Background Knee pain is the most common manifestation of knee osteoarthritis (OA) and is generally thought to get worse with time. Risk factors for knee pain have been extensively investigated; however, whether the risk factors are associated with a specific pain trajectory has not yet been comprehensively explored. Furthermore, knee structural pathology on MRI, such as bone marrow lesions (BMLs), effusion-synovitis and cartilage defects, are thought to be the origin of knee pain. However, its underlying mechanisms remain to be elucidated and hampered by a large individual variation of pain course.

Objectives To identify distinct trajectories of knee pain over 10.7 years in an older population, to describe risk factors with identified trajectories, and to explore MRI-detected structural pathology with the trajectories.

Methods 1,099 participants (mean age 63 years) from a population-based cohort study were recruited at baseline. 875, 768 and 563 participants attended years 2.6, 5.1 and 10.7 follow-up, respectively. Demographic, psychological, lifestyle and comorbidities data were obtained at baseline. Knee pain was assessed using Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) at each time-point. T1-weighted or T2-weighted MRI of the right knee was performed to measure knee structural pathology. Knee radiographic OA was assessed by X-ray at baseline. Group-based trajectory modelling was applied to identify pain trajectories. Multi-nominal logistic regression was used for the analyses with adjustment for potential confounders.

Results 1,099 participants were included for the identification of pain trajectories and three distinct pain trajectories were defined. Participants in Group 1 (“Mild pain” n=568, 51.7%) had relatively stable mild pain over time. Participants in Group 2 (“Moderate pain”, n=366, 33.2%) had moderate pain over time. Participants in Group 3 (“Severe pain”, n=165, 15.1%) developed or displayed fluctuating severe pain over time. Compared with the “Mild pain”, higher BMI, emotional problems, and musculoskeletal diseases were significantly associated with both “Moderate pain” and “Severe pain” trajectories. Also, younger age, lower education level and unemployment status were associated with “Severe pain” trajectory. The presence of cartilage defects and BMLs were associated with increased risk of “Moderate pain” and “Severe pain” trajectories before or after adjustment for potential confounders. Effusion-synovitis was not statistically associated with “Moderate pain” (P=0.082), but associated with “Severe pain” trajectory. Furthermore, a dose-response relationship was observed between number of knee structural abnormalities, and “Moderate pain” and “Severe pain” trajectories (both P for trend<0.001).

Conclusions This is the first long-term study to identify pain trajectories and their risk factors, and explore the associations between structural pathology and pain trajectories. Three distinct pain trajectory groups were identified, suggesting that homogeneous subgroups exist and follow a specific trajectory over time despite large individual variation of pain course. Significant associations between structural pathology and pain trajectories suggest that peripheral stimuli may play a role in the development and maintenance of pain severity.

Disclosure of Interest None declared