Bidirectional longitudinal relationships between MRI-detected structural changes and contralateral knee pain in knee osteoarthritis: Data from the Osteoarthritis Initiative.

Objective: To examine bidirectional longitudinal associations between MRI-detected structural changes and pain between knees (within a person) with or at risk for osteoarthritis (OA). Design: Using data from the Osteoarthritis Initiative (OAI) with a bidirectional knee-based design, each participant contributed two observations by alternating one knee as the exposure knee and the other as the contralateral (outcome) knee. MRI-based composite metrics of disease activity and cumulative damage were assessed over two years. Knee pain was measured using The Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC). Multinomial logistic regression estimated associations between baseline or changes in MRI from one knee and changes in the contralateral knee pain. Linear regression assessed associations between baseline or changes in structural metrics of one knee and changes in the contralateral knee. Results: The sample included 625 knee pairs (1250 knees) from 303 participants, with some contributing data from multiple intervals. Baseline measures or longitudinal changes in disease activity or cumulative damage in the one knee were not associated with two-year changes in pain in the contralateral knee. Two-year changes in disease activity (adjusted β = 0.16, 95% CI: 0.05 to 0.27) and cumulative damage (adjusted β = 0.16, 95% CI: 0.03 to 0.30) were associated with contralateral knee disease activity and cumulative damage changes, respectively. Conclusions: Structural worsening in one knee was not associated with contralateral pain but correlated with concurrent structural worsening in the other knee, suggesting parallel bilateral progression and the need to consider both knees in OA studies and therapeutic trials.