Dual Energy X Ray Absorptiometry Biology Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

The quantitative in vivo measurement of bone mineral density and body composition has been an area of interest for both clinicians and researchers. These values are useful for assessing diseases affecting both bone mineral density (BMD) and body composition. BMD measurement is currently the most accurate method for predicting risk of osteoporotic fractures (Grainger et al 2001). The BMD measurements are also useful in monitoring response of treatments for prevention of bone loss and fractures. Body composition measurements are used to quantify soft tissue to assess changed associated with growth, disease, exercise and etc. Today, these measurements are most commonly taken by using dual energy X-ray absorptiometry (DXA). DXA is a type of dual energy imaging technique based on exploiting the difference in the attenuation of soft tissue and bone tissue at different energy levels. Since the discovery of dual energy imaging, its technology has advanced very quickly and its practical applications have expanded far beyond the original intent, which was to quantify the bone mineral content. With minor adjustments and modifications to the DXA system, its uses are now beyond the field of bone health and tissue composition measurements.

X-ray technology has been involved in medical investigations since the early 1900s. At first, radiographs were taken using X-rays to examine bone structures in patients. This was possible because bones being denser than other soft tissues, absorb X-rays more readily, hence producing an image. With improvements in the technique, X-ray is used to detect lung diseases, obstructions in the abdominal organs, identification of stones in gallbladder and the urinary system. Today the use of X-ray in medicine has been expanded to real time imaging such as angiography and surgical interventions.


Plain radiographs have been used to assess bone mineral density. This was done by examining the radiograph of lateral view of thoracolumbar spine, the information can be used to determine osteopaenic changes or to establish osteoporosis. However, on a radiograph the bone would appear normal until a deficit of bone mineral content (BMC) in the order of 20% occurs (Ardran 1951). Also, plain radiographs cannot be used to monitor the change of mineral content because the results are not easily repeated.

Single photon absorptiometry

In 1963, single photon absorptiometry (SPA) was introduced (Cameron and Sorenson 1963). This allowed bone mineral density to be measured in vivo and a quantitative value is obtained. This methods works by measuring the absorption by bone of a monochromatic, low energy photon beam. This beam is produced by a radioactive isotope source, iodine-125 is commonly used and emits photons at 27.3 keV (Cameron and Sorenson 1963). This technique was used to quantify bone loss due to osteoporosis in the forearm. To compensate for varying thickness of soft tissues around the bone, the patient is required to place the arm in a water bath of known width, as the absorption coefficients of water and soft tissue virtually the same, bone mineral content and areal density can be quantified accurately. The beam of photons is scanned across the water bath and forearm and the transmission at each point is noted.

Dual photon absorptiometry

Later, dual photon absorptiometry was developed (Reed 1966). This uses two photon beams of different energy to compensate for soft tissue thickness and eliminating the need of a water bath to measure BMC. Scanning of the axial skeleton became possible, more clinically relevant osteoporotic fracture sites such as the hip and the spine can be measured. Both photon absorptiometry techniques have a number of draw backs. The radioactive isotopes have short half-lives, they needed to be replaced often and the decay can also cause problem in follow up measurements. The scanning time can be up to 30 minutes, for certain groups of population for example the elderly, it can be very difficult to keep relatively still for this period of time and can cause discomfort. The image quality of this method is relatively poor and resulted in a low precision, which is the ability to produce the same results under the same circumstances.

Dual energy X-ray absorptiometry

The principle of DXA technology is very similar to that of DPA. DXA uses beams of X-rays emitted from an X-ray tube rather than radioactive isotope sources to provide the dual energies. The energies produced can be rapidly switching or by using a constant potential X-ray tube with rare earth filters to produce different energy levels. (Felsenberg et al 1995, Wahner et al 1988, Truscott et al 1996) An X-ray source increase the number of photon passing through the subject in a given time period, thus reducing scanning period and running cost.

Hologic introduced the first commercial DXA system in 1987 (Truscott et al 1996), and then closely followed by other companies. The first generation of DXA scanners used collimated pencil beam, reducing unnecessary radiation exposure to the subject. The subject lies on a bed, beneath the bed is the X-ray source and above the subject is the detector. The source and the detector are held on the can arm and the arm move these two parts synchronously and scan the subject in a rectilinear pattern in the region of interest (ROI).

