A. Mathur

Correlation of Trabecular Bone Structure with Age, Bone Mineral Density, and Osteoporotic Status: In Vivo Studies in the Distal Radius Using High Resolution Magnetic Resonance Imaging

High resolution magnetic resonance (MR) images of the distal radius were obtained at 1.5 Tesla in premenopausal normal, postmenopausal normal, and postmenopausal osteoporotic women. The image resolution was 156 μm in plane and 700 μm in the slice direction; the total imaging time was ∼16 minutes. An intensity‐based thresholding technique was used to segment the images into trabecular bone and marrow, respectively. Extensions of standard stereological techniques were used to derive measures of trabecular bone structure from these segmented images. The parameters calculated included apparent measures of trabecular bone volume fraction, trabecular thickness, trabecular spacing, and trabecular number. Fractal‐based texture parameters, such as the box‐counting dimension, were also derived. Trabecular bone mineral density (BMD) and cortical bone mineral content (BMC) were measured in the distal radius using peripheral quantitative computed tomography (pQCT). In a subset of patients, spinal trabecular BMD was measured using quantitative computed tomography (QCT). Correlations between the indices of trabecular bone structure measured from these high‐resolution MR images, age, BMD, and osteoporotic fracture status were examined. Cortical BMC and trabecular BMD at the distal radius, spinal BMD, trabecular bone volume fraction, trabecular thickness, trabecular number, and fractal dimension all decreased with age. Trabecular spacing showed the greatest percentage change and increased with age. In addition, significant differences were evident in spinal BMD, radial trabecular BMD, trabecular bone volume fraction, trabecular spacing, and trabecular number between the postmenopausal nonfracture and the postmenopausal osteoporotic subjects. Trabecular spacing and trabecular number showed moderate correlation with radial trabecular BMD but correlated poorly with radial cortical BMC. High resolution MR imaging, a potentially useful tool for quantifying trabecular structure in vivo, may have applications for understanding and evaluating skeletal changes related to age and osteoporosis.

Comparisons of Noninvasive Bone Mineral Measurements in Assessing Age‐Related Loss, Fracture Discrimination, and Diagnostic Classification

The purpose of this study was to examine the commonly available methods of noninvasively assessing bone mineral status across three defined female populations to examine their interrelationships, compare their respective abilities to reflect age‐ and menopause‐related bone loss, discriminate osteoporotic fractures, and classify patients diagnostically. A total of 47 healthy premenopausal (age 33 ± 7 years), 41 healthy postmenopausal (age 64 ± 9 years), and 36 osteoporotic postmenopausal (age 70 ± 6 years) women were examined with the following techniques: (1) quantitative computed tomography of the L1–L4 lumbar spine for trabecular (QCT TRAB BMD) and integral (QCT INTG BMD) bone mineral density (BMD); (2) dual X‐ray absorptiometry of the L1–L4 posterior‐anterior (DXA PA BMD) and L2–L4 lateral (DXA LAT BMD) lumbar spine, of the femoral neck (DXA NECK BMD) and trochanter (DXA TROC BMD), and of the ultradistal radius (DXA UD BMD) for integral BMD; (3) peripheral QCT of the distal radius for trabecular BMD (pQCT TRAB BMD) and cortical bone mineral content (BMC) (pQCT CORT BMC); (4) two radiographic absorptiometric techniques of the metacarpal (RA METC BMD) and phalanges (RA PHAL BMD) for integral BMD; and (5) two quantitative ultrasound devices (QUS) of the calcaneus for speed of sound (SOS CALC) and broadband ultrasound attenuation (BUA CALC). In general, correlations ranged from (r = 0.10−0.93) among different sites and techniques. We found that pQCT TRAB BMD correlated poorly (r ≤ 0.46) with all other measurements except DXA UD BMD (r = 0.62, p ≤ 0.0001) and RA PHAL BMD (r = 0.52, p ≤ 0.0001). The strongest correlation across techniques was between QCT INT BMD and DXA LAT BMD (r = 0.87, p ≤ 0.0001), and the weakest correlation within a technique was between pQCT TRAB BMD and pQCT CORT BMC (r = 0.25, p ≤ 0.05). Techniques showing the highest correlations with age in the healthy groups also showed the greatest differences among groups. They also showed the best discrimination (as measured by the odds ratios) for the distinction between healthy postmenopausal and osteoporotic postmenopausal groups based on age‐adjusted logistic regression analysis. For each anatomic site, the techniques providing the best results were: (1) spine, QCT TRAB BMD (annual loss, −1.2% [healthy premenopausal and healthy postmenopausal]); Students t‐value [not the T score], 5.4 [healthy postmenopausal vs. osteoporotic postmenopausal]; odds ratio, 4.3 [age‐adjusted logistic regression for healthy postmenopausal vs. osteoporotic postmenopausal]); (2) hip, DXA TROC BMD (−0.46; 3.5; 2.2); (3) radius, DXA UD BMD (−0.44; 3.3; 1.9) and pQCT, CORT BMC (−0.72; 2.9; 1.7); (4) hand, RA PHAL (−0.51; 3.6; 2.0); and (5) calcaneus, SOS (−0.09; 3.4; 2.1) and BUA (−0.52; 2.6; 1.7). Despite these performance trends, the differences among sites and techniques were statistically insignificant (p > 0.05) using age‐adjusted receiver operating characteristic (ROC) curve analysis. Nevertheless, kappa score analysis (using −2.0 T score as the cut‐off value for osteopenia and −2.5 T score for osteoporosis) showed that in general the diagnostic agreement among these measurements in classifying women as osteopenic or osteoporotic was poor, with kappa scores averaging about 0.4 (exceptions were QCT TRAB/INTG BMD, DXA LAT BMD, and RA PHAL BMD, with kappa scores ranging from 0.63 to 0.89). Often different patients were estimated at risk by using different measurement sites or techniques.

