Nuclear Oncology, 1 Ed.



William D. Leslie


Osteoporosis is defined by the World Health Organization (WHO) as “a systemic skeletal disease characterized by low bone mass and microarchitectural deterioration of bone tissue, with a consequent increase in bone fragility and susceptibility to fracture.”1 The presence of osteoporosis is a major risk factor for the development of fractures of the hip, proximal humerus, vertebra, and forearm but most other skeletal sites are also at increased risk of fracture.2Worldwide, the number of fracture sufferers in 2000 was estimated at 56 million with approximately 9 million new osteoporotic fractures each year.3 Older age and female sex are primary risk factors for osteoporosis, with the global burden of osteoporosis is projected to increase markedly over the next few decades as the number of elderly individuals increases.4 Cancer is the second leading cause of death in the United States and is expected to surpass heart disease over the next few years.5 Cancer in industrialized nations is a disease of aging, with a median age of 67 at diagnosis. Improvements in outcome imply that more elderly cancer patients will survive and be at risk for osteoporosis, a consequence of some of the treatment regimens (e.g., aromatase inhibitors (AIs) for breast cancer, androgen deprivation therapy for prostate cancer, and TSH suppression for differentiated thyroid cancer).

Bone mass is the primary determinant of bone strength, as studies of excised bone have demonstrated that approximately 80% of bone strength is determined by the amount of bone. Bone mineral density (BMD) is an integral element in fracture risk assessment and in the diagnosis of osteoporosis.6,7 Dual-energy x-ray absorptiometry (DXA) provides an objective and quantifiable index of skeletal strength that is widely used in clinical practice guidelines and recommendations from expert bodies,812 notwithstanding increasing recognition of the importance and independent contribution of clinical risk factor (CRF) evaluation as part of treatment decision-making (see Fracture Risk Assessment).13 In the absence of a defining fracture, the diagnosis of osteoporosis is based on the measurement of BMD by DXA (Table 30.1). The WHO has provided an operational definition of osteoporosis given as a BMD that lies 2.5 standard deviations (SDs) or more below the average mean value for young healthy women (T-score ≤−2.5 SD) based upon a standardized reference site (the femoral neck) and a standard reference range for both men and women (the NHANES III data for women aged 20 to 29 years).1,14,15 Microarchitectural deterioration also has an important effect on bone strength. Typical changes in trabecular bone include reduction in trabecular thickness and number, and perforation of trabeculae by deep resorption pits. This results in loss of trabecular connectivity and is currently believed to be irreversible. These microarchitectural changes may not be reflected by reductions in bone density but still contribute to skeletal fragility and fracture susceptibility.

TABLE 30.1


The consequences of fracture include increased mortality, morbidity, institutionalization, and economic costs.16,17 An individual with a hip fracture has a markedly increased risk of death within the first 6 months that is partly attributable to the fracture itself. Both hip and vertebral fractures are associated with an excess risk of death beyond the first year,1820 and this is largely attributed to underlying comorbidity.21,22 Moreover, all osteoporosis-related fractures can lead to significant long-term disability and decreased quality of life.23,24

The economic impact of fractures is also noteworthy. Studies have consistently shown hip fractures to be associated with the highest costs, typically exceeding $25,000 US in the first year.16,2527 Spine fractures show costs up to $14,977 US in the first year post fracture.25,27 Fractures other than hip and spine showed costs in the first year post fracture are still quite significant ($9,183 US for age 50 to 64 years, $6,106 US for age 65 years and older).27Because of their high prevalence, the total first-year costs of nonhip nonspine fractures are greater for those age 50 to 64 years than for hip and spine fractures combined.27,28


Bone is a dynamic living tissue consisting of cellular, organic, and inorganic components with a complex internal structure that undergoes continuous remodeling throughout life. The mature skeleton consists of a mixture of cortical bone (85%) and trabecular bone (15%), the relative amounts of which vary widely between different anatomic sites (Fig. 30.1).

Bone reacts to stress and injury through a well-orchestrated sequence for removing old bone and building new tissue. Bone remodeling is carried out by the basic multicellular unit (BMU), which consists of both osteoclasts and osteoblasts. The BMU typically takes 3 to 6 months to complete a cycle (Fig. 30.2). Bone remodeling affects 3% to 5% of cortical bone per year, but up to 25% of trabecular bone due in part to its greater relative surface area. The osteoclast, a multinucleated cell of monocyte origin, resorbs bone through the release of acid and enzymes such as cathepsin K from its ruffled border. Osteoblasts, derived from mesenchymal cells, enter the resorption pit and lay down organic matrix (osteoid). The osteoblast then dies or enters a dormant stage and is known as an osteocyte (if trapped within calcified tissue) or a lining cell (if found on the surface of calcified tissue). The osteoid is subsequently mineralized over a period lasting several months. There is close coupling in osteoclast and osteoblast activities, although the intercellular signaling involved is incompletely understood. It is clear, however, that processes which stimulate (or suppress) one cell type result in stimulation (or suppression) of the other. For example, after menopause, osteoblast activity increases in an attempt to compensate for increased osteoclastic resorption. On the other hand, antiresorptive treatments targeted at suppressing osteoclast activity are only able to achieve a slight gain in bone mass as there is a parallel reduction in osteoblast activity. The bone remodeling cycle is regulated by a myriad of growth factors and cytokines. A recently recognized system has recently been discovered and shown to be the primary regulator of osteoclastogenesis activity. Three molecules, the receptor activator of NF-κ B (RANK), its ligand RANK-L, and the decoy receptor of RANK-L, osteoprotegerin (OPG), play a pivotal role as central regulators of osteoclast function.29 RANK/RANK-L signaling is required for osteoclast development, whereas OPG inhibits RANK/RANK-L signaling and osteoclast activity.

FIGURE 30.1. Percent of trabecular bone in skeletal sites commonly assessed with bone densitometry. (Reproduced with permission from Leslie WD, Roe BE. Bone densitometry. In: Leslie WD, Greenberg D, eds. Nuclear Medicine. Austin: Landes Bioscience; 2003:93–120.)

Bone cell activity can be evaluated through the measurement of biochemical markers. Osteoblasts produce type I collagen (the primary collagen of bone tissue), noncollagenous proteins (such as osteocalcin), and enzymes (such as alkaline phosphatase). Many of these are measured in the serum as indices of bone formation. Collagen is a triple-helical molecule that undergoes extensive posttranslational modifications including hydroxylation, glycosylation, and covalent cross-linking. The degradation products of cross-linked type I collagen, which include C-telopeptide, N-telopeptide, and deoxypyridinoline, can be measured in the urine or serum as an index of osteoclast activity.

Skeletal mass accumulates rapidly during childhood, especially during the years of most rapid growth in early adolescence, with peak bone mass achieved by 25 years. Determinants of peak bone mass included genetics, physical activity, and diet (especially calcium intake). The major genes that influence peak bone mass have yet to be elucidated, but twin studies suggest that 60% to 90% of the variation in peak bone mass and the majority of variation in bone loss is determined by genetic factors.30,31 Ethnicity is also important. Blacks have higher average bone mass than Caucasians who in turn have higher average bone mass than Asians.14,32

After early adulthood, both men and women experience a slow, progressive decline in bone mass that continues until death. Bone turnover accelerates in women at the time of menopause, especially in trabecular bone, and often results in the loss of 5% to 15% of bone mass over the first 5 years after menopause. After early menopause, age-related bone loss continues at a rate of 0.5% to 1% per year and may actually accelerate again in later life.33 The pathogenesis of age-related bone loss is unclear, but may be related in part to changes in calcium absorption and vitamin D availability.


Dual Energy X-Ray Absorptiometry

DXA is the most widely used technique for measuring BMD in clinical practice. DXA has a well-established role in the evaluation of individuals at risk of osteoporosis, and in helping clinicians advise patients about the appropriate use of antifracture treatment. DXA examinations have three major roles: The diagnosis of osteoporosis, the assessment of a patient’s risk of fracture, and for monitoring response to treatment. DXA evolved from dual-photon absorptiometry (DPA) which used gadolinium-153 as the photon source, and suffered from the need for radionuclide source changes, poor image resolution and reproducibility, and has been largely replaced by DXA.

DXA uses an x-ray tube to generate two different x-ray energies. Bone blocks (or attenuates) x-rays to a greater degree than soft tissue, and lower x-ray energies are attenuated more than higher energies. An x-ray detector records the amount of attenuation for the two energies and can calculate both the amount of soft tissue (using an estimated soft tissue attenuation value from the nonbone pixels) and the amount of bone calcium (using hydroxyapatite as the reference material) in the path of the beam. The x-ray tube and detector scan over the area of interest, building up an image of bone mineral content (BMC) expressed in grams of calcium. The densitometer’s software identifies the projected bone area using an edge-detection algorithm. Dividing BMC (grams of calcium) by the bone area (cm2) yields BMD as g/cm2. DXA has the advantage of being rapid (particularly with newer scanners that use higher output x-ray tubes and a fan-beam configuration) and is able to scan the structures of greatest clinical interest such as the spine, hip (Fig. 30.3), forearm, and total body. The effective radiation dose from a lumbar spine scan is much less than 10 μSv (1 mRem), a value similar to one day’s normal background radiation and of negligible risk. Effective doses from scans of the hip, forearm, and total body are even lower.

FIGURE 30.2. Bone remodeling cycle. Proceeding from left to right the sequence is (A) osteoclast resorption, (B) osteoblast proliferation, (C) osteoid matrix deposition, and (D) mineralization. (Reproduced with permission from Leslie WD, Roe BE. Bone densitometry. In: Leslie WD, Greenberg D, eds. Nuclear Medicine. Austin: Landes Bioscience; 2003:93–120.)

FIGURE 30.3. Lumbar spine and left hip dual-energy x-ray absorptiometry scans in a 76-year-old white female with papillary thyroid cancer. Based on total hip T-score of −1.6 she would be designated as having low bone mass (not osteopenia). BMC, bone mineral content; BMD, bone mineral density; YA, young adult.

Limitations and Errors in Dual Energy X-Ray Absorptiometry

Measurement errors in DXA of the spine and hip are typically on the order of 5% to 7%, but in individual cases may be much larger.34,35 The principal source of error relates to the inherent assumption that the body is composed of two tissue types: Bone and soft tissue. In reality, soft tissue can be decomposed into lean and fat components, with the former have significantly greater density than the latter. If the composition of the soft tissue overlying the bone region of interest (ROI) is not known, or is not correctly estimated from the nonbone pixels, then this will cause an error in the BMD measurement. Because of the large thickness of tissue in the abdomen, the areal soft tissue mass for a spine DXA scan is considerably greater than that of bone mineral (range: 15 to 25 g/cm2 compared with a typical BMD value of 1 g/cm2), and therefore even small differences in x-ray attenuation between lean and adipose tissue discussed above can generate clinically significant measurement errors in the BMD results.34,35 Figure 30.4shows the error that can be introduced in the calculation of BMD when an incorrect soft tissue attenuation value is generated by the inclusion of a staghorn calculus in the soft tissue region, even when bone edge detection is unaffected. Assessment of bone and soft tissue typing is an essential component of DXA quality control.

Because most bone density techniques give an area measurement based upon a two-dimensional projection of bone (in g/cm2), larger bones will have a higher apparent bone density than smaller bones because of the increased depth (Fig. 30.5). Many of the reported area BMD differences related to sex, ethnicity, childhood/adolescence, and conditions associated with short stature (e.g., dwarfism, Turner syndrome) relate to the effect of bone size on areal bone density measurements.3644 The confounding effect of skeletal size has also been seen among a relatively homogeneous population of 16,205 White women aged 50 years and older, in whom total hip areal BMD categorized a substantially higher fraction of women with smaller bone area as being osteoporotic despite similar incident fractures and paradoxically lower prevalent fractures.45 Although techniques have been developed to try to address this (such as estimations of skeletal volume to provide a “volumetric” bone mineral apparent density [BMAD] reported in g/cm3),46 only quantitative CT (QCT) provides a true volumetric measurement.

FIGURE 30.4. A large left renal staghorn calculus is seen (left). If this is not excluded from the soft tissue map (middle), then the lumbar spine bone mineral density (BMD) measurement is 0.756 g/cm2. After correct exclusion from the soft tissue map (right), the lumbar spine BMD measurement is 0.834 g/cm2 (9.1% difference).

Apparent discrepancies between hip and spine BMD measurements are common and emphasize the complexity of skeletal metabolism. As a “systemic skeletal disorder,” osteoporosis affects all bones but the degree is modified by local determinants of bone metabolism that include bone composition (trabecular bone undergoes more rapid turnover and loss), mechanical loading (weight bearing enhances osteoblast activity), and age-related artifacts (usually elevating spine BMD). Together these factors help to explain why differences between spine and hip BMD measurements are so common.

Attention to correct patient positioning is paramount in ensuring reliable BMD measurements (Fig. 30.6). Bone density may be overestimated in anteroposterior measurements of the lumbar spine because of the presence of degenerative sclerosis or osteophytes, acquired conditions (e.g., ankylosing spondylitis, Paget’s diseases, compression fractures, metastases), superimposed vascular calcification, or other dense materials (e.g., barium, iodinated contrast or undissolved calcium tablets) (Fig. 30.7). Discrepancies between spine measurements and other skeletal sites are more typically seen in older subjects (more than 60 years of age) and those with known spine disease. Hip measurements are affected by patient positioning and the degree of hip rotation which make it critical for technologists to adopt a standardized technique. Hip measurements are less susceptible to degenerative changes, but thickening of the medial cortex of the femoral neck (“buttressing”) will be reflected in bone density measurements (Fig. 30.8). The trochanteric region appears to be relatively unaffected by these changes. Overlying artifact, previous fracture, surgery, or Paget’s disease (see Fig. 30.8) can affect hip results and the contralateral hip should be measured under such circumstances.

FIGURE 30.5. Bone volume strongly affects measured bone density using areal techniques such as dual-energy x-ray absorptiometry (DXA). Assume that two cubes are constructed from hydroxyapatite (1 g/cm3). The larger cube will have twice the areal density of the smaller cube because of its greater depth because this is not measured in DXA.

