Abstract
-
Background
Dual energy X-ray absorptiometry (DXA) measurements vary among instruments from different manufacturers. This study aimed to assess the correlation between bone mineral density (BMD) and body composition measured by the GE Lunar Prodigy and Hologic Horizon W DXA systems and to develop conversion equations.
-
Methods
A total of 120 healthy volunteers (59 males, 61 females) aged 20 to 70 years were scanned on both DXA systems on the same day. BMD was measured at the lumbar spine (L1–L4) and bilateral femurs. Body composition parameters were measured for the whole body and regional sites. Bland-Altman analysis, Pearson correlation, and concordance correlation coefficient (CCC) were used to assess agreement. Linear regression was performed to derive conversion equations.
-
Results
The Hologic system measured BMD values 14% to 20% lower at the lumbar spine and 8% to 17% lower at femoral sites compared to the GE Lunar system (all P<0.0001). Lean mass values were 4% to 10% lower on the Hologic system. Very strong correlations were observed for both BMD (r=0.944–0.980) and lean mass (r=0.963–0.984). Cross-calibration equations were developed for all measurement sites.
-
Conclusions
Despite significant systematic differences, there was very high correlation between the two DXA systems. The conversion equations can facilitate comparison of measurements between these systems in clinical practice and research.
-
Key words: Absorptiometry, photon · Body composition · Bone density · Calibration
GRAPHICAL ABSTRACT
INTRODUCTION
Osteoporosis is a systemic skeletal disorder characterized by decreased bone mass and microarchitectural deterioration of bone tissue, leading to enhanced bone fragility and an increased risk of fractures.[
1]
Dual energy X-ray absorptiometry (DXA) has been established as the gold standard technique for measuring bone mineral density (BMD) since its introduction in the late 1980s.[
2] DXA offers several advantages including high precision, low radiation exposure, rapid acquisition time, and the ability to measure BMD at clinically important skeletal sites.[
3]
Beyond bone densitometry, DXA has emerged as a valuable tool for body composition assessment. The technique can accurately quantify lean mass, fat mass, and bone mineral content at both whole-body and regional levels.[
4] This capability has become increasingly important with the recognition of sarcopenia—the age-related loss of skeletal muscle mass and function—as a significant contributor to adverse health outcomes in older adults.[
5] The European Working Group on Sarcopenia in Older People (EWGSOP) has endorsed DXA as the preferred method for assessing muscle mass in clinical practice and research.[
6]
The two dominant manufacturers of DXA systems are GE Healthcare (Madison, WI, USA) and Hologic Inc. (Bedford, MA, USA), which together account for the majority of DXA installations worldwide.[
7] Although both manufacturers employ similar dual energy X-ray technology, significant differences exist in their calibration standards, proprietary algorithms for calculating BMD and body composition, edge detection methods, and reference databases.[
8] Previous studies have demonstrated that Hologic systems typically report BMD values approximately 10% to 15% lower than GE Lunar systems at the spine and hip.[
9]
These inter-manufacturer differences pose significant challenges in clinical practice and research. Early efforts to address inter-manufacturer differences led to the development of standardized BMD equations by the International Committee for Standards in Bone Measurement (ICSBM) in 1994.[
10] However, these equations were derived from first-generation pencil-beam DXA systems, and subsequent studies have shown that they may not adequately eliminate differences between modern fan-beam systems. [
10,
11] Additionally, standardization equations have not been widely developed for body composition measurements, particularly for newer DXA models.[
12]
The GE Lunar Prodigy and Hologic Horizon W represent current-generation DXA systems widely used in clinical practice. While several cross-calibration studies have examined older models from these manufacturers, data comparing these specific systems remain limited, particularly in Asian populations.[
13] Given the differences in body composition and bone geometry between Asian and Western populations, population-specific conversion equations may be necessary for optimal accuracy.[
14]
Therefore, the aims of this study were: (1) to investigate the correlation and agreement between BMD and body composition measurements obtained using the GE Lunar Prodigy and Hologic Horizon W DXA systems in healthy Korean adults; and (2) to develop cross-calibration equations that can facilitate the comparison and conversion of measurements between these two systems.
