Review Article Osteoporosis - Springer Link

Osteoporos Int (1998) 8:299–310 ß 1998 European Foundation for Osteoporosis and the National Osteoporosis Foundation

Osteoporosis International

Review Article Quantitative Bone Mineral Assessment at the Forearm: A Review P. Augat, T. Fuerst and H. K. Genant Osteoporosis and Arthritis Research Group, Department of Radiology, University of California San Francisco, CA, USA

Abstract. Bone mineral density and geometric properties of the human forearm can be measured to determine the amount of bone or bone loss at the scanning site and to predict the risk of forearm fractures. These forearm mesurements are also used to estimate bone mass at remote anatomical locations and thereby estimate the risk for spine, hip and other fractures. The peripheral location of the human forearm, with its relatively small amount of surrounding soft tissue, improves the accuracy and the precision of bone mass measurement and has made this site an early choice for the assessment of a subject’s bone mineral status. Furthermore, the anatomy of the human radius enables the examination of both cortical and cancellous bone. This review describes the procedures for non-invasive bone assessment at peripheral sites including some of the more recently developed systems dedicated to assessment of the distal radius. The accuracy, precision and normative values they provide are presented. Responses to different forms of therapies as well as the ability to discriminate or predict osteoporotic fractures are also assessed. Low radiation dose, comfortable and fast handling, moderate cost, and a strong association with the risk of non-spine fractures, promote the use of forearm scanning as a widely applied screening procedure for the detection of generalised osteoporotic bone loss. However, a higher accuracy of fracture risk prediction at the spine or at the hip can be achieved by a direct bone density measurement at these sites. The monitoring of treatment at the distal forearm appears to require a longer follow-up time due to its

Correspondence and offprint requests to: Peter Augat, PhD, Department of Orthopaedic Research and Biomechanics, University of Ulm, Helmholtzstrasse 14, D-89081 Ulm, Germany. Tel: +49 (731) 502 3496. Fax: +49 (731) 502 3498. e-mail: [email protected]. uni-ulm.de.

decreased responsiveness compared with such highly trabecular load-bearing sites as the spine and the proximal femur. Keywords: Bone mineral density; Distal radius; Dual Xray absorptiometry (DXA); Osteoporosis; Peripheral quantitative computed tomography (pQCT)

Introduction Bone mineral densities and geometric properties of the forearm can be measured to determine peak bone mass, the degree of disease, or age-related bone loss. The purpose of these measurements is not only to determine the amount of bone or bone loss at the scanning site and to predict the risk of forearm fractures but also to estimate bone mass at other appendicular sites and at the axial skeleton. The peripheral location of the forearm and the relatively small amount of surrounding soft tissue have made this site an obvious early choice for the assessment of a subject’s bone mineral. The limited amount of surrounding tissue increases the accuracy and the precision of bone mass measurement due to reduced beam hardening and negligible soft tissue correction. The peripheral scanning site reduces the radiation dose to the gonads and makes equipment requirements simpler and less expensive. The anatomy of the radius – thin cortex with mainly cancellous bone at the distal end and pure cortical bone along the diaphysis – enables the examination of both cortical and cancellous bone. The metabolic response is much faster in cancellous bone and therapeutic effects as well as postmenopausal bone loss can be detected earlier in regions of cancellous bone. However, appendicular cancellous bone contains more fatty marrow than does the axial skeleton with its large amount of haematopoietic

300

P. Augat et al.

marrow. Although the metabolic activity of appendicular bone is still higher than in cortical bone the difference might not be as large as in the axial skeleton [1]. Cortical bone, on the other hand, contributes substantially to the strength of long bones. The quantification of the cortical bone mass and its architectural structure adds information for fracture risk assessment. Quantitative assessment of bone mineral status at the forearm has been used to screen for osteoporosis [2,3], and study age-related changes [4–6], the impact of hormone replacement therapy [7,8], rheumatoid arthritis [9,10], renal osteodystrophy [11] and hyperparathyroidism [12,13]. The effects of exercise and different levels of activity have been investigated successfully [14,15]. Children have been scanned at the forearm to verify bone mineral disorders [16,17] or to investigate age, sex and body size effects [18–20]. A variety of non-invasive scanning procedures have been applied for measurements at peripheral skeletal sites during the past several decades. Some of the more recent systems have been especially developed for the assessment of the distal radius. These procedures are described and the normative values they provide are presented. We further assessed the relation of appendicular measurements to osteoporosis-related fractures as well as their ability as regards fracture prediction.

needed. Hydroxyapatite (Ca10(PO4)6(OH)2), dipotassium hydrogen phosphate (K2HPO4), calcium chloride (CaCl2) and aluminium are widely used as bone mineral equivalent materials [21]. Measurements that are based on projected images (SPA, SXA, DPA, DXA) compute the projected bone mineral density (BMD) or more accurately bone mineral content per projected area in g/cm2. Bone mineral content (BMC) is therefore calculated by summing the BMD values over the projected area. Sometimes in SPA and SXA the BMC is normalised to the width of the scanned region, resulting in bone mineral content per length in g/cm. Quantitative computed tomography (QCT), in contrast, measures a volumetric density and the results are given in g/cm3. Terminology is often applied inconsistently and requires particular attention on the part of the reader [21]. Radiation dose and scan time are very similar among the different techniques (Table 1). The peripheral location of the forearm dramatically reduces the gonadal dose and the effective absorbed radiation dose of 1 mSV or less is small compared with the yearly background dose of 2400 mSv [22].

SPA

Principles Quantitative bone mass measurement is based on the principle of linear attenuation of radiation in tissue due to photoelectric absorption, coherent scattering and Compton scattering. The amount of absorbed radiation per length can be described by a material-specific parameter, the attenuation coefficient. If the types of tissue contributing to the attenuation process and their specific coefficients of attenuation are known the relative proportion of two different materials can be calculated. A quantitative determination of bone mineral density requires calibration with materials of known coefficients of attenuation. To account for problems with polychromatic radiation (beam hardening), beam geometry, resolution, and imaging artifacts (partial volume effects, exponential edge gradient), anthropometric phantoms are highly recommended. To assess these effects separately specific phantoms of simple geometry are

Single photon absorptiometry (SPA) for bone densitometry was developed by Cameron and Sorenson in 1963 [23]. The method uses a collimated 125I (iodine) gamma source with a 27 keV peak and a crystal photomultiplier tube detector [24]. The source and detector form a rigidly coupled assembly that is motor-driven to traverse the longitudinal axis of the bone. The measurement site is placed between the source and the detector and single or multiple scan paths can be performed. A tissueequivalent substance, usually water, must provide a constant thickness of absorbing material [25]. The tissue-equivalent absorbing material provides a baseline absorption that can be substracted from the absorption in bone tissue. The fat content at the distal forearm can be determined with a precision of less than 2% and this information is used to minimise the influence of subcutaneous fat on the quantitative measures of bone mineral [26].

Table 1. Scanning characteristics for the different bone mineral measurement techniques Technique

Radiation dosea

Scan time

Resolution

SPA SXA DPA DXA

50–150 mSv [30,54] 17 mSv [27] 30–40 mSv [55,69] 20–30 mSv [46,55] 0.07 mSV* [131] 30–100 mSv [40,133]

10–15 min [30,51,52,54] 4 min [27] 25 min [130] 3–6 min [46,51,52]

0.4 mm [27]

pQCT a

6–10 min

Radiation dose is given in skin entrance dose or effective absorbed dose (*).