Vertebral Fracture Assessment

The majority of spine fractures are not clinically diagnosed, but may still have health consequences and economic implications.47 Significant vertebral fractures (usually showing >25% height loss and end-plate interruption) unrelated to trauma are associated with a fivefold increased risk for recurrent vertebral fractures; mild spinal deformities (<25% height loss without definite end-plate fracture) are not strong predictors of future osteoporotic fractures or low bone density.48 Vertebral Fracture Assessment (VFA) is a scanning and software option on modern DXA instruments which use a fan-beam scanning technology and can detect fracture (Fig. 30.9) and nonfracture abnormalities in the thoracolumbar spine (Fig. 30.10). It can be easily performed at the time of BMD measurement, allowing integration of BMD and vertebral fracture information in the clinical care of patients evaluated for osteoporosis. VFA is associated with low radiation exposure (2–50 μSv versus 600 μSv for lateral spine radiographs). When Genant’s SQ approach for vertebral fracture grading is used (Fig. 30.11), caution should be exercised when diagnosing Grade 1 deformities as fractures, particularly when such deformities are observed in the thoracic spine as Grade 1 fractures are less predictive of future fractures, and are more difficult to detect on VFA.49 Therefore, only Grade 2 and 3 deformities should generally be considered as clear fractures. Like radiographic fractures, vertebral fracture identified on VFA predict future osteoporotic and hip fractures independently of age, weight, and BMD.48,50A proposed set of indications for VFA are listed in Table 30.2.51

An additional benefit of VFA is the detection of abdominal aortic calcification (AAC) (see Fig. 30.10), an important marker of subclinical CVD.52 AAC from VFA has been shown to predict subsequent vascular events independent of the Framingham risk score.53 Among 408 women (aged >75 years) who sustained an MI or stroke during a median 4-year follow-up period and randomly selected 408 controls, the odds ratio (OR) of incident MI or stroke for those in the top tertile compared with the bottom tertile of AAC score were 1.74 (95% confidence interval (CI): 1.19 to 2.56) for a 24-point scale and 1.77 (95% CI: 1.22 to 2.55) for an 8-point scale, adjusted for age, high-density lipoprotein and low-density lipoprotein cholesterol, triglycerides, blood pressure, smoking, renal function, health status, and baseline diagnoses of diabetes mellitus, hypertension, angina, and prior stroke.

FIGURE 30.6. Incorrect (left) and correct (right) positioning in dual-energy x-ray absorptiometry. The lumbar spine should be aligned with the table center line. The femur should be internally rotated and the shaft aligned with the table center line.

TABLE 30.2


FIGURE 30.7. Spine dual-energy x-ray absorptiometry artifacts. A: Severe degenerative disk disease (L1–L4 T-score −0.8, total hip T-score −4 for the same patient). B: Ankylosing spondylitis (L1–L4 T-score +4.5). C: Paget’s disease affecting L1 (L1 T-score +2.1 versus L2–L3 −0.8). D: Metastastic prostate cancer involving T12, L1, and L3 (T-scores L1 +5.7, L2 +1, L3 +6.5, L4 +1.4).

Other Bone Measurement Technologies

Conventional DXA (also known as central DXA) is able to measure all skeletal structures, including those in the thicker body regions such as lumbar spine and hip. Unfortunately, conventional DXA equipment is relatively expensive and heavy. Therefore, a variety of compact, portable devices have been developed for measuring bone density in the extremities such as the forearm and calcaneus. Single photon absorptiometry (SPA) used a radionuclide source (iodine-125) but required periodic source replacements and immersion of the body part in water. Peripheral DXA (pDXA) devices avoid these limitations and impart an exceedingly small radiation dose of less than 0.1 μSv (0.01 mRem).

Ultrasound has emerged as another tool for characterizing bone strength and has the advantages of being radiation-free and using relatively inexpensive, portable devices. Ultrasound penetrates bone poorly, and higher frequencies are attenuated more than lower frequencies. Two measures are typically derived from quantitative ultrasound (QUS). One is the speed of sound (SOS), a measure of the speed with which sound travels from one transducer to the other through the bone (m/s). The other is broad beam ultrasound attenuation (BUA) which is the slope of the relationship between attenuation and frequency (dB/kHz). Because of the ultrasound’s difficulty in penetrating deep structures, most devices measure the more accessible bones such as the calcaneus, phalanges, and tibia. Although x-ray–based techniques are calibrated against calcium content, there is considerable controversy over the physical properties measured by bone ultrasound. Whether QUS measures bone quality independent of bone density is currently an area of controversy.

CT scanners are capable of measuring spine and hip bone density by using a calibrated phantom and specialized software. Such measurements are known as QCT and give bone density in terms of volume (mg/cm3). Although QCT is expensive and has a relatively high radiation dose (1500–3000 μSv for multidetector CT), it has the advantage of providing a true volumetric measure of bone density (rather than areal as with DXA). This can be advantageous when skeletal size deviates markedly from average or when there are dense artifacts (such as heavily calcified aorta or osteophytosis of the lumbar spine) that preclude accurate DXA measurements. Smaller CT devices have been developed for studying the distal radius and distal tibia, known as peripheral QCT (pQCT). High-resolution devices (HR-pQCT) are capable of imaging the structure of bone and provide insights into alterations in cortical and trabecular microarchitecture.

FIGURE 30.8. Hip dual-energy x-ray absorptiometry artifacts. A: Advanced hip osteoarthritis produced “buttressing” with overestimation in femoral neck bone mineral density (right hip T-scores femoral neck +1.5, trochanter –3.2; left hip T-scores femoral neck –1, trochanter –3.2). B: Credit card artifact (T-score +0.9 before, –2.2 after removal). C: Previous left hip pinning with pin removed (left total hip T-score 0, right total hip T-score –2.7).

Although conventional x-rays are not quantitative, they are still an important component in the assessment of osteoporosis because the presence of fragility fractures (such as vertebral compression fractures) indicates osteoporosis. Plain radiographs of the hand can also be used to measure cortical width in the fingers, but this is a relatively insensitive technique. With the introduction of aluminum calibration wedges, however, there is renewed interest in plain radiographs of the hands as an accessible and low-cost alternative to the methods previously listed.

Accuracy and Precision

Bone density measurement has a primary clinical role in the initial diagnostic and fracture risk assessment of osteoporosis,7,54 and is also widely used for serial monitoring of patients with suspected or confirmed osteoporosis.55 A formal quality assurance program is an essential component of a bone density program. In the case of DXA, this should minimally consist of a daily calibration check (usually with an anthropomorphic spine phantom) which is compared with predefined tolerance limits. The cumulative data must be inspected regularly to look for subtle changes or drifts in performance signaling the need for corrective action (Fig. 30.12).

FIGURE 30.9. Vertebral fracture assessment in three different patients. Normal (left), moderate T12 fracture (arrowmiddle) (same case as Figure 28.3), and multiple severe fractures (right).

FIGURE 30.10. Nonfracture vertebral fracture assessment abnormalities. Ankylosing spondylitis (left), metastasis from breast cancer (arrowmiddle), and aortic calcification (right).

FIGURE 30.11. Genant’s semiquantitative classification of vertebral fractures based upon severity and shape. Mild (Grade 1) refers to 20% to 25% height reduction, moderate (Grade 2) 26% to 40% height reduction, and severe (Grade 3) >40% height reduction. Vertebral fracture is diagnosed when reduction of height in anterior, middle, or posterior dimensions of vertebral body exceeds 20%. The approximate degree of height reduction determines assignment of grade to vertebra. Fractures are classified as wedge, biconcave, or crush, depending on whether anterior, middle, or posterior portion of vertebral body is most diminished in height. (Reproduced with permission from Genant HK, Wu CY, van KC, et al. Vertebral fracture assessment using a semiquantitative technique. J Bone Miner Res. 1993;8(9):1137–1148.)

Accuracy refers to how closely a measured result approximates the “true” value and is of critical importance when comparing an individual patient to a reference population. The accuracy of bone mineral measurements is determined by comparison with dry- or ash-weight of bone samples. DXA is the predominant technology used for evaluating bone density, and has a measurement error of 5% to 7%. This error is small relative to the range of values in the population, enabling its use as a tool to diagnose osteoporosis and assess fracture risk.

Precision (also referred to as reproducibility) is the ability of a system to obtain the same results in repeated measurements of the same individual. A technique must have good precision if serial measurements are to be used in following an individual. Greater precision makes it possible to detect smaller changes in a subject. Current methodologies typically demonstrate precision errors that are larger than annual changes in bone density. Thus, in an individual patient, it may be difficult to determine whether a small change in the bone mass measurement reflects precision error or true change. DXA reproducibility is influenced by instrument-, operator- and subject-dependent factors. These last two tend to be much more important than the instrument itself, and patient positioning is the single most important determinant (see Fig. 30.6). Although DXA instruments are ultimately calibrated against excised bone samples, methodologic differences in how this is performed have led to large discrepancies in patient measurements when performed on instruments from different vendors. Calibration differences between otherwise identical machines also occur. Such differences are usually small (1% to 2%) but on occasion can be clinically significant (exceeding instrument reproducibility). Therefore, measurements from different machines are very difficult to compare, and whenever possible follow-up examinations should be performed on the same machine.

FIGURE 30.12. Quality assurance plots from three hypothetical systems with mean 1, standard deviation 0.005 (coefficient of variation 0.5%). The upper plot indicates a stable machine, the middle plot indicates an abrupt shift in baseline (regression lines plotted to the data prior to and after the shift), and the lower plot indicates a continuous drift in baseline (regression line plotted to all time points). The latter is particularly insidious and would be difficult to identify by visual inspection of the data alone. (Reproduced with permission from Leslie WD, Roe BE. Bone densitometry. In: Leslie WD, Greenberg D, eds. Nuclear Medicine. Austin: Landes Bioscience, 2003:93–120.)

FIGURE 30.13. Effect of a large soft tissue fold (pannus) on hip dual-energy x-ray absorptiometry assessment is illustrated in this 64-year-old Caucasian woman (height 162.56 cm [64 in], weight 118 kg [261 lb], body mass index 45 kg/m2). The upper images show the scan before displacement of the pannus (arrowheads) and the lower images after displacement of the pannus (bone display on the left, tissue display on the right). The total hip measured bone mineral density decreased from 0.867 to 0.779 g/cm2 (8.5% difference).

Assessment of precision errors in BMD testing is a prerequisite for characterizing longitudinal changes (see Monitoring with Bone Densitometry).56,57 The International Society for Clinical Densitometry (ISCD) has proposed a standardized methodology for such precision studies.58,59 The ISCD procedure states that precision error (SD) should be obtained from an assessment with 30 degrees of freedom (e.g., 30 individuals with two scans) drawn from the patient referral population and using the root mean square (RMS) approach:

where m subjects have paired measurements with dj the difference in the first and second measurements. (Slightly different formulae are used for 15 individuals with three scans or where subjects have unequal numbers of scans.56)

Subjects’ scans are commonly performed after simple repositioning (arising from the scanner table between the repeat measurements), though this is known to underestimate day-to-day measurement error,60,61 and can lead to a significant rate of over categorization of change (up to 19.3% for the lumbar spine and up to 18.3% for the total hip).61 Other factors shown to have an adverse effect on BMD precision are obesity,62 an abdominal pannus overlying the hip63 (Fig. 30.13), and the presence of focal structural defects with the lumbar spine necessitating vertebral exclusions.64,65

Precision can be stated as either SD or percent coefficient of variation (%CV, defined as 100 • √SD/mean). Although precision error is commonly stated as a %CV, error is independent of bone mass and will therefore be underestimated in the lower (osteoporotic) range.66 It is therefore preferable to express precision error as the absolute SD (g/cm2), and to use this as the basis for determining whether change in BMD is significant or simply because of the measurement error.66 The smallest change that must be present before one can conclude (with 95% confidence) that the change in BMD is not related to measurement error is 2.77 • SD,56,57 and a value known as the least significant change (LSC). Vendors frequently cite in vivo precision as ~1% for DXA instruments, but this significantly underestimates the error seen in nonresearch, clinical populations. The minimum acceptable precision is: Lumbar spine: 1.9% (LSC = 5.3%), total hip: 1.8% (LSC = 5%), femoral neck: 2.5% (LSC = 6.9%).67 Reproducibility is optimized through a systemic process that includes careful quality control of the instrument, scanning technique, and analysis. Reproducibility is further compromised when examining smaller regions of interest. Femoral neck precision is much worse than the total hip region, and Ward’s region (a subregion within the femoral neck that contains little trabecular bone) is so variable that it is of no clinical value.

FIGURE 30.14. An 80-year-old woman’s bone mineral density (BMD) of the hip that is average for her age (648 mg/cm2) will have a Z-score of –0.5 but a T-score of –2.5. A 30-year-old woman with exactly the same bone density measurement would have a T-score and Z-score that are both –2.5. WHO, World Health Organization. (Reproduced with permission from Leslie WD, Roe BE. Bone densitometry. In: Leslie WD, Greenberg D, eds. Nuclear Medicine.Austin: Landes Bioscience, 2003:93–120.)

T-Scores and Z-Scores

Absolute measurements of bone density are of limited value. They are influenced by the site chosen for measurement, calibration used by the equipment manufacturer, and even the particular instrument. Because bone density follows a bell-shaped (Gaussian) distribution, measurements are conventionally described in terms of the number of SDs that a value deviates from the population mean (Fig. 30.14). Age-related changes in bone density, described earlier, must be taken into account.

The Z-score refers to the number of SD above or below the mean for an age- and sex-matched population (if reference data are available then race/ethnicity should also be considered):

The T-score refers to the number of SD above or below the mean for a young adult population (age 20 to 30 years):

In interpreting an individual patient’s test results, the following question is confronted: Should you compare the patient with someone of the same age or with a young adult? The former masks the increasing prevalence of osteoporosis with advancing age, while judging a 90-year-old against the same standard used in a 30-year-old seems unreasonable. In reality, both approaches have merits and are complementary. An age-adjusted measurement indicates whether the individual is average for their age and, if not, how markedly they deviate from the expected value. On the other hand, bone strength depends upon bone mass and not the age of the subject, therefore predictions in terms of fracture risk are best based upon comparison with an absolute standard (young adult).

