Quantitative microcomputed tomography assessment of intratrabecular, intertrabecular, and cortical bone architecture in a rat model of severe renal osteodystrophy.

OBJECTIVE: To determine the effects of renal osteodystrophy (ROD) on bone microarchitecture in growing rats. METHODS: A total of 24 rats underwent 5/6 nephrectomy (NX) and were fed a high-phosphorus diet to induce ROD; another 6 underwent sham NX. In vitro microcomputed tomography images (GEMS, London, Ontario, Canada) were obtained in the femoral metaphysis and midshaft. RESULTS: Trabecular and cortical bone volume/total volume (BV/TV) were significantly lower in NX specimens because of pores within the trabeculae and along the endosteal surface. Topological analysis using component labeling in 3-dimensions verified that trabecular pores connected to the marrow space. After the trabecular pores were filled using a morphological filter, trabecular thickness was significantly increased in NX. In contrast, cortical thickness was significantly decreased in NX compared with controls; however, after filling the endocortical pores, thickness did not differ. CONCLUSIONS: The ROD resulted in decreased cortical and trabecular BV/TV, increased porosity, and increased trabecular thickness. Advanced image processing algorithms demonstrated the effects of cortical and trabecular porosity on BV/TV and structure in ROD.

Experimental and computational analyses of the effects of slice distortion from a metallic sphere in an MRI phantom.

Susceptibility artifacts due to metallic prostheses are a major problem in clinical magnetic resonance imaging. We theoretically and experimentally analyze slice distortion arising from susceptibility differences in a phantom consisting of a stainless steel ball bearing embedded in agarose gel. To relate the observed image artifacts to slice distortion, we simulate images produced by 2D and 3D spin-echo (SE) and a view angle tilting (VAT) sequence. Two-dimensional SE sequences suffer from extreme slice distortion when a metal prosthesis is present, unlike 3D SE sequences for which--since slices are phase-encoded--distortion of the slice profile is minimized, provided the selected slab is larger than the region of interest. In a VAT sequence, artifacts are reduced by the application of a gradient along the slice direction during readout. However, VAT does not correct for the excitation slice profile, which results in the excitation of spins outside the desired slice location and can lead to incorrect anatomical information in MR images. We propose that the best sequences for imaging in the presence of a metal prosthesis utilize 3D acquisition, with phase encoding replacing slice selection to minimize slice distortion, combined with excitation and readout gradient strengths at their maximum values.

MRI can identify high intensity bands around implants that correspond to radiolucent lines on radiographs: an ex vivo study of sheep acetabula.

We investigated the use of magnetic resonance imaging in categorizing the tissue interface between bone and bone cement after total hip replacement. In an ex vivo sheep model, we compared correlations between the thicknesses of radiolucent lines measured from plain radiographs and corresponding high signal intensity bands in magnetic resonance imaging scans. Correlations obtained for the three main Gruen Zones were in the range between r2 = 0.58 (superior zone), and r2 = 0.86 (inferior zone). In two specimens magnetic resonance imaging was able to detect high signal intensity bands that were not visible on the radiographs. The average thickness of the high signal intensity bands measured on the magnetic resonance imaging scans ranged from 14.6% (Zone 2) to 39.9% (Zone 3) larger than the corresponding radiolucent lines measured on the plain radiographs. Magnetic resonance imaging has the advantage over current techniques in that it is noninvasive and does not use ionizing radiation. If problems associated with metal artifacts can be reduced, these potential advantages may make magnetic resonance imaging the preferred method for longitudinal studies assessing osteointegration and analysis of fibrous membranes.