METHODS
1. Study population
We conducted a prospective cross-calibration study at Chung-Ang University Hospital between August 2021 and October 2021. A total of 120 healthy volunteers (59 males and 61 females) who provided written informed consent were enrolled. The inclusion criteria were: (1) age between 20 and 70 years; (2) body mass index (BMI) between 18.5 and 35 kg/m2; and (3) ability to provide written informed consent. The exclusion criteria were: (1) underlying skeletal pathology such as scoliosis, osteoarthritis, diffuse idiopathic skeletal hyperostosis, or other clinical spinal deformities that could result in inaccurate BMD assessment; (2) previous history of adverse reaction after DXA scan or radiography; (3) pregnant or lactating women; (4) persons with artificial or implanted cardiac pacemakers; and (5) patients with mental illnesses such as severe depression that could affect their ability to cooperate during the examination. This study was approved by the Institutional Review Board (IRB) of Chung-Ang University Hospital (IRB No. 2021-023), and all participants provided written informed consent before enrollment.
2. DXA measurements
All participants underwent DXA scanning on both the GE Lunar Prodigy DXA system (GE Lunar Prodigy Advance; GE Medical Systems) and the Hologic Horizon W DXA system (Horizon W; Hologic Inc.) on the same day. The order of scanning on each system was randomized to minimize any potential systematic bias. All scans were performed according to each manufacturer’s standard positioning protocols.
For all scans, participants were positioned supine with lateral pronated hands by their side and both feet in 15 to 25 degrees of internal rotation. BMD measurements were obtained at the lumbar spine (L1–L4 vertebrae), bilateral femoral neck, trochanter, and total femur. Whole-body composition analysis was performed to obtain lean mass and fat mass values for bilateral upper extremities, bilateral lower extremities, trunk, and whole body. All measurements were automatically analyzed using each manufacturer’s standard software: GE Lunar enCORE version 17.0 for the GE system and Hologic Apex version 5.6 for the Hologic system. Daily quality assurance calibrations were performed on both systems according to manufacturer recommendations throughout the study period.
3. Statistical analysis
Descriptive statistics were presented as mean±standard deviation (SD) for continuous variables. Paired
t-tests were used to compare measurements between the two DXA systems. The agreement between the two systems was assessed using multiple methods. First, Bland-Altman analysis was performed by plotting the differences between the two measurements against their mean values, with the calculation of bias (mean difference) and 95% limits of agreement (±1.96 SD).[
15] Second, Pearson correlation coefficients were calculated to assess the linear relationship between measurements from the two systems, with correlations interpreted as weak (0.2–0.4), moderate (0.4–0.6), strong (0.6–0.8), or very strong (0.8–1.0).[
16] Third, Lin’s concordance correlation coefficient (CCC) [
17] was calculated to evaluate both the precision and accuracy of agreement, with CCC values graded as poor (<0.90), moderate (0.90–0.95), substantial (0.95–0.99), or almost perfect (>0.99).
Linear regression analysis was performed to derive conversion equations between the two systems, with measurements from one system as the independent variable and measurements from the other system as the dependent variable. Separate equations were developed for converting GE Lunar values to Hologic values and vice versa. All statistical analyses were performed using SPSS software version 26.0 (IBM Corp., Armonk, NY, USA). Statistical significance was set at P value less than 0.05.
RESULTS
1. Participant characteristics
The demographic characteristics of the 120 study participants are summarized in
Table 1. The mean age was 48.6± 14.4 years, with a range from 20 to 70 years. The mean weight, height, and BMI were 65.43±12.36 kg, 165.45± 8.88 cm, and 23.78±3.28 kg/m
2, respectively. Male participants (N=59) had a mean age of 47.6±13.9 years, weight of 72.70±11.75 kg, height of 172.16±5.30 cm, and BMI of 24.48±3.51 kg/m
2. Female participants (N=61) had a mean age of 49.4±14.9 years, weight of 58.40±8.22 kg, height of 158.95±6.46 cm, and BMI of 23.11±2.90 kg/m
2.
