Comparison of narrow-angle fan-beam and pencil-beam densitometers: in vivo and phantom study of the effect of bone density, scan mode, and tissue depth on spine measurements.

This study compared the in vivo and in vitro performances of the Lunar MD and Prodigy dual-energy X-ray absorptiometers (DXAs). Ten volunteers and three different spine phantoms were studied to determine the effect of scan mode, tissue depth, and bone density on measures of spine bone area (BA), bone mineral content (BMC), and areal bone mineral density (BMD). These studies demonstrated that the choice of scan mode was most important for the Prodigy and for subjects who were thin, obese, or had low BMD. Increase in tissue depth caused an increase in measured BMC and BMD for the MD but had a small effect on Prodigy results if the appropriate scan mode was selected. BA was dependent on the BMD for both DXA systems. Results using a hydroxyapatite phantom demonstrated that after correcting for the calibration of Lunar systems, the BMC measured by the MD and Prodigy was similar to the calculated hydroxyapatite content of the phantom. In vivo studies confirmed the in vitro findings and demonstrated that even when the appropriate scan mode was selected, the BMC, BMD, and T-scores were significantly higher on the Prodigy than MD.     

Bone Mineral Measurements: A Comparison of Delayed Gamma Neutron Activation, Dual-Energy X-ray Absorptiometry and Direct Chemical Analysis

A system in vitro consisting of a femur from a cadaver and soft-tissue equivalent material was used to test the agreement between several techniques for measuring bone mineral. Calcium values measured by delayed gamma neutron activation (DGNA) and bone mineral content (BMC) by Lunar, Hologic and Norland dual-energy X-ray absorptiometers (DXA) were compared with calcium and ash content determined by direct chemical analysis. To assess the effect of soft-tissue thickness on measurements of bone mineral, we had three phantom configurations ranging from 15.0 to 26.0 cm in thickness, achieved by using soft-tissue equivalent overlays. Chemical analysis of the femur gave calcium and ash content values of 61.83 g ± 0.51g and 154.120 ± 0.004 g, respectively. Calcium measured by DGNA did not differ from the ashed amount of calcium at any of the phantom configurations. The BMC measured by DXA was significantly higher, by 3–5%, than the amount determined by chemical analysis for the Lunar densitometer and significantly lower, by 3–6%, for the Norland densitometer (p<0.001–0.024), but only 1% lower (not significant) for the Hologic densitometer. DXA instruments showed a decreasing trend in BMC as the thickness increased from 20.5 to 26.0 cm (p<0.05). However, within the entire thickness range (15.0–26.0 cm), the overall influence of thickness on BMC by DXA was very small. These findings offer insight into the differences in these currently available methods for bone mineral measurement and challenge the comparability of different methods. Received: 27 July 1998 / Accepted: 9 January 1999     

DEXA Measurement of spine density in the lateral projection. I: Methodology

Summary Bone mineral content and bone mineral density (BMC in g and BMD in g/cm²) were measured using dualenergy X-ray absorptiometry (DEXA). DEXA scans in the lateral decubitus position required about 12 minutes for the L2−L4 sequence at 0.75 mA (dose 5 mrem) and 4 minutes at 4.75 mA (7 mrem). The former scans were done with the Lunar DPX densitometer and the latter with the Lunar DPX-L One test of the algorithms used for measurement is the equality of BMC in both AP and lateral projections. BMC in the lateral projection averaged about 1% lower than in the AP projection in phantoms and for L2+L3 in 8 subjects, but the difference was not significant. Additional tests were done on the effects of tissue thickness and position from the tabletop. There was little or no influence of tissue thickness from 18 to 30 cm on BMD results, but there was a small influence of thickness below 18 cm (0.01 g/cm²;P=0.01) and of distance from the tabletop at extremes of positioning (0.02 g/cm²;P=0.06). The precisionin vivo was similar for both 4- and 12-minute scans; the standard deviation of repeat measurements was about 0.02 g/cm², which was about 2% relative to the mean BMD for a region within the vertebral body. The latter region included half the BMC of the body, or 24% of the entire vertebra. Results of 4-minute scans on the DPX and 12-minute scans on the DPX-L in 9 subjucts were highly correlated (r=0.98;P<0.001).     