In 1994, the WHO formulated diagnostic ranges for osteoporosis based upon T-score which have been refined more recently.1,14,15 Fracture risk is continuous – there is no “fracture threshold”68 – and these ranges were originally intended to be used to provide a framework for collection of epidemiologic data. The classification was based upon data derived almost exclusively from postmenopausal Caucasian women, but in recent years have been extended to all menopausal women and also men after the age of 50 (Table 30.1). The operational definition of osteoporosis is a BMD T-score that lies 2.5 SDs or more below the average mean value for young healthy women (T-score ≤−2.5 SD) based upon a standardized reference site (the femoral neck) and a standard reference range for both men and women (the NHANES III data for women aged 20 to 29 years). Classification of T-scores between −1 and −2.5 is more controversial.69 In the past this was referred to as “osteopenia,” terminology that created problems as a word commonly used by radiologists to describe the radiographic appearance of bones, and confused with “osteoporosis” by patients and physicians as an equally serious medical disorder requiring treatment. For these reasons, the term “osteopenia” is best avoided, and the descriptive term “low bone mass” or “low bone density” is now preferred for people with T-scores between −1 and −2.5.

The occurrence of fragility fractures, especially minimal trauma vertebral compression fractures, is sufficient for a clinical diagnosis of osteoporosis in any group providing that other pathologic causes (such as neoplasm) have been excluded. The criteria for diagnosing osteoporosis from bone density alone are more controversial when applied to groups other than postmenopausal Caucasian women. For example, men have higher average bone density than women but also have a lower fracture rate. The absolute likelihood of fracture in men seems to show the same relationship to absolute bone density as in women.70,71 This has led to uncertainty over whether the osteoporosis threshold in men should be based upon a male or female reference group.72 Similar concerns arise over comparison to non-Caucasian groups.69 Black subjects have significantly higher bone density and lower fracture rates than Caucasians, suggesting a similar relationship between absolute bone density and absolute fracture risk. In contrast, Asians have lower hip bone density than Caucasians but paradoxically have a much lower fracture likelihood. The reasons for this are still debated but include shorter stature, likelihood of falling, and possibly shorter femoral neck (resulting in shorter hip axis length).

Finally, osteoporosis should not be diagnosed by BMD alone in premenopausal women, men younger than the age of 50, or children.69 The recommendation to use Z-scores in young adults is, in part, intended to prevent misapplication of WHO criteria, which are based upon the T-score, and to avoid inappropriate treatment. The WHO criteria were developed for postmenopausal females, and extrapolation to other groups may not identify people at equivalent levels of fracture risk. Age is a strong predictor of fracture risk that is independent of BMD. Because healthy young adults have low short-term fracture risk across the BMD spectrum an arbitrary T-score (or Z-score) diagnostic cutoff does not have the same significance in a 30-year-old as in a 70-year-old. BMD measurement in healthy premenopausal women and men younger than the age of 50 cannot be used to assess short-term fracture risk and is only recommended to assess the skeletal impact of specific conditions. Conceptually, the Z-score is best suited to this purpose and avoids incorrectly labeling patients according to WHO criteria. Figure 30.15 shows DXA results for a 28-year-old healthy male who had unnecessary BMD testing. He was concerned to be told that he had osteoporosis and high fracture risk based upon the lumbar spine T-score of −2.6, highlighting the harm that can be done by inappropriate testing and reporting. A more accurate interpretation would be that his spine BMD is “below the expected range for age” (hip BMD is “within the expected range for age”) and his fracture risk is currently low. Reduced BMD in healthy young people does not predict high fracture risk (young age is a stronger BMD-independent protective factor).73 Reduced BMD in healthy young people usually reflects reduced attainment of peak bone mass, not BMD loss or high bone turnover (which are characteristic features of postmenopausal and age-related osteoporosis). Drug therapy is usually not warranted, though optimization of diet, exercise, and lifestyle would be reasonable, in addition to limited investigation to exclude an undiagnosed medical condition. The reduced BMD in this case was attributed to being very slim with small skeletal size that did not provide an accurate measurement of his true volumetric BMD (see Limitations and Errors in Dual Energy X-ray Absorptiometry).

FIGURE 30.15. Dual-energy x-ray absorptiometry (DXA) results for a 28-year-old healthy male who had unnecessary bone mineral density (BMD) testing.

Pediatric Bone Densitometry

It is inappropriate to generate T-scores or apply WHO diagnostic criteria to children because they have not yet achieved peak bone mass. The interpretation of BMD in children is challenging as BMD is expected to increase (not remain stable as in adults), and BMD measurements by DXA are strongly affected by height status. Delayed skeletal growth and maturation interferes with the use of age-matched reference data. Erroneous interpretation can arise when appropriate adjustments are not made for children with growth or maturational delay. Many methods to adjust BMC/BMD Z-scores for height give biased results, though the height-age Z-score (HAZ) method show the least biased relative to HAZ and age and can be used to evaluate the effect of short or tall stature on BMC/BMD Z-scores.74,75

FIGURE 30.16. Total-body dual-energy x-ray absorptiometry assessment in a young female scanned initially at age 10.5 and again at age 16.6 (see text). BMC, bone mineral content; BMD, bone mineral density.

Figure 30.16 shows the DXA assessment in a young female with steroid-dependent lupus scanned initially at the age of 10.5 and again at the age of 16.6. BMD measurements were virtually identical (0.884 g/cm2 and 0.896 g/cm2) but because of the changing age-specific reference data the Z-scores were very different (initial −0.2 was within the expected range for age, final −2.9 was below the expected range for age). Height at the time of the final scan was 147.2 cm (58 in) which is well below the average age (HAZ −2.4). When BMD is HAZ-adjusted, the Z-score improves to −1.7 which is now within the expected range for age (though still slightly below average), indicating the importance of considering and adjusting for impaired skeletal growth. For a detailed review of pediatric bone densitometry the reader is encouraged to read the reviews and Official Positions from the 2007 International Society for Clinical Densitometry Pediatric Position Development Conference.7679

FIGURE 30.17. Body composition analysis for a 50-year-old woman being assessed prior to bariatric surgery. The body mass index (BMI) is in the obese range, with a total-body fat mass 53.8 kg and a total-body lean tissue mass 47.6 kg.

Body Composition

Overweight and obesity are reaching epidemic proportions in countries around the world, predicted to result in a high population burden of diabetes, cardiovascular and cerebrovascular diseases.80,81 Body mass index (BMI) is most frequently used to assess overweight and obesity, but may erroneously categorize some individuals.82 For example, it does not distinguish increased weight related to muscle mass in athletes from adiposity. It also does not account for the distribution of excess weight, and may not accurately reflect intra-abdominal fat83,84 which is more closely linked to metabolic syndrome and its complications than other fat depots.85,86 The greater importance of abdominal fat is highlighted in data from 2,739 US women who participated in the Heart and Estrogen/progestin Replacement Study which found that larger waist circumference was associated with increased mortality whereas higher BMI (adjusted for waist circumference) was associated with decreased risk of total mortality, independent of cardiac risk factors.87

DXA can estimate soft tissue composition by decomposing the amount of attenuation from the two x-ray energies in bone-free pixels to calculate the amount of lean tissue and fat tissue in the path of the beam. DXA is most widely used for assessment of osteoporosis but is also a well-validated technique for body composition analysis (Fig. 30.17).88 DXA measures account for 80% of the variation in intra-abdominal fat as measured by CT.89 Data indicate that DXA-derived body composition can in turn be used to predict overall mortality and cardiovascular-related deaths.90 Whole-body fat mass and lean mass can be measured and specific subregions, such as the trunk, can be extracted from these scans. Regional fat tissue composition can be measured from spine and hip DXA, and estimates total-body composition more accurately than with BMI.91 DXA-derived abdominal fat measurement has been shown to predict risk for subsequent diabetes diagnosis in 30,252 nondiabetic women aged 40 years and older referred for baseline osteoporosis assessment.92 During mean 5.2 years of observation, 1,252 (4.1%) women met the case definition for diabetes. Greater proportion of abdominal fat from spine DXA was strongly related to subsequent diabetes diagnosis in models adjusted for age, BMI, and other comorbidities. Those in the highest quintile had 3.56 (95% CI: 2.67 to 4.75) times the risk for subsequent diabetes diagnosis compared with those in the lowest quintile.

DXA-derived body composition is still largely a clinical research tool, but may assume increasing clinical relevance for cancer survivors. AIs used as adjuvant therapy in postmenopausal breast cancer induce a reduction in bioavailable estrogens with alterations in body composition.93 In a subanalysis of a 2-year randomized clinical trial of 82 women with nonmetastatic breast cancer, newly menopausal following chemotherapy, women on AIs gained a significant amount of lean body mass compared to baseline as well as to non-AI users. Women not on an AI gained total body fat compared to baseline and AI users. There are epidemiologic data linking overweight, diabetes, and postmenopausal breast cancer.94 The RR for postmenopausal breast cancer is around 1.5 for overweight women and >2 for obese women; diabetes is associated with postmenopausal breast cancer with summary RRs from meta-analyses of 1.15 to 1.20. A systematic review and meta-analysis concluded that women with breast cancer and pre-existing diabetes have a greater risk of death, tend to present at more advanced stages and receive altered treatment regimens.95 Changes in body composition are also relevant to men with prostate cancer undergoing androgen deprivation therapy (ADT). A systematic review and meta-analysis found that ADT for prostate cancer is associated with increased fat and decreased lean mass96 which likely contributes to the reported increased risk for diabetes and cardiovascular disease in men with prostate cancer receiving ADT.97 Pooling data from 16 studies showed that ADT increased percent body fat by on average 7.7% (95% CI: 4.3, 11.2, from seven studies, p < 0.0001) and decreased percent lean body mass by on average −2.8% (95% CI: −3.6, −2, from six studies, p < 0.0001), with increased body weight (2.1%, p < 0.0001 from nine studies) and BMI (2.2%, p < 0.0001, from eight studies).96

FIGURE 30.18. A 60-year-old woman has bone density measurements of the hip with a T-score of –2 and a Z-score of –3. The relative risk (RR) of hip fracture (compared with an average 60-year-old woman) increases with an RR of 2.6 for each standard deviation below average. Therefore, the RR of hip fracture is increased 18-fold (2.63), placing the woman at markedly increased risk. (Reproduced with permission from Leslie WD, Roe BE. Bone densitometry. In: Leslie WD, Greenberg D, eds. Nuclear Medicine. Austin: Landes Bioscience, 2003:93–120.)


Predicting Fractures from Bone Density Measurements

The mechanical strength of excised bone is strongly related to the amount of bone mineral. Although bone density is on average significantly lower in fracture patients than in nonfracture patients, there is considerable overlap between the two groups. Risk of fracture shows a continuous gradient relationship with bone density: There is no “fracture threshold.” Bone density measurement provides a relative gradient of risk (usually expressed as a rate ratio [RR] per SD) that is as good as other commonly used risk stratification measures such as blood pressure for stroke and serum cholesterol for cardiovascular disease.68 The increase in fracture risk with decreasing bone density is exponential (not simply additive). The effect of a progressive reduction in hip BMD on hip fracture risk is shown in Figure 30.18, where a reduction by 3 SDs translates into a 17.6-fold increase in fracture risk. Prospective studies show that all studies measuring bone density at any site had similar predictive ability for a decrease of 1 SD in bone density (RR per SD ∼1.5), except for measurements at hip and spine, which have better predictive ability for fractures in hip (RR per SD 2.6) and spine (RR per SD 2.3), respectively. Figure 30.19 illustrates why a larger gradient of risk improves fracture risk stratification.

Absolute Fracture Risk Assessment Tools

The ability to accurately determine fracture risk is critical in identifying cost-effective thresholds for intervention.13,98 Recently, there has been a shift from risk assessment based upon T-score categories to absolute fracture risk based on 10-year absolute fracture probability.13 Although reduced bone mass is an important and easily quantifiable measurement, studies have shown that most fractures occur in individuals with a BMD T-score above the operational threshold for osteoporosis.2,99 The use of new fracture risk prediction systems that integrate multiple CRFs has been shown to enhance the performance of BMD in the prediction of hip and other major osteoporotic fractures.100 This has initiated a paradigm shift in the field of osteoporosis. One such example, the fracture risk assessment tool (FRAX,, was developed by the WHO Collaborating Centre for Metabolic Bone Diseases for estimation of individual 10-year osteoporotic and hip fracture probability (Fig. 30.20).101 The output of FRAX is the 10-year probability of a major fracture (hip, clinical spine, humerus, or wrist fracture) and the 10-year probability of hip fracture. In addition to a prior fragility fracture, age, sex, BMI, and additional risk factors for fractures were identified including the prior use of glucocorticoids, secondary osteoporosis, rheumatoid arthritis, a parental history of hip fracture, current cigarette smoking, and alcohol intake of three or more units daily (Table 30.3).102 Femoral neck BMD can be optionally entered to enhance fracture risk prediction. Unlike other algorithms,103105 fracture probability is computed taking both the risk of fracture and the risk of death into account. This is important because some of the risk factors affect the risk of death as well as the fracture risk. Examples include increasing age, low BMI, low BMD, glucocorticoids, and smoking. Population-specific FRAX tools are customized to the fracture and mortality epidemiology in a specific region.101 Table 30.4 shows the marked variation in fracture probabilities that occur by using a different FRAX calculation tool. The highest and lowest calculated probabilities differ by more than an order in magnitude, consistent with the large global variation in fracture rates, and highlighting the importance of using the appropriate FRAX tool. At present more than 50 FRAX models are available, and others are being developed. The FRAX system has been endorsed and integrated into CPGs by several national bodies.106–114Despite the wide acceptance of FRAX, it should not be uncritically used in the management of patients without an appreciation of its limitations as well as its strengths. Some of these limitations were highlighted in a recent set of reports from joint IOF–ISCD Task Forces.115 For example, measurements other than BMD or T-score at the femoral neck by DXA cannot be used in FRAX. The FRAX algorithm was calibrated for use with femoral neck BMD based upon the strength of the association with subsequent fractures (particularly hip fractures) in the FRAX derivation cohorts and the availability of a standardized young adult reference database (NHANES III white female) for calculation of T-scores. Lumbar spine BMD is also strongly associated with future fracture risk, especially spine fractures.68 Therefore, FRAX may underestimate or overestimate major osteoporotic fracture risk when lumbar spine T-score is much lower or higher (>1 SD discrepancy) than femoral neck T-score. It is not uncommon to find situations where T-scores in the lumbar spine and femoral neck show “discordance,” given the modest correlation in BMD between these two sites (typically R = 0.6 to 0.7).116,117 One report found that approximately one in eight women had discordance exceeding 2 SD based upon lumbar spine, femoral neck, total hip, and trochanter.118 It may seem intuitively obvious to practicing clinicians that when two individuals differ in their spine measurements (e.g., Patient 1 with lumbar spine T-score = −1.5 versus Patient 2 with lumbar spine T-score = −3.5) but who are identical in all other respects (Patient 1 and Patient 2 both have femoral neck T-score = −1.5), the individual with the lower lumbar spine measurement (Patient 2) would be at higher fracture risk. These two individuals would generate identical fracture probabilities under FRAX, though available data suggest that Patient 2 in the scenario above does indeed have higher fracture risk than Patient 1.119 Observed fracture rates exceeded those predicted by FRAX when the lumbar spine T-score was much lower than the femoral neck T-score (>1 SD discrepancy), and conversely, fracture rates were lower than predicted when the lumbar spine T-score was much higher than the femoral neck T-score.120

FIGURE 30.19. Relative fracture risk according to bone mineral density T-score and gradient of risk (relative risk [RR] per standard deviation reduction).

