Article Text

A5.01 The relationship between total knee replacement and 3D MRI knee bone shape: Data from the osteoarthritis initiative
  1. AJ Barr1,
  2. B Dube1,
  3. EMA Hensor1,
  4. SR Kingsbury1,
  5. G Peat2,
  6. MA Bowes3,
  7. LD Sharples4,
  8. PG Conaghan1
  1. 1Leeds Institute of Rheumatic and Musculoskeletal Medicine, University of Leeds and NIHR Leeds Musculoskeletal Biomedical Research Unit, Leeds, UK
  2. 2Arthritis Research UK Primary Care Centre, Research Institute for Primary Care & Health Sciences, Keele University
  3. 3Imorphics Ltd, Kilburn House, Manchester, UK
  4. 4Leeds Institute of Clinical Trials Research, University of Leeds, Leeds, UK


Background and objectives Imaging biomarkers of osteoarthritis (OA) structural progression are more accurate when derived from MRI than conventional radiography. Accurate 3D quantification of MRIs is achieved using active appearance modelling (AAM). Subchondral bone shape changes are integral to OA structural progression and are predictive of incident radiographic knee OA. We aimed to determine the relationship between scalar 3D bone shape and total knee replacement (TKR).

Materials and methods This case-control analysis, nested within the osteoarthritis initiative (OAI) cohort, matched 1:1 control knees that “survived” 6 years of follow up with case knees undergoing TKR, using stratification (propensity) score matching based upon baseline age, BMI, gender, knee side, ipsilateral knee pain severity and recruiting centre. AAM of the patella, tibia and femur were used to identify vectors that best classified knees as having OA vs. no OA, scaled such that -1 and +1 represented the mean non-OA and mean OA shapes, respectively. Within matched case-control pairs, the vector values were compared using paired t-tests and using conditional logistic regression, the odds of TKR per unit of 3D bone shape were obtained.

Results Case-control pairs (n = 311) of knees were well matched based upon propensity scores. Amongst the 311 well matched pairs, mean baseline 3D bone shape vector of TKR cases was more positive (more advanced OA) relative to controls, for the femur [mean 0.98 vs. -0.16; difference (95% CI) 1.14 (0.92, 1.37)], tibia [mean 0.86 vs. -0.05; difference (95% CI) 0.90 (0.69, 1.12)] and patella [mean 0.95 vs. -0.07; difference (95% CI) 1.02 (0.74, 1.31)]. Unadjusted conditional odds ratios (95% CI) for the femur, tibia and patella revealed increased odds of TKR with increasingly positive 3D bone shape vector values (increasing OA structural severity). After Kellgren Lawrence (KL) grade adjustment in a multivariable analysis, femoral 3D shape vector was independently associated with TKR [OR 1.21 (1.01, 1.45)] with an improvement in model fit (AIC) compared with KL grade.

Conclusions 3D bone shape is associated with TKR. Femur shape has the greatest association with TKR. This provides evidence of predictive validity of 3D bone shape and its potential utility in trials of prospective disease modifying OA drugs.

Statistics from

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.