Bone mineral and body fat measurements by two absorptiometry systems: Comparisons with neutron activation analysis

In 185 adults (68 white and 31 black males, 66 white and 20 black females), total body bone mineral density and content and body fat% were measured by two dual energy X-ray absorptiometry (DXA) systems—Norland XR-26, software version 2.4, and Lunar DPX, software versions 3.4 and 3.6. In a subgroup of 18 males (10 white, 8 black), body fat% and total body calcium were also measured by in vivo neutron activation analysis (IVNA). For total body calcium, the DPX 3.4 system gave the highest (1239 g), IVNA the lowest (1195 g), and the XR-26 (1226 g) was not significantly different from the DPX 3.6 results. For fat%, the XR-26 system gave the highest estimate (23.5%), whereas measurements by the DPX 3.4 and 3.6 systems (17.4 and 18.2%) were similar to the IVNA measurements (18.3%). BMD and BMC measurements by the two DXA systems were highly correlated but significantly different for the entire studied population except in the case of BMC in black males.     

Anomalies in the Measurement of Changes in Bone Mineral Density of the Spine by Dual-Energy X-ray Absorptiometry

Dual-energy X-ray absorptiometry (DXA) is frequently used for longitudinal studies of bone mineral status because of the high precision obtained, but evidence is emerging that the accuracy of measurements of changes may be a limitation because of artefacts of the analysis procedure, in particular, a dependence of the measured bone area (BA) on the bone mineral content (BMC). Results of spine bone mineral measurements taken at intervals with two DXA scanners, a Hologic QDR 1000W, and a Norland XR 26 HS, were examined. There was a consistent correlation between changes in BA and in BMC, with a slope of approximately 0.25 when expressed as percentages. A real change of BA of the magnitude observed is not feasible. There were no differences among the correlations for different instruments, genders, ages, or weight changes. There would appear to be an underestimation of changes in bone mineral density (BMD), but there is a possibility that some of the anomaly is manifested as an overestimation of a change in BMC. Phantom measurements were undertaken with the DXA scanners mentioned above and with a Lunar DPX. The phantoms consisted of simulations of the spine cut from aluminium sheet, so that the effective BMD could be varied. The dependence of the measured BA on BMC varied with the phantom outline, particularly the thickness of the transverse processes. Evidence was obtained of both an underestimate of BMD changes and an overestimate of BMC changes. There are errors in measuring spine changes, but these do not seem to be as serious as a previous report suggests for the Hologic scanner and are not likely to lead to misinterpretation of results. Received: 17 June 1997 / Accepted: 23 January 1998     

A Cross-Calibration Study of GE Lunar iDXA and GE Lunar DPX Pro for Body Composition Measurements in Children and Adults