FIGURE 30.20. Screen page for input of data and format of results for a US Caucasian model using the fracture risk assessment tool known as FRAX. Results are for a hypothetical 65-year-old Caucasian woman who smokes cigarettes and has a femoral neck T-score of −2.3 which is above the osteoporotic threshold. Under the National Osteoporosis Foundation (NOF) and American Association of Clinical Endocrinologists (AACE) guidelines, she would qualify for treatment based upon a 10-year hip fracture probability exceeding 3%. (US FRAX tool version 3.4,

TABLE 30.3


A simple procedure for adjusting the FRAX estimation of major osteoporotic fracture probability based upon the T-score difference (offset) between the lumbar spine and femoral neck has been endorsed with the ISCD–IOF Task Force.120 For every offset SD difference there was an approximately 10% change in fracture risk that was higher when the lumbar spine T-score was less than the femoral neck T-score and lower when the lumbar spine T-score was greater than the femoral neck T-score. The offset adjustment was found to reclassify fracture probability in a relatively small proportion overall (less than 10%) but reclassified a larger number of individuals with moderate risk and an offset greater than 1 SD (one in four). The rule that was developed was, “Increase/decrease FRAX estimate for a major fracture by one-tenth for each rounded T-score difference between the lumbar spine and femoral neck.” An example of the rule application follows: “Consider an individual with femoral neck T-score −1.7 and major osteoporotic FRAX probability 18%. If the lumbar spine T-score is −3.5 then this indicates an offset of −1.8 (3.5 minus −1.7). This is rounded to the nearest whole number (−2). One-tenth of the FRAX estimate based upon the femoral neck is 1.8%, which is multiplied by the rounded offset value (giving 3.6%). This is then added (because lumbar spine T-score is lower than femoral neck T-score) to the original FRAX estimate (18%) giving a final (rounded) probability of 22% (= 18% + 3.6%).

TABLE 30.4



Although primary osteoporosis is most commonly seen in postmenopausal women and the elderly, a number of secondary causes of bone loss should be considered in the evaluation of an individual with low bone mass (Table 30.3). These factors may exist in isolation, or accelerate bone loss in primary osteoporosis. The clinical assessment should be directed toward elucidating these potential causes and any fracture history. Assessment of the risk factors for falls is an important adjunct to measuring bone density. Laboratory testing can be limited to the measurement of serum calcium, alkaline phosphatase, creatinine, and a complete blood count. In individuals with postmenopausal or age-related osteoporosis, all of these indices should be within the normal range. These investigations may be further expanded to include serum TSH, parathyroid hormone (PTH), serum 25-hydroxyvitamin D, protein electrophoresis, and 24-hour urinary calcium determination, as guided by clinical judgment. Although not routinely required, imaging of the thoracolumbar spine helps to determine the number and type of pre-existing vertebral fractures, and their presence indicates high risk for further fractures. Biochemical markers of bone metabolism provide an indirect method to evaluate the rate of bone turnover.121 Osteocalcin, bone-specific alkaline phosphatase, and procollagen I peptide can assess the level of bone formation whereas urinary or serum measurements of type 1 collagen cross-links (such as deoxypyridinoline, N-telopeptide, and C-telopeptide) reflect the level of bone resorption. These markers have not been useful for diagnosing osteoporosis or for predicting bone loss, but some studies suggest that they independently predict fracture risk.122

Nonpharmacologic Therapy

General measures to reduce the risk of fractures should include assessment of hazards in the home environment, sedative use, muscle weakness, postural hypotension, and uncorrected visual deficits. Exercise should also be encouraged in an attempt to preserve bone mass and to maintain or improve muscular conditioning. Calcium supplementation is often necessary in postmenopausal and elderly women to reach recommended targets (1,200 mg/day). Calcium carbonate is the most commonly used supplement and is recommended for those who do not have achlorhydria (common in the very elderly and those on acid suppressing medication). Vitamin D supplementation (minimum recommended 600 to 800 IU daily, maximum 4,000 IU daily) is also advised, even in those on antiresorptive therapy.123 The importance of calcium and vitamin D supplementation is even greater in the elderly as calcium absorption is impaired and sunlight exposure is reduced, especially in winter months.

Pharmacologic Therapy

There are expanding evidence-based options for osteoporosis therapy.124 Estrogen replacement has largely been replaced by bisphosphonates, nasal calcitonin, selective estrogen receptor modulators (SERMs), and denosumab. These agents act primarily through inhibition of osteoclast number and activity. In contrast, there is only one group of approved anabolic agents that act by stimulating osteoblast activity. For vertebral fracture prevention, the following agents have good evidence to support their use for individuals at high risk of fracture: Alendronate, risedronate, ibandronate, zoledronic acid, teriparatide, raloxifene, denosumab, and estrogen.124 There is fair evidence for the use of calcitonin in vertebral fracture prevention. For hip fracture prevention, the following therapies have good evidence: Alendronate, risedronate, zoledronic acid, denosumab, and estrogen. For nonvertebral fracture prevention, there is good evidence for alendronate, zoledronic acid, risedronate, teriparatide, denosumab, and estrogen.124 Therapeutic benefit is reduced or eliminated if there is suboptimal adherence to the regimen, including frequently missed doses, failing to take the medication correctly to optimize absorption and action, or discontinuation of therapy.125127 Compliance rates at 1 year in the range 25% to 50% with oral anti-osteoporosis agents are commonly reported, and are only marginally better with less frequent dosing regimens.125,127


This class of drugs inhibits osteoclast number and activity, have few extra-skeletal effects and have become the first-line treatment for osteoporosis. Alendronate is a potent aminobisphosphonate that is capable of increasing bone density at the lumbar spine by 8% over 3 years. More importantly, a meta-analysis of 11 studies representing 12,068 women receiving at least 1 year of alendronate for postmenopausal osteoporosis128 was associated with significant reductions in vertebral fractures (RR: 0.55; 95% CI: 0.43 to 0.69) and for the secondary prevention of nonvertebral fractures (RR: 0.77; 95% CI: 0.64 to 0.92), wrist fractures (RR: 0.50; 95% CI: 0.34 to 0.73), and hip fractures (RR: 0.47; 95% CI: 0.26 to 0.85).128 Similarly, a meta-analysis assessing the efficacy of risedronate in the prevention of osteoporotic fracture in postmenopausal women found a 37% (CI: 0.51 to 0.77) reduction for vertebral fractures, a 20% reduction for nonvertebral fractures (RR: 0.80; 95% CI: 0.72 to 0.90), and a 26% reduction (RR: 0.74; 95% CI: 0.59 to 0.94) for hip fractures.129 In two trials with zoledronic acid there was evidence of vertebral (RR: 0.33; CI: 0.274 to 0.4), nonvertebral (RR: 0.75; CI: 0.66 to 0.85), and hip fracture (RR: 0.62; CI: 0.47 to 0.83) reduction.130


Hormone replacement therapy (HRT) is effective at preventing bone loss in postmenopausal women and is associated with a bone density increase of 4% to 6% in the first 2 years of therapy. HRT reduces overall fractures by about 30%, with benefit seen for vertebral fractures (RR: 0.67; CI: 0.48 to 0.93), nonvertebral fractures (RR: 0.73, CI: 0.64 to 0.81), and hip fractures (RR: 0.60; CI: 0.42 to 0.93).130 HRT is also associated with increased risk of breast cancer, cardiovascular events, and thromboembolism. Therefore, HRT is no longer seen as a first-line treatment for osteoporosis, but can be beneficial in women with moderate-to-severe vasomotor symptoms.

Selective Estrogen Receptor Modulators

New agents have been developed that retain estrogen’s positive effects on bone, while attempting to minimize the associated risks and side effects. This class of drugs is termed SERMs and includes raloxifene and tamoxifen. Raloxifene reduces the risk of vertebral fractures by approximately 40% (RR: 0.64; CI: 0.54 to 0.78), and produces increase in bone density of approximately 3% at the spine in its first 3 years, but has not been shown to protect against nonvertebral fractures.131 It also lowers total and LDL cholesterol, does not stimulate the endometrium, and reduces the risk of breast cancer.


Salmon calcitonin has been used subcutaneously for many years. The intranasal form is more acceptable for long-term use. Despite modest effects on bone density, vertebral fracture risk is reduced by approximately 35% (RR: 0.65; CI: 0.48 to 0.88) though there is no effect for nonvertebral fractures.130 Intranasal calcitonin has been shown to reduce pain associated with acute vertebral fractures. A possible association between calcitonin use and cancer has recently led to a call for discontinuing its use in long-term osteoporosis treatment.132


Denosumab is a human monoclonal antibody to the RANK-L that blocks its binding to RANK, inhibiting the development and activation of osteoclasts. In an RCT of 7,868 women, denosumab given twice yearly reduced the risk of hip fracture by 40% compared to placebo (hazard ratio: 0.60; 95% CI: 0.37 to 0.97; ARR 0.5%) and also reduced the risk of nonvertebral fracture by 20% (hazard ratio:, 0.80; 95% CI: 0.67 to 0.95; ARR: 1.5%).133

Anabolic Agents

Strategies which directly stimulate osteoblast activity have the potential to produce larger increases in bone density and potentially larger effects on fracture prevention. One such strategy is the use of intermittent PTH therapy which has been demonstrated to produce large increases in BMD. Teriparatide, a recombinant form of PTH (amino acid sequence 1 through 34), is approved for the treatment of osteoporosis in men and postmenopausal women who are at high risk for fracture. A meta-analysis concluded that both vertebral fractures (RR: 0.36; CI: 0.23 to 0.57) and nonvertebral fractures (RR: 0.49; CI: 0.27 to 0.87) were reduced by teriparatide.130 However, this agent is contraindicated in those diagnosed with cancer because of an association between the drug and osteosarcomas in laboratory rats.

Adverse Events

Although generally well tolerated, there are a number of uncommon or rare adverse effects that have been associated with approved treatments, though a causal link has not always been clear.134,135 An infrequent adverse effect of oral aminobisphosphonate use is erosive esophagitis, especially in patients with prior esophageal disease, gastroesophageal reflux, or when directions are not carefully followed.134 Osteonecrosis of the jaw (ONJ) was initially described in 2001 in oncology patients receiving high-dose intravenous bisphosphonate therapy. Large epidemiologic studies have described an incidence of ONJ ranging from 2% to 11%; concomitant use of glucocorticoids, several chemotherapies, and recent dental extractions appears to enhance the risk of ONJ. Among osteoporosis patients the risk is much less clear, with an incidence between one in 10,000 and one in 100,000.134,135 Severe bone pain, esophageal cancer, and atrial fibrillation have been described among bisphosphonate users, but their relationship is questionable. Of great concern are emerging reports of atypical subtrochanteric or diaphyseal femoral fractures in osteoporosis patients that have received long-term bisphosphonate therapy.136 Major features include a transverse or short oblique orientation, minimal or no associated trauma, a medial spike when the fracture is complete, and absence of comminution; minor features include cortical thickening, a periosteal reaction of the lateral cortex, prodromal pain, bilaterality, delayed healing, comorbid conditions, and concomitant drug exposures, including bisphosphonates, other antiresorptive agents, glucocorticoids, and proton pump inhibitors. Recent observations suggest that the risk rises with increasing duration of exposure.137 It is important to keep these reports in perspective: The incidence of atypical femoral fractures associated with bisphosphonate therapy for osteoporosis appears to be much smaller than the number of vertebral, hip, and other fractures that are prevented by bisphosphonates.

Who to Treat?

Fracture rates are known to vary by more than an order of magnitude worldwide, therefore a single approach cannot be universally applied to all countries. National considerations must reflect the burden of osteoporosis, available resources, the disease costs to the individual and society, and how these relate to competing health and other societal priorities. Recent developments in terms of diagnosis, fracture risk prediction, and therapeutic options have prompted many countries to review and update their clinical practice guidelines for the prevention and management of osteoporosis intended for use in primary care in the general adult population. A recent review examined updated guidelines from the following countries: Australia, Belgium, Canada, Germany, the United Kingdom, and the United States.138

The American Association of Clinical Endocrinologists (AACE) updated US guidelines for prevention of and treatment of osteoporosis and related fractures in 2010.139 The scope of the guideline was limited to postmenopausal women. Bone densitometry was recommended for all women aged 65 and older, and for postmenopausal women younger than the age of 65 with one or more additional risk factors for fracture. The diagnosis of osteoporosis by BMD criteria was limited to BMD at the lumbar spine, total hip, femoral neck, and one-third radius site. Recommendations are also given for work-up of individuals with osteoporosis by bone density criteria or fragility fractures for causes of secondary osteoporosis, indications for lateral spine imaging to detect prevalent vertebral fractures, for calcium and vitamin D intakes, and for lifestyle interventions to reduce risk of fractures.

The AACE endorsed the U.S. National Osteoporosis Foundation 2008 guidelines, and on these aspects the two guidelines are identical. Pharmacotherapy (with an oral or IV bisphosphonate, calcitonin, denosumab, raloxifene, or teriparatide) is recommended for those who have had a hip or vertebral fracture (radiographic or clinical), T-score ≤−2.5 at femoral neck, total hip, or lumbar spine, or T-score −1 to −2.5 with hip or major osteoporotic 10-year fracture risks, respectively, of ≥3% or ≥20%. This guideline recommends pharmacologic therapy for a substantially wider proportions of the population than other guidelines; for example, 91% of non-Hispanic white women aged 80 and older with T-scores between −1 and −2.5 would be offered drug therapy using this guideline.140 Because of their proven hip fracture reduction efficacy, alendronate, risedronate, zoledronic acid, and denosumab are recommended as the first-line agents. Simultaneous use of two or more agents is not recommended. However, sequential therapy is stated to be appropriate at times, especially use of antiresorptive agents after a treatment course of teriparatide. The duration of drug therapy considered to be safe is explicated in the guideline. Drug holidays after many years of bisphosphonate therapy are thought to be reasonable so long as BMD is followed and treatment reinitiated if bone loss occurs.

