RSNA 2009 

Abstract Archives of the RSNA, 2009


Quantitative Dynamic Contrast-enhanced MRI in Vertebral Bone Marrow

Scientific Papers

Presented on November 30, 2009
Presented as part of SSE15: Musculoskeletal (Quantitative Imaging)


Andreas Biffar, Presenter: Nothing to Disclose
Gerwin Paul Schmidt MD, Abstract Co-Author: Nothing to Disclose
Steven Sourbron PhD, Abstract Co-Author: Nothing to Disclose
Olaf Dietrich, Abstract Co-Author: Nothing to Disclose
Maximilian F. Reiser MD, Abstract Co-Author: Nothing to Disclose
Andrea Baur-Melnyk PhD, Abstract Co-Author: Nothing to Disclose


DCE-MRI in the spine has a large clinical potential in a variety of pathologies (e.g. osteoporosis). Previous studies are based on simple descriptive indices, suffering from well-known limitations, which can be overcome by the use of quantitative methods. Here we develop a fully quantitative approach for the analysis of DCE-MRI in vertebral bone marrow, generalizing existing approaches by allowing for an unknown fraction of fatty tissue. A first evaluation in 10 patients is presented.  


Tracer concentrations C are derived via numerical inversion from measured signals, assuming that the M is caused by a fraction fwat of water and (1-fwat) of fat, while the tracer only affects the R1 of the water protons. Perfusion parameters are then determined by fitting C(t) to a 2-compartment model.10 patients were examined at 1.5T (Siemens Avanto). DCE-MRI was performed with a 2D-saturation-recovery TFL, measuring 4 slices/s (3 sagittal, 1 axial) for 5 min (300 dynamics, matrix=192x144, FOV=300x225x50mm3, TR/TE/flip angle=3.1ms/1.37ms/12). Precontrast relaxation rates were determined with a SR-technique using a HASTE sequence. fwat was quantified as the ratio of water and fat proton densities. To analyze the effect of the fat correction, parameters were also calculated with the conventional approach neglecting the fat component fwat=1.


Four perfusion parameters (PF,PV,IV,EF) have been calculated from concentration time curves obtained in normal BM and in benign and malignant lesions. Differences up to 220 % are found between both approaches, most severe for healthy bone marrow (which has the lowest fwat = 59%) and decreasing with higher fwat. fwat  is dependent on lesion type and has high inter-patient variability (11 % in normal BM). PF- and EF-values were significantly different (p<0.05) in osteoporotic (n=5) and metastatic compression fractures (n=5).


Perfusion imaging in vertebral BM is feasible. Correcting for fat component has a strong effect on perfusion parameters. Correction depends on water/fat ratio, being significantly different between patients and pathologies. This suggests that the method improves the accuracy of quantitative perfusion metrics, and has the potential for lesion characterization.


Data indicates potential for differentiation between osteoporotic and pathologic vertebral compression fractures, which remains to be verified in larger studies.

Cite This Abstract

Biffar, A, Schmidt, G, Sourbron, S, Dietrich, O, Reiser, M, Baur-Melnyk, A, Quantitative Dynamic Contrast-enhanced MRI in Vertebral Bone Marrow.  Radiological Society of North America 2009 Scientific Assembly and Annual Meeting, November 29 - December 4, 2009 ,Chicago IL.