Vertebral Morphometry: Repeat Scan Precision Using the Lunar Expert-XL and the Hologic 4500A. A Study for the “WISDOM” RCT of Hormone Replacement Therapy

On radiation safety grounds there is concern about the morbidity attributable to routine radiographs of the spine for the identification of new fractures in large-scale trials of fracture prevention. However, the role of the potentially safer low-radiation-dose technique of vertebral morphometry performed by third generation dual-energy X-ray absorptiometry equipment requires evaluation for use in clinical trials. We have therefore investigated the short-term inter-scan imprecision as well as the imprecision attributable to different-day analyses by the same operator and differences in analyses by different operators. The volunteer subjects were participants in a pilot study for a randomized controlled trial of hormone replacement therapy (Women”s International Study of long Duration Oestrogen after Menopause, WISDOM). Each subject had two morphometric X-ray analysis scans separated by 2–4 weeks. Exclusions were women with densitometrically defined osteoporosis, as defined by the WHO criterion, and women with a body mass index exceeding 30.9 kg/m². On average, the women were 58.7 years of age and had bone mineral density values in the lumbar spine which were about 0.7 SD units higher than a reference US female age-matched population. Scans were assessed from vertebrae T7 through L4. In the study there were no clinically significant differences in performance between the Hologic QDR 4500A and the Lunar Expert XL equipment. Between-scan imprecision was significantly worse than imprecision attributable to reanalysis of the same scan by a different operator or the same operator after an interval. Vertebral level had an effect on measurement uncertainty, especially at the level of the diaphragm and at T7. Coefficients of variation, expressed as percentages of mean values, were better for absolute height measurements than for height ratios, ranging from 1.75% to 3.40% for the three heights measured on three separate machines and from 2.34% to 4.11% for the two height ratios. These results compared favorably with the equivalent figures from a parallel study of morphometry precision undertaken using standard lateral radiographs of the thoracic and lumbar spine (3.1–3.6% and 3.8–3.9%, respectively). We conclude that in trials of prevention therapy in women (or men) selected for not having osteoporosis, low-dose vertebral morphometry using the Hologic 4500A, the Lunar Expert XL or similar equipment is preferable on safety grounds to the classical technique based on standard radiographs, although conventional radiology may still be required in those with prevalent or incident deformities to exclude causes other than osteoporosis. The place of this low-dose technique in trials performed on patients with osteoporosis requires further study. Received: 29 June 1999 / Accepted: 15 December 1999     

Interest of a Prescreening Questionnaire to Reduce the Cost of Bone Densitometry

Bone mineral density (BMD) measurement is widely recognized as the best single tool to identify patients with a high lifetime risk of developing an osteoporosis-related fracture. However, the cost/benefit value of screening the whole population has been repeatedly challenged and demonstrated to be rather poor. In many countries, BMD scan is not or no longer reimbursed because of lack of validated criteria to identify patients who should benefit from this procedure. Based on the proposals of a nationwide expert panel, a simple questionnaire identifying historical, clinical and behavioral risk factors for osteoporosis was developed. The aim of this study was to assess the diagnostic accuracy of the proposed criteria; to determine the extent to which this questionnaire could be useful for optimizing the use of densitometry tests; and, more specifically, to estimate the diagnostic costs per osteoporotic or osteopenic patient detected. For this purpose, we applied the questionnaire to 3998 consecutive individuals at least 20 years old, of both genders, either consulting spontaneously or referred for a BMD measurement to an outpatient osteoporosis center. BMD was measured by dual-energy X-ray absorptiometry (DXA) at the lumbar spine and at the hip (both total hip and femoral neck). Diagnostic accuracies were evaluated through measures of sensitivity, specificity, and positive and negative predictive values. After determining a benchmark value for age, different strategies were compared in order to identify the most cost-effective one in terms of cost per patient detected. According to the WHO operational definition of osteoporosis (T-score <–2.5), 31% of the subjects were classified as osteoporotic at one or more of the measured sites. If only patients with at least one of the proposed risk factors had been referred for scans, 33.3% of the BMD measurements would have been avoided. Among those, less than 5% were missclassified as they did have osteoporosis at the total hip and up to 23% at one or more of the considered sites. On the other hand, of the subjects who would be recommended for a densitometry test, only a small fraction were identified correctly (the positive predictive values varied from 11.3% at the total hip to 34.8% at any site). In this first setting, the suggested criteria seem useful chiefly for excluding subjects who do not need a DXA scan rather than selecting osteoporotic patients. When applied only to patients aged 61 years or more, the positive predictive values rose to 15.1% (total hip) and 42.9% (any site), whereas the corresponding negative predictive values were set at 93% and 68.6%. In comparison, with a mass screening scenario the estimated diagnostic costs (costs associated with the DXA procedure) per osteoporotic patient detected at any of the considered sites would be reduced by more than 9% (59.4 instead of 65.3 Euros) if the suggested indications are taken into account for prescreening patients. And when the questionnaire is applied only to women over the age of 60 years these costs would be further reduced to 50.6 Euros, representing a 23% decrease. Then, a prescreening strategy based on these indications concomitantly with an age-selective criterion could represent a promising way toward a more rational use of BMD measurement. Received: 12 April 2001 / Accepted: 29 November 2001     

