Radiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Link, T. M.
Right arrow Articles by Majumdar, S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Link, T. M.
Right arrow Articles by Majumdar, S.
(Radiology. 2000;217:855-862.)
© RSNA, 2000

Musculoskeletal Imaging

Changes in Calcaneal Trabecular Bone Structure after Heart Transplantation: An MR Imaging Study1

Thomas M. Link, MD, Albrecht Lotter, PhD, Florian Beyer, MS, Stefan Christiansen, MD, David Newitt, PhD, Ying Lu, PhD, Christof Schmid, MD and Sharmila Majumdar, PhD

1 From the Departments of Clinical Radiology (T.M.L., A.L., F.B.) and Cardiac and Thoracic Surgery (S.C., C.S.), University of Muenster, Germany; and the Department of Radiology, University of California, San Francisco (T.M.L., D.N., Y.L., S.M.). From the 1999 RSNA scientific assembly. Received November 4, 1999; revision requested December 17; final revision received March 22, 2000; accepted April 20. Supported by German Research Society grant LI 710, 2-1. Address correspondence to T.M.L., Department of Radiology, Klinikum rechts der Isar, Technical University of Munich, Ismaninger Str. 22, D 81675, Germany (e-mail: tmlink@roe.med.tu-muenchen.de).


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
PURPOSE: To use high-spatial-resolution magnetic resonance (MR) imaging to analyze the trabecular bone structure of the calcaneus in patients before and after heart transplantation and to compare this technique with bone mineral density (BMD) measurement in predicting therapy-induced bone loss and vertebral fracture status.

MATERIALS AND METHODS: High-spatial-resolution 1.5-T MR imaging of the calcaneus was performed in 40 men 11–120 months after heart transplantation, in 11 men before heart transplantation, and in 10 age-matched male volunteers. Sagittal and transverse T1-weighted spin-echo images with a voxel size of 0.195 x 0.195 x 1.000 mm were obtained, and structure measurements analogous to bone histomorphometric values were calculated. In addition, the BMD of the lumbar spine was determined in the transplant recipients pre- and postoperatively by using quantitative computed tomography, and vertebral fracture status was assessed.

RESULTS: Significant differences in structure and BMD measurements were found between patients before and after heart transplantation (P <. 05). In 17 (42%) of 40 transplant recipients, vertebral fractures were found. Although structure measurements were significantly different between patients with and those without fractures (P < .05), BMDs were not. Correlations between time after transplantation and some structure measurements were moderately significant (P <. 05), but such correlations with BMD measurements were not.

CONCLUSION: MR imaging–derived structure measurements in the calcaneus are useful for monitoring bone changes after heart transplantation and assessing vertebral fracture status.

Index terms: Bones, absorptiometry • Bones, CT, 4642.12111 • Bones, effects of drugs on, 4642.419, 4642.569 • Bones, fractures, 4642.419 • Bones, MR, 4642.121411, 4642.12146 • Heart, transplantation, 51.459 • Osteoporosis, 4642.569


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Increased longevity after heart transplantation is due in part to improved immunosuppressive medications, which include cyclosporine A, prednisolone, and azathioprine. However, these medications are associated with the development of a form of osteoporosis that becomes clinically apparent within the 1st few years after transplantation. Study results have shown osteoporotic fractures in up to 50% of transplant recipients (1–5). Identification of patients at risk for fractures would permit preventive treatment (6). One of the most important parameters in assessing fracture risk is bone mineral density (BMD). However, Lee et al (1) found limitations in the use of BMD measurements for differentiating patients before and after heart transplantation, and Shane et al (5) observed a substantial overlap of BMD measurements in heart transplant recipients with and without osteoporotic fractures.

The results of several studies (7–10) have indicated that in addition to BMD, trabecular architecture is equally important in assessing bone strength. To our knowledge, however, up to now, trabecular bone structure has not been analyzed in a larger number of patients after heart transplantation. In immunosuppressed heart transplant recipients, invasive bone biopsy does not appear to be justified for research purposes, and to our knowledge, only one study (11) involving the analysis of transiliac bone biopsy, performed in six heart transplant recipients, has been published. Noninvasive assessment of trabecular architecture with a number of radiologic imaging techniques has been introduced, and some of these techniques involve the use of magnetic resonance (MR) imaging with high spatial resolution (12–18).

Thus, the purposes of this study were to (a) analyze trabecular bone structure, as seen on high-spatial-resolution MR images of the calcaneus in patients before and after heart transplantation and compare the bone structure with that seen on the MR images obtained in healthy volunteers, (b) assess differences in bone structure between transplant recipients with and those without vertebral spine fractures, and (c) compare these apparent structure measurements in the calcaneus with BMD of the lumbar spine.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Patients
Fifty-one white men before and after heart transplantation, as well as 10 age-matched healthy white male volunteers were prospectively examined in accordance with the regulations of the committee of human research at the University of Muenster. Informed consent was obtained from all patients and volunteers after the nature of the examinations had been fully explained.

Three groups were examined: 11 men who were scheduled for heart transplantation (mean age ± SD, 51.9 years ± 9.8; age range, 32–63 years), 40 men 11–120 months after heart transplantation (mean age ± SD, 56.0 years ± 10.8; age range, 24–76 years), and 10 healthy age-matched male volunteers (mean age ± SD, 52.1 years ± 11.5; age range, 34–66 years). The differences in age among the three groups were not statistically significant. All patients were ambulatory, had not experienced a period of inactivity longer than that related to the transplantation, and were able to walk freely without walkers or the help of others. None of the patients or volunteers received hormone replacement therapy, bisphosphonates, calcitonin, fluoride, or parathyroid hormones, and they had not received substantial amounts of glucocorticoids (ie, prednisolone dose >10 mg/day for >3 months) before transplantation. Paget disease, multiple myeloma, and malignant disease with metastatic tumor involvement were further exclusion criteria. Heart transplantation was due to ischemic heart disease in 22 patients, congestive cardiomyopathy in 14 patients, and valvular disease in four patients.