The second generation of DXA systems uses a fan beam (Figure 1) of X-rays collimated by a slit-beam in conjunction with an array of detectors (Kelly et al 1991). This allows the scan to be completed with a single sweep of the scan which reduced the scan time to around 30 seconds. The image quality is near radiographic which gives a much improved precision and accuracy, which is the ability to give the true result compared to an accepted gold standard method. The system can measure the BMD of lumbar spine, hip, total body and body composition. With the new introduction of rotating C-arm scanner, a lateral lumbar spine BMD can be measured and spinal morphometry can also be performed.

Figure 1. Illustration of a modern fan-beam DXA scanner. Depending on different systems, X-ray can be emitted from the top or the bottom of the C-arm.

Comparison with other methods

Many of the advantages of DXA comparing to earlier dual energy imaging techniques have been mentioned above. DXA has higher accuracy and precision, the scan time and running time are greatly reduced.

DXA measurement of BMD in vivo is inevitably contaminated with false readings. DXA methodology can only return an accurate value of the areal bone mineral density only if the "two component" condition is met. The scan ROI can only contain two X-ray absorptiometrically disparate materials. As we know, the ROI contains a number of different types of soft tissues, for example: adipose tissue, muscle tissue, different organs such as intestines, blood vessel and intraosseous soft tissues (Bolotin 2007). Although modern systems provide algorithms in the software to compensate the different composition of soft tissue, DXA cannot provide 100% accurate BMD readings because of this inherent flaw in the methodology.

Apart from dual energy technique, there are other methods for measuring bone mineral density. One of which is quantitative computed tomography (QCT). QCT of the spine and the hip are available on standard CT units. The advantage of the QCT technique is its ability to assess the three-dimensional bone density (g/cm3) and DXA can only assess the two dimensional density (g/cm2), which is affected by the size of the bone. Two bones of equal density with different size, the bigger bone will give a greater BMD reading using DXA. QCT also allows the separate measurements of trabecular bone density, which is more sensitive to effects of hormones and drug treatments. The scan time of QCT is rather long, 20 minutes comparing to 30 seconds of DXA. The radiation dose of QCT could be up to 100 times higher than that of DXA, 2 μSv of radiation produced by DXA is equivalent of less than one day of background radiation (according to Report of the United Nations Scientific Committee on the Effects of Atomic Radiation to the General Assembly). Furthermore, CT scanners are in great demand for its many applications and they cost 10 times more than DXA scanners.


Accuracy (%)

Precision (%)

Effective Radiation Dose (μSv)

Scan Time (min)

DXA Pencil Beam





DXA Fan Beam















Table 1. Comparison of different BMD measurement techniques. Adapted from Grainger et al 2001 p186.

Current DXA systems

Currently there are two main competitors in the market for DXA scanners: Hologic and GE-Lunar. Hologic systems employ a single pass wide-angle fan beam, while the GE-Lunar systems use a multi-pass narrow-angle fan beam with some overlap between passes (Fan et al 2009). Hologic claims that by using wide-angle beam, it eliminates overlap errors and image distortion found in rectilinear acquisition techniques and results in superior image quality and data stability with subsequent scans. GE-Lunar pointed out that the wide-angle beam causes more magnification depending on the distance away from the detector (Boudousq et al 2005).

The two systems have different analysis algorithms, region of interest definitions and calibration standards so they present systematically different readings from the same patient. This causes difficulties in follow up scans of the same patient and also comparison of different studies when different scanners are used. With adequate standardisation methods of BMD measurements, it is possible to make the results from two systems comparable (Ozdemir et al 2007).

DXA applications

DXA technology was originally intended to measure the density of the bone in order to assess the risk of fracture. Now, it has a wide range of applications. Because its high precision, it is a very useful tool to monitor the progress of drug treatment, which is vital in drug trials. The algorithm behind the principle of dual energy imaging was designed to measure density of bone, but by changing the subject of the equation to soft tissue we can measure the composition of soft tissue. Many of the latest DXA scanners have a rotating C-arm, which can acquire a lateral image of the lumbar spine. This allows a morphometric assessment of the lumbar spine to identify fractures and it can also be used to identify cardiovascular diseases by analyses of the abdominal aortic calcification.

Diagnosing osteoporosis

Osteoporosis is defined as a progressive systemic skeletal disease characterised by a low bone mass and microarchitectural deterioration of bone tissue, leading to an increase in bone fragility and susceptibility to fracture (Grainger et al 2001). Clinical risk factors alone cannot predict low bone density and the risk of osteoporotic fractures, and bone mineral density is highly correlated to bone strength (Turner 1991). Men have a higher BMD than women but it reduces for both sexes after the fourth decade. In women, the rate of loss in BMD is accelerated at the time of menopause and continues do so for the next 10-15 years (Mosekilde and Mosekilde 1990). DXA is a non-invasive, accurate and precise method to measure bone mineral density in a variety of anatomical sites.

