Certain 3D Knee MRI Bone Shapes Predictive of TKR for OA

Three-dimensional (3D) bone shape appears to be associated with the important patient outcome of total knee replacement (TKR) and the femur may have the greatest association, according to British researchers. They are reporting that quantitative bone measures may be useful in future therapeutic trials which seek to assess disease-modifying osteoarthritis (OA) drugs.

“We’ve had MRI for many years and it has enormously expanded our knowledge of OA structural pathology. However, we haven’t had the technology to extract all the information from MR images. Statistical shape modeling (SSM) provides a novel tool for automated highly accurate 3-dimensional assessment of joint tissues, but it needs to be validated against outcomes that are important for people and health systems. So, this study was about a novel SSM bone shape outcome and whether it predicted total joint replacement,” study author Philip Conaghan, MBBS, PhD, professor of musculoskeletal medicine at the University of Leeds, Leeds, UK, said in an interview with Rheumatology Advisor.

He said this study is not suggesting clinicians should use this imaging biomarker to predict who needs a joint replacement. Dr Conaghan said that remains a clinical decision between the treating clinician and the OA patient. The researchers conducted this investigation to analyze the relationship between 3D MRI bone shape and TKR. The researchers compared 310 matched controls to 310 TKR cases. The team collected baseline demographic data and MRI scans were performed in all the subjects.

The researchers found that those in the TKR arm had a more positive mean baseline 3D bone shape vector, indicating more advanced structural OA. Overall, the study demonstrated that the mean baseline 3D bone shape scalar variable indicated significantly greater structural severity in cases of TKR compared to the controls. 

The mean in the femur cases was 0.98 (1.51) compared to a mean of 0.11 (1.40) for the controls. This was found to be a statistically significant difference of 1.10. The mean in the patella cases was 0.95 (1.84) compared to 0.03 (1.83), resulting in a statistically significant difference of 0.92. The odds ratios (ORs) per normalized unit of 3D bone shape vector for the femur were 1.85, for the tibia they were 1.64 and for the patella they were 1.36.

Summary & Clinical Applicability

Dr Conaghan and his colleagues note that currently OA is one of the most common causes of global disability and the numbers are only expected to rise as people live longer and become more obese. The researchers write that currently there are no licensed disease-modifying agents in OA, but 3D assessments may be a valuable tool in helping test novel new agents for OAs.

“These findings are important because they are part of a body of work validating bone shape as a biomarker and prefacing a new generation of automated, accurate imaging biomarkers,” Dr Conaghan told Rheumatology Advisor. 

Limitations and Disclosures

 “At present there is still no need to use MRI in routine OA clinical practice as it does not guide therapy. However, new highly accurate quantification of pathology will inform patient phenotypes for interventions and unblock clinical trials for potential [disease modifying osteoarthritis drugs].”

One study author discloses employment at Imorphics Ltd. 

Reference

Andrew J. Barr, Bright Dube, Elizabeth M. A. Hensor et al. The relationship between three-dimensional knee MRI bone shape and total knee replacement—a case control study: data from the Osteoarthritis Initiative. Rheumatology. 2016; doi:10.1093/rheumatology/kew191.