The glucocorticoid regimen intraoperatively and during the 1st days after heart transplantation consisted of doses of methylprednisolone and prednisolone as high as 100–500 mg per day but was rapidly decreased to doses of 10–30 mg. The long-term immunosuppressive regimen for all patients following the 1st months after transplantation consisted of prednisolone (10 mg/day), cyclosporine (serum levels, 200–300 ng/mL), and azathioprine (1–4 mg/kg/day with hemoglobin level >9 g/dL [90 g/L] and white blood cell count >1,000/mL [1 cell x 109/L]). Rejection was managed with high-dose intravenously or orally administered glucocorticoids and depended on the severity of rejection. Rejection occurred in five patients.

In all patients, we examined the calcaneus of the nondominant leg, or, in cases of history of foot or ankle trauma, the calcaneus of the contralateral leg. The fracture status of the thoracic and lumbar spine was assessed in the patients before and after heart transplantation by using the spinal fracture index, as previously described by Genant et al (19). In addition, serum calcium, phosphate, alkaline phosphatase, and creatinine levels were measured by using standard autoanalyzer techniques. BMD measurements were obtained in only the heart transplantation recipients pre- and postoperatively. The healthy volunteers did not undergo BMD measurement, because there are well-established normal BMD data on age-matched male and female patients and the ethical committee at the University of Muenster would not approve this radiation exposure in healthy volunteers.

MR Imaging
Transverse and sagittal high-spatial-resolution MR images of the calcaneus with a section thickness of 1 mm were obtained by using a 1.5-T MR imaging unit (Vision; Siemens, Erlangen, Germany) equipped with 22 mT/m gradients and a flexible surface coil (Flex Small; Siemens). A two-dimensional spin-echo sequence was used with the following parameters: 450/14 (repetition time msec/echo time msec), two acquisitions, an acquisition time of 5 minutes, 48 seconds, a 384 x 512 matrix, and a 7.5 x 10.0-cm field of view to yield a pixel size of 0.195 x 0.195 mm2. When the calcaneus was large, the field of view had to be adjusted to 10 x 10 cm, which increased the imaging time to 8 minutes but did not change the spatial resolution.

Fifteen sections were acquired in a standardized manner: To determine the position of the sagittal images, the transverse localizer image of the calcaneus was divided into three parts, and the sections were aligned parallel to the center points of the lines between the posterior and middle parts of the calcaneus and between the middle and anterior third of the calcaneus, as shown in Figure 1a. The chosen alignment of the transverse sections was parallel to the caudal border of the anterior part of the calcaneus and adjacent to the talar anterior articular surface of the calcaneus, as shown in Figure 1b.



View larger version (135K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1a. (a) Transverse, two-dimensional, T1-weighted spin-echo MR image (450/14) of the calcaneus in a 50-year-old man 3 years after transplantation shows the location of the sagittal images used for analysis. The two shorter lines perpendicular to the axis of the calcaneus represent the border lines between the posterior and middle parts of the calcaneus and between the middle and anterior parts of the calcaneus. The two longer lines along the axis of the calcaneus are aligned parallel to the center points of the short lines and represent the orientation and area of the sagittal sections. (b) Corresponding sagittal MR image (450/14) shows the location of the transverse images used for analysis. The top line was placed adjacent to the talar anterior articular surface of the calcaneus, and the bottom line was aligned parallel to the caudal border of the anterior part of the calcaneus. Thus, the orientation as well as the inferior and superior borders of the sections are defined.

 


View larger version (147K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1b. (a) Transverse, two-dimensional, T1-weighted spin-echo MR image (450/14) of the calcaneus in a 50-year-old man 3 years after transplantation shows the location of the sagittal images used for analysis. The two shorter lines perpendicular to the axis of the calcaneus represent the border lines between the posterior and middle parts of the calcaneus and between the middle and anterior parts of the calcaneus. The two longer lines along the axis of the calcaneus are aligned parallel to the center points of the short lines and represent the orientation and area of the sagittal sections. (b) Corresponding sagittal MR image (450/14) shows the location of the transverse images used for analysis. The top line was placed adjacent to the talar anterior articular surface of the calcaneus, and the bottom line was aligned parallel to the caudal border of the anterior part of the calcaneus. Thus, the orientation as well as the inferior and superior borders of the sections are defined.

 
Structure Analysis
On the transverse and sagittal images, the 10 central sections were analyzed by using a posterior circular region of interest (ROI) on the images (Fig 2). On the transverse images, the posterior ROI was chosen to be the largest possible circle, which fit the inner perimeter of the posterior portion of the calcaneus on the central section (section 5). Then, all 10 central sections were checked, and on all sections the circle was reduced in size to exclude cortical bone. On the sagittal images, again the central section was used to determine the initial size of the ROI, which was fit to the inner rim of cortical bone in the posterior part of the calcaneus, and then the ROI diameter was reduced to exclude cortical bone on the other sections. To improve reproducibility, all ROIs were placed by one of the authors (F.B.). The signal intensity homogeneity on the images was assessed by using a homogeneous oil phantom. The mean signal intensity (± SD) in the center of a ROI on the oil phantom that corresponded to the posterior calcaneal ROI on the sagittal and transverse images was 3,687 (± 310), and the mean signal intensity at the periphery of the ROI closest to the surface coil was 3,806 (± 272). Thus, given the small variation in signal intensity, coil correction algorithms were not used.



