Synergistic effect of cumulative corticosteroid dose and immunosuppressants on avascular necrosis in patients with systemic lupus erythematosus. (September 2018)

• Record Type: Journal Article
• Authors: Kwon, H H; Bang, S Y; Won, S; Park, Y; Yi, J H; Joo, Y B; Lee, H S; Bae, S C
• Is Part Of: Lupus. Volume 27:Number 10(2018)
• Page Start: 1644
• Page End: 1651
• Publication Date: 2018-09
• Subjects: Systemic lupus erythematosus -- avascular necrosis -- cumulative corticosteroid -- immunosuppressant; Systemic lupus erythematosus -- Periodicals; 616.772005
• DOI: 10.1177/0961203318784648
• Languages: English
• ISSNs: 0961-2033

Abstract

Objectives: Avascular necrosis (AVN) is one of the most common causes of organ damage in patients with systemic lupus erythematosus (SLE) and often causes serious physical disability. The aims of this study were to investigate clinical risk factors associated with symptomatic AVN and to analyze their synergistic effects in a large SLE cohort in Korea. Methods: Patients with SLE were enrolled and followed from 1998 to 2014 in the Hanyang BAE Lupus cohort, and damage was measured annually according to the Systemic Lupus International Collaborating Clinics/American College of Rheumatology Damage Index (SDI). AVN was confirmed by imaging study if patients had symptoms. To determine risk factors for AVN, clinical, laboratory and therapeutic variables were analyzed by logistic regression. Relative excess risk due to interaction (RERI), attributable proportion (AP), and synergy index (S) were calculated to measure interactions between significant variables. Results: Among 1219 SLE patients, symptomatic AVN was the most common type of musculoskeletal damage (10.8%, n = 132). SLE patients with AVN showed an earlier onset age, demonstrated AVN more commonly in conjunction with certain other clinical manifestations such as renal and neuropsychiatric disorders, and received significantly higher total cumulative corticosteroid dose and immunosuppressive agents than did patients without AVN. However, in multivariable analysis, only two variables including use of a cumulative corticosteroid dose greater than 20 g (odds ratio (OR) 3.62, p = 0.015) and use of immunosuppressants including cyclophosphamide or mycophenolate mofetil (OR 4.51, p < 0.001) remained as significant risk factors for AVN. Patients with cumulative corticosteroid dose > 20 g and immunosuppressant use had a 15.44-fold increased risk for AVN, compared with patients without these risk factors ( p < 0.001). RERI, AP and S, which define the strength of interactions between two risk factors, were 9.01 (95% confidence interval (CI) 1.30–16.73), 0.58 (95% CI 0.36–0.81) and 2.66 (95% CI 1.42–4.99), respectively, supporting the presence of synergistic interactions in the development of symptomatic AVN in our Korean lupus cohort. Conclusions: An individual risk assessment for AVN development should be made prior to and during treatment for SLE, especially in patients with high-dose corticosteroid and immunosuppressant use regardless of clinical manifestations and disease activity.