Volumetric quantitative computed tomography of the proximal femur : Precision and relation to bone strength

We have developed a three-dimensional computed tomography (CT) scanning and image analysis method for measurement of trabecular and integral bone mineral density (BMD) and geometry in automatically determined femoral-neck and trochanteric subregions of the proximal femur. We measured the correlation of the density and geometry variables to femoral strength assessed in vitro under loading simulating a single-limb condition and a fall to the side. While BMD alone accounted for 48%-77% of the variability in strength for the stance loading configuration, femoral neck cross-sectional area (minCSA) and femoral neck axis length (FNAL) also contributed independently to femoral strength, and a combination of BMD and geometry variables explained 87%-93% of the variance in the data. For the fall loading configuration, trochanteric trabecular BMD alone explained 87% of the variability of strength. The reproducibility in vivo of the technique was assessed in a group of seven postmenopausal women, who underwent repeat scans with repositioning. For trabecular BMD, the precision was 1.1% and 0.6% for the femoral neck and trochanteric subregions, respectively, compared to 3.3% and 1.6% for the corresponding integral envelopes. Thus, trabecular BMD measurements were reproducible and highly correlated to biomechanical strength measurements. These results support further exploration of quantitative CT for assessment of osteoporosis at the proximal femur.

A Comparative Study of Trabecular Bone Properties in the Spine and Femur Using High Resolution MRI and CT

The purpose of this study was to use high resolution (HR) magnetic resonance (MR) and computed tomography (CT) images combined with texture analysis to investigate the trabecular structure of human vertebral and femoral specimens and to compare these techniques with bone mineral density (BMD) in the prediction of bone strength. Twenty‐nine bone cubes were harvested from 12 proximal femur cadaver specimens and 29 from 8 spines. HR MR and CT images were obtained, and texture analysis techniques were used to assess trabecular structure. Additionally, BMD, elastic modulus (EM), and maximum compressive strength were determined. R2 for EM versus texture measures computed in the MR images was higher (R2 = 0.27–0.64, p < 0.01) in the spine than in the femur specimens (R2 = 0.12–0.22, p < 0.05). R2 values were similar in the CT images. R2 for EM versus BMD was 0.66 (p < 0.01) in the spine and 0.61 (p < 0.01) in the femur specimens. In the MR images, texture measures combined with BMD in a multivariate‐regression model significantly increased R2, while improvement was less significant in the CT images. Thus, texture analysis may provide additional information needed to analyze bone strength and quality.