Glucocorticoids are often used in the management of oncologic and inflammatory conditions, but are associated with increased risk for osteoporosis, with resultant fractures.141 A rapid decline in BMD begins within the first 3 months of glucocorticoid initiation and peaks at 6 months, followed by a slower steady loss with continued use. An increased risk of both vertebral and nonvertebral fractures occurs with dosages of prednisone or equivalent as low as 2.5 to 7.5 mg daily. In 2010, the American College of Rheumatology (ACR) published recommendations for the prevention and treatment of glucocorticoid-induced osteoporosis.141 Updated approaches to identify patients at highest risk for fracture using the 10-year absolute probability of fracture calculated by the FRAX tool are incorporated into these guidelines.

Monitoring with Bone Densitometry

Monitoring BMD in patients not receiving active treatment can help in the identification of individuals with rapid bone loss (“fast losers”). Repeat testing may also be useful in confirming a positive treatment response, although some evidence suggests that much of the antifracture effect of current antiresorptive therapies is mediated through mechanisms other than increasing bone mass.142 Stable BMD is consistent with successful treatment. Therefore, the major objective of repeat testing in treated patients is to identify individuals with continued BMD loss, despite appropriate osteoporosis treatment. Measurement error must be considered when interpreting serial BMD assessments to determine whether the change is real and not simply random fluctuation or artifact. Continued BMD loss exceeding the LSC may reflect poor adherence to therapy, failure to respond to therapy, or previously unrecognized secondary causes of osteoporosis (e.g., vitamin D insufficiency).

Patient-related factors also need to be considered at the time of repeat testing, including the average rate of expected bone loss and the maximum rate of loss that is likely to be encountered. The latter is critical because follow-up bone mass measurements should ideally identify patients who are failing treatment before substantial bone loss develops or fractures occur. Average rates of bone loss are greater in untreated early postmenopausal women (approximately 2% per year) than in older women (less than 1% per year). The site of most rapid bone loss also changes with age. Loss of trabecular bone from the spine exceeds that of the hip in early postmenopausal women. Similarly, increase in skeletal mass from antiresorptive treatment is usually most evident in the spine because of the relatively faster turnover of trabecular bone. For untreated older subjects the decline in the hip generally exceeds that of the spine because of the development of age-related degenerative artifacts in the spine. It should be emphasized that measurement imprecision makes it difficult to accurately assess loss rates in individuals. These have been stated to exceed 5% per year in some cases but the frequency of such rapid loss in the absence of major medical factors (such as high-dose steroid therapy) is low. It is likely that active treatment induces a “shift to the right” of the bone loss distribution, decreasing the number of cases in which rapid loss occurs.

No randomized trials have directly assessed the value of repeat BMD testing on persistence with medication or fracture reduction. Notwithstanding the lack of conclusive data, many patients and clinicians find value in an objective measurement that documents the effect of treatment.143 With careful attention to factors that can affect comparisons (Table 30.5), serial BMD testing can be a helpful clinical tool.144 Depending on the clinical situation, BMD scans are usually repeated every 1 to 3 years, with a decrease in testing once therapy is shown to be effective. In those at low risk without additional risk factors for rapid BMD loss, a longer testing interval (5 to 10 years) may be sufficient.33,145 If BMD is stable then less frequent monitoring can be considered.

TABLE 30.5



Breast Cancer

Breast cancer is the most common cancer of women in both the developing and developed world.146 Breast cancer survivors are at increased risk for osteoporosis147 and fractures.148 This has been attributed to the effects of chemotherapy, ovarian failure, early menopause, and more recently AIs. In postmenopausal women, the use of AIs increases bone turnover and induces bone loss at sites rich in trabecular bone at an average rate of 1% to 3% per year leading to an increase in fracture incidence compared to that seen during tamoxifen use.149 The bone loss is much more marked in young women with treatment-induced ovarian suppression followed by AI therapy (average: 7% to 8% per annum). Pretreatment with tamoxifen for 2 to 5 years may reduce the clinical significance of the adverse bone effects associated with AIs, particularly if this leads to a shortening in the duration of exposure to an AI. The rate of bone loss in women who experience a premature menopause before the age of 45 or are receiving ovarian suppression therapy is accelerated by the concomitant use of AIs. The Arimidex, Tamoxifen, Alone or in Combination (ATAC) trial included a BMD substudy; 108 patients treated with monotherapy AI or tamoxifen were included in the primary analysis.150 Among anastrozole-treated patients, there was a decrease in median BMD from baseline to 5 years in lumbar spine (−6.08%) and total hip (−7.24%) compared with the tamoxifen group (lumbar spine, +2.77%; total hip, +0.74%). Importantly, no patients with normal BMD at baseline became osteoporotic at 5 years.

Studies in breast cancer patients receiving adjuvant AIs versus tamoxifen have documented an increased risk of fracture (increase 15% to 113%).149,151155 It is notable that placebo arms were not included so that the effect of AIs alone is less clear, as tamoxifen has a favorable effect on BMD in postmenopausal women156 and this appears to translate into a reduced incidence of osteoporotic fracture.157 There has been no large adjuvant study of AIs versus placebo except for NCIC MA17.158 After 30 months of letrozole there was no difference in the incidence of fractures but because of the prior exposure to tamoxifen in all the patients in this study, it remains unclear what the effect the AI would be on the incidence of fracture compared to untreated patients. Support for the possibility that some of the differences in fracture risk between tamoxifen and AIs occur through a protective effect of tamoxifen comes from population-based study of 2,748 older breast cancer patients (mean age 73 years, 28.2% AI users, 27.8% tamoxifen users, the remainder not using hormone therapy).159

Several randomized trials have confirmed that antiresorptive treatments can preserve BMD in women with primary breast cancer under AI treatment, however reduction in fractures remains to be established (see review by Theriault160). For example, zoledronic acid and risedronate have been demonstrated to reduce AI-associated BMD loss. Zoledronic acid prevents bone loss in postmenopausal women with osteoporosis or low bone mass starting letrozole.161 Up-front zoledronic acid prevented AI-associated BMD loss with early breast cancer more effectively than delaying therapy until BMD loss or fracture occurs.162 As well, the addition of zoledronic acid to adjuvant endocrine therapy improves disease-free survival in premenopausal patients with estrogen-responsive early breast cancer.163 For patients taking adjuvant anastrozole for early breast cancer, risedronate resulted in significant increase in lumbar spine and total hip BMD.164 A small trial has also shown that the RANK ligand inhibitor, denosumab, increased lumbar spine BMD by 5.5% and 7.6% at 12 and 24 months versus placebo (p < 0.0001 at both time points); increases were also observed at the total hip, total body, femoral neck, and the one-third radius whereas bone turnover markers decreased.165

Serum estrogen is a risk factor for breast cancer166 and serum estrogen is expected to confer a reduced risk for osteoporotic fracture because of its effect on BMD.167 Interestingly, there are data showing that BMD measurements are associated with increased risk of overall and ER-positive breast cancer, and this risk is independent of the Gail score.168170

FIGURE 30.21. Baseline dual-energy x-ray absorptiometry scans for a 60-year-old Caucasian woman with breast cancer starting aromatase inhibitor therapy (anastrozole). She had natural menopause at age 53, body mass index 23.4 kg/m2, and no additional risk factors for fracture. Her 10-year major osteoporotic fracture probability is 14% and hip fracture probability 4%. BMC, bone mineral content; BMD, bone mineral density. (US FRAX tool version 3.4,

The National Comprehensive Cancer Network (NCCN) guideline on breast cancer (version 2.2011) recommends BMD monitoring at baseline and periodically in AI recipients. The optimal testing interval is unclear; a UK expert group suggested that postmenopausal women with normal BMD at baseline do not require monitoring beyond the usual recommendation for healthy postmenopausal women.150 Women who experience premature menopause are at greater risk for rapid BMD loss, and periodic monitoring is warranted in those receiving AIs even when baseline BMD is normal.150 The NCCN also suggests that women treated with a bisphosphonate should undergo a dental examination with preventive dentistry prior to the initiation of therapy, and should take supplemental calcium and vitamin D. A UK expert group concluded that bisphosphonates, along with a healthy lifestyle and adequate intake of calcium and vitamin D are the treatments of choice to prevent bone loss.149 Because of the rate of bone loss associated with breast cancer treatments, and uncertainties about the interaction between AI use and BMD for fracture risk, the threshold for intervention was set at a higher level than that generally recommended for postmenopausal osteoporosis.

Figures 30.21 and 30.22 show two representative cases of women undergoing AI therapy for breast cancer. The case in Figure 30.21 shows a woman starting anastrozole who is osteoporotic on her baseline BMD test with 10-year major osteoporotic fracture probability 14% and hip fracture probability 4%. Under the National Osteoporosis Foundation (NOF) and AACE guidelines, she would qualify for osteoporosis treatment (in addition to optimizing calcium and vitamin D) even in the absence of AI therapy based upon her osteoporotic BMD and hip fracture probability exceeding 3%. The case in Figure 30.22 indicates BMD loss in a woman who has completed 3 years of anastrozole. Despite a 9.2% decrease in spine BMD and a 10.7% decrease in hip BMD, measurements remain average to above average for age and her 10-year fracture risk is low. Continued monitoring and nonpharmacologic management without drug treatment for osteoporosis would be appropriate.

Prostate Cancer

ADT, achieved by bilateral orchiectomy or administration of luteinizing hormone-releasing hormone (LHRH) agonists, is the mainstay of treatment for advanced prostate cancer. Serum testosterone and estrogen fall to subnormal levels, and these hormones are important for maintaining bone mass because they exert antiapoptotic effects on osteoblasts and osteocytes and proapoptotic effects on osteoclasts. Use of ADT can improve survival in men with locally advanced prostate cancer but its prolonged use can lead to significant bone loss that may affect health-related quality of life in these men. More than 70% of men with prostate cancer are older than 65 and already at risk for osteoporosis or fragility fracture.171 Studies on bone health in men with prostate cancer receiving ADT and the current evidence regarding bone-health monitoring and management have been reviewed.172

FIGURE 30.22. Repeat (US FRAX tool version 3.4, scans for a 62-year-old Caucasian woman with breast cancer after 3 years of aromatase inhibitor therapy (anastrozole). She had natural menopause at age 53, BMI 34.9 kg/m2 and current smoking as an additional risk factor. Her 10-year major osteoporotic fracture probability is 5.2% and hip fracture probability 0.1%, which is low despite a significant decrease in spine BMD (0.103 g/cm2, 9.2%) and total hip BMD (0.120 g/cm2, 10.7%). BMC, bone mineral content; BMD, bone mineral density. (US FRAX tool version 3.4,

Bone loss in men who are receiving ADT is not a trivial issue. Men with nonmetastatic prostate cancer receiving continuous or intermittent ADT can have significant BMD loss as early as the first 6 to 12 months after starting ADT, with annual BMD decrements of –1.4% to –4.6% at the lumbar spine, –0.6% to –3.3% at the total hip, and –0.7% to –3.9% at the femoral neck.173 Several cohort studies have shown that men who receive ADT for prostate cancer are also at much higher risk for fracture.172,173 In a linked database of the Surveillance, Epidemiology, and End Results (SEER) program and Medicare, significantly more fractures occur in the 50,613 men with prostate cancer surviving at least 5 years after diagnosis in 1992 to 1997 who received ADT with LHRH agonists (19.4%) than in those who did not (12.5%, p < 0.001).174 In more recent population-based studies, adjusted odds ratios (ORs) for fragility fracture with current ADT were 1.71 (95% CI: 1.13 to 2.58)175 and 1.65 (95% CI: 1.53 to 1.77).176 Independent predictors of fragility and any fracture were increasing age, prior bone thinning medications, chronic kidney disease, prior dementia, prior fragility fracture, and prior osteoporosis diagnosis or treatment.175

A number of studies have investigated the effectiveness of antiosteoporosis therapies in men with nonmetastatic prostate cancer receiving ADT. A systematic review and meta-analysis of bisphosphonates in the treatment of osteoporosis in patients with prostate adenocarcinoma under ADT identified 15 published, randomized, placebo-controlled trials (2,634 participants). Treatment with bisphosphonates showed a substantial effect in preventing fractures (RR, 0.80; p = 0.005) and BMD-defined osteoporosis (RR: 0.39; p < 0.00001). Zoledronic acid showed the lowest number needed to treat (NNT), compared with placebo, in relation to fractures and osteoporosis (NNT = 14.9 and NNT = 2.68, respectively). The between-group difference (bisphosphonates versus placebo) in lumbar spine and femoral neck BMD were 5.2% and 2.4%. Denosumab has been evaluated for men receiving ADT in a double-blind, randomized trial.177 At 24 months, lumbar spine BMD had increased by 5.6% in the denosumab group (loss of 1% in the placebo group, p < 0.001) and had also significantly increased at the total hip, femoral neck, and distal third of the radius. Patients who received denosumab had a decreased incidence of new vertebral fractures at 36 months (1.5%, versus 3.9% with placebo, RR: 0.38; 95% CI: 0.19 to 0.78; p = 0.006). There are insufficient fracture data in studies with SERMs and salmon calcitonin.

A cost-effectiveness study concluded that universal alendronate use without a BMD test was not justifiable in a hypothetical cohort of men 70 years of age receiving a 2-year course of ADT for locally advanced or high-risk localized prostate cancer.178 The comparison looked at three patient groups: No BMD test and no alendronate therapy; a BMD test before ADT with selective alendronate therapy for 5 years in patients with osteoporosis; and universal alendronate therapy for 5 years without a baseline BMD test. Universal alendronate use was justifiable in men 80 years of age and older with a previous low-trauma fracture or with low BMD at baseline (femoral neck T-score −1.8 SDs or lower) and that alendronate can be considered for men in whom BMD test finds osteoporosis.

Under the NCCN guideline for prostate cancer (version 4.2011), screening and treatment for osteoporosis are advised according to guidelines for the general population from the NOF ( This includes recommendations for supplemental calcium (1,200 mg daily) and vitamin D3 (800 to 1,000 IU daily) for all men over the age of 50 years, and additional treatment for men when the 10-year probability of hip fracture is 3% or the 10-year probability of a major osteoporosis-related fracture is 20% according to the WHO–FRAX system. A baseline BMD study should be considered for men with surgical or medical ADT. When men with prostate cancer are diagnosed with low BMD or osteoporosis, their management should follow the guidelines set out for men with osteoporosis.