Objective: To cross-calibrate dual energy X-ray absorptiometry machines when replacing GE Lunar DPX-Pro with GE Lunar iDXA. Methods: A cross-sectional study was conducted in 126 children (3–19 years) and 135 adults (20–66 years). Phantom cross calibration was carried out using aluminum phantom provided with each of the machines on both machines. Total body less head (TBLH), lumbar spine (L2–L4) and left femoral neck bone mineral density (BMD), bone mineral content (BMC), and bone area were assessed for each patient on both machines. TBLH lean and fat mass were also measured. Bland-Altman analysis, linear regressions, and independent sample t test were performed to evaluate consistency of measurements and to establish cross-calibration equations. Results: iDXA measured 0.33% lower BMD and 0.64% lower BMC with iDXA phantom as compared to DPX-Pro phantom (p < 0.001). In children, TBLH-BMC, femoral BMC and area were measured 10%–14% lesser, TBLH area was higher by 1%–2% and L2–L4 area by 10%–14% by iDXA as compared to DPX-Pro. iDXA measured higher TBLH fat [15% (girls), 31% (boys)] than DPX-Pro. In adults, TBLH-BMD (1.7%–3.4%), BMC (6.0%–10.9%) and area (4.2%–7.6%) were measured lesser by iDXA than DPX-Pro. L2–L4 BMD was higher [2.7% (men), 1.8% (women)] by iDXA than DPX-Pro. Femoral BMC was 2.11% higher in men and 4.1% lower in women by iDXA as compared to DPX-Pro. In children, R² of cross-calibration equations, ranged from 0.91 to 0.96; in adults, it ranged from 0.93 to 0.99 (p < 0.01). After the regression equations were applied, differences in BMD values between both machines were negligible. Conclusion: A strong agreement for bone mass and body composition was established between both machines. Cross-calibration equations need to be applied to transform DPX-Pro measurements into iDXA measurements to avoid errors in assessment. This study documents a need for use of cross-calibration equations to transform DPX-Pro body composition data into iDXA values for clinical diagnosis.     

Comparative evaluation of the DPX and DPX-IQ lunar densitometer systems following software upgrade.

Bone densitometry research departments perform system and software upgrades infrequently in order to maintain high precision. This study compares the results obtained on a Lunar densitometer with DPX, and DPX-IQ installed to achieve year 2000 compliance. The DPX-IQ provides an improved femur edge detection algorithm with an expanded reference database. Two hundred data files for each measurement site acquired on DPX were reanalyzed on DPX-IQ. There was no change to the bone mineral density (BMD), bone mineral content (BMC), T-scores or Z-scores for the L2-L4 spine, radius (ultradistal and 33%), and total body. There was a significant high correlation for the femoral neck BMD (r = 0.98; p < 0.05). The mean differences in BMD, BMC, T-scores, and Z-scores at the femoral neck and Ward's and trochanteric regions were not significant (p > 0.05). The limits of agreement within the 95% confidence interval for the femoral neck BMD using the Bland and Altman method was between -0.057 and 0.063 g/cm(2). This order of magnitude magnifies the long-term precision error and alters the usual confidence limits for interpretation of true change in densitometry practice. Therefore, it is important for reanalysis of DPX data files with the DPX-IQ to be performed so that longitudinal changes in BMD can be accurately assessed.     

The influence of aortic calcification on spinal bone mineral densityin vitro

We examined the influence of aortic calcification on the spine phantom bone mineral density (BMD). Soft X-ray photographs of human aortae were taken to calculate the percent calcification of aortic tissues. Human aorta laid on lumber spine phantom was placed in the bottom of 15 cm of water and BMD and bone mineral content (BMC) were measured in the anteroposterior view. Samples with severe aortic calcification (over 30%) caused a 2.5% increase of BMD. There might be a relatively small influence of aortic calcification on the value of L2-L4 BMD, but changes over time in a patient could falsely elevate values.     

Bone mineral density in prepubertal obese and control children: relation to body weight,lean mass,and fat mass

The aim of the study was to determine the influence of obesity on bone status in prepubertal children. This study included 20 obese prepubertal children (10.7 +/- 1.2 years old) and 23 maturation-matched controls (10.9 +/- 1.1 years old). Bone mineral area, bone mineral content (BMC), bone mineral density (BMD), and calculation of bone mineral apparent density (BMAD) at the whole body and lumbar spine (L1-L4) and body composition (lean mass and fat mass) were assessed by DXA. Broadband ultrasound attenuation (BUA) and speed of sound (SOS) at the calcaneus were measured with a BUA imaging device. Expressed as crude values, DXA measurements of BMD at all bone sites and BUA (69.30 versus 59.63 dB/MHz, P < 0.01) were higher in obese children. After adjustment for body weight and lean mass, obese children displayed lower values of whole-body BMD (0.88 versus 0.96 g/cm2, P < 0.05) and BMC (1190.98 versus 1510.24 g, P < 0.01) in comparison to controls. When results were adjusted for fat mass, there was no statistical difference between obese and control children for DXA and ultrasound results. Moreover, whole-body BMAD was lower (0.086 versus 0.099 g/cm3, P < 0.0001), whereas lumbar spine BMAD was greater (0.117 versus 0.100 g/cm3, P < 0.001) in obese children. Thus, it was observed that, in obese children, cortical and trabecular bone displayed different adaptation patterns to their higher body weight. Cortical bone seems to enhance both size and BMC and trabecular bone to enhance BMC. Finally, considering total body weight and lean mass of obese children, these skeletal responses were not sufficient to compensate for the excess load on the whole body.     