View larger version (132K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2a. (a) Transverse, two-dimensional, T1-weighted spin-echo MR image (450/14) of the calcaneus in the same patient as in Figure 1 and (b) corresponding sagittal MR image (450/14) show the circular region of interest in the posterior part of the calcaneus.

 


View larger version (146K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2b. (a) Transverse, two-dimensional, T1-weighted spin-echo MR image (450/14) of the calcaneus in the same patient as in Figure 1 and (b) corresponding sagittal MR image (450/14) show the circular region of interest in the posterior part of the calcaneus.

 
Image analysis was performed by first binarizing the high-spatial-resolution MR images according to a dual-threshold algorithm, as previously described by Majumdar et al (14): A histogram-based approach, in which bone and marrow phases were defined in an empirical model, was used. Structure measurements were obtained in the ROIs. All bone structure measurements were calculated on SUN/SPARC workstations (Sun Microsystems, Mountain View, Calif) with software developed by using an interactive display language (Research Systems, Boulder, Colo) interface. The average time to perform the structure analysis in an individual patient or volunteer was approximately 15 minutes. Analysis was blinded—that is, without knowledge of the transplantation status, age, or other patient data. By using images of a similar spatial resolution, an in vivo precision of 2%–6% was achieved for calcaneus measurements in a previous study (20).

Morphologic Parameters
Parameters synonymous to those obtained from standard histomorphometry (21,22) were determined. Because histomorphometric analysis is performed at a much higher spatial resolution (5 µm), the parameters determined on the MR images were defined as apparent structure measurements. The total number of bright pixels contributing to the bone phase on the binarized image relative to the total number of pixels in the region of interest was used to compute the apparent bone volume to total volume ratio (BV/TV). The total number of black and white pixel edges that crossed a set of parallel rays at a given angle {Theta} through the image were counted, and then a measurement of the mean intercept length was computed as the ratio between the total area of the bright pixels and half the number of edges. The mean value of the intercept length for all angles was the width of the bright pixels and was defined as the apparent trabecular thickness. From measurements of the apparent BV/TV and apparent trabecular thickness, two other morphologic parameters were determined: the apparent trabecular number, which is equal to the area fraction of bright pixels divided by the apparent trabecular thickness, and the apparent trabecular separation, which is equal to the quotient of 1 divided by the apparent trabecular number, minus the apparent trabecular thickness.

Fractal Dimension
Fractal dimension was obtained from the binarized images by using a box-counting algorithm, as previously described by Ishida et al (23), Majumdar et al (24), and Waldt et al (25). A grid consisting of a box of a given size {varepsilon} was superimposed on the boundary of the trabecular bone network to be quantified. The number of boxes of {varepsilon} that contained the boundary points, N({varepsilon}), was computed. This procedure was repeated for different {varepsilon}, with the sides ranging from 2 to 128 pixels. The logarithm of {varepsilon} versus the logarithm of N({varepsilon}) was plotted. The linear portion of this curve was identified, and the fractal dimension D was defined as the negative slope of the curve.

BMD Measurements
Quantitative computed tomography (CT) (Tomoscan LX; Philips Medical Systems, Best, the Netherlands) of the L2 through L4 vertebrae was performed in all transplant recipients pre- and postoperatively by using a solid Cann-Genant calibration phantom (Image Analysis, Columbia, Ky). Midvertebral images of the L2 through L4 vertebrae were obtained with a collimation of 10 mm and an exposure dose of 200 mAs and 120 kVp. Trabecular BMD was measured by using an elliptical ROI. All ROIs were placed by radiologic technicians and checked by one of the co-authors (T.M.L.), who supervises BMD analysis.

Data Analysis
The means and SDs of structure measurements were calculated in the healthy volunteers, pre- and postoperatively in the heart transplant recipients, and postoperatively in the heart transplant recipients with and without vertebral fractures. The means and SDs of BMD measurements were determined in only the heart transplant recipients pre- and postoperatively. Differences between pairs of individual groups were calculated by using the Student t test. The Bonferroni method was used to adjust for the effect of multiple comparisons. To remove age differences between the two cohorts of heart transplant recipients—those with and without fractures—we compared the structure measurements by using an analysis of covariance test at the 5% significance level, with age as a covariate. Although such an adjustment may not completely remove age as a confounding factor, it at least reduced the confounding effect of age in this analysis.

Correlations between structure and BMD measurements and between age and these measurements were assessed by using Pearson correlation coefficients and two-tailed Student t tests of significance (26). The number of months after heart transplantation versus the structure and BMD measurements was correlated, with adjustments made for patient age. In addition, receiver operating characteristic analyses of the BMD and structure measurements were performed to estimate the power of these assessments in predicting vertebral fractures (27). Combined area under the curve values for structure and BMD measurements were calculated by using logistic regression analysis, with adjustments made for age. All statistical computations were processed by using STATVIEW, version 4.1 (Abacus, Berkeley, Calif), and JMP (SAS Institute, Cary, NC) software. Receiver operating characteristic analysis was performed by using S-plus, 4.5, software (MathSoft, Seattle, Wash).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 
Figures 3 and 4 are representative images obtained in patients before and after heart transplantation and in two healthy volunteers. The images depict the differences in the appearance of solidity of the trabecular network among the three individual groups. The marrow spaces between the trabeculae appeared to be widest on the images obtained in the patients after heart transplantation. Furthermore, the trabecular number was lower on the images obtained in the patients after heart transplantation compared with that on the images obtained in the patients before heart transplantation and in the healthy volunteers.



