Diagnostic performance of optical spectral transmission compared to magnetic resonance imaging in patients with inflammatory arthritis

Abstract

Optical spectral transmission (OST) is a modern diagnostic method capable of quantifying inflammation in the finger and wrist joints of arthritis patients by assessing the blood-specific absorption of light transmitted through a tissue. The diagnostic performance of this modality has not been adequately examined and data regarding OST associations with magnetic resonance imaging (MRI) are limited. Aim of this study was therefore to investigate the performance of OST in assessing joint inflammation as compared to MRI in patients with inflammatory arthritis (IA). Data from patients who underwent MRI and OST for suspected IA were analyzed. For comparison, a historical healthy control (HC) group with OST was also accounted. MRI findings were quantified using the Rheumatoid Arthritis MRI Score (RAMRIS). Diagnostic accuracy of OST was evaluated using Receiver Operating Characteristics (ROC), while correlation analyses were conducted to explore relationships between OST and MRI, as well as disease activity markers. Overall, 71 patients with known rheumatic diseases (n = 1,542 wrist and finger joints) and 114 HC (n = 2,508 joints) subjects were included. 51 patients showed inflammatory signs on MRI (MRI+). These also showed significantly higher OST scores (16.41 ± 5.53) than subjects without MRI inflammation (MRI-) (11.52 ± 5.03) or HC (10.78 ± 4.19) (all; p < 0.001). OST showed significant correlations with RAMRIS-synovitis and tenosynovitis scores in the MRI + group (rho = 0.541, p < 0.001; rho = 0.341, p = 0.01, respectively). Significant correlations were observed between OST and clinical parameters for disease activity. Using MRI as a reference, the best diagnostic value of OST was observed at the wrist level in the MRI + group, by an AUC of 0.833 (95%CI 0.700-0.966). OST showed an excellent performance compared to MRI and correlated significantly with RAMRIS scores and clinical parameters in IA patients, also differentiating IA from HC.