Biological measures of synovial inflammation predict pain, symptoms, and ultrasound measures of synovitis in knee OA

Abstract

Purpose: Synovial inflammation (synovitis or effusion-synovitis) measured via clinical imaging (i.e., magnetic resonance imaging (MRI), ultrasound (US)) has been shown to be related to worse outcomes in people with knee OA, including worse pain, higher risk of OA incidence and progression, and increased risk of requiring a total knee arthroplasty (TKA). Recent studies have shown associations between synovial histopathology and structural features of knee OA and MRI measures of synovitis. However, the relationship between histopathology and patient symptoms remains unclear. Histopathology measures of synovial inflammation represent a set of measures that are related to biological disease activity in OA. Therefore, the aim of this study was to establish the link between synovial inflammation and clinical outcomes. Methods: Patients were recruited from the Western Ontario Registry for Early Osteoarthritis (WOREO): Knee Study. Ninety-two people with severe symptoms and radiographic damage (Kellgren-Lawrence (KL) grades 3 or 4) undergoing surgery for end-stage knee OA were included. Patients completed the Knee Injury and Osteoarthritis Outcome Score (KOOS) questionnaire specific to each knee and a self-reported patient global assessment (PGA). An investigator global assessment (IGA) was also completed for each patient. Musculoskeletal-US was used to measure effusion-synovitis in suprapatellar mid-line, lateral, and medial windows (0-3; None-Severe; summed (0-9) and maximum effusion depth (mm). Power Doppler signal was also assessed as absent or present (0 or 1). Lateral suprapatellar synovial biopsies were obtained at time of surgery. Routine histopathology was scored in six domains, on 5 slides per patient; mean scores were binned into categories 0, 0.5-1.5, and 2-3 (none, mild, moderate-severe). Composite histopathology scores were created based on previously published literature, principal component analysis (Eigen values ≥ 1.0), and biological rationale. We fitted a series of multivariate linear or logistic regression models to evaluate the association of histopathology (individual and composite) scores with clinical outcomes and US imaging measures of synovitis. All analyses were adjusted for age, sex, and body mass index (BMI). Robust sandwich estimators were used. Results include unstandardized β coefficients or odds ratios (OR) with 95% confidence intervals (CIs). All analyses completed using StataIC 15.1. Results: A total of 92 patients were included (Table 1). Of the 6 histopathology domains, more severe perivascular edema was most strongly associated with lower (worse) KOOS pain, ADL, and QoL subscale scores (Table 2). Synovial lining and sub-synovial infiltrate scores were significantly associated with more severe effusion-synovitis measured by US grading and maximum effusion depth (Table 3). Vascularization was associated with higher synovitis score, while more severe fibrosis was associated with high QoL scores (Table 2) and lower US synovitis score (Table 3). Synovial histopathology principal components accounted for approximately 70% of the variance in the dataset but there were minimal associations with clinical and imaging outcomes (Table 4). The histopathology inflammatory composite scores were most strongly associated with KOOS subscale scores (Table 4). Conclusions: Inflammation measured with synovial histopathology is associated with knee-specific pain, symptoms and US imaging measures of inflammation in patients with end-stage knee OA. These findings may help us to better understand OA overall. We provide biological evidence supporting the known relationship between knee inflammation and OA outcomes, which is largely based on imaging assessment in the literature.View Large Image Figure ViewerDownload Hi-res image Download (PPT)View Large Image Figure ViewerDownload Hi-res image Download (PPT)View Large Image Figure ViewerDownload Hi-res image Download (PPT)