Most tests differ according to which bones are measured to determine the BMD result. DXA is currently the most widely used, but quantitative ultrasound QUS has been described as a more cost-effective approach to measure bone density. The density of these bones is then compared with an average index based on age, sex, and size. The resulting comparison is used to determine risk for fractures and the stage of osteoporosis if any in an individual. Results are generally scored by two measures, the T-score and the Z-score. Scores indicate the amount one's bone mineral density varies from the mean.
Negative scores indicate lower bone density, and positive scores indicate higher. The T-score is the relevant measure when screening for osteoporosis. It is the bone mineral density BMD at the site when compared to the young normal reference mean. It is a comparison of a patient's BMD to that of a healthy year-old. The US standard is to use data for a year-old of the same sex and ethnicity, but the WHO recommends using data for a year-old white female for everyone. The Z-score is the comparison to the age-matched normal and is usually used in cases of severe osteoporosis.
This is the number of standard deviations a patient's BMD differs from the average BMD of their age, sex, and ethnicity. This value is used in premenopausal women, men under the age of 50, and in children. In this setting, it is helpful to scrutinize for coexisting illnesses or treatments that may contribute to osteoporosis such as glucocorticoid therapy, hyperparathyroidism, or alcoholism. From Wikipedia, the free encyclopedia. J Am Osteopath Assoc. Archived from the original on Retrieved University of Washington Bone Physics. American College of Radiology.
A total of 22 subjects enrolled as placebo controls in an osteoporosis treatment study at the University of California, San Francisco UCSF comprised the precision study group. Ten of the 22 women received three hip BMD scans over a two year interval while 12 of the women received four hip BMD scans over a three year interval. The interobserver variability for BMD estimates using CTXA was estimated by comparing results obtained independently on the same in vivo data set by two trained operators.
Data from a single clinical trial site were used for this analysis. Twenty eight studies were analyzed, and the results for estimated BMD of the femoral neck and total hip regions of interest were compared.
Interest in bone densitometry methods has recently experienced a resurgence within the medical community. Physicians have become more interested than. The objective of this paper is to propose the methodology which determines the . dual X-ray absorptiometry (DXA) and peripheral quantitative computed.
Thirty patients were recruited at one center and 39 patients at the other. Seven patients were men and 62 were women.
Ages ranged from 20—80 years, but the patients were predominantly postmenopausal women. All CT scanners were maintained as specified by the manufacturers. Subjects were positioned supine on the CT scanner table, lying on top of a K 2 HPO 4 CT calibration phantom and bolus bags so that the calibration phantom extended from the lumbar vertebrae to mid-thigh, to cover the pelvis and proximal femur region. Positioning was used so that the pelvis was as straight as possible and the knees were flat on the scanner table.
Subjects were asked to put their feet together and remain still, but the feet were not restrained by a positioner.
An anterior-posterior computed radiograph was obtained by the scanner from the iliac crest to mid-thigh, and the top of the femoral head to approximately 1 cm below the inferior extent of the lesser trochanter was defined graphically to define the scanning region. A contiguous series of scans was obtained, 3 mm thick every 3 mm, with a 40 cm display field-of-view 0. Typically 40 images were obtained, with the time to acquire this image set approximately 3.
Scanning parameters varied slightly depending on the capabilities of the CT scanner used and patient size, and were 80 kVp, mAs for the GE, and kVp, — mAs for the Philips and ProSpeed scanners. All subject and QA data were sent to Mindways where analysis was centralized. DXA image data were acquired and analyzed at each site according to standard procedures used at those sites, including daily calibrations for quality control. A single individual at each site was responsible for all DXA analyses.
The DXA systems were calibrated and maintained in accordance with the manufacturer's specifications.
The DXA data were acquired and analyzed according to the manufacturer's instructions. Each QA study of 8—10 QA images acquired using the same technique as for subject scans was analyzed to determine CT scanner performance characteristics, and any deviations from expected performance were identified by the software. Any degradation of scanner performance identified by the QA software was resolved before subsequent subject data were analyzed.
Subject results were referenced to the appropriate CT scanner QA results. CT image data were analyzed in a standardized fashion with the CTXA Hip software, using the left proximal femur unless pathology prevented this. A square box region of interest was centered over the femoral neck as identified on the axial images, and a volumetric region of interest containing the proximal femur was extracted from the CT image data set for analysis.
Segmentation of bone from surrounding non-bone tissue was performed using an adaptive algorithm controlled by three parameters. The second parameter defined the size of the neighborhood of pixels considered when adaptively modifying the local threshold. A default range corresponds to a neighborhood with a width of approximately five pixels. This 3D data set of bone voxels was then rotated such that the femoral shaft was vertical in the coronal and sagittal planes and the femoral neck was horizontal in the axial plane Figure 2.
