Abstract
BACKGROUND: Osteoporosis is characterized by low bone mineral density (BMD), which predisposes individuals to frequent fragility fractures. Quantitative BMD measurements can potentially help distinguish bone pathologies and allow clinicians to provide disease-relieving therapies. Our group has developed non-invasive and non-ionizing magnetic resonance imaging (MRI) techniques to measure bone mineral density quantitatively. Dual-energy X-ray Absorptiometry (DXA) is a clinically approved non-invasive modality to diagnose osteoporosis but has associated disadvantages and limitations. PURPOSE: Evaluate the clinical feasibility of phosphorus ((31)P) MRI as a non-invasive and non-ionizing medical diagnostic tool to compute bone mineral density to help differentiate between different metabolic bone diseases. MATERIALS AND METHODS: Fifteen ex-vivo rat bones in three groups [control, ovariectomized (osteoporosis), and vitamin-D deficient (osteomalacia - hypo-mineralized) were scanned to compute BMD. A double-tuned ((1)H/(31)P) transmit-receive single RF coil was custom-designed and in-house-built with a better filling factor and strong radiofrequency (B(1)) field to acquire solid-state (31)P MR images from rat femurs with an optimum signal-to-noise ratio (SNR). Micro-computed tomography (μCT) and gold-standard gravimetric analyses were performed to compare and validate MRI-derived bone mineral densities. RESULTS: Three-dimensional (31)P MR images of rat bones were obtained with a zero-echo-time (ZTE) sequence with 468 μm spatial resolution and 12-17 SNR on a Bruker 7 T Biospec having multinuclear capability. BMD was measured quantitatively on cortical and trabecular bones with a known standard reference. A strong positive correlation (R = 0.99) and a slope close to 1 in phantom measurements indicate that the densities measured by (31)P ZTE MRI are close to the physical densities in computing quantitative BMD. The (31)P NMR properties (resonance linewidth of 4 kHz and T(1) of 67 s) of ex-vivo rat bones were measured, and (31)P ZTE imaging parameters were optimized. The BMD results obtained from MRI are in good agreement with μCT and gravimetry results. CONCLUSION: Quantitative measurements of BMD on ex-vivo rat femurs were successfully conducted on a 7 T preclinical scanner. This study suggests that quantitative measurements of BMD are feasible on humans in clinical MRI with suitable hardware, RF coils, and pulse sequences with optimized parameters within an acceptable scan time since human femurs are approximately ten times larger than rat femurs. As MRI provides quantitative in-vivo data, various systemic musculoskeletal conditions can be diagnosed potentially in humans.