Thyroid Cancer

Most guidelines recommend the use of levothyroxine to maintain low TSH levels in treatment of patients with papillary, follicular, or Hürthle cell thyroid carcinoma.179 Thyroid hormone is known to stimulate the bone remodeling cycle, and biochemical markers of bone turnover are elevated in women with excess thyroid hormone.180 This has created the concern that excess thyroid hormone as used for TSH suppression may be associated with detrimental effects on bone, even in asymptomatic persons.

Clinically diagnosed hyperthyroidism has been associated with reduced BMD and increased hip fracture risk in older women,181 but not premenopausal women.182 Whether subclinical hyperthyroidism is a major risk factor for osteoporosis and/or fractures is less clear. The Study of Osteoporotic Fractures (SOF) prospectively measured serum TSH and serial BMD measurements in a subgroup of 487 women (including 198 thyroid hormone users) older than 65 years of age.180 Bone turnover markers (serum osteocalcin and bone-specific ALP) were elevated in women with low TSH, suggesting accelerated bone remodeling, but there was no consistent evidence that low TSH was associated with low baseline BMD or accelerated bone loss in older ambulatory women. A separate report from the SOF cohort assessed serum TSH in a nested case-control subgroup of women older than 65 years of age (148 women with new hip fractures, 149 women with new vertebral fractures, and 398 randomly selected women).183 Women with a low TSH level (<0.1 mU/L) had increased risk for hip fracture (adjusted relative hazard: 3.6; [95% CI: 1 to 12.9]) and vertebral fracture (adjusted odds ratio: 4.5 [CI: 1.3 to 15.6]) compared with women who had normal TSH levels. Unfortunately, serum thyroxine or triidothyronine levels were not measured and it was not possible to determine the independent effect of subclinical versus overt hyperthyroidism. Use of thyroid hormone itself was not associated with increased risk for fracture if TSH levels were normal. The Cardiovascular Health Study (CHS) is a prospective community-based cohort study of community-dwelling men and women aged 65 years and older of whom 3,567 had biochemically defined subclinical thyroid dysfunction or euthyroidism with 39,952 person-years of follow-up (median, 13 years).184 Among the 2,195 women, 359 (16.4%) had subclinical hypothyroidism and 142 (6.5%) had subclinical hyperthyroidism; no clear association between subclinical thyroid dysfunction and hip fracture was observed. In contrast, 18 of 184 men with subclinical hypothyroidism sustained a hip fracture (multivariable-adjusted hazard ratio: 2.31; 95% CI: 1.25 to 4.27) compared with 4 of 29 men with subclinical hyperthyroidism (adjusted hazard ratio: 3.27, 95% CI: 0.99 to 11.30).

The largest study to date used population record-linkage technology to identify patients over 18 years old with subclinical hyperthyroidism (identified from biochemistry, prescription, admission, and radioactive iodine treatment records) and five matched comparators from the general population.185 There were a total of 2,004 patients (77.4% female, mean age 66 years) with subclinical hyperthyroidism (1,491 had serum TSH from 0.1 to 0.4 mU/L, 414 had TSH concentrations <0.1 mU/L). During the total follow-up period of 76,124 years (median, 5.6 years), subclinical hyperthyroidism was associated with a slightly increased risk of osteoporotic fracture (HR: 1.25; 95% CI: 1.04 to 1.50), but this was no longer significant when patients who developed overt hyperthyroidism during follow-up were excluded (though there was increased risk of cardiovascular morbidity [HR = 1.36 (1.19 to 1.57)], dysrhythmia [HR = 1.39 (1.02 to 1.90)], and dementia [HR = 1.79 (1.28 to 2.51)]).

The NCCN guidelines (version 2.2012) recognize the potential toxicities associated with TSH-suppressive doses of levothyroxine—including bone demineralization (particularly in postmenopausal women)— and that patients whose TSH levels are chronically suppressed should be counseled to ensure adequate daily intake of calcium (1,200 mg/day) and vitamin D (1,000 U/day). There are no specific recommendations regarding BMD testing and monitoring though this would appear to be prudent in selected cases (e.g., women aged 65 years or older, and younger postmenopausal women or with additional risk factors, especially when serum TSH suppression <0.1 mU/L is required). An individualized approach to TSH management should consider the benefits and potential for adverse effects induced by iatrogenic subclinical hyperthyroidism.186 More aggressive TSH suppression is indicated in patients with high-risk disease or recurrent tumor, whereas less aggressive TSH suppression is reasonable in low-risk patients. Normalization of serum TSH is appropriate for long-term treatment of disease-free elderly patients with differentiated thyroid cancer and significant comorbidities.

Figure 30.3 illustrates these principles applied to a 76-year-old woman with papillary thyroid cancer who requires aggressive TSH suppression for presumed residual disease based upon a markedly elevated serum thyroglobulin (>500 ug/L) that could not be localized on extensive imaging including FDG PET/CT and was resistant to empirical radioiodine. Her fracture risk is relatively low based upon BMD measurements (10-year major osteoporotic fracture probability 10% and hip fracture probability 1.9% from the US FRAX tool version 3.4). However, VFA done at the time of BMD testing showed a moderate asymptomatic T12 vertebral compression fracture (Fig. 30.9[middle]), not reported on the prior CT scan but confirmed on review with benign features. Under the NOF and AACE guidelines, she would qualify for osteoporosis treatment (in addition to optimizing calcium and vitamin D) based upon the vertebral fracture.


More cancer patients are surviving into old age and are at risk for osteoporosis. Osteoporosis is also a consequence of some cancer treatments (e.g., AI for breast cancer, ADT for prostate cancer, and TSH suppression for differentiated thyroid cancer). Bisphosphonates, along with a healthy lifestyle and adequate intake of calcium and vitamin D, are currently the treatments of choice to prevent bone loss. Randomized clinical trials for some of these conditions show that targeted treatment can prevent the BMD loss and the accelerated bone turnover associated with oncologic therapy. Treatment initiation is based on an assessment of fracture risk that integrates CRFs and BMD levels obtained from DXA. These risk assessment systems are still evolving.

Research is needed to confirm that fracture risk assessment tools designed for the general population (e.g., the WHO FRAX tool) also accurately predict fracture risk in the oncology population. Because the increased fracture risk associated with AI and ADT is at least partially captured by alterations in BMD, it is uncertain whether these continue to independently affect fracture risk once BMD is known. Finally, the optimal frequency of BMD monitoring needs to be better defined.


1. Kanis JA, Melton LJ III, Christiansen C, et al. The diagnosis of osteoporosis. J Bone Miner Res. 1994;9(8):1137–1141.

2. Stone KL, Seeley DG, Lui LY, et al. BMD at multiple sites and risk of fracture of multiple types: Long-term results from the Study of Osteoporotic Fractures. J Bone Miner Res. 2003;18(11):1947–1954.

3. Johnell O, Kanis JA. An estimate of the worldwide prevalence and disability associated with osteoporotic fractures. Osteoporos Int. 2006;17(12):1726–1733.

4. Melton LJ III. Epidemiology worldwide. Endocrinol Metab Clin North Am. 2003;32(1):1–13.

5. Muss HB. Cancer in the elderly: A societal perspective from the United States. Clin Oncol (R Coll Radiol). 2009;21(2):92–98.

6. NIH Consensus Development Panel on Osteoporosis Prevention, Diagnosis, and Therapy. Osteoporosis prevention, diagnosis, and therapy. JAMA. 2001; 285(6):785–795.

7. Cummings SR, Bates D, Black DM. Clinical use of bone densitometry: Scientific review. JAMA. 2002;288(15):1889–1897.

8. Development Committee of the National Osteoporosis Foundation. Osteoporosis: Review of the evidence for prevention, diagnosis and treatment and cost-effectiveness analysis. Introduction. Osteoporos Int. 1998;8(suppl 4):S7–S80.

9. Brown JP, Josse RG. 2002 clinical practice guidelines for the diagnosis and management of osteoporosis in Canada. CMAJ. 2002;167(10 suppl):S1–S34.

10. Nelson HD, Helfand M, Woolf SH, et al. Screening for postmenopausal osteoporosis: A review of the evidence for the U.S. Preventive Services Task Force. Ann Intern Med. 2002;137(6):529–541.

11. Consensus development conference: Diagnosis, prophylaxis, and treatment of osteoporosis. Am J Med. 1993;94(6):646–650.

12. Binkley N, Bilezikian JP, Kendler DL, et al. Official positions of the International Society for Clinical Densitometry and Executive Summary of the 2005 Position Development Conference. J Clin Densitom. 2006;9(1):4–14.

13. Kanis JA, Borgstrom F, De Laet C, et al. Assessment of fracture risk. Osteoporos Int. 2005;16(6):581–589.

14. Looker AC, Wahner HW, Dunn WL, et al. Updated data on proximal femur bone mineral levels of US adults. Osteoporos Int. 1998;8(5):468–489.

15. Kanis JA, McCloskey EV, Johansson H, et al. A reference standard for the description of osteoporosis. Bone. 2008;42(3):467–475.

16. Wiktorowicz ME, Goeree R, Papaioannou A, et al. Economic implications of hip fracture: Health service use, institutional care and cost in Canada. Osteoporos Int. 2001;12(4):271–278.

17. Papaioannou A, Adachi JD, Parkinson W, et al. Lengthy hospitalization associated with veretebral fractures despite control for comorbid conditions. Osteoporos Int. 2001;12(10):870–874.

18. Center JR, Nguyen TV, Schneider D, et al. Mortality after all major types of osteoporotic fracture in men and women: An observational study. Lancet. 1999; 353(9156):878–882.

19. Johnell O, Kanis JA, Oden A, et al. Mortality after osteoporotic fractures. Osteoporos Int. 2004;15(1):38–42.

20. Ioannidis G, Papaioannou A, Hopman WM, et al. Relation between fractures and mortality: Results from the Canadian Multicentre Osteoporosis Study. CMAJ. 2009;181(5):265–271.

21. Kanis JA, Oden A, Johnell O, et al. The components of excess mortality after hip fracture. Bone. 2003;32(5):468–473.

22. Kanis JA, Oden A, Johnell O, et al. Excess mortality after hospitalisation for vertebral fracture. Osteoporos Int. 2004;15(2):108–112.

23. Adachi JD, Ioannidis G, Berger C, et al. The influence of osteoporotic fractures on health-related quality of life in community-dwelling men and women across Canada. Osteoporos Int. 2001;12(11):903–908.

24. Hallberg I, Rosenqvist AM, Kartous L, et al. Health-related quality of life after osteoporotic fractures. Osteoporos Int. 2004;15(10):834–841.

25. De Laet CE, van Hout BA, Burger H, et al. Incremental cost of medical care after hip fracture and first vertebral fracture: The Rotterdam study 1. Osteoporos Int. 1999;10(1):66–72.

26. Pike C, Birnbaum HG, Schiller M, et al. Direct and indirect costs of non-vertebral fracture patients with osteoporosis in the US. Pharmacoeconomics. 2010;28(5):395–409.

27. Shi N, Foley K, Lenhart G, et al. Direct healthcare costs of hip, vertebral, and non-hip, non-vertebral fractures. Bone. 2009;45(6):1084–1090.

28. Leslie WD, Metge CJ, Azimaee M, et al. Direct costs of fractures in Canada and trends 1996-2006: A population-based cost-of-illness analysis. J Bone Miner Res. 2011;26(10):2419–2429.

29. Leibbrandt A, Penninger JM. RANK(L) as a key target for controlling bone loss. Adv Exp Med Biol. 2009;647:130–145.

30. Duncan EL, Brown MA. Clinical review 2: Genetic determinants of bone density and fracture risk–state of the art and future directions. J Clin Endocrinol Metab. 2010;95(6):2576–2587.

31. Zhai G, Andrew T, Kato BS, et al. Genetic and environmental determinants on bone loss in postmenopausal Caucasian women: A 14-year longitudinal twin study. Osteoporos Int. 2009;20(6):949–953.

32. Barrett-Connor E, Siris ES, Wehren LE, et al. Osteoporosis and fracture risk in women of different ethnic groups. J Bone Miner Res. 2005;20(2):185–194.

33. Berger C, Langsetmo L, Joseph L, et al. Change in bone mineral density as a function of age in women and men and association with the use of antiresorptive agents. CMAJ. 2008;178(13):1660–1668.

34. Blake GM, Fogelman I. How important are BMD accuracy errors for the clinical interpretation of DXA scans? J Bone Miner Res. 2008;23(4):457–462.

35. Bolotin HH, Sievanen H, Grashuis JL. Patient-specific DXA bone mineral density inaccuracies: Quantitative effects of nonuniform extraosseous fat distributions. J Bone Miner Res. 2003;18(6):1020–1027.

36. Alekel DL, Peterson CT, Werner RK, et al. Frame size, ethnicity, lifestyle, and biologic contributors to areal and volumetric lumbar spine bone mineral density in Indian/Pakistani and American Caucasian premenopausal women. J Clin Densitom. 2002;5(2):175–186.

37. Bhudhikanok GS, Wang MC, Eckert K, et al. Differences in bone mineral in young Asian and Caucasian Americans may reflect differences in bone size. J Bone Miner Res. 1996;11(10):1545–1556.

38. Looker AC, Beck TJ, Orwoll ES. Does body size account for gender differences in femur bone density and geometry? J Bone Miner Res. 2001;16(7):1291–1299.

39. Melton LJ III, Khosla S, Achenbach SJ, et al. Effects of body size and skeletal site on the estimated prevalence of osteoporosis in women and men. Osteoporos Int. 2000;11(11):977–983.

40. Garnero P, Sornay-Rendu E, Chapuy MC, et al. Increased bone turnover in late postmenopausal women is a major determinant of osteoporosis. J Bone Miner Res. 1996;11(3):337–349.

41. Seeman E. Growth in bone mass and size–are racial and gender differences in bone mineral density more apparent than real? J Clin Endocrinol Metab. 1998;83(5):1414–1419.

42. Cauley JA, Zmuda JM, Wisniewski SR, et al. Bone mineral density and prevalent vertebral fractures in men and women. Osteoporos Int. 2004;15(1):32–37.

43. Benbassat CA, Eshed V, Kamjin M, et al. Are adult patients with Laron syndrome osteopenic? A comparison between dual-energy X-ray absorptiometry and volumetric bone densities. J Clin Endocrinol Metab. 2003;88(10):4586–4589.

