Absolute Relative and Attributable Risk
As noted previously, incidence risk is also known as absolute risk. The term absolute risk will be used here. This is the number of individuals developing a disease within a specified period of time divided by the number of individuals at risk for the disease at the beginning of the time interval. In the context of this discussion, a certain level of bone density determines risk. The disease is fracture. For example, assume that 1000 women with the same level of bone density in the femoral neck...
Info Vgg
Reprinted from ref. 12 with permission from the American Society for Bone and Mineral Research. Reprinted from ref. 12 with permission from the American Society for Bone and Mineral Research. Knowledge of the frequency of anomalous vertebral segmentation, the characteristic shapes created by the posterior lumbar elements on a PA lumbar spine study and the expected incremental change in BMC and BMD can be used to label the vertebrae correctly. If the vertebrae are mislabeled, comparisons to the...
Standardization of DXA BMD Results for the Femoral Neck Trochanter and Wards
In 2001, Lu et al. 13 developed equations for an sBMD for the femoral neck, trochanter, and Ward's area based on information obtained from studies of the same 100 women whose data was previously used to create formulas for the sBMD of the spine and total femur 10,12 . The authors developed site-specific standardization formulas for the hip subre-gions. They compared the utility of the subregion formulas in reducing the disparity in BMD results among the three manufacturers' devices to that of...
Quantitative Data
Quantitative data is basically numerical data. Within this broad category, numerical data may be continuous or discrete. Bone density data, for example, is continuous quantitative data, whereas the number of bone density measurements that a patient may have had is discrete quantitative data. Continuous data is numerical data that can be any fractional value on an interval scale. Height, weight, and blood pressure, as well as bone density, are all types of continuous quantitative data. Any of...
Qualitative or Categorical Data
If data are classified as belonging to categories, the data are considered qualitative data. Such data are discrete and generally consist of counts or the number of individuals belonging to each category. This type of data is also called categorical data. Categorical data can be ordinal or nominal, depending on whether there is a logical order to the categories. For example, individuals might be placed into height categories of tall, medium or short remember that the actual measurement of...
Hologic DXA Lunar DXA and Lunar DPA
The general relationship between Lunar DP4 values, Hologic QDR-1000, and Lunar DPX values is summarized in a study from McClung and Roberts 6 . Ninety-three subjects underwent bone density measurement on all three machines at the PA spine and proximal femur. The ratio of mean values obtained for each combination of machines is shown in Table 5-1. Note that the PA spine values obtained with either DXA device are lower than those obtained with the DP4. The three regions in the proximal femur are...
Ankylosing Spondylitis
Low bone density has been frequently observed in ankylosing spondylitis although its etiology remains uncertain. For 2 years Maillefert et al 23 followed 54 patients with ankylosing spondylitis to determine the prevalence of osteopenia and osteoporosis and the relationship of any observed bone loss to therapy, physical impairment, or inflammation. There were 35 men and 19 women in the study with an average age of 37.3 years and average disease duration of 12.4 years. In 23 patients, the disease...
Artifacts in PA or AP Spine Densitometry
The PA lumbar spine has been, and continues to be, used extensively in densitometry for diagnosis, fracture prediction, and monitoring. Unfortunately, it is also the skeletal site most often affected by structural changes and artifacts that may limit its utility. The BMD of a fractured vertebra will be increased because of the fracture itself. This increase in density could erroneously lead the physician to conclude that the bone strength is better and the risk for fracture, lower, than is the...
Info Ioc
Fig. 3-6. The relationship between absolute risk, relative risk, and the standard deviation score. Relative risk is the ratio of two absolute risks. way of putting this would be to say what is the second group's risk relative to the first group's risk As shown in Fig. 3-6, the group with the SD score of -1 has an absolute risk for fracture of 4 . The group with the SD score of 0 has an absolute risk for fracture of 2 . The relative risk is therefore 4 2 or 2. An appropriate interpretation of...
Standardization of Central DXA Absolute BMD Values
In November 1990, the major manufacturers of DXA equipment agreed to work together in the area of standards as part of an international committee, known as the International Committee for Standards in Bone Measurement. Under the auspices of this committee, a study of 100 healthy women 10 was performed in which each of the women underwent PA spine and proximal femur studies on the Hologic QDR-2000, the Norland XR-26 Mark II, and the Lunar DPX-L. The women ranged in age from 20 to 80, with an...
dxa from lunar to hologic to norland
The central DXA devices from all the major manufacturers have inherently superb accuracy and precision in quantifying the bone density at virtually any skeletal site. The BMD obtained on any one machine, however, will not be identical to the BMD obtained on a device from a different manufacturer. Comparison studies using combinations of machines provide conversion equations for the measurement of bone density on one device to the anticipated value on another device. The Hologic QDR-1000, the...