Poor bone health in underprivileged Indian girls: an effect of low bone mass accrual during puberty

A socio-economic gradient exists for most reasons of morbidity and mortality including delayed puberty in lower (LSES) as compared to higher (HSES) socio-economic stratum and puberty is an important factor affecting bone status in children and adolescents. Thus, a cross-sectional study was conducted on 195 age-matched pairs of girls (8-17years) from LSES and HSES in Pune City, India to assess the hypothesis that socio-economic factors working through late puberty would have a negative association with bone status of adolescents. Height, weight and Tanner stage were assessed. Total body bone mineral content (TBBMC), total body bone area (TBBA), total body bone mineral density (TBBMD), lean body mass (LBM) and total body fat mass (TBFM) were measured using GE Lunar DPX Pro Pencil Beam DXA (Wisconsin, USA) scanner. Mean TBBMC (1172±434g), TBBA (1351±356cm(2)), TBBMD (0.846±0.104g/cm(2)), LBM (21,622±5306g) and TBFM (7746±5194g) in LSES girls were significantly lower than that of HSES girls [TBBMC (1483±525g), TBBA (1533±380cm(2)), TBBMD (0.942±0.119g/cm(2)), LBM (24,308±5829g) and TBFM (12,196±7404g)] (p<0.01). There was a significant effect of age and puberty on all bone parameters. The differences in TBBMC, TBBA, LBM and TBFM between the 2 socio-economic strata at Tanner stage I were not significant (p>0.1) whereas there were significant differences in these parameters from Tanner stages II to V (p<0.05). The percentage difference between LSES and HSES girls in TBBMC, TBBA, TBBMD, LBM and TBFM was 3.4%, 0%, 3.7%, 0.2% and 17.3% respectively at Tanner stage I which increased to 19.1%, 9.7%, 10.4%, 8.8% and 31.2% respectively at Tanner stage V. In conclusion, our results suggest that pubertal years may provide a window of opportunity to promote bone health in adolescent girls from the lower socio-economic stratum.     

A Newly Recognized DXA Confounder: The Potassium-Binding Medication Sodium Zirconium Cyclosilicate

Objective: Radio-dense artifacts, including contrast material, alter dual-energy X-ray absorptiometry (DXA) results. An apparent diffuse artifact was identified during spine DXA acquisition in a patient without recent radiographic procedures. The patient reported taking sodium zirconium cyclosilicate (SZC; Lokelma®) 10 g 1 h before scanning. SZC is a potassium-binding agent recently marketed to treat hyperkalemia. Given the chemical composition, we hypothesized that SZC may alter DXA results. This study evaluated if SZC affects DXA results using an encapsulated spine and a total body phantom. Methodology: An encapsulated spine and total body phantom were scanned using a Lunar iDXA. Each phantom was scanned 5 times serially without repositioning in 5 configurations: (1) Bare, (2) 45 mL tap water, (3) 90 mL water, (4) 10 g SZC in 45 mL of water, and (5) 30 g SZC in 90 mL of water. Water and SZC was contained in plastic quart bags, folded, and placed over L2-3 on the spine phantom and flat over the pelvis/torso of the total body phantom. Results: Tap water did not change spine phantom measurements, but did increase (p < 0.05) total body phantom lean mass 46 g and 89 g with 45 mL and 90 mL, respectively. SZC 10 g or 30 g increased (p < 0.001) L2 and L3 bone mineral density (BMD) 18%–110%, mean 0.295 and 0.924 g/cm², respectively, while L1 and L4 BMD was statistically, but not clinically, altered by <0.010 g/cm². A dose-dependent change (p < 0.001) in total body phantom trunk measurements was demonstrated. The 10 g dose increased lean mass 16.8% and BMC 1%; fat mass was reduced 16.6%, while 30 g increased lean 41.9%, BMC 3.2%, and decreased fat 42.9%. Conclusion: SZC confounds BMD and body composition phantom measurements. It is likely that SZC alters DXA results in humans. DXA technologists and interpreters should be aware of this confounder.     

