Miriam Antoinette Bredella MD, Presenter: Nothing to Disclose

Scott M. Daley, Abstract Co-Author: Nothing to Disclose

Mannudeep K. S. Kalra MD, Abstract Co-Author: Nothing to Disclose

Martin Torriani MD, Abstract Co-Author: Nothing to Disclose

Keenan Brown, Abstract Co-Author: Employee, Mindways Software, Inc Stockholder, Mindways Software, Inc

Karen K. Miller MD, Abstract Co-Author: Nothing to Disclose

Quantification of marrow fat has been proposed as a predictor of bone weakening, independent of bone mineral density (BMD). 1H-MRS is able to reliably quantify bone marrow fat fraction in a single voxel, however, the ability of measuring larger areas of heterogeneous marrow is limited. Advances in DECT allow quantification of BMD of the entire axial and appendicular skeleton with no additional radiation exposure compared to standard QCT. The purpose of our study was to test the performance of DECT in assessing marrow fat content of the lumbar spine, using 1H-MRS as a reference standard.

The study was IRB approved and complied with HIPAA guidelines. Written informed consent was obtained from all subjects. Seven healthy men (mean age: 39±13 y) who participated in a clinical obesity trial underwent single voxel 1H-MRS at 3T (Siemens Trio) of the L2 vertebra using a PRESS sequence without water suppression. DECT (Siemens Definition Flash) of the L2 vertebra (80 kVp @ 210 mAs, 140 kVp @ 80 mAs) was performed with use of a calibration phantom. Average basis material composition relative to the phantom was estimated within an elliptical cylinder (3 cm3) positioned in the anterior cancellous bone region of L2. The resulting basis material composition was then fit to a cancellous bone model yielding a description in terms of volumes of model mineralized collagen, yellow and red marrow from which the volume fraction of yellow marrow within the marrow space was derived. Pearson correlation coefficient and Bland-Altman 95% limit of agreement of fat fraction obtained from 1H-MRS and DECT were calculated. 

There was a strong correlation between marrow fat fraction obtained by 1H-MRS and DECT (r=0.90, p= 0.006). Using Bland-Altman analysis, there was good agreement between 1H-MRS and DECT without evidence of bias. The mean difference in fat fraction between the techniques was 0 with a 95% confidence interval between -0.26 and 0.25.

DECT is a reliable method to measure marrow fat content of the lumbar spine. DECT provides data that closely correlate with 1H-MRS. Therefore, DECT could potentially be used to assess both BMD and marrow fat content in a single examination.

DECT is a novel imaging technique that can assess BMD and marrow fat content in a single examination, thereby providing important information on skeletal integrity. 

Bredella, M, Daley, S, Kalra, M, Torriani, M, Brown, K, Miller, K, Bone Marrow Fat Quantification of the Lumbar Spine Using Dual Energy CT (DECT) and Single Voxel 1H-MR Spectroscopy (1H-MRS)—A Feasibility Study.  Radiological Society of North America 2014 Scientific Assembly and Annual Meeting, - ,Chicago IL. http://archive.rsna.org/2014/14045529.html