A axial, B sagittal and C coronal images of segmented bone in proximal femur, rotated into a standard projection. The CTXA Hip software generates a 2-dimensional image similar to a DXA image from the rotated 3D data set by summing all the bone voxels along lines perpendicular to the coronal plane. Each pixel of the resulting image represented the mass of mineral summed along that line, and was further characterized by a known pixel area, and a total volume of bone along the line. The lower extent of the Intertrochanter ROI was set at the lower junction of the lesser trochanter and the femoral shaft.
The angle of the femoral neck axis, and the position and size of the femoral neck box ROI, were adjusted by the operator as required. CTXA projected image with standard regions of interest used for BMD calculations femoral neck, trochanter, intertrochanter, and Total Hip as sum of these three regions.
Position of femoral neck box and intertrochanter limit line at base of lesser trochanter, and rotation of femoral neck axis, are adjustable by user. Ward's Triangle ROI is displayed but not used in comparisons.
Compartmental analysis of the ROIs was performed for the group of 22 women recruited for the precision study as a homogeneous population of postmenopausal women identified as osteoporotic by DXA. The mean and standard deviation for the 3 or 4 measurements for each patient were calculated without regard for the rate of bone change. Long-term precision was then estimated by calculating the root-mean-square average of the set of standard deviation estimates for the group of patients . For interobserver variability, significance of difference of means was tested using a two-tailed t-test.
Results were first compared by individual site. Similarity of the distributions suggested pooling the results. Pooling of these results was objectively justified based on a two-sample t-test, assuming unequal variances, of the means of the bias distribution for each site. To ascertain the validity of t-tests, an Anderson-Darling test was used to detect significant deviations from normality in measurement distributions.
No substantial evidence was found for rejecting the hypothesis that the data measurements are reasonably described by a normal sampling process using the Anderson-Darling test to detect significant deviations from normality in measurement distributions. Figures 4 and 5 show the correlations for total hip and femoral neck graphically. Correlation coefficients of 0.
This is particularly important in the case of follow-up examinations. As shown below, by the time women reach age 80, very few are considered normal. Quantitative Computed Tomography. Table model DXA machines can measure BMD at the hip or spine but can also be used to measure the total amount of mineral in the whole skeleton or forearm. Email this NetForum article. The Official Positions that are new or revised since are in bold type. T Score The difference in number of SDs between the value for an individual and the mean value of a group of young usually to year-old adults of the same sex.
For the study population, the total hip region of interest was found to contain For the femoral neck the proportions were The focus on measurements of bone density at the proximal femur as a standard reference  has meant DXA has become the gold standard technology used to make these measurements. Long term in vivo precision is an important parameter for clinical practice, and we obtained results for CTXA Hip essentially identical to those from a Hologic QDR system for osteoporotic patients studied under controlled conditions over a 2—3 year period.
Our long-term precision in osteoporotic subjects is similar to short-term precision of 2. Our precision results are consistent with the observations of Khoo et al  where CTXA short-term precision estimates that were either non-inferior to or superior to DXA were reported. However, a limitation of the present study is that the results from only two observers where available for the reproducibility analysis.
This bias is well modeled by an additive negative bias term as shown in Figure 4 where the observed slope in the linear regression analysis was found to be not significantly different from unity. In this case, however, linear regression analysis indicated the bias was better explained by a model slope significantly different than one with an additive offset not significantly different from zero. Biases in hip BMD estimates between bone densitometers from various manufacturers have been reported in numerous studies.
Biases of the same magnitude we report here have been observed in comparison studies of DXA devices from Hologic, Lunar and Norland  , . As noted by those authors, these differences may be due to technical differences in the way data are acquired and analyzed. The difference is more pronounced when working at higher densities. Essentially the soft-tissue component is subtracted in DXA. Next, DXA standardizes measurement positioning by controlling foot positioning during scanning.
While the standard DXA foot positioning does result in turning the femoral neck axis outward such that the femoral neck axis is more nearly orthogonal to the x-ray projection direction, the orthogonality of the femoral neck axis to the projection direction is not what is controlled with DXA. This orthogonality is, however, what is being controlled within CTXA. Another cause of differences is projection geometry.
CTXA uses a parallel beam projection geometry. While older DXA units also use a parallel projection geometry, most DXA units today use fan beam projection geometries that include a depth-dependent magnification attribute. Depending upon DXA unit design, there can also be a residual magnification component related to how far a patient's bone sits above, say, the DXA device table top. All of these are on top of algorithmic variations in core image processing steps used to define various anatomical landmarks and reference lengths used to standardize BMD measurement on a particular device.
Well-established biases in BMD estimates from different devices are currently handled in clinical densitometry practice by reporting normalized BMD estimates and interpreting normalized proximal-femur BMD scores from all DXA units using the same guidelines.