Efficiency of Quantitative Ultrasound Measurements as Compared with Dual-Energy X-ray Absorptiometry in the Assessment of Corticosteroid-Induced Bone Impairment

Bone loss due to corticosteroid treatment differs from that of postmenopausal osteoporosis with regard to bone structure. Corticosteroids affect both horizontal and vertical trabeculae while horizontal trabeculae are damaged in postmenopausal osteoporosis. Dual-energy X-ray absorptiometry (DXA) is the gold standard to evaluate bone loss. The place of quantitative ultrasound (QUS), a technique that could theoretically provide information on bone structure, is not well established in corticosteroid-induced bone impairment. The aim of the study was to determine the usefulness of QUS in the assessment of corticosteroid-induced bone impairment. We hypothesized that the relationship between QUS and DXA could be influenced by changes in bone structure and thus differ with regard to corticosteroid treatment. Seventy-seven women with inflammatory diseases chronically treated with corticosteroids (dose: 7.5–15 mg/day), 29 without corticosteroids and 100 controls were investigated. Bone mineral density at the lumbar spine (BMDL) was measured by DXA and QUS parameters were measured at the calcaneus. Both the QUS parameters (SOS, BUA, Stiffness) and BMDL were significantly lower (by 1.3% for SOS, 5.8% for BUA, 12.7% for Stiffness and 11% for BMDL) in patients treated with corticosteroids compared with patients not taking corticosteroids and with controls (p<0.001, ANCOVA, with age and height as covariates). Multiple linear regressions of Stiffness, SOS and BUA as dependent variables on age, BMDL, corticosteroid treatment and a computed new variable designed to test the interaction between BMDL and the treatment group showed that Stiffness, SOS and BUA were dependent on age and BMDL (p<0.001); BUA and Stiffness were dependent on treatment group. Taking into account the age of the patients, a significant difference was observed in the relation between BUA and BMDL according to treatment with corticosteroids. A similar difference was found in the subgroup of patients without fractures. SOS and BUA were strongly correlated but their relation did not differ according to treatment. Thus, QUS is useful in the assessment of corticosteroid-associated bone loss. Furthermore, the observation of a significant difference in the relationship between BUA and BMDL with regard to corticosteroid treatment might support the hypothesis that QUS, especially BUA, could give additional information about bone structure. Received: 24 August 1998 / Accepted: 4 March 1999     

