Background The clinical relevance of pain sensitization in knee osteoarthritis (OA) has been highlighted, whereby individuals with more severe knee pain demonstrate widespread hyperalgesia and increased temporal summation report. Physiotherapy is a frequently recommended conservative intervention for knee OA, however not all patients achieve clinically meaningful improvements in pain and function. To date the relationship between QST parameters and subsequent response to physiotherapy has not been prospectively investigated.
Objectives The purpose of this prospective cohort study was to compare the somatosensory characteristics of people with knee OA who responded to physiotherapy with those who did not show a clinically meaningful response. We hypothesized that the non-responders to physiotherapy would demonstrate more evidence of pain sensitization at baseline.
Methods A comprehensive battery of Quantitative Sensory Testing (QST) was conducted in 122 participants with knee OA referred for physiotherapy. QST measures included pressure pain thresholds (PPTs) measured at local, distal and remote sites, temporal summation (TS) at local and remore sites and conditioned pain modulation (CPM), in addition to measuring vibration threshold and mechanical detection thresholds.
Following physiotherapy participants were classified into responder and non-responder groups based on their change in WOMAC pain and function scores, and participants' global rating of change using the Outcome Measures in Rheumatology- Osteoarthritis Research Soceity International (OMERACT-OARSI) criteria.
A one-way between-groups multivariate analysis of variance (MANOVA) was performed to investigates effects of group (responder/non-responder) differences on QST parameters using a Bonferroni adjusted alpha level (p=.0083).
Results 43 participants (35%) were responders to physiotherapy and 79 participants (65%) were non-responders. There were no significant differences in gender and age between the two groups. Likewise there were no significant differences between the groups with respect to sensory hypoesthesia (vibration and light touch).
The responder group demonstrated significantly higher PPTs at the arm (M =375kPa, SD =93.5) and tibia (M =378 kPa, SD =116) compared to the non-responder group (arm PPT, M =330 kPa, SD =85.2; tibia PPT, M =318.5, SD =96). Significantly lower temporal summation was also demonstrated in the responder group at the arm (M=1.1, SD=1.2) and at the knee (M =1.9, SD =1.6) compared to the non-responder group (TS arm M =1.9; SD =1.3; TS knee M =2.8, SD =1.46). The responder group also showed a more efficient CPM response (103% CPM change) compared to the non-responder group (99% CPM change), however these differences were not statistically significant (p=.056).
Conclusions In knee OA non-responders to physiotherapy have a different somatosensory profile, showing increased hyperalgesia and enhanced temporal summation. While the results do not address predictors of treatment response, participants with greater indications of pain sensitization were more likely to be “non-responders” to physiotherapy. Patients identified at baseline as less likely to benefit from physiotherapy may warrant an alternate treatment approach.
Disclosure of Interest None declared