3-dimensional bone shape and knee osteoarthritis: What have we learned?

The concept that multiple joint tissues are involved in the osteoarthritis (OA) disease process is now widely accepted. There have been significant and important insights over the past two decades in the understanding of bone as a tissue undergoing pathological changes in OA. The specific bony changes of osteophyte growth and “bone attrition” associated with OA have been recognized for many years with several semi-quantitative radiographic and magnetic resonance imaging (MRI) grading systems designed to capture the magnitude of these changes. Over the past decade, there has been significant and important progress in the quantitative measurement of these changes. Manual methods for measuring bone area from 3D MR images have been improved with automation which offers both superior precision and a more responsive measurement that has been applied in several DMOAD randomized controlled trials. Measurement of true 3D bone shape, as opposed to simple geometric measures such as curvature and length, depends on automated methods of segmentation. In this field, important developments have taken place in the statistical parameterization of shape and the construction of OA vs non-OA shape metrics. Work has demonstrated that bone shape may provide an indication of OA status, may predict future OA onset, and is associated with clinical markers of OA such as pain, function and total joint replacement (TKR). Thus, bone shape may be a useful imaging biomarker for OA.