Differential diagnosis between psoriatic arthritis and hand osteoarthritis using indocyanine green-based fluorescence optical imaging

Benedict Drude^1*Øystein Maugesten²Stephanie Gabriele Werner³Jens Klotsche⁴Gerd Burmester¹Gerhard Krönke¹Marina Backhaus⁵Jörn Berger⁶Ida K. Haugen²Sarah Ohrndorf¹

• ¹Department of Rheumatology and Clinical Immunology, Charité University Medicine Berlin, Berlin, Germany
• ²Department of Rheumatology, Diakonhjemmet Hospital, Oslo, Norway
• ³RHIO (Rheumatology, Immunology, Osteology) Duesseldorf, Düsseldorf, Germany
• ⁴German Rheumatism Research Center (DRFZ), Berlin, Baden-Württemberg, Germany
• ⁵Park Clinic Weißensee, Berlin, Baden-Württemberg, Germany
• ⁶Xiralite GmbH, Berlin, Baden-Württemberg, Germany

Fluorescence optical imaging (FOI) visualizes enhanced microcirculation in the hands as a marker for inflammation. The correct diagnosis of psoriatic arthritis (PsA) and erosive hand osteoarthritis (EHOA) can be challenging. The aim of this study was to differentiate active PsA from EHOA using FOI.An atlas with FOI images of different grades of enhancement (FOIAS 0-3) and typical morphologic patterns ('streaky signals', 'green/blue nail sign', 'Werner sign', and 'Bishop's crozier sign') for PsA and hand EHOA was created. Twenty FOI sequences of patients with PsA and EHOA were randomly mixed and scored by two blinded readers. All images were scored twice by one of the two readers. Inter-and intra-reader reliability for joint enhancement, morphologic patterns and diagnosis (PsA vs. EHOA) was calculated. Subsequently, one reader blinded to the diagnosis scored the remaining PsA (n=54) and EHOA (n=47) images using the same atlas.Inter-reader reliability on joint enhancement was overall substantial (κ=0.74), with substantial to almost perfect intra-reader reliability (κ=0.88). Inter-reader reliability on all morphological patterns was fair (κ=0.36) and substantial (κ=0.68) in the intra-reader exercise with variation between the different patterns. Inter-(κ=0.3) and intra-reader reliability (κ=0.4) on diagnosis was fair. In analyses of all 101 cases (47 EHOA, 54 PsA), persons with EHOA presented significantly higher mean FOI sum scores in the PIP (38.98 vs 20.00) and DIP joints (16.45 vs 8.40) compared to the PsA patients. Regarding morphology, the 'Werner sign' was significantly more often detected in PsA than in EHOA (55.6% vs. 21.3%; p<0.01).Joint enhancement in FOI can be reliably assessed using a predefined scoring method. The stronger enhancement in PIP and DIP joints for EHOA cases and the occurrence of 'Werner sign' in PsA cases may facilitate the differential diagnosis between the two diagnoses.