Difficult-to-treat concept in Psoriatic Arthritis: analysis of two potential definitions in a large group of patients. A cross-sectional study

Objective: The main aim of the study was to evaluate the performance of two proposed criteria for difficult-to-treat (D2T) Psoriatic arthritis (PsA) in a group of patients and to evaluate the agreement between the two sets of criteria. Methods: We performed a cross-sectional analysis of two longitudinal cohorts of PsA patients, fulfilling the CASPAR criteria with at least 1-year follow-up. A detailed medical history and physical examination were collected for all recruited patients. The proposed criteria for D2T PsA patients were applied in our group. To test the performance of the two sets of criteria, we use an external validator (the absence of patient acceptable symptom state (PASS no)+ physician global assessment (PhGA)≥6 cm). Finally, the agreement between the two sets of criteria was assessed. Results: we evaluated 378 patients with PsA (M/F: 219/159), mean age (range) 58 (19-75) years, median follow-up 6 (4-8) years. Seventy-five (19.8%) patients fulfilled the D2T criteria proposed by Perrotta et al. and 58 (15.3%) the D2T criteria proposed by Kumthekar et al. Both criteria showed a low sensitivity (Perrotta et al.: 37.8%, Kumthekar et al.: 29.7%) but good specificity (Perrotta et al.: 82.1%, Kumthekar et al.: 86.2%) with a comparable performance. Finally, the agreement between the two sets of criteria is substantial (Fleiss's kappa: 0.72) suggesting that both criteria identifies nearly the same group of patients. Conclusion: Our study compared two published sets of criteria showing comparable performance and substantial agreement. The study may pave the way for other researches in this field.