2. Comparison of BMD measurements
The BMD measurements from both DXA systems are presented in
Table 2. All BMD values measured by the Hologic system were significantly lower than those measured by the GE Lunar system (all
P<0.0001). At the lumbar spine, the mean BMD values ranged from 0.91±0.14 g/cm
2 (L1) to 1.01±0.17 g/cm
2 (L3) on the Hologic system, compared with 1.06±0.17 g/cm
2 (L1) to 1.22±0.20 g/cm
2 (L4) on the GE Lunar system. The mean difference for total L1–L4 BMD was 0.19 g/cm
2 (16.4%), with the Hologic system measuring lower.
At the femoral sites, the Hologic BMD values ranged from 0.65±0.09 g/cm2 (left trochanter) to 0.89±0.12 g/cm2 (right total femur), while the GE Lunar values ranged from 0.78±0.13 g/cm2 (bilateral trochanter) to 0.97±0.14 g/cm2 (total femur). The mean differences between the two systems at the femoral sites ranged from 0.078 g/cm2 (8.0%) at the total femur to 0.156 g/cm2 (17.1%) at the femoral neck.
Bland-Altman analysis revealed consistent negative bias (Hologic − GE Lunar) across all BMD sites, with the mean differences ranging from −0.21 to −0.15 g/cm
2 at the lumbar spine and from −0.16 to −0.08 g/cm
2 at the femoral sites. The 95% limits of agreement for total L1–L4 BMD were −0.09 to −0.29 g/cm
2 (
Fig. 1,
2).
3. Comparison of body composition measurements
Lean mass measurements from both systems are shown in
Table 3. Similar to BMD, the lean mass values measured by the Hologic system were systematically lower than those measured by the GE Lunar system at all sites (all
P<0.0001). The mean whole-body lean mass was 41,055±8,681 g on the Hologic system compared with 44,143±9,279 g on the GE Lunar system, representing a mean difference of 3,088 g (7.0%). Regional differences were observed, with the greatest relative difference at the lower extremities (mean difference of 730 g or 10.4% for the right leg and 529 g or 7.7% for the left leg) and the smallest at the trunk (mean difference of 805 g or 3.8%).
Bland-Altman analysis showed that the magnitude of the difference between the two systems increased with increasing mean lean mass values, suggesting proportional bias. This pattern was consistent across all body regions measured.
4. Correlation and agreement analysis
The correlation and concordance statistics are presented in
Table 4. Very strong Pearson correlations (r=0.944–0.984) were observed between the two systems for all BMD and lean mass measurements. For BMD, the correlations ranged from 0.944 (left femur trochanter) to 0.980 (L3 vertebra and right total femur). For lean mass, the correlations ranged from 0.963 (left lower extremity) to 0.984 (whole body).
Despite the strong correlations, the CCC showed greater variability, reflecting the systematic bias between the two systems. For BMD measurements, CCC values ranged from 0.505 (left femur trochanter) to 0.829 (right total femur), indicating poor agreement. The CCC was generally lower at the femoral neck and trochanter sites compared with the total femur and spine. For lean mass measurements, CCC values ranged from 0.854 (right upper extremity) to 0.946 (trunk), indicating poor to moderate agreement. The agreement was better for larger body regions (trunk and whole body) compared with smaller regions (extremities).
5. Cross-calibration equations
The conversion equations derived from linear regression analysis are presented in
Table 5. For converting GE Lunar BMD values to Hologic values, the equations followed the general form: Hologic=(slope×GE Lunar)+intercept. The slopes ranged from 0.63 to 0.86, reflecting the proportionally lower values obtained with the Hologic system. For example, the equation for total L1–L4 BMD was: Hologic=(0.81×GE Lunar)+0.037.
