Rheumatologic manifestations of sarcoidosis and increased risk of spondyloarthritis occurrence. A retrospective single center case-control study

International audienceAn association between sarcoidosis and spondyloarthritis (SpA) has already been reported. The prevalence of sacroiliitis in sarcoidosis was estimated between 6.6% and 23.4%. All patients with sacroiliitis were diagnosed with SpA according to the Assessment of SpondyloArthritis International Society (ASAS) classification criteria. However, sacroiliac joint involvement has been described in sarcoidosis without associated SpA.This association led us to determine a clinical form of sarcoidosis predictive of SpA development and to evaluate their respective management.We conducted a retrospective, single-center, multiparameter, case-control study that enrolled patients with a sarcoidosis/SpA sequence, followed up between 1992 and 2018, who were randomly matched to patients with sarcoidosis without SpA development during the same period. Identification of patients was performed via a semantic query in the database of the Normandy Clinical Data Warehouse (NCDW). The diagnoses of sarcoidosis and SpA were based on ATS/ERS/WASOG and ASAS diagnostic and classification criteria, respectively. The number of events was gathered under one entity based on sarcoidosis organ involvement. The predictive effect of different variables, collected at the time of sarcoidosis diagnosis, on SpA occurrence was studied using conditional logistic regression models (PROC PHREG). This study has been approved by the local Institutional Review Board (E2020-91). Twenty-two patients with a sarcoidosis/SpA sequence were identified; 66 controls were selected. The proportion of organ involvement related to sarcoidosis was similar between the two groups except for rheumatologic manifestations. In cases, the mean (SD) delay to SpA diagnosis was 7.7 years [range 0–25]. The occurrence of SpA was significantly associated with the rheumatologic involvement of sarcoidosis (OR 6.42 95%CI [2.30–19.59] P < 0.01 after Bonferroni's correction). This association remained significant in our age- and sex-adjusted model (OR 5.36 95%CI [1.80–17.19] P < 0.05). No specific rheumatologic clinical picture could be identified, but axial involvement, corresponding to inflammatory back pain and alternating buttock pain, was more frequently observed in cases, resulting in a higher proportion of patients presenting a combined pattern of axial and peripheral involvement. SpA characteristics of patients with sarcoidosis developing SpA are summarized in Appendix A