Accuracy and Precision of 62 Bone Densitometers Using a European Spine Phantom

Dual-energy absorptiometry (DXA) is widely used for bone mineral density measurements. Different types of devices are available. Differences between devices from either the same manufacturer or different manufacturers can lead to difficulties in clinical practice when patients are followed on different machines. We calculated the accuracy and precision of 62 DXA devices from two manufacturers (51 Hologic, 11 Lunar) using a European Spine Phantom (ESP, semi-anthropomorphic). The ESP was measured 5 times on each device without repositioning. Accuracy was assessed by comparing bone mineral density (BMD, g/cm²) values measured on each device with the actual value of the phantom. Precision was assessed by the coefficient of variation (CVsd), using the root mean square average. The limits of agreement were estimated from the differences between each replicate measurement of BMD and the estimated true value for a particular manufacturer, according to Bland and Altman. The results confirm the difference between devices from different manufacturers (18.5%). Mean CVsd values were 0.57% and 0.64% for Hologic and Lunar respectively. The limits of agreement among devices from the same manufacturer were 0.026 g/cm² and 0.025 g/cm² for Hologic and Lunar respectively. Differences in extreme results between devices from the same manufacturer were on average 5.4% and 3.6% for Hologic and Lunar respectively. Results of different devices from the same manufacturer are highly comparable, although unpredictable differences exist that may be clinically relevant. Received: 12 June 1998 / Accepted: 20 November 1998     

Does Osteoporosis Classification Using Heel BMD Agree Across Manufacturers?

The lack of standardization in bone mineral density (BMD) measurements is known. Several studies have been carried out to cross-calibrate the axial dual X-ray absorptiometry (DXA) devices. Recently, a number of peripheral DXA (pDXA) densitometers have been introduced. In this study we evaluated the agreement between two heel DXA devices on BMD and T-scores. A total of 99 females aged 21–78 years (ca. 16 per decade) had their non-dominant heel BMD measured using the PIXI (Lunar Inc.) and the Apollo (Norland Medical) pDXA scanners. The mean BMD values were 0.492 and 0.607 g/cm² and the mean T-scores using manufacturers’ specified reference data were −0.07 and −0.25 for the PIXI and Apollo, respectively. Both the BMD and T-score intermachine relationships were highly correlated but showed significant nonidentity slopes and non-zero offsets. The diagnostic comparison on T-scores resulted in 86% agreement between the instruments (weighted kappa score of 0.550). Normalizing the reference peaks and SDs using this study’s young adult population BMD results removed the systematic T-score disagreement. We found that PIXI and Apollo are highly correlated. Differences in BMD values are mainly due to different region of interest (ROI) definitions and additional T-score disagreement reflects the difference in normative databases. Received: 9 July 2001 / Accepted: 12 March 2002     

Femur BMD Bias Associated with DPX-IQ Reanalysis of DPX-Acquired Scan Files

Bone mineral density values calculated by DPX-IQ software reanalysis of original DPX software analyzed scans were compared at the femur region. Based on the results for 24 subjects each scanned on 12 different occasions, a significant bias at all sites – neck, Ward’s triangle and trochanter – ranging from −0.7% (neck) to −2.1% (trochanter) was found. For the trochanter site 16 of the 24 subjects had significantly lower mean BMD values after DPX-IQ reanalysis, a trend also noted at the neck and Ward’s triangle. Although it has been recommended that scans analyzed with DPX software do not require DPX-IQ reanalysis, our findings strongly indicate that DPX-IQ reanalysis is essential to minimize total error during follow-up BMD determinations. Received: 9 April 1999 / Accepted: 20 July 1999     

How the New Hologic Hip Normal Reference Values Affect the Densitometric Diagnosis of Osteoporosis

In February 1997, Hologic supplied new software to all QDR dual-energy X-ray absorptiometry (DXA) machines replacing the previous femoral normative reference database with the NHANES III normative data. In addition to changing the normative database (and therefore T-scores) for all regions of the hip, the new software has changed the primary region of interest from the femoral neck to the total hip. In the present study we examined how these changes influence the densitometric diagnosis of osteoporosis in a large clinical referral population (n= 2311, mean age 62.7 years). The patients had spine and hip DXA performed at either of two centers using a Hologic QDR-2000 over a 4-year period. T-scores were derived for each patient using both previous and current young normal reference databases. Intraindividual differences in T-scores were calculated. The prevalence of osteoporosis based on the two normative databases and the difference between the prevalence was calculated for each skeletal site. The average paired difference between current and previous T-scores at femoral neck is 0.64, the difference increasing with age. Using the new normative database, the percentage of osteoporotic patients decreases from 49% of all patients at the femoral neck to 28% at the femoral neck and 20% at the total hip. In conclusion, the densitometric diagnosis of osteoporosis will be affected in a significant proportion of women as a result of the implementation of the new hip normative database supplied by Hologic. Whether this will translate into fewer patients being treated remains to be seen. Received: 23 July 1997 / Accepted: 21 January 1998     