For lean mass conversions, the slopes ranged from 0.84 to 0.93 for converting GE Lunar to Hologic values. The equation for whole-body lean mass was: Hologic= (0.92×GE Lunar)+398. Reverse equations (Hologic to GE Lunar) are also provided for clinical scenarios requiring conversion in the opposite direction.
DISCUSSION
This prospective cross-calibration study provides comprehensive data on the relationship between BMD and body composition measurements obtained using the GE Lunar Prodigy and Hologic Horizon W DXA systems in healthy Korean adults. Our findings confirm the existence of significant systematic differences between these two widely used DXA systems while demonstrating very strong correlations that support the feasibility of cross-calibration. The conversion equations derived in this study can facilitate the comparison and transfer of DXA data between these systems in both clinical practice and research settings.
The observed differences in BMD measurements between the GE Lunar and Hologic systems are consistent with previous cross-calibration studies, although the magnitude of differences varies depending on the specific models and skeletal sites examined. In our study, the Hologic system consistently measured BMD values approximately 14% to 20% lower than the GE Lunar system at the lumbar spine and 8% to 17% lower at the femoral sites. These findings align with the well-documented observation that Hologic systems typically report lower BMD values than GE Lunar systems.[
9,
10]
The clinical implications of these inter-manufacturer differences are substantial. Given that the World Health Organization diagnostic criteria for osteoporosis are based on T-score thresholds derived from BMD measurements, the use of different DXA systems could lead to different diagnostic classifications for the same patient.[
18] A recent study by Analay et al. [
19] demonstrated that among patients diagnosed with osteoporosis on a Hologic system, up to 51% would be reclassified as having only osteopenia when measured on a Lunar system. This discrepancy underscores the critical importance of using appropriate cross-calibration when patients are monitored on different systems or when DXA data from different sources are pooled for research purposes.
An important finding of our study is the differential magnitude of between-system differences across anatomical sites. The smallest relative differences were observed at the total femur (8.0%), while the largest differences were at the femoral neck and trochanter (14%–17%) and lumbar spine (16%–18%). This pattern has clinical relevance because different skeletal sites are often used for different purposes—the femoral neck is typically used for fracture risk assessment with fracture risk assessment tool, while the spine may be preferred for monitoring treatment response due to its higher metabolic activity.[
20] The site-specific nature of the between-system differences highlights the importance of applying the appropriate conversion equation for each measurement site.
With the growing recognition of sarcopenia as a significant health concern in aging populations, the accurate assessment of lean mass has become increasingly important. [
5,
6] Our study demonstrates that the Hologic system consistently measures lean mass values 4% to 10% lower than the GE Lunar system across all body regions. These differences are clinically significant because they could affect the diagnosis of sarcopenia, which is based on cutoff values for appendicular lean mass.[
6] Previous studies have reported similar magnitude of differences in lean mass measurements between different DXA systems.[
12,
13] Park et al. [
13] recently published conversion equations between the same two DXA systems in Korean adults, and our results are generally consistent with their findings, providing cross-validation for the reliability of these conversion equations.
The observation that the between-system differences in lean mass increased with increasing mean values (proportional bias) has important implications. This pattern suggests that the relationship between the two systems is not simply a constant offset but involves both systematic and proportional components.
The very strong Pearson correlations (r>0.94 for all measurements) observed in our study indicate that the two systems are measuring the same underlying constructs with high fidelity. However, the CCC, which account for both precision and accuracy, were notably lower for several sites, reflecting the systematic bias between the systems. This discrepancy between correlation and concordance illustrates why simple correlation analysis is insufficient for method comparison studies.[
15,
16] The Bland-Altman approach and CCC provide complementary information about agreement that is essential for proper interpretation of cross-calibration data.
Several limitations of this study should be acknowledged. First, although our sample size of 120 participants is adequate for developing regression equations and is consistent with previous cross-calibration studies, larger samples would provide narrower confidence intervals for the conversion coefficients. Second, our study population consisted of healthy Korean adults, and the conversion equations may not be directly applicable to populations with different body compositions, ethnicities, or disease states. Third, while we included participants across a wide age range (20–70 years) and BMI range (18.5–35 kg/m2), extremely elderly individuals and those with morbid obesity were not represented. Fourth, we did not include T-score comparisons, which would require knowledge of the specific reference databases used by each manufacturer. Finally, our study examined only two specific DXA models, and the conversion equations may not be applicable to other models from the same manufacturers.