Dual X-Ray absorptiometry in pediatric studies: changing scan modes alters bone and body composition measurements.

The use of dual X-ray absorptiometry (DXA) for measurement of bone mineral and body composition in pediatric subjects faces a major technical issue: body size dictates choice of scan mode. However, different scan modes change results in the same subject, thus affecting the accuracy of bone/body composition measurements and especially the capacity to measure changes owing to either growth or intervention. To evaluate the effect of scan mode selections on measurements of bone mineral and body composition, 13 children with weights at the cutoff point between the pediatric large and adult medium scan modes of Lunar DPX or DPXL (Lunar, Madison, WI) with software 3.6 g (35.3 +/- 0.9 kg or 77.7 +/- 2.0 lb) were scanned by both modes. Adult medium mode gave significantly higher results than pediatric large mode for total body fat mass (11.1%), fat% (10.5%), bone mineral content (8.1%), and bone area (1.3%) (p < 0.02). The differences between pediatric large and adult medium modes in fat measurements increased with increasing body mass index ([BMI], kg/m(2)), body surface area ([BSA], m(2)), and trunk size (mm), whereas the differences in bone mineral measurement tended to be greater only with increasing BMI and BSA. None of the differences were correlated to body weight. This study suggests that scan mode selections based on trunk size, BMI, or BSA instead of body weight may improve continuity of bone and body composition measurements by the DXA technique in pediatric subjects.     

Dual-energy x-ray absorptiometry measurements of total-body bone mineral during weight change.

Bone mineral measurements were made using dual-energy X-ray absorptiometry during a multicenter diet trial.There were five centers, two using Hologic QDR4500 fan-beam scanners, two using Lunar Prodigy fan-beam scanners,and one using a pencil-beam Lunar DPX. Measurements were made at 0, 2.5, and 6 mo. The mean weight loss was 7.9 kg, but there was a wide range. With the Lunar instruments, the total-body bone mineral density reduced with weight loss, but with the Hologic scanners, it appeared to increase. This anomaly is similar to that observed previously with a Hologic QDR1000 pencil-beam scanner. It was shown that changes of fat distribution can lead to alterations in bone measurement without any real change in the skeleton. With all of the scanners, there was a strong correlation between the change in the bone mineral content and bone area, with some values of the latter being quite implausible. There was an associated worsening of long-term precision compared with that derived from short-term duplicated scans, more marked with the Lunar scanners. It is concluded that measurement artifacts preclude the valid assessment of total-body bone mineral during weight change.     

