Osteonecrosis in SLE: prevalence, patterns, outcomes and predictors

Abstract

Objective Osteonecrosis is a serious comorbidity in patients with systemic lupus erythematosus. The aims of this study were to describe the prevalence of symptomatic osteonecrosis, determine the pattern of joint involvement, identify the outcomes and investigate predictive factors in a large cohort of patients with systemic lupus erythematosus followed prospectively. Methods At the Toronto Lupus Clinic patients have been followed prospectively according to a standard protocol since 1970. Osteonecrosis is recorded if patients are symptomatic and is confirmed by imaging. The site of osteonecrosis is recorded and whether or not surgery was performed. For determination of prevalence, pattern and outcome of osteonecrosis a longitudinal cohort design was performed. For the predictive factors, only patients with incident osteonecrosis were included and were matched for gender, year of entry to clinic (within 5 years), year of birth (within 5 years) and disease duration (within 3 years) with systemic lupus erythematosus patients without osteonecrosis. Results Of 1729 patients with systemic lupus erythematosus registered in the database, 234 (13.5%) developed symptomatic osteonecrosis in 581 sites. Hips and knees were most commonly affected and 47% of the patients had multiple sites involved. More than half of the joints involved at first occurrence of osteonecrosis had surgery. Univariate analysis identified black race, damage, elevated cholesterol and glucocorticosteroids as predictive factors, but glucocorticosteroids remained as the primary predictor for the development of osteonecrosis on multivariable analysis. Conclusion Despite advancements in the assessment and treatment of systemic lupus erythematosus, symptomatic osteonecrosis continues to be a significant comorbidity. Strategies to minimize glucocorticosteroid use are necessary to prevent this serious complication.

Keywords: Systemic lupus erythematosus; osteonecrosis; outcomes; predictive factors.