Figure 2. Illustration of a DXA scanner in operation. Adapted from Hologic brochure.

During scanning, the patient lays flat on the motorised bed, fully clothed (Figure 2). The operator should check for any metallic items such as jewellery in the region of interest that can interfere with the density readings. A positioning guidance laser crosshair is emitted down from the scanner and the operator can move the patient into the correct position by moving the bed. For hip BMD measurement a position block is placed at the feet of the patient and the leg (typically the left) is secured into position. This is important because the leg can adopt many positions depending the position of the joint and this can change the BMD readings for the hip slightly. Similar approach is done when the lumbar spine BMD is measured, the patient rests his/her legs on a block to ensure the spine is as straight as possible.

When the patient is in position, the operator can initiate the scan on the computer, the scanner then makes a single sweep up the patient. This usually takes 30s and an image or ROI and BMD readings are produced instantaneously. No shielding was required during the scan because the X-ray scattering is minimised by collimation, plus the radiation dose is minute. Although the radiation dose is very small, pregnant woman are not scanned due to possible damage to the fetus. A print out (Figure 3) can be produced and this information can be passed on to the patient's clinician or a specialist.

Figure 3. Sample DXA report print out, contains BMD value, T and Z scores for both lumbar spine and hip.

Hip BMD is considered the gold standard by many for assessing the risk of fracture and making diagnosis of osteoporosis. Hip BMD is a stronger predictor of hip fracture than BMD of other sites and predicts the risk of other fracture equally well as other measurements (Marshall et al 1996). Furthermore, unlike BMD of the lumbar spine, the hip is not affected by the calcification of the abdominal aorta. The DXA scanner can measure the BMD of total hip and its many subregions including the femoral neck the trochanter and the Ward's triangle (which is not a true anatomic area but is considered to be the area having the lowest BMD in the femoral head).

The World Health Organisation defined osteoporosis as a T-score of -2.5 or lower at any site of measurement and osteopaenia (low bone mass) as a T-score between -1 and -2.5 (Kanis et al 1994). A T-score is the number of standard deviations below or above the mean BMD for young adults between the ages of 20 to 30. A T-score of 0 means the patient has a BMD value that is exactly the mean for young adults. Another term commonly mentioned in BMD measurement is the Z-scores. A Z-score is the number of standard deviation below or above the mean BMD adjusted for the patient's age and sex. This cut off point off was originally designed to compare the prevalence of osteoporosis in different countries, it was not intended as the sole deciding factor of the management of patients. The WHO cut off points are also rather arbitrary. If a patient has a T-score of -2.4 but has a number of relevant clinical risk factors (age, previous fractures, parent hip fractures, current smoking and drinking habits, glucocorticoids use and rheumatoid arthritis), should the patient not be considered to start treatment? When receiving a diagnosis of osteopaenia, patients can be unnecessarily distressed. Osteopaenia is neither a cause for alarm nor a reason to start treatment. About half of all postmenopausal women have a bone density value in the osteopaenic range and most normal women over 65 could be labelled as osteopaenic (Cummings et al 2002).

While BMD is a good predictor of fracture risk, density combined with clinical risk factors (those mentioned above) for fracture is a better method than either of those used alone. The Fracture Risk Assessment Tool (FRAX) estimates the 10 year probability of fracture on the basis of clinical risk factors for fracture and the BMD of the femoral neck. The FRAX model was developed through an analysis of almost 60,000 men and women in 12 population-based cohorts with about 250,000 peron-years of observation, and externally validated in an additional 11 cohorts with 230,000 men and women and more than 1.2 million person-years of observation (Lewiecki 2009).

FRAX is available both electronically (www.shef.ac.uk/FRAX/) and in paper form (Figure 4). With a relatively accurate fracture probability, it is much easily for clinicians to decide an adequate treatment plan for the individual patients.

Figure 4. Section of the FRAX table for the UK women. Numbers indicate the 10-year probability of an osteoporotic fracture.