Bone mineral density in diabetic children and adolescents: a follow-up study

OBJECTIVE: To establish whether T1DM can affect bone mineral density (BMD) in children and adolescents. RESEARCH DESIGN AND METHODS: We performed a cross-sectional and longitudinal study of 57 diabetic children and adolescents and 57 normal controls. Total body and lumbar BMD and bone mineral content (BMC) were assessed by DXA (Lunar DPX) and volumetric transformation was calculated using the Katzman formula for total body BMD (BMAD) and using the Kroger formula for Lumbar BMD (L2L4BMDvol). BMC, BMAD, BMDspine, and L2L4BMDvol were adjusted for confounding factors such as age, gender, BMI, height, weight, and pubertal stage. RESULTS: BMDspine in the control group increased by 0.006 (g/cm(2))/year; while in the 39 diabetic patients longitudinally studied, it dropped by 0.006 (g/cm(2))/year during a follow-up period of 51 +/- 27 months. The average time spent weekly doing physical activity resulted in T1DM group directly correlated to BCM (P < 0.001) and inversely correlated with BMDspine (P < 0.05) and L2L4BMDvol (P < 0.01). L2L4BMDvol resulted significantly correlated with previous BMD spine (R = 0.63; P < 0.0001) and BMC evaluation (R = 0.42; P < 0.01) but not with BMAD. A second lumbar DXA evaluation performed in 38 patients after 1.00 +/- 0.16 years confirmed a small but significant decrease of 1.6% per year in L2L4BMDvol. The percentage of variation of L2L4BMDvol between the two evaluations was not correlated with the level of metabolic control, insulin requirement, and duration of the disease. Patients with complications showed similar L2L4BMDvol to patients without complications. CONCLUSIONS: Diabetic children and adolescents show a slight negative pattern of spine mineralization, which does not depend on metabolic control and microvascular complications.     

The relative contributions of lean tissue mass and fat mass to bone density in young women

Although obesity is associated with increased risk of many chronic diseases including cardiovascular disease, diabetes, hypertension, and cancer, there is little evidence to suggest that obesity increases risk of osteoporosis. In fact, both weight and body mass index (BMI) are positive predictors of bone mass in adults, suggesting that those who are overweight or obese may be at lower risk of osteoporosis. However, recent evidence suggests that in children and adolescents, obesity may be associated with lower rather than higher bone mass. To understand the relation of fat mass to bone mass, we examined data gathered from an ethnically diverse group of 921 young women, aged 20-25 years (317 African Americans, 154 Asians, 322 Caucasians, and 128 Latinas) to determine how fat mass (FM) as well as lean tissue mass (LTM) is associated with bone mass. Bone mass, FM, and LTM were measured using dual energy X-ray absorptiometry (GE Lunar Corp, Madison, WI). Bone mass was expressed as bone mineral density (BMD; g/cm2) and bone mineral apparent density (BMAD; g/cm3) for the spine and femoral neck, and as BMD and bone mineral content (BMC; g) for the whole body. Regression techniques were used to examine the following: (1) in separate equations, the associations of LTM and FM with each bone mass parameter; and (2) in the same equation, the independent contributions of LTM and FM to bone mass. LTM and FM were positively correlated with BMD at all skeletal sites. When the contributions of FM and LTM were examined simultaneously, both FM and LTM continued to be positively associated with bone mass parameters but the effect of FM was noted to be smaller than that of LTM. We conclude that in young women, LTM has a greater effect than fat mass on bone density per kg of tissue mass.     

Axial and Total-Body Bone Densitometry Using a Narrow-Angle Fan-Beam

We assessed a new dual-energy bone densitometer, the PRODIGY, that uses a narrow-angle fan-beam (4.5°) oriented parallel to the longitudinal axis of the body (i.e., perpendicular to the usual orientation). High-resolution scans across the body can be stepped at 17 mm intervals. The energy-sensitive array detector uses cadmium zinc telluride, which allowed rapid photon counting. Spine and femur scans required 30 s, and total-body scans required 4–5 min; the dose was only 3.7 mrem and 0.04 mrem respectively, or about 5 to 10 times lower than conventional fan-beam densitometry. We found only a small influence of soft-tissue thickness on bone mineral density (BMD) results. There was also a small ( ± 1%) influence of height above the tabletop on BMD results. A software correction for object height allowed a first-order correction for the large magnification effects of position on bone mineral content (BMC) and area. Consequently, the results for BMC and area, as well as BMD, with PRODIGY corresponded closely to those obtained using the predecessor DPX densitometer, both in vitro and in vivo; there was a generally high correlation (r= 0.98–0.99) for BMD values. Spine and femur values for BMC, area and BMD averaged within 0.5% in vivo (n= 122), as did total-body BMC and BMD (n= 46). PRODIGY values for total-body lean tissue and fat also corresponded within 1% to DPX values. Regional and total-body BMD were measured with 0.5% precision in vitro and 1% precision in vivo. The new PRODIGY densitometer appears to combine the low dose and high accuracy of pencil-beam densitometry with the speed of fan-beam densitometers. Received: 2 April 1999 / Accepted: 27 July 1999     

