What makes osteoarthritis painful? Evidence of central sensitization in a cohort of subjects with hand osteoarthritis

Nidhi Sofat1, Cori Smee1, Monika Hermansson1, Matthew Howard1,2, Emma Baker1, Franklyn Howe1, Tom Barrick1. 1St George’s, University of London, London, UK, 2Institute of Psychiatry, King’s College London, London, UK

Background: Hand osteoarthritis (OA) is a prevalent condition for which treatments are based on analgesia and physical therapies. However, the majority of patients continue to have symptoms of pain and reduced function despite optimisation of current available treatments. We hypothesised that inflammation in the hand joints due to osteoarthritis enhances sensitivity and firing of peripheral nociceptors, thereby causing chronic pain.

Objective: We aimed to evaluate the characteristics of pain perception in a cohort of participants with hand OA to evaluate whether central sensitization contributes to pain perception in hand OA.

Methods: Participants with proximal and distal interphalangeal joint hand OA (n = 13) and non-OA controls (n = 13) were recruited. Clinical parameters of local joint disease including VAS (visual analogue score) for pain (0-100 mm scale), HAQ (health assessment questionnaire), Kellgren-Lawrence scores measuring radiological severity (scale 0-4) and pain thresholds using an algometer were assessed in Newtons. Central brain pain processing in all participants in the cohort was then evaluated using a standardised finger flexion-extension task. Subjects underwent brain functional magnetic resonance imaging (fMRI) whilst they performed the finger flexion-extension task to measure central components of pain processing. Activation of central pain processing pathways was then evaluated using group analyses with FMRIB software (www.fmrib.ox.ac.uk/fsl).

Results: All hand OA participants reported pain despite 92% taking oral analgesic drugs. The mean VAS in hand OA participants was 59.31 mm +/- 8.19 compared with 4.00 mm +/- 1.89 in the control group. Hand OA participants also had HAQ scores 8-fold higher than controls, indicating high levels of functional impairment. Objective measures of pain using algometers on 30 hand joint regions per subject (15 joint regions per hand) showed lower pain thresholds in the OA group versus controls (p<0.0001). There was a global reduction in pain threshold in the hand OA group despite the main radiological changes being found in the proximal and distal interphalangeal joints. Pain threshold in the OA group did not vary significantly with increasing radiological severity. Brain functional MRI during the painful finger flexion-extension task demonstrated increased activation of the thalamus, cingulate gyrus, frontal and somatosensory cortex in the hand OA group that was not observed in the control group (p < 0.05). The activated regions observed by fMRI in the hand OA group also showed a positive correlation with patient-reported VAS scores that mapped to pre-frontal limbic regions of the brain.

Conclusions: Our data demonstrate that hand OA subjects are sensitized to pain. Hand OA subjects demonstrate lower pain thresholds globally in their hands compared with controls. We found that hand OA subjects also have enhanced central brain sensitization as demonstrated by increased activation of central brain pain processing pathways. Our findings suggest that pain processing in OA is associated with activation of brain regions mediating emotion and fear as evidenced by upregulation of the pre-frontal limbic regions in the hand OA group. Development of future therapies could be targeted at modulating activation of central pain processing centres in OA to improve pain management in patients.