An Update on the Diagnosis and Assessment of Osteoporosis with Densitometry

In 1994 the WHO proposed guidelines for the diagnosis of osteoporosis based on measurement of bone mineral density. They have been widely used for epidemiological studies, clinical research and for treatment strategies. Despite the widespread acceptance of the diagnostic criteria, several problems remain with their use. Uncertainties concern the optimal site for assessment, thresholds for men and diagnostic inaccuracies at different sites. In addition, the development of many new technologies to assess the amount or quality of bone poses problems in placing these new tools within a diagnostic and assessment setting. This review considers the recent literature that has highlighted the strengths and weaknesses of diagnostic thresholds and their use in the assessment of fracture risk, and makes recommendations for actions to resolve these difficulties.     

Development and Evaluation of a Phantom for Morphometric X-ray Absorptiometry

Morphometric X-ray absorptiometry (MXA) provides the potential to assess vertebral deformity using a technique with much lower radiation dose to the patient than standard radiographic procedures. MXA overcomes many other limitations such as cone beam distortion observed in conventional plane radiographs. A phantom has been designed to assess the accuracy of the MXA technique, to monitor long-term precision and to assess inter- and intra-operator variability. The phantom consists of two columns of 12 cylinders representing the vertebral bodies, one of regular components and one representing vertebral deformities. Each column may be inserted, as required, into a Perspex torso-mimicking block. Initial assessment on the Lunar Expert-XL demonstrates that the phantom provides image parameters reflecting those found clinically. Measurement of vertebral height was found to be consistently underestimated by 4.9%. Operator precision ranged from 0.6% for posterior height measurement to 1.0% for middle height measurement of the regular component column. The corresponding precision range for the column representing vertebral deformation was 0.6% (posterior) and 1.1% (middle). Analysis of 10 scans of each column by two independent operators demonstrated a few significant differences in height assessment confined to the ‘thoracic’ region of the regular column. However, inter-operator variability was found to increase with increasing complexity of vertebral shape producing several differences, particularly in posterior height assessment of the deformed column. Received: 11 June 1997 / Accepted: 16 April 1998     

Comparison and Cross-Calibration of DXA Systems: ODX-240 and Sophos L-XRA versus Hologic QDR-4500, for Spinal Bone Mineral Measurement. Translation of a Reference Database

Replacement of dual-energy X-ray densitometry equipment may be necessary in time as a result of upgrading systems or new equipment. The lack of standardization in bone mineral density (BMD) measurements is known. Standardization efforts have been made for several years by the European Union under its organization COMAC-BME (Comité d'Actions Concertés–BioMedical Engineering) and by the International Committee for Standards in Bone Measurement. Cross-calibration is generally considered to be the result of linear regression between the measurements obtained with two densitometers. A major disadvantage of the regression method is the noncompatibility of the two formulae of calibration (Y versus X and X versus Y). In this study we considered cross-calibration in terms of a structural model that produced circular equations when, for example, three systems were cross-calibrated. Cross-calibration in this study was calculated from the measurement of the lumbar BMD of a population of 204 patients, with Hologic QDR-4500, ODX-240 and Sophos L-XRA systems. In vitro accuracy and short-term reproducibility of the three systems were studied. Using the structural calibration equation we transformed a reference database for L2–4 BMD obtained from a population of 983 French females, aged 11–47 years, on an ODX-240 to a reference database for a Hologic QDR-4500. A new young adult reference was obtained and consequently a new evaluation of the T-score for the Hologic QDR-4500. Received: 17 July 1997 / Revised: 9 March 1998     

Quantitative Ultrasound of the Tibia Depends on Both Cortical Density and Thickness