Do Textiles Impact DXA Bone Density or Body Composition Results?

External artifacts can confound dual-energy X-ray absorptiometry (DXA) measurements. It is often accepted that garments free of metal do not affect DXA results; however, little data exist in this regard. It is plausible that some textiles absorb radiation and thereby alter DXA results. We hypothesized that some dense or synthetic textiles, for example, reflective materials, might alter DXA-measured bone and soft tissue mass. Hologic and GE Lunar spine phantoms and a Bioclinica prototype total body phantom were imaged on a GE Lunar iDXA and Prodigy densitometer. Each phantom was scanned 10 times to establish mean values. Subsequently, 2 layers of various fabrics were placed over the entire top surface of the phantom, and 10 scans were performed without repositioning. Samples of natural, synthetic, or embellished fabric (including those with reflective material) and of varying thickness were used. Wilcoxon signed rank tests were used to compare the means between bare phantom and textile-covered phantom. Significant differences were demonstrated often, depending on the scanner, phantom, and textile used. A polyester fabric with reflective strip consistently altered measurements. For example, this fabric increased measured mean lumbar spine bone mineral density and total body bone mineral content by 0.008?g/cm² and 3.6?g, respectively (p?<?0.01). Similarly, mean total body fat decreased (?173 g) and lean mass increased (+213 g; p?<?0.01). Fat and lean mass were also affected by metallic thread, wool, blend denim, and shiny polyester (p?<?0.05), and lean mass was affected by cotton denim and sweatshirt material (p?<?0.0003). In conclusion, textiles can affect DXA-measured bone mineral density and body composition results. Even small amounts of reflective material could alter mass measurements by ~25% of the least significant change. Clothing made of dense textiles (e.g., wool and denim) or those with reflective material and metallic thread should be avoided during DXA scanning.     

Precision and sensitivity of dual-energy x-ray absorptiometry in spinal osteoporosis.

This study evaluated the performance of dual-energy x-ray absorptiometry (DEXA) with regard to (1) the correlation with dual-photon absorptiometry (DPA), (2) the ability to discriminate between normal and osteoporotic patients, and (3) long-term reproducibility. The bone mineral density (BMD) of the spine in 112 subjects, both normal and osteoporotic, was measured with DPA and DEXA (Lunar Corporation, Madison, Wisconsin) of the spine. The femur BMD of 22 cases was also measured with both machines. The results for the two techniques were highly correlated (r greater than 0.9, SEM = 0.02 to 0.04 g/cm2). BMD was measured using DEXA in 80 women (mean age = 61 years) with established spinal osteoporosis and 110 normal age-matched controls. The osteoporotic patients had significantly reduced spine and femur BMDs compared to the controls: -23% for L2-4 BMD (Z score = 2.6) and -13 to -20% for femur BMD (Z score = 1.1-1.3). L2-4 BMD had the best discriminative value, with an area under the ROC curve of 94%; the Ward's triangle BMD had an area of 84%. The precision error in vitro in a phantom over a 1-year period was 0.7%. The measured precision in vivo with young adults was approximately 1% (SD = 0.012 g/cm2) for L2-4 BMD and 1.7-2.3% (SD = 0.015-0.022 g/cm2) for femur over the 1 year period. The reproducibility was not as good for osteoporotic patients (SD = 0.017 g/cm2).     