Hologic DXA and Norland DXA
In vivo comparisons of spine measurements made using the Norland XR-26 and the Hologic QDR-1000 were conducted by Lai et al. 8 in 65 subjects. The correlation for BMD at the spine was 0.990 and was highly significant. BMDs obtained on the Norland XR-26 tended to be lower than those obtained on the QDR-1000. The equation for predicting the Hologic BMD from the measurement of BMD on the Norland XR-26 was Hologic QDR 1000 SpineBMD -0.1 1.09 x Norland XR-26 SpineBMD
regression analysis
Regression analysis is the development of a mathematical model based on real-world observations or measurements. The model is intended to describe mathematically the relationship between an independent variable or variables and a dependent variable. There are different types of regression analyses and a variety of reasons to employ them. No matter what the circumstance, an important caveat to remember about such mathematical models is, All models are wrong. Some are useful 9 . The model is,...
reference databases
Two of the most common applications of bone densitometry are the diagnosis of osteoporosis and the assessment of fracture risk. These applications depend on comparisons of the BMD to the reference database that is supplied by the manufacturer of the bone densitometry equipment. The diagnosis of osteopenia or osteoporosis using WHO criteria depend on comparing the patient's BMD to the average peak BMD of the young adult and noting how many SDs below this value the patient's value lies. In other...
J.bone Miner Res 1994
1. Kelly TL, Slovik DM, Schoenfeld DA, Neer RM. Quantitative digital radiography versus dual photon absorptiometry of the lumbar spine. J Clin Endocrinol Metab 1988 67 839-844. 2. Pacifici R, Rupich R, Vered I, Fischer KC, Griffin M, Susman N, Avioli LV. Dual energy radiography DER a preliminary comparative study. Calcif Tissue Int 1988 43 189-191. 3. Holbrook TL, Barrett-Connor E, Klauber M, Sartoris D. A population-based comparison of quantitative dual-energy X-ray absorptiometry with...
Acromegaly
Forty-five subjects 24 women and 21 men with acromegaly for an average of 11.4 years underwent PA spine, proximal femur, and total body bone density studies with DXA Lunar DPX 21 . The subjects ranged in age from 21 to 77 years with a median age of 43 years. Twenty-five individuals were Caucasian and 20 were Black. Twenty percent of the individuals had age-matched z-scores in the spine of -1 or poorer, whereas only 8.8 had similar age-matched z-scores in the proximal femur. Osteopenia in the...
Standardization of Forearm DXA Results
The difficulty in creating an sBMD for forearm bone density is compounded by the multitude of potential sites in the forearm and the different definitions of those sites among the various manufacturers of forearm measurement devices. Shepherd and colleagues 14 have developed standardization equations for the ultradistal, mid, and proximal forearm for six devices, five of which employ DXA. The sixth device utilized radiogrammetry. The study was commissioned by the International Committee for...
Prevalence and Incidence
Both prevalence and incidence rate can be considered as measures of a disease experience within a population. They differ principally in that prevalence is derived from observations of a population made at a single point in time, whereas the incidence rate is derived from observations that are made only after observing a population over a period of time. There are two ways of expressing prevalence point prevalence and period prevalence. In this context, it is not terribly important to...
from pencilbeam to fanarray dxa data
The latest generation of central DXA scanners are fan-array scanners that were first introduced with the Hologic QDR-2000 20 . The terminology reflects a change in design in these scanners in which a fan-shaped beam is projected through an entire scan line and captured by an array of detectors. This is markedly different from the earlier pencil-beam devices in which a very narrow X-ray beam was projected in a plane that was perpendicular to the region of interest. This narrowed beam moved in...
The Utility of the sBMD
The sBMD is attractive as a means of comparing a BMD value obtained on one manufacturer's device with a BMD value obtained on another. This is useful in large population studies and clinical trials in which devices from several manufacturers must be used. The root-mean-square RMS error in the calculation of the sBMD is estimated to be 4 for the PA spine and total femur and even larger for the proximal femur subregion 13 . This error is simply too large to base important clinical decisions on...
Info Msm
Characterizing the Skeleton in Densitometry The Metacarpals, Phalanges, and Calcaneus Densitometry is primarily a quantitative measurement technique rather than a skeletal imaging technique. Nevertheless, there are unique aspects of skeletal anatomy in densitometry that must be appreciated to properly utilize the technology and interpret the quantitative results as well as the skeletal images. characterizing the skeleton in densitometry The bones of the skeleton can be described by four...
Hologic DXA and Lunar DPA
Kelly et al. 1 evaluated the relationship between BMD values in the spine in 85 individuals ranging in age from 21 to 78 years using the Lunar DP4, a 153Gd DPA device, and the Hologic QDR-1000, a pencil-beam DXA device. The correlation for the measurement of spine BMD between the two devices was extremely good with r 0.98. Not surprisingly, however, the two instruments did not give exactly the same results. Values obtained on the QDR-1000 were consistently lower than those obtained on the DP4....
using the phantom to create control tables and charts
Most daily quality assurance procedures to detect mechanical failures on today's densitometers are automated. The program will indicate a passing or failing condition. Before outright mechanical failure occurs, however, regular scanning of the quality control phantom and the application of Shewhart charts and rules or CUSUM charts can detect drifts or shifts in machine values that require correction in order to ensure continued accuracy and precision. Abrupt shifts in values are generally easy...