This study investigated whether tibial speed of sound (SOS; SoundScan 2000, Myriad Ultrasound Systems, Israel) reflects not only bone mineral density (BMD) but also tibial cortical thickness, as assessed by dual-energy X-ray absorptiometry (DXA) and Quantitative CT (QCT) at a site-matched location. The secondary focus of the study was how tibial SOS compares with BMD at the spine and the hip, the most widely used locations for densitometry. Twenty-two young normal (N) and 23 postmenopausal women with spinal fractures (Fx) (mean (SD) age 35 (8) and 70 (5) years) underwent quantitative ultrasound (QUS) SOS measurement at the left tibial midshaft. From site-matched QCT scans (three 3-mm slices spaced along the QUS measurement region), BMD and cortical thickness were computed (QCT-cBMD, QCT-cTh). The cortex in the CT images was then subdivided into three concentric and equally spaced bands, and QCT-cBMD was computed separately for each band. DXA was performed at the mid-tibia (TIB BMD), at the spine (SPINE BMD) and the hip (total hip, HIP BMD). Correlation coefficients between parameters were determined with least-square linear fits. Intergroup differences were assessed by analysis of covariance, whose r ² value reflects the percentage variation in the data explained by group assignment. SOS correlated significantly with site-matched parameters (QCT-cBMD, QCT-cTh and TIB BMD, all r= 0.6, p < 0.001), SPINE BMD and HIP BMD (both r= 0.5, p < 0.001). Multiple regression with both QCT-cBMD and QCT-cTh against SOS yielded r= 0.7 with both parameters contributing significantly. For the cortex band subdivision, SOS correlated better with QCT-cBMD in the outermost band of the cortex (r= 0.67) than with the more central bands (r= 0.59 and r= 0.53). Group assignment could best explain SPINE BMD (r ²= 0.62) and HIP BMD (r ²= 0.51). SOS was comparable to TIB BMD (r ²= 0.3 vs. r ²= 0.35).: Our findings suggest that the tibial SOS measurement depends on both the thickness and density of the tibia, but is more strongly influenced by the density of the cortex near the surface than by its interior parts. The power of tibial ultrasound to discriminate between normal and fracture patients was less than that of spinal and femoral DXA BMD and comparable to site-matched DXA BMD. Received: 25 February 2000 / Accepted: 5 July 2000     

Discordant Normal Ranges for Proximal Femur Bone Density in Australia

The aim of the study was to investigate the reference ranges currently in use for femoral neck bone mineral density (BMD) in Australia. Representative scans were obtained of the proximal femur from 78 dual-energy X-ray absorptiometry centers across Australia. Femoral neck BMD was standardized using published conversion equations and corresponding T-scores calculated using an Australian reference range (A_RR) and using the NHANES-III reference range (NH3_RR). Reported T-scores (T_R) from the DXA centers were compared with T-scores derived from the NH3_RR (T_NH) and from the A_RR (TA). Examination of the regression lines comparing T_R with T_NH, and T_R with T_A, demonstrated that the reference ranges used by centers with Lunar, and some of the Norland and Hologic instruments, differed only slightly from both the NH3_RR and the A_RR. A number of centers, however, with Norland or Hologic scanners, were using reference ranges which differed from both the NH3_RR and A_RR. The reference ranges used with these latter Hologic or Norland scanners could result in T-score differences of up to 1 standard deviation in the clinically relevant T-score range. The NH3_RR was compared with the A_RR using a set of BMD values covering the clinical range. There was close agreement between T_NH and corresponding T_A, confirming close agreement between the NH3_RR and A_RR . There are thus clinically important differences in the reference ranges currently used within Australia for interpretation of femoral neck BMD. For consistency, and in view of its validation, the NH3_RR is recommended as the most appropriate Caucasian normal range in Australia. The A_RR for femoral neck BMD does not differ significantly from the NH3_RR and is also acceptable. Problems relating to the use of discordant BMD reference ranges potentially may exist in other countries. Received: 22 August 2000 / Accepted: 1 February 2001     

Longitudinal Study of Bone Loss in Pre- and Perimenopausal Women: Evidence for Bone Loss in Perimenopausal Women