Impact of using the ultradistal radius region of interest on diagnostic classification.

World Health Organization (WHO) criteria using T-scores for classifying patients as normal, osteopenic, or osteoporotic are based on bone mineral density (BMD, g/cm2) of the lumbar spine and hip and bone mineral content (BMC) (BMC, g) at the distal and midradius. There is no consensus on whether other forearm regions of interest (ROIs) can be used with the WHO criteria. Because the ultradistal radius region of interest (UDR) has a greater ratio of trabecular to cortical bone than midshaft portions of the radius, it is possible that more patients would be classified as osteoporotic if the UDR is measured. The objective of this study was to determine the prevalence of osteoporosis when using T-scores from the UDR in addition to PA lumbar spine, proximal femur (hip), and the radius 33% ROI. Retrospective data were obtained from three centers with differing patient demographics, thus reducing bias as a result of patient characteristics. Data were used only from patients who had a spine, hip, and forearm scan on the same day. Central dual-energy X-ray absorptiometry (DXA) systems included a GE Lunar DPX-L, DPX IQ, and Prodigy and a Hologic Delphi. Hologic data were for the ultradistal radius + ulna ROI (UDRU). Diagnostic classification (using the WHO T-score criteria) was made excluding and including the UDR and UDRU T-scores, in addition to lumbar spine (L2-L4 or L1-L4), hip (femoral neck, greater trochanter, or total), and the radius 33% ROI. The lowest T-score from any ROI determined the classification. For all GE Lunar patients (n = 409 women; age range: 20-96 yr), the distribution of normal, osteopenic, osteoporotic not using the UDR was 94 (23%), 170 (42%), and 145 (36%), respectively. The distribution when using the UDR was 67 (16%), 137 (33%), and 205 (50%), respectively. The difference in the ratio of normal + osteopenic versus osteoporotic when excluding and including the UDR T-scores was significant (p < 0.0001; two-tailed Fisher's exact test). For all Hologic patients (n = 153 women; age range: 44-93 yr), the distributions were 32 (21%), 66 (43%), and 55 (36%) not using and 31 (20%), 64 (42%), and 58 (38%), respectively, using the UDRU (not statistically significantly different). The group mean T-scores were lowest for the UDR compared to the spine and hip with GE Lunar but not Hologic patients.     

Performance evaluation of a dual-energy X-ray bone densitometer

Summary We tested a dual-energy bone densitometer (LUNAR DPX) that uses a stable x-ray generator and a K-edge filter to achieve the two energy levels. A conventional scintillation detector in pulse-counting mode was used together with a gain stabilizer. The densitometer normally performs spine and femur scans in about 6 minutes and 3 minutes, respectively, with adequate spatial resolution (1.2×1.2mm). Total body scans take either 10 minutes or 20 minutes. The long-term (6 months, n=195) precision of repeat measurement on an 18-cm thick spine phantom was 0.6% at the medium speed. Precision errorin vivo was about 0.6, 0.9 and 1.5% for spine scans (L2-L4) at slow, medium and fast speeds, while the error was 1.2 and 1.5 to 2.0%, respectively, for femur scans at slow and medium speed. The precision of total body bone density was 0.5%in vitro andin vivo. The response to increasing amounts of calcium hydroxyapatite was linear (r=0.99). The densitometer accurately indicated (within 1%) the actual amount of hydroxyapatite after correction for physiological amounts of marrow fat. The measured area corresponded exactly (within 0.5%) to that of known annuli and to the radiographic area of spine phantoms. There was no significant effect of tissue thickness on mass, area, or areal density (BMD) between 10 and 24cm of water. The BMD values for both spine and femurin vivo correlated highly (r=0.98, SEE=.03 g/cm²) with those obtained using conventional¹⁵³Gd DPA. Similarly, total body BMD correlated highly (r=0.96, SEE=.02g/cm²) with DPA results.