Arthritis & Rheumatism, Volume 60,
October 2009 Abstract Supplement

The 2009 ACR/ARHP Annual Scientific Meeting
Philadelphia October 16-21, 2009.

Reproducibility of MR Biomarkers of Cartilage Structure and Composition in An Osteoarthritis Population at Risk of Progression

Bowes1,  Mike A., Hunter2,  D.J., Eaton3,  Charles B., Kwoh4,  C. K., Maciewicz5,  RA., Samuels6,  Jonathan, Taylor7,  Christopher J.

Imorphics, Manchester, United Kingdom
NEBH, Boston, MA
Alpert Medical School of Brown University, Pawtucket, RI
UPitt, Pittsburgh, PA
AstraZeneca, Macclesfield, United Kingdom
New York University/Hospital for Joint Diseases, New York, NY
University of Manchester, United Kingdom


Osteoarthritis (OA) is a slowly-progressing disease, and imaging biomarkers of cartilage structure have failed to detect short-term (< 6 months) change. An imaging biomarker of cartilage quality (i.e., tissue composition and function) such as magnetic resonance transverse relaxation time of hyaline cartilage water protons (T2), may be more a responsive biomarker of OA progression. T2 is a proxy for disorder and hydration, and is easily measured without a contrast agent, but there are few data on variability of this marker over time in OA. The purpose of this study was to determine the short-term (1 week) reproducibility of cartilage T2 and thickness obtained from an anatomically-corresponded regional analysis using statistical shape modelling.


We conducted a multi-centre, non-randomized study at 4 sites with a sample at risk of medial tibiofemoral progression including women, BMI>=25 kg/m2, symptomatic radiographic evidence of medial tibiofemoral OA (K&L grade 2–3, medial JSN>= lateral JSN), varus malalignment >=-2° (anatomic axis), and pain. As part of a larger study, eligible participants had MRI scans of the same knee at baseline and 1 week. The OAI protocol for the index knee was deployed on 3T Siemens Trio systems. A trained operator, blind to time-point but not subject, manually segmented the cartilage from the DESSwe MR images using EndPoint (Imorphics). Anatomically corresponding regions of interest were identified on each image by fitting a bone model, and mean cartilage thickness (with areas denuded of cartilage included as having zero thickness - ThCtAB) within each region was calculated. Voxelwise transverse relaxation rates were calculated from a linear least-squares fit of the log of the signal values against echo time. Mean T2 values were also recorded in each region from the 50% most exochondral and 50% most endochondral voxels. Coefficients of Variation (CoV) were calculated.


The 29 participants had a mean age of 62 years, mean BMI of 36 kg/m2, with 8 index knees graded as K&L =2 and 21 as K&L=3. Anatomical mal-alignment ranged from -1.9° to 6.3°, with mean 0.9°, where varus mal-alignment is measured in the positive direction. 28 subjects provided data for reproducibility of ThCtAB and 20 for T2.

Table 1. Reproducibility MR Measures

 Mean T2/msecCoV T2Mean ThCtAB/mmCoV ThCtAB
Medial Femur49.53.1%1.592.2%
Lateral Femur49.02.6%1.692.8%
Medial Tibia38.72.8%1.352.5%
Lateral Tibia38.82.6%1.663.0%
T2 was approximately 10% higher in the exochondral vs. endochondral layer in each region.


T2 can be measured reproducibly in an OA population in multicentre studies and with similar variability to a cartilage structural assessment. These results support further evaluation of T2 as a candidate biomarker of cartilage composition for assessing interventions.

To cite this abstract, please use the following information:
Bowes, Mike A., Hunter, D.J., Eaton, Charles B., Kwoh, C. K., Maciewicz, RA., Samuels, Jonathan, et al; Reproducibility of MR Biomarkers of Cartilage Structure and Composition in An Osteoarthritis Population at Risk of Progression [abstract]. Arthritis Rheum 2009;60 Suppl 10 :768
DOI: 10.1002/art.25848

Abstract Supplement

Meeting Menu