View larger version (155K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3a. Transverse, two-dimensional, T1-weighted spin-echo MR images (450/14) of the calcaneus (voxel size, 0.195 x 0.195 x 1 mm) in (a) a 65-year-old healthy man (apparent BV/TV, 0.47; apparent trabecular separation, 0.28 mm), (b) a 60-year-old man before cardiac transplantation (apparent BV/TV, 0.32; apparent trabecular separation, 0.44 mm; BMD, 177 mg/cm3), and (c) a 56-year-old man 96 months after heart transplantation with no history of fractures (apparent BV/TV, 0.25; apparent trabecular separation, 0.570 mm; BMD, 106 mg/cm3). In a, the thickest trabecular structure, with the highest bone fraction and smallest marrow spaces between the individual trabeculae in the posterior part of the calcaneus, is seen. In b, the trabeculae are more scarce and not as numerous. In c, increasingly more scarce trabeculae, with the lowest bone fraction and largest marrow spaces between the trabeculae, are seen.

 


View larger version (181K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3b. Transverse, two-dimensional, T1-weighted spin-echo MR images (450/14) of the calcaneus (voxel size, 0.195 x 0.195 x 1 mm) in (a) a 65-year-old healthy man (apparent BV/TV, 0.47; apparent trabecular separation, 0.28 mm), (b) a 60-year-old man before cardiac transplantation (apparent BV/TV, 0.32; apparent trabecular separation, 0.44 mm; BMD, 177 mg/cm3), and (c) a 56-year-old man 96 months after heart transplantation with no history of fractures (apparent BV/TV, 0.25; apparent trabecular separation, 0.570 mm; BMD, 106 mg/cm3). In a, the thickest trabecular structure, with the highest bone fraction and smallest marrow spaces between the individual trabeculae in the posterior part of the calcaneus, is seen. In b, the trabeculae are more scarce and not as numerous. In c, increasingly more scarce trabeculae, with the lowest bone fraction and largest marrow spaces between the trabeculae, are seen.

 


View larger version (169K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3c. Transverse, two-dimensional, T1-weighted spin-echo MR images (450/14) of the calcaneus (voxel size, 0.195 x 0.195 x 1 mm) in (a) a 65-year-old healthy man (apparent BV/TV, 0.47; apparent trabecular separation, 0.28 mm), (b) a 60-year-old man before cardiac transplantation (apparent BV/TV, 0.32; apparent trabecular separation, 0.44 mm; BMD, 177 mg/cm3), and (c) a 56-year-old man 96 months after heart transplantation with no history of fractures (apparent BV/TV, 0.25; apparent trabecular separation, 0.570 mm; BMD, 106 mg/cm3). In a, the thickest trabecular structure, with the highest bone fraction and smallest marrow spaces between the individual trabeculae in the posterior part of the calcaneus, is seen. In b, the trabeculae are more scarce and not as numerous. In c, increasingly more scarce trabeculae, with the lowest bone fraction and largest marrow spaces between the trabeculae, are seen.

 


View larger version (155K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4a. Sagittal, two-dimensional, T1-weighted spin-echo MR images (450/14) of the calcaneus (voxel size, 0.195 x 0.195 x 1 mm) in (a) a 40-year-old healthy man (apparent BV/TV, 0.44; apparent trabecular separation, 0.38 mm), (b) a 50-year-old man before cardiac transplantation (apparent BV/TV, 0.37; apparent trabecular separation, 0.41 mm; BMD, 119 mg/cm3), and (c) a 55-year-old man 12 months after heart transplantation with three vertebral fractures (apparent BV/TV, 0.20; apparent trabecular separation, 0.842 mm; BMD, 50 mg/cm3). In a, the thickest and most complex trabecular structure is seen. In b, the trabecular architecture is less complex and the individual trabeculae are thinner. In c, the trabeculae are more scarce and the marrow spaces between the trabeculae are the widest.

 


View larger version (155K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4b. Sagittal, two-dimensional, T1-weighted spin-echo MR images (450/14) of the calcaneus (voxel size, 0.195 x 0.195 x 1 mm) in (a) a 40-year-old healthy man (apparent BV/TV, 0.44; apparent trabecular separation, 0.38 mm), (b) a 50-year-old man before cardiac transplantation (apparent BV/TV, 0.37; apparent trabecular separation, 0.41 mm; BMD, 119 mg/cm3), and (c) a 55-year-old man 12 months after heart transplantation with three vertebral fractures (apparent BV/TV, 0.20; apparent trabecular separation, 0.842 mm; BMD, 50 mg/cm3). In a, the thickest and most complex trabecular structure is seen. In b, the trabecular architecture is less complex and the individual trabeculae are thinner. In c, the trabeculae are more scarce and the marrow spaces between the trabeculae are the widest.

 


View larger version (159K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4c. Sagittal, two-dimensional, T1-weighted spin-echo MR images (450/14) of the calcaneus (voxel size, 0.195 x 0.195 x 1 mm) in (a) a 40-year-old healthy man (apparent BV/TV, 0.44; apparent trabecular separation, 0.38 mm), (b) a 50-year-old man before cardiac transplantation (apparent BV/TV, 0.37; apparent trabecular separation, 0.41 mm; BMD, 119 mg/cm3), and (c) a 55-year-old man 12 months after heart transplantation with three vertebral fractures (apparent BV/TV, 0.20; apparent trabecular separation, 0.842 mm; BMD, 50 mg/cm3). In a, the thickest and most complex trabecular structure is seen. In b, the trabecular architecture is less complex and the individual trabeculae are thinner. In c, the trabeculae are more scarce and the marrow spaces between the trabeculae are the widest.