Bone loss before and around the time of menopause is not well characterized by longitudinal studies. We measured bone mineral density at various skeletal sites – total body, femoral neck, trochanter, anteroposterior (AP) and lateral spine, and forearm – with dual-energy X-ray absorptiometry in a large prospective cohort of 272 untreated pre- and perimenopausal women aged 31–59 years, at 1 year intervals for 3 years. Sex steroids and the following markers of bone remodeling were measured: serum osteocalcin (OC), procollagen I carboxyterminal extension peptide, bone alkaline phosphatase (BAP) and urinary crosslinks (CTX and NTX). Seventy-six women were classified as perimenopausal and 196 as premenopausal. Over the 3 years, premenopausal women had no significant bone loss at any site and a small but significant increase in bone mineral density at the trochanter, total hip, AP spine and radius. Perimenopausal women significantly lost bone from cancellous and cortical sites, i.e., the femoral neck, trochanter and lumbar spine. In perimenopausal women with increased follicle stimulating hormone, the rate of bone loss at the femoral neck correlated negatively with OC and BAP. In perimenopausal women, serum estradiol levels decreased during the 3 years of follow-up and bone loss from the trochanter and the AP spine was correlated with serum estradiol after 3 years. In conclusion, among premenopausal women there is no bone loss. In contrast, there is a rapid and diffuse bone loss in perimenopausal women, related to decreased estrogen secretion. Bone markers may be useful to identify these women losing bone. Received: 13 October 1997 / Accepted: 19 October 1998     

DXA of the Hip and Heel Ultrasound but not Densitometry of the Fingers Can Discriminate Female Hip Fracture Patients from Controls: A Comparison Between Four Different Methods

Dual-energy X-ray absorptiometry (DXA) of the proximal femur and in more recent years quantitative ultrasound (QUS) of the heel are the most established methods for assessing hip fracture risk. Measurement of the fingers offers a new approach. We performed DXA of the proximal femur, QUS of the heel and fingers, and radiographic absorptiometry (RA) of the fingers in 87 non-institutionalized women, 65–85 years of age, with a first hip fracture and compared them with 195 randomly selected age-matched controls. Bone mineral density (BMD) of the femoral neck and heel Stiffness Index were significantly lower among cases than among controls (by 15% and 17%, respectively; p<0.0001), whereas no significant differences were found for finger measurements. When applying the WHO criterion of osteoporosis, 62–98% of the patients were classified as osteoporotic, compared with 19–85% of the controls, depending on method and site. The risks of hip fracture, estimated as odds ratios for every 1 SD reduction in femoral neck BMD, heel Stiffness Index, finger QUS and finger RA, were: 3.6 (95% CI 2.4–5.5), 3.4 (95% CI 2.2–5.0), 1.0 (95% CI 0.7–1.3) and 1.2 (95% CI 0.8–1.6), respectively. Compared with women with normal BMD of the femoral neck, those classified as osteopenic had an odds ratio of hip fracture of 14 (95% CI 2-110), whereas those classified as osteoporotic had an odds ratio of 63 (95% CI 8–501). We conclude that hip DXA and heel QUS have similar capacities to discriminate the risk of a first hip fracture, whereas QUS and RA of the phalanges seem inferior techniques for differentiating female hip fracture patients from controls. Received: 10 March 2000 / Accepted: 21 September 2000     

The Contribution of Cortical and Cancellous Bone to Dual-Energy X-ray Absorptiometry Measurements in the Female Proximal Femur

Dual-energy X-ray absorptiometry (DXA) is the most common method for determining bone mineral density (BMD) in the proximal femur. However, there remain questions concerning the contribution of cortical and cancellous bone to this technology in the proximal femur. The purpose of this investigation was to identify structural and compositional characteristics of human bone in the proximal femur that significantly influence DXA BMD measurements. Twenty-four femora were obtained at autopsy from Caucasian females ranging in age from 17 to 92 years (mean ± SD, 61 ± 25 years). DXA scans were performed on each specimen with a Hologic QDR-2000 densitometer. Direct measurements were determined from proximal femoral sections for cancellous bone (volume fraction, ash fraction, cancellous cross-sectional area and percent cancellous cross-sectional area), cortical bone (thickness, ash fraction, porosity, cortical cross-sectional area and percent cortical cross-sectional area) and anteroposterior thickness. These parameters were compared with the associated DXA measurements by means of simple and multiple regressions. Cancellous volume fraction was the best predictor of variability of DXA measurements for both the neck and trochanter, with an R ² of 0.87 and 0.76, respectively (p<0.0001). There was only a minor influence of cortical factors such as thickness (neck and trochanter R ²= 0.51 and 0.42, respectively, p<0.001) and trochanteric cross-sectional area (R ²= 0.21, p<0.05). Although the accuracy for determining specific components of the proximal femur was low, the DXA BMD measurement was a strong predictor of cancellous bone factors, but not cortical bone factors that have been shown to change significantly with age. Received: 2 February 2000 / Accepted: 7 September 2000     

