We developed an alternative steroid-associated osteonecrosis (ON) rabbit model using a combination of a single injection of low-dose lipopolysaccharide (LPS) and three subsequent injections of pulsed high-dose methylprednisolone (MPS). The usefulness of this experimental ON model was evaluated using both conventional and advanced bio-imaging techniques, including contrast-enhanced dynamic MRI and a high-resolution micro-CT. Details on establishment of methodology are described, which were adopted into an efficacy study on a herbal Epimedium-derived phytoestrogenic extract (HEPE) developed for prevention of steroid-associated ON using an established rabbit model. The underlying mechanisms of HEPE for prevention of steroid-associated ON were found to be associated with inhibition of both intravascular thrombosis and extravascular bone marrow lipid deposition, the two known mechanistic pathways in pathogenesis of ON. Our experimental results provide for potential clinical trials or applications of HEPE in the prevention of ON among highrisk patients undergoing steroid treatment.

Bone marrow was assumed with little effect on the bone mechanical property. As a result, there is no simulation model of bone strength considering bone marrow. While, it has been reported that fat marrow content increased in osteoporotic bone. Whether the fat marrow influences the bone mechanical property is unknown yet. Therefore, the objective of this study was to figure out the contribution of bone marrow fat in bone strength. A finite element model was established by using quantitative computed tomographic (QCT) data from a 63-yearold female's L3 vertebrae. With mechanical load applied on the finite element model, a significant higher maximum stress was observed in the model without bone marrow, compared to the one with fat marrow. The results indicated that bone marrow fat diminish the maximum stress on the trabecular bone and may function as a force buffer in a bone.

The purpose of this study was to establish a new pharmacokinetic model to investigate bone marrow and muscle perfusion without using arterial input function. Based on the reference region model, erector spinae muscle around the vertebral body was selected as the reference region. A muscle-based model was established for erector spinae muscle perfusion with a personalized algorithm. Seventy-six female subjects (age 72.5±3.4 yrs) were recruited in this study and classified into three groups (normal, osteopenia, and osteoporosis) according to the T-score. Bone mineral density was measured by dual-energy X-ray absorptiometry. Quantitative parameters were extracted from the pharmacokinetic model, K trans,TOI (contrast agent extravasation rate constants for blood perfusion of the tissue of interest) showed a significant reduction in subjects with lower bone mineral density, which were consistent with previous studies. However, muscle perfusion parameters remained unchanged among different groups. The results indicated that muscle-based model was stable for bone marrow perfusion modeling. Meanwhile, the maintained muscle perfusion indicated that the decreased perfusion in the bone marrow happened in the bone itself but not in the supply arteries. Keywords-muscle-based model, bone marrow perfusion, erector spinae muscle perfusion, bone mineral density

The pharmacokinetic model has been widely used in tissue perfusion analysis, such as bone marrow perfusion. In the modeling process, the arterial input function is important to guarantee the reliability of the fitting result. However, the arterial input function is variable and hard to control, which makes it difficult to compare results across different studies. The purpose of this study was to establish a muscle-based pharmacokinetic model for bone marrow perfusion without using arterial input function. Erector spinae muscle around the vertebral body was selected as the reference region. The study was carried out in elderly females with different bone mineral densities (normal, osteopenia, and osteoporosis). Quantitative parameters were extracted from the pharmacokinetic model. Parameter K trans,BM (contrast agent extravasation rate constants for blood perfusion of the bone marrow) showed a significant reduction in subjects with lower bone mineral density, which is consistent with previous studies. However, muscle perfusion parameters remained unchanged among different groups. The results indicated that the muscle-based model was stable for bone marrow perfusion modeling. Additionally, nonsignificant change in muscle parameters indicated that the diminished perfusion is only a local rather than a systematic change in the bone marrow for osteoporosis.