POS1096 THE EFFECT OF ONE DRY NEEDLING SESSION ON PAIN AND CENTRAL PAIN PROCESSING IN PATIENTS WITH KNEE OSTEOARTHRITIS: A RANDOMIZED CONTROLLED TRIAL

Abstract

Background: Osteoarthritis (OA) is the leading cause of chronic disability in the elderly1,2. Abnormal central pain processing (CPP) is present in around 30% of the knee OA patients3 and can be partly induced by peripheral nociception through long term potentiation4. An attempt to resolve abnormal CPP can be to eliminate this nociception5. Myofascial trigger points (MTrPs) are often present in knee OA6, can lead to nociception7 and therefore abnormal CPP if prolonged present8. These are usually defined as hypersensitive tender spots within taut bands of a muscle9. Additionally, both MTrPs and knee OA can induce disturbed motor control, increased co‐antagonist activation and modified gait pattern10,11. Dry needling (DN) is often used to deactivate the MTrP and thus resolve the source of nociception, which normally results in reducing pain and restoring muscle dysfunction12. However, studies about the effect of DN on CPP and other outcomes than pain are very limiting. Therefore, more high‐quality studies concerning DN and its effects on CPP, muscle features and gait are needed13,14. Objectives: The aim of this randomized controlled trial is to assess the effect of one DN session compared to one sham needling (SN) session on pain (processing), muscle activity and gait in patients with knee OA. Methods: 61 patients participated of which 31 were allocated to the DN and 30 to the SN group. Each patient underwent one treatment session. Primary outcomes were pain intensity, measured with questionnaires; and CPP, measured with quantitative sensory testing. Secondary outcomes included muscle co‐activation, measured with electromyography; and spatiotemporal parameters, measured with gait analysis. Patients were assessed at baseline, 15 minutes (post 1) and 3 days after intervention (post2‐only for the outcome pain). Linear mixed models was used to identify the possible differences over time between the two therapy modalities. Results: The following significant within group differences were observed: decreased pain, stride‐and step time and increased widespread pain pressure threshold and step length. A significant between group difference of the conditioned pain modulation score was found, whereas the SN group showed a decrease in difference between the pain pressure threshold scores (with and without conditioning stimulus) and the DN group remained stable. No other significant between or within group differences were detected. However, if we compared both interventions, the change over time for the visual analogue scale (VAS) behaved different in the DN (p<0.05, post 2 ‐baseline) and SN group (p<0.05, post 1 ‐baseline and post 2 ‐post 1). Conclusion: One DN session has no larger effect on all outcome measurements compared to SN. Both therapies seem to be useful to improve pain and widespread pain pressure threshold in short term, however the improvement of pain differs in the groups. Although improvements in some spatiotemporal parameters were observed, it is uncertain they are of clinical relevance or related to treatment. More research is necessary to reveal the ideal number of sessions of DN to improve outcomes and to reveal the effect of DN compared to no treatment, as SN could have hide the real treatment effect. (Figure Presented).