44. Neely EK, Marcus R, Rosenfeld RG, et al. Turner syndrome adolescents receiving growth hormone are not osteopenic. J Clin Endocrinol Metab. 1993;76(4): 861–866.

45. Leslie WD, Tsang JF, Lix LM. The effect of total hip bone area on osteoporosis diagnosis and fractures. J Bone Miner Res. 2008.

46. Carter DR, Bouxsein ML, Marcus R. New approaches for interpreting projected bone densitometry data. J Bone Miner Res. 1992;7(2):137–145.

47. Nevitt MC, Ettinger B, Black DM, et al. The association of radiographically detected vertebral fractures with back pain and function: A prospective study. Ann Intern Med. 1998;128(10):793–800.

48. Lentle BC, Brown JP, Khan A, et al. Recognizing and reporting vertebral fractures: Reducing the risk of future osteoporotic fractures. Can Assoc Radiol J. 2007;58(1):27–36.

49. Vokes T, Bachman D, Baim S, et al. Vertebral fracture assessment: The 2005 ISCD Official Positions. J Clin Densitom. 2006;9(1):37–46.

50. McCloskey EV, Vasireddy S, Threlkeld J, et al. Vertebral fracture assessment (VFA) with a densitometer predicts future fractures in elderly women unselected for osteoporosis. J Bone Miner Res. 2008;23(10):1561–1568.

51. Lewiecki EM, Gordon CM, Baim S, et al. International Society for Clinical Densitometry 2007 Adult and Pediatric Official Positions. Bone. 2008;43(6): 1115–1121.

52. Schousboe JT, Wilson KE, Kiel DP. Detection of abdominal aortic calcification with lateral spine imaging using DXA. J Clin Densitom. 2006;9(3):302–308.

53. Schousboe JT, Taylor BC, Kiel DP, et al. Abdominal aortic calcification detected on lateral spine images from a bone densitometer predicts incident myocardial infarction or stroke in older women. J Bone Miner Res. 2008;23(3):409–416.

54. Lewiecki EM, Watts NB, McClung MR, et al. Official positions of the international society for clinical densitometry. J Clin Endocrinol Metab. 2004;89(8): 3651–3655.

55. Lenchik L, Kiebzak GM, Blunt BA. What is the role of serial bone mineral density measurements in patient management? J Clin Densitom. 2002;5(suppl): S29–S38.

56. Gluer CC, Blake G, Lu Y, et al. Accurate assessment of precision errors: How to measure the reproducibility of bone densitometry techniques. Osteoporos Int. 1995;5(4):262–270.

57. Gluer CC. Monitoring skeletal changes by radiological techniques. J Bone Miner Res. 1999;14(11):1952–1962.

58. Baim S, Wilson CR, Lewiecki EM, et al. Precision assessment and radiation safety for dual-energy X-ray absorptiometry: Position paper of the International Society for Clinical Densitometry. J Clin Densitom. 2005;8(4):371–378.

59. Bonnick SL, Johnston CC Jr, Kleerekoper M, et al. Importance of precision in bone density measurements. J Clin Densitom. 2001;4(2):105–110.

60. Fuleihan GE, Testa MA, Angell JE, et al. Reproducibility of DXA absorptiometry: A model for bone loss estimates. J Bone Miner Res. 1995;10(7):1004–1014.

61. Leslie WD. Factors affecting short-term bone density precision assessment and the effect on patient monitoring. J Bone Miner Res. 2008;23(2):199–204.

62. Patel R, Blake GM, Rymer J, et al. Long-term precision of DXA scanning assessed over seven years in forty postmenopausal women. Osteoporos Int. 2000;11(1):68–75.

63. Binkley N, Krueger D, Vallarta-Ast N. An overlying fat panniculus affects femur bone mass measurement 1. J Clin Densitom. 2003;6(3):199–204.

64. Blank RD, Malone DG, Christian RC, et al. Patient variables impact lumbar spine dual energy X-ray absorptiometry precision. Osteoporos Int. 2006;17(5): 768–774.

65. Slosman D, Provvedini D, Meunier P, et al. The use of different dual x-ray absorptiometry brands in a multicenter clinical trial. J Clin Densitom. 1998; 2(1):37–44.

66. Leslie WD, Manitoba Bone Density Program. The importance of spectrum bias on bone density monitoring in clinical practice. Bone. 2006;39(2):361–368.

67. Shepherd JA, Lu Y, Wilson K, et al. Cross-calibration and minimum precision standards for dual-energy X-ray absorptiometry: The 2005 ISCD Official Positions. J Clin Densitom. 2006;9(1):31–36.

68. Marshall D, Johnell O, Wedel H. Meta-analysis of how well measures of bone mineral density predict occurrence of osteoporotic fractures. BMJ. 1996;312(7041): 1254–1259.

69. Leslie WD, Adler RA, El-Hajj FG, et al. Application of the 1994 WHO classification to populations other than postmenopausal Caucasian women: The 2005 ISCD Official Positions. J Clin Densitom. 2006;9(1):22–30.

70. De Laet CE, Van Hout BA, Burger H, et al. Hip fracture prediction in elderly men and women: Validation in the Rotterdam study. J Bone Miner Res. 1998; 13(10):1587–1593.

71. Langsetmo L, Leslie WD, Zhou W, et al. Using the same bone density reference database for men and women provides a simpler estimation of fracture risk. J Bone Miner Res. 2010;25(10):2108–2114.

72. Khosla S, Amin S, Orwoll E. Osteoporosis in men. Endocr Rev. 2008;29(4):441–464.

73. Hui SL, Slemenda CW, Johnston CC Jr. Age and bone mass as predictors of fracture in a prospective study. J Clin Invest. 1988;81(6):1804–1809.

74. Zemel BS, Leonard MB, Kelly A, et al. Height adjustment in assessing dual energy x-ray absorptiometry measurements of bone mass and density in children. J Clin Endocrinol Metab. 2010;95(3):1265–1273.

75. Zemel BS, Kalkwarf HJ, Gilsanz V, et al. Revised reference curves for bone mineral content and areal bone mineral density according to age and sex for black and non-black children: Results of the bone mineral density in childhood study. J Clin Endocrinol Metab.2011;96(10):3160–3169.

76. Bishop N, Braillon P, Burnham J, et al. Dual-energy X-ray aborptiometry assessment in children and adolescents with diseases that may affect the skeleton: The 2007 ISCD Pediatric Official Positions. J Clin Densitom. 2008; 11(1):29–42.

77. Baim S, Leonard MB, Bianchi ML, et al. Official Positions of the International Society for Clinical Densitometry and executive summary of the 2007 ISCD Pediatric Position Development Conference. J Clin Densitom. 2008;11(1):6–21.

78. Rauch F, Plotkin H, DiMeglio L, et al. Fracture prediction and the definition of osteoporosis in children and adolescents: The ISCD 2007 Pediatric Official Positions. J Clin Densitom. 2008;11(1):22–28.

79. Gordon CM, Bachrach LK, Carpenter TO, et al. Dual energy X-ray absorptiometry interpretation and reporting in children and adolescents: The 2007 ISCD Pediatric Official Positions. J Clin Densitom. 2008;11(1):43–58.

80. Singh R, Shaw J, Zimmet P. Epidemiology of childhood type 2 diabetes in the developing world. Pediatr Diabetes. 2004;5(3):154–168.

81. Wild S, Roglic G, Green A, et al. Global prevalence of diabetes: Estimates for the year 2000 and projections for 2030. Diabetes Care. 2004;27(5):1047–1053.

82. Litwin SE. Which measures of obesity best predict cardiovascular risk? J Am Coll Cardiol. 2008;52(8):616–619.

83. Storti KL, Brach JS, FitzGerald SJ, et al. Relationships among body composition measures in community-dwelling older women. Obesity (Silver Spring). 2006;14(2):244–251.

84. Leslie WD, Weiler HA, Nyomba BL. Ethnic differences in adiposity and body composition: The First Nations bone health study. Appl Physiol Nutr Metab. 2007;32(6):1065–1072.

85. von Eyben FE, Mouritsen E, Holm J, et al. Intra-abdominal obesity and metabolic risk factors: A study of young adults. Int J Obes Relat Metab Disord. 2003;27(8):941–949.

86. Jensen MD. Role of body fat distribution and the metabolic complications of obesity. J Clin Endocrinol Metab. 2008;93(11 suppl 1):S57–S63.

87. Kanaya AM, Vittinghoff E, Shlipak MG, et al. Association of total and central obesity with mortality in postmenopausal women with coronary heart disease. Am J Epidemiol. 2003;158(12):1161–1170.

88. Albanese CV, Diessel E, Genant HK. Clinical applications of body composition measurements using DXA. J Clin Densitom. 2003;6(2):75–85.

89. Svendsen OL, Hassager C, Bergmann I, et al. Measurement of abdominal and intra-abdominal fat in postmenopausal women by dual energy X-ray absorptiometry and anthropometry: Comparison with computerized tomography. Int J Obes Relat Metab Disord. 1993;17(1):45–51.

90. Krakauer JC, Franklin B, Kleerekoper M, et al. Body composition profiles derived from dual-energy X-ray absorptiometry, total body scan, and mortality. Prev Cardiol. 2004;7(3):109–115.

91. Leslie WD. Prediction of body composition from spine and hip bone densitometry. J Clin Densitom. 2009;12(4):428–433.

92. Leslie WD, Ludwig SM, Morin S. Abdominal fat from spine dual-energy x-ray absorptiometry and risk for subsequent diabetes. J Clin Endocrinol Metab. 2010;95(7):3272–3276.

93. van Londen GJ, Perera S, Vujevich K, et al. The impact of an aromatase inhibitor on body composition and gonadal hormone levels in women with breast cancer. Breast Cancer Res Treat. 2011;125(2):441–446.

94. La VC, Giordano SH, Hortobagyi GN, et al. Overweight, obesity, diabetes, and risk of breast cancer: Interlocking pieces of the puzzle. Oncologist. 2011; 16(6):726–729.

95. Peairs KS, Barone BB, Snyder CF, et al. Diabetes mellitus and breast cancer outcomes: A systematic review and meta-analysis. J Clin Oncol. 2011;29(1):40–46.

96. Haseen F, Murray LJ, Cardwell CR, et al. The effect of androgen deprivation therapy on body composition in men with prostate cancer: Systematic review and meta-analysis. J Cancer Surviv. 2010;4(2):128–139.

97. Smith MR. Androgen deprivation therapy and risk for diabetes and cardiovascular disease in prostate cancer survivors. Curr Urol Rep. 2008;9(3):197–202.

98. Kanis JA, Oden A, Johnell O, et al. The burden of osteoporotic fractures: A method for setting intervention thresholds. Osteoporos Int. 2001;12(5):417-427.

99. Cranney A, Jamal SA, Tsang JF, et al. Low bone mineral density and fracture burden in postmenopausal women. CMAJ. 2007;177(6):575–580.

100. Kanis JA, Oden A, Johnell O, et al. The use of clinical risk factors enhances the performance of BMD in the prediction of hip and osteoporotic fractures in men and women. Osteoporos Int. 2007;18(8):1033–1046.

101. Kanis JA, Oden A, Johansson H, et al. FRAX and its applications to clinical practice. Bone. 2009;44(5):734–743.

102. Kanis JA, McCloskey EV, Johansson H, et al. How to decide who to treat. Best Pract Res Clin Rheumatol. 2009;23(6):711–726.

103. Nguyen ND, Frost SA, Center JR, et al. Development of a nomogram for individualizing hip fracture risk in men and women. Osteoporos Int. 2007;18(8): 1109–1117.

104. Nguyen ND, Frost SA, Center JR, et al. Development of prognostic nomograms for individualizing 5-year and 10-year fracture risks. Osteoporos Int. 2008; 19(10):1431–1444.

105. Hippisley-Cox J, Coupland C. Predicting risk of osteoporotic fracture in men and women in England and Wales: Prospective derivation and validation of QFractureScores. BMJ. 2009;339:b4229.

106. Dawson-Hughes B. A revised clinician’s guide to the prevention and treatment of osteoporosis. J Clin Endocrinol Metab. 2008;93(7):2463–2465.

107. Dawson-Hughes B, Tosteson AN, Melton LJ III, et al. Implications of absolute fracture risk assessment for osteoporosis practice guidelines in the USA. Osteoporos Int. 2008;19(4):449–458.

108. Kanis JA, Johnell O, Oden A, et al. FRAX and the assessment of fracture probability in men and women from the UK. Osteoporos Int. 2008;19(4):385–397.

109. Kanis JA, McCloskey EV, Johansson H, et al. Case finding for the management of osteoporosis with FRAX((R))-assessment and intervention thresholds for the UK. Osteoporos Int. 2008;19(10):1395–1408.

110. Lippuner K, Johansson H, Kanis JA, et al. FRAX assessment of osteoporotic fracture probability in Switzerland. Osteoporos Int. 2010;21(3):381–389.

111. Kanis JA, Burlet N, Cooper C, et al. European guidance for the diagnosis and management of osteoporosis in postmenopausal women. Osteoporos Int. 2008; 19(4):399–428.

112. Fujiwara S, Kasagi F, Masunari N, et al. Fracture prediction from bone mineral density in Japanese men and women. J Bone Miner Res. 2003;18(8):1547–1553.

113. Neuprez A, Johansson H, Kanis JA, et al. [A FRAX model for the assessment of fracture probability in Belgium]. Rev Med Liege. 2009;64(12):612–619.

114. Papaioannou A, Morin S, Cheung AM, et al. 2010 clinical practice guidelines for the diagnosis and management of osteoporosis in Canada: Summary. CMAJ. 2010;182(17):1864–1873.

115. Kanis JA, Hans D, Cooper C, et al. Interpretation and use of FRAX in clinical practice. Osteoporos Int. 2011;22(9):2395–2411.

116. Blake GM, Knapp KM, Spector TD, et al. Predicting the risk of fracture at any site in the skeleton: Are all bone mineral density measurement sites equally effective? Calcif Tissue Int. 2006;78(1):9–17.

117. Leslie WD, Tsang JF, Caetano PA, et al. Number of osteoporotic sites and fracture risk assessment: A cohort study from the Manitoba Bone Density Program. J Bone Miner Res. 2007;22(3):476–483.

118. Leslie WD, Lix LM, Tsang JF, et al. Single-site vs multisite bone density measurement for fracture prediction. Arch Intern Med. 2007;167(15):1641–1647.