 
The means and SDs of the morphologic measurements and fractal dimensions are shown in Table 1. All of the measurements, except fractal dimension, were significantly different among the three groups (P < .05). With Bonferroni adjustment for multiple comparisons, however, the differences in apparent trabecular thickness between the pretransplantation and posttranplantation groups were no longer significant, whereas all other differences remained significant. As was expected, the apparent trabecular separation was highest in the posttransplantation group, whereas the apparent BV/TV, apparent trabecular thickness, and apparent trabecular number were lowest in this group. It should be noted that there were also significant differences in some of the structure measurements between the healthy volunteers and the pretransplantation group, whereas the mean age-matched normal reference BMD value (± SD) for 50-year-old men is 122.3 mg/mL (± 26.3) (28), which approximates the BMD value observed in the pretransplantation group in our study (119.2 mg/mL ± 32.8). The differences in BMD between the patients before and after heart transplantation, however, were significant (P < .05) (Table 1). The differences in serum calcium, phosphate, alkaline phosphatase, and creatinine levels between the patients before and after heart transplantation were not significant (P > .05).


View this table:
[in this window]
[in a new window]

  		TABLE 1. Morphologic Parameters, Fractal Dimensions, and BMDs of Patients before and after Heart Transplantation
 
With the exception of correlations with apparent trabecular numbers on the sagittal images, statistically significant but moderate correlations between the structure measurements of apparent BV/TV, apparent trabecular thickness, apparent trabecular separation, and apparent trabecular number and the BMD measurements of the lumbar spine were found in all patients (P < .05) (Table 2). The correlation between fractal dimension and BMD was not significant (P > .05).


View this table:
[in this window]
[in a new window]

  		TABLE 2. Correlations between Structure Measurements and Vertebral BMDs in 51 Patients before and after Heart Transplantation
 
In the healthy volunteers and the pretransplantation patient group, correlations between age and the structure measurements of apparent BV/TV and apparent trabecular thickness were moderately significant on both the sagittal and transverse images and correlations between age and apparent trabecular separation were moderately significant on the sagittal images (r = 0.40–0.59; P < .05). The correlation (r) between BMD and age in the pretransplantation group was 0.61 (P = .022).

When the apparent structure measurements versus the number of months after heart transplantation were correlated, moderate yet significant correlations were found for apparent BV/TV (r = -0.37; P = .03) and apparent trabecular thickness (r = -0.42; P = .012) on the transverse images. The time after transplantation was correlated with the BMD; however, this correlation was not statistically significant (P > .05).

Vertebral fractures were found in 17 (42%) of the 40 transplant recipients postoperatively. By using the spinal fracture index (19), five fractures were classified as grade 1; 10 fractures, grade 2; and three fractures, grade 3. The mean age of the transplant recipients with fractures postoperatively was significantly older than that of the patients without fractures (P < .05). We therefore adjusted the calculation of the P values for structure and BMD measurements for age. Figure 5 shows representative images of heart transplantation recipients postoperatively—one with and one without vertebral fractures; the trabecular structure is more scarce in the patient with vertebral fractures. Table 3 shows the means (± SD) of morphologic measurements, fractal dimensions, and BMDs in the heart transplant recipients with and without vertebral fractures. Differences in morphologic measurements between the two groups were statistically significant, but differences in fractal dimensions and BMDs were not. Differences in serum calcium, phosphate, alkaline phosphatase, and creatinine levels, as well as in cumulative doses of corticosteroids also were not significant between the fracture and nonfracture groups (P > .05).



View larger version (170K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5a. Sagittal, two-dimensional, T1-weighted spin-echo MR images (450/14) of the calcaneus (voxel size, 0.195 x 0.195 x 1 mm) in (a) a 56-year-old man 16 months after heart transplantation (apparent BV/TV, 0.33; apparent trabecular separation, 0.546 mm; BMD, 112 mg/cm3) without fractures and (b) a 65-year-old man 14 months after heart transplantation with three vertebral fractures (apparent BV/TV, 0.24; apparent trabecular separation, 0.692 mm; BMD, 93 mg/cm3). In a, the trabecular structure is scarce and the trabeculae are thin. In b, however, the trabecular structure is basically gone and the individual trabeculae are very thin.

 


View larger version (172K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5b. Sagittal, two-dimensional, T1-weighted spin-echo MR images (450/14) of the calcaneus (voxel size, 0.195 x 0.195 x 1 mm) in (a) a 56-year-old man 16 months after heart transplantation (apparent BV/TV, 0.33; apparent trabecular separation, 0.546 mm; BMD, 112 mg/cm3) without fractures and (b) a 65-year-old man 14 months after heart transplantation with three vertebral fractures (apparent BV/TV, 0.24; apparent trabecular separation, 0.692 mm; BMD, 93 mg/cm3). In a, the trabecular structure is scarce and the trabeculae are thin. In b, however, the trabecular structure is basically gone and the individual trabeculae are very thin.

 

View this table:
[in this window]
[in a new window]

  		TABLE 3. Morphologic Parameters, Fractal Dimensions, and BMDs in Heart Transplant Recipients with and without Vertebral Fractures
 
By using receiver operating characteristic analysis, the highest diagnostic performance in differentiating fracture from nonfracture transplant recipients was calculated for the measurements of apparent BV/TV, apparent trabecular separation, and apparent trabecular number. There was a lower area under the receiver operating characteristic curve for BMD measurements (Table 4). Because of the limited sample size, the differences between the individual area under the curve values, however, were not statistically significant. By using logistic regression analysis and combining all structure measurements, the area under the curve value increased to 0.93, and adding BMD measurements to the structure measurements caused an increase in the area under the curve values to only a minor extent (Table 4).