Despite these limitations, our study has several strengths. The prospective design with same-day scanning on both systems minimizes the confounding effects of biological variation that could occur if measurements were obtained on different days. The inclusion of both BMD and body composition measurements provides comprehensive cross-calibration data relevant to current clinical practice. The use of multiple statistical methods (paired t-tests, Bland-Altman analysis, Pearson correlation, and CCC) provides a thorough characterization of the agreement between the two systems.
This study demonstrates a very high correlation between BMD and body composition measurements obtained using the GE Lunar Prodigy and Hologic Horizon W DXA systems, despite the presence of significant systematic differences. The Hologic system consistently measures lower values for both BMD and lean mass compared to the GE Lunar system. The conversion equations derived in this study can facilitate the comparison and transfer of DXA data between these systems, which is valuable for longitudinal patient monitoring, multicenter clinical trials, and epidemiological research.
DECLARATIONS
-
Funding
This research was supported by a fund (2021-11-007) by Research of Korea Disease Control and Prevention Agency.
-
Ethics approval and consent to participate
The study protocol conformed to the ethical guidelines of the 1975 Declaration of Helsinki and was approved by the Institutional Review Board of the Chung-Ang University Hospital (IRB No. 2021-023). Written informed consent was obtained from all participants.
-
Conflict of interest
No potential conflict of interest relevant to this article was reported.
Fig. 1Bland-Altman plots showing the agreement between GE Lunar Prodigy and Hologic Horizon W dual energy X-ray absorptiometry systems for lumbar spine bone mineral density (BMD) measurements. The y-axis represents the difference (Hologic − GE Lunar) and the x-axis represents the mean of the two measurements. The solid line indicates the mean difference (bias), and the dashed lines indicate the 95% limits of agreement (±1.96 standard deviation [SD]). L, lumbar vertebra.
Fig. 2Bland-Altman plots showing the agreement between GE Lunar Prodigy and Hologic Horizon W dual energy X-ray absorptiometry systems for femoral bone mineral density (BMD) measurements. The y-axis represents the difference (Hologic − GE Lunar) and the x-axis represents the mean of the two measurements. The solid line indicates the mean difference (bias), and the dashed lines indicate the 95% limits of agreement (±1.96 standard deviation [SD]). U/E, upper extremity; L/E, lower extremity.
Table 1Demographic characteristics of the study participants
