Reproducibility of sodium MRI measures of articular cartilage of the knee in osteoarthritis

SummaryObjectiveTo determine the stability and reproducibility of the sodium magnetic resonance imaging (MRI) signal measured in the articular cartilage of the knee in both healthy volunteers and osteoarthritis (OA) patients.DesignThis was a prospective Research Ethics Committee approved study that acquired sodium and proton MRI data from 15 subjects with OA (three males, age 64 ± 10) and five healthy controls age and sex matched over the group. Each subject underwent standing planar radiographs of their knees for radiological scoring as well as symptomatological assessment questionnaires. In two MRI sessions on the same day, high resolution double-echo steady state (DESS) and 3D short echo time sodium MRI images of the most diseased knee were acquired and co-registered in each session. A blinded reader (LT) manually delineated the articular cartilage into four discrete regions, and two combined regions, on the DESS images. These regions were applied to the sodium images, and a median sodium signal from each reported. Within-subject and between-subject coefficients of variation were estimated and intraclass correlation coefficients for the healthy control group, OA subject group, and all pooled subjects group were calculated.ResultsWithin-subject variability of sodium MRI at 3 T was 3.2% overall, and 2.0% in healthy age-matched volunteers compared to a reproducibility of 3.6% on OA subjects.ConclusionsThe reproducibility of sodium MRI was similar in both healthy controls and OA subjects. Researchers piloting techniques in healthy controls thus may expect a similar reproducibility in a controlled trial involving subjects with American College of Rheumatology (ACR)-defined OA of the knee.