Bone Marrow Lipid Profiles from Peripheral Skeleton as Potential Biomarkers for Osteoporosis: A 1H-MR Spectroscopy Study.

RATIONALE AND

OBJECTIVES

To characterize the lipidic profile of bone marrow in the calcaneus and femoral neck of healthy, osteopenic, and osteoporotic women, by using magnetic resonance spectroscopy (MRS) at 3T. The final goal was to identify specific metabolites with the potential ability to discriminate between healthy, osteopenic, and osteoporotic subjects.

MATERIALS AND

METHODS

Sixty-two and thirty three postmenopausal women recruited to investigate calcaneus and femoral neck, respectively, underwent a bone mineral density (BMD) measurement to be classified as healthy subjects (n = 22), osteopenic (n = 45), or osteoporotic (n = 28) patients. MRS spectra were used to quantify and compare bone marrow fat resonances between the three BMD groups. Between-group differences were tested using a Welch analysis of variance. Multiple comparisons were made with the Games-Howell correction. Relationships between pairs of variables were assessed with linear correlation analysis. Reproducibility analysis was performed for all the lipid resonances in both sites.

RESULTS

The reproducibility was satisfactory. In femoral neck, methylene (L13), glycerol (L41, L43), and total lipid resonances were significantly lower in healthy as compared to osteoporotic subjects. On the other hand, in calcaneus, L13/glycerol significantly discriminated between osteopenic and osteoporotic subjects whereas L13/(unsaturated lipid) discriminated between healthy and osteopenic group. However, the reproducibility of both unsaturated lipid and glycerol resonances were less optimal.

CONCLUSIONS

MRS of bone marrow lipid profiles from peripheral skeletal sites may be a promising tool for screening of large population to identify individuals with or at risk for developing osteoporosis. Moreover, it provides information about the metabolic changes occurring in bone marrow with the development of osteoporosis, which are skeletal site dependent.