.2e-16) and lower lean mass (Lean Mass index: by -18%, p < 2.2e-16) than men. On the other hand, men have higher android FM than women (Android/gynoid FM ratio: by 56%, p < 2.2e-16). Interesting differences also emerged among countries. Polish elderly have higher FM (Fat/Lean mass ratio: by 52%, p < 2.2e-16) and lower lean mass (Skeletal Mass index: by -23%, p < 2.2e-16) than elderly from the other four countries. At variance, French elderly show lower FM (Fat/Lean mass ratio: by -34%, p < 2.2e-16) and higher lean mass (Skeletal Mass index: by 18%, p < 2.2e-16). Moreover, five BC profiles in women and six in men have been identified by a cluster analysis based on BC parameters. Finally, these data can serve as reference for normative average and variability of BC in the elderly populations across Europe.

Interpretation of Dual-Energy X-Ray Absorptiometry-Derived Body Composition Change in Athletes: A Review and Recommendations for Best Practice

Dual-energy X-ray absorptiometry (DXA) is a medical imaging device which has become the method of choice for the measurement of body composition in athletes. The objectives of this review were to evaluate published longitudinal DXA body composition studies in athletic populations for interpretation of "meaningful" change, and to propose a best practice measurement protocol. An online search of PubMed and CINAHL via EBSCO Host and Web of Science enabled the identification of studies published until November 2016. Those that met the inclusion criteria were reviewed independently by 2 authors according to their methodological quality and interpretation of body composition change. Twenty-five studies published between 1996 and November 2016 were reviewed (male athletes: 13, female athletes: 3, mixed: 9) and sample sizes ranged from n = 1 to 212. The same number of eligible studies was published between 2013 and 2016, as over the 16 yr prior (between 1996 and 2012). Seven did not include precision error, and fewer than half provided athlete-specific precision error. There were shortfalls in the sample sizes on which precision estimates were based and inconsistencies in the level of pre-scan standardization, with some reporting full standardization protocols and others reporting only single (e.g., overnight fast) or no control measures. There is a need for standardized practice and reporting in athletic populations for the longitudinal measurement of body composition using DXA. Based on this review and those of others, plus the official position of the International Society for Clinical Densitometry, our recommendations and protocol are proposed as a guide to support best practice.

Tracking of Areal Bone Mineral Density From Age Eight to Young Adulthood and Factors Associated With Deviation From Tracking: A 17-Year Prospective Cohort Study

We have previously shown that bone mineral density (BMD) tracks strongly from age 8 to 16 years. This study aimed to describe whether this strong tracking continued to age 25 years and describe factors associated with deviation from tracking. Ninety-nine participants were followed from age 8 to 25 years and 197 participants from age 16 to 25 years. Outcomes measured were BMD at the spine, hip, and total body (by dual-energy X-ray absorptiometry [DXA]). Other factors measured were anthropometrics, inhaled corticosteroids (ICS) use, history of being breastfed, sports participation, fitness (by physical work capacity [PWC₁₇₀ ]), lean mass (LM), and fat mass (FM) (by DXA). There was moderate to strong tracking of BMD from age 8 to 25 years (correlation coefficients: males, 0.59 to 0.65; females, 0.70 to 0.82) and strong tracking from age 16 to 25 years (males, 0.81 to 0.83; females, 0.84 to 0.88) after adjustment for change in body size. From age 8 to 25 years, 54% to 56% of participants kept their BMD tertile position. PWC₁₇₀ at age 8 years, relative and absolute change in LM, and sports participation at age 25 years predicted males would improve their tertile position or remain in the highest tertile of spine or hip BMD. However, relative and absolute change in FM had the opposite association in males while absolute change in FM predicted positive deviation in females. From age 16 to 25 years, LM, PWC₁₇₀ , sports participation at age 16 years, and change in LM, PWC₁₇₀ , and sports participation at age 25 years predicted positive deviation in males. LM at age 16 years was positively associated and PWC₁₇₀ negatively associated with positive deviation in females. BMD tracks from childhood to early adulthood in both males and females. There appears to be greater capacity to alter tracking before age 16 years. Increasing LM in both sexes and improving fitness and sports participation in males during growth might be effective strategies to improve BMD in early adulthood. © 2017 American Society for Bone and Mineral Research.

Amalgamated Reference Data for Size-Adjusted Bone Densitometry Measurements in 3598 Children and Young Adults-the ALPHABET Study

The increasing use of dual-energy X-ray absorptiometry (DXA) in children has led to the need for robust reference data for interpretation of scans in daily clinical practice. Such data need to be representative of the population being studied and be "future-proofed" to software and hardware upgrades. The aim was to combine all available pediatric DXA reference data from seven UK centers to create reference curves adjusted for age, sex, ethnicity, and body size to enable clinical application, using in vivo cross-calibration and making data back and forward compatible. Seven UK sites collected data on GE Lunar or Hologic Scanners between 1996 and 2012. Males and females aged 4 to 20 years were recruited (n = 3598). The split by ethnic group was white 2887; South Asian 385; black Afro-Caribbean 286; and mixed heritage 40. Scans of the total body and lumbar spine (L₁ to L₄ ) were obtained. The European Spine Phantom was used to cross-calibrate the 7 centers and 11 scanners. Reference curves were produced for L₁ to L₄ bone mineral apparent density (BMAD) and total body less head (TBLH) and L₁ to L₄ areal bone mineral density (aBMD) for GE Lunar Prodigy and iDXA (sex- and ethnic-specific) and for Hologic (sex-specific). Regression equations for TBLH BMC were produced using stepwise linear regression. Scans of 100 children were randomly selected to test backward and forward compatibility of software versions, up to version 15.0 for GE Lunar and Apex 4.1 for Hologic. For the first time, sex- and ethnic-specific reference curves for lumbar spine BMAD, aBMD, and TBLH aBMD are provided for both GE Lunar and Hologic scanners. These curves will facilitate interpretation of DXA data in children using methods recommended in ISCD guidelines. The databases have been created to allow future updates and analysis when more definitive evidence for the best method of fracture prediction in children is agreed. © 2016 American Society for Bone and Mineral Research.