Send to

Choose Destination
Bone. 2013 Jul;55(1):7-15. doi: 10.1016/j.bone.2013.03.004. Epub 2013 Mar 16.

Diffusion tensor imaging and magnetic resonance spectroscopy assessment of cancellous bone quality in femoral neck of healthy, osteopenic and osteoporotic subjects at 3T: Preliminary experience.

Author information

Department of Diagnostic and Interventional Radiology, Molecular Imaging and Radiotherapy, PTV Foundation, Tor Vergata University of Rome, Viale Oxford 81, 00133 Rome, Italy.


We assessed the potential of diffusion tensor imaging (DTI) in combination with proton magnetic resonance spectroscopy (1H-MRS), in cancellous bone quality evaluation of the femoral neck in postmenopausal women.


DTI allows for non-invasive microarchitectural characterization of heterogeneous tissue. In this work we hypothesized that DTI parameters mean diffusivity (MD) and fractional anisotropy (FA) of bone marrow water, can provide information about microstructural changes that occur with the development of osteoporosis disease. Because osteoporosis is associated with increased bone marrow fat content, which in principal can alter DTI parameters, the goal of this study was to examine the potential of MD and FA, in combination with bone marrow fat fraction (FF), to discriminate between healthy, osteopenic and osteoporotic subjects, classified according to DXA criteria.


Forty postmenopausal women (mean age, 68.7 years; range 52-81 years), underwent a Dual-energy X-ray absorptiometry (DXA) examination in femoral neck, to be classified as healthy (n=12), osteopenic (n=14) and osteoporotic (n=14) subjects. 1H-MRS and DTI (with b value=2500 s/mm2) of femoral neck were obtained in each subject at 3T. The study protocol was approved by local Ethics Committee. MD, FA, FF and MD/FF, FA/FF were obtained and compared among the three bone-density groups. One-way ANOVA with multiple comparisons Bonferroni test and Pearson correlation analysis were applied. Receiver operating characteristic (ROC) curve analysis was also performed.


Reproducibility of DTI measures was satisfactory. CV was approximately 2%-3% for MD and 4%-5% for FA measurements. Moreover, no significant difference was found in both MD and FA measurements between two separate sessions (median 34 days apart) comprised of six healthy volunteers. FF was able to discriminate between healthy and osteoporotic subjects only. Conversely MD and FA were able to discriminate healthy from osteopenic and healthy from osteoporotic subjects, but they were not able to discriminate between osteopenic and osteoporotic patients. A significant correlation between MD and FF was observed in healthy group only. A moderate correlation was found between MD and T-score when all groups together are considered. No significant correlation was found between MD and T-score within groups. A significant positive correlation between FA and FF was found in both osteopenic and osteoporotic groups. Vice-versa no correlation between FA and FF was observed in healthy group. A high significant positive correlation was found between FA and T-score in all groups together, in healthy and in osteoporotic groups. MD/FF and FA/FF are characterized by a higher sensitivity and specificity compared to MD and FA in the discrimination between healthy, and osteoporotic subjects. MD/FF vs. FA/FF graph extracted from femoral neck, identify all healthy individuals according to DXA results.


DTI-(1)H-MRS protocol performed in femoral neck seems to be highly sensitive and specific in identifying healthy subjects. A MR exam is more expensive when compared to a DXA investigation. However, even though DXA BMD evaluation has been the accepted standard for osteoporosis diagnosis, DXA result has a low predictive value on patients' risk for future fractures. Thus, new approaches for examining patients at risk for developing osteoporosis would be desirable. Preliminary results showed here suggest that future studies on a larger population based on DTI assessment in the femoral neck, in combination with 1H-MRS investigations, might allow screening of high-risk populations and the establishment of cut-off values of normality, with potential application of the method to single subjects.

[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Elsevier Science
Loading ...
Support Center