1–4 mm [130] 1.5 mm [132] 0.25–0.68 mm [44,105,126]

Quantitative Bone Mineral Assessment at the Forearm

301

SXA In single X-ray absorptiometry (SXA) an X-ray beam replaces the radiation source of SPA. To reduce the effect of beam hardening a narrow energy band is produced by filtering a polychromatic 40 kVp X-ray beam with a thin tin foil. The X-ray detection system consists of a sodium iodide crystal coupled to a photomultiplier tube. Multiple scan paths are used to sample a larger region of bone and improve the precision of the measurement. Calibration is achieved automatically by sampling an aluminium reference material at the end of every scan line [27].

DPA Dual photon absorptiometry (DPA) was developed in the late 1970s to overcome some limitations of SPA and enable scanning of the axial skeleton [28,29]. To distinguish between absorption in soft tissue and in bone and to eliminate the need for a constant path length a radiation source with two different energy peaks is used. Typically a DPA system consists of a collimated 153 Gd (gadolinium) source with photon energy peaks at 44 keV and 100 keV. A soft tissue equivalent is not necessary because the two different energy peaks are used to calculate the contribution of soft tissue mass and bone mass. Quantitative DPA requires a number of corrections, performed by computer software for the effect of radiation scattering, background radiation, bone edge detection and soft tissue baselines [30].

DXA Dual X-ray absorptiometry (DXA) is based on the principle of DPA. Instead of the photon emission source two X-ray beams generate a dual-energy radiation. Two energetically different X-ray beams are either produced by a single X-ray potential followed by a K-edge filtration (Cerium Filter, Samarium Filter) or by operating the X-ray unit with two different potentials and differential filtration [31]. Low- and high-level potentials of X-ray units vary between different machines and are typically 39 kVp and 70 kVp or 70 kVp and 140 kVp respectively. The higher photon flux generated by the X-ray source compared with the photon source allows the use of smaller collimators for higher resolution, reduced scatter and reduced scanning time. On DXA images the anatomy can be clearly visualized and regions of interest can easily be delineated (Fig. 1). Only recently dedicated systems for forearm measurements (peripheral DXA) have been introduced with reduced X-ray energies and simplified design.

Fig. 1. Projectional image of a right forearm by dual X-ray absorptiometry (DXA). The scan locations ultra-distal (UD), middistal (MID) and proximal (1/3) are defined by measurement of the ulnar length and placement of the distal reference line at the tip of the ulnar styloid.

CT In computed tomography (CT) a thin cross-sectional image of the distal forearm is reconstructed by a computer from the data of many X-ray projections [32,33]. The attenuation coefficients at each point of the cross-sectional image must be reconstructed from the projected data [34]. These CT numbers can be quantified for bone mineral assessment by a simultaneous exposure of calibration phantoms usually containing dipostassium hydrogen phosphate or hydroxyapatite [35,36]. The usefulness of CT lies in its ability to provide measures for cancellous, cortical and integral bone separately [36]. Special-purpose CT scanners have been developed for imaging the peripheral skeleton [37,38]. Peripheral quantitative CT (pQCT) scanners are in general smaller, less expensive, more mobile and do not need additional radiation shielding as compared with multipurpose CT machines. The first pQCT systems employed 125I photon sources for the generation of radiation while the newer devices employ X-ray tubes [38–41]. After selection of the scan location on a projectional scout view single slice or multiple slice CT scans are performed with slice thicknesses ranging form less than 1 mm up to 2.5 mm.

302

P. Augat et al.

Fig. 2. The ultra-distal scan region in peripheral quantitative computed tomography (pQCT) is defined by placement of a reference line at the radial endplate on an anterior–posterior scout view and automatic calculation of 4% of the ulnar length. Threshold-based image analysis enables the separation of cortical and trabecular bone in the distal radius region of interest.

Multiple slice techniques are advantageous with respect to measurement precision by evaluating a larger scan volume and employing algorithms for location matching [42,43]. The development of higher resolution scanning modes and reductions in slice thickness led to higher geometric resolution and, secondarily, by reduction of the partial volume effect, to high-contrast images of trabecular bone on which morphometric analysis can be performed [44,45].

Scan Location The scan locations at the forearm are based on the ulnar length measured between the ulnar styloid and the olecranon process, which can be easily palpated [25]. The scanning sites in forearm protocols are usually the ultra-distal site, the distal or mid-distal site, and the proximal or shaft site [46,47]. Furthermore, the analysis can be performed for both forearm bones or for the radius and the ulna individually. However, the definition of the specific locations varies among the different modalities and the manufacturers of the scanning devices. The placement of the regions of interest is usually performed semi-automatically on the scan image itself or, in the case of pQCT, on an anterior–posterior projectional scout view. The ultra-distal region usually includes a 10–15 mm length of the radius and the ulna which begins at a certain distance proximal to the base of the ulnar styloid process or the radial endplate, or is defined by the bifurcation point of radius and ulna. In single slice pQCT the ultra-distal scan location is centred at 4% of the forearm length proximal to the radial endplate (Fig. 2). In multiple slice pQCT a larger volume is scanned that usually extends between the 4% and 10% scanning sites. The ultra-distal scanning site contains up to 80%

Fig. 3. Scan locations at the distal forearm, defined with respect to the length of the forearm measured between the olecranon and the ulnar styloid. Reference lines are placed at the tip of the ulnar styloid or at the radial endplate.

trabecular bone and is used for the examination of the trabecular compartment [48]. The distal or mid-distal site covers a region that is defined with respect to a distance of 5–8 mm from the bifurcation point of radius and ulna, or can be defined by the edges of the adjacent ultra-distal and shaft scanning regions (Fig. 3). The shaft region or proximal region is usually located at 1/3 of the length of the forearm and covers between 10 and 15 mm [49]. It contains less than 10% trabecular bone and the bone mineral content remains fairly constant between 10% and 90% of the radius length [48,50], making this site preferable for measurements of cortical properties.

Comparison Between Techniques Comparisons between different techniques for quantitative bone assessment at the distal forearm have been performed extensively since DXA was invented as a faster, more precise and more accurate method. The introduction of DXA simplified the measurement procedures of single-energy scans. The dual-energy