Magnetic resonance imaging of trabecular bone structure in the distal radius: Relationship with X-ray tomographic microscopy and biomechanics

The contribution of trabecular bone structure to bone strength is of considerable interest in the study of osteoporosis and other disorders characterized by changes in the skeletal system. Magnetic resonance (MR) imaging of trabecular bone has emerged as a promising technique for assessing trabecular bone structure. In this in vitro study we compare the measures of trabecular structure obtained using MR imaging and higher-resolution X-ray tomographic microscopy (XTM) imaging of cubes from human distal radii. The XTM image resolution is similar to that obtained from histomorphometric sections (18 µm isotropic), while the MR images are obtained at a resolution comparable to that achievable in vivo (156×156×300 µm). Standard histomorphometric measures, such as trabecular bone area fraction (synonymous with BV/TV), trabecular width, trabecular spacing and trabecular number, texture-related measures and three-dimensional connectivity (first Betti number/volume) of the trabecular network have been derived from these images. The variation in these parameters as a function of resolution, and the relationship between the structural parameters, bone mineral density and the elastic modulus are also examined. In MR images, because the resolution is comparable to the trabecular dimensions, partial volume effects occur, which complicate the segmentation of the image into bone and marrow phases. Using a standardized thresholding criterion for all images we find that there is an overestimation of trabecular bone area fraction (∼3 times), trabecular width (∼3 times), fractal dimension (∼1.4 times) and first Betti number/ volume (∼10 times), and an underestimation of trabecular spacing (∼1.6 times) in the MR images compared with the 18-µm XTM images. However, even for a factor of 9 difference in spatial resolution, the differences in the morphological trabecular structure measures ranged from a factor of 1.4 to 3.0. We have found that trabecular width, area fraction, number, fractal dimension and Betti number/volume measured from the XTM and MR images increases, while trabecular spacing decreases, as the bone mineral density and elastic modulus increase. A preliminary bivariate analysis showed that in addition to bone mineral density alone, the Betti number, trabecular number and spacing contributed to the prediction of the elastic modulus. This preliminary study indicates that measures of trabecular bone structure using MR imaging may play a role in the study of osteoporosis.

Assessment of Trabecular Structure Using High Resolution Ct Images and Texture Analysis

PURPOSE Our goal was to use high resolution (HR) CT images combined with texture analysis to investigate the trabecular structure of human vertebral specimens and to compare these techniques with bone mineral density (BMD) in the prediction of bone strength. METHOD HR CT images with a slice thickness of 1 mm were obtained of 28 bone cubes. Four different groups of texture analysis techniques were used to assess these images. In addition, quantitative CT (QCT) was performed and elastic modulus (EM) was determined biomechanically. RESULTS R2 between EM and BMD was 0.78 (p < 0.01). R2 values for EM versus most of the texture measures were also significant. Texture measures in addition to measures of BMD in a multivariate regression model significantly increased R2 up to 0.87. CONCLUSION In an experimental setting, texture parameters calculated using HR CT images correlated significantly with EM. Combining texture measures with BMD improved the prediction of EM significantly.

Comparative calibration without a gold standard.

Comparative calibration is the broad statistical methodology used to assess the calibration of a set of p instruments, each designed to measure the same characteristic, on a common group of individuals. Different from the usual calibration problem, the true underlying quantity measured is unobservable. Many authors have shown that this problem, in general, does not have a unique solution. Most commonly used assumptions to obtain a unique solution are (i) one instrument is the gold standard (that is, unbiased) and (ii) the measurement errors of the p instruments are independent. Such constraints, however, may not be valid for many clinical applications, for example, the universal standardization project for dual X-ray absorptiometry (DXA) scanners. In this paper, we propose a new approach to resolve the comparative calibration problem when a gold standard is unavailable. Instead of the usual assumptions, we use external information in addition to data from the p instruments, to solve the problem. We address statistical estimation, hypothesis testing and missing data problems. We apply the new method specifically to the universal standardization project data where a group of individuals have been measured for bone mineral density (BMD) by three DXA scanners. We compare the results of the new method to currently used methods and show that they have better statistical properties.

Survival tree analysis for fracture risk

In the mexican group osteopenia was found in 33% of lumbar spine determinations and 35% of femoral necks studied. Osteoporosis was found 18% and .p% respectively. Duration of breast feeding and years after menopause were negasve correlated with BMD at lumbar spine and years after menopause and age were negative correlated with BMD al femoral neck in both countries. Time of estrogen use was also negatively correlated with BMD at lumbar spine in the mexican gmt~p. Number of pregnancies and alcohol consumption were negatively correlated with BMD at lumbar spine and femoral neck in the venezuelan group. CONCLUSION. Our results define a risk profile and are useful in decision making at the clinical practice as well as to characterize preventive intervention in patients and populations.