Article Text

THU0452 The Relationship Between 3D MRI Knee Bone Shape and Prevalent and Incident Knee Pain: Data from the Osteoarthritis Initiative
  1. A. Barr1,
  2. B. Dube1,
  3. E. Hensor1,
  4. S. Kingsbury1,
  5. G. Peat2,
  6. L. Sharples3,
  7. M. Bowes4,
  8. P.G. Conaghan1
  1. 1Leeds Institute of Rheumatic and Musculoskeletal Medicine, University of Leeds, Leeds
  2. 2Arthritis Research UK Primary Care Centre, Research Institute for Primary Care & Health Sciences, Keele University, Keele
  3. 3Leeds Institute of Clinical Trials Research, University of Leeds, Leeds
  4. 4Imorphics Ltd, Kilburn House, Manchester, United Kingdom


Background Incident frequent knee symptoms may represent early knee OA. MRI-detected OA bone pathology is associated with prevalent frequent knee OA symptoms (PFKS) and incident frequent knee OA symptoms (IFKS) in pre-radiographic knee OA. Active appearance modelling (AAM) enables accurate, 3D quantification of MRIs. Changes in 3D subchondral bone shape derived from AAMs are predictive of incident radiographic knee OA. However, the association of 3D subchondral bone shape with onset of knee symptoms is unknown.

Objectives To determine the relationship between scalar 3D bone shape and PFKS and IFKS in individuals at increased risk of knee OA.

Methods AAMs of the femur, tibia and patella and linear discriminant analysis identified vectors best classifying knees having OA vs. no OA. Vectors were scaled such that -1 and +1 represented the mean non-OA and mean OA shapes, respectively. Using a subcohort of 1114 persons with Kellgren Lawrence (KL) zero in both knees at the 12 month visit from the osteoarthritis initiative (OAI) we assessed whether 3D bone shape vector was associated with PFKS or IFKS. We defined PFKS as (pain, aching or stiffness) or medication use for knee symptoms most days of 1 month in the past 12 months. IFKS was defined as those lacking PFKS at baseline but reporting PFKS at any two consecutive annual OAI visits between the 12 and 60-month visits. Logistic regression, using one knee per individual, was used to evaluate the association between 3D bone shape vectors of the femur, tibia and patella and each of PFKS at the 12 month visit and IFKS. All models were adjusted for age, sex, BMI, previous knee injury and previous surgery.

Results The 3D bone shape vectors of the femur, tibia and patella were not associated with PFKS in the univariable models at the 12 month visit. Adjusted odds ratios (95% CI) for the femur, tibia and patella were; 1.00 (0.71, 1.26), 1.12 (0.90, 1.41), 0.84 (0.66, 1.08) respectively, all p>0.05. The 3D bone shape vector was significantly associated with IFKS in the univariable model for the tibia only. The 3D bone shape vectors of the femur, tibia and patella were not associated with IFKS after adjustment for covariates (Table 1). However BMI, previous knee surgery and knee injury were significantly associated with PFKS and IFKS.

Table 1.

The association between 3D bone vector measures at the 12 month visit and risk of incident frequent knee symptoms by the 60-month visit

Conclusions 3D bone shape is not associated with prevalent or incident frequent knee pain in individuals without radiographic OA but at increased risk of knee OA.

Disclosure of Interest None declared

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.