Background Mechanisms of pain in osteoarthritis (OA) include mechanical factors, inflammation and sensitisation to pain. However, the relation of OA joint structural abnormalities to pain sensitisation in people presenting for total knee replacement (TKR) are not well understood.
Objectives To establish if specific components of tissue injury in OA, including cartilage damage, synovitis, effusion and bone marrow changes, detected by magnetic resonance imaging (MRI), could provide the necessary peripheral nociceptive inputs to induce pain sensitisation in people with knee OA.
Methods The Pain Perception in Osteoarthritis Study (PAPO) is a longitudinal cohort of people undergoing TKR (UKCRN ID 15707). Participants had WOMAC scoring, MRIs (3T Phillips) of the target knee and standardised somatosensory assessment of pain pressure thresholds (PPT) before TKR and after surgery. PPT were measured with algometry (Somedic) (1 cm2 tip, 0.5 kg/sec) recorded at the point where subjects reported pressure change to pain. Recordings (x3) were taken at 13 points including the target knee, contralateral knee and distal forearms. The MRI Osteoarthritis Knee Score (MOAKS) assessed degree of synovitis, effusion, cartilage damage and bone marrow lesions (BMLs). MOAKS scoring was by 2 independent Consultant Radiologists and a consensus score reached. The scoring assessed presence of each lesion at a level of 1 (<33%) of subregional volume (SV), 2 (33-66%) of SV and 3 (>66%) of SV. The relation of synovitis/effusion, cartilage damage, BMLs and PPT was assessed by Mann Whitney testing for specific MOAKS grades.
Results A total of 60 knees were analysed with 1170 points tested for PPTs. Demographics showed a mean age 68.5, with 86.7% female. The BMI range was 22.9 - 49.1, mean 32.3, SD 6.7. The WOMAC pain score range was 70.5 - 480 out of 500, mean 306.4, SD 121. All participants had a mean lesion size detected by MOAKS scoring for synovitis, cartilage damage and bone lesions, graded 1, 2 & 3 (see Table for prevalence). Our group had globally reduced PPTs in the symptomatic knee, contra-lateral knee and distal sites compared with published normal control data. Results (shown in Table) demonstrated higher severity of synovitis/effusion scores associated with a significantly higher PPT (p=0.04) i.e. less sensitised. Greater BML severity was associated with lower PPTs i.e. more sensitised, but did not reach significance (p=0.33).
Conclusions People with severe knee OA presenting for TKR have significant pain reported subjectively by WOMAC and globally reduced PPTs compared with non-OA populations. Synovitis/effusion may represent more inflammatory components of OA pain, with BMLs contributing to sensitisation components of OA pain perception.
Acknowledgements We gratefully acknowledge NIHR, Pfizer Neusentis and The Rosetrees' Trust for funding. The funders had no role in the study concept, design, conduct or data interpretation.
Disclosure of Interest None declared