303

technique made soft tissue equivalents or water baths useless. Improved mechanics and sophisticated scanning and detecting techniques accelerated the image acquisition. Comparisons with phantom measurements have revealed excellent agreement between SPA and DXA [51]. In vivo forearm measurements are comparable only if the anatomical location of the scanning site is similar. For well-matched scan sites an intermachine crosscalibration can provide almost identical results for quantitative bone mineral measurement [16, 52, 53]. Different sizes and reference points for the scanned regions of interest reduce the correlation between the two methods and undermine the precise follow-up for an individual measured by both techniques [54]. SXA and DXA have shown excellent correlations in vivo at the 8 mm distal forearm site (r = 0.97 for BMC and r = 0.96 for BMD) [27]. The higher photon flux available from X-ray sources provides several advantages of DXA over DPA. DXA offers better precision, improved spatial resolution, reduced influence of soft issue thickness and decreased scan times [55]. Correlations between the two types of instruments of phantoms and of measurements in the spine and the femur have been close to unity [56]. Comparison between pQCT and the projectional methods is confounded due to differences in imaging principles and dissimilar scan locations. Correlations between DXA and pQCT at the ultra-distal site have been not higher than r = 0.75 when measurements of DXA-BMD have been directly compared with pQCTBMD [57]. A considerable improvement has been achieved with the approximation of projectional densities using pQCT volumetric densities and geometric properties. In vivo correlations between DXA and derived pQCT measurements have ranged between r = 0.89 and r = 0.98 with standard error of estimates of the order of 3–7% [58]. Scanning instabilities and technical malfunctions influence quantitative results of bone mineral measurements. Rigorous quality control is mandatory in the application of quantitative densitometry. Quality control as well as cross-calibration between different machines [59] and even different methods can be performed with

special-purpose scan phantoms [60]. For measurements at the forearm a semi-anthropomorphic phantom has been developed that mimics the distal forearm [61]. Measurement of this forearm phantom on different SPA, DXA and CT machines has demonstrated that bone mineral measurements are only comparable after machine-to-machine cross-calibration [60]. Differences in radiation energy, scan geometry, correction for fat and soft tissue, edge detection algorithms, and region of interest (ROI) definition are responsible for the differences observed between scanners of different make.

Accuracy and Precision Accuracy can be defined by the difference between the measured value and the true physical property. The accuracy of absorption methods can be determined by measuring excised bones or phantoms whose mineral content or geometric properties are known (Table 2). Precision can be defined as the ability to make reproducible measurements without regard to their accuracy. Sources of imprecision and inaccuracy in quantitative bone mineral assessment are machine and measurement variables (Poisson statistics, source stability, temperature, humidity) and positioning. The precision of a method can be determined by repeated measurement of either phantoms, excised bones, or patients with complete repositioning between individual scans. Precise patient measurements require exact repositioning to make sure that the same region of bone is assessed. This is especially important in the ultra-distal region where bone density changes rapidly with position along the length of the forearm. Imprecision due to repositioning errors can be reduced by methods that automatically select scanning locations and regions of interest. The peripheral location of the forearm and the relatively small amount of surrounding soft tissue minimise errors due to beam hardening or fat and soft tissue corrections. Successful efforts have been undertaken during the development of quantitative bone assessment to maximise the in vivo precision of the

Table 2. Accuracy and precision of the different modalities for forearm densitometry Technique Accuracy (r)

In vitro precision (%CV)

In vivo precision (%CV)

SPA

SEE = 2–3% [30,52]

BMC: 1% [121] BMD: 2%–3% [30] BMC: 0.5% [27]

Ash content: r40.99, SEE 55% [123] Ash content: r40.97 [59] Geometry: r40.96 [102] Calcium content: r40.97 [124] Ash content: r40.99 [124] Total mineral: r40.86 (SEE 518%) [126] Cortical mineral: r40.91 (SEE 411%) [127] Cortical thickness: r40.88 [128]

BMD: 0.2%–1.7% [111 BMD: 0.4%–0.9% [52,58]

BMC: 0.6%–1.9% [49] BMD: 0.8%–1.4% [51,53,54] BMC: 0.6% [27] BMD: 1%–2% [122] BMD: 0.5%–1.5% [81,111] BMC: 0.7%–1.3% [46,53,54]

SXA DPA DXA CT pQCT

BMD: 2.1% [125] Geometry: 0.5% [125] BMC: 0.4%–1.0% [58] BMD: 0.03%–0.3% [2,39,43] Geometry: 1.1%–4.6% [2,57]

SEE, standard error of estimates; %CV, coefficient of variation.

BMD: 2% [124] BMC: 1.1%–2.2% [38,58] BMD: 0.3–2.2% [2,43,57,129] Geometry: 1.1%–4.6% [2,57]

304

P. Augat et al.

different scanning devices. The specific solutions depend on the measurement method but have resulted in an in vivo precision error that is less than 2% for all currently available devices (Table 2).

Factors Affecting Numerical Values The numerical values reported for bone mass measurements depend on several factors and, therefore, can vary considerably. A direct comparison of absolute values, which depend on the scanner as well as the calibration reference, is difficult without careful cross-calibration. Differences between populations, however, can be compared within the same study. Table 3 enables comparison within the same study by looking at individual lines. The most important factors that influence the results of bone mass measurements are scanning location, age, sex

and menopausal status. The contribution of cortical and trabecular bone is known to vary considerably along the lengths of the radius and the ulna [48], resulting in a strong dependency of bone mass and geometric properties on the location of the forearm [50]. The different regions of interest examined with different systems account for variations in bone mass measurements and aggravate intermodality comparison. Measurements of cortical bone are more likely to be affected by anthropometric factors such as height and weight, while measurements of trabecular bone are more closely related to age and menstrual status [62]. Age dependency of bone mineral measurements at the distal forearm usually shows a biphasic or even triphasic behaviour. The mass of bone increases during early adulthood until it reaches the so-called peak bone mass value at an age of about 30 years. When the peak bone mass is reached the values remain stable with age and begin to decrease at the age of about 40 years [2,63] or

Table 3. Numerical values for bone mineral measurements at the different forearm locations as a function of modality and menopausal status Technique

Location

Premenopausal BMC/W (g/cm)

Postmenopausal BMC (g)

BMD (g/cm2)*

BMC/W (g/cm)

0.95+0.16b

0.36+0.05b

0.42+0.06c

Osteoporotic BMC (g)

BMD (g/cm2)*

BMC/W (g/cm)

BMC (g)

BMD (g/cm2)*

0.31+0.06a SPA

Distal 0.51+0.07c 0.58+0.13p

0.26+0.06j 0.33+0.07j,p 0.47+0.05j 0.42+0.10p

Mid-distal 0.60+0.16p

0.38+0.07j 0.45+0.08a

0.89+0.10b

Proximal 1/3 0.86+0.11m 0.63+0.18p SXA

0.71+0.05b 0.57+0.05j 0.69+0.06m

0.79+0.14g 0.49+0.08j 0.53+0.09m

0.69+0.12m

0.47+0.04e

Ultra-distal

0.67+0.14j Mid-distal Proximal DPA

Ultra-distal

DXA

Ultra-distal

3.2 +0.4h

2.5 +0.4h

0.48+0.04h 0.71+0.12

l

0.56+0.13l

0.31+0.06e 0.21+0.05l

0.61+0.10l

0.42+0.07l 0.3 +0.05r 0.23+0.06s

0.43j 0.44+0.05q

1.7 +1.1o

0.48+0.06q

0.69+0.05q

0.53+0.06q 0.62+0.09a 0.59i 0.59+0.06n 0.60+0.07q

0.16+0.04f 0.18+0.03k 0.19+0.03q 0.35+0.05f 0.24+0.03k 0.36+0.05q

0.13+0.05f 0.16+0.02k 0.17+0.05q 0.26+0.07f 0.22+0.03k 0.32+0.05q

0.59+0.05q

1.1 +0.6o

Proximal 1/3 0.70j

0.29+0.05q

0.28+0.06n 0.34+0.06q

0.54d

Ultra-distal Trabecular Bone Ultra-distal Total bone

0.40+0.07a 0.35i 0.37+0.05n 0.38+0.04q 0.53+0.07a 0.47d 0.50+0.06n

Mid-distal

pQCT

0.36+0.05e 0.26+0.06l 0.36+0.05h 0.49+0.08l

0.24+0.03q

*In the case of pQCT an the units for BMD are g/cm3. a [54]; b[112]; c[5]; d[4]; e[115]; f[2]; g[114]; h[27]; i[46]; j[49]; k[40]; l[66]; m[119]; n[113]; o[120]; p[102]; q[57]; r[81]; s[82].

