Ab0039 expression of inflammatory markers in patients with painful knee osteoarthritis: a multivariate analysis of presurgical serum samples

Background Knee osteoarthritis (KOA) is a pathological condition prevalent in the general population. Exercise and pharmaceutical approaches are first-line treatments and the end-stage treatment is total knee replacement (TKR). Interestingly, around 20% of patients experience chronic postoperative pain and the reason for this is not fully understood. Pain in OA is multifactorial and low-grade chronic inflammation has been indicated as a potential cause. Pre-clinical evidence suggests that pro-inflammatory mediators, such as interleukin 6, can sensitize the peripheral and central nerves and these molecules might be associated to clinical pain. Clinical evidence has demonstrated differences in pro-inflammatory mediators when comparing patients with OA and healthy individuals. A recent study has linked these profiles to chronic postoperative pain after TKR but an in-depth analysis, as seen for neuropathic and widespread pain is needed to advance the field. Currently, no specific biomarkers have been identified, despite initiatives on focused molecular inflammatory mediators. In an attempt to advance the field, a comprehensive analysis of an extensive network of cytokines, chemokines, and growth factors could identify the role of low-grade inflammation in patients with OA and potentially stratify patients more prone to experiencing pain. Objectives This study aimed 1) to evaluate preoperative serum levels of 92 inflammatory biomarkers in KOA patients compared to healthy controls, 2) to investigate preoperative differences of inflammatory biomarkers within different subgroups of patients with KOA and link these subgroups to clinical pain before and after TKR surgery. Methods Blood samples from preoperative patients with KOA scheduled for TKR (n=200) and healthy participants (n=39) were collected. After centrifugation of the serum was frozen at -80C until analysis. Serum samples were analyzed for inflammatory markers using the OLINK inflammation panel, which included 92 protein markers. Clinical pain was assessed using a Visual Analog Scale (VAS). Moreover, patients completed the Knee injury and Osteoarthritis Outcome Score (KOOS) questionnaire before and 12 months after TKR. Multivariate data analysis was performed to identify differences between patients and controls. Hierarchical cluster analysis (HCA) and Orthogonal Partial least squares discriminant analysis (OPLS-DA) was used for comparing groups (patients vs controls) and to identify subgroups within patients. T-tests were used to evaluate difference within the KOA cohort in terms of VAS and KOOS scores before and 12 months after TKR. Results Multivariate analysis showed that 12 proteins were differentially expressed between patients and controls (P<0.05). Hierarchical cluster and OPLS-DA analysis identified two patient subgroups (pat-1, n = 46; pat-2, n= 72) and 23 proteins were dysregulated comparing these two groups (p<0.01). Postoperative VAS and KOOS assessments were significantly different between the two subgroups (p-<0.05). Conclusion The present study suggest a low-grade inflammation in patients with KOA when compared to healthy pain free subjects. Additionally, this study suggests that a high inflammatory subgroup for patients with KOA exist and this group is likely to have more clinical and worst function 12-months after TKR. Acknowledgements Supported by the Danish National Research Foundation (DNRF121). Disclosure of Interests None Declared.