Incidence and Risk Factors for Steroid-associated Osteonecrosis in Children and Adolescents: A Systematic Review of the Literature.

Objective: Steroid-associated osteonecrosis in pediatric patients with inflammatory and oncologic disease is an uncommon yet debilitating condition causing significant functional disability. Pediatric orthopaedic surgeons encounter this population during stages in which surgical intervention may be necessary for joint preservation. Various risk factors for steroid-associated osteonecrosis have been suggested, but a comprehensive systematic review of the literature has not been performed. The purpose of this systematic review is to investigate incidence and risk factors for steroid-associated osteonecrosis in pediatric, adolescent, and young adult patients to help guide clinical decision-making.

Methods: We conducted a systematic review of the literature according to the preferred reporting items for systematic reviews and meta-analyses guidelines. MEDLINE, Embase, PubMed databases, and the Cochrane Central Registry of Controlled Trials were used to search for studies assessing risk factors for osteonecrosis in patients 0 to 21 years of age with systemic corticosteroid exposure. Two reviewers independently screened titles, abstracts, and full texts of retrieved studies for inclusion. Quality assessment of retrospective and prospective nonrandomized case-control and cohort studies was completed using the MINORS criteria. Outcomes and variables of interest included reported incidence and demographic, clinical, radiographic, and genetic risk factors for steroid-associated osteonecrosis. Reported statistics were deemed significant if P <0.05. Due to heterogeneous and limited reporting, data were not combined in a meta-analysis.

Results: The literature search revealed 895 articles and 37 articles were included. Of the included studies, 47% were retrospective cohort studies, and 39% were prospective cohort studies. There were 3 randomized controlled trials included. of the included studies, 95% were conducted in patients with leukemia and/or lymphoma. The overall prevalence of steroid-associated osteonecrosis ranged from 1% to 39%. Osteonecrosis was diagnosed with a mean or median of 1 to 2 years after the start of steroid therapy, and the most frequently involved joints were knees, followed by hips. Age older than 10 years, female gender, greater body mass index, and white and non-Hispanic race were the most reported risk factors for steroid-associated osteonecrosis. Core decompression was a frequent operative treatment with variable improvement in outcomes. For pediatric leukemia patients, those stratified as High risk and Intermediate risk were at the greatest risk for steroid-associated osteonecrosis.

Conclusion: This systematic review summarizes specific risk factors and demographics of steroid-associated osteonecrosis and helps lay the foundation for future studies to delineate the causal role of risk factors and guides clinical decision-making for current and proposed screening techniques. Steroid-associated osteonecrosis is often asymptomatic with clinical symptoms frequently lagging presentation on advanced imaging. The development of standard clinical pathways that incorporate screening for osteonecrosis may become necessary to improve outcomes through early detection and interventions such as core decompression to reduce pain and prevent progression to early osteoarthritis.