0.41+0.06n

0.49+0.07n 0.55+0.06q

0.06+0.02q

0.16+0.04q

0.22+0.04q

0.29+0.05q


even earlier [64]. In women the hormonal changes at the end of the menopause cause a steep decrease in bone mass known as postmenopausal bone loss (Table 3). Due to the large variation of the ratio of trabecular and cortical bone, age-related changes of BMD measurements at the distal radius depend on the location of the scan line and can vary considerably (Table 3). At the shaft site, which contains predominantly cortical bone, reports on annual bone loss vary between 0.7%/year (SPA and DXA) [46,64,65] and 1.1%/year (SXA) [66]. At the mid-distal site the annual loss of bone seems to be slightly higher and has been estimated as between 0.9% (SPA) [65], 1.2% (SPA) [7,27] and 1.5% (SXA) [66]. Reports on bone loss at the ultra-distal site report annual losses between 0.4% [57] and 0.7% [46]. Price et al. [49] found a steep decrease in bone density in the early postmenopausal phase at all forearm locations ranging from 3.6%/year at the ultra-distal site to 1.8%/year at the shaft site. Using pQCT age-related bone loss can be studied separately for cortical and trabecular bone at the same site. Butz et al. [2] found that the loss of trabecular bone in women (0.85%/year) was more pronounced as in men (0.59%/year) and comparable to the total loss of bone in the distal radius (1.08%/year for women and 0.54% in men). Grampp et al. [57] found, cross-sectionally, that the annual loss of bone mass in healthy women was twice as much for cortical bone (0.5%/year) as the loss in trabecular bone (0.22% year), while Boomen et al. [67] found the opposite in an elderly female population. In children no differences in appendicular bone mass could be found between males and females [18,19]. However, later in life bone mass in men exceeds that in women, principally due to the larger bone size in men. Gender differences in bone mineral status become more pronounced when the bone mass in women decreases dramatically at the menopause [18,68,69]. Differences in bone mineral properties between different ethnic groups are still significant after the adjustment for body size [70,71]. The highest values for bone mineral densities have been reported for African-Americans, while lowest values are associated with Asian heritage [63,70,71].

Response to Therapy Precise assessment of bone mineral is one of the most important criteria for depicting responses to therapy. Its precision, ease of use and wide availability have made forearm densitometry a popular tool for monitoring treatment in a variety of studies. However, longitudinal sensitivity does not depend solely on precision but has to take into account the responsiveness as well. The responsiveness not only depends on the form of treatment but also differs between measurement sites, types of bone and scanning technique. This results in a complex relationship between responsiveness and form of treatment, a detailed description of which is beyond

305

the scope of this paper. The following overview merely illustrates the wide applicability of forearm densitometry. Hormone replacement therapy (HRT) by oestrogen has shown to be effective in the prevention of bone loss at the peripheral skeleton in a variety of studies [72–75]. There are some indications that short periods of oestrogen treatment are insufficient in the non-loadbearing forearm [8,73], while long-term HRT seemed to be more effective on the forearm and the spine compared with the femur [73,76]. Forearm densitometry has been successfully used to demonstrate the efficacy of bisphosphonates in osteoporosis [77–79] and Paget’s disease [80]. Other osteoporotic treatments that have been monitored at the forearm include ipriflavone [81,82], oral calcium [83,84] and calcitonin [85,86]. The responsiveness to bisphosphonate therapies, however, is significantly lower at the distal forearm than at the hip or the spine [87–89]. It is not entirely clear whether these differences between different bones are due to different amounts of trabecular and cortical bone or to different amounts of weight-bearing at different anatomical locations. While these factors result in variable rates of bone turnover it might also by speculated that some forms of treatment affect different bone compartments differently. Finally, some forms of treatment not only affect bone mineral but also alter soft tissue and fat composition and thereby induce differential errors in bone density measurement [90,91].

Prediction of Fracture Risk Measurements of BMD at the forearm have been shown to predict osteoporotic fractures at the forearm itself as well as at the hip and the spine. Fractures at the distal forearm are strongly related to the future risk of vertebral fractures [92,93] and femoral neck fractures [92,94]. Likewise, patients with osteoporotic fractures of the forearm are most likely to have decreased BMD at the spine [95,96] and the hip. The predictive accuracy for all non-spine fractures together is very similar for bone mass measurements at the femur, the spine and the distal radius [97,98]. At an individual level the degree of heterogeneity in the extent of osteoporosis at different sites strongly influences the consequences of bone mass measurement [63]. A higher sensitivity for assigning fracture risks related to osteopenia may perhaps be the combination of bone density measurements at multiple sites [99].

Forearm Fractures The strong relationship between the mechanical strength of the forearm and BMD measurements at that location has been shown in various biomechanical studies. BMD, BMC, cortical area and moment of inertia all showed a strong relationship to the strength of the radius.

306

Compression tests of the ultra-distal region revealed the highest correlations between ultimate strength and BMC (r = 0.83–0.87) [100,101]. The loads to produce distal radius fractures by simulating a fall on to the outstretched hand [102,103] have been accurately predicted by density measurements with SXA or DXA and geometric measurements with CT. The most accurate predictors of fracture strength were SPA-BMC at the ultra-distal site (r = 0.94) [104], area of cortical bone at the shaft site (r = 0.84–0.89) [103,104] and combinations of BMD with moments of inertia (r = 0.93) [105]. Cross-sectional studies have shown that BMD at the distal radius is generally decreased in patients with distal forearm fractures or Colles’ fractures [7,93,106–108]. None of the different measurement sites conferred consistent advantages in fracture discrimination over the other sites [49]. Studies using CT suggested that the cortical geometry differed between patients with fractures and healthy controls [106]. In prospective studies the risk of distal forearm fractures was inversely related to BMD measurements at the distal site [109] and the shaft site [3]. The predictive ability of measurements at the shaft site was found to be slightly superior to that of the ultra-distal site [110]. The relative risk associated with a 1 standard deviation decrease in BMD ranged between 1.6 and 1.8 [97,109].

P. Augat et al.

Hip Fractures BMD measurements at the distal radius have been found to be moderately correlated with BMD measurements at the hip. In vitro studies have revealed moderate correlations between trabecular BMD (r = 0.73) at the distal radius and the femoral neck [105]. The strongest in vivo correlations have been found between the distal or mid-distal sites and the trochanteric region of the femur (r = 0.66) [54] and between the trabecular compartment at the ultra-distal site and Ward’s triangle (r = 0.71) [2]. The total bone densities at the ultra-distal radius site (SPA), however, correlated rather moderately with all the different regions in the proximal femur (r50.53– 0.59) [65,112]. Measurements at the distal forearm have been able to differentiate between postmenopausal women who sustained a fracture of the hip and those without any long bone fractures [93]. Distal forearm BMD has enabled the prediction of future fractures of the hip, although the predictive power has not been as high as for BMD measurements at the hip itself [116,117]. In the Study of Osteoporotic Fractures, an analysis of 8134 women of whom 65 had a fracture of their hip, a 1 standard deviation decrease in distal or mid-distal radius BMD (as measured by SPA) has been associated with a 1.5–1.6 times increased risk of hip fracture, while low BMD values at the hip itself have increased the fracture risk by a factor of about 2.8 [116].