119. Fink HA, Harrison SL, Taylor BC, et al. Differences in site-specific fracture risk among older women with discordant results for osteoporosis at hip and spine: Study of osteoporotic fractures. J Clin Densitom. 2008;11(2):250–259.

120. Leslie WD, Lix LM, Johansson H, et al. Spine-hip discordance and fracture risk assessment: A physician-friendly FRAX enhancement. Osteoporos Int. 2011; 22(3):839–847.

121. Brown JP, Albert C, Nassar BA, et al. Bone turnover markers in the management of postmenopausal osteoporosis. Clin Biochem. 2009;42(10–11):929–942.

122. Garnero P, Sornay-Rendu E, Claustrat B, et al. Biochemical markers of bone turnover, endogenous hormones and the risk of fractures in postmenopausal women: The OFELY study. J Bone Miner Res. 2000;15(8):1526–1536.

123. Institute of Medicine (US) Committee to Review Dietary Reference Intakes for Vitamin D and Calcium. In: Ross AC, Taylor CL, Yaktine AL, Del Valle HB, eds. Dietary Reference Intakes for Calcium and Vitamin D. Washington (DC): National Academies Press (US); 2011.

124. MacLean C, Newberry S, Maglione M, et al. Systematic review: Comparative effectiveness of treatments to prevent fractures in men and women with low bone density or osteoporosis. Ann Intern Med. 2008;148(3):197–213.

125. Silverman S. Adherence to medications for the treatment of osteoporosis. Rheum Dis Clin North Am. 2006;32(4):721–731.

126. McCombs JS, Thiebaud P, Laughlin-Miley C, et al. Compliance with drug therapies for the treatment and prevention of osteoporosis. Maturitas. 2004; 48(3):271–287.

127. Gold DT, Silverman S. Review of adherence to medications for the treatment of osteoporosis. Curr Osteoporos Rep. 2006;4(1):21–27.

128. Wells GA, Cranney A, Peterson J, et al. Alendronate for the primary and secondary prevention of osteoporotic fractures in postmenopausal women. Cochrane Database Syst Rev. 2008;(1):CD001155.

129. Wells G, Cranney A, Peterson J, et al. Risedronate for the primary and secondary prevention of osteoporotic fractures in postmenopausal women. Cochrane Database Syst Rev. 2008;(1):CD004523.

130. National Institute for Health and Clinical Excellence. Osteoporosis: Assessment of fracture risk and the prevention of osteoporotic fracture in individuals at high risk. National Collaborating Centre for Nursing and Supportive Care. 2008.

131. Ensrud KE, Stock JL, Barrett-Connor E, et al. Effects of raloxifene on fracture risk in postmenopausal women: The raloxifene use for the heart trial. J Bone Miner Res. 2008;23(1):112–120.

132. Overman RA, Borse M, Gourlay ML. Salmon calcitonin use and associated cancer risk. Ann Pharmacother. 2013;47(12):1675–1684.

133. Cummings SR, Martin SJ, McClung MR, et al. Denosumab for prevention of fractures in postmenopausal women with osteoporosis. N Engl J Med. 2009; 361:756–765.

134. Solomon DH, Rekedal L, Cadarette SM. Osteoporosis treatments and adverse events. Curr Opin Rheumatol. 2009;21(4):363–368.

135. Rizzoli R, Reginster JY, Boonen S, et al. Adverse reactions and drug-drug interactions in the management of women with postmenopausal osteoporosis. Calcif Tissue Int. 2011;89(2):91–104.

136. Shane E, Burr D, Abrahamsen B, et al. Atypical subtrochanteric and diaphyseal femoral fractures: Second report of a task force of the American Society for Bone and Mineral Research. J Bone Miner Res. 2014;29:1–23.

137. Park-Wyllie LY, Mamdani MM, Juurlink DN, et al. Bisphosphonate use and the risk of subtrochanteric or femoral shaft fractures in older women. JAMA. 2011;305(8):783–789.

138. Leslie WD, Schousboe JT. A review of osteoporosis diagnosis and treatment options in new and recently updated guidelines on case finding around the world. Curr Osteoporos Rep. 2011;9(3):129–140.

139. Watts NB, Bilezikian JP, Camacho PM, et al. American Association of Clinical Endocrinologists Medical Guidelines for Clinical Practice for the diagnosis and treatment of postmenopausal osteoporosis. Endocr Pract. 2010;16(suppl 3):1–37.

140. Dawson-Hughes B, Looker AC, Tosteson AN, et al. The potential impact of new National Osteoporosis Foundation guidance on treatment patterns. Osteoporos Int. 2010;21(1):41–52.

141. Grossman JM, Gordon R, Ranganath VK, et al. American College of Rheumatology 2010 recommendations for the prevention and treatment of glucocorticoid-induced osteoporosis. Arthritis Care Res (Hoboken). 2010;62(11): 1515–1526.

142. Chen P, Krege JH, Adachi JD, et al. Vertebral fracture status and the World Health Organization risk factors for predicting osteoporotic fracture risk. J Bone Miner Res. 2009;24(3):495–502.

143. Lewiecki EM, Watts NB. Assessing response to osteoporosis therapy. Osteoporos Int. 2008;19(10):1363–1368.

144. Siminoski K, Leslie WD, Frame H, et al. Recommendations for bone mineral density reporting in Canada. Can Assoc Radiol J. 2005;56(3):178–188.

145. Frost SA, Nguyen ND, Center JR, et al. Timing of repeat BMD measurements: Development of an absolute risk-based prognostic model. J Bone Miner Res. 2009;24(11):1800–1807.

146. Bray F, McCarron P, Parkin DM. The changing global patterns of female breast cancer incidence and mortality. Breast Cancer Res. 2004;6(6):229–239.

147. Ramaswamy B, Shapiro CL. Osteopenia and osteoporosis in women with breast cancer. Semin Oncol. 2003;30(6):763–775.

148. Chen Z, Maricic M, Bassford TL, et al. Fracture risk among breast cancer survivors: Results from the Women’s Health Initiative Observational Study. Arch Intern Med. 2005;165(5):552–558.

149. Reid DM, Doughty J, Eastell R, et al. Guidance for the management of breast cancer treatment-induced bone loss: A consensus position statement from a UK Expert Group. Cancer Treat Rev. 2008;34(suppl 1):S3–S18.

150. Eastell R, Adams JE, Coleman RE, et al. Effect of anastrozole on bone mineral density: 5-year results from the anastrozole, tamoxifen, alone or in combination trial 18233230. J Clin Oncol. 2008;26(7):1051–1057.

151. Eastell R, Hannon RA, Cuzick J, et al. Effect of an aromatase inhibitor on bmd and bone turnover markers: 2-year results of the Anastrozole, Tamoxifen, Alone or in Combination (ATAC) trial (18233230). J Bone Miner Res. 2006;21(8): 1215–1223.

152. Baum M, Buzdar A, Cuzick J, et al. Anastrozole alone or in combination with tamoxifen versus tamoxifen alone for adjuvant treatment of postmenopausal women with early-stage breast cancer: Results of the ATAC (Arimidex, Tamoxifen Alone or in Combination) trial efficacy and safety update analyses. Cancer. 2003;98(9):1802–1810.

153. Coates AS, Keshaviah A, Thurlimann B, et al. Five years of letrozole compared with tamoxifen as initial adjuvant therapy for postmenopausal women with endocrine-responsive early breast cancer: Update of study BIG 1-98. J Clin Oncol. 2007;25(5):486–492.

154. Coombes RC, Kilburn LS, Snowdon CF, et al. Survival and safety of exemestane versus tamoxifen after 2-3 years’ tamoxifen treatment (Intergroup Exemestane Study): A randomised controlled trial. Lancet. 2007;369(9561):559–570.

155. Jakesz R, Jonat W, Gnant M, et al. Switching of postmenopausal women with endocrine-responsive early breast cancer to anastrozole after 2 years’ adjuvant tamoxifen: Combined results of ABCSG trial 8 and ARNO 95 trial. Lancet. 2005;366(9484):455–462.

156. Ding H, Field TS. Bone health in postmenopausal women with early breast cancer: How protective is tamoxifen? Cancer Treat Rev. 2007;33(6):506–513.

157. Cooke AL, Metge C, Lix L, et al. Tamoxifen use and osteoporotic fracture risk: A population-based analysis. J Clin Oncol. 2008;26(32):5227–5232.

158. Goss PE, Ingle JN, Martino S, et al. Randomized trial of letrozole following tamoxifen as extended adjuvant therapy in receptor-positive breast cancer: Updated findings from NCIC CTG MA.17. J Natl Cancer Inst. 2005;97(17): 1262–1271.

159. Neuner JM, Yen TW, Sparapani RA, et al. Fracture risk and adjuvant hormonal therapy among a population-based cohort of older female breast cancer patients. Osteoporos Int. 2011;22(11):2847–2855.

160. Theriault RL. Bisphosphonates: Ready for use as adjuvant therapy of breast cancer? Curr Opin Obstet Gynecol. 2010;22(1):61–66.

161. Hines SL, Sloan JA, Atherton PJ, et al. Zoledronic acid for treatment of osteopenia and osteoporosis in women with primary breast cancer undergoing adjuvant aromatase inhibitor therapy. Breast. 2010;19(2):92–96.

162. Brufsky AM, Bosserman LD, Caradonna RR, et al. Zoledronic acid effectively prevents aromatase inhibitor-associated bone loss in postmenopausal women with early breast cancer receiving adjuvant letrozole: Z-FAST study 36-month follow-up results. Clin Breast Cancer.2009;9(2):77–85.

163. Gnant M, Mlineritsch B, Schippinger W, et al. Endocrine therapy plus zoledronic acid in premenopausal breast cancer. N Engl J Med. 2009;360(7):679–691.

164. Van Poznak C, Hannon RA, Mackey JR, et al. Prevention of aromatase inhibitor-induced bone loss using risedronate: The SABRE trial. J Clin Oncol. 2010;28(6): 967–975.

165. Ellis GK, Bone HG, Chlebowski R, et al. Randomized trial of denosumab in patients receiving adjuvant aromatase inhibitors for nonmetastatic breast cancer. J Clin Oncol. 2008;26(30):4875–4882.

166. Cummings SR, Duong T, Kenyon E, et al. Serum estradiol level and risk of breast cancer during treatment with raloxifene. JAMA. 2002;287(2):216–220.

167. Cauley JA, Gutai JP, Sandler RB, et al. The relationship of endogenous estrogen to bone density and bone area in normal postmenopausal women. Am J Epidemiol. 1986;124(5):752–761.

168. Zhang Y, Kiel DP, Kreger BE, et al. Bone mass and the risk of breast cancer among postmenopausal women. N Engl J Med. 1997;336(9):611–617.

169. Chen Z, Arendell L, Aickin M, et al. Hip bone density predicts breast cancer risk independently of Gail score: Results from the Women’s Health Initiative. Cancer. 2008;113(5):907–915.

170. Grenier D, Cooke AL, Lix L, et al. Bone mineral density and risk of postmenopausal breast cancer. Breast Cancer Res Treat. 2011;126(3):679–686.

171. Edwards BK, Brown ML, Wingo PA, et al. Annual report to the nation on the status of cancer, 1975–2002, featuring population-based trends in cancer treatment. J Natl Cancer Inst. 2005;97(19):1407–1427.

172. Lee CE, Leslie WD, Czaykowski P, et al. A comprehensive bone-health management approach for men with prostate cancer receiving androgen deprivation therapy. Curr Oncol. 2011;18(4):e163–e172.

173. Higano CS. Androgen-deprivation-therapy-induced fractures in men with nonmetastatic prostate cancer: What do we really know? Nat Clin Pract Urol. 2008;5(1):24–34.

174. Shahinian VB, Kuo YF, Freeman JL, et al. Risk of fracture after androgen deprivation for prostate cancer. N Engl J Med. 2005;352(2):154–164.

175. Lau YK, Lee E, Prior HJ, et al. Fracture risk in androgen deprivation therapy: A Canadian population based analysis. Can J Urol. 2009;16(6):4908–4914.

176. Alibhai SM, Duong-Hua M, Cheung AM, et al. Fracture types and risk factors in men with prostate cancer on androgen deprivation therapy: A matched cohort study of 19,079 men. J Urol. 2010;184(3):918–923.

177. Smith MR, Egerdie B, Hernandez TN, et al. Denosumab in men receiving androgen-deprivation therapy for prostate cancer. N Engl J Med. 2009; 361(8):745–755.

178. Ito K, Elkin EB, Girotra M, et al. Cost-effectiveness of fracture prevention in men who receive androgen deprivation therapy for localized prostate cancer. Ann Intern Med. 2010;152(10):621–629.

179. American Thyroid Association (ATA) Guidelines Taskforce on Thyroid Nodules and Differentiated Thyroid Cancer, Cooper DS, Doherty GM, et al. Revised American Thyroid Association management guidelines for patients with thyroid nodules and differentiated thyroid cancer. Thyroid.2009;19(11):1167–1214.

180. Bauer DC, Nevitt MC, Ettinger B, et al. Low thyrotropin levels are not associated with bone loss in older women: A prospective study. J Clin Endocrinol Metab. 1997;82(9):2931–2936.

181. Cummings SR, Nevitt MC, Browner WS, et al. Risk factors for hip fracture in white women. Study of Osteoporotic Fractures Research Group. N Engl J Med. 1995;332(12):767–773.

182. Faber J, Galloe AM. Changes in bone mass during prolonged subclinical hyperthyroidism due to L-thyroxine treatment: A meta-analysis. Eur J Endocrinol. 1994;130(4):350–356.

183. Bauer DC, Ettinger B, Nevitt MC, et al. Risk for fracture in women with low serum levels of thyroid-stimulating hormone. Ann Intern Med. 2001; 134(7):561–568.

184. Lee JS, Buzkova P, Fink HA, et al. Subclinical thyroid dysfunction and incident hip fracture in older adults. Arch Intern Med. 2010;170(21):1876–1883.

185. Vadiveloo T, Donnan PT, Cochrane L, et al. The Thyroid Epidemiology, Audit, and Research Study (TEARS): Morbidity in patients with endogenous subclinical hyperthyroidism. J Clin Endocrinol Metab. 2011;96(5):1344–1351.

186. Biondi B, Cooper DS. Benefits of thyrotropin suppression versus the risks of adverse effects in differentiated thyroid cancer. Thyroid. 2010;20(2):135–146.