Standardization of Bone Mineral Density at Femoral Neck, Trochanter and Ward’s Triangle

The International Committee for Standards in Bone Measurement (ICSBM) has published standardization formulas for total hip bone mineral density (BMD). In many applications, however, BMD of hip subregions, such as femoral neck (FN), trochanter (TR), and Ward’s triangle (WT), are commonly measured. This paper addresses whether the standardization formulas for total hip BMD can be adequately used for hip subregions. We used data from 100 healthy women, from 20 to 80 years old, who had hip BMD measured in both the total hip and hip subregions by a Hologic QDR 2000, a Lunar DPX, and a Norland XR26 Mark II. The same women were used by ICSBM for the standardization of total hip BMD. In addition, we used data of 3139 patients from a clinical trial to validate our results. We derived standardization formulas for FN, TR, and WT using the same statistical method as that used for total hip BMD. We applied both total hip calibration formulas and subregion-specific formulas to the data and compared the corresponding effect. We found that the total hip calibration formulas can partially reduce BMD differences between the Hologic and Lunar as well as the Lunar and Norland scanners in hip subregions, but increase differences between the Hologic and Norland scanners. The subregion calibration formulas are most appropriate for calibrating subregion BMD values and their absolute changes, and should be adopted. Standardization is unnecessary for BMD percentage changes in our clinical trial data. Received: 9 September 1999 / Accepted: 11 October 1999     

Do Markers of Bone Resorption Add to Bone Mineral Density and Ultrasonographic Heel Measurement for the Prediction of Hip Fracture in Elderly Women? The EPIDOS Prospective Study

We have previously shown that hip bone mineral density (BMD), heel broadband ultrasound attenuation (BUA) and bone resorption markers are independent predictors of hip fracture in elderly women. We investigated whether a combination of these three parameters could improve the predictive value of a single test in a nested case–control analysis (75 hip fractures and 228 age-matched controls) of the EPIDOS prospective study comprising 7598 healthy women 75 years of age and older followed prospectively for a mean 22 months. At baseline, prior fracture, femoral neck BMD by dual-energy X-ray absorptiometry (DXA), heel BUA and urinary type I collagen C-telopeptide breakdown products (CTX) were assessed. The area under the receiver operating characteristic curve was significant for the three diagnostic tests, heel BUA being the best single predictor. The added value of urinary CTX to either BMD or BUA depends on the cutoff point chosen to define patients at risk and on the therapeutic strategy that is considered. Defining patients at risk as those with low BMD (or low BUA) or high CTX resulted in a significant increase in the sensitivity compared with BMD or BUA alone – a strategy that could be applied when a broad treatment is considered. However, this increased sensitivity was also obtained simply by increasing the BMD and BUA cutoffs, suggesting that a combination of CTX with BMD/BUA is not useful for that type of treatment strategy. Conversely, defining patients at risk as those with both low BMD and high CTX increases the specificity (88% vs 78%) with a similar number of hip fracture patients being identified (30% vs 32%) – a combination that could be useful when the strategy is to target treatment to a subset of high-risk patients. This strategy appears to be more cost-effective than bone mass measurement alone as indicated by the 37% fewer patients who need to be treated to avoid one fracture per year. If DXA or ultrasound is not available, the combination of a bone resorption marker with a history of any type of fracture after the age of 50 years gave a predictive value similar to that obtained with femoral neck BMD or heel BUA alone, for both types of treatment strategy. We conclude that the combination of urinary CTX with hip BMD could be useful for the identification of elderly women at high risk for hip fracture, resulting in higher specificity for a given sensitivity threshold than BMD measurement alone. If DXA is not available, the combination of history of fracture and urinary CTX performs as well as hip BMD to assess hip fracture risk in elderly women. Received: 24 November 1997 / Revised: 3 March 1998     

