Abstract
Background: Isotretinoin, a frequently prescribed treatment for severe acne vulgaris, is associated with rare but clinically important musculoskeletal adverse effects, notably sacroiliitis. Although previous studies reported an association, comprehensive long-term data on clinical outcomes, imaging progression, and risk of evolution to axial spondyloarthritis (axSpA) remain limited. Basically, previous studies commonly excluded patients with spondyloarthritis-related risk factors such as chronic inflammatory back pain, but this limits the generalization. This study uniquely evaluated the clinical course and progression of isotretinoin-induced sacroiliitis using objective magnetic resonance imaging (MRI) without excluding patients presenting these risk factors. Methods: In this historical cohort study, patients with MRI-confirmed sacroiliitis during isotretinoin treatment were assessed. Clinical data including spondyloarthritis-associated features were recorded. Follow-up MRI performed at least 6 months later assessed inflammation resolution. Patients were classified according to Assessment of SpondyloArthritis International Society axSpA criteria. Statistical analyses were descriptive. Results: Among the 16 patients, 14 underwent follow-up MRI; of these, 2 (14.3%) demonstrated persistent inflammation. Additionally, 1 patient met the Assessment of SpondyloArthritis International Society axSpA criteria clinically without repeat imaging. Overall, 3 patients (18.8%) progressed to axSpA, all with chronic inflammatory back pain, and 1 patient also had psoriasis. Half of the patients initially presenting with inflammatory back pain (3/6) progressed to axSpA during follow-up. Conclusion: Although most isotretinoin-induced sacroiliitis resolves spontaneously, approximately one-fifth of cases may progress to axSpA, especially in patients with inflammatory back pain or other spondyloarthritis features. Clinicians should closely monitor isotretinoin-treated patients developing new or worsening back pain. Longitudinal studies are warranted to establish risk factors and optimize screening strategies.

This publication has 31 references indexed in Scilit: