POS0130 NEAR INFRARED FLUORESCENCE OPTICAL IMAGING (NIR-FOI) CAN SUPPORT THE DIFFERENTIAL DIAGNOSES OF CONNECTIVE TISSUE DISEASES

Abstract

BackgroundDifferentiation between degenerative osteoarthritis (OA) and autoimmune caused connective tissue diseases (CTD), or rheumatoid arthritis (RA) is important for effective therapy. Anamnesis, clinical examination, and analysis of laboratory parameters are helpful for diagnosis, but do not always lead to clear results. One problem is that patients often show overlapping symptoms. Particularly, imaging techniques as NIR-based fluorescence optical imaging (FOI) are playing an increasingly role as quantifiable and sensitive diagnostic methods. After injection of a fluorescent dye, changes in the microcirculation caused by inflammation can be detected due to altered blood flow and accumulation in anatomical structures of the hands. Analyses of enrichment features can be used to identify pathological changes and to support the diagnosis of degenerative and autoimmune diseases.ObjectivesThe present work investigates the potential of NIR-FOI feature assessment to differentiate between various rheumatic diseases, which are accompanied by pathological changes in connective tissues.MethodsNIR fluorescence imaging comprised 360 images/6 min. after injection of the dye. Analyses of patients with the clinical diagnosis of OA (n = 236) were compared with RA patients (n = 235), and patients diagnosed with autoimmune CTD (collagenoses) (n = 149). 17 features known to occur in FOI images were analyzed in both hands in three time periods (P1-3) after injection in two independent passes. In some cases, Prima Vista images (PVI), representing the sum of images 1-240 were also analyzed (n = 45, 33, 33 for OA, RA, CTD). Signals compared to the background were assessed as positive or negative (Figure 1). The data were analyzed statistically.Figure 1.Examples of significantly different features in NIR-FOI images of patients with clinically diagnosed OA, RA, or CTD. Top: the whole hands, also with nail bed injuries in phase 1 (P1) for a CTD patient (right), bottom: increased intensity in muscle-tendon junction in the forearm in phase 2 (P2) for an OA patient, increased intensity in MCP joint in phase 1 (P1) for an RA patient and a secondary Raynaud syndrome for a CTD patient in PVI.ResultsComparing the analyses of all phases of the OA cohort with those of the RA cohort, 14 of 51 features were identified as significantly different (p<0.05). The dye accumulation in metacarpophalangeal joints (MCP) has high specificity and diagnostic odds ratio (DOR) for RA patients (91%, 4.32). Patients in the OA cohort showed an increase in the signal at the muscle-tendon junction in the forearm (80% specificity, DOR 3.11). A comparison of the RA and CTD cohorts revealed 24 significant differences, most prominent changes in the nail bed in P1 (100% specificity, DOR 8.27) and a punctate accumulation pattern in P2 (98% specificity, DOR 2.61) for the CTD cohort. Comparing the RA and CTD cohorts, 22 features were significantly different. The strongest differences were found in the PIP joints of RA patients (78% sensitivity, DOR 2.97) and in the nail bed of CTD patients (100 % specificity, DOR 8.18). In the cumulative PV images a high specificity for a secondary Raynaud’s syndrome in CTD patients was found as compared to RA (97%, DOR 17.7) and OA (91%, DOR 5.9). Further, DIP signals for OA comparing to CTD show significant differences (76 % sensitivity, DOR 2,73).ConclusionThe present work demonstrates the detection and localization of specific, significant features in NIR-FOI of patients with different rheumatic diseases and can thus make an important contribution to diagnosis and optimization of therapy. In future, multivariate analysis and artificial intelligence algorithms can combine these features to further improve the diagnostic value.AcknowledgementsFunded by the Federal Ministry of Education and Research (grant nb.; 13GW0341A)Disclosure of InterestsDenise Kiesel Employee of: Xiralite GmbH, Jörn Berger Employee of: Xiralite GmbH, Egbert Gedat Employee of: Xiralite GmbH, Sarah Ohrndorf: None declared, Andreas Briel Shareholder of: Xiralite GmbH, Employee of: Xiralite Gmbh, Vieri Failli Employee of: nanoPET GmbH, Pia Welker Employee of: nanoPET GmbH