Table 1
|
Variables |
Total (N=120) |
Male (N=59) |
Female (N=61) |
|
Age (yr) |
48.6±14.39 |
47.6±13.90 |
49.4±14.91 |
|
Weight (kg) |
65.43±12.36 |
72.70±11.75 |
58.40±8.22 |
|
Height (m) |
165.45±8.88 |
172.16±5.30 |
158.95±6.46 |
|
BMI (kg/m2) |
23.78±3.28 |
24.48±3.51 |
23.11±2.90 |
Table 2Comparison of bone mineral density measurements between Hologic Horizon W and GE Lunar Prodigy dual energy X-ray absorptiometry systems
Table 2
|
BMD |
Hologic Horizon W |
GE Lunar Prodigy |
P-value |
|
|
|
Mean±SD |
Range |
Mean±SD |
Range |
|
L1 |
0.91±0.14 |
0.59–1.39 |
1.06±0.17 |
0.69–1.58 |
<0.0001a)
|
|
|
L2 |
0.97±0.15 |
0.61–1.53 |
1.14±0.18 |
0.72–1.74 |
<0.0001a)
|
|
|
L3 |
1.01±0.17 |
0.60–1.68 |
1.21±0.2 |
0.77–1.98 |
<0.0001a)
|
|
|
L4 |
1.00±0.17 |
0.64–1.58 |
1.22±0.2 |
0.78–1.91 |
<0.0001a)
|
|
|
Total L1–4 |
0.97±0.15 |
0.65–1.55 |
1.16±0.18 |
0.78–1.81 |
<0.0001a)
|
|
|
Right femur neck |
0.76±0.12 |
0.51–1.17 |
0.91±0.13 |
0.63–1.35 |
<0.0001a)
|
|
|
Right femur trochanter |
0.67±0.09 |
0.45–0.91 |
0.78±0.13 |
0.48–1.14 |
<0.0001a)
|
|
|
Right femur total |
0.89±0.12 |
0.65–1.20 |
0.97±0.14 |
0.71–1.32 |
<0.0001a)
|
|
|
Left femur neck |
0.76±0.12 |
0.50–1.08 |
0.91±0.13 |
0.58–1.29 |
<0.0001a)
|
|
|
Left femur trochanter |
0.65±0.09 |
0.43–0.89 |
0.78±0.13 |
0.50–1.14 |
<0.0001a)
|
|
|
Left femur total |
0.88±0.12 |
0.65–1.16 |
0.97±0.14 |
0.68–1.31 |
<0.0001a)
|
Table 3Comparison of lean mass measurements between Hologic Horizon W and GE Lunar Prodigy dual energy X-ray absorptiometry systems
Table 3
|
Lean mass |
Hologic Horizon W |
GE Lunar Prodigy |
P-value |
|
|
|
Mean±SD |
Range |
Mean±SD |
Range |
|
Right upper extremity |
2,067.76±690.57 |
994–4,489.3 |
2,451.93±802.57 |
1,208–5,307 |
<0.0001a)
|
|
|
Left upper extremity |
1,986.89±681.77 |
940–4,191.4 |
2,296.21±777.20 |
1,173–5,241 |
<0.0001a)
|
|
|
Right lower extremity |
6,294.36±1,561.85 |
3,902–10,436.6 |
7,024.79±1,642.30 |
4,222–10,953 |
<0.0001a)
|
|
|
Left lower extremity |
6,384.86±1,550.80 |
3,938–10,798.9 |
6,913.41±1,599.82 |
4,476–10,943 |
<0.0001a)
|
|
|
Trunk |
20,438.58±4,043.51 |
13,196–31,429 |
21,243.10±4,246.83 |
13,559–31,472 |
<0.0001a)
|
|
|
Whole body |
41,055.2±8,680.72 |
26,223–62,911.1 |
44,143.32±9,278.78 |
28,303–67,684 |
<0.0001a)
|
Table 4Correlation and concordance between GE Lunar Prodigy and Hologic Horizon W dual energy X-ray absorptiometry systems
Table 4
|
Pearson correlation (r) |
CCC |
95% CI |
|
BMD |
|
L1 |
0.959 |
0.652 |
0.585–0.711 |
|
L2 |
0.964 |
0.636 |
0.567–0.695 |
|
L3 |
0.98 |
0.594 |
0.526–0.654 |
|
L4 |
0.967 |
0.559 |
0.488–0.622 |
|
Total L1–4 |
0.975 |
0.584 |
0.515–0.645 |
|
Right femur neck |
0.956 |
0.537 |
0.466–0.602 |
|
Right femur trochanter |
