The aims of this study were to look for clinical or serological markers able to predict the use of biological disease-modifying anti-rheumatic drugs (bDMARDs) in patients with adult-onset Still's disease (AOSD) and to evaluate the efficacy and safety of bDMARDs in AOSD. In a single-center retrospective study, 39 patients with AOSD were divided into two groups according to whether they were ever treated with bDMARDs or not. Literature was searched for articles dealing with possible predictors of the use of bDMARDs in AOSD. Among the 18 AOSD patients who received at least one bDMARD, the prevalence of pericarditis was higher than that in the other patients [p = 0.014, odds ratio (OR) = 13.4, 95 % confidence interval (CI) = 1.45 to 122]. Literature search retrieved another paper dealing with predictors of bDMARDs need in AOSD: the analysis pooling data from our series and this previous report confirmed pericarditis at disease onset as a predictor of bDMARDs need (p = 0.028, OR = 3.62, 95 % CI = 1.22 to 10.7). A complete remission was observed in 17 out of 18 patients treated with bDMARDs, allowing withdrawal or tapering of corticosteroid therapy (p < 0.001), but because of inefficacy or adverse events, some patients received more than one bDMARD during the course of the disease and 31 different trials of bDMARDs were needed. Pericarditis at disease onset may be a predictor of bDMARDs need in AOSD. These drugs have a good efficacy and safety profile and should be considered for patients not responding to conventional therapy.
Keywords: Adult-onset Still’s disease; Autoinflammatory diseases; Biological disease-modifying anti-rheumatic drugs; Pericarditis.