Vertebral Fractures

Summary The predominance of trabecular bone at the distal radius and in vertebral bodies might suggest that BMD measurements at these two locations are closely correlated. While in vitro correlations have been rather strong (r = 0.81–0.85) [111], additional sources of error in in vivo measurements further decreased the correlations considerably (r = 0.4–0.7) [54,57,66,111,112]. It is not entirely clear to what extent degenerative changes at the spine, different rates of bone loss, and different correlations with age are responsible for these differences. Low in vivo correlations indicate that peripheral measurements are of minor significance for the prediction of the vertebral status, especially in the assessment of individual patients. In cross-sectional studies vertebral fractures have been associated with decreased BMD at the distal radius [7,66,93,113] while measurements at the shaft site have revealed conflicting results [113,114]. SXA of the middistal region of the non-dominant forearm has demonstrated diagnostic ability comparable to antero-posterior and lateral lumbar spine DXA for the detection of radiographically confirmed fractures of the lumbar vertebrae [115]. The risk of developing vertebral deformities has been demonstrated to increase by approximately 1.5 per standard deviation decrease in BMD or BMC at the forearm with no significant differences between distal or proximal sites [66].

A variety of modalities enable quantitative assessment of bone mineral at the forearm. The most important goals of these measurements are: (1) evaluation of perimenopausal women for initiation of therapy, (2) detection of osteoporosis and assessment of its severity, (3) evaluation of patients with metabolic diseases that affect the skeleton, and (4) monitoring of treatment and evaluation of disease course [118]. Low radiation doses, comfortable and fast handling, moderate cost, and a strong association with the risk of non-spine fractures, promote the use of forearm scanning as a widely applied screening procedure for the detection of generalized osteoporotic bone loss. However, a higher accuracy of fracture risk prediction at the spine or the hip can be achieved by a direct bone density measurement at these sites. The monitoring of treatment at the distal forearm requires a longer follow-up time due to the decreased responsiveness compared with such highly trabecular load-bearing sites as the spine and the proximal femur.

References 1. Little K. Bone behaviour. London: Academic Press, 1973. 2. Butz S, Wu¨ster C,Scheidt-Nave C, Go¨tz M, Ziegler R. Forearm BMD as measured by peripheral quantitative computed tomography (pQCT) in a German reference population. Osteoporos Int 1994;4:179–84.

Quantitative Bone Mineral Assessment at the Forearm 3. Ga¨rdsell P, Johnell O, Nilsson BE. Predicting fractures in women by using forearm bone densitometry. Calcif Tissue Int 1989;44:235–42. 4. Bjarnason K, Hassager C, Ravn P, Christiansen C. Early postmenopausal diminution of forearm and spinal bone mineral density: a cross-sectional study. Osteoporos Int 1995;5:35–8. 5. Bagni B, Bagni I, Scutellari PN, et al. [Normal ranges of bone mineral content of the distal radius determined with single photon absorptiometry: study of 1521 cases]. Radiol Med (Torino) 1991;82:508–11. 6. Luisetto G, Zangari M, Tizian L, et al. Influence of aging and menopause in determining vertebral and distal forearm bone loss in adult healthy women. Bone Miner 1993;22:9–25. 7. Gilfillan CP, Silberberg S, Scrivenor P, et al. Determinants of forearm mineral density and its correlation with fracture history in women. Maturitas 1994;20:199–208. 8. Kohrt WM, Birge SJ Jr. Differential effects of estrogen treatment on bone mineral density of the spine, hip, wrist and total body in late postmenopausal women. Osteoporos Int 1995;5:150–5. 9. Lane NE, Pressman AR, Star VL, Cummings SR, Nevitt MC. Rheumatoid arthritis and bone mineral density in elderly women. The Study of Osteoporotic Fractures Research Group. J Bone Miner Res 1995;10:257–63. 10. Toyoda T, Inokuchi S, Saito S, Horie Y, Tomita S. Bone loss of the radius in rheumatoid arthritis: comparison between 34 patients and 40 controls. Acta Orthop Scand 1996;67:269–73. 11. Sakurai K, Marurno F, Iwanami S, Uchida H, Matsubayashi T. Quantitative computed tomographic evaluation of femoral bone mineral content in renal osteodystrophy compared with radial photon absorptiometry. Invest Radiol 1989;24:375–82. 12. Amato G, Izzo G, La Montagna G, Bellastella A. Low dose recombinant human growth hormone normalises bone metabolism and cortical bone density and improves trabecular bone density in growth hormone deficient adults without causing adverse effects. Clin Endocrinol 1996;45:27–32. 13. Minisola S, Rosso R, Romagnoli E, et al. Trabecular bone mineral density in primary hyperparathyroidism: relationship to clinical presentation and biomakers of skeletal turnover. Bone Miner 1993;20:113–23. 14. Karlsson MK, Johnell O, Obrant KJ. Bone mineral density in weight lifters. Calcif Tissue Int 1993;52:212–5. 15. Sandler RB, Cauley JA, Hom DL, Sashin D, Kriska AM. The effects of walking on the cross-sectional dimensions of the radius in postmenopausal women. Calcif Tissue Int 1987;41:65– 9. 16. Shore RM, Langman CB, Donovan JM, Conway JJ, Poznanki AK. Bone mineral disorders in children: evaluation with dual xray absorptiometry. Radiology 1995;196:535–40. 17. Zionts LE, Nash JP, Rude R, Ross T, Stott NS. Bone mineral density in children with mild osteogenesis imperfecta. J Bone Joint Surg Br 1995;77:143–7. 18. Sugimoto T, Nishino M, Tsunenari T, et al. Radial bone mineral content of normal Japanese infants and prepubertal children: influence of age, sex and body size. Bone Miner 1994;24:189– 200. 19. Zanchetta JR, Plotkin H, Alvarez Filgueira ML. Bone mass in children: normative values for the 2–20-year-old population. Bone 1995;16:S393–9. 20. Exner GU, Prader A, Elsasser U, Ruëgsegger P, Anliker M. Bone densitometry using computed tomography. I. Selective determination of trabecular bone density and other bone mineral parameters: normal values in children and adults. Br J Radiol 1979;52:14–23. 21. Jonson R. Mass attenuation coefficients, quantities and units for use in bone mineral determination. Osteoporos Int 1993;3:103– 6. 22. Kalender WA. Effective dose values in bone mineral measurements by photon absorptiometry and computed tomography. Osteoporos Int 1992;2:82–7. 23. Cameron JR, Sorenson JA. Measurement of bone mineral in vivo: an improved method. Science 1963;142:230–2. 24. Goldsmith NF, Johnston JO, Picetti G, Garcia C. Bone mineral

307

25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39.