0.958 |
0.597 |
0.527–0.659 |
|
Right femur total |
0.98 |
0.829 |
0.785–0.865 |
|
Left femur neck |
0.955 |
0.528 |
0.457–0.593 |
|
Left femur trochanter |
0.944 |
0.505 |
0.433–0.570 |
|
Left femur total |
0.977 |
0.756 |
0.701–0.802 |
|
|
Lean mass |
|
Right upper extremity |
0.978 |
0.854 |
0.814–0.886 |
|
Left upper extremity |
0.972 |
0.884 |
0.849–0.911 |
|
Right lower extremity |
0.967 |
0.875 |
0.836–0.904 |
|
Left lower extremity |
0.963 |
0.911 |
0.880–0.934 |
|
Trunk |
0.965 |
0.946 |
0.925–0.961 |
|
Whole body |
0.984 |
0.927 |
0.904–0.945 |
Table 5Cross-calibration equations between GE Lunar Prodigy and Hologic Horizon W dual energy X-ray absorptiometry systems
Table 5
|
Conversion equation (GE Lunar Prodigy to Hologic Horizon W) |
Conversion equation (Hologic Horizon W to GE Lunar Prodigy) |
|
BMD |
|
L1 |
Hol_L1=(0.82×GE_L1)+0.048 |
GE_L1=(1.13×Hol_L1)+0.030 |
|
L2 |
Hol_L2=(0.81×GE_L2)+0.050 |
GE_L2=(1.15×Hol_L2)+0.022 |
|
L3 |
Hol_L3=(0.83×GE_L3) − 0.004 |
GE_L3=(1.15×Hol_L3)+0.052 |
|
L4 |
Hol_L4=(0.83×GE_L4) − 0.009 |
GE_L4=(1.13×Hol_L4)+0.090 |
|
Total |
L1–4 Hol_L1–4=(0.81×GE_L1–4)+0.037 |
GE_L1–4=(1.18×Hol_L1–4)+0.014 |
|
Right femur neck |
Hol_RFneck=(0.86×GE_RFneck) − 0.026 |
GE_RFneck=(1.18×Hol_RFneck)+0.107 |
|
Right femur trochanter |
Hol_RFtro=(0.68×GE_RFtro)+0.137 |
GE_RFtro=(1.35×Hol_RFtro) − 0.121 |
|
Right femur total |
Hol_RFTotal=(0.86×GE_RFTotal)+0.062 |
GE_RFTotal=(1.12×Hol_RFTotal) − 0.031 |
|
Left femur neck |
Hol_LFneck=(0.85×GE_LFneck) − 0.019 |
GE_LFneck=(1.07×Hol_LFneck)+0.101 |
|
Left femur trochanter |
Hol_LFtro=(0.63×GE_LFtro)+0.154 |
GE_LFtro=(1.41×Hol_LFtro) − 0.131 |
|
Left femur total |
Hol_LFTotal=(0.82×GE_LFTotal)+0.081 |
GE_LFTotal=(1.16×Hol_LFTotal) − 0.050 |
|
|
Lean mass |
|
Right upper extremity |
Hol_RUE=(0.84×GE_RUE)+4 |
GE_RUE=(1.14×Hol_RUE)+102 |
|
Left upper extremity |
Hol_LUE=(0.85×GE_LUE)+29 |
GE_LUE=(1.11×Hol_LUE)+95 |
|
Right lower extremity |
Hol_RLE=(0.92×GE_RLE) −169 |
GE_RLE=(1.02×Hol_RLE)+622 |
|
Left lower extremity |
Hol_LLE=(0.93×GE_LLE) − 70 |
GE_LLE=(0.99×Hol_LLE)+569 |
|
Trunk |
Hol_Trunk=(0.92×GE_Trunk)+915 |
GE_Trunk=(1.01×Hol_Trunk)+522 |
|
Whole body |
Hol_WB=(0.92×GE_WB)+398 |
GE_WB=(1.05×Hol_WB)+941 |
REFERENCES
- 1. Kanis JA. Diagnosis of osteoporosis and assessment of fracture risk. Lancet 2002;359:1929-36.
https://doi.org/10.1016/s0140-6736(02)08761-5.
- 2. Blake GM, Fogelman I. The role of DXA bone density scans in the diagnosis and treatment of osteoporosis. Postgrad Med J 2007;83:509-17.
https://doi.org/10.1136/pgmj.2007.057505.
- 3. Bazzocchi A, Ponti F, Albisinni U, et al. DXA: Technical aspects and application. Eur J Radiol 2016;85:1481-92.
https://doi.org/10.1016/j.ejrad.2016.04.004.
