Clinical Predictors in Chronic Articular Adult-Onset Still's Disease: Comment on the Article by Ichida et al


We read with great interest the article by Ichida et al published recently in Arthritis Care & Research regarding the clinical manifestations of a large cohort of patients with adult-onset Still's disease (AOSD) ([1]). We fully agree with the authors, and we would like to make a brief comment on both the possibility of erosive arthritis and the need for reliable clinical predictors of outcome to treat arthritis aggressively as soon as possible. The evolution of AOSD may be self-limited, intermittent, and chronic ([2]); the latter having the worst prognosis quoad valetudinem because of the erosion associated with articular involvement. In contrast to rheumatoid arthritis (RA), erosions are not an early event. The characteristics of osteoarticular involvement in AOSD are both erosions and joint space narrowing (JSN). Because of this, we focused on this topic using the Simple Erosions Narrowing Score (SENS) ([3]); in particular, we observed that an active polyarthritis persisting over 6 months from the onset, irrespective of the involved joints, was strongly (P < 0.001) associated with the development of a chronic course of the disease. This could be considered tautological, but it is worth considering that, as in early RA, the persistence of joint symptoms over time is the most reliable chronicity index.

Contrary to what was claimed by Ichida et al, in our study, higher levels of ferritin were significantly (P < 0.002) associated with a chronic disease course. Moreover, comparing serum ferritin levels in patients with a chronic systemic and chronic articular course at the time of diagnosis and 6 months later, we concluded that the persistence of high ferritin levels after appropriate treatment may predict a chronic articular course ([4]). In our study, we also identified factors associated with the rate of progress of the articular disease, as measured with the SENS. The mean annual increase in the SENS was associated with ferritin level (P < 0.001) and the Disease Activity Score in 28 joints (P < 0.001); in particular, the SENS increased faster for higher levels of these 2 predictors. The same applied separately to erosions (P < 0.001 and P < 0.08, respectively) and JSN (P < 0.001 for both). The mean annual increase slowed down with time (measured by the number of years following diagnosis) for JSN (P < 0.001) and the SENS (P = 0.002).

As far as our study is concerned ([3]), the clinical findings not only exhibited some inconsistencies with the case series by Ichida et al, but also with previous studies. Some possible explanations include a selection bias and also different genetic backgrounds.

As far as treatments with biologic agents are concerned, according to our experience and data from studies in the literature, we would like to point out that anti–tumor necrosis factor α (anti-TNFα) medications may be helpful in controlling both systemic and articular symptoms of refractory AOSD. TNFα blockade would not be as effective in slowing down the articular damage as in improving systemic and osteoarticular symptoms. This leads to the awareness that anti-TNFα agents should not be a conceivable first-line therapy in AOSD. Owing to a more targeted action and a more reassuring safety profile, as well as more evidence-based effectiveness, the interleukin-1 and interleukin-6 receptor antagonists anakinra and tocilizumab could actually be considered a first-choice therapy in AOSD patients.