Background/Purpose: Significant challenges remain in mitigating OA pain and despite growing evidence of sex differences, including in pain and in response to NSAIDs, sex has received little attention in clinical studies of potential OA drug treatment strategies. In this study, we investigated the association between knee-specific OA pain and systemic markers of inflammation, focusing on whether associations differed by sex.

Methods: Study participants were 231 patients with knee OA scheduled for total joint arthroplasty at a tertiary care hospital in Toronto, Canada. Eligibility: 35+ years of age; English fluency. Exclusions: acute trauma/injury or inflammatory arthritides. Health questionnaires were completed and blood samples drawn in-clinic prior to surgery. Questionnaires collected data on knee pain (WOMAC pain subscale) and sex, age, height and weight (used for BMI calculation), comorbidity, depressive symptoms (HADS depression scale), symptomatic joint count, and pain medication use. Blood cytokine analyses included IL-6, -8, -10, -1β and TNF-α using Luminex bead-based ELISA assays. A series of linear regression models were estimated with knee pain as the outcome. The final model included sex, age, BMI, comorbidity count, depressive symptom score, symptomatic joint count, log-transformed cytokine concentrations, and interaction terms between sex and any factors found to have sex-specific influences. Sensitivity analyses: final model a) in individuals aged 55+ only (age-based proxy for post-menopausal status) and b) with consideration for medication use.

Results: Participants were of mean age 65 years (range 43-89); women comprised 57.6% of the sample. Women had higher comorbidity and joint counts, and worse depressive symptoms and knee pain scores than men. No sex differences were found in median cytokine concentrations, except for higher TNF-α in men (8.1 vs. 6.9 pg/ml; p=0.006). In adjusted linear regression, initially not allowing for sex-specific effects, only 2 cytokines were significantly associated with knee pain scores; higher IL-10 with lower pain (p=0.002), and higher TNF-α with worse pain (p=0.031). With consideration of potential sex-specific effects, the association between knee pain and IL-10 and TNF-α remained unchanged and was similar for women and men. However, the association between knee pain and IL-8 (interaction term p<0.001), IL-6 (interaction term p=0.023) and IL-1β (interaction term p=0.047) differed significantly between women and men. In addition, higher BMI for men, and higher depressive symptoms and joint count for both sexes were significantly associated with worse knee pain scores. Findings from sensitivity analyses were consistent with primary analyses.

Conclusion: While cytokine-targeted treatments exist, indiscriminate use in OA populations may limit effectiveness. Our findings provide evidence of sex-specific associations between individual inflammatory cytokines and knee OA pain, and suggest that sex-specific targets of anti-inflammatory treatments for OA need to be considered.

« Back to 2018 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/sex-specific-associations-between-systemic-inflammatory-cytokines-and-osteoarthritis-knee-pain/