- 4. Messina C, Albano D, Gitto S, et al. Body composition with dual energy X-ray absorptiometry: From basics to new tools. Quant Imaging Med Surg 2020;10:1687-98.
https://doi.org/10.21037/qims.2020.03.02.
- 5. Cruz-Jentoft AJ, Bahat G, Bauer J, et al. Sarcopenia: Revised European consensus on definition and diagnosis. Age Ageing 2019;48:16-31.
https://doi.org/10.1093/ageing/afy169.
- 6. Cawthon PM. Assessment of lean mass and physical performance in sarcopenia. J Clin Densitom 2015;18:467-71.
https://doi.org/10.1016/j.jocd.2015.05.063.
- 7. Shepherd JA, Fan B, Lu Y, et al. A multinational study to develop universal standardization of whole-body bone density and composition using GE Healthcare Lunar and Hologic DXA systems. J Bone Miner Res 2012;27:2208-16.
https://doi.org/10.1002/jbmr.1654.
- 8. Kendler DL, Borges JL, Fielding RA, et al. The official positions of the international society for clinical densitometry: Indications of use and reporting of DXA for body composition. J Clin Densitom 2013;16:496-507.
https://doi.org/10.1016/j.jocd.2013.08.020.
- 9. Genant HK, Grampp S, Glüer CC, et al. Universal standardization for dual X-ray absorptiometry: Patient and phantom cross-calibration results. J Bone Miner Res 1994;9:1503-14.
https://doi.org/10.1002/jbmr.5650091002.
- 10. Fan B, Lu Y, Genant H, et al. Does standardized BMD still remove differences between Hologic and GE-Lunar state-of-the-art DXA systems? Osteoporos Int 2010;21:1227-36.
https://doi.org/10.1007/s00198-009-1062-3.
- 11. Hull H, He Q, Thornton J, et al. iDXA, prodigy, and DPXL dual-energy X-ray absorptiometry whole-body scans: A cross-calibration study. J Clin Densitom 2009;12:95-102.
https://doi.org/10.1016/j.jocd.2008.09.004.
- 12. Pearson D, Horton B, Green DJ. Cross calibration of Hologic QDR2000 and GE Lunar Prodigy for whole body bone mineral density and body composition measurements. J Clin Densitom 2011;14:294-301.
https://doi.org/10.1016/j.jocd.2011.03.008.
- 13. Park SS, Lim S, Kim H, et al. Comparison of two DXA systems, Hologic Horizon W and GE Lunar Prodigy, for assessing body composition in healthy Korean adults. Endocrinol Metab (Seoul) 2021;36:1219-31.
https://doi.org/10.3803/EnM.2021.1274.
- 14. Ha YC, Yoo JI. Cross-calibration of bone mineral densities and body composition between GE Lunar Prodigy and osteosys primus. J Bone Metab 2021;28:215-21.
https://doi.org/10.11005/jbm.2021.28.3.215.
- 15. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986;1:307-10.
- 16. Lin LI. A concordance correlation coefficient to evaluate reproducibility. Biometrics 1989;45:255-68.
- 17. McBride GB. A proposal for strength-of-agreement criteria for Lin’s concordance correlation coefficient. Hamilton, NZ: National Institute of Water and Atmospheric Research; 2005. Report No.: NIWA Client Rep: HAM2005-062.
- 18. Kanis JA, McCloskey EV, Johansson H, et al. A reference standard for the description of osteoporosis. Bone 2008;42:467-75.
https://doi.org/10.1016/j.bone.2007.11.001.
- 19. Analay P, Kara M, Sertçelik A, et al. Diagnosing (severe) osteoporosis by Hologic vs. Lunar measurements: A single-center retrospective study. J Clin Densitom 2024;27:101509.
https://doi.org/10.1016/j.jocd.2024.101509.
- 20. Jankowski LG, Warner S, Gaither K, et al. Cross-calibration, least significant change and quality assurance in multiple dual-energy X-ray absorptiometry scanner environments: 2019 ISCD official position. J Clin Densitom 2019;22:472-83.
https://doi.org/10.1016/j.jocd.2019.09.001.