View this table:
[in this window]
[in a new window]

  		TABLE 4. Diagnostic Performance in Assessing Fracture Status at Receiver Operating Characteristic Analysis in Patients after Transplantation
 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
REFERENCES
 
The results of this study indicate that structure measurements assessed on high-spatial-resolution MR images of the calcaneus provide a distinction between patients before and after heart transplantation and between transplant recipients with and without vertebral fractures, whereas spine BMD measurements alone do not. Figures 3–5 depict the differences in trabecular structure between healthy control subjects and patients before and after heart transplantation, whereas Tables 1 and 3 show these differences quantitatively.

The identification of heart transplant recipients who are at higher risk for osteoporotic fractures and the monitoring of drug-induced changes in bone structure are important for preventing osteoporotic fractures and permitting adequate treatment (6). In a recent study, Shane et al (29) treated 18 heart transplant recipients with pamidronate intravenously, etidronate orally, and calcitriol orally and compared these patients with 52 patients who had undergone transplantation previously and not received antiresorptive therapy. With this therapy, the rate of incident fractures (mainly vertebral) in the first 12 months was substantially reduced—from 33% to 12%. To optimize therapy, treatment of only those patients with a higher fracture risk is required.

The prevalence of vertebral fractures in heart transplant recipients is high (1–5); fracture rates are 18%–50% and compare well to the rate of fractures in our study (42%). Unfortunately, the overlap of individual BMD measurements among the groups of patients with and without fractures suggests that BMD does not reliably predict the patients who are at risk of having fractures (5). This is particularly true for men, many of whom develop fractures despite having normal bone mass. Again, these results correspond to those in our study, in which there was no significant difference in lumbar BMD (at quantitative CT) between the patients with and those without fractures. Thus, the development of new techniques to assess fracture risk and bone loss seems to be important. One such technique is the noninvasive analysis of trabecular bone structure.

At present, techniques to assess trabecular bone structure noninvasively are considerably less developed than are those to measure BMD, which are the standard methods for the evaluation of osteoporosis. Several imaging modalities have been used in vivo, including conventional radiography of the spine (30,31) and calcaneus (32) and thin-section CT of the spine (33–35). The findings of these studies showed the potential of structure analysis to provide information in addition to BMD in the prediction of osteoporotic fracture status. More recent studies (14–16,36–38) have focused on high-spatial-resolution MR imaging of the distal radius and the calcaneus, which has superior spatial resolution compared with clinical CT and does not involve radiation exposure.

To our knowledge, structure measurements obtained on high-spatial-resolution MR images have not yet been used to assess trabecular bone structure in patients with transplants and immunosuppressive drug regimens. In this study, we used structure measurements that are based on standard bone histomorphometry, as well as fractal dimension. Because standard histomorphometry is performed on microscopic images with much higher spatial resolution, it has to be considered that the trabecular structure (average thickness of the trabeculae, 80–200 µm), as depicted by using a spatial resolution of 195 x 195 x 1,000 µm, is subjected to major partial volume effects. Nevertheless, it has been shown that structure measurements obtained at these spatial resolutions predict the biomechanical strength of bone specimens (18) and that the structure shown on these images correlates with that seen on true histomorphometric images (38–40).

Structure analysis of high-spatial-resolution MR images has been used to assess postmenopausal patients with and without osteoporotic fractures. Wehrli et al (16) analyzed high-spatial-resolution MR images (voxel size, 137 x 137 x 500 µm3) of the distal radius in 20 postmenopausal women and found that neither BMD nor any of the structural parameters individually correlated significantly with the vertebral deformity fraction, but a simple function involving two of the structure measures did. In another study (36) involving the analysis of MR images (voxel size, 195 x 195 x 500 µm3) of the calcaneus in postmenopausal patients with and without osteoporotic hip fractures, results similar to those in this study were found: "Apparent" structure measurements had superior diagnostic performance compared with proximal femur BMD measurements.

Another important result of this study was that some of the bone structure parameters showed a significant yet moderate correlation with the number of months after heart transplantation, even after correction for age, whereas BMD showed no significant correlation. Thus, structure measurements may be superior to BMD measurements in monitoring long-term drug-induced deterioration of bone structure in transplant recipients. This hypothesis, however, has to be verified by using longitudinal studies.

The substantial differences in some structure measurements between the healthy control subjects and the transplant recipients preoperatively may be explained by inactivity and long-term use of diuretic agents, which affect calcium metabolism. Differences in BMD measurements of the axial skeleton between the transplant recipients preoperatively and an age-matched healthy population, however, were not significant.

The higher rate of fractures in the older patients after heart transplantation may be explained by preexistent lower bone mass and inferior bone structure due to age. It may not be explained by a longer time after transplantation, because the number of months after transplantation in the two groups was not substantially different. To avoid this potentially confounding age factor, the statistical significance of the differences in BMD and structure parameters between both groups was calculated with age adjustment.

Because both transverse and sagittal images could be used to distinguish between heart transplant recipients pre- and postoperatively, as well as between patients with and without vertebral fractures, the question may be raised whether it is required to obtain both orientations. Limiting the examination to only one sequence would reduce examination time substantially—to the time required for quantitative CT. Thus, this examination would be more feasible for routine use.

The limitations of this study included the cross-sectional study design; however, to our knowledge, this is the first study in which the bone structure in heart transplant recipients was analyzed by using high-spatial-resolution MR imaging. It is clear that longitudinal studies are required in the future. A section thickness of 1 mm may not be optimal to depict trabecular bone structure, but with the use of standard equipment, coil design, and sequence protocols, signal-to-noise ratios limit image quality and the section thickness cannot be reduced further. To improve analysis of trabecular bone architecture, an optimized coil design is required. In addition, three-dimensional sequences should be used and volumes rather than two-dimensional images should be assessed; these protocols are works in progress at our institutions.