Ward S Triangle
Hologic q R-4500A Left Hip U8.15a 3 Hologic q R-4500A Left Hip U8.15a 3 Fig. 2-23. DXA proximal femur studies. Five regions of interest are defined. A Hologic QDR 4500 DXA study. Case courtesy of Hologic Inc., Bedford, MA. B Lunar Prodigy. Four regions of interest are labeled for emphasis on this study. The total region of interest, which is not outlined, includes the neck, trochanter, and shaft. Fig. 2-23. DXA proximal femur studies. Five regions of interest are defined. A Hologic QDR 4500 DXA...
Lunar DXA and Hologic DXA
The Lunar DPX and the Hologic QDR-1000 were compared in a study of 46 women by Pocock et al. 9 . These women underwent lumbar spine and proximal femur studies on both machines on the same day. The correlations were extremely good with r values of 0.98, 0.94, 0.96, and 0.96 for the lumbar spine, femoral neck, Ward's area, and the trochanter, respectively. The absolute BMD values were 16 lower on the QDR-1000 Conversion Formulas for BMDs of the AP Spine Between DXA Devices Hologic QDR-2000...
CUSUM Charts
CUSUM charts are not as easy to use as Shewhart charts and rules, but these are the types of charts employed by many professional densitometry quality control centers. This technique was originally developed for use in industry 13 and was subsequently adapted for use in bone densitometry 11,14,15 . The principle underlying CUSUM charts is the expected random variation in the phantom measurement. Remember that even in a perfectly functioning machine, daily phantom BMD values are expected to...
The European Spine Phantom
The development of the European Spine Phantom ESP resulted from efforts to develop a more perfect spine phantom that could be used on all central devices. It was developed independently of any manufacturer under the direction of the Committee d'Actions Concertes-Biomedical Engineering COMAC-BME 6 . The ESP is a semi-anthropomorphic phantom. Its three vertebrae are made of hydroxyapatite and vary in density, with standardized BMDs of 500, 1000, and 1500 mg cm2 7 . The vertebrae are encased in a...
The Lunar Spine Phantom
The Lunar spine phantom is a rectangular aluminum bar which is intended to replicate the lower half of T12, all of L1, L2, L3, and L4, and the upper half of L5. Each vertebra has a different density that is achieved by varying the thickness of the aluminum. The area of each vertebra is different as well. The densities of L1 to L4 are 0.92, 1.076, 1.239, and 1.403 g cm2, respectively. The L2-L4 BMD is 1.256 g cm2. To simulate soft tissue, the aluminum phantom is submerged in a water bath with a...
Shewhart Rules And Cusum Charts
Automated Quality Control Procedures Although much has been written about quality control procedures in densitometry, much of this literature has been concerned with data collection in clinical research rather than patient data collected as part of medical care. Quality control, although absolutely necessary in clinical research, is no less necessary in clinical practice. The original indications for bone mass measurements from the National Osteoporosis Foundation published in 1988 and the...
Estimated Time To Complete
We are pleased to award category 1 credit s toward the AMA Physician's Recognition Award. By completing the Review in the CD-ROM Companion in Appendix XIV, you are eligible for up to 30 hours of category 1 credit. After answering all of the questions correctly, complete the review evaluation and enter the required identifying information on the certificate of course completion. This certificate is not valid until signed with authorized signature at the Foundation for Osteoporosis Research. The...
Effect of Leg Dominance on BMD in the Proximal Femur
In general, there does not seem to be a significant difference in the BMD in the regions of the proximal femur between the right and left legs of normal individuals 32-35 . Leg dominance, unlike arm dominance, does not appear to exert a significant effect on the bone densities in the proximal femur and is not used to determine which femur should be studied. When proximal femur bone density studies first became available, the default or automatic positioning mode for the proximal femur was the...
Effect of Scoliosis Osteoarthritis Osteophytes Surgery and Fracture on BMD in
Structural changes and artifacts that interfere with DXA proximal femoral BMD measurements occur less often than at the spine. Osteoarthritic change in the hip joint may cause thickening of the medial cortex and hypertrophy of the trabeculae in the femoral neck, which may increase the BMD in the femoral neck and Ward's area 36 . The trochanteric region is not apparently affected by such change and has been recommended as the preferred site to evaluate in patients with osteoarthritis of the hip...
Image Analysis Inc Columbia
Fig. 1-8. An early Lunar DP3 dual-photon absorptiometer. This device utilized 153Gd to generate photon energy. Photo courtesy of GE Medical Systems, Madison, WI. Fig. 1-8. An early Lunar DP3 dual-photon absorptiometer. This device utilized 153Gd to generate photon energy. Photo courtesy of GE Medical Systems, Madison, WI. DPA studies of the spine required approximately 30 minutes to complete. Studies of the proximal femur took 30 to 45 minutes to perform. Total body bone density studies with...