Improved Imaging of Osteoarthritis using Fluctuating Equilibrium Magnetic Resonance Imaging

Gary Gold, MD; S. Vasanawala, MD, PhD; D. Nishimura, PhD; J. Pauly, PhD; Gary S. Beaupré, PhD

Objectives: The overall purpose of our program is to develop new methods for imaging articular cartilage and studying cartilage biomechanics. Osteoarthritis is a disease of cartilage degeneration, and is extremely common in aging veterans. Magnetic resonance imaging (MRI) is a promising non-invasive tool for evaluation of cartilage. The purpose of this study is to evaluate the use of Fluctuating Equilibrium MR (FEMR) to image cartilage.

Methods: Current MRI methods lack sufficient spatial resolution and contrast for detailed evaluation of cartilage. Three-dimensional spoiled gradient recalled echo (3D-SPGR) offers high resolution, but has relatively long scan times and suboptimal cartilage/fluid contrast. T2-weighted Fast Spin Echo (FSE) produces bright synovial fluid, but has limited resolution and low cartilage signal. Proton density FSE provides good contrast, but blurring limits detail.

Proton Density and FEMR images of a patient with cartilage damage. Note the sharp detail in the area of damage (arrow) using FEMR.

We have developed a rapid 3D fat/water separating technique called fluctuating equilibrium magnetic resonance (FEMR). The sequence refocuses magnetization over a repetition time (TR), and uses a TR of 6.6 ms for fat/water separation at 1.5 T. FEMR shows excellent synovial fluid/cartilage contrast and cartilage signal. The knee joints of ten healthy volunteers and one patient were imaged on a GE Signa 1.5 T scanner with CV/i gradients (150 T/m/s), using an extremity coil. The FEMR sequence was followed by a 3D-SPGR (50 ms TR), T2-weighted FSE (3.2 s TR) and a proton density FSE (4 s TR), all with fat saturation. Scans were preceded by a field shimming sequence.

Results: Contrast-to-noise measurements for cartilage using FEMR were superior to T2W-FSE. Proton density FSE images showed bright synovial fluid with limited cartilage detail. 3D-SPGR had comparable resolution to FEMR, but suboptimal cartilage/fluid contrast and longer scan times (8 minutes versus 2 minutes). Cartilage surface detail, outlined by bright synovial fluid, was best seen on the FEMR images.

Conclusion: FEMR obtains high-resolution 3D images of the entire knee in two minutes with excellent cartilage/fluid contrast. FEMR is sensitive to field inhomogeneity and requires shimming. Surface defects are outlined by bright synovial fluid, and cartilage has higher contrast-to-noise compared with T2W-FSE.

Funding Acknowledgment: This study was funded by the VA Rehabilitation Research and Development Service, Research Career Development Grant #B2376V.