FRAX has not been validated in patients outside the specified age range (40-90) or in children. The seven risk factors are entered as yes or no responses, but the actually risk of the patient may depend on the actual dose of hormone or the severity of other factors. Other important risk factors such as falling, rate of bone loss and high bone turn over are not included in the algorithm. FRAX remains a work in progress, with continuing updates expected as new information becomes available. Future FRAX may include more country specific fracture rates and bone density measurements with technologies other than DXA.

Treatment monitoring

The spine has more trabecular bone than other sites, spine BMD is more sensitive to effects of hormones and drugs. Spine BMD measures the lumbar vertebrae, L1 to L4. The spine BMD is the total mineral content of the whole lumbar spinal column, including the vertebral bodies, the posterior elements and the facet joints. The spine BMD is also artifactually increased by degenerative arthritis and aortic calcification, both of which become common after age of 65 (Cummings et al 2002). The spine DXA measurement can be affected if one of more of the vertebral bodies included in the measurement is fractured or affected by changes due to arthritis, therefore well trained operators are needed to exclude those vertebrae from the results.

Once treatment has been initiated, an important part of osteoporosis management is to monitor the efficacy of treatments. This will help to identify the patients who do not respond to the current therapy and those who are non-compliant. When using BMD for subsequent treatment follow-up, the measurements should be done on the same machine and if machine is changed a cross calibration is needed. (International Society for Clinical Densitometry, 2007 ISCD Official Positions Brochure). According to the International Society for Clinical Densitometry, measurements should be performed before and one year after the initiation of therapy, with longer intervals once efficacy has been established.

Responding to treatment does not necessarily mean a gain in BMD. A patient can lose BMD after treatment but without the treatment, the loss could be greater. So a patient can be responding to treatment despite a loss in BMD. Many clinicians believe that by monitoring the therapy it can improve compliance. Most problems with adherence occur within the first 3 months of starting treatment so BMD measurements done 1 or 2 years into treatment is not likely to influence adherence. Other biochemical markers are used to determine patient response to therapy. Studies shown antiresorptive agents reduce the levels of biochemical markers of bone turnover 3 to 6 months after treatment initiation and these markers can be useful in determining response to therapy (Bonnick and Shulman 2006). Bone resorption markers include: N-telopeptide of type 1 collagen, C-telopeptide of type 1 collagen, pyridinoline and deoxypyridinoline, should be measured at baseline and at 3 months and 6 months after initiation. Bone formation markers (bone alkaline phosphatase, osteocalcin and propeptides of type 1 collage) should be assessed at baseline and at 6 month after starting therapy (Nishizawa et al 2005). Changes in these markers are valid intermediate endpoint for the efficacy of treatment, these could provide early data to indicate the success of treatment particularly before change in BMD is apparent.

However, the level of change of any given marker that indicates therapeutic efficacy remains undecided between the major organisations. Therefore, in spite of their rapid assessment of therapeutic efficacy, biochemical markers currently have limited applicability and for most patients DXA spine is still the most appropriate diagnostic tool for monitoring treatment efficacy.

Body composition

Measuring body composition is used for monitoring growth and to assess the effect of nutrition and exercise. Currently available methods such as neutron activation analysis, isotope dilution, anthropomorphic measurements and underwater weighing present a number of technical, ethical, precision and availability problems (Grainger et al 2001). DXA is now one of the most frequently used techniques for body composition measurements as a result of the increasing worldwide availability of the scanner. It is non-invasive, easily applied to both healthy people and patients, and the radiation dose is extremely low. Newer scanners have also reduced its scanning time and are capable of providing regional body composition analysis. This technique is increasingly being viewed as a reference method for the estimation of total body fat, and since its introduction it has became the gold standard method for body fat measurement (Plank 2005).

The DXA body composition approach assumes that the body consists of three components: fat, lean soft tissue and bone. Those are distinguishable by their X-ray attenuation properties. Within any pixel the proportions of only two components can be resolved by the differential absorption of two energies. Soft tissues and bone are relative easy to distinguish. In areas where bone is not present, calibration allows fat and lean fractions to be resolved from soft tissue and the composition of these areas is extrapolated to the soft tissue overlying bone to produce a total body fat and lean soft tissue.

Figure 5. An example of composition read out. DXA is capable of total and regional fat and lean soft tissue measurement. Adapted from http://www.leanresearch.com/

Most DXA machines are calibrated to measure over a thickness of 12-28cm, so any deviation from this range may affect the accuracy of the measurement, for example for severely obese patients (Laskey et al 1992).