40.

41.

42.

43. 44.

45.

46.

in the radius and vertebral osteoporosis in an insured population: a correlative study using 125-I photon absorption and miniature roentgenography. J Bone Joint Surg Am 1973;55:1276–93. Thorson LM, Wahner HW. Single- and dual-photon absorptiometry techniques for bone mineral analysis. J Nucl Med Technol 1986;14:163–71. Hassager C, Borg J, Christiansen C. Measurement of the subcutaneous fat in the distal forearm by single photon absorptiometry. Metabolism 1989;38:159–65. Kelly TL, Crane G, Baran DT. Single X-ray absorptiometry of the forearm: precision, correlation, and reference data. Calcif Tissue Int 1994;54:212–8. Reed GW. The assessment of bone mineralisation for the relative transmission of 241Am and 137Cs radiations. Phys Med Biol 1966;11:174. Peppler WW, Mazess RB. Total body bone mineral and lean body mass by dual-photon absorptiometry. I. Theory and measurement procedure. Calcif Tissue Int 1981;33:353–9. Mazess RB, Barden HS, Single- and dual-photon absorptiometry for bone measurement in osteoporosis. San Francisco: Radiology Research and Education Foundation, 1987. Greenfield MA. Current status of physical measurements of the skeleton. Med Phys 1992;19:1349–57. Ruëgsegger P, Elsasser U, Anliker M, et al. Quantification of bone mineralisation using computed tomography. Radiology 1976;121:93–7. Hounsfield GN. Computerized transverse axial scanning (tomography). 1. Description of system. Br J Radiol 1973;46:1016–22. Hermann GT. Image reconstruction from projections: the fundamentals of computerized tomography. Orlando: Academic Press, 1980. Cann CE. Quantitative CT applications: comparison of current scanners. Radiology 1987;162:257–61. Genant HK, Boyd D. Quantitative bone mineral analysis using dual energy computed tomography. Invest Radiol 1977;12:545– 51. Ruëgsegger P. Quantitative computed tomography at peripheral measuring sites. Ann Chir Gynaecol 1988:77:204–7. Hosie CJ, Smith DA. Precision of measurement at bone density with a special purpose computed tomography scanner. Br J Radiol 1986;59:345–50. Lehmann R, Wapniarz M, Kvasnicka HM, et al. [Reproducibility of bone density measurements of the distal radius using a high resolution special scanner of peripheral quantitative computed tomography (sinbgle energy PQCT)]. Radiologe 1992;32:177–81. Schneider P, Bo¨rner W. [Peripheral quantitative computed tomography for bone mineral measurement using a new special QCT-scanner: methodology, normal values, comparison with manifest osteoporosis]. Rofo Fortschr Geb Rontgenstr Neuen Bildgeb Verfahr 1991;154:292–9. Hangartner TN, Overton TR, Harley CH, van den Berg L, Crockford PM. Skeletal challenge: an experimental study of pharmacologically induced changes in bone density in the distal radius, using gamma-ray computed tomography. Calcif Tissue Int 1985;37:19–24. Mu¨ller R, Hildebrand T, Hauselmann HJ, Ruëgsegger P. In vivo reproducibility of three-dimensional structural properties of noninvasive bone biopsies using 3D-pQCT. J Bone Miner Res 1996;11:1745–50. Hangartner TN. The OsteoQuant: an isotope-based CT scanner for precise measurement of bone density. J Comput Assist Tomogr 1993;17:798–805. Mu¨ller R, Hahn M, Vogel M, Delling G, Ruëgsegger P. Morphometric anlaysis of noninvasively assessed bone biopsies: comparison of high-resolution computed tomography and histologic sections. Bone 1996;18:215–20. Gordon CL, Webber CE, Adachi JD, Christoforou N. In vivo assessment of trabecular bone structure at the distal radius from high-resolution computed tomography images. Phys Med Biol 1996;41:495–508. Leboff MS, Fuleihan GE, Angell JE, Chung S, Curtis K. Dual-

308

47. 48.

49.

50.

51. 52. 53. 54.

55. 56.

57.

58.

59. 60.

61. 62.

63. 64.

65. 66. 67.

P. Augat et al. energy x-ray absorptiometry of the forearm: reproducibility and correlation with single-photon absorptiometry. J Bone Miner Res 1992;7:841–6. Wahner HW, Eastell R, Riggs BL. Bone mineral density of the radius: where do we stand? J Nucl Med 1985;26:1339–41. Schlenker RA, VonSeggen WW. The distribution of cortical and trabecular bone mass along the lengths of the radius and ulna and the implications for in vivo bone mass measurements. Calcif Tissue Res 1976;20:41–52. Price RI, Barnes MP, Gutteridge DH, et al. Ultradistal and cortical forearm bone density in the assessment of postmenopausal bone loss and nonaxial fracture risk. J Bone Miner Res 1989;4:149–55. Hsu ES, Patwardhan AG, Meade KP, Light TR, Martin WR. Cross-sectional geometrical properties and bone mineral contents of the human radius and ulna. J Biomech 1993;26:1307–18. Weinstein RS, New KD, Sappington LJ. Dual-energy X-ray absorptiometry versus single photon absorptiometry of the radius. Calcif Tissue Int 1991;49:313–6. Larcos G, Wahner HW. An evaluation of forearm bone mineral measurement with dual-energy X-ray absorptiometry. J Nucl Med 1991;32:2101–6. Nelson D, Feingold M, Mascha E, Kleerekoper M. Comparison of single-photon and dual-energy x-ray absorptiometry of the radius. Bone Miner 1992;18:77–83. Nieves JW, Cosman F, Mars C, Lindsay R. Comparative assessment of bone mineral density of the forearm using single photon and dual X-ray absorptiometry. Calcif Tissue Int 1992;51:352–5. Mazess RB, Barden HS. Measurement of bone by dual-photon absorptiometry (DPA) and dual-energy X-ray absorptiometry (DEXA). Ann Chir Gynaecol 1988;77:197–203. Wahner HW, Dunn WL, Brown ML, Morin RL, Riggs BL. Comparison of dual-energy x-ray absorptiometry and dual photon absorptiometry for bone mineral measurements of the lumbar spine. Mayo Clin Proc 1988;63:1075–84. Grampp S, Lang P, Jergas M, et al. Assessment of the skeletal status by peripheral quantitative computed tomography of the forearm: short-term precision in vivo and comparison to dual Xray absorptiometry. J Bone Miner Res 1995;10:1566–76. Overton TR, Wheeler GD. Bone mass measurements in the distal forearm using dual-energy x-ray absorptiometry and gamma-ray computed tomography: a longitudinal, in vivo comparative study. J Bone Miner Res 1992;7:375–81. Hagiwara S, Engelke K, Yang SO, et al. Dual x-ray absorptiometry forearm software: accuracy and intermachine relationship. J Bone Miner Res 1994;9:1425–7. Pearson J, Ruëgsegger P, Dequeker J, et al. European semianthropomorphic phantom for the cross-calibration of peripheral bone densitometers: assessment of precision accuracy and stability. Bone Miner 1994;27:109–20. Ruëgsegger P, Kalender WA. A phantom for standardization and quality control in peripheral bone measurements by pQCT and DXA. Phys Med Biol 1993;38:1963–70. Vico L, Prallet B, Chappard D, et al. Contributions of chronological age, age at menarche and menopause and of anthropometric parameters to axial and peripheral bone densities. Osteoporos Int 1992;2:153–8. Davis JW, Novotny R, Ross PD, Wasnich RD. The peak bone mass of Hawaiian, Filipino, Japanese, and white women living in Hawaii. Calcif Tissue Int 1994;55:249–52. Sowers MR, Clark MK, Hollis B, et al. Radial bone mineral density in pre- and perimenopausal women: a prospective study of rates and risk factors for loss. J Bone Miner Res 1992;7:647– 57. Steiger P, Cummings SR, Black DM, Spencer NE, Genant HK. Age-related decrements in bone mineral density in women over 65. J Bone Miner Res 1992;7:625–32. Ross P, Huang C, Davis J, et al. Predicting vertebral deformity using bone densitometry at various skeletal sites and calcaneus ultrasound. Bone 1995;16:325–32. Boonen S, Cheng XG, Nijs J, et al. Factors associated with

68. 69.

