Background Fluorescence optical imaging is a novel tool for assessment of disease activity of inflammatory joint disorders of the hands (1). Initially developed for assessment of inflammatory activity in the hands, represented by an altered microcirculation, it became quickly apparent, that different underlying diseases appear with a variation of image patterns. Optical Evaluation of Rheumatoid Arthritis (OPERA1) is a prospective, multicenter use study with optical imaging in a variety of inflammatory joint disorders of the hands.
Objectives With this analysis on a small sample of patients from the OPERA study, the feasibility of fluorescence optical imaging to provide differential diagnostic patterns was evaluated. The hypothesis under evaluation was that rheumatoid arthritis (RA), psoriatic arthritis (PsA), and osteoarthritis (OA) have distinct image patterns in fluorescence optical imaging.
Methods From the OPERA1 study, 120 patients with RA, with PsA, and with OA (40 each) were randomly selected. All image sets were evaluated by a single, experienced radiologist to a previously defined set of variables representing image patterns and dynamic image characteristics. First, univariate multinomial regression analyses were performed to determine the significance of factors to differentiate between at least two of the three diagnoses (RA, PsA, OA). Significant factors were included into a multivariate model using backward selection of factors to obtain a final model using p-values ≥0.05 as exclusion criterion. Significant results suggest that patterns of significant factors may provide potential for differentiating between the diagnoses.
Results From the factors evaluated, the image pattern (p<0.0001, fig. 1) and signal continuity between DIP and nailbed (p=0.0027) were found to significantly differentiate between RA, PSA, and OA in the final multivariate analysis. In the univariate analyses, also late phase signal (p=0.0016), overall signal intensity (p=0.031), a discontinuous signal in the phalanx (p=0.0056), and an arcuate signal pattern in the phalanx (p=0.0106) were found as significant factors. Early enhancement (p=0.295) and duration of the enhancement (p=0.064) were not found to significantly differentiate between the three diagnoses.
Conclusions This single reader feasibility study shows, that the hypothesis of distinct imaging patterns in fluorescence optical imaging depending on the underlying disease may hold true. A well designed prospective, blinded multi-reader analysis needs to be performed for further evaluation of this hypothesis and its proof.
Werner SG et al. Inflammation assessment in patients with arthritis using novel in vivo fluorescence optical imaging technology. Ann Rheum Dis 2012, doi:10.1136/ard.2010.148288
Disclosure of Interest M. Bahner Shareholder of: mivenion GmbH, Employee of: mivenion GmbH, C. Schwenke: None Declared, M. Schirner Shareholder of: mivenion GmbH, Employee of: mivenion GmbH