Image Analysis of the Distal Radius Trabecular Network Using Computed Tomography

Bone texture analysis might provide information about bone structure in a noninvasive manner. In a prospective case–control cross-sectional study we investigated the value of computed tomography (CT) image analysis of the distal radius in the assessment of osteoporosis. Twenty patients suffering from postmenopausal osteoporosis were studied and compared with 21 age-matched controls. Eight slices were selected in each patient: four consecutive coronal slices and four consecutive cross-sectional slices. Bone texture analysis was performed using statistical, fractal and structural methods leading to the measurement of 32 features. Structural variables derived from histomorphometric parameters were measured after segmentation from a binary or a skeletonized image. Bone mineral density was measured by dual-energy X-ray absorptiometry both at the lumbar spine and the femoral neck. Eight of the 9 statistical features were significantly different in osteoporotic women as compared with controls (coronal slices, p < 0.05). Seven structural variables were statistically different between the two groups on coronal slices (p < 0.05): valley surface area, bone volume/tissue volume, trabecular partition, Euler’s number, trabecular bone pattern factor, node-to-node strut count and terminus-to-terminus strut count. The most significant results on coronal slices (p < 0.01) concerned 4 structural features: trabecular partition, Euler’s number, trabecular bone pattern factor and terminus-to-terminus strut count. Three features were statistically different (p < 0.01) between the two groups on cross-sectional slices (skeletonization from gray levels). A few features yielded by texture analysis were correlated with both lumbar spine and femoral neck bone mineral density, but the level of these correlations was weak (r < 0.5). In conclusion, CT image analysis of the distal radius is a useful tool for characterizing bone texture alterations in osteoporotic women. These findings are in keeping with microarchitectural osteoporosis-related changes diagnosed on bone biopsies. Received: 8 April 1998 / Accepted: 14 September 1998     

Effects of Cigarette-Smoking on Bone Mass as Assessed by Dual-Energy X-ray Absorptiometry and Ultrasound

In order to elucidate the influence of nicotine smoking on bone mass in elderly women, bone mass was cross-sectionally assessed by dual energy X-ray absorptiometry (DXA) in total body, hip and lumbar spine, as well as with ultrasound of calcaneus and phalanges of the hand. Subjects were 1,042, 75-year old women, recruited on a population basis (Osteoporosis Prospective Risk Assessment (OPRA) study). We found bone mineral density (BMD) to be lower in hip (0.71 vs. 0.76 g/cm², p<0.0001 for femoral neck) and total body (0.96 vs. 1.02 g/cm², p<0.0001) in current smokers compared to never-smokers. There was no difference in BMD of the lumbar spine between current smokers and never-smokers. Bone mass as assessed by ultrasound of the calcaneus was lower for speed of sound (p<0.01), broadband ultrasound attenuation (p<0.0001) and stiffness (p<0.0001) in current smokers than in never-smokers. No differences were found for ultrasound measurements of the phalanges between smokers and never-smokers. Also, weight and current physical activity as assessed by a questionnaire differed significantly between current smokers and never-smokers.  There was no evident difference between former smokers and never-smokers in any of the skeletal regions assessed by DXA or ultrasound.  After correcting for differences in weight and physical activity, current smokers had lower BMD in all hip sites (p<0.05) and total body (p<0.01) compared to never-smokers. Ultrasound and BMD spine did not differ between these two groups after correction for weight and physical activity.  We conclude that nicotine smoking has a negative influence on bone mass independent of differences in weight and physical activity. This difference is detected by DXA but not by ultrasound measurements of the calcaneus or the phalanges. The present data are encouraging since no bone mass differences were found between former and never-smokers. Received: 29 March 2002 / Accepted: 2 July 2002