The total body measurement read out (Figure 5) gives a comparison with the normal age and sex matched peers together with regional assessments and total tissue mass. Body composition measurement has a range of clinical applications from assessing the change in composition due to disease or therapy to the body composition monitoring for sports professionals.

Morphometric X-ray absorptiometry

Vertebral fractures are often asymptomatic or not diagnosed. The diagnosis of vertebral fractures requires lateral radiographs of both the thoracic and lumbar spine. The quantitative measurement of vertebral size is known as vertebral morphometry. Vertebral fractures are associated with increased disability and morbidity and they are the most common osteoporotic fractures (Genant et al 2000). Women with vertebral fractures have been shown to have a 2-fold increase in the likelihood of fracture and a 5-fold increase in their risks for subsequent vertebral fracture. Plus, one out of five women who have an incident of vertebral fractures will suffer a subsequent fracture within the following 12 months (Lindsay et al 2001).

Measurements of the anterior, middle and posterior vertebral heights of the T4-L4 vertebrae are classically done on lateral radiographs. The disadvantages of this technique include distortion associated with patient positioning and the time taken to measure the vertebrae (Grainger et al 2001). With the new rotating C-arm DXA scanners (Figure 6), a lateral scan of T4-L4 can be obtained while the patient remains in a supine position. The scanner takes a single sweep (about 15 seconds) of the thoracic and lumbar spine, the software is designed to measure the vertebral heights automatically and a print out is provided (Figure 7).

Figure 6. Illustration of a rotating C-arm in operation. The leg block is used to straighten the spine. Adapted from Hologic brochure.

Figure 7. Sample vertebral fracture assessment print out. Anterior, middle and posterior vertebral heights and percentage deformation are automatically calculated. From http://avoidboneloss.com/hologic

DXA vertebral fracture assessment provides a convenient, cheap and safe way to assess vertebral fracture. In general it can detect all severe fractures but it can miss some more moderate fractures that can be seen on radiographs (Cummings et al 2002).

Abdominal aortic calcification

The abdominal aorta is situation closely anterior to the lumbar spine. Calcification in the abdominal aortic wall has been a hindrance to accuracy measure spine BMD by DXA, because it would make the lumbar vertebrae appear to have a higher BMD. The close association of the lumbar spine and the abdominal has now been exploited by the lateral vertebral imaging technology provided by DXA.

Similar to the coronary circulation, aortic calcification is likely to influence subsequent cardiovascular events such as aortic occlusion, aneurysm development and distal embolisation. Calcification within the aorta can also impact on medical and surgical treatment by impairing the outcome of aortic stenting and aneurysmal repairs (Tintut and Demer 2001). One quick and simple method called AAC-8 scale is used to estimate the total length of calcification of the anterior and posterior aortic walls in front of vertebrae L1-L4 (Schousboe et al 2006). The anterior and posterior aortic walls are assigned a score between 0-4 depending on the number of vertebra is spans. The sum of the two scores for the anterior and posterior walls gives the AAC-8 score, a score greater than two is considered moderate on this scale.

Figure 8. Image showing severe abdominal aortic calcification. From www.hologic.com

Abdominal aortic calcification (AAC) can be a particularly valuable measurement since it contributes independently of traditional clinical risk factors such as cholesterol, blood pressure, diabetes, and age to the prediction of heart attack risk. Detection of moderate to severe AAC is equivalent to risk of an additional 160 mg/dL of total cholesterol (Wilson et al 2001). With the new rotating C-arm DXA scanner, AAC can be viewed in the same scan used for a vertebral fracture assessment. AAC seems to be particularly strong risk factor in postmenopausal women, which is the population also in greatest need of BMD scans.


Dual energy X-ray absorptiometry provides a very useful non-invasive clinical tool to evaluate and monitor bone mineral density. DXA measurements of BMD have become an integral part for diagnosing osteoporosis and monitoring treatment efficacy. Its body composition measurement function is much more capable of other available methods and it is accepted as the gold standard for measuring body fat. The new rotating C-arm scanners have opened a whole new field of applications. It can detect deformities of the lumbar vertebrae and abdominal aortic calcifications. A new development in the field is 3D DXA scans taking full advantage of the rotating C-arms. It would be natural to consider the bone structures as 3D objects, the next generation DXA devices will take multiple projections and combine the data to form a tomographic representation of the structures. This will give an even more accurate picture of bone density, geometry and strength (www.hologic.com). DXA system scan produce results of an exceedingly high standard, have low running cost, low radiation dose to subjects and ease of use, I think this technology will remain in clinical practice for many years to come.