In summary, our study results indicate that structure measurements, as determined by using high-spatial-resolution MR imaging of the calcaneus, may be used to characterize the trabecular skeleton in patients with osteoporosis induced by immunosuppressive drug regimens. All statistical computations, including receiver operating characteristic analysis, yielded similar or better results with the structure measurements than with the BMD measurements in the differentiation of patients before and after heart transplantation and of transplant recipients with and without osteoporotic vertebral fractures. Thus, high-spatial-resolution MR imaging may be suitable for use in the prediction of fracture in heart transplant recipients.


   ACKNOWLEDGMENTS
 
The authors acknowledge the help of Simone Waldt, MD, in coordinating the patient examinations.


   FOOTNOTES
 
Abbreviations: BMD = bone mineral density, BV/TV = bone volume to total volume ratio, ROI = region of interest

Author contributions: Guarantors of integrity of entire study, T.M.L., S.M., A.L.; study concepts, T.M.L., S.M., A.L., C.S., D.N.; study design, T.M.L., S.M.; definition of intellectual content, T.M.L., S.M., Y.L.; literature research, T.M.L., F.B., S.M.; clinical studies, T.M.L., F.B., S.C., C.S.; data acquisition, T.M.L., F.B., S.C.; data analysis, F.B., T.M.L., A.L., D.N., Y.L., S.M.; statistical analysis, Y.L., T.M.L., F.B.; manuscript preparation, T.M.L., S.M.; manuscript editing, S.M., D.N., Y.L., T.M.L.; manuscript review, S.M., Y.L., T.M.L.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 

