Investigating the efficacy of intra-articular corticosteroid injections in osteoarthritis versus inflammatory arthritis: impact on inflammatory pain and sensitisation outcome measures

      Purpose: Intra-articular corticosteroid injections are widely used in clinical practice for the treatment of flare-ups in osteoarthritis (OA) and inflammatory arthritis (IA). However, different studies have reported varying efficacy of intra-articular corticosteroid injections for knee flare in distinct forms of arthritis. A meta-analysis study by Mikhail et al. (2020) evaluated the efficacy of intra-articular knee corticosteroid injections in OA participants. With 8 studies included, Mikhail et al. concluded there was a reduction on pain severity for up to 3 months after intervention. In contrast, a 2009 systematic review by Hepper et al. concluded that intra-articular corticosteroid ‘decreases pain by one third...but provides that benefit for only 1 week’. Addressing rheumatoid arthritis, a type of IA, there are limited studies directly addressing the potential benefit of intra-articular corticosteroid injections for pain management in these patients. However, a 2006 Cochrane review by Wallen and Gillies investigating the efficacy of intra-articular knee injections for the management of RA, concluded that intra-articular injections improved pain and morning stiffness for up to 22 weeks, compared to 1 week in the placebo group. Furthermore, the impact of intra-articular corticosteroids on inflammatory and pain sensitization components of pain are not well understood. We hypothesised that subjects with OA and IA will have different outcomes to corticosteroid injections for knee flares.
      Methods: We recruited subjects presenting to a London Teaching Hospital for a knee flare with full informed consent ( Identifier: NCT03533569). Participants were recruited from Rheumatology or Orthopaedic services by clinician referral when a corticosteroid knee injection was clinically indicated as standard of care. All subjects underwent corticosteroid injection with 40mg depomedrone and 1-2% lignocaine with full aseptic technique. Injections were performed by a qualified clinician in clinic or under ultrasound guidance. Criteria for entry into the study included a Visual Analogue Scale (VAS) for pain in the target knee of at least 5/10, swelling or stiffness, fulfilment of ACR criteria for osteoarthritis (OA) or ACR/EULAR criteria for rheumatoid arthritis (RA), psoriatic arthritis (PsA), spondyloarthritis (SpA) respectively depending on the diagnosis. Clinical measures for pain were recorded at baseline before intra-articular injection and then 3 months after corticosteroid injection. These included the VAS for pain 0-10, the painDETECT scale for assessment of pain sensitisation (ranging 0-35) and the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) for pain, function and stiffness subscales. Graphpad Prism software Version 8 was used for all statistical analyses. Planned analyses included initial comparison to detect any significant differences between baseline and 3 months after corticosteroid injection for VAS pain, painDETECT and WOMAC outcomes (Mann Whitney U test). Differences between the two groups was assessed by ANOVA. A p value < 0.05 was considered to be statistically significant.
      Results: A total of 64 subjects with distinct forms of arthritis were recruited. Of these, 46 subjects had knee OA and 18 subjects had IA. The mean age (SD) for the OA group was 62.8 (± 9.9) and for the IA group was 62.7 (± 14.1). The OA group had a higher mean BMI of 33.2 (± 9.8) compared with the IA group mean of 29.7 (± 5.5). Pain outcome results are summarised in Figure 1. Participants with knee OA reported a significant improvement in pain measured by VAS 3 months after injection (p<0.0001) compared with the inflammatory arthritis group (p=0.7). Measurement of pain sensitisation using the painDETECT questionnaire suggested sensitisation components in both groups, with mean (95% CI) reported as 15.4 (13.2 to 17.6) in the OA group versus 14.7 (11.0 to 18.3) in the inflammatory arthritis group. The painDETECT score for sensitisation improved significantly in the OA group 3 months after corticosteroid injection (p=0.002) but not in the inflammatory arthritis group (p=0.6). For WOMAC outcomes, significant improvements in knee pain were observed 3 months after steroid injection in the OA group (p=0.005) and inflammatory arthritis group (p=0.03). For WOMAC function, a significant improvement in outcome was observed in the OA group (p=0.003) but not in the inflammatory arthritis group (p=0.23). Similarly, for WOMAC stiffness, a significant improvement in outcome was observed in the OA group (p=0.02) but not in the IA group (0.82) after corticosteroid injection.
      Conclusions: Our data shows a significant improvement in knee pain measured by VAS, WOMAC pain and function in participants treated with corticosteroid injection for flares of knee OA. Subjects with knee OA also had pain sensitisation measured by painDETECT that improved significantly after intra-articular corticosteroid injection. In contrast, in participants with IA, the overall VAS pain or painDETECT score did not improve significantly 3 months after corticosteroid injection, but knee specific pain assessed by WOMAC showed an improvement in OA and IA subjects. Traditionally knee flares are triggered by knee effusion and synovitis. Our data show that inflammatory and sensitisation pain components in knee OA are improved with intra-articular corticosteroid injection. In contrast, in subjects with IA, pain measures were not significantly improved, suggesting that interventions in systemic inflammatory arthritides require additional treatment to intra-articular knee injections in order to have an impact on improving outcomes for disease flares.