70.

71.

72. 73. 74. 75.

76. 77.

78.

79. 80.

81. 82. 83.

84.

85. 86. 87.

cortical and trabecular bone loss as quantified by peripheral computed tomography (pQCT) at the ultradistal radius in aging women. Calcif Tissue Int 1997;60:164–70. Kelly PJ, Twomey L, Sambrook PN, Eisman JA. Sex differences in peak adult bone mineral density. J Bone Miner Res 1990;5:1169–75. Felson DT, Zhang Y, Hannan MT, Anderson JJ. Effect of weight and body mass index on bone mineral density in men and women: the Framingham study. J Bone Miner Res 1993;8:567– 73. Luckey MM, Meier DE, Mandeli JP, et al. Radial and vertebral bone density in white and black women: evidence for racial differences in premenopausal bone homeostasis. J Clin Endocrinol Metab 1989;69:762–70. Nelson DA, Jacobsen G, Barondess DA, Parfitt AM. Ethnic differences in regional bone density, hip axis length, and lifestyle variables among healthy black and white men. J Bone Miner Res 1995;10:782–7. Naesseń T, Mallmin H, Ljunghall S. Heel ultrasound in women after long-term ERT compared with bone densities in the forearm, spine and hip. Osteoporos Int 1995;5:205–10. Felson DT, Zhang Y, Hannan MT, et al. The effect of postmenopausal estrogen therapy on bone density in elderly women. N Engl J Med 1993;329:1141–6. Sowers MR, Clark MK, Jannausch ML, Wallace RB. Body size, estrogen use and thiazide diuretic use affect 5-year radial bone loss in postmenopausal women. Osteoporos Int 1993;3:314–21. Davis JW, Ross PD, Johnson NE, Wasnich RD. Estrogen and calcium supplement use among Japanese-American women: effects upon bone loss when used singly and in combination. Bone 1995;17:369–73. Felson DT, Zhang Y, Hannan MT, et al. The effect of postmenopausal estrogen therapy on bone density in elderly women. N Engl J Med 1993;329:1141–6. Erdtsieck RJ, Pols HA, Valk NK, et al. Treatment of postmenopausal osteoporosis with a combination of growth hormone and pamidronate: a placebo controlled trial. Clin Endocrinol 1995;43:557–65. Thie´baud D, Burckhardt P, Melchior J, et al. Two years’ effectiveness of intravenous pamidronate (APD) versus oral fluoride for osteoporosis occurring in the postmenopause. Osteoporos Int 1994;4:76–83. Smith ML, Fogelman I, Hart DM, et al. Effect of etidronate disodium on bone turnover following surgical menopause. Calcif Tissue Int 1989;44:74–9. Price RI, Gutteridge DH, Stuckey BG, et al. Rapid, divergent changes in spinal and forearm bone density following short-term intravenous treatment of Paget’s disease with parnidronate disodium. J Bone Miner Res 1993;8:209–17. Agnusdei D, Adami S, Cervetti R, et al. Effects of ipriflavone on bone mass and calcium metabolism in postmenopausal osteoporosis. Bone Miner 1992;19:S43–8. Passeri M, Biondi M, Costi D, et al. Effect of ipriflavone on bone mass in elderly osteoporotic women. Bone Miner 1992;19:S57–62. Ruëgsegger P, Keller A, Dambacher MA. Comparison of the treatment effects of ossein-hydroxyapatite compound and calcium carbonate in osteoporotic females. Osteoporos Int 1995;5:30–4. Cepollaro C, Orlandi G, Gonnelli S, et al. Effect of calcium supplementation as a high-calcium mineral water on bone loss in early postmenopausal women. Calcif Tissue Int 1996; 59:238–9. Sambrook P, Birmingham J, Kelly P, et al. Prevention of corticosteroid osteoporosis: a comparison of calcium, calcitriol and calcitonin. N Engl J Med 1993;328:1747–52. Arnala I, Saastamoinen J, Alhava EM. Salmon calcitonin in the prevention of bone loss at perimenopause. Bone 1996;18:629– 32. Liberman UA, Weiss SR, Broll J, et al. Effect of oral alendronate on bone mineral density and the incidence of fractures in postmenopausal osteoporosis: the Alendronate Phase


88. 89.

90. 91.

92. 93.

94. 95.

96. 97. 98.

99. 100. 101.

102. 103. 104. 105. 106. 107. 108. 109.

110.

