FRI0664 Influence of joint pathology on optical spectral transmission imaging, assessing inflammation in hand and wrist joints of rheumatoid arthritis patients

Abstract

BackgroundRheumatoid arthritis (RA) patients benefit from treat-to-target strategies, aiming for remission or low disease activity1. Clinical disease activity measures like the Disease Activity Score (DAS28) have questionable reproducibility and lack sensitivity for low disease activity states; MRI and ultrasound (US) are sensitive, but scanning multiple joints is time-consuming. Optical spectral transmission (OST) is stronger than DAS28 associated with inflammation assessed by both US and MRI2. OST measures the blood-specific absorption of light transmitted through tissue, which is reduced in presence of joint inflammation, but also influenced by other joint pathology.ObjectivesEvaluating the influence of joint pathology on the misclassification of joint inflammation, in the hand and wrist joints of RA patients, determined by OST, as compared to US, the reference standard.MethodsFifty RA patients with at least one swollen joint, generally with low disease activity were included in this cross-sectional study. Assessments were US, OST, and DAS28, performed according to established guidelines3 by separate experienced examiners, blinded for other study outcomes. US joint inflammation was defined as a gray-scale score>1 or a power Doppler score>0 (scales 0–3), assessed in MCP, (P)IP, and wrist joints. Using US as reference, diagnostic performance of OST in detecting inflammation at joint level was evaluated using receiver operating characteristic (ROC) analyses; at patient level, DAS28 and OST were correlated to US. Joint pathology potentially influencing misclassification of OST (erosions, osteophytes, tendon (sheet) inflammation, abnormal vasculature, and triangular fibrocartilage complex injuries) were evaluated for significance in a multivariate nominal logistic regression model.ResultsOST performed well at joint level, separately for the MCP (ROC-AUC:0.85), PIP (ROC-AUC:0.79) and wrist (ROC-AUC:0.72) joints and for all joints together (ROC-AUC:0.83). On patient level, DAS28 correlated poorly with US (r=0.29), but OST correlation was good (r=0.72). The presence of joint pathologies per misclassification group is shown in table 1. In the regression model, inflammation in MCP and PIP joints had a higher risk of false negative misclassification in the presence of dorsal bone erosions (OR:3.5, 95% CI:1.7–7.3), volar erosions (OR:5.0, 95% CI:1.8–14.1), flexor tenosynovitis (OR:2.5, 95% CI:1.4–4.5), osteophytes (OR:1.9, 95% CI:1.2–2.8), and extensor tendonitis (OR:3.7, 95% CI:1.6–8.5), and a higher risk of false positive misclassification in the presence of osteophytes (OR:2.3, 95% CI:1.6–3.2).ConclusionsOST is a sensitive and specific technique to assess inflammation in hand and wrist joints of RA patients with low disease activity, nonetheless, joint pathology like erosions, tendonitis, and osteophytes increase the risk of misclassification of inflammation by OST.References Bijlsma J et al. Early rheumatoid arthritis treated with tocilizumab, methotrexate, or their combination. Lancet 2016;388:343–55.Van Onna M et al....