  1. Lee A, Mull R, Keenan G, et al. Osteoporosis and bone morbidity in cardiac transplant recipients. Am J Med 1994; 96:35-41.[Medline]
  2. Muchmore J, Cooper D, Ye Y, Schlegel V, Zudhi N. Loss of vertebral bone density in heart transplant recipients. Transplant Proc 1991; 23:1184-1185.[Medline]
  3. Meys E, Terreaux-Duvert F, Beaume-Six T, Dureau G, Meunier P. Effects of calcium, calderol and monofluorophosphate on lumbar bone mass and parathyroid function in patients after cardiac transplantation. Osteoporos Int 1993; 3:329-332.
  4. Rich G, Mudge G, Laffel G, LeBoff M.. Cyclosporine A and prednisone-associated osteoporosis in heart transplant recipients. J Heart Lung Transplant 1992; 11:950-958.[Medline]
  5. Shane E, Rivas M, Staron R, et al. Fracture after cardiac transplantation: a prospective longitudinal study. J Clin Endocrinol Metab 1996; 81:1740-1746.[Abstract]
  6. Rodino M, Shane E. Osteoporosis after organ transplantation. Am J Med 1998; 104:459-469.[Medline]
  7. Mosekilde L. Vertebral structure and strength in vivo and in vitro. Calcif Tissue Int 1993; 53:121-126.
  8. Goldstein R, Goulet R, McCubbrey D. Measurement and significance of three-dimensional architecture to the mechanical integrity of trabecular bone. Calcif Tissue Int 1993; 53(suppl 1):S127-S133.
  9. Odgaard A, Kabel J, van Rietbergen B, Dalstra M, Huiskes R. Fabric and elastic principal directions of cancellous bone are closely related. J Biomech 1997; 30:487-495.[Medline]
  10. Dalstra M, Huiskes R, Odgaard E. Mechanical and textural properties of pelvic trabecular bone. J Biomech 1993; 27:375-389.
  11. Glendenning P, Kent G, Adler B, et al. High prevalence of osteoporosis in cardiac transplant recipients and discordance between biochemical turnover markers and bone histomorphometry. Clin Endocrinol 1999; 50:347-355.[Medline]
  12. Majumdar S, Genant H. Magnetic resonance imaging in osteoporosis. Eur J Radiol 1995; 20:193-197.[Medline]
  13. Majumdar S, Genant H. A review of the recent advances in magnetic resonance imaging in the assessment of osteoporosis. Osteoporos Int 1995; 5:79-92.[Medline]
  14. Majumdar S, Genant H, Grampp S, et al. 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. J Bone Miner Res 1997; 12:111-118.[Medline]
  15. Majumdar S, Link T, Augat P, et al. Trabecular bone architecture in the distal radius using MR imaging in subjects with fractures of the proximal femur. Osteoporos Int 1999; 10:231-239.[Medline]
  16. Wehrli F, Hwang S, Ma J, Song H, Ford J, Haddad J. Cancellous bone volume and structure in the forearm: noninvasive assessment with MR microimaging and image processing. Radiology 1998; 206:347-357.[Abstract/Free Full Text]
  17. Wehrli F, Ford J, Haddad J. Osteoporosis: clinical assessment with quantitative MR imaging in diagnosis. Radiology 1995; 196:631-641.[Abstract/Free Full Text]
  18. Link T, Majumdar S, Lin J, et al. A comparative study of trabecular bone properties in the spine and femur using high resolution MRI and CT. J Bone Miner Res 1998; 13:122-132.[Medline]
  19. Genant HK, Wu CY, van Kuijk C, Nevitt MC. Vertebral fracture assessment using a semiquantitative technique. J Bone Miner Res 1993; 8:1137-1148.[Medline]
  20. Ouyang X, Selby K, Lang P, Majumdar S, Genant H. High resolution MR imaging of the calcaneus: age-related changes in trabecular structure and comparison with DXA measurements. Calcif Tissue Int 1997; 60:139-147.[Medline]
  21. Parfitt M, Drezner M, Glorieux F, et al. Bone histomorphometry: standardization of nomenclature, symbols and units—report of the ASBMR Histomorphometry Nomenclature Committee. J Bone Miner Res 1987; 2:595-610.[Medline]
  22. Parfitt M. Bone Histomorphometry: standardization of nomenclature, symbols and units—summary of a proposed system. Bone Miner 1988; 4:1-5.[Medline]
  23. Ishida T, Kazuya Y, Takigawa A, Kariya K, Itoh H. Trabecular pattern analysis using fractal dimension. Jpn J Appl Phys Vol 1993; 32:1867-1871.
  24. Majumdar S, Weinstein R, Prasad R. The fractal dimension of trabecular bone: a measure of trabecular structure. Calcif Tissue Int 1993; 52:168.
  25. Waldt S, Meier N, Renger B, Fiebich M, Rummeny E, Link T. Strukturanalyse hochauflösender Computertomogramme als ergänzendes Verfahren in der Osteoporosediagnostik: in vitro Untersuchungen an Wirbelsäulensegmenten. Rofo Fortschr Geb Röntgenstr Neuen Bildgeb Verfahr. 1999; 171:136-142.[Medline]
  26. Hill A. Principles of medical statistics New York, NY: Oxford University Press, 1971.
  27. DeLong E, DeLong D, Clarke-Pearson D. Comparing the areas under two or more correlated receiver operating characteristics curves: a nonparametric approach. Biometrics 1988; 44:837-845.[Medline]
  28. Kalender W, Felsenberg D, Louis O, et al. Reference values for trabecular and cortical vertebral bone density in single and dual-energy quantitative computed tomography. Eur J Radiol 1989; 9:75-80.[Medline]
  29. Shane E, Rodino M, McMahon D, et al. Prevention of bone loss after heart transplantation with antiresorptive therapy: a pilot study. J Heart Lung Transplant 1998; 17:1089-1096.[Medline]
  30. Caligiuri P, Giger M, Favus M, Jia H, Doi K, Dixon L. Computerized radiographic analysis of osteoporosis: preliminary evaluation. Radiology 1993; 186:471-474.[Abstract/Free Full Text]
  31. Caligiuri P, Giger ML, Favus M. Multifractal radiographic analysis of osteoporosis. Med Phys 1994; 21:503-508.[Medline]
  32. Pothuaud L, Lespessailles E, Harba R, et al. Fractal analysis of trabecular bone texture on radiographs: discriminant value in postmenopausal osteoporosis. Osteoporos Int 1998; 8:618-625.[Medline]
  33. Chevalier F, Laval-Jeantet A, Laval-Jeantet M, Bergot C. CT image analysis of the vertebral trabecular network in vivo. Calcif Tissue Int 1992; 51:8-13.[Medline]
  34. Ito M, Ohki M, Hayashi K, Yamada M, Uetani M, Nakamura T. Trabecular texture analysis of CT images in the relationship with spinal fracture. Radiology 1995; 194:55-59.[Abstract/Free Full Text]
  35. Mundinger A, Wiesmeier B, Dinkel E, Helwig A, Beck A, Schulte-Moenting J. Quantitative image analysis of vertebral body architecture: improved diagnosis in osteoporosis based on high-resolution computed tomography. Osteoporos Int 1993; 3:138-147.[Medline]
  36. Link TM, Majumdar S, Augat P, et al. In vivo high resolution MRI of the calcaneus: differences in trabecular structure in osteoporosis patients. J Bone Miner Res 1998; 13:1175-1182.[Medline]
  37. Majumdar S, Kothari M, Augat P, et al. High resolution magnetic resonance (MR): three dimensional (3D) trabecular bone architecture and biomechanical properties. Bone 1998; 22:445-454.[Medline]
  38. Majumdar S, Newitt D, Kothari M, et al. Measuring 3D trabecular structure and anisotropy using magnetic resonance. Osteoporos Int 1997; 7:272.
  39. Hipp J, Jansujwicz A, Simmons C, Snyder B. Trabecular bone morphology from micro-magnetic resonance imaging. J Bone Miner Res 1996; 11:286-292.[Medline]
  40. Kothari M, Keaveny T, Lin J, Newitt D, Genant H, Majumdar S. Impact of spatial resolution on the prediction of trabecular architecture parameters. Bone 1998; 22:437-443.[Medline]



This article has been cited by other articles:


Home page
 RadiologyHome page
C. M. Phan, M. Matsuura, J. S. Bauer, T. C. Dunn, D. Newitt, E. M. Lochmueller, F. Eckstein, S. Majumdar, and T. M. Link
Trabecular Bone Structure of the Calcaneus: Comparison of MR Imaging at 3.0 and 1.5 T with Micro-CT as the Standard of Reference
Radiology, May 1, 2006; 239(2): 488 - 496.
[Abstract] [Full Text] [PDF]

Home page
 J. Clin. Endocrinol. Metab.Home page
S. Boutroy, M. L. Bouxsein, F. Munoz, and P. D. Delmas
In Vivo Assessment of Trabecular Bone Microarchitecture by High-Resolution Peripheral Quantitative Computed Tomography
J. Clin. Endocrinol. Metab., December 1, 2005; 90(12): 6508 - 6515.
[Abstract] [Full Text] [PDF]

Home page
 RadiologyHome page
T. M. Link, B. B. Koppers, T. Licht, J. Bauer, Y. Lu, and E. J. Rummeny
In Vitro and in Vivo Spiral CT to Determine Bone Mineral Density: Initial Experience in Patients at Risk for Osteoporosis
Radiology, June 1, 2004; 231(3): 805 - 811.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Link, T. M.
Right arrow Articles by Majumdar, S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Link, T. M.
Right arrow Articles by Majumdar, S.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
RADIOLOGY RADIOGRAPHICS RSNA JOURNALS ONLINE