III Osteoporosis Treatment Study Group. N Engl J Med 1995;333:1437–43. Tucci JR, Tonino RP, Emkey RD, et al. Effect of three years of oral alendronate treatment in postmenopausal women with osteoporosis. Am J Med 1996;101:488–501. Storm T, Thamsborg G, Steiniche T, Genant HK, Sorensen OH. Effect of intermittent cyclical etidronate therapy on bone mass and fracture rate in women with postmenopausal osteoporosis. N Engl J Med 1990;322:1265–71. Gluër CC, Genant HK. Impact of marrow fat on accuracy of quantitative CT. J Comput Assist Tomogr 1989;13:1023–35. Svendsen OL, Hassager C, Skodt V, Christiansen C. Impact of soft tissue on in vivo accuracy of bone mineral measurements in the spine, hip, and forearm: a human cadaver study. J Bone Miner Res 1995;10:868–73. Nordin BEC, Crilly RG, Smith DA, Osteoporosis. New York: Chruchill Livngstone, 1984. Jensen GF, Christiansen C, Boesen J, Hegedu¨s V, Transbol I. Epidemiology of postmenopausal spinal and long bone fractures: a unifying approach to postmenopausal osteoporosis. Clin Orthop 1982;166:75–81. Owen RA, Melton LJ III, Ilstrup DM, Johnson KA, Riggs BL. Colles’ fracture and subsequent hip fracture risk. Clin Othrop 1982;171:37–43. Krølner B, Tøndevold E, Toft B, Berthelsen B, Nielsen SP, Bone mass of the axial and the appendicular skeleton in women with Colles’ fracture: its relati0n to physical activity. Clin Physiol 1982;2:147–57. Peel NF, Barrington NA, Smith TW, Eastell R. Distal forearm fracture as risk factor for vertebral osteoporosis. BMJ 1994;308:1543–4. Black DM, Cummings SR, Genant HK, et al. Axial and appendicular bone density predict fractures in older women. J Bone Miner Res 1992;7:633–8. Kro¨ger H, Huopio J, Honkanen R, et al. Prediction of fracture risk using axial bone mineral density in a perimenopausal population: a prospective study. J Bone Miner Res 1995;10:302– 6. Davis J, Ross P, Wasnich R. Evidence for both generalized and regional low bone mass among elderly women. J Bone Miner Res 1994;9:305–9. Horsman A, Currey JD. Estimation of mechanical properties of the distal radius from bone mineral content and cortical width. Clin Orthop 1983;176:298–304. Louis O, Boulpaep F, Willnecker J, Van den Winkel P, Osteaux M. Cortical mineral content of the radius assessed by peripheral QCT predicts compressive strength on biomechanical testing. Bone 1995;16:375–9. Myers ER, Sebeny EA, Hecker AT, et al. Correlations between photon absorption properties and failure load of the distal radius in vitro. Calcif Tissue Int 1991;49:292–7. Myers ER, Hecker AT, Rooks DS, Hipp JA, Hayes WC. Geometric variables from DXA of the radius predict forearm fracture load in vitro. Calcif Tissue Int 1993;52:199–204. Spadaro JA, Werner FW, Brenner RA, et al. Cortical and trabecular bone contribute strength to the osteoporotic distal radius. J Orthop Res 1994;12:211–8. Augat P, Reeb H, Claes L. Prediction of fracture load at different skeletal sites by geometrical properties of the cortical shell. J Bone Miner Res 1996;11:1356–63. Hesp R, Klenerman L, Page L. Decreased radial bone mass in Colles’ fracture. Acta Orthop Scand 1984;55:573–5. Eastell R, Riggs BL, Wahner HW, et al. Colles’ fracture and bone density of the ultradistal radius. J Bone Miner Res 1989;4:607–13. Nordin BE, Chatterton BE, Walker CJ, Wishart J. The relation of forearm mineral density to peripheral fractures in postmenopausal women. Med J Aust 1987;146:300–4. Kelsey JL, Browner WS, Seeley DG, Nevitt MC, Cummings SR. risk factors for fractures of the distal forearm and proximal humerus. The study of Osteoporotic Fractures Research Group. Amm J Epidemiol 1992;135:477–89. Du¨ppe H, Gardsell P, Nilsson B. A single bone density

309

111.

112. 113.

114.

115.

116. 117. 118. 119.

120. 121. 122.

123. 124. 125.

126. 127. 128.

129. 130.

131.

measurement can predict fractures over 25 years. Calcif Tissue Int 1997;60:171–4. Hansen MA, Riis BJ, Overgaard K, Hassager C, Christiansen C. Bone mass measured by photon absorptiometry: comparison of forearm, heel, and spine. Scand J Clin Lab Invest 1990;50:517– 23. Mazess RB, Barden HS. Interrelationships among bone densitometry sites in normal young women. Bone Miner 1990;11:347–56. Ryan PJ, Blake GM, Herd R, Parker J, Fogelman I. Postmenopausal vertebral osteoporosis: can dual energy X-ray absorptiometry forearm bone density substitute for axial measurements? Br J Rheumatol 1994;33:546–9. Heuck AF, Block J, Glueer CC, Steiger P, Genant HK. Mild versus definite osteoporosis: comparison of bone densitometry techniques using different statistical models. J Bone Miner Res 1989;4:891–900. Bjarnason K, Nilas L, Hassager C, Christiansen C. Dual energy X-ray absorptiometry of the spine: decubitus lateral versus anteroposterior projection in osteoporotic women. Comparison to single energy X-ray absorptiometry of the forearm. Bone 1995;16:255–60. Cummings SR, Black DM, Nevitt MC et al. Bone density at various sites for prediction of hip fractures. The Study of Osteoporotic Fractures Research Group. Lancet 1993;341:72–5. Melton LJ III, Atkinson EJ, O’Fallon WM, Wahner HW, Riggs BL. Long-term fracture prediction by bone mineral assessed at different skeletal sites. J Bone Miner Res 1993;8:1227–33. Genant HK, Engelke K, Fuerst T, et al. Noninvasive assessment of bone mineral and structure: state of the art. J Bone Miner Res 1996;11:707–30. Bouxsein ML, Myburgh KH, Vandermeulen MCH, Lindenberger E, Marcus LR. Age-related differences in cross-sectional geometry of the forearm bones in healthy women. Calcif Tissue Int 1994;54:113–8. Myers ER, Hecker AT, Rooks DS, Hipp JA, Hayes WC. Geometric variables from DXA of the radius predict forearm fracture load in vitro. Calcif Tissue Int 1993;52:199–204. Mazess RB, Cameron JR, Miller H. Direct readout of bone mineral content using radionuclide absorptiometry. Int J Appl Rad Isotopes 1972;23:471–9. Rey P, Sornay-Rendu E, Garnero P, Vey-Marty B, Delmas PD. [Measurement of bone density in the wrist using X-ray absorptiometry: comparison with measures of other sites]. Rev Rhum Ed Fr 1994;61:619–26. Erman J, Ott SM. Accuracy of dual photon absorptiometry in excised femurs. J Nucl Med 1988;29:1853–5. Posner I, Griffiths HJ. Comparison of CT scanning with photon absorptiometric measurement of bone mineral content in the appendicular skeleton. Invest Radiol 1977;12:542–4. Cheng S, Toivanen JA, Suominen H, Toivanen JT, Timonen J. Estimation of structural and geometrical properties of cortical bone by computerized tomography in 78-year-old women. J Bone Miner Res 1995;10:139–48. Takada M, Engelke K, Hagiwara S, Grampp S, Genant HK. Accuracy and precision study in vitro for peripheral quantitative computed tomography. Osteoporos Int 1996;6:207–12. Louis O, Soykens S, Willnecker J, Van den Winkel P, Osteaux M. Cortical and total bone mineral content of the radius: accuracy of peripheral QCT. Bone 1996;18:467–72. Louis O, Willnecker J, Soykens S, Vandenwinkel P, Osteaux M. cortical thickness assessed by peripheral quantitative computed tomography: accuracy evaluated on radius specimens. Osteoporos Int 1995;5:446–9. Mu¨ller A, Ruëgsegger E, Ruëgsegger P. Peripheral QCT: a low risk procedure to identify women predisposed to osteoporosis. Phys Med Biol 1989;34:741–9. von Wowern N, Storm TL, Olgaard K. Bone mineral content by photon absorptiometry of the mandible compared with that of the forearm and the lumbar spine. Calcif Tissue Int 1988;42:157–61. Lewis MK, Blake GM, Fogelman I. Patient dose in dual x-ray absorptiometry. Osteoporos Int 1994;4:11–5.

310 132. Seo GS, Shiraki M, Aoki C, et al. Assessment of bone density in the distal radius with computer assisted X-ray densitometry (CXD). Bone Miner 1994;27:172–82.

P. Augat et al. 133. Ruëgsegger P, Durand E, Darnbacher MA. Localization of regional forearm bone loss form high resolution computed tomographic images. Osteoporos Int 1991;1:76–80.

